Annual Reports in
MEDICINAL CHEMISTRY VOLUME
44 Sponsored by the Division of Medicinal Chemistry of the American Chemical Society Editor-in-Chief
JOHN E. MACOR Neuroscience Discovery Chemistry Bristol-Myers Squibb Wallingford, CT, United States Section Editors ROBICHAUD STAMFORD BARRISH MYLES PRIMEAU LOWE DESAI
Amsterdam Boston Heidelberg London New York Oxford Paris San Diego San Francisco Singapore Sydney Tokyo Academic Press is an imprint of Elsevier
ACADEMIC PRESS
Academic Press is an imprint of Elsevier Linacre House, Jordan Hill, Oxford OX2 8DP, UK 84 Theobald’s Road, London WC1X 8RR, UK Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA
First edition 2009 Copyright r 2009 Elsevier Inc. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email:
[email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://www.elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material. Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. ISBN: 978-0-12-374766-2 ISSN: 0065-7743 For information on all Academic Press publications visit our web site at www.elsevierdirect.com
Printed and bound in USA 09 10 11 12 13
10 9 8 7 6 5 4 3 2 1
CONTRIBUTORS
Neel K. Anand
339
James N. Leonard
149
Jan D. Andersson
501
Shawn Maddaford
27
Subhash C. Annedi
27
Ju¨rgen Maibaum
105
Gregory S. Bisacchi
379
Carl L. Manthey
211
Robert M. Borzilleri
301
Dennis J. McCarthy
501
Bruce D. Car
555
Nicholas A. Meanwell
397
Karen Miller-Moslin
323
Terry W. Moore
443
Bernard P. Murray
535
379
John M. Nuss
339
Charles S. Elmore
515
Brian T. O’Neill
David L. Feldman
105
Stefan Peukert
323
William R. Foster
555
M. Edward Pierson
501
Robert G. Gentles
3
William J. Pitts
247
Christer Halldin
501
Mark R. Player
211
Sangdon Han
227
Mimi L. Quan
189
51
Suman Rakhit
27
Jotham W. Coe John Gaetano D’Angelo Gene M. Dubowchik Jacques Dumas
Simon N. Haydar
71 359 3
71
Shridhar Hegde
577
Jailall Ramnauth
27
Christopher Paul Hencken
359
Hans Rollema
71
Warren D. Hirst
51
Stefan U. Ruepp
555
Shuanghua Hu
3
Thomas Ryckmans
129
Arthur T. Sands
475
397
Michelle Schmidt
577
Alvin Solomon Kalinda
359
Paul M. Scola
397
Rao Kalla
265
Joanne M. Smallheer
189
John A. Katzenellenbogen
443
Roland G.W. Staal
Robert M. Jones John F. Kadow
149, 227
51
xv
xvi
Contributors
Nicole van Straten
171
Dolatrai M. Vyas
281
Jayant Thatte
227
Stephen T. Wrobleski
247
Thomas N. Thompson
459
Mark D. Wittman
281
Marco Timmers
171
Heedong Yun
51
Amy Lew Tsuhako
339
Jeff Zablocki
265
Upender Velaparthi
281
Brian P. Zambrowicz
475
Gregory D. Vite
301
PREFACE
Annual Reports in Medicinal Chemistry has reached Volume 44. I hope that it continues to be the review resource for medicinal chemists. Volume 44 continues the traditions of Annual Reports in Medicinal Chemistry with 28 chapters covering the themes of central nervous system disease, cardiovascular and metabolic diseases, inflammation/pulmonary/gastrointestinal (GI), oncology, infectious disease, topics in biology, topics in drug design and discovery and finally our review of new drugs introduced in 2008 in the ‘‘To Market, To Market’’ section. I am particularly pleased that Volume 44 contains three case histories detailing the discovery and development of varenicline for smoking cessation, aliskerin for hypertension and ixabepilinone for cancer. We will continue to seek case histories for future volumes because I believe these successes are some of the most instructive stories in medicinal chemistry. It is also gratifying to see the breadth of the sources of the reviews in Volume 44. My colleagues at Bristol-Myers Squibb continue to embrace the series with seven contributions in Volume 44. Scientists from Pfizer and AstraZeneca each contributed three chapters to Volume 44, while Arena Pharmaceuticals and Novartis each contributed two chapters. Wyeth, Neuraxon, Schering-Plough, Johnson&Johnson, CV Therapeutics, Exelixis, Lexicon Pharmaceuticals and Gilead scientists each contributed one chapter to Volume 44. Finally, chapters from Alfred University, the University of Illinois and an individual consultant completed the line up for Volume 44. I will continue to look to increase the diversity of contributing organizations and urge those organizations with significant medicinal chemistry resources who have not contributed recently to ‘‘step up to the plate’’ and contribute to this community resource known as Annual Reports in Medicinal Chemistry. Although we all worry about the mergers and resulting contractions of medicinal chemistry departments, we remain a vibrant science, and I am confident that Annual Reports in Medicinal Chemistry will continue to receive the quality contributions that have defined this series for over 40 years. Putting together Annual Reports in Medicinal Chemistry is a team effort of volunteers, starting with the chapter authors themselves. I thank the contributors to Volume 44 for their dedication and talent. Helping bring
xvii
xviii
Preface
this all together are the Section Editors: Joel Barrish, Manoj Desai, John Lowe, David Myles, John Primeau, Albert Robichaud and Andrew Stamford. I thank them for yet another seamless quality effort. Helping them and me were a team of reviewers/proof readers that have done a spectacular job behind the scenes as well. I acknowledge these reviewers/proof readers by listing their names below as a demonstration of our appreciation for their time and effort. AstraZeneca — Mike Barbachyn and Brian Sherer Bristol-Myers Squibb — Stephen Adams, Joanne Bronson, James Corte, Andrew Degnan, Murali Dhar, Douglas Dischino, James Duan, Carolyn Dzierba, Rick Ewing, Matthew Hill, John Hynes, George Karageorge, Lawrence Marcin, Ivar McDonald, Harold Mastalerz, Michael Miller, Natesan Murugesan, Richard Olson, Kenneth Santone, Michael Sinz, Lawrence Snyder, John Starrett, Drew Thompson, Dolatrai Vyas, Michael Walker and Christopher Zusi Gilead Sciences — Randall Halcomb, Richard Mackman and Will Watkins Novartis — Lawrence Hamann Pfizer — Joe Brady and Joel Morris Schering-Plough — Brian McKittrick Wyeth — Jonathan Gross, Steven O’Neil and Dane Springer Finally, I also acknowledge Shridhar Hedge and Michelle Schmidt for compiling the ‘‘To Market, To Market’’ review once again. I see that section as one of the consistent highlights of the book. And last, I also thank Ms. Catherine Hathaway, my Administrative Assistant, who always keeps things on an even keel. In summary, I hope that Volume 44 of Annual Reports in Medicinal Chemistry continues to be a key reference for your medicinal chemistry pursuits. As Editor-in-Chief, I continue to look for ways to optimize, improve and evolve the series. Please do not hesitate to contact me with suggestions for improving the series (
[email protected]). Thank you. John E. Macor Bristol-Myers Squibb, R&D, Wallingford, CT, USA
CHAPT ER
1 Recent Advances in the Discovery of GSK-3 Inhibitors and a Perspective on their Utility for the Treatment of Alzheimer’s Disease Robert G. Gentles, Shuanghua Hu and Gene M. Dubowchik
Contents
1. 2. 3. 4.
Introduction Etiology of Alzheimer’s Disease Principal Functions of GSK-3 Additional Neuroprotective Indications for GSK-3 Inhibition 5. Selective Functional Inhibition of GSK-3 6. GSK-3 Isoforms 7. GSK-3 Crystal Structures 8. Challenges in the Development of Effective GSK-3 Inhibitors 9. Advances in GSK-3 Inhibitors 9.1 Indirubins 9.2 Maleimides 9.3 Organometallic GSK-3 inhibitors 9.4 Thiadiazolidinones: noncompetitive GSK-3 inhibitors 9.5 Established drugs with GSK-3 inhibitory properties 9.6 Miscellaneous chemotypes 10. Conclusion References
3 4 5 7 8 9 9 9 10 11 13 14 18 19 20 22 22
Bristol Myers Squibb Company, 5 Research Parkway, Wallingford, CT 06492, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04401-7
r 2009 Elsevier Inc. All rights reserved.
3
4
Robert G. Gentles et al.
1. INTRODUCTION Alzheimer’s disease (AD) is the seventh leading cause of death in the United States and the fifth leading cause of mortality in people over the age of 65. Currently more than 5 million suffer from the condition and this number is expected to grow significantly with progressive aging of the population, rising to between 11 and 16 million affected individuals by 2050 [1]. It is estimated that the cost of treating AD and other dementias currently exceeds $148 billion annually in direct and indirect costs, and this is expected to increase dramatically with projected demographic changes. Current therapies are mostly palliative [2], and there exists a significant unmet medical need in the treatment of this devastating condition. AD is a dementia characterized symptomatically by progressive agerelated memory loss and cognitive impairment. Histopathologically, the disease is defined by neuronal loss, the presence of intracellular neurofibrillary tangles (NFTs), the formation of extracellular senile plaques [3], and the occurrence of cerebral amyloid angiopathy (CAA) [4]. NFTs are composed of aggregated hyperphosphorylated forms of the microtubule-associated protein tau [5]. This aggregation is thought to impair intracellular transport mechanisms [6] and may result in neuronal dystrophy [7]. Extracellular plaques are composed of precipitated amyloid beta-peptide (Ab) and are frequently associated with activated microglia, inflammation, and neuronal atrophy [8]. In CAA, Ab is deposited in cerebral arteries where it is associated with increased risk of hemorrhagic stroke [9,10].
2. ETIOLOGY OF ALZHEIMER’S DISEASE The most long-considered theory on the cause of AD is the ‘‘amyloid cascade hypothesis’’ [3,11]. This posits that Ab overproduction leads directly to the formation of senile plaques and CAA [12,13]. The presence of significant degenerative neuronal processes in the area of plaques, and the death of smooth muscle cells in the vicinity of the Ab deposits in cerebral arteries, is highly suggestive of the toxicity of the Ab peptide [14,15]. Additionally, areas of the brain most significantly affected in AD show co-localization between Ab plaques and neuronal cell death [16]. There has been much discussion on the exact form of Ab that may be noxious [17]. Experiments have shown that under some conditions fibrillation of Ab is required to observe neurotoxic effects [18,19]. However, at present, the exact toxic species of Ab has yet to be unambiguously identified.
Recent Advances in the Discovery of GSK-3 Inhibitors
5
An additional aspect of Ab pathology is its induction of hyperphosphorylation of tau [20]. This is thought to be primarily mediated through the activation of glycogen synthase kinase-3 (GSK-3), which has been shown to be responsible for the phosphorylation of key tau epitopes known to be relevant in AD [21,22]. Hyperphosphorylated tau disengages from microtubules, resulting in their structural destabilization with concomitant negative effects on intracellular structures and transport mechanisms [6]. In addition, the uncomplexed hyperphosphorylated tau assembles into paired helical filaments (PHFs) that aggregate to produce the stereotypic intracellular NFTs associated with AD [23,24]. It should be noted that other kinases such as CDK5 are also known to hyperphosphorylate tau and have also been pursued as therapeutic targets [25]. A recently proposed alternative theory to the amyloid cascade hypothesis postulates that GSK-3 may play a more instigative role in the etiology of AD [26]. It has been suggested that aberrant wnt or insulin signaling results in increased GSK-3 function [27–29] and this is responsible for the observed hyperphosphorylation of tau and the formation of PHFs and NFTs. In addition, elevated GSK-3 activity may induce increased Ab formation through its action on g-secretase [30,31], and thereby give rise to the primary stereotypic pathology observed in AD. GSK-3 has also been demonstrated to be involved in mechanisms underlying memory and learning, and dysregulation of GSK-3 function may explain some of the early cognitive deficiencies observed in AD [26]. However, some recent research has suggested the potential for the induction of behavioral deficits if constitutive GSK-3 activity is overly suppressed [32]. While these hypotheses continue to be challenged, refined, and even reconsidered, there are currently no definitive refutations of either theory. Consequently, it seems reasonable to pursue therapeutic approaches focused both on the direct modulation of Ab levels and on the inhibition of GSK-3. Ultimately, the successful treatment of AD may require that multiple treatment options be explored in different patient populations, and that differently targeted therapeutics be exploited either sequentially or in combination.
3. PRINCIPAL FUNCTIONS OF GSK-3 GSK-3 is a proline-directed serine/threonine kinase. It effects the phosphorylation of a range of substrates and is involved in the regulation of numerous diverse cellular functions, including metabolism [33,34], differentiation, proliferation, and apoptosis [35]. GSK-3 is constitutively active, with its basal level of activity being positively modulated by
6
Robert G. Gentles et al.
phosphorylation on Tyr216 [36]. GSK-3 has a unique substrate selectivity profile that is distinguished by the strong preference for the presence of a phosphorylated residue optimally located four amino acids C-terminal to the site of GSK-3 phosphorylation [37]. Most commonly, GSK-3 activity is associated with inducing a loss of substrate function, such that GSK-3 inhibition will frequently result in increased downstream activity. GSK-3 is widely distributed in the brain [38,39] and as discussed above, a principal consequence of its dysregulation is the hyperphosphorylation of the microtubule-associated protein tau [20], as depicted in the bottom-right panel of Figure 1. This function of GSK-3 has been demonstrated both in cell culture [40] and in in vivo studies looking at tau phosphorylation [41] and NFT formation [22,42]. The exact form of GSK3 (complexed or free) responsible for the phosphorylation of tau has not been characterized.
GSK-3: Discrete Cellular Pools Wnt Pathway Non-Stimulated
P
P
P
P
P
P
β-Catenin P
IRS
P
P
AXIN
P
P
CK1
GSK-3
P P
β-Catenin β-Catenin β-Catenin
Z
Glycogen Z Synthase
APC
AXIN
P
GSK-3 Glycogen Synthase
Tau
PKB
Z Phosphatases
Phosphorylation of Tau
Microtubule
P
GSK-3 FRAT
PDK1
PI3K
APC
GSK-3
P
Wnt Pathway Stimulated
Insulin Signaling Pathway
Insulin
β-Catenin
GSK-3
P
Microtubule
Microtubule
P
P
CK1 Disassembles
PHF’s
NFT’s
Figure 1 Key elements in the primary cellular pathways involving GSK-3 discussed in this review. APC, adenomatous polyposis coli; CK1, casein kinase 1; IRS, insulin receptor substrate; PI3K, phosphoinositide 3-kinase; PKB, protein kinase B; PDK1, phosphoinositide-dependent kinase-1. (See Color Plate 1.1 in Color Plate Section.)
Recent Advances in the Discovery of GSK-3 Inhibitors
7
GSK-3 is also known to play a key role in glucose metabolism, and was first identified as the enzyme responsible for effecting the inhibitory phosphorylation of glycogen synthase [43], the result of which is to reduce the rate of conversion of glucose to glycogen, giving rise to elevations in blood glucose levels. This function of GSK-3 is controlled by insulin, operating through the signaling pathway [44] shown in the topright panel of Figure 1. The net effect of this cellular cascade is inhibition of GSK-3 by a process that exploits features of the substrate selectivity mechanism described above. Binding of insulin to its receptor leads indirectly to the activation of protein kinase B (PKB) and subsequent phosphorylation of a key serine residue in the N-terminal domain of GSK-3 [45]. The phosphorylated N-terminus functions as a pseudosubstrate, folding in a manner such that it occupies the substrate-binding cleft [46,47]. This conformation is stabilized by interactions between the phosphorylated serine and residues on the enzyme that constitute the substrate selectivity pocket. An unrelated mechanism of GSK-3 inhibition operates in the wntsignaling cascade, a cellular pathway involved in controlling cell-fate, differentiation, and proliferation [48,49]. In this system, GSK-3 is complexed with APC, axin, and b-catenin, as well as other proteins [50], as depicted in the top-left panel of Figure 1. This protein assembly is occasionally termed the ‘‘destruction complex’’. When the wnt system is in a non-stimulated state, GSK-3 phosphorylates axin and APC, the effect of which is to create a more tightly associated complex. CK1, which is also associated with this protein assembly, functions in one capacity as a priming kinase for GSK-3 by performing an initial phosphorylation of b-catenin [51]. This protein is then poly-phosphorylated by GSK-3, resulting in its dissociation from the complex and its subsequent ubiquitination and destruction by the proteosome [52], thereby regulating cellular levels of b-catenin [53,54]. However, following binding of wnt ligands to their receptors, axin is displaced from the complex with GSK-3 as a result of binding of the latter to FRAT (frequently rearranged in advanced T-cell lymphomas) peptide [55], see bottom-left panel of Figure 1. This leads to the dissociation of the destruction complex, the consequence of which is that b-catenin is no longer effectively phosphorylated and degraded. Consequently, intracellular concentrations of the protein rise, and there is increased trafficking of b-catenin to the nucleus [56] where it interacts with transcription factors that induce increased expression of wnt regulated genes, which, among other things, are operative in certain neoplastic transformations. It is interesting to note that no crossover in these inhibition mechanisms is thought to occur: GSK-3 mediated phosphorylation of axin and b-catenin is not inhibited by a phosphorylated hexapeptide that is effective at inhibiting GSK-3 phosphorylation of primed substrates
8
Robert G. Gentles et al.
[37], suggesting that N-terminal serine phosphorylation is not operative in controlling GSK-3 activity within the destruction complex. In addition, wnt signaling does not affect the phosphorylation state of uncomplexed GSK-3 [57]. Correspondingly, the effects of insulin or wnt signaling would appear to be limited to discrete cellular pools of GSK-3, the significance of which is discussed below.
4. ADDITIONAL NEUROPROTECTIVE INDICATIONS FOR GSK-3 INHIBITION GSK-3 has been demonstrated to be involved in apoptosis regulation, and it has been suggested that its inhibition can have neuroprotective effects distinct from prevention of tau hyperphosphorylation. Mitochondria, for example, have been found to be susceptible to unregulated GSK-3 activity, resulting in Parkinson’s disease (PD)-like effects [58]. In addition, GSK-3 inhibition can attenuate apoptosis resulting from MPP+ and rotenone challenges that mimic PD neuronal pathology [58]. GSK-3 inhibition has also been implicated as a target for amyotrophic lateral sclerosis (ALS). Highly phosphorylated tau protein, along with upregulated GSK-3 has been observed in ALS with cognitive impairment [59]. A recent study has shown that chronic lithium dosing (vide infra) to ALS patients (2 150 mg/day) significantly delayed disease progression and increased survival time in comparison with those taking riluzole, a neuroprotective agent. In addition, lithium was shown by the same group to be protective in a mutant superoxide dismutase 1 (SOD1) model of ALS [60]. Clinical trials with lithium (alone, and in combination with riluzole) in ALS are ongoing [61].
5. SELECTIVE FUNCTIONAL INHIBITION OF GSK-3 It is apparent that inhibition of GSK-3 may be associated with significant mechanism-based toxicities, potentially ranging from hypoglycemia to tumorigenesis. To successfully develop a GSK-3 inhibitor it may be necessary to identify agents that can selectively inhibit specific cellular functions of GSK-3. How might this be realized? One potential mechanism could involve targeting or obviating the inhibition of the discrete cellular pools described above. It is conceivable that this could be achieved by exploiting subtle differences in the structures of GSK-3 that might arise when the enzyme is associated within different protein assemblies, or when the enzyme is in an uncomplexed form. Alternatively, if inhibitors can be identified that interact in the vicinity of
Recent Advances in the Discovery of GSK-3 Inhibitors
9
the ATP-binding site it may be possible to achieve substrate-selective inhibition as has been observed in the development of gamma secretase inhibitors [62]. Hypothetically, the identification of allosteric inhibitors [63] could afford very different and potentially unique functionally selective inhibition profiles [63], as could the development of isoform-selective GSK-3 inhibitors (vide infra). To date, the primary literature is devoid of such information but it might safely be assumed that acquisition of data of this type is a feature of many preclinical programs.
6. GSK-3 ISOFORMS GSK-3 exists in two isoforms, GSK-3a (51 kDa) and GSK-3b (47 kDa), that share 84% overall identity and greater than 98% identity within their respective catalytic domains [64,65]. A minor splice variant of GSK-3b, denoted as GSK-3b2, has been reported [66]. This isoform has a 13-amino-acid insert within its kinase domain and displays reduced activity toward tau protein. It has been found by immunohistochemistry to be localized predominantly in cell soma, unlike GSK-3b that is found extensively in neuronal processes. Both primary isoforms are ubiquitously expressed [67,68], with particularly high levels observed in the brain and testes. In most brain areas, GSK-3b is the predominant isoform and has become of primary interest as a CNS target. However, it should be noted that some studies suggest that GSK-3a and GSK-3b share very similar, if not entirely redundant functions in a number of cellular processes [69]. Therefore, the utility of an isoform-selective inhibitor could be determined by the particular cellular function targeted, as well as the relative isoform expression levels in the tissue of interest.
7. GSK-3 CRYSTAL STRUCTURES The crystal structure of GSK-3 has been reported [46], as have structures of GSK-3 complexed to components of the wnt-signaling pathway [50]. All these data have provided insight into the mechanism of action of this kinase, the nature of its substrate selectivity, and the unique mechanism of inhibition effected through phosphorylation of the N-terminal serine residue [70]. In addition, a number of co-crystal structures have been published of GSK-3 bound to a diverse set of active site inhibitors [71–73]. These data have been exploited in improving the intrinsic potency of these ligands, as well as suggesting pathways for improved kinase selectivity profiles. More recently, these crystal structures have
10
Robert G. Gentles et al.
been employed in a variety of virtual screening protocols directed at the identification of novel GSK-3 inhibitors [74,75].
8. CHALLENGES IN THE DEVELOPMENT OF EFFECTIVE GSK-3 INHIBITORS In common with all kinase-targeted therapeutics, a key issue to be addressed relates to selectivity. This must be sufficient to avoid off-target toxicities, and this may be more important for CNS-targeted drugs than compounds designed for use in oncology, where targeting multiple kinases can occasionally be advantageous. In general, the broader the kinase counter screen used in determining selectivity, the greater the confidence that this problem can be adequately assessed. However, in the extant GSK-3 literature, limited selectivity data are disclosed beyond that related to phylogenetically related kinases. Historically, this can be attributed to technological limitations in large-scale parallel kinase screening. However, recent developments [76] have largely eliminated such restrictions and more extensive kinase counter-screening data are now beginning to be disclosed [77]. A specific issue for the use of GSK-3 inhibitors in AD is the need to access the CNS compartment. Given the polar nature of many kinase inhibitors it would be desirable to include early assessments of blood brain barrier (BBB) permeability and P-glycoprotein (Pgp) substrate potential. In addition, having suitable brain-to-plasma ratios may be a determinant in achieving an adequate therapeutic index by limiting the peripheral exposure required to drive CNS efficacy. Again, few data in this area have been offered, and most reports are limited to the discussion of biochemical activities with limited disclosures of in vivo studies. With regard to mechanism-based toxicities, the primary concern for GSK-3-targeted therapeutics relates to their potential to induce transformation of nonmalignant cells or exacerbate preexisting malignancies through their actions on b-catenin (vide supra). It is worth recalling however, that lithium has been used as a standard therapeutic for the treatment of bipolar disorder since the 1950’s. This agent is a weak GSK-3 inhibitor, exerting its effect through a mixed mechanism of direct inhibition [78] and activation of PKC-a [79], the latter of which leads to increased GSK-3 Ser9/21 phosphorylation. At therapeutic doses, lithium is estimated to achieve B25% inhibition of total GSK-3 activity, and this degree of inhibition has not yet been associated with increased levels of tumorigenesis or deaths from cancer. This is perhaps the most compelling argument for the potential safety of GSK-3 inhibitors.
Recent Advances in the Discovery of GSK-3 Inhibitors
11
9. ADVANCES IN GSK-3 INHIBITORS Over the past 20 years a number of chemical classes of GSK-3 inhibitors have been discovered [80]. More recently, the structural diversity of compounds reported in this area has expanded considerably, especially in the patent literature. However, in light of the preceding discussion we limit ourselves here to a review of GSK-3 inhibitors on which additional data beyond that of the primary enzyme activity have been disclosed, such that the utility of these compounds in the treatment of AD might be better assessed.
9.1 Indirubins The indirubins are a class of natural product bis-indoles that can be isolated from a variety of plant and animal sources, or alternatively, can be relatively easily synthesized from appropriately functionalized indoles and isatins. They have been demonstrated to possess potent kinase inhibition activities, and considerable efforts over a number of years have resulted in the identification of GSK-3-selective analogs [81]. An issue constraining their biological evaluation is their characteristically low aqueous solubility. Recent work has focused on addressing this limitation while attempting to maintain required potency and selectivity profiles. A number of co-crystal structures of indirubins complexed with various kinases have been reported. These have proved valuable in understanding the molecular features that determine affinity and selectivity, and have contributed to the prioritization of vectors that could be exploited for the modulation of off-target and physicochemical properties. In the case of the co-crystal structure of 6-bromo-indirubin-3u-oxime (6BIO) with GSK-3, the principal interactions between the ligand and the enzyme have been identified [82,83]. From the extant structure activity relationship (SAR) data, it is known that bromine at the 6-position of the indirubin core can be a key determinant of selectivity for GSK-3 over the phylogenetically related cyclin-dependent kinases, CDK1 and CDK5. From the co-crystal data this can be rationalized by noting that bulky substituents at position 6 take advantage of the sterically more accommodating ‘‘gatekeeper’’ residue of GSK-3 (Leu132) compared to the more demanding related residue (Phe80) for the cyclin-dependent kinases (CDKs). Furthermore, from this structure it is apparent that the 3u-oxime projects out from a solvent-accessible cavity and would appear to offer a vector from which to extend solubilizing functionality. This supposition was supported in a series of analogs that incorporated a basic moiety appended with a hydrocarbon linker bound to the oxygen atom of the oxime [84]. A diverse range of groups were examined and the molecules were evaluated for potency against GSK-3, CDK1/cyclinB, and CDK5/p25 and for cytotoxicity
12
Robert G. Gentles et al.
in an SH-SY5Y cell line. Compound 1 (GSK-3, IC50 5 nM; CDK1/cyclinB, IC50 600 nM; and CDK5/p25, IC50 500 nM), compound 2 (GSK-3, IC50 11 nM; CDK1/cyclinB, IC50 2,800 nM, and CDK5/p25, IC50 300 nM), and compound 3 (GSK-3, IC50 14 nM; CDK1/cyclinB, IC50 900 nM; and CDK5/ p25, IC50 310 nM), all displayed similar or improved potency and selectivity profiles relative to 6BIO (GSK-3, IC50 5 nM; CDK1/cyclinB, IC50 320 nM; and CDK5/p25, IC50 83 nM), and all were significantly less cytotoxic against an SH-SY5Y cell line. N Br
HO
N
Br
O
N
N HO NH
N H
N H
O 6BIO
O
(1)
N
N N
NH
Br
O
N
Br
O N
N O
O N H (2)
NH O
OH
N H
NH O
(3)
The cellular activity of this compound class was confirmed in a b-catenin reporter assay where 1 (IC50 700 nM) was demonstrated to be approximately equipotent to 6BIO (IC50 300 nM). Additionally, selected compounds also displayed activity in a cell culture luminescence assay of GSK-3-mediated circadian rhythmicity. Importantly, all of the above analogs were dramatically more soluble (0.1–4.2 mg/mL) than the parent bromo-indirubin (o5 mg/mL), a feature that should significantly facilitate the biological assessment of this compound class. In an unrelated study, a series of indirubins were prepared that were independently functionalized at positions 5- and 3u-, as depicted in the structures given below [85].
13
Recent Advances in the Discovery of GSK-3 Inhibitors
4′
O
HNAc
O2N
5
HO
6
4
O
N 7
5′
3′
6′ 7′
N H
NH 1
O
(4)
N H (5)
NH O
N H
NH O
(6)
Of the many reported analogs, only compounds 5 (GSK-3, IC50 2 nM; CDK1/cyclinB, IC50 19 nM; and CDK5/p25, IC50 6 nM) and 6 (GSK-3, IC50 7 nM; CDK1/cyclinB, IC50 50 nM; and CDK5/p25, IC50 18 nM) displayed single-digit nanomolar activity, and none of the analogs were particularly selective for GSK-3. Interestingly however, there appeared to be interdependent SARs at the positions explored, with the optimal group at one vector being dependent on the functionality incorporated at the other positions explored. This would argue that a matrix-based optimization of this chemotype could lead to further enhancements in potency and selectivity.
9.2 Maleimides The natural product staurosporine (7) [86] is a broadly promiscuous polycyclic kinase inhibitor. The related bis(indolyl)maleimide ruboxistaurin (8), although less conformationally constrained, is relatively specific in its inhibition of PKC-b, having a reported IC50 of 5 nM and no other significant activity in a broad kinase counter screen [87]. This selectivity has been attributed in part to the nonplanar disposition of the two indole moieties in 8, and recently reported work [73] has focused on the synthesis of analogs of 8 that exploit a cyclophane structure, as shown in compounds 9 through 14. These molecules also present a nonplanar arrangement of the bis-indolyl moiety, but in a more constrained ‘‘pyridinophane’’ cycle, and it was anticipated that this might favorably impact the potency and selectivity profiles previously observed with other macrocyclic bis(indolyl)maleimides.
14
Robert G. Gentles et al.
H N
H N
O
O
O R1
N
N
N
N
O R3
Ru
R2 O
MeO HN (7)
(15) R1 = R2 = R3 = H (16) R1 = OH, R2 = Br, R3 = CO2CH3
Expansion of the 14-membered macrocycle in ruboxistaurin to 16–22membered rings has previously been shown to not only retain excellent PKC-b inhibitory activity but also introduce significant activity against GSK-3. Contrastingly, the larger ring pyridinophanes 9 through 14 all display potent GSK-3 activity and are highly selective against PKC-b. For example, compound 13 inhibited GSK-3b and PKC-bII with IC50 values of 3 and 1,400 nM, respectively. This potency and selectivity profile had not previously been reported for any of the known bis(indolyl)maleimides. To further assess the broader kinase selectivity of this novel series of pyridinophanes, compounds 9, 10, and 12 were screened against a panel of 100 kinases. With the exception of the GSK-3a and GSK-b isozymes, the only other significant activities observed were against MSK1, PKC-y, and Rsk3. More specifically, compound 13 had IC50 values against these kinases of 510, 98, and 48 nM, respectively. A co-crystal structure of 13 with GSK-3b was obtained from which it was noted that the maleimide moiety makes key hydrogen bonds with the hinge residues, Asp133 and Val135, and an additional hydrogen bond through a bridging water molecule to Asp200. A number of hydrophobic interactions were observed, but interestingly, the 2-dimethylaminopyridine group is largely solvent exposed and does not interact directly with the enzyme, suggesting a site for further modification [73].
9.3 Organometallic GSK-3 inhibitors Organometallics have been used for a considerable time as pharmaceutical agents, especially in the oncology arena [88]. More recently, interest has grown in the use of transition metal complexes in chemical genetics, where it is believed that they may access unexplored chemical space, and thereby present opportunities for the design of small molecules with novel biological properties [89]. More specifically, given their unique
Recent Advances in the Discovery of GSK-3 Inhibitors
15
geometries, these compounds may be useful additions to the tools available to address some of the unanswered questions raised in the introduction of this review. In the area of kinase inhibitors, metal complexes are an expanding area of interest, and several highly potent and selective GSK-3 inhibitors have recently been disclosed. In all of the complexes discussed, the metals are thought to be biologically inert and function solely by preorganizing their ligands to present a complementary surface to cognate proteins. This can be appreciated by comparing the structurally complex staurosporine 7, with the simpler ruthenium complex 15, in which the indolocarbazole alkaloid scaffold has been replaced with a somewhat simpler bidentate pyridocarbazole ligand that retains the main features of the indolocarbazole aglycone [90]. The remaining metal ligands project into areas occupied by the glycoside element of staurosporine, and both structures have related overall topologies. Importantly, these molecular geometries are significantly easier to explore with transition metal complexes than they are with many organic macrocycles that are typically accessed by extended syntheses. Additionally, the complexes reviewed here have been shown to be stable in air and aqueous solution, and are tolerant of millimolar concentrations of thiols. H N
O
N O
Ru
O
N Cl NH
(17)
The racemate of the ruthenium complex 15 has IC50 values for GSK-3a and GSK-3b of 20 and 50 nM (100 mm ATP), respectively. In addition, it was demonstrated to inhibit Pim-1 kinase with an IC50 of 3 nM, as well as MSK1 (IC50 120 nM), nMRsk1 (IC50 300 nM), and TrkA (IC50 70 nM), but did not significantly inhibit other kinases in a panel of 50 isozymes explored, including the phylogenetically related CDKs [90]. To address these cross-reactivities, a limited SAR study resulted in the identification of compound 16 that was shown to be selective for GSK-3
16
Robert G. Gentles et al.
when screened in a panel of 57 kinases. This compound was resolved by chiral HPLC and the R isomer was highly potent against GSK-3a and GSK-3b (IC50 350 and 550 pM, respectively), and the previously noted activities against Pim-1, MSK1, nMRsk1, and TrkA were all significantly attenuated. Interestingly, the S isomer has the reverse selectivity for Pim1 over GSK-3 [90]. The complex 16 is configurationally stable in solution when stored in the freezer, but can be expected to slowly epimerize under physiological conditions. Of note is the fact that the uncomplexed pyridocarbazole ligand of 16 is a more potent Pim-1 inhibitor (IC50 10 nM) than GSK-3a (IC50 30 nM) or GSK-3b inhibitor (IC50 50 nM). Consequently, complexation of this ligand to the ruthenium half-sandwich fragment of 16 increases the potency for GSK-3 by more than an order of magnitude, and reverses the selectivity profile for GSK-3 over Pim-1 [89]. The cellular activity of (R)-16 was determined in a b-catenin-dependent luminescence reporter assay, in which the degree of luminescence relates to cellular b-catenin levels. Activity was observed at 30 nM and a luminescence increase factor of 700 was determined, indicative of significant b-catenin stabilization. This level of potency was significantly greater than that observed with the longer-studied GSK-3 inhibitor 6BIO, and confirmed the ability of these complexes to permeate cells and inhibit GSK-3, including the fraction of GSK-3 bound in the b-catenin destruction complex. Zebrafish embryos that were exposed to micromolar concentrations of (R)-16 demonstrated a number of developmental abnormalities, strongly suggesting that effective functional inhibition of GSK-3 had been achieved in vivo [89]. In an interesting extension to this work, a series of octahedral ruthenium complexes incorporating a bidentate pyridocarbazole in combination with a diversity of other ligands were explored for activity against GSK-3a, Pim-1, MSK1, and CDK2/cyclinA [91]. When bound to a target kinase, the pyridocarbazole makes key H-bond contacts with the target enzyme, and the metal and its remaining ligands occupy the approximately spherical ribose pocket of the enzyme’s active site. In the case of octahedral complexes, the choices of substituents on the four valences projecting into the sugar pocket are key determinants of selectivity and, as such, achieve this in a different fashion compared to many organic inhibitors. Of the octahedral complexes explored, compound 17 was identified as a potent GSK-3a inhibitor (IC50 8 nM) and was 10-fold selective over Pim-1.
Recent Advances in the Discovery of GSK-3 Inhibitors
H N
O
17
O
HO F O
N
N Ru
NH
O
O OH (18)
A space-filling model of 17 showed that the ruthenium atom is buried within the complex and is not able to directly interact with target proteins. As previously noted, the metal functions primarily to organize its coordinated ligands into specific and well-defined geometries. In addition, complexes such as 17 are relatively rigid and kinetically inert under physiological conditions. The most interesting feature of compound 17 is its observed selectivity for GSK-3a (IC50 8 nM) over GSK3b (IC50 50 nM). Selectivity between these isozymes is somewhat surprising given the high homology of their active sites (W97%) and has not been previously reported. Understanding the structural features contributing to this differential activity would be a valuable addition to the field. More recent studies directed at the further optimization of ruthenium-based GSK-3 inhibitors resulted in the identification of an extremely high-affinity GSK-3 inhibitor [92]. The complex (RRu)-NP549 (18) inhibits GSK-3b with a Ki of 5 pM and is one of the most potent kinase inhibitors yet reported. H N
O
O
H N
O
O
OH N
N Ru
N
N Os
O
(19)
OH
O
(20)
A co-crystal of (RRu)-NP549 and GSK-3b has been obtained and shows a remarkable complementarity between the surfaces of the
18
Robert G. Gentles et al.
ruthenium complex and the ATP-binding site of GSK-3. Nearly all of the polar functionality in 18 makes productive H-bonds with the enzyme. Interestingly, the CO ligand contacts the glycine-rich loop and is buried in a small pocket, contributing to the affinity and selectivity of this compound. In an unrelated study, the activities of a pair of ruthenium 19 and osmium 20 congeners of 18 were compared [93]. Both analogs were evaluated in a variety of biochemical and cell assays and had essentially indistinguishable properties. Both 19 and 20 were highly active at GSK-3b (IC50 1.4 and 0.6 nM, respectively), and both induced strong and essentially identical apoptotic effects in 1205 Lu melanoma cells. This effect is thought to be attributable to GSK-3 inhibition that produces p53-induced apoptosis. The equivalence of these ruthenium and osmium complexes in a cytotoxicity assay using 1205 Lu cells is noteworthy, as such exchange of one metal with its higher periodic table congener is not normally tolerated in many classes of cytotoxics. R2 N
S O
O
N R1
(21) R1 = Bn, R2 = Me: NP031112 (22) R1 = Ph, R2 = Et: NP01138 (23) R1 = Bn, R2 = Bn: NP00111
A co-crystal structure of the osmium analog 20 complexed to Pim-1 was obtained and displayed a binding mode similar to that for related ruthenium complexes. Not surprisingly, when 19 was modeled into this structure, the two analogs were essentially indistinguishable. This is consistent with their almost identical biological properties and with the hypothesis that the metals in these complexes function primarily as structural components and do not participate in direct interactions with target proteins.
9.4 Thiadiazolidinones: noncompetitive GSK-3 inhibitors Kinase inhibitors that are non-ATP-competitive are potentially attractive for several reasons. By not having to compete with endogenous ATP, they may show better cellular and in vivo potency in comparison with competitive inhibitors having comparable ADME properties. In addition, much better kinase selectivity may be expected from inhibitors that bind outside of the ATP pocket. This may be especially attractive for CNS
Recent Advances in the Discovery of GSK-3 Inhibitors
19
indications where target overlap is probably less desirable than it can be for anticancer therapies. A series of 1,2,4-thiadiazole-3,5-diones (TDZDs) have been described, examples of which appear to be noncompetitive with ATP [63]. The initial lead compound in this series, TDZD-8 (NP031112, 21), showed modest GSK-3b inhibition (IC50 2 mM), noncompetitive enzyme kinetics, and good selectivity against a limited set of other kinases (CDK1, CK2, PKA, and PKC). Putative binding sites, both inside and outside the ATP pocket, have been suggested through a combination of 3D-SAR and molecular docking analyses [94]. Dose-dependent reduction of tau phosphorylation in human SHSY5Y neuroblastoma cells has been reported in conference presentations (IC50 ¼ 51 mM). NP031112 was tested in GSK-3b conditional, and hAPPx tau transgenic mouse models (100–200 mg/kg/day, PO for 3 weeks and 3 months, respectively). Following these prolonged treatments, cognitive function was significantly improved and neuronal loss, amyloid deposition, tau phosphorylation, and neuroinflammation were reduced [95]. O H3CO H3CO N (24)
Further studies suggest that NP031112, as well as related compounds NP01138 (22) and NP00111 (23), may have more pronounced neuroprotective effects that may operate through alternate mechanisms, such as peroxisome proliferator-activated receptor g (PPARg) modulation under excitotoxic conditions [96]. Company reports state that NP031112 began phase I clinical trials for AD in April 2006.
9.5 Established drugs with GSK-3 inhibitory properties Donepezil (Aricepts, 24) is an acetylcholinesterase inhibitor that is widely used for the treatment of cognitive and behavioral symptoms of mild-to-moderate AD. Although its effect on disease progression has been a source of controversy, a clinical study has suggested that it may have benefit in severe AD patients [97]. A recent study has demonstrated GSK-3 inhibition in primary cortical neuron cultures challenged with Ab (20 mM) and treated with donepezil, resulting in neuroprotection (dosedependent, from 0.1 to 10 mM) and reduced tau phosphorylation (10 mM) [98]. The presence of a PI3 kinase inhibitor (LY294002) abolished activity,
20
Robert G. Gentles et al.
while a nicotinic acetylcholine blocker, mecamylamine, only partially reversed neuroprotection, suggesting that GSK-3 inhibition results from blockade of inhibitory phosphorylation. CN N
S
N NH N
NH
N
S
HN
N
N (25)
(26)
Olanzapine (25) is an atypical antipsychotic that is associated with weight gain and disturbances in glucose metabolism. A recent study demonstrated GSK-3 inhibition in brains of mice treated with atypical antipsychotics [99]. One group decided to look at direct inhibition of GSK-3 by 25, first by in silico molecular docking experiments utilizing the published co-crystallized structure of the known GSK-3 inhibitor AR-A014418, and then by enzyme assay [100]. Modeling suggested a good fit in the ATP pocket and, indeed, olanzapine was a fairly potent inhibitor of GSK-3b (IC50 91 nM). O F
OH
OH
N
HN
N
N
N
O Cl
N
NH2
(27)
O
(28)
N
Recent Advances in the Discovery of GSK-3 Inhibitors
21
O O HN S
NH2 N N
O
N
(29)
The same group carried out wider in silico docking studies on a structural database of established drugs and identified the following as potent inhibitors [101]: cimetidine (26) (IC50 13 nM), hydroxychloroquine (27) (IC50 33 nM), and gemifloxacin (28) (IC50 88 nM).
9.6 Miscellaneous chemotypes 9.6.1 Furopyrimdines A furopyrimidine scaffold that had previously yielded potent VEGFR2 and Tie-2 inhibitors was modified to provide a potent series of GSK-3 inhibitors [102]. This was done by rational design following analysis of the published binding modes (revealed by X-ray crystallography) of analogous aza- and diazaindazoles. An exemplary compound 29 demonstrated good binding affinity (IC50 30 nM), with good-to-excellent selectivity against a panel of nine other kinases including CDK2. Critical to GSK-3 selectivity and potency was the 3-pyridyl group whose nitrogen is expected to form a hydrogen bond with K85.
CN
F
H3CO
OCH3
S
O O
S
N
O
O
N
N N
N N (30)
(31)
9.6.2 1,3,4-Oxadiazoles A high-throughput screening effort resulted in the identification of a 2-aryl-1,3,4-oxadiazole hit, 30, with double-digit nanomolar potency against GSK-3b (IC50 65 nM) [71]. Subsequent optimization, guided by an
22
Robert G. Gentles et al.
X-ray structure of 30 in the GSK-3 ATP pocket, led to 31, a highly potent inhibitor (IC50 2.3 nM) with W1,000-fold selectivity against a panel of 23 other kinases, including CDK2. In a rat oral PK study, compound 31 showed low oral bioavailability (F : 1.7%), reportedly due to limited intestinal absorption.
F3C HN
N N
NH
HO OH (32)
9.6.3 2,5-Diaminopyrimidines One group’s recent screening efforts led to identification of a purinone scaffold that proved difficult to progress. However, investigation of synthetic intermediates led to a small series of adamantylaminopyrimidines that showed good GSK-3 inhibition. Compound 32 (GSK-3b IC50 41 nM) also showed good oral bioavailability in rats (F : 34%) [103]. N N
H N
Cl
N H
OH
N
N
(33)
9.6.4 Amino-1,3,5-triazines In a recent study, a series of biheteroaryl triazine CDK inhibitors were prepared and examined as potential antitumor agents [104]. The most potent compound in this class, 33 (CDK1, IC50 21 nM; CDK2, IC50 7 nM; and CDK5, IC50 3 nM), was similarly active against GSK-3b (IC50 20 nM), but fairly selective against a panel of 12 other kinases.
Recent Advances in the Discovery of GSK-3 Inhibitors
23
10. CONCLUSION In addition to the compounds discussed above, the patent literature is replete with examples of structures reported to be active at GSK-3, and it would appear reasonable to assume that compounds with sufficient potency, kinase-selectivity, and appropriate physicochemical properties can be identified to evaluate their use in the treatment of AD. However, the key outstanding issue to be addressed remains mechanism-based toxicity, and whether this can be avoided. This is unlikely to be known until a number of analogs (with a variety of inhibitory profiles) are comprehensively tested in vivo or significant clinical progress is reported [80].
REFERENCES [1] [2] [3] [4]
[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
2009 Alzheimer’s Disease Facts and Figures. Alzheimer’s & Dementia, 2009, 5, 30. R. Gilstad John and E. Finucane Thomas, J. Am. Geriatr. Soc., 2008, 56, 1556. J. Hardy, J. Alzheimer’s Dis., 2006, 9(Suppl. 3), 151. J. Zaghi, B. Goldenson, M. Inayathullah, A. S. Lossinsky, A. Masoumi, H. Avagyan, M. Mahanian, M. Bernas, M. Weinand, M. J. Rosenthal, A. Espinosa-Jeffrey, J. Vellis, D. B. Teplow and M. Fiala, Acta Neuropathol., 2009, 117, 111. I. Grundke-Iqbal, K. Iqbal, M. Quinlan, Y. C. Tung, M. S. Zaidi and H. M. Wisniewski, J. Biol. Chem., 1986, 261, 6084. H. Stebbings, Cytoskeleton, 1996, 2, 113. M. L. Michaelis, S. Ansar, Y. Chen, E. R. Reiff, K. I. Seyb, R. H. Himes, K. L. Audus and G. I. Georg, J. Pharmacol. Exp. Ther., 2005, 312, 659. K. Vehmas Anne, H. Kawas Claudia, F. Stewart Walter and C. Troncoso Juan, Neurobiol. Aging, 2003, 24, 321. C. G. Dotti and B. De Strooper, Nat. Cell Biol., 2009, 11, 114. K. Jellinger, J. Neurol., 1977, 214, 195. J. A. Hardy and G. A. Higgins, Science, 1992, 256, 184. G. G. Glenner and C. W. Wong, Biochem. Biophys. Res. Commun., 1984, 120, 885. C. L. Masters, G. Simms, N. A. Weinman, G. Multhaup, B. L. McDonald and K. Beyreuther, Proc. Natl. Acad. Sci. U.S.A., 1985, 82, 4245. J. W. Geddes, K. J. Anderson and C. W. Cotman, Exp. Neurol., 1986, 94, 767. M. Kawai, P. Cras and G. Perry, Brain Res., 1992, 592, 278. J. Rogers and J. H. Morrison, J. Neurosci., 1985, 5, 2801. J. B. Standridge and E. M. Welsh (eds), Trends in Alzheimer’s Disease Research, Nova Science Publishers, Hauppauge, NY, 2006, pp. 53–96. B. A. Yankner, Nat. Med., 1996, 2, 850. A. Lorenzo and B. A. Yankner, Ann. N.Y. Acad. Sci., 1996, 777, 89, Neurobiology of Alzheimers Disease. A. Ferreira, Q. Lu, L. Orecchio and K. S. Kosik, Mol. Cell. Neurosci., 1997, 9, 220. P. Delobel, S. Flament, M. Hamdane, A. Delacourte, J. P. Vilain and L. Buee, FEBS Lett., 2002, 516, 151. K. Leroy, Z. Yilmaz and J. P. Brion, Neuropathol. Appl. Neurobiol., 2007, 33, 43. K. Iqbal and I. Grundke-Iqbal, J. Alzheimer’s Dis., 2006, 9(Suppl. 3), 219.
24
Robert G. Gentles et al.
[24] C. M. Wischik, M. Novak, P. C. Edwards, A. Klug, W. Tichelaar and R. A. Crowther, Proc. Natl. Acad. Sci. U.S.A., 1988, 85, 4884. [25] M. A. Glicksman, G. D. Cuny, M. Liu, B. Dobson, K. Auerbach, R. L. Stein and K. S. Kosik, Curr. Alzheimer Res., 2007, 4, 547. [26] C. Hooper, R. Killick and S. Lovestone, J. Neurochem., 2008, 104, 1433. [27] G. J. Biessels and L. J. Kappelle, Biochem. Soc. Trans., 2005, 33, 1041. [28] G. V. De Ferrari, A. Papassotiropoulos, T. Biechele, F. W. De-Vrieze, M. E. Avila, M. B. Major, A. Myers, K. Saez, J. P. Henriquez, A. Zhao, M. A. Wollmer, R. M. Nitsch, C. Hock, C. M. Morris, J. Hardy and R. T. Moon, Proc. Natl. Acad. Sci. U.S.A., 2007, 104, 9434. [29] E. M. Reiman, J. A. Webster, A. J. Myers, J. Hardy, T. Dunckley, V. L. Zismann, K. D. Joshipura, J. V. Pearson, D. Hu-Lince, M. J. Huentelman, D. W. Craig, K. D. Coon, W. S. Liang, R. H. Herbert, T. Beach, K. C. Rohrer, A. S. Zhao, D. Leung, L. Bryden, L. Marlowe, M. Kaleem, D. Mastroeni, A. Grover, C. B. Heward, R. Ravid, J. Rogers, M. L. Hutton, S. Melquist, R. C. Petersen, G. E. Alexander, R. J. Caselli, W. Kukull, A. Papassotiropoulos and D. A. Stephan, Neuron, 2007, 54, 713. [30] C. J. Phiel, C. A. Wilson, V. M. Y. Lee and P. S. Klein, Nature (London), 2003, 423, 435. [31] X. Sun, S. Sato, O. Murayama, M. Murayama, J. M. Park, H. Yamaguchi and A. Takashima, Neurosci. Lett., 2002, 321, 61. [32] S. Hu, N. Begum Aynun, R. Jones Mychica, S. Oh Mike, K. Beech Walter, H. Beech Beverly, F. Yang, P. Chen, J. Ubeda Oliver, C. Kim Peter, P. Davies, Q. Ma, M. Cole Greg and A. Frautschy Sally, Neurobiol. Dis., 2009, 33, 193. [33] G. I. Welsh, C. M. Miller, A. J. Loughlin, N. T. Price and C. G. Proud, FEBS Lett., 1998, 421, 125. [34] G. I. Welsh and C. G. Proud, Biochem. J., 1993, 294(Pt 3), 625. [35] S. Kaku, S. Chaki and M. Muramatsu, Curr. Signal Transduct. Ther., 2008, 3, 195. [36] R. V. Bhat, J. Shanley, M. P. Correll, W. E. Fieles, R. A. Keith, C. W. Scott and C.-M. Lee, Proc. Natl. Acad. Sci. U.S.A., 2000, 97, 11074. [37] S. Frame, P. Cohen and R. M. Biondi, Mol. Cell, 2001, 7, 1321. [38] J. J. Pei, T. Tanaka, Y. C. Tung, E. Braak, K. Iqbal and I. Grundke-Iqbal, J. Neuropathol. Exp. Neurol., 1997, 56, 70. [39] J.-J. Pei, E. Braak, H. Braak, I. Grundke-Iqbal, K. Iqbal, B. Winblad and R. F. Cowburn, J. Neuropathol. Exp. Neurol., 1999, 58, 1010. [40] B. R. Sperber, S. Leight, M. Goedert and V. M. Y. Lee, Neurosci. Lett., 1995, 197, 149. [41] K. Spittaels, C. Van den Haute, J. Van Dorpe, H. Geerts, M. Mercken, K. Bruynseels, R. Lasrado, K. Vandezande, I. Laenen, T. Boon, J. Van Lint, J. Vandenheede, D. Moechars, R. Loos and F. Van Leuven, J. Biol. Chem., 2000, 275, 41340. [42] T. Engel, P. Goni-Oliver, J. J. Lucas, J. Avila and F. Hernandez, J. Neurochem., 2006, 99, 1445. [43] N. Embi, D. B. Rylatt and P. Cohen, Eur. J. Biochem., 1980, 107, 519. [44] D. R. Alessi, Biochem. Soc. Trans., 2001, 29, 1. [45] D. A. E. Cross, D. R. Alessi, P. Cohen, M. Andjelkovich and B. A. Hemmings, Nature (London), 1995, 378, 785. [46] R. Dajani, E. Fraser, S. M. Roe, N. Young, V. Good, T. C. Dale and L. H. Pearl, Cell, 2001, 105, 721. [47] A. R. Saltiel and C. R. Kahn, Nature, 2001, 414, 799. [48] M. Ilyas, J. Pathol., 2005, 205, 130. [49] T. C. Dale, Biochem. J., 1998, 329(Pt 2), 209. [50] R. Dajani, E. Fraser, S. M. Roe, M. Yeo, V. M. Good, V. Thompson, T. C. Dale and L. H. Pearl, EMBO J., 2003, 22, 494. [51] V. H. Bustos, A. Ferrarese, A. Venerando, O. Marin, J. E. Allende and L. A. Pinna, Proc. Natl. Acad. Sci. U.S.A., 2006, 103, 19725.
Recent Advances in the Discovery of GSK-3 Inhibitors
[52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71]
[72]
[73] [74] [75] [76]
[77]
[78] [79]
25
H. Aberle, A. Bauer, J. Stappert, A. Kispert and R. Kemler, EMBO J., 1997, 16(13), 3797. J. Huelsken and J. Behrens, J. Cell Sci., 2002, 115, 3977. R. Nusse, Cell Res., 2005, 15, 28. G. M. Thomas, S. Frame, M. Goedert, I. Nathke, P. Polakis and P. Cohen, FEBS Lett., 1999, 458, 247. E.-K. Suh and B. M. Gumbiner, Exp. Cell Res., 2003, 290, 447. V. W. Ding, R.-H. Chen and F. McCormick, J. Biol. Chem., 2000, 275, 32475. T. D. King, B. Clodfelder-Miller, K. A. Barksdale and G. N. Bijur, Neurotox. Res., 2008, 14, 367. W. Yang, C. Leystra-Lantz and M. J. Strong, Brain Res., 2008, 1196, 131. N. Vanacore and F. Galeotti, Proc. Natl. Acad. Sci. U.S.A., 2008, 105, E35, author reply E36 Clinicaltrials.gov, Identifiers: NCT00790582 and NCT00818389. T. Yang, D. Arslanova, Y. Gu, C. Augelli-Szafran and W. Xia, Mol. Brain, 2008, 1, 15. A. Martinez, M. Alonso, A. Castro, C. Perez and F. J. Moreno, J. Med. Chem., 2002, 45, 1292. J. R. Woodgett, EMBO J., 1990, 9, 2431. J. R. Woodgett, Methods Enzymol., 1991, 200, 564. F. Mukai, K. Ishiguro, Y. Sano and S. C. Fujita, J. Neurochem., 2002, 81, 1073. B. Schaffer, M. Wiedau-Pazos and D. H. Geschwind, Gene, 2003, 302, 73. K. F. Lau, C. C. J. Miller, B. H. Anderton and P. C. Shaw, J. Pept. Res., 1999, 54, 85. B. W. Doble, S. Patel, G. A. Wood, L. K. Kockeritz and J. R. Woodgett, Dev. Cell, 2007, 12, 957. E. ter Haar, J. T. Coll, D. A. Austen, H.-M. Hsiao, L. Swenson and J. Jain, Nat. Struct. Biol., 2001, 8, 593. M. Saitoh, J. Kunitomo, E. Kimura, Y. Hayase, H. Kobayashi, N. Uchiyama, T. Kawamoto, T. Tanaka, C. D. Mol, D. R. Dougan, G. S. Textor, G. P. Snell and F. Itoh, Bioorg. Med. Chem., 2009, 17, 2017. D. A. Heerding, N. Rhodes, J. D. Leber, T. J. Clark, R. M. Keenan, L. V. Lafrance, M. Li, I. G. Safonov, D. T. Takata, J. W. Venslavsky, D. S. Yamashita, A. E. Choudhry, R. A. Copeland, Z. Lai, M. D. Schaber, P. J. Tummino, S. L. Strum, E. R. Wood, D. R. Duckett, D. Eberwein, V. B. Knick, T. J. Lansing, R. T. McConnell, S. Y. Zhang, E. A. Minthorn, N. O. Concha, G. L. Warren and R. Kumar, J. Med. Chem., 2008, 51, 5663. H.-C. Zhang, L. V. R. Bonaga, H. Ye, C. K. Derian, B. P. Damiano and B. E. Maryanoff, Bioorg. Med. Chem. Lett., 2007, 17, 2863. I. M. Withers, M. P. Mazanetz, H. Wang, P. M. Fischer and C. A. Laughton, J. Chem. Inf. Model., 2008, 48, 1448. H.-J. Kim, H. Choo, Y. S. Cho, K. T. No and A. N. Pae, Bioorg. Med. Chem., 2008, 16, 636. M. A. Fabian, W. H. Biggs, D. K. Treiber, C. E. Atteridge, M. D. Azimioara, M. G. Benedetti, T. A. Carter, P. Ciceri, P. T. Edeen, M. Floyd, J. M. Ford, M. Galvin, J. L. Gerlach, R. M. Grotzfeld, S. Herrgard, D. E. Insko, M. A. Insko, A. G. Lai, J.-M. Lelias, S. A. Mehta, Z. V. Milanov, A. M. Velasco, L. M. Wodicka, H. K. Patel, P. P. Zarrinkar and D. J. Lockhart, Nat. Biotechnol., 2005, 23, 329. M. W. Karaman, S. Herrgard, D. K. Treiber, P. Gallant, C. E. Atteridge, B. T. Campbell, K. W. Chan, P. Ciceri, M. I. Davis, P. T. Edeen, R. Faraoni, M. Floyd, J. P. Hunt, D. J. Lockhart, Z. V. Milanov, M. J. Morrison, G. Pallares, H. K. Patel, S. Pritchard, L. M. Wodicka and P. P. Zarrinkar, Nat. Biotechnol., 2008, 26, 127. M. Hong, D. C. R. Chen, P. S. Klein and V. M. Y. Lee, J. Biol. Chem., 1997, 272, 25326. N. Kirshenboim, B. Plotkin, B. Shlomo Shani, O. Kaidanovich-Beilin and H. EldarFinkelman, J. Mol. Neurosci., 2004, 24, 237.
26
Robert G. Gentles et al.
[80] M. Medina and A. Castro, Curr. Opin. Drug Discov. Dev., 2008, 11, 533. [81] L. Meijer,, N. Guyard,, L. A. Skaltsounis, and G. Eisenbrand (eds), Life in Progress, Station Biologique, Roscoff, France, 2006, p. 297. [82] P. Polychronopoulos, P. Magiatis, A.-L. Skaltsounis, V. Myrianthopoulos, E. Mikros, A. Tarricone, A. Musacchio, S. M. Roe, L. Pearl, M. Leost, P. Greengard and L. Meijer, J. Med. Chem., 2004, 47, 935. [83] L. Meijer, A.-L. Skaltsounis, P. Magiatis, P. Polychronopoulos, M. Knockaert, M. Leost, X. P. Ryan, C. A. Vonica, A. Brivanlou, R. Dajani, C. Crovace, C. Tarricone, A. Musacchio, S. M. Roe, L. Pearl and P. Greengard, Chem. Biol., 2003, 10, 1255. [84] K. Vougogiannopoulou, Y. Ferandin, K. Bettayeb, V. Myrianthopoulos, O. Lozach, Y. Fan, C. H. Johnson, P. Magiatis, A.-L. Skaltsounis, E. Mikros and L. Meijer, J. Med. Chem., 2008, 51, 6421. [85] M. J. Moon, S. K. Lee, J.-W. Lee, W. K. Song, S. W. Kim, J. I. Kim, C. Cho, S. J. Choi and Y.-C. Kim, Bioorg. Med. Chem., 2006, 14, 237. [86] T. Tamaoki and H. Nakano, Biotechnology, 1990, 8, 732. [87] M. R. Jirousek, J. R. Gillig, C. M. Gonzalez, W. F. Heath, J. H. McDonald, III, D. A. Neel, C. J. Rito, U. Singh, L. E. Stramm, et al., J. Med. Chem., 1996, 39, 2664. [88] J. Respondek and J. Engel, Drugs Future, 1996, 21, 391. [89] E. Meggers, G. E. Atilla-Gokcumen, H. Bregman, J. Maksimoska, S. P. Mulcahy, N. Pagano and D. S. Williams, Synlett, 2007, 1177. [90] G. E. Atilla-Gokcumen, D. S. Williams, H. Bregman, N. Pagano and E. Meggers, ChemBioChem., 2006, 7, 1443. [91] J. Maksimoska, L. Feng, K. Harms, C. Yi, J. Kissil, R. Marmorstein and E. Meggers, J. Am. Chem. Soc., 2008, 130, 5764. [92] G. E. Atilla-Gokcumen, N. Pagano, C. Streu, J. Maksimoska, P. Filippakopoulos, S. Knapp and E. Meggers, ChemBioChem., 2009, 10, 198. [93] J. Maksomiska, D. S. Williams, G. E. Atilla-Gokcumen, K. S. M. Smalley, P. J. Carroll, R. D. Webster, P. Filippakopoulos, S. Knapp, M. Herlyn and E. Meggers, Chem. Eur. J., 2008, 14, 4816. [94] A. Martinez, M. Alonso, A. Castro, I. Dorronsoro, J. L. Gelpi, F. J. Luque, C. Perez and F. J. Moreno, J. Med. Chem., 2005, 48, 7103. [95] E. Martin-Aparicio, A. Fuertes, M. J. Perez-Puerto, D. Perez-Navarro, M. Alonso, A. Martı´nez and M. Medina, Neurobiol. Aging, 2004, 25(Suppl. 2), Abst P4–428. [96] R. Luna-Medina, M. Cortes-Canteli, S. Sanchez-Galiano, J. A. Morales-Garcia, A. Martinez, A. Santos and A. Perez-Castillo, J. Neurosci., 2007, 27, 5766. [97] B. Winblad, L. Kilander, S. Eriksson, L. Minthon, S. Batsman, A. L. Wetterholm, C. Jansson-Blixt and A. Haglund, Lancet, 2006, 367, 1057. [98] M. Y. Noh, S. H. Koh, Y. Kim, H. Y. Kim, G. W. Cho and S. H. Kim, J. Neurochem., 2009, 108, 1116. [99] X. Li, K. M. Rosborough, A. B. Friedman, W. Zhu and K. A. Roth, Int. J. Neuropsychopharmacol., 2007, 10, 7. [100] M. K. Mohammad, I. M. Al-masri, M. O. Taha, M. A. S. Al-Ghussein, H. S. AlKhatib, S. Najjar and Y. Bustanji, Eur. J. Pharmacol., 2008, 584, 185. [101] M. O. Taha, Y. Bustanji, M. A. Al-Ghussein, M. Mohammad, H. Zalloum, I. M. Al-Masri and N. Atallah, J. Med. Chem., 2008, 51, 2062. [102] Y. Miyazaki, Y. Maeda, H. Sato, M. Nakano and G. W. Mellor, Bioorg. Med. Chem. Lett., 2008, 18, 1967. [103] C. Lum, J. Kahl, L. Kessler, J. Kucharski, J. Lundstrom, S. Miller, H. Nakanishi, Y. Pei, K. Pryor, E. Roberts, L. Sebo, R. Sullivan, J. Urban and Z. Wang, Bioorg. Med. Chem. Lett., 2008, 18, 3578.
CHAPT ER
2 Advancements in the Development of Nitric Oxide Synthase Inhibitors Shawn Maddaford, Subhash C. Annedi, Jailall Ramnauth and Suman Rakhit
Contents
1. Introduction 2. NOS — Structure and Function 2.1 Introduction 2.2 Isoform-selective inhibitors — structural basis 3. Selective NOS Inhibitors 3.1 Introduction 3.2 Selective nNOS inhibitors 3.3 Selective iNOS inhibitors 4. Clinical Findings with NOS Inhibitors 5. Future Directions — Dual Action NOS Inhibitors 5.1 Introduction 5.2 Dual action nNOS inhibitors 5.3 Dual action iNOS inhibitors 6. Conclusions References
27 28 28 29 32 33 33 38 41 43 43 44 45 46 47
1. INTRODUCTION Nitric oxide (NO) is perhaps one of the most intensively studied neurotransmitters since its discovery by Ignarro [1] and Furchgott [2] in the early 1980s as a mediator of vascular tone. This early work revealed NeurAxon Inc, 480 University Ave, Suite 900, Toronto, ON M5G 1V2, Canada Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04402-9
r 2009 Published by Elsevier Inc.
27
28
Shawn Maddaford et al.
that endothelium-derived relaxing factor (EDRF), the agent found in the endothelium that was responsible for dilation of blood vessels, was in fact NO. Since then, the unraveling of the diverse biological functions of NO has continued unabated as evidenced by more than 80,000 publications on nitric oxide synthase (NOS) and NO described in the literature. In addition to its role in regulating blood pressure, NO is important in platelet aggregation, bone remodeling, inflammation, and neurotransmission, wherein NO initiates changes in neuronal excitability and synaptic strength by acting at pre- and/or post-synaptic locations. The reduction in pathophysiological levels of NO through inhibition of NOS has the potential to be therapeutic in a multitude of indications including the treatment of septic shock, stroke, neurodegenerative disorders (e.g., Parkinson’s, amyotrophic lateral sclerosis (ALS), multiple sclerosis (MS)), and in the treatment of pain [e.g., migraine, chronic tension-type headache (CTTH), visceral, and neuropathic] [3]. However, the therapeutic control of NO synthesis has, until recently, been unattainable due to the difficulties in achieving isoform-selective inhibition. The selective inhibition of the neuronal NOS (nNOS) enzyme and/or the inducible (iNOS) over the endothelial NOS (eNOS) enzymes for the treatment of pain or migraine would be required to avoid the cardiovascular liabilities associated with eNOS inhibition [4]. Earlier reviews in this publication have focused on the enzymology and synthesis of NOS inhibitors [5,6]. This review focuses on recent advances in the design of selective inhibitors of NOS and some of the emerging preclinical and clinical developments of these newer inhibitors.
2. NOS — STRUCTURE AND FUNCTION 2.1 Introduction NO is synthesized by three isoforms of NOS, which catalyzes the fiveelectron oxidation of L-arginine to L-citrulline [7]. The neuronal or brain NOS (nNOS or NOS1) and endothelial form (eNOS or NOS3) are constitutively expressed, whereas the inducible form (iNOS or NOS2) is expressed under conditions of stress or upon the release of inflammatory mediators such as tumor necrosis factor-a, interleukin-1, or lipopolysaccharides (LPS). These homodimeric proteins consist of a C-terminal reductase domain that transfers electrons from nicotinamide adenine dinucleotide phosphate (NADPH) through the flavin adenine dinucleotide (FAD) and flavin mononucleotide (FMN) prosthetic groups to the Nterminal oxygenase domain that binds the arginine substrate, (1uR, 2uR, 6R)-5,6,7,8-tetrahydrobiopterin (BH4) and heme [8]. The two constitutive forms are activated by increasing Ca2+ concentration and the binding of a Ca2+/calmodulin complex while the activity of iNOS appears to be
29
Advancements in the Development of NOS Inhibitors
independent of Ca2+ concentration due to the tight binding of the complex at the dimer interface [9]. The active dimeric form of NOS is stabilized by a structural zinc binding at the oxygenase dimer interface [10], which forms the BH4-binding site. The closest related enzymes are the cytochrome (CYP) P450 enzymes, wherein both the function and amino acid sequence of the reductase domains and the function but not the sequence of the oxygenase domains are similar. Reduction of the BH4 and heme iron allows for the activation of O2 and subsequent CYP-like oxidation of L-arginine to No-hydroxy-L-arginine and finally to citrulline, ultimately releasing NO. In the absence of BH4, NOS reduces oxygen to form superoxide [11]. The inhibitor design reviewed herein has primarily focused on targeting the arginine or BH4-binding sites to prevent the overproduction of NO or the uncoupled reduction of O2 to superoxide, thus minimizing their deleterious effects.
2.2 Isoform-selective inhibitors — structural basis The three NOS isoforms possess approximately 50% overall sequence homology and identical structural architecture. With the exception of the human nNOS enzyme, the crystal structures for all three isoforms have been determined allowing for the application of structure-based inhibitor design [10,12–17]. However, analysis of the oxygenase domains reveals striking active site structural conservation across the three isoforms with ˚ of the substrate-binding site being identical 16 of 18 residues within 6 A [18]. Despite the high similarity of the binding pockets, inhibitors 1–4 derived from L-NG-nitroarginine (L-NNA) show selectivity for rat nNOS versus bovine eNOS (Table 1). These compounds make interactions with the L-amino acid-specific binding pocket demonstrating hydrogen (H) bonds to Gln249, Tyr359, Glu363, and Asn368 (rat nNOS) [19]. In all cases, the guanidine and aminopyridine groups exhibit similar bifurcated H-bonds with the conserved glutamate (e.g., Glu363 or 592) across the three isoforms. However, the nitro group of inhibitors 1–3 provides Table 1 Peptide, reduced amide, and aminopyridine-based substrate inhibitor selectivity for 1–4 Compound Rat nNOS
1 2 3 4
0.30 0.10 0.15 0.38
Note: Ki values are in mM.
Bovine eNOS
Murine iNOS
e/n
i/n
D597N nNOS
N368D eNOS
107 110 80 434
25 29 39 58
1538 1280 2617 1114
192 290 325 150
67 34 21 –
9.5 4.6 5.1 –
30
Shawn Maddaford et al.
additional hydrogen bonding and non-bonded contacts with the protein backbone, which results in enhanced binding affinity over L-arginine [20]. Inspection of the X-ray complexes suggests that the selectivity of 1–3 for nNOS over eNOS arises from a single amino acid substitution of a neutral Asn368 (eNOS) residue for a charged Asp597 (nNOS) residue (Figure 1). These inhibitors adopt a curled conformation in the nNOS active site placing the a-amino group between Glu592 and Asp597, ˚ ) and potentially a so that it makes a direct H-bond to Glu592 (2.8 A favorable electrostatic interaction with the charged Asp597 (nNOS). In contrast, the inhibitors bound to eNOS were in a fully extended conformation with the a-amino group in 1, reorienting to make a H-bond with Gln249 rather than Glu592. Mutagenesis studies confirmed that this single residue difference was responsible for the two orders of magnitude difference in potency between the two isoforms (Table 1). NO2 HN
H2N
NO2
NH
HN
NO2 NH
HN
N
NH
NH NH
NH
NH
O H N H2N
NH2 O
H N
O
H2N
H2N O
1
NH2
NH
HN
H N NH2
NH2
NH2
4
3
2
ASP597 GLN478 ASN368 ASN569
GLU592
GLN249
ASN340
GLU363
Figure 1 Active site poses of dipeptide amide 1 in the active site of nNOS (left) and eNOS (right). The a-amino group makes hydrogen bonding and electrostatic interaction with Glu592 in nNOS, whereas 1 displays an extended conformation in eNOS and makes an H-bond with Gln249 as opposed to Glu363. (See Color Plate 2.1 in Color Plate Section.)
31
Advancements in the Development of NOS Inhibitors
The D597N mutation in nNOS resulted in a drop in potency (1; w.t. ¼ 0.3 mM, D597 N ¼ 67 mM), whereas the corresponding eNOS mutant N368D resulted in an increase in potency (1; w.t. ¼ 107 mM, N368D ¼ 9.5 mM). Similar trends were observed for compounds 2 and 3. Finally, a crystal structure of the nNOS D597N mutant complexed with 1 revealed an extended conformation, whereas the eNOS mutant N368D complexed to 1 adopted a curled conformation [19]. Recently, it has been suggested that the selectivity for a series of quinazoline and aminopyridine inhibitors for the iNOS enzyme is due to an isozyme-specific residue plasticity, wherein residues distant from the active site modulate conformational changes of invariant residues in contact with the inhibitors [17]. In this ‘‘anchored plasticity approach,’’ an inhibitor core anchored in a conserved binding pocket and a rigid bulky substituent extend into remote specificity pockets upon conformational changes of ‘‘plastic residues.’’ This approach is conceptually different to the approach described earlier where the ligand changes its conformation to interact with variant residues within the immediate active site. Thus, non-selective ligands such as S-ethylisothiourea (SEITU), aminopyridine 5, or the selective aminopyridine 6 bound to human iNOS (Figure 2) are anchored through bidendate hydrogen bonds with Glu377. The bulky side chain of 6 induces an opening of invariant
Phe
Val Cα
Phe286
Val305 Cα
Leu288 Cα O
O Asn H2N
H2N
Asn
H2N
Asn
O Arg
HN
Ile269 Cα
NH
H HN N
Arg
Arg
NH2
NH2
H2N
H HN N
O
O O H2N hiNOS Gln-closed (A)
Gln
H2N hiNOS Gln-open
Gln
H2N
Gln
heNOS Gln-open
(B)
Figure 2 (A) X-ray crystal structures of human iNOS in the Gln-closed form (gray carbons-SEITU bound) and Gln-open form (green carbon atoms; compound 6). The bulky inhibitor 6 avoids steric clashes by creation of a new pocket as a result of rotation of first shell residue Gln263 about its w1 and w2 torsion angles. (B) Schematic representation of the movement of residues in iNOS. In eNOS, the Gln-open form is not possible due to the bulky third shell Leu and Ile residues. (See Color Plate 2.2 in Color Plate Section.)
32
Shawn Maddaford et al.
first-shell residues Gln263 and Arg373 inducing a cascading effect of second-shell residue Asn283 toward third-shell residues Phe286 and Val305. Small non-selective inhibitors do not induce this conformational change and thus bind to a first-shell Gln-closed state, whereas the bulky ligand 6 binds to the Gln-open state in a newly created binding pocket. Compound 6 binds weakly to eNOS in a Gln-closed state. In human eNOS, the invariant first (Gln246 and Arg249) and second shell residues (Asn266) are unable to undergo this conformational change required for binding since the final third shell residues are the much bulkier isoleucine (Ile269) and leucine residues (Leu288) compared with the smaller Phe and Val residues found in human iNOS.
O N N
NH2
N
N H
5: iNOS: 0.12 μM 6: iNOS: 0.37 μM eNOS: 0.32 μM eNOS: 100 μM nNOS: 0.16 μM nNOS: 23 μM
O
O
OEt
O
O N
N N
N H 7: iNOS: 0.52 μM eNOS: >100 μM nNOS: 15 μM
N
N H NH2 8: iNOS: 24 μM eNOS: >100 μM nNOS: 39 μM
Although the use of crystal structures can provide valuable insights into the understanding of drug–ligand interactions, caution must be used when interpreting structural data. In general, modelers and medicinal chemists make several assumptions about the protein crystal structure: 1) the protein structure and amino acid sequence is correct; 2) the structure of the ligand and its interactions with the protein are correct; and 3) the protein–ligand structure is relevant for drug discovery. Issues often arise when the resolution of the structure is insufficient to determine the identity or orientation of a ligand in the active site or the position of side chain residues. In addition, crystal structures do not provide information on the thermodynamics of ligand binding and desolvation contributions to the overall binding. A recent example illustrates these limitations of X-ray crystallography in the design of iNOS inhibitors related to structures 6 and 7 [21]. The co-crystal structure of mouse iNOS with 7 revealed a single amino acid difference in contact with the inhibitor (Asp376 in iNOS substituted for Asn in eNOS). Compound 8 was designed in attempts to exploit this difference through an interaction of the basic nitrogen with the Asp residue and displacement of a bound water molecule H-bonded to this residue. Both docking studies and a ˚ ) seemed to confirm this interaction. low resolution X-ray structure (3.3 A However, the in vitro NOS assay revealed a substantial reduction in potency for iNOS (24 mM). It is possible that the loss of hydrogen bonds between the displaced water molecule with the protein or an energy
Advancements in the Development of NOS Inhibitors
33
penalty associated with desolvation of the primary amine group accounts for this loss in potency. In another example, Silverman and co-workers [22] designed an N-hydroxy analog of 3 to displace a structural water molecule hydrogen-bonded between the heme propionate groups of the nNOS enzyme. A crystallographic analysis revealed that indeed the water molecule was successfully displaced, but this change failed to improve the in vitro potency. The successful use of NOS X-ray structures in drug design are discussed in Section 3.2.
3. SELECTIVE NOS INHIBITORS 3.1 Introduction Numerous attempts have been made toward the design of isozymeselective NOS inhibitors targeting both the L-arginine and BH4-binding sites. Early inhibitors were based on modifications of mono- and dipeptides, with subsequent inhibitor design and synthesis utilizing co-crystal structures of all NOS isozymes. Several review articles covering progress toward the development of selective nNOS inhibitor syntheses [23,24], selective iNOS inhibitor syntheses [25], and computational studies [26] have already appeared in the literature. The application of selective NOS inhibitors for various therapeutic indications such as the treatment of shock [27], asthma [28], migraine, tension-type and cluster headache [29], neurological diseases [30], and inflammatory joint diseases [31] were recently reviewed. Many early NOS inhibitors were modified substrate (L-arginine) or product (L-citrulline) analogs. First-generation arginine-based inhibitors such as NG-methyl-L-arginine (L-NMMA), NG-nitro-L-arginine (L-NNA) and its L-nitroarginine methyl ester (L-NAME), NG-amino-L-arginine (LNAA), NG-allyl-L-arginine (L-ALA), NG-cyclopropyl-L-arginine (L-NCPA), and NG-propyl-L-arginine (L-NPA) are reasonably potent but are poorly selective toward the NOS isoforms [23]. The most commonly used arginine-based NOS inhibitors for in vitro and in vivo assays are L-NMMA, L-NNA, and L-NAME, a consequence of commercial availability, chemical stability, water solubility, and low toxicity. Owing to the peptide nature and poor drug-like properties of these inhibitors, non-peptidic small molecule drug design and synthesis are warranted.
3.2 Selective nNOS inhibitors 3.2.1 Peptide-based inhibitors The amino group appears to be very important, whereas the carboxylic group seems unnecessary for enzyme inhibition. In an attempt to
34
Shawn Maddaford et al.
increase isozyme selectivity, the carboxylic acid can be functionalized with amino acids to form dipeptides. Most of the arginine-based dipeptides display excellent potency and selectivity (such as compounds 1 and 2) for rat nNOS over bovine eNOS and murine iNOS [24]. However, the activity of these compounds with human isoforms has not been reported. In an attempt to improve brain penetration, various aminopyridine analogs have been prepared using a de novo design strategy called ‘‘fragment hopping’’ and demonstrate a significant improvement in nNOS activity and selectivity over eNOS [32]. This strategy is based on pharmacophore requirements such as hydrophobic and steric interactions, as demonstrated with compound 4, which revealed lipophilic pockets near the 4-position of the aminopyridine ring and the terminal aminomethyl group. Compounds 9 and 10, derived from 4, are much more potent at nNOS (Ki ¼ 0.08 and 0.01 mM) and selective over eNOS (1000- and 2000-fold) than the corresponding unsubstituted parent compounds [18]. Both compounds prevented hypoxia-ischemia-induced death in a rabbit model for cerebral palsy, also reducing the number of newborn kits exhibiting symptoms of cerebral palsy without affecting blood pressure. Maternal administration of compounds 9 or 10 showed 83% and 69% normal kits, respectively, compared to 9% of saline-treated kits. H N
N H
R
H N
N n
H2N 9: n = 1, R = 4-Cl 10: n = 2, R = 3-F
3.2.2 Non-peptide-based inhibitors 7-Nitroindazole (7-NI, 11) and 1-(2-trifluoromethylphenyl)imidazole (TRIM) (12) [33] are two of the most extensively studied non-peptidic NOS inhibitors, due to a high in vivo selectivity for nNOS despite modest selectivity in vitro (7-NI; rat nNOS IC50 ¼ 0.47 mM and bovine eNOS IC50 ¼ 0.7 mM) [23]. The inhibitory mechanism and selectivity is the result of a reversible competitive interaction with both the L-arginine and the BH4-binding sites in nNOS [34], which is limited to the pterin cofactor site with iNOS and eNOS. In addition to NOS knockout studies, these early ‘‘brain selective’’ inhibitors provided much of the foundation for
35
Advancements in the Development of NOS Inhibitors
elucidating the pharmacological role of nNOS including nociception [35,36], antidepressant and anxiolytic properties [37], modification of opioid-induced side effects, and synergistic enhancement of serotonergic antidepressants [38].
CF3
H N
N N H
Cl N
NH
N
N H
NO2
11
12
S
13
Compound 13 (ARL17477) is a selective nNOS inhibitor with IC50 values of 0.035, 5.0, and 3.5 mM for human nNOS, iNOS, and eNOS, respectively [39]. Evaluation of ARL17477 in a global ischemia model produced a significant reduction (52% protection) in ischemia-induced hippocampal damage following global ischemia dosed immediately (50 mg/kg i.p.) post-occlusion [40]. In addition, ARL17477 was reported to provide greater neuroprotection (44%) than L-NAME (19%), 7-NI (22%), or TRIM (8%) in a gerbil model of global cerebral ischemia. In an endothelin-1 model of focal ischemia, a 1 mg/kg i.v. administration (0, 1, or 2 h post-endothelin-1) of ARL17477 significantly attenuated the infarct volume when compared to pre-treatment with the NMDA antagonist MK-801, post-treatment with the iGluR5 antagonist LY377770, or the immunophilin FK-506. In another study, a combination of ARL17477 with MK-801 or LY293558 provided 78% and 71% greater neuroprotection than the calculated additive effects of the individual treatments [41]. ARL17477 also inhibited electrically induced nitrergic relaxations in pig gastric fundus strips and gastric fundic compliance in conscious pigs in a dose-dependent manner [42]. On the basis of the knowledge that incorporation of a basic side chain could improve potency and selectivity in amidine compounds such as ARL17477, a series of related fused 2-aminodihydroquinoline compounds were designed [43]. Thus, incorporation of the amine side chain into 14a to give 15a improved the potency and selectivity (nNOS over eNOS) up to 224-fold. To improve brain penetration through a reduction in basicity, fluoro analogs 14b and 15b were synthesized. Incorporation of fluorine into the unsubstituted analog 14a decreased the pKa from 9.7 to 7.9 (14a and 14b, respectively) without affecting the potency or selectivity (nNOS ¼ 0.16 mM, 0.1 mM, eNOS ¼ 3.3 mM, 2.7 mM). However, fluorination of the corresponding amine side chain–containing compound 15a to give 15b led to a similar reduction in pKa (9.9 to 8.0) but
36
Shawn Maddaford et al.
was accompanied by a loss in potency relative to the parent compound 15a [44].
H
H X
Cl
X N H
NH
NHR N
NH 16a R = 1-(4-F-Benzylpiperidin-4-yl) 16b R = 2-(1H-Imidazol-5-yl)ethyl
15a: X = H, 15b: X = F
14a: X = H,14b: X = F
S
NH
HN N H
H N
S
A series of substituted 2-aminobenzothiazole inhibitors were shown to be selective toward nNOS over eNOS and iNOS [45]. By varying substitution at the 2-position, up to 40–50-fold selectivity was achieved for human nNOS over eNOS (16a; nNOS ¼ 0.2 mM, eNOS ¼ 8 mM, e/n ¼ 40). It is notable that substantial species differences were observed in some cases such as 16b (IC50 values for human nNOS ¼ 0.3 mM, eNOS ¼ 0.7 mM and iNOS ¼ 6 mM; rat nNOS ¼ 4.7 mM, bovine eNOS ¼ 362 mM). Compound 16b showed neuroprotection in a concentrationdependent manner, when preincubated for 60 min before NMDA challenge in rat cortical cultures. At 50 and 100 mM concentration, increases in cell survival rates of 80% and 90%, respectively, were observed, compared to 30% in NMDA-treated controls [46].
NH
N S H N
S NH
N
N H
NH
N
N
N H 17
H N
S
N 18
19
Various di-substituted indole compounds were investigated for their activity in all three human NOS isozymes and showed submicromolar potencies for nNOS and good selectivity for nNOS over both eNOS and iNOS [47,48]. In this study, the basic amine side chain was varied at both the 1- and the 3-positions on the indole and the amidine group between the 5- and the 6-positions. The most potent compound in the series was obtained with the basic amine group at the 1-position and with the amidine at the 5-position (IC50: human nNOS ¼ 0.02 mM; eNOS ¼ 1.92; iNOS ¼ 17 mM). Alternatively, the best selectivity was achieved with a quinuclidine-based system and the basic side chain at the 3-position
Advancements in the Development of NOS Inhibitors
37
(17; human IC50: nNOS ¼ 0.76 mM, eNOS ¼ 103 mM, iNOS ¼ 89; e/n ¼ 136, i/n ¼ 117). In further studies, compound 18 (human IC50: nNOS ¼ 1.2 mM, eNOS ¼ 15 mM, iNOS ¼ 60; e/n ¼ 12.5, i/n ¼ 50) was highly neuroprotective and demonstrated efficacy in multiple pain models. Pre-incubation of rat cortical cells at 25 mM concentration of 18 followed by an NMDA challenge resulted in 95% cell survival when compared to only 50% in NMDA-treated controls. In a separate experiment, 18 was able to reduce hippocampal cell death up to 95% in a concentration-dependent manner after oxygen–glucose deprivation. The antinociceptive effect of compound 18 was measured by formalininduced hyperalgesia and inflammation in an experimental model of sustained inflammatory nociception associated with long-term intracellular changes of nociceptive processing at the level of the spinal cord. Pre-administration of compound 18 (5 and 10 mg/kg i.p.) or 7-NI (2.5 and 5 mg/kg i.p.) reduced paw licking in a concentration-dependent manner for both compounds indicating efficacy for the treatment of inflammatory pain. In the Chung Spinal Nerve Ligation (SNL) neuropathic pain model, 18 demonstrated a complete reversal of both thermal and tactile hyperalgesia in rats after 3 and 20 mg/kg i.p. administration, respectively. A series of indoline, oxindole, tetrahydroquinoline, quinolone, and benzazepine derivatives were explored in an attempt to identify selective human nNOS inhibitors [49]. In this study, nanomolar potency for human nNOS (IC50 ¼ 0.01 mM) and more than 1,000-fold selectivity over eNOS (IC50 ¼ 14.2 mM) was achieved. Compound 19 (nNOS ¼ 0.34 mM, eNOS ¼ 48 mM, e/n ¼ 141) was able to completely reverse tactile allodynia in the sciatic nerve cuff model of neuropathic pain after i.p. administration at 30 mg/kg. At the same time, compound 19 was able to reduce the frequency of paw lifts up to 50% on a cold platform in the same model after multiple dose administration. The 2-aminopyridine group serves as an effective isosteric replacement of the guanidinium group and even simple derivatives such as 4-methyl-2-aminopyridine exhibit potent but non-selective inhibition of NOS. Analogs such as 20 are potent nNOS inhibitors with modest selectivity over eNOS (nNOS ¼ 140 nM, eNOS ¼ 887 nM, e/n ¼ 6) with minimal off-target activity at m2 and m4 receptors (Ki ¼ 0.32 mM and 0.59 mM, respectively) [50]. Pharmacokinetic studies with 20 after subcutaneous administration (24 mg/kg) gave a Cmax of 1.7 mg/mL in plasma, 0.45 mg/mL in cerebrospinal fluid (CSF), and 11 mg/g in whole brain, indicating sufficient brain levels to inhibit nNOS in vivo. Although the compound did not increase blood pressure in rats at doses of up to 100 mg/kg, the rat eNOS values were not reported. Further modifications of this scaffold by incorporating bulk onto the phenyl ring further increased the selectivity for nNOS (e.g., 21, e/n ¼ 50) [51]. Compound 21
38
Shawn Maddaford et al.
was active in the harmaline-induced cGMP model in rat cerebellum (ED50 ¼ 7 mg/kg s.c.). In contrast to 20, compound 21 inhibited phencyclidine (PCP)-induced hypermotility (2.8 mg/kg s.c.) in rats, a model used to assess antipsychotic agents.
N O
NH2
N
NH2
N
N
O
N Ph
20
21
A set of 41 pteridine nNOS-selective inhibitors employing four scaffolds was developed by a combination of ligand- and structure-based design [52]. The X-ray crystal structure for rat NOS-dimeric oxygenase domain with BH4 and L-arginine was used to develop a human isoform homology model. Substitution on the 4-, 5-, 6-, and 7-positions are necessary for both affinity and selectivity. Bulky and hydrophobic substituents at the 5- or 6-position increased the selectivity for nNOS over iNOS and eNOS. The tetrahydro antipterins were more active and selective for nNOS over eNOS; up to 58-fold selectivity was observed with compound 22. Ph Cl 5 H N
HN 4 3N
6
2 H2N
N H 8
N 1
7
22
3.3 Selective iNOS inhibitors Inducible NOS has been implicated in a number of inflammatory diseases such as septic shock, multiple sclerosis, rheumatoid and osteoarthritis, ulcerative colitis, and asthma. Not surprisingly, many research institutions have developed programs to design inhibitors of iNOS. In addition to
Advancements in the Development of NOS Inhibitors
39
potency, selectivity against the other two isoforms, especially eNOS, is required since they are important in normal physiology. Selective inhibitors of iNOS have been reviewed previously [25]. Early inhibitors of iNOS were simple analogs of the substrate L-arginine in which the guanidine group or the side-chain had been modified. Compound 23 (GW274150), a selective and modestly potent inhibitor of human iNOS (IC50 ¼ 1.4 mM) has completed a phase II clinical trial for the acute and prophylactic treatment of migraine headache and asthma [53]. HN
O S
HO NH2
O N
N H N H H
23
O
H
24
NH
N
N H
CN 25
In general, due to their physicochemical properties, amino acid arginine-like inhibitors will rely on active transport processes for their absorption and distribution. Consequently, a great deal of effort has been invested in discovering NOS inhibitors with more drug-like properties. Amidines and isothiourea derivatives have been developed to replace the guanidino group, a key anchor point in the active site. Generally, the isothioureas are more selective for nNOS while amidines tend to be selective for iNOS. Considerable effort has been expended in the design and synthesis of selective amidine iNOS inhibitors. Notably, the fused bicyclic amidine, 24, has been reported to inhibit human iNOS with an IC50 ¼ 22 nM and greater than 30-fold selectivity over eNOS, but only two-fold over nNOS [54]. The ability of compound 24 to inhibit iNOS activity in vivo was measured in a rat endotoxin assay. This model measures the plasma nitrate/nitrite (NOx) levels due to the induction of iNOS following LPS administration. An oral dose of 24 (10 mg/kg) given at 1 or 16 h before LPS challenge showed complete inhibition or 40% reduction of NOx respectively, at the 5-h time points. This demonstrates that efficacious levels of 24 can be maintained in circulation for more than 16 h. The 2-aminopyridine group used as an isosteric replacement for amidines is generally less basic and should therefore be more membrane permeable. The N-alkylated aminopyridine 25 is a potent (71 nM) and selective (W1,000- and 100-fold selective against eNOS and nNOS, respectively) inhibitor of iNOS [55]. Reported in a later publication, compound 25 was tested in a rat model of LPS-induced NO production and showed an oral dose-dependent inhibition of elevated plasma NOx levels (IC50 ¼ 1.8 mM) measured 4 h after LPS administration [17].
40
Shawn Maddaford et al.
Most competitive NOS inhibitors possess a minimum pharmacophore of a basic cis-amidine (a guanidine isostere) required for a bidentate interaction with the carboxyl group of Glu371. The imidazopyridine 26 and the spirocyclic quinazoline 27 represent novel classes of potent and selective iNOS inhibitors bearing a masked cis-amidine moiety. Compound 26, a potent iNOS arginine-competive inhibitor (86 nM) exhibiting more than 1,500-fold selectivity over eNOS and 80-fold selectivity over nNOS, was identified by screening a corporate compound library [56]. Inhibition of iNOS-derived NOx formation in various intact cells following LPS treatment was observed for compound 26: nitrite generation after induction of iNOS was inhibited with IC50 values of 3.1 mM in murine macrophage cell line RAW, 33 mM in rat mesangial cell line RMC, and 13 mM in a human HEK293/iNOS assay. The spirocyclic quinazoline 27 inhibits purified recombinant human iNOS (IC50 ¼ 37 nM) and NO synthesis in a whole cell assay (human DLD-1 cells, IC50 ¼ 0.9 mM) [57]. The compound was shown to be selective over nNOS (25-fold) and eNOS (W2,000-fold). When given orally to rats, 27 produced a dose-dependent inhibition of NO production induced by LPS (ID50 ¼ 3 mmol/kg, 4 h post-dose). In addition, 27 showed efficacy in the Freund’s complete adjuvant-induced polyarthritis model in rats. Administration of 27 (10–100 mmol/kg p.o., twice daily) commenced on the day of challenge delayed the onset of observable symptoms (inflammation, edema, pain, and joint destruction) and reduced their severity in a dose-dependent manner (ED50 at 20 d timepoint B10 mmol/kg). At a higher dose (100 mmol/kg), the compound completely abolished all indications of developing arthritis for the full 20 days of the experiment. O
O F N
N
N
F
N N
N
HN
26
NH2 N
H N
S
CN HN
O
O
NH2
27
28
The structure-based design of compound 28 was based on the ‘‘anchored plasticity’’ approach (vide ante) [17]. Compound 28, with a cyclic amidine, inhibits human iNOS with an IC50 ¼ 0.4 mM and greater than 100-fold selectivity over eNOS. However, the selectivity over nNOS was only 2-fold. Compound 29 is an L-arginine competitive human iNOS inhibitor (IC50 ¼ 90 nM) that was optimized from a high-throughput screening (HTS) hit [58]. The compound does not contain a guanidine mimic that
Advancements in the Development of NOS Inhibitors
41
distinguishes it from other compounds described in the literature. It appears to interact primarily through p-stacking and non-bonded interactions. Compound 29 inhibits rat nNOS (IC50 ¼ 0.56 mM) and bovine eNOS (IC50 ¼ 11.1 mM). A new series of aminopiperidines with potent human iNOS activity (IC50 ¼ 0.33 mM; e/iW25, n/i ¼ 16) represented by compound 30 was reported recently [59]. Again as in compound 29, the optimized HTS hit 30 does not contain a guanidine-like functionality, and docking studies reveal that 30 p-stacks with the heme.
CN O
NH2
Cl
O
H N
O
N O
N N
O
29
O
NH
Cl
N
N
30
N
31
A unique class of iNOS inhibitors was obtained by screening a combinatorial library using a whole-cell assay [60]. These compounds inhibit the dimerization of iNOS monomers, thus preventing the formation of the active dimeric form of the enzyme. Optimization led to the identification of compound 31 (IC50 ¼ 14 nM) with selectivity greater than 500 for eNOS and 300 for nNOS. This compound is orally available and was shown to significantly ameliorate adjuvant-induced arthritis in a rat model.
4. CLINICAL FINDINGS WITH NOS INHIBITORS Early clinical trials [61] shed light on the role of NO in the mechanism of migraine pain. In several double-blind studies [62–64], it was shown that after intravenous administration of nitroglycerine (GTN), non-migraineurs rapidly developed a headache that subsided after removal of GTN, whereas migraineurs developed an initial headache, followed by a delayed (B5–6 h) migraine attack [65], suggesting that NO may be partially or completely responsible for migraine pain. A similar observation was made by using sublingual isosorbide dinitrate, another NO donor [65]. The first clinical trial with a NOS inhibitor was conducted using the non-selective NOS inhibitor L-NMMA (32) for the treatment of migraine headache. In a randomized double-blind study, 15 migraine patients with
42
Shawn Maddaford et al.
single spontaneous attacks were infused with 6 mg/kg L-NMMA or placebo (5% dextrose) over a period of 15 min and monitored at 0, 30, 60, and 120 min post-dosing for headache severity, clinical disability (scored on a scale of 0–3), nausea, photophobia, and phonophobia. Heart rate, blood pressure, and ECG were also monitored continuously. A significant antimigraine effect was observed in patients relative to placebo, 67% vs. 25%, albeit with an increase in blood pressure by 17% and decrease in heart rate by 21%. The antimigraine effect was treatment related, most likely the result of specific inhibition of NOS [66–68] and not due to a direct vasoconstriction effect. Migraine relief associated with vasoconstriction was ruled out because of a lack of effect of L-NMMA observed on transcranial Doppler-determined velocity in the middle cerebral artery of the patients. A similar study in the reduction of CTTH by the infusion of L-NMMA was reported [69,70], although the effect was not as pronounced as in the case of migraine headache. NH N H
N H
CO2H NH2
32
In a recent study [71], the effect of infusion of 6 mg/kg L-NMMA on basal and acetazolamide induced changes in cerebral arteries and regional blood flow, indicating that the non-selective NOS inhibitor decreased regional cerebral blood flow only to a minor degree. The authors inferred that in migraine with aura where cerebral blood flow is reduced, treatment of migraine with aura with non-specific inhibitors of NOS might be problematic. Most recently, at the European Headache and Migraine Trust International Congress 2008, three companies presented the results of the effect of NOS inhibitors for the treatment of either acute migraine or migraine prophylaxis. In a study conducted with the selective iNOS inhibitor GW274150 (23) using up to 120 mg daily for 12 weeks for the prophylactic treatment of migraine, the drug was found to be no more effective than the placebo although it did not manifest any intolerable side effects [72]. In a related study utilizing an adaptive clinical trial of GW274150 in the treatment of acute migraine, no significant differences between any of the doses, ranging from 5 to 180 mg, and placebo were observed in the proportion of subjects who became pain free at 2 h after treatment [73]. In addition, no treatment-related benefit was observed for
Advancements in the Development of NOS Inhibitors
43
other migraine symptoms including nausea, photophobia, phonophobia, and allodynia. Since GW274150 is known to reduce exhaled NO levels in patients [74], these studies indicate that the antimigraine effect of L-NMMA is not related to inhibition at the iNOS isoform. In a placebocontrolled double-blind study, the non-selective NOS inhibitor S-alkyl isothiourea MTR-106 was found to be effective in female migraine patients without aura [75]. The study showed statistically significant pain reduction after 2 h with 23% of subjects remaining pain free after 24 h. Although there was no dose-dependent relationship (25, 50, 75, and 100 mg), the drug exhibited linear pharmacokinetics and more adverse effects with higher doses. Recent phase I clinical trials with NXN-188, a novel dual action nNOS inhibitor with 5-HT1B/D agonism, assessed the safety and tolerability of the drug after once daily or multiple twice daily oral doses [76]. The compound exhibited an initial rapid absorption, followed by a more prolonged absorption phase with a Tmax of 4–5 h. The drug displayed near linear pharmacokinetics over the dose range studied (0.027–16 mg/kg). Of the 168 patients that received active drug, no serious adverse events were reported up to the maximally administered dose of 800 mg. Importantly, it was noted that NXN-188 had no effect on blood pressure in the treated patients unlike the non-selective NOS inhibitor L-NMMA. NXN-188 was also studied for the treatment of acute migraine in a phase II study wherein the drug produced extended relief in patients [77]. Future studies will be required to fully elucidate the pharmacological effects of combined nNOS inhibition and 5HT1D/B agonism on the pharmacological profile relative to the existing triptans. Besides inhibition of NOS, another strategy to reduce the amount of NO in migraine patients is to use a NO scavenger. In an open trial, the NO scavenger hydroxocobalamin was assessed in migraine prophylaxis. Daily treatment with intranasal hydroxocobalamin (1 mg) for a period of 3 months demonstrated W50% reduction in the frequency of migraine attacks in 10 of 19 patients. Given the percentage of responders was 35–40%, it is unlikely that this was a placebo effect [78].
5. FUTURE DIRECTIONS — DUAL ACTION NOS INHIBITORS 5.1 Introduction Before the development of molecular biology, traditional drug discovery relied on the use of in vivo animal models of disease to seek out new medicines, often without knowledge of the molecular mechanisms of
44
Shawn Maddaford et al.
their action. Many of the older ‘‘dirty’’ drugs discovered by this means, albeit very efficacious, suffered from undesirable off-target-related side effects. This resulted in a paradigm shift in drug discovery toward optimization of compounds against a single molecular target. Although this approach has delivered many new drugs with potentially fewer side effects, many complex, multifactorial diseases such as cancer, diabetes, and pain are still inadequately treated with single-action drugs. It is likely that complex diseases arise from a complicated network of interdependent biological changes occurring in multiple organs over a period of time. Still, prediction of efficacy or off-target side effects for a single mechanism compound must be put in the context of interspecies differences in target potency and selectivity and the usefulness of acute animal models of disease for evaluating complex, chronic human diseases. To address this biological complexity, there is an increasing readiness to develop agents that modulate multiple targets simultaneously (polypharmacology) and to develop new animal models of chronic disease, with the aim of enhancing efficacy or improving safety relative to drugs that address only a single target.
5.2 Dual action nNOS inhibitors It has been shown that a combination of a nNOS inhibitor and an antioxidant in a model of focal ischemia was synergistic in reducing neuronal damage [79]. A novel strategy for the treatment of stroke consists of designing hybrid molecules such as compound 33, which possess a NOS inhibitor pharmacophore linked to an antioxidant fragment to scavenge reactive oxygen species (ROS). This compound displayed potent nNOS inhibition (Ki ¼ 0.12 mM) and inhibition of lipid peroxidation induced by ROS (IC50 ¼ 0.4 mM). Scavenging ROS and inhibiting NOS simultaneously has been shown to enhance neuronal survival after cerebral ischemia in animal models [80]. Given that NO reacts with O2 to form peroxynitrite, it is perhaps not surprising that the two mechanisms work synergistically. The inhibition of nNOS is known to enhance opioid analgesia and reduce the development of tolerance. It has been shown that the development of morphine-induced hyperalgesia after chronic administration in rats can be delayed or reversed by the addition of a selective nNOS inhibitor [46]. Thus, a dual nNOS inhibitor and m-opioid receptor agonist would be beneficial in the treatment of various pain states. Benzimidazole 34 inhibits nNOS (IC50 ¼ 0.44 mM) and binds to the m-opioid receptor with an IC50 ¼ 13 nM [81]. In a functional assay, the compound displayed much weaker m-opioid receptor agonist activity (EC50 ¼ 0.44 mM).
45
Advancements in the Development of NOS Inhibitors NEt2
N S
H N N H HO
NH
S
S
N
NH
H N
N
N H
NH
N H
OEt
33
34
35
NO derived from nNOS appears to modify the activity of many neurotransmitter systems including glutamate (NMDA), opioid, GABA, and serotonin, likely through spatiotemporal interactions of the nNOS protein with these systems. As an example, nNOS appears to co-localize with the serotonin transporter (SERT) in serotonergic neurons through PDZ-binding domains (PDZ is an acronym combining the first letters of three proteins — post synaptic density protein (PSD95), Drosophila disc large tumor suppressor (DlgA), and zonula occludens-1 protein (zo-1)). This negatively regulates SERT activity, and SERT-mediated 5-HT uptake enhances nNOS activity. Conversely, administration of selective 5-HT reuptake inhibitors or tricyclic antidepressants into the hippocampus reduces nNOS activity. In an effort to reduce the cardiovascular side effect of the triptans, hybrid compounds with both selective nNOS inhibition and 5HT1D/1B agonism were designed. The triptan-like 35 inhibits nNOS (IC50 ¼ 0.89 mM, 40-fold selective over eNOS) and binds to both the 5HT1D and the 5HT1B receptors (5HT1D IC50 ¼ 130 nM and 5HT1B IC50 ¼ 310 nM) [48].
5.3 Dual action iNOS inhibitors In an inflammatory setting or in the presence of endotoxin and cytokines, the inducible NOS isoform is expressed in numerous cell types. Cyclooxygenase (COX), an enzyme which converts arachidonic acid to prostaglandins (PG), is another critical enzyme in many inflammatory diseases. Studies have shown that production of high levels of PG can be augmented in the presence of NO. Early amino acid–selective inhibitors of iNOS such as N-iminoethyl-L-lysine (L-NIL) have been shown to not only block NO formation but also attenuate the elevated release of prostaglandins [82]. The c-Src kinase was identified as a proto-oncogene, and in some abnormal cases, such as mutation or over-expression, these enzymes can become hyperactivated, resulting in uncontrolled cell proliferation. The role of iNOS during tumor development is highly complex and incompletely understood, but there is evidence to show that the malignant transformation, angiogenesis, and metastasis effects of tumors are all modulated by iNOS. A potential anticancer strategy involves the dual inhibition of c-Src and iNOS, two key enzymes in tumorigenesis.
46
Shawn Maddaford et al.
Modification of a protein tyrosine kinase inhibitor to incorporate an NOS pharmacophore led to compound 36. This compound potently inhibits c-Src kinase (IC50 ¼ 9.2 nM) and is a modest inhibitor for iNOS of mouse macrophage ANA-1 (IC50 ¼ 2.2 mM) [83]. Studies have shown that iNOS and peroxisome proliferator-activated receptor g (PPARg) play important roles in neuroinflammation. For instance, the PPARg agonists Rosiglitazone and Pioglitazone, which are approved for the treatment of diabetes, preserve cognitive function in early Alzheimer’s patients and attenuate learning and memory deficits in transgenic Alzheimer mouse models. In a rat spinal cord injury model, treatment of rats with thiazolidinedione PPARg agonists prevented neuronal damage, motor dysfunction, myelin loss, and the development of neuropathic pain [84]. Similarly, increased iNOS function is important in neuroinflammation and pain. A series of 2-phenyl-ethenesulfonic acid phenyl esters were synthesized and shown to suppress NO production in LPS/interferon g-stimulated RAW 264.7 cells and activate PPARg in a cell-based transactivation assay [85]. Compound 37 inhibits iNOS activity (IC50 ¼ 1.8 mM) in a whole cell assay and binds competitively to the PPARg receptor (IC50 ¼ 1.4 mM).
O
Cl
N
O
N
O S
N O
CN
O Cl
NH
S
Cl
N
36
37
6. CONCLUSIONS The preceding sections of this review exemplify the connection between NO and various central nervous system (CNS) disorders. Therefore, targeting isoforms of NOS may be a valid strategy to produce novel mechanism-based therapeutics. Despite the enormous efforts by multiple groups to obtain isoform-selective compounds, it is surprising that no candidates have progressed into the clinic until recently. One of the reasons may be that in most cases, the enzymes chosen for primary assay were of rat, mouse, or bovine origin and not of human origin. Although
Advancements in the Development of NOS Inhibitors
47
there is considerable homology between the enzymes from different species, differences in the inhibitory potencies exist between species (e.g., compound 16b). Therefore, a rat isoform-selective compound showing the desired pharmacological effect in rats (usual animal species for initial screening) may not translate into humans because of low potency or non-selectivity in human forms of the enzymes. Still, tremendous progress has been achieved in the design and synthesis of selective nNOS or iNOS inhibitors using various techniques including structure-based design and HTS. These recent compounds are more drug-like as demonstrated by preclinical animal studies and recent clinical trials. Novel chemical entities that combine several selective and additive mechanisms into one molecule (designed multiple ligands) [86] may also address the failure of current analgesics to control hyperalgesia and allodynia. It can be hoped that NOS inhibitors with appropriate potency and selectivity, and dual action selective nNOS or iNOS small molecule inhibitors incorporating additional mechanisms of action will ultimately improve the current standard of care for pain management [87] and the treatment of CNS disorders.
REFERENCES [1] L. J. Ignarro, G. M. Buga, K. S. Wood, R. E. Byrns and G. Chaudhuri, Proc. Natl. Acad. Sci. U.S.A., 1987, 84, 9265. [2] R. F. Furchgott, in Mechanisms of Vasodilation (ed. P. M. Vanhoutte), Raven, New York, 1988, pp. 401–404. [3] P. Vallance and J. Leiper, Nat. Rev. Drug. Disc., 2002, 1, 939. [4] A. J. Cayatte, J. J. Palacino, K. Horten and R. A. Cohen, Arterioscler. Thromb., 1994, 14, 753. [5] J. M. Fukuto and Y. Komori, Annu. Rep. Med. Chem., 1994, 29, 83. [6] J. E. MacDonald, Annu. Rep. Med. Chem., 1996, 31, 221. [7] D. J. Stuehr, Annu. Rev. Pharmacol. Toxicol., 1997, 37, 339. [8] D. S. Bredt, P. M. Hwang, C. E. Glatt, C. Lowenstein, R. R. Reed and S. H. Snyder, Nature, 1991, 351, 714. [9] H. J. Cho, Q. W. Xie, J. Calaycay, R. A. Mumford, K. M. Swiderek, T. D. Lee and C. Nathan, J. Exp. Med., 1992, 176, 599. [10] T. O. Fischmann, A. Hruza, X. D. Niu, J. D. Fossetta, C. A. Lunn, E. Dolphin, A. J. Prongay, P. Reichert, D. J. Lundell, S. K. Narula and P. C. Weber, Nat. Struct. Biol., 1999, 6, 233. [11] S. Pou, L. Keaton, W. Surichamorn and G. M. Rosen, J. Biol. Chem., 1999, 274, 9573. [12] B. R. Crane, A. S. Arvai, R. Gachhui, C. Wu, D. K. Ghosh, E. D. Getzoff, D. J. Stuehr and J. A. Tainer, Science, 1997, 278, 425. [13] B. R. Crane, A. S. Arvai, D. K. Ghosh, C. Wu, E. D. Getzoff, D. J. Stuehr and J. A. Tainer, Science, 1998, 279, 2121. [14] C. S. Raman, H. Li, P. Marta´sek, V. Kral, B. S. S. Masters and T. L. Poulos, Cell, 1998, 95, 939. [15] H. Li, C. S. Raman, C. B. Glaser, E. Blasko, T. A. Young, J. F. Parkinson, M. Whitlow and T. L. Poulos, J. Biol. Chem., 1999, 274, 21276.
48
Shawn Maddaford et al.
[16] H. Li, H. Shimizu, M. Flinspach, J. Jamal, W. Yang, M. Xian, T. Cai, E. Z. Wen, Q. Jia, P. G. Wang and T. L. Poulos, Biochemistry, 2002, 41, 13868. [17] E. D. Garcin, A. S. Arvai, R. J. Rosenfeld, M. D. Kroeger, B. R. Crane, G. Andersson, G. Andrews, P. J. Hamley, P. R. Mallinder, D. J. Nicholls, S. A. St-Gallay, A. C. Tinker, N. P. Gensmantel, A. Mete, D. R. Cheshire, S. Connolly, D. J. Stuehr, A. Aberg, A. V. Wallace, J. A. Tainer and E. D. Getzoff, Nat. Chem. Biol., 2008, 4, 700. [18] J. Haitao, L. Huiying, P. Marta´sek, L. J. Roman, T. L. Poulos and R. B. Silverman, J. Med. Chem., 2009, 52, 779. [19] M. L. Flinspach, H. Li, J. Jamal, W. Yang, H. Huang, J.-M. Hah, J. A. Go´mez-Vidal, E. A. Litzinger, R. B. Silverman and T. L. Poulos, Nat. Struct. Mol. Biol., 2004, 11, 54. [20] C. S. Raman, H. Li, P. Marta´sek, G. Southan, B. S. S. Masters and T. L. Poulos, Biochemistry, 2001, 40, 13448. [21] A. M. Davis, S. A. St-Gallay and G. J. Kleywegt, Drug. Disc. Today, 2008, 13, 831. [22] J. Seo, J. Igarashi, H. Li, P. Marta´sek, L. J. Roman, T. L. Poulos and R. B. Silverman, J. Med. Chem., 2007, 50, 2089. [23] E. P. Erdal, E. A. Litzinger, J. Seo, Y. Zhu, H. Ji and R. B. Silverman, Curr. Top. Med. Chem., 2005, 5, 603. [24] R. B. Silverman, Acc. Chem. Res., 2009, 42, 439. [25] A. C. Tinker and A. V. Wallace, Curr. Top. Med. Chem., 2006, 6, 77. [26] A. Tafi, L. Angeli, G. Venturini, M. Travagli, F. Corelli and M. Botta, Curr. Med. Chem., 2006, 13, 1929. [27] L. G. Howes and D. G. Brillante, Exp. Opin. Invest. Drugs, 2008, 17, 1573. [28] G. Folkerts and F. P. Nijkamp, Curr. Pharm. Des., 2006, 12, 3221. [29] J. Olesen, Pharmacol. Ther., 2008, 120, 157. [30] P.-E. Chabrier, C. Demerle-Pallardy and M. Auguet, Cell. Mol. Life Sci., 1999, 55, 1029. [31] D. O. Stichtenoth and J. C. Frolish, Br. J. Rheumatol., 1998, 37, 246. [32] H. Ji, B. Z. Stanton, J. Igarashi, H. Li, P. Marta´sek, L. J. Roman, T. L. Poulos and R. B. Silverman, J. Am. Chem. Soc., 2008, 130, 3900. [33] R. L. C. Handy, P. Wallace, Z. A. Gaffen, K. J. Whitehead and P. K. Moore, Br. J. Pharmacol., 1995, 116, 2349. [34] R. L. C. Handy and P. K. Moore, Life Sci., 1997, 60, 389. [35] M. Boulouard, P. Schumann-Bard, S. Butt-Gueulle, E. Lohou, S. Stiebing, V. Collot and S. Rault, Bioorg. Med. Chem. Lett., 2007, 17, 3177. [36] B. Cottyn, F. Acher, B. Ramassamy, L. Alvey, M. Lepoivre, Y. Frapart, D. Stuehr, D. Mansuy, J.-L. Boucher and D. Vichard, Bioorg. Med. Chem., 2008, 16, 5962. [37] V. Volke, G. Wegener, M. Bourin and E. Vasar, Behav. Brain Res., 2003, 140, 141. [38] G. Ulak, O. Mutlu, F. Y. Akar, F. I. Komsuog˘lu, P. Tanyeri and B. F. Erden, Pharmacol. Biochem. Behav., 2008, 90, 563. [39] L. Salerno, V. Sorrenti, C. Di Giacomo, G. Romeo and M. A. Siracusa, Curr. Pharm. Des., 2002, 8, 177. [40] M. J. O’Neill, T. K. Murray, D. R. McCarty, C. A. Hicks, C. P. Dell, K. E. Patrick, M. A. Ward, D. J. Osborne, T. R. Wiernicki, C. R. Roman, D. Lodge, J. H. Fleisch and J. Singh, Brain Res., 2000, 871, 234. [41] C. A. Hicks, M. A. Ward, J. B. Swettenham and M. J. O’Neill, Eur. J. Pharmacol., 1999, 381, 113. [42] R. A. Lefebvre, J. M. C. Dick, S. Guerin and C.-H. Malbert, Eur. J. Pharmacol., 2005, 525, 143. [43] S. Jaroch, P. Holscher, H. Rehwinkel, D. Sulzle, G. Burton, M. Hillmann and F. M. McDonald, Bioorg. Med. Chem. Lett., 2003, 13, 1981. [44] S. Jaroch, H. Rehwinkel, P. Holscher, D. Sulzle, G. Burton, M. Hillmann, F. M. McDonald and H. Miklautz, Bioorg. Med. Chem. Lett., 2004, 14, 743. [45] J. Patman, N. Bhardwaj, J. Ramnauth, S. C. Annedi, P. Renton, S. P. Maddaford, S. Rakhit and J. S. Andrews, Bioorg. Med. Chem. Lett., 2007, 17, 2540.
Advancements in the Development of NOS Inhibitors
49
[46] J. Ramnauth, N. Bhardwaj, S. Rakhit and S. Maddaford, US Patent 7141595 B2, 2006. [47] S. Maddaford, J. Ramnauth, S. Rakhit, J. Patman, P. Renton and S. C. Annedi, PCT Patent WO 2007118314 A1, 2007. [48] S. Maddaford, J. Ramnauth, S. Rakhit, J. Patman, P. Renton and S. C. Annedi, US Patent US2006258721 A1, 2006. [49] S. Maddaford, J. Ramnauth, S. Rakhit, J. Patman, S. C. Annedi, J. Andrews, P. Dove, S. Silverman and P. Renton, PCT Patent WO 2008116308 A1, 2008. [50] J. A. Lowe III, W. Qian, R. A. Volkmann, S. Heck, J. Nowakowski, R. Nelson, C. Nolan, D. Liston, K. Ward, S. Zorn, C. Johnson, M. Vanase, W. S. Faraci, K. A. Verdries, J. Baxter, S. Doran, M. Sanders, M. Ashton, P. Whittle and M. Stefaniak, Bioorg. Med. Chem. Lett., 1999, 9, 2569. [51] J. A. Lowe III, W. Qian, S. E. Drozda, R. A. Volkmann, D. Nason, R. B. Nelson, C. Nolan, D. Liston, K. Ward, S. Faraci, K. Verdries, P. Seymour, M. Majchrzak, A. Villalobos and W. F. White, J. Med. Chem., 2004, 47, 1575. [52] H. Matter, H. S. A. Kumar, R. Fedorov, A. Frey, P. Kotsonis, E. Hartmann, L. G. Froehlich, A. Reif, W. Pfleiderer, P. Scheurer, D. K. Ghosh, I. Schlichting and H. H. H. W. Schmidt, J. Med. Chem., 2005, 48, 4783. [53] R. J. Young, R. M. Beams, K. Carter, H. A. R. Clark, D. M. Coe, C. L. Chambers, P. I. Davies, J. Dawson, M. J. Drysdale, K. W. Franzman, C. French, S. T. Hodgson, H. F. Hodson, S. Kleanthous, P. Rider, D. Sanders, D. A. Sawyer, K. J. Scott, B. G. Shearer, R. Stocker, S. Smith, M. C. Takley and R. G. Knowles, Bioorg. Med. Chem. Lett., 2000, 10, 597. [54] R. N. Guthikonda, S. K. Shah, S. G. Pacholok, J. L. Humes, R. A. Mumford, S. K. Grant, R. M. Chabin, B. G. Green, N. Tsou, R. Ball, D. S. Fletcher, S. Luell, D. E. MacIntyre and M. MacCoss, Bioorg. Med. Chem. Lett., 2005, 15, 1997. [55] S. Connolly, A. Aberg, A. Arvai, H. G. Beaton, D. R. Cheshire, A. R. Cook, S. Cooper, D. Cox, P. Hamley, P. Mallinder, I. Millichip, D. J. Nicholls, R. J. Rosenfeld, S. A. St-Gallay, J. Tainer, A. C. Tinker and A. V. Wallace, J. Med. Chem., 2004, 47, 3320. [56] A. Strub, W.-R. Ulrich, C. Hesslinger, M. Eltze, T. Fuchss, J. Strassner, S. Strand, M. D. Lehner and R. Boer, Mol. Pharmacol., 2006, 69, 328. [57] A. C. Tinker, H. G. Beaton, N. Boughton-Smith, T. R. Cook, S. L. Cooper, L. Fraser-Rae, K. Hallam, P. Hamley, T. McInally, D. J. Nicholls, A. D. Pimm and A. V. Wallace, J. Med. Chem., 2003, 46, 913. [58] S. A. Jackson, S. Sahni, L. Lee, Y. Luo, T. R. Nieduzak, G. Liang, Y. Chiang, N. Collar, D. Fink, W. He, A. Laoui, J. Merrill, R. Boffey, P. Crackett, B. Rees, M. Wong, J.-P. Guilloteau, M. Mathieu and S. S. Rebello, Bioorg. Med. Chem., 2005, 13, 2723. [59] B. Le Bourdonnec, L. K. Leister, C. A. Ajello, J. A. Cassel, P. R. Seida, H. O’Hare, M. Gu, G.-H. Chu, P. A. Tuthill, R. N. DeHaven and R. E. Dolle, Bioorg. Med. Chem. Lett., 2008, 18, 336. [60] D. D. Davey, M. Alder, D. Arnaiz, K. Eagen, S. Erickson, W. Guilford, M. Kenrick, M. M. Morrissey, M. Ohlmeyer, G. Pan, V. M. Paradkar, J. Parkinson, M. Polokoff, K. Saionz, C. Santos, B. Subramanyam, R. Vergona, R. G. Wei, M. Whitlow, B. Ye, Z. Zhao, J. J. Devlin and G. Phillips, J. Med. Chem., 2007, 50, 1146. [61] J. Olesen, H. K. Iversen and L. L. Thomsen, NeuroReport, 1993, 4, 1027. [62] J. Olesen, L. L. Thomsen and H. Iversen, Trends in Pharmacol. Sci., 1994, 15, 149. [63] L. L. Thomsen, H. K. Iversen, T. A. Brinck and J. Olesen, Cephalalgia, 1993, 13, 395. [64] S. K. Afridi, H. Kaube and P. J. Goadsby, Pain, 2004, 110, 675. [65] P. Bellantonio, G. Micieli, M. G. Buzzi, S. Marcheselli, A. E. Castellano, F. Rossi and G. Nappi, Cephalalgia, 1997, 17, 183. [66] L. H. Lassen, M. Ashina, I. Christiansen, V. Ulrich and J. Olesen, Lancet, 1997, 349, 401. [67] L. H. Lassen, H. K. Iversen and J. Olesen, Eur. J. Clin. Pharmacol., 2003, 59, 499.
50
Shawn Maddaford et al.
[68] L. H. Lassen, M. Ashina, I. Christiansen, V. Ulrich, R. Grover, J. Donaldson and J. Olesen, Cephalalgia, 1998, 18, 27. [69] M. Ashina, L. H. Lassen, L. Bendtsen, R. Jensen and J. Olesen, Lancet, 1999, 353, 287. [70] M. Ashina, Exp. Opin. Pharmacother., 2002, 3, 395. [71] L. H. Lassen, B. Sperling, A. R. Andersen and J. Olesen, Cephalalgia, 2005, 25, 344. [72] K. Hoye, B. E. Laurijssens, L. O. Harnisch, C. K. Twomey, R. M. Dixon, A. Kirkham, P. M. Williams and A. L. Wentz, Cephalalgia, 2009, 29, 132. [73] J. E. Palmer, F. L. Guillard, B. E. Laurijssens, A. L. Wentz, R. M. Dixon and P. M. Williams, Cephalalgia, 2009, 29, 124. [74] D. Singh, D. Richards, R. G. Knowles, S. Schwartz, A. Woodcock, S. Langley and B. J. O’Connor, Am. J. Respir. Crit. Care Med., 2007, 176, 988. [75] A. Mosek, S. Groppa and R. Barkan, Cephalalgia, 2009, 29, 118. [76] D. Vaughan, J. Speed, R. Medve and J. S. Andrews, unpublished results. [77] R. A. Medve and J. S. Andrews, Cephalalgia, 2009, 29, 126. [78] P.-H. M. van der Kuy, F. W. H. M. Merkus, J. J. H. M. Lohman, J. W. M. ter Berg and P. M. Hooymans, Cephalalgia, 2002, 22, 513. [79] S. Auvin, M. Auguet, E. Navet, J. J. Harnett, I. Viossat, J. Schulz, D. Bigg and P.-E. Chabrier, Bioorg. Med. Chem. Lett., 2003, 13, 209. [80] P.-E. Chabrier, M. Auguet, B. Spinnewyn, S. Auvin, S. Cornet, C. Demerle-Pallardy, C. Guilmard-Favre, J.-G. Marin, B. Pignol, V. Gillard-Roubert, C. Roussillot-Charnet, J. Schulz, I. Viossat, D. Bigg and S. Moncada, Proc. Natl. Acad. Sci. U.S.A., 1999, 96, 10824. [81] P. Renton, S. Maddaford, S. Rakhit and J. Andrews, PCT Int. Appl., WO2007017764 A2, 2007. [82] D. Salvemini, P. T. Manning, B. S. Sweifel, K. Seibert, J. Connor, M. G. Currie, P. Needleman and J. L. Masferrer, J. Clin. Invest., 1995, 96, 301. [83] X. Cao, Q.-D. You, Z.-Y. Li, Q.-L. Guo, J. Shang, M. Yan, J.-W. Chern and M.-L. Chen, Bioorg. Med. Chem., 2008, 16, 5890. [84] S.-W. Park, J.-H. Yi, G. Miranpuri, I. Satriotomo, K. Bowen, D. K. Resnick and R. Vemuganti, J. Pharm. Exp. Ther., 2007, 320, 1002. [85] Y.-Z. Lee, C.-W. Yang, I.-J. Kang, S.-H. Wu, Y.-S. Chao, J.-H. Chern and S.-J. Lee, Bioorg. Med. Chem. Lett., 2008, 18, 5676. [86] R. Morphy and Z. Rankovic, J. Med. Chem., 2005, 48, 1. [87] J. Woodcock, J. Witter and R. A. Dionne, Nat. Rev. Drug Discovery, 2007, 6, 703.
CHAPT ER
3 Small-Molecule Protein–Protein Interaction Inhibitors as Therapeutic Agents for Neurodegenerative Diseases: Recent Progress and Future Directions Simon N. Haydar*, Heedong Yun*, Roland G.W. Staal** and Warren D. Hirst**
Contents
1. Introduction 2. Ab Aggregation and Oligomers in Alzheimer’s Disease 2.1 Ab aggregation and neurotoxic oligomers 2.2 Small-molecule inhibitors of Ab aggregation 3. Tau Aggregation in Alzheimer’s Disease 3.1 Tau pathophysiology 3.2 Small-molecule inhibitors of tau aggregation 4. a-Synuclein Aggregation in Parkinson’s Disease 4.1 Biochemistry of a-synuclein aggregation 4.2 Small-molecule inhibitors of a-synuclein aggregation 5. Conclusion References
51 52 52 53 58 58 59 63 63 64 65 66
* Chemical Sciences, Wyeth Research, CN 8000, Princeton, NJ 08543 ** Neuroscience, Wyeth Research, CN 8000, Princeton, NJ 08543 Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04403-0
r 2009 Elsevier Inc. All rights reserved.
51
52
Simon N. Haydar et al.
1. INTRODUCTION Alzheimer’s disease (AD) and Parkinson’s disease (PD) are the two most common chronic, progressive neurodegenerative diseases affecting an estimated 10% and 1%, respectively, of the elderly population [1,2], with financial costs of hundreds of billions of dollars per year. AD is characterized by the presence of extracellular parenchymal and vascular amyloid deposits containing b-amyloid peptide (Ab) and intracellular neuronal tangles composed of hyperphosphorylated tau. a-Synuclein containing Lewy bodies, spherical inclusions found in the cytoplasm of surviving neurons, are the cardinal hallmark of PD. Despite their distinct pathologies, these neurodegenerative diseases are increasingly being realized to have common cellular and molecular mechanisms including protein aggregation and inclusion body formation. Compelling evidence strongly supports the hypothesis that accumulation of misfolded proteins leads to synaptic dysfunction, neuronal apoptosis, brain damage, and disease. However, the mechanisms by which protein misfolding and aggregation trigger neurodegeneration and the identity of the neurotoxic structures are still unclear. Current hypotheses propose that, in the aggregation process, there is an accumulation of small soluble oligomeric intermediates, which leads to the neuropathology, whereas the large insoluble deposits that make up the inclusion bodies might function as reservoirs of these toxic, soluble oligomers [3]. As we increase our knowledge of the role of oligomeric, fibrillar, and higher-order molecular entities of the misfolded proteins in neurodegenerative diseases, new approaches may offer themselves for therapeutic intervention. Over the past few years, there has been significant interest in developing therapeutics and chemical probes that inhibit these specific protein–protein interactions. This effort has been hampered by the size and the geometry of the protein interaction interface, which are devoid of defined ‘‘pockets’’ into which a small molecule can bind in an energetically favorable manner [4]. Despite the challenges of developing compounds that are capable of specifically inhibiting protein–protein interactions, there are a number of examples of small molecules that achieve this with reasonable potency [5]. This was made possible because of the discovery of ‘‘hot spots’’ on the protein interaction surfaces [6]. These ‘‘hot spots’’ are small regions on the protein interaction interface that are responsible for a disproportionate contribution to the binding energy of the two proteins. This review will highlight the recent progress in the development of small-molecule protein–protein interaction inhibitors that have applications in furthering the mechanistic understanding of neurodegenerative diseases and will potentially lead to the development of rational therapeutics.
Therapeutic Agents for Neurodegenerative Diseases
53
2. Ab AGGREGATION AND OLIGOMERS IN ALZHEIMER’S DISEASE 2.1 Ab aggregation and neurotoxic oligomers Substantial genetic and physiological evidence suggest that the Ab plays a central role in AD pathogenesis. Ab is a 39- to 42-amino acid peptide derived from the proteolytic processing of the amyloid precursor protein (APP) by secretases [7]. Gradual changes in the steady-state levels of Ab in the brain are thought to initiate the amyloid cascade [8,9]. Since the elucidation of the Ab sequences [10–12], investigators have used synthetic Ab to examine aggregation and its effects on physiology. Many in vitro studies have suggested that aggregation of Ab is essential for toxicity, but characterization of the Ab species that formed was limited. However, amyloid plaque number does not correlate well with severity of dementia [13–15], and instead, there is a stronger link between soluble Ab levels and the extent of synaptic loss and the severity of cognitive impairment [16–18]. Therefore, more recent studies have focused on various soluble forms of synthetic Ab1-40 and Ab1-42, ranging from monomeric to protofibrils [19]. A number of reports have described the biochemical characterization of the soluble Ab extracted from human AD brain. The presence of sodium dodecyl sulfate (SDS)-stable dimers and trimers in the soluble fraction of human brain and in extracts of amyloid plaques suggests that SDS-stable, low n oligomers of Ab are the fundamental building blocks of insoluble amyloid deposits and could be the earliest mediators of neuronal dysfunction [20]. Recent studies have described the physiological characterization of Ab dimers isolated from AD brains that inhibit long-term potentiation (LTP), a physiological correlate of memory, and reduce dendritic spine density in normal rodent hippocampus [21]. In addition, the dimers disrupt memory of a learned behavior when directly injected into the brains of normal rats. As outlined earlier, a number of Ab assemblies have been proposed to exert neurotoxic effects, but with evidence only recently emerging on which forms that are the pathological species in vivo, and the scarcity of structural data on the oligomers complicates a rational search for compounds that could inhibit Ab aggregation and toxicity. Despite these obstacles, a number of different compounds that interfere with Ab aggregation, in one way or another, have been described; these are reviewed in the following sections.
2.2 Small-molecule inhibitors of Ab aggregation 2.2.1 Scyllo-inositol Recent in vitro studies with scyllo-inositol (1) have shown that it can interact with Ab42 peptide promoting a conformational change from
54
Simon N. Haydar et al.
random coil to b-sheet structure and stabilized it in small, nonfibrillar complexes, blocking fibril formation [22]. These stabilized complexes were significantly less toxic to neuronal cell lines and primary neuronal cultures than untreated Ab42 or chiro-inositol-treated Ab42 [22]. Moreover, in in vivo studies in the TgCRND8 mouse model of AD, it was shown that compound 1 inhibited Ab aggregation, decreased Ab-induced impairments in spatial memory, reduced the cerebral Ab pathology, and attenuated the rate of mortality [23]. To explore the molecular details of the inositol–Ab42 interaction, a series of scylloinositols were prepared in which one or two hydroxyl groups were replaced with fluoro, chloro, methoxy, or hydrogen substituents [24]. After incubation with Ab42 for 7 days, the activity of the derivatives was measured by electron microscopy to monitor the formation of Ab42 fibers. Despite the synthesis of numerous analogs, only single hydroxyl substitutions such as 1-deoxy-1-fluoro-scyllo-inositol (2) and the disubstituted analog 1,4-dimethoxy-scyllo-inositol (3) were shown to have similar activity to the parent compound 1. Compound 3 was shown to exhibit the most pronounced effect on Ab42 aggregation, in that it produced a more homogenous population of small amorphous aggregates and no fibers were detected [24]. OH
HO HO HO
OH OH 1
OH
HO HO HO
OH F 2
OH
MeO HO HO
OH OMe 3
2.2.2 Tricyclic pyrones and pyridinones Hua et al. [25] used a neuronal cell line overexpressing a C-terminal fragment of APP (MC65 cells) to identify inhibitors of toxicity related to intracellular Ab and discovered a class of tricyclic pyrones (TP). In particular, a TP 4 that contains an adenine moiety (at N-3u) attached at the C7-alkyl side chain of the ring system showed significant protection in the MC65 cell assay with an EC50 value of 0.31 mM [25]. Further characterization of compound 4 using surface plasmon resonance (SPR) spectroscopy, atomic force microscopy (AFM), and protein quantification studies showed it binds to Ab42 oligomers, inhibits Ab aggregation, and disaggregates Ab42 oligomers and protofibrils [26]. Transgenic mice treated for 2 weeks with compound 4, administered i.c.v., resulted in 40% and 50% decreases in non-fibrillar and fibrillar Ab species, respectively. [26]. To further investigate the structure activity relationship, various heterocycles and nitrogen-containing TP were prepared [27]. It was concluded that attachment of N3u-adenine at C7 side chain in compound
Therapeutic Agents for Neurodegenerative Diseases
55
4 provides the strongest MC65 protective activity with an EC50 of 0.31 mM. However, tricyclic pyranoisoquinolinones lacking the adenine moeity as in compounds 5 and 6 still possess protective activity with EC50 values of 2.49 and 1.25 mM, respectively. The 6u-amino group in compound 4 enhances potency but is not required for activity. O
O 1
10
H
O N
7
NH2
N 14
12
N
3′
N 9′
4
OMe O
O
O H O
N
N 5
6
Among the various C7 side chain heterocycles prepared, none showed better protective activity than the adenine in compound 4. It is noteworthy that the most potent analog prepared in this class was a 2-aminopurine derivative 7 with EC50 of 0.86 mM. Replacement of the oxygen at position 2 with a nitrogen, as in compound 8, provides similar protective activity (EC50 ¼ 0.35 mM) as that of compound 4 [27]. O
O
OMe
1 10
H
O
O
H O
N 7
N
N
N 14
12
3′
N 7
N NH2
N
N N
NH2 8
56
Simon N. Haydar et al.
2.2.3 Indoles Highly ordered p-stacking interactions between aromatic ring systems are important in self-assembly of complex biological and chemical supramolecular structures [28]. Several studies showed that aromatic interactions may play a critical role in early-stage amyloid formation by providing stability, as well as order and directionality in the formation of amyloid fibrils, presumably facilitated by restricted geometry of interaction between planar aromatic systems [29]. In particular, in amyloid peptide fragments, a high frequency of aromatic residues was noted. When these aromatic residues were replaced with hydrophobic amino acids, a decrease in the amyloidogenic propensity was observed [30]. Gazit et al. screened various indole derivatives for their ability to prevent formation of amyloid fibrils using fluorescence spectroscopy, AFM, and electron microscopy. Three inhibitors were identified: indole-3-carninol (9), 3hydroxyindole (10), and 4-hydroxyindole (11) with IC50 values of 85, 100, and 200 mM, respectively. These simple indoles effectively inhibited Ab fibril formation and prevented cell death induced by 5 mM Ab40 in PC12 cells in culture. Curiously, analog 9 inhibited Ab fibrillization only at high concentration, and it did not show a dose dependency on inhibition of fibril formation. The inhibitory mechanism of these compounds remains unclear; however, the authors suggest that the hydroxyl group interacts with the backbone of the peptides preventing the ability of the Ab peptide to create a p-stacking interaction, which limit the fibrillogenesis process. OH
OH
HO
N H
N H 9
N H 10
11
Recent adsorption studies of Ab solutions on poly(tetrafluoroethylene) surfaces showed that the fluorinated surface strongly promoted a-helix re-formation [31]. A similar effect was observed with solution of Ab in CF3-containing solvents [32]. On the basis of these findings, Torok et al. [33] demonstrated the design and application of a new class of trifluoroethylindoles against amyloid fibrillogenesis. Analogs 12a–c showed significant inhibitory effect with IC50 values of 0.53, 0.23, and 0.36 molinhibitor/molAb. Further structure–activity relationship demonstrated that the CF3 and OH groups are necessary for binding to Ab peptide. It was suggested that the acidity of the hydroxyl group plays a key role in binding to one or both lysine residues of the Ab peptide. Interestingly, removing the ester group slightly diminished the inhibitory activity.
Therapeutic Agents for Neurodegenerative Diseases
57
HO CO Et 2 F3C X N H 12a, b, c a; X = Cl b; X = Br c; X = I
Tramiprosate (3-amino-1-propanesulfonic acid; 3APS; Alzhemedt) (13) was found to maintain Ab in a non-fibrillar form, decrease Ab42-induced cell death in neuronal cell cultures, and also inhibit amyloid deposition [34]. Treatment of TgCRND8 mice with Tramiprosate resulted in significant reduction (B30%) in the brain amyloid plaque load and a significant decrease in the cerebral levels of soluble and insoluble Ab40 and Ab42 (B20–30%) [34]. Although Tramiprosate ultimately failed in a phase III clinical trial, it provided a proof of concept that small-molecule inhibitors of Ab protofibril formation may be a viable approach to AD treatment [5].
H2N
O O S OH
13, Tramiprosate
Another recent development in Ab aggregation inhibitors was the development of Memoquin (14). This compound was shown to be a multifunctional therapy to AD, acting as an acetyl cholinesterase (AchE) inhibitor (Ki ¼ 2.6 nM), a free radical scavenger, and an inhibitor of Ab aggregation [35,36].
N O
O
H N
N H
O
N
O 14, Memoquin
Additional small molecules such as Congo red, curcumin, and galantamine have been described in the literature as inhibitors of Ab aggregation, recently reviewed by Hawkes et al. (2009) [37]. These molecules were also found to be inhibitors of a-synuclein, and we will report on their activity in Section 4.2.
58
Simon N. Haydar et al.
3. TAU AGGREGATION IN ALZHEIMER’S DISEASE 3.1 Tau pathophysiology Tau is predominantly expressed in neurons where its main function is thought to be stabilizing microtubules, particularly in axons. The tau gene consists of 16 exons and alternative splicing results in six isoforms of tau protein ranging in size from 352–441 amino acids [38]. Tau stabilizes microtubules by binding to them through an interaction with the three or four microtubule-binding domains at the C-terminus of the protein. Stabilization of microtubules by tau in neurons is important for maintenance of cellular morphology and transport of molecules and organelles over long distances [39]. Binding of tau to microtubules is also regulated post-translationally, primarily through phosphorylation, although other modifications such as glycosylation, ubiquitylation, and proteolysis have been reported for the tau protein [40]. Tau contains W80 serine and threonine residues, which are potential phosphorylation sites. The phosphorylation state, which is controlled by a balance of kinase and phosphatase activity, affects the microtubule-binding affinity. Hyperphosphorylation of tau at many sites, as seen in tauopathies, of which AD is the most common, leads to reduced affinity for microtubules, which causes disruption of cellular trafficking leading to degeneration of synaptic terminals. This loss of function may be exacerbated by a toxic gain of function, where higher than normal concentrations of tau increase the chances of pathogenic conformational changes, which in turn lead to the aggregation and fibrillization, which might block transport and cause cell death [40]. Phosphorylation of certain residues on tau, specifically S396 and S404, has been shown to increase the fibrillogenic nature of tau and contribute to its accumulation into paired helical filaments [41–43]. Similarly, the removal of the C-terminus of the protein increases tau aggregation [44,45]. It is thought that tau aggregation occurs in a multi-step process whereby tau is phosphorylated and dissociates from microtubules. The unbound hyperphosphorylated tau abnormally localizes to the somatodendritic compartment of the cell, undergoes conformational changes and further phosphorylation. Finally, the hyperphosphorylated tau forms fibrils that aggregate into neurofibrillary tangles (NFTs) [46]. Tau aggregates also form in axons and dendrites, called neuropil threads. Both NFTs and neuropil threads are postulated to have a toxic gain of function. In cell models, tau aggregation in the cell body causes cell death [47]. In axons, neuropil threads are thought to be toxic because they might physically impair transport, which would be toxic to synaptic terminals [48]. Similar to Ab, oligomeric forms of tau, which are promoted by phosphorylation [49] and are observed in aging and early AD [50], could
59
Therapeutic Agents for Neurodegenerative Diseases
also be the toxic species leading to neurodegeneration. In animal models of tauopathy, there is indirect evidence that soluble, not aggregated, forms of tau are toxic: flies expressing human tau can exhibit neurodegeneration without fibril formation [51], and in some mouse models of tauopathy, overexpression of tau causes neuronal loss in areas without extensive neurofibrillary pathology [52]. If oligomeric tau is toxic, formation of large aggregates could be viewed as protective because oligomers are sequestered into insoluble neurofibrillary pathology [3,49]. The two principal current strategies targeting tau in neurodegenerative disease are (i) reducing tau phosphorylation through inhibition of specific protein kinases [53] and (ii) anti-aggregation approaches [54]. The issue of whether phosphorylation of tau precedes or follows tau aggregation remains a subject of debate, but reducing tau phosphorylation is regarded by many as the preferred target, and some transgenic animal studies have shown this to be a valid strategy [53]. In the following sections, structurally diverse small-molecule inhibitors of tau aggregation are described.
3.2 Small-molecule inhibitors of tau aggregation 3.2.1 Phenylthiazolyl hydrazides In an effort to develop small-molecule inhibitors of tau aggregation, Mandelkow et al. [55,56] identified compounds related to structure 15 from a high throughput screen of a collection of 200,000 compounds. To establish the structure–activity relationship (SAR), a series of thiazolylhydrazides were prepared by synthetic derivatization of R1, R2, R3, and R4 [56]. Structure–activity relationship of phenylthiazolyl hydrazides demonstrated that two aromatic rings at R1 and R4, a hydrophobic region on the thiazole ring, and a hydrogen bonding acceptor on the carboxyl amide are essential for inhibitory effect of tau aggregation. Notably, compound 16 showed superior potency with IC50 value of 1.6 mM for inhibiting tau aggregation and reduced toxicity when tested in an N2A cell model of tau aggregation [57]. The potency of compound 16 is believed to be due to the hydrogen bonding capacity of the nitro group and to the p-stacking interactions with the indazole group as confirmed by saturation transfer difference (STD) NMR spectroscopy experiments. O
R2 R1
N H
H N
N
N
R4
S R3 15
N H
H N
N S
N H 16
NO2
60
Simon N. Haydar et al.
3.2.2 Rhodanine-based inhibitors Mandelkow’s group has also described the SAR of substituted rhodanines (2-thioxothiazolidin-4-ones) [58]. Extensive SAR studies resulted in the preparation of an interesting biphenyl rhodanine derivative 17, which displayed an IC50 of 170 nM for assembly inhibition and DC50 of 130 nM for disassembly induction. It is noteworthy that the presence of an aromatic side chain appeared necessary, supporting hydrophobic p-stacking interactions of this fragment [59]. Unfortunately, compound 17 and many other analogs showed a large discrepancy between in vitro and cell-based activity. Poor physiochemical properties were implicated as the likely cause of poor cellular results, reflecting a need for further optimization of this series. O
Ph
S
HO
N
S
O
O
17
3.2.3 Cationic thiacarbocyanine dye Thiacarbocyanine dye N744 (compound 18) has been shown to inhibit recombinant tau fibrillization in the presence of anionic surfactant aggregation inducers with an IC50 value of 0.3 mM [60]. In an effort to increase potency, a cyclic bis-thiacarbocyanine 19 was synthesized and characterized with respect to tau fibrillization inhibition by electron microscopy and ligand aggregation state by absorbance spectroscopy [61]. Data showed that the inhibitory activity of the bis-thiacarbocyanine 19 was similar to a monomeric cyanine dye, but was more potent with an IC50 of 80 nM. Data reported for these two compounds further suggest that the inhibitory activity of bis-thiacarbocyanine 19 results from multivalency. This finding might offer an new mode of interaction for design of tau aggregation inhibitors [61]. S
S Br
O
S
O
S
N
N
N
Br
N N
N
HO
OH 18
S
S 19
61
Therapeutic Agents for Neurodegenerative Diseases
3.2.4 N-Phenylamines and anthraquinones Additional small molecule tau aggregation inhibitors were reported in the literature derived from N-phenylamines (20) and anthraquinones such as daunorubicin (21) and adriamycin (22) [55,62].
NO2
O 2N
O
OH
A
B
C
D
O
O
OH O
H N N
O OH
OH
20
OH
A
B
C
O
O
OH O
NH2
21
D
O
O
O
O
HO OH
O OH
OH
NH2
22
Compared to the two compound classes discussed earlier (rhodamines and phenylthiazolylhydrazides), the N-phenylamines displayed far lower potencies in vitro and in cells, which precluded their use in in vivo models [54]. The b-hydroxyenone moiety in anthraquinone, also observed in other inhibitor classes, such as flavonoids and naphthoquinones, may play a significant role in the inhibitory potency of these chemotypes. Compounds 21 and 22 were able to inhibit the aggregation of the K19 tau construct and induced the disaggregation of preformed aggregates [55]. Substitution on the ring A in compound 21 does not appear to play a critical role of inhibitory activity. Ring D in compounds 21 and 22 bearing the sugar moiety is moderately sensitive to the substitutions on that ring, which indicates further opportunity for structural modifications to improve the inhibitory potency. Despite the observed activity of anthraquinones, it should be noted that they are known cytostatics and present a hazardous toxicological profile, which preclude them from being desired therapeutics for the chronic treatment of AD.
3.2.5 Polyphenols, phenothiazines, and porphyrins Taniguchi et al. [63] reported three classes of compounds (phenols, phenothiazines, and porphyrins) that were able to inhibit aggregation of human tau 46. Polyphenols such as compound 23 shared the SAR described in the previously section for anthraquinones with characteristic b-hydroxyenone moieties. Phenothiazines possess a tricyclic core that incorporates sulfur and nitrogen atoms on the central ring system. They possess positively charged atoms and hydrophobic
62
Simon N. Haydar et al.
aromatic groups similar to the benzothiazole inhibitors described previously. The charge-neutral phenothiazines are reported to have good blood–brain barrier permeability. The planarity and aromaticity of the central heterocyclic core appears to play a critical role of tau aggregation inhibitory activity. In particular, compound 24 (MTC, methylthioninium chloride) also known as methylene blue was reported as a potent in vivo tau aggregation inhibitor with sub-micromolar potency in cells [64,65]. Data from the phase II clinical trial with methylene blue reported a significantly lower rate of decline of cognitive functions compared with a placebo (81%, po0.0001). Although these data are preliminary and require further confirmation, it does present a promising approach for the management of AD and a potential proof of concept for the strategy of inhibiting tau aggregation.
OH HO
N O
HO
OH N
S
N
HO O
OH
23
24
Porphyrins such as compound 25 are the only organometallic compounds that bind differently than other reported tau aggregation inhibitors. The inhibitory activity depends on a central metal (iron or zinc) as phthalocyanine compound lacking the central metal has a weaker inhibitory potency [63]. It has been suggested that the coordination of the metal center with histidine on the protein–protein interface plays a significant role in tau aggregation inhibition [66]. HOOCC2H4
HOOCC2H4
N N Fe N
25
N
Therapeutic Agents for Neurodegenerative Diseases
63
4. a-SYNUCLEIN AGGREGATION IN PARKINSON’S DISEASE 4.1 Biochemistry of a-synuclein aggregation a-Synuclein is a small 140 amino acid, natively unfolded protein with little secondary or tertiary structure whose aberrant aggregation has been linked to the etiology of PD. a-Synuclein can assume either an a-helical conformation upon binding to lipids and membranes or a b-sheet conformation upon aggregation at high concentrations, elevated temperature, agitation, or in the presence of metals, simple alcohols, or detergents [67,68]. Increases in b-sheet content are associated with formation of small soluble oligomers, larger protofibrils, and the macromolecular fibrils. Dimers are the smallest oligomeric aggregates with limited b-sheet structure and have been proposed to act as seeds in the nucleation-dependent aggregation of a-synuclein [69,70]. Both oligomers and protofibrils have increased amounts of b-sheet structure compared with monomeric or lipid bound a-synuclein. The predominantly observed structure of the protofibrils is globular, although they are also able to form rod-like filaments as well as annular rings that can insert into lipid membranes enabling leakage of small molecules [67]. These annular protofibrils have been implicated in the pathogenesis of PD by the virtue that the A53T and A30P mutations either increase the propensity of a-synuclein to form annular protofibrils or stabilize them. The annular protofibrils can insert into lipid membranes and enable leakage of ions, neurotransmitters, and small dyes [67]. Although these properties make the oligomers and protofibrils attractive targets, their size and morphological continuum present a tremendous logistical hurdle for assay development and structure activity relationship analysis. Eventually, a-synuclein aggregates into insoluble fibrils that have a very high b-sheet content. It is the insoluble fibril that is deposited in Lewy bodies in the brains of patients suffering from PD, which are the defining pathological hallmark of the disease. A long-standing issue has been whether Lewy bodies are markers of a neurodegenerative process or a protective mechanism, serving as a means to sequester toxic oligomers and protofibrils [3]. Still, fibrils, as the end point of aggregation and the hallmark of pathological PD, have been targeted extensively in attempts to discover small-molecule inhibitors of a-synuclein aggregation as a treatment to slow down or halt the progression of PD. Many molecules have been shown to inhibit fibril formation (curcumin, Congo red, epigallocatechin gallate, peptide-mimetics, and non-steroidal anti-inflammatories) including the flavonoid, baicalein, and the neurotransmitter, dopamine [71–77]. More detailed studies with the latter two molecules reveal that while these compounds are inhibiting
64
Simon N. Haydar et al.
fibril formation, they are also stabilizing oligomers/protofibrils [77]. In the case of dopamine, the aggregates are much more stable than the oligomers/protofibrils produced under more ‘‘conventional’’ conditions, in that they are SDS stable. Caution should also be exercised when targeting disruption of fibril b-sheet as the tertiary structure of many proteins contains b-sheets. If inhibitors of fibril formation are to be developed as a therapeutic for PD, appropriate screens should be developed to assess their ability to stabilize soluble oligomers/protofibrils, to assess the toxicity of any stabilized species and the ability to disrupt the b-sheet structure in other proteins.
4.2 Small-molecule inhibitors of a-synuclein aggregation Conway et al. [77] have shown that catecholamines such as compound 26 can inhibit the formation of a-synuclein fibrils by stabilizing oligomeric intermediates. Li et al. [78] reported that the oxidation state of catecholamine affected the inhibitory activity. As such, dopaminochrome (27), one of the oxidation products of dopamine, was shown to be more potent at inhibiting a-synuclein fibril formation than the parent dopamine [78,79]. This also raises the possibility that the protein may be covalently modified by the dopaminochrome or dopamine under oxidative conditions. H
N O
HO
N H
O
HO 26
27
Various polyphenolic compounds such as flavonoids have also been shown to be effective inhibitors of a-synuclein aggregation [80]. One of these inhibitors is baicalein, 28, isolated from the Chinese skullcap plant (Scutellaraia baicalensis), has been shown to directly bind to a single site on a-synuclein with submicromolar affinity and, as such, inhibit formation of fibrils through the stabilization of oligomers [71,81]. It has been suggested that the quinone oxidation by-product (29) of baicalein is responsible for the observed inhibitory activity. HO
O
HO
O
O
O OH O
OH O
28
29
Therapeutic Agents for Neurodegenerative Diseases
65
Other small-molecule inhibitors of a-synuclein have been identified in the literature such as Congo red.[80]. Polyphenolic compounds such as rifampicin, curcumin, and tetracycline are capable of inhibiting both a-synuclein [79] and Ab aggregation [82] in a concentration-dependent manner with some potency (IC50o10 mM). Li et al. [83] point out that many of these polyphenolic anti-fibrillogenic compounds have antioxidant activities and readily oxidize in the presence of atmospheric oxygen to form quinones and thus may all act through oxidative modification of peptides. This modification, formation of quinone adducts or formation of Schiff-base, may act to inhibit fibril formation by constraining the peptides to a conformation not compatible with the tight orderly packing of b-sheets found in fibrils, but at the same time stabilizing soluble oligomers. Although some of these compounds have been demonstrated to protect cells against a-synuclein overexpression, in agreement with the biochemistry of a-synuclein aggregation [84], many reports either attribute the efficacy of the compounds to metal chelation and antioxidant activity or do not show the mechanism to be inhibition of protein aggregation. Furthermore, caution in pursuing these polyphenolic and catecholamine compounds as aggregation inhibitors is urged, however, as one long-standing hypothesis postulates that it is actually the soluble oligomers, not the insoluble fibrils that are neurotoxic in PD. Still, these polyphenolic/ anti-oxidant types of compounds appear to share a common mechanism: oxidation-dependent modification of a-synuclein, which inhibits fibril formation. If the issues, raised earlier, are resolved in the drug discovery process, then small-molecule inhibitors of a-synuclein aggregation could be key therapeutics for PD [82]. However, in general, it has been noted from the current literature that there are few reports and limited efforts to develop structural activity relationship around the compounds listed earlier, suggesting that drug discovery strategies to identify small-molecule inhibitors of a-synuclein are still evolving.
5. CONCLUSION In summary, the literature reviewed in this chapter imply two general assumptions regarding the inhibition mechanism of amyloid protein fibril formation by small molecules: (a) specific structural conformation is necessary for b-sheet interaction and stabilization of the inhibition– protein complex; (b) aromatic interaction between the inhibitor molecule and the aromatic residues in the amyloidogenic sequence, potential ‘‘hot spots,’’ may direct the inhibitor to the amyloidogenic core blocking
66
Simon N. Haydar et al.
the protein–protein interaction. These assumptions are highly relevant for future design of small-molecule inhibitors as therapeutic agents for the treatment of amyloid-associated diseases. In conclusion, drug development for AD and PD remains highly active, and there is a realistic expectation that new therapies will be evaluated in clinical trials in the near future.
REFERENCES [1] B. L. Plassman, K. M. Langa, G. G. Fisher, S. G. Heeringa, D. R. Weir, M. B. Ofstedal, J. R. Burke, M. D. Hurd, G. G. Potter, W. L. Rodgers, D. C. Steffens, R. J. Willis and R. B. Wallace, Neuroepidemiology, 2007, 29, 125. [2] L. M. Lau and M. M. Breteler, Lancet Neurology, 2006, 5, 525. [3] C. Haass and D. J. Selkoe, Nat. Rev. Mol. Cell. Biol., 2007, 8, 101. [4] M. R. Arkin and J. A. Wells, Nat. Rev. Drug Discov., 2004, 3, 301. [5] L. L. Blazer and R. R. Neubig, Neuropsychopharmacology, 2009, 34, 126. [6] A. A. Bogan and K. S. Thorn, J. Mol. Biol., 1998, 280, 1. [7] D. J. Selkoe, Physiol. Rev., 2001, 81, 741. [8] D. J. Selkoe, Neuron, 1991, 6, 487. [9] J. A. Hardy and G. A. Higgins, Science, 1992, 256, 184. [10] G. G. Glenner and C. W. Wong, Biochem. Biophys. Res. Commun., 1984, 120, 885. [11] C. L. Masters, G. Simms, N. A. Weinman, G. Multhaup, B. L. McDonald and K. Beyreuther, Proc. Natl. Acad. Sci. U.S.A., 1985, 82, 4245. [12] D. J. Selkoe, C. R. Abraham, M. B. Podlisny and L. K. Duffy, J. Neurochem., 1986, 46, 1820. [13] R. Katzman, N. Engl. J. Med., 1986, 314, 964. [14] R. D. Terry, E. Masliah, D. P. Salmon, N. Butters, R. DeTeresa, R. Hill, L. A. Hansen and R. Katzman, Ann. Neurol., 1991, 30, 572. [15] D. W. Dickson, H. A. Crystal, C. Bevona, W. Honer, I. Vincent and P. Davies, Neurobiol. Aging, 1995, 16, 285. [16] L. F. Lue, Y. M. Kuo, A. E. Roher, L. Brachova, Y. Shen, L. Sue, T. Beach, J. H. Kurth, R. E. Rydel and J. Rogers, Am. J. Pathol., 1999, 155, 853. [17] C. A. McLean, R. A. Cherny, F. W. Fraser, S. J. Fuller, M. J. Smith, K. Beyreuther, A. I. Bush and C. L. Masters, Ann. Neurol., 1999, 46, 860. [18] J. Wang, D. W. Dickson, J. Q. Trojanowski and V. M. Lee, Exp. Neurol., 1999, 158(2), 328. [19] M. A. Findeis, Pharmacol. Ther., 2007, 116, 266. [20] D. M. Walsh and D. J. Selkoe, J. Neurochem., 2007, 101, 1172. [21] G. M. Shankar, S. Li, T. H. Mehta, A. Garcia-Munoz, N. E. Shepardson, I. Smith, F. M. Brett, M. A. Farrell, M. J. Rowan, C. A. Lemere, C. M. Regan, D. M. Walsh, B. L. Sabatini and D. J. Selkoe, Nature Med., 2008, 14, 837. [22] J. McLaurin, R. Golomb, A. Jurewicz, J. P. Antel and P. E. Fraser, J. Biol. Chem., 2000, 275, 18495. [23] J. McLaurin, M. E. Kierstead, M. E. Brown, C. A. Hawkes, M. H. L. Lambermon, A. L. Phinney, A. A. Darabie, J. E. Cousins, J. E. French, M. F. Lan, F. Chen, S. S. N. Wong, H. T. J. Mount, P. E. Fraser, D. Westaway and P. George-Hyslop, Nat. Med., 2006, 12, 801.
Therapeutic Agents for Neurodegenerative Diseases
67
[24] Y. Sun, G. Zhang, C. A. Hawkes, J. E. Shaw, J. McLaurin and M. Nitz, Bioorg. Med. Chem., 2008, 16, 7177. [25] D. H. Hua, X. Huang, M. Tamura, Y. Chen, M. Woltkamp, L.-W. Jin, E. M. Perchellet, J.-P. Perchellet, P. K. Chiang, I. Namatame and H. Tomoda, Tetrahedron, 2003, 59, 4795. [26] H.-S. Hong, S. Rana, L. Barrigan, A. Shi, Y. Zhang, F. Zhou, L.-W. Jin and D. H. Hua, J. Neurochem., 2009, 108, 1097. [27] S. Rana, H.-S. Hong, L. Barrigan, L.-W. Jin and D. H. Hua, Bioorg. Med. Chem. Lett., 2009, 19, 670. [28] A. Aggeli, M. Bell, N. Boden, J. N. Keen, P. F. Knowles, T. C. McLeish, M. Pitkeathly and S. E. Radford, Nature, 1997, 386, 259. [29] E. Gazit, FASEB J., 2002, 16, 77. [30] T. Cohen, A. Frydman-Marom, M. Rechter and E. Gazit, Biochemistry, 2006, 45, 4727. [31] C. E. Giacomelli and W. Norde, Biomacromolecules, 2003, 4, 1719. [32] E. P. Vieira, H. Hermel and H. Mohwald, Biochim. Biophys. Acta, Proteins Proteomics, 2003, 1645, 6. [33] M. Torok, M. Abid, C. Mhadgut Shilpa and B. Torok, Biochemistry, 2006, 45, 5377. [34] F. Gervais, J. Paquette, C. Morissette, P. Krzywkowski, M. Yu, M. Azzi, D. Lacombe, X. Kong, A. Aman, J. Laurin, W. A. Szarek and P. Tremblay, Neurobiol. Aging, 2007, 28, 537. [35] A. Cavalli, M. L. Bolognesi, S. Capsoni, V. Andrisano, M. Bartolini, E. Margotti, A. Cattaneo, M. Recanatini and C. Melchiorre, Angew. Chem., Int. Ed., 2007, 46, 3689. [36] A. Cavalli, M. L. Bolognesi, A. Minarini, M. Rosini, V. Tumiatti, M. Recanatini and C. Melchiorre, J. Med. Chem., 2008, 51, 347. [37] C. A. Hawkes, V. Ng and J. McLaurin, Drug Dev. Res., 2009, 70, 111. [38] A. Andreadis, W. M. Brown and K. S. Kosik, Biochemistry, 1992, 31, 10626. [39] P. W. Baas and L. Qiang, Trends Cell Biol., 2005, 15, 183. [40] C. Ballatore, V. M. Lee and J. Q. Trojanowski, Nat. Rev. Neurosci., 2007, 8, 663. [41] J. Leger, M. Kempf, G. Lee and R. Brandt, J. Biol. Chem., 1997, 272, 8441. [42] J. Eidenmuller, T. Fath, T. Maas, M. Pool, E. Sontag and R. Brandt, Biochem. J., 2001, 357(Pt 3), 59. [43] T. Fath, J. Eidenmuller and R. Brandt, J. Neurosci., 2002, 22, 9733. [44] A. Abraha, N. Ghoshal, T. C. Gamblin, V. Cryns, R. W. Berry, J. Kuret and L. I. Binder, J. Cell Sci., 2000, 113(Pt 21), 3737. [45] R. W. Berry, A. Abraha, S. Lagalwar, N. LaPointe, T. C. Gamblin, V. L. Cryns and L. I. Binder, Biochemistry, 2003, 42, 8325. [46] J. Kuret, E. E. Congdon, G. Li, H. Yin, X. Yu and Q. Zhong, Microsc. Res. Tech., 2005, 67, 141. [47] I. Khlistunova, J. Biernat, Y. Wang, M. Pickhardt, M. Bergen, Z. Gazova, E. Mandelkow and E. M. Mandelkow, J. Biol. Chem., 2006, 281, 1205. [48] O. Katsuse, W. L. Lin, J. Lewis, M. L. Hutton and D. W. Dickson, Neurosci. Lett., 2006, 409, 95. [49] W. Chun and G. V. Johnson, J. Biol. Chem., 2007, 282, 23410. [50] S. Maeda, N. Sahara, Y. Saito, S. Murayama, A. Ikai and A. Takashima, Neurosci. Res., 2006, 54, 197. [51] C. W. Wittmann, M. F. Wszolek, J. M. Shulman, P. M. Salvaterra, J. Lewis, M. Hutton and M. B. Feany, Science, 2001, 293, 711.
68
Simon N. Haydar et al.
[52] T. L. Spires, J. D. Orne, K. SantaCruz, R. Pitstick, G. A. Carlson, K. H. Ashe and B. T. Hyman, Am. J. Pathol., 2006, 168, 1598. [53] M. P. Mazanetz and P. M. Fischer, Nat. Rev. Drug Disc., 2007, 6, 464. [54] B. Bulic, M. Pickhardt, B. Schmidt, E.-M. Mandelkow, H. Waldmann and E. Mandelkow, Angew. Chem., Int. Ed., 2009, 48, 1740. [55] M. Pickhardt, Z. Gazova, M. Von Bergen, I. Khlistunova, Y. Wang, A. Hascher, E.-M. Mandelkow, J. Biernat and E. Mandelkow, J. Biol. Chem., 2005, 280, 3628. [56] G. Larbig, M. Pickhardt, D. G. Lloyd, B. Schmidt and E. Mandelkow, Curr. Alzheimer Res., 2007, 4, 315. [57] M. Pickhardt, G. Larbig, I. Khlistunova, A. Coksezen, B. Meyer, E.-M. Mandelkow, B. Schmidt and E. Mandelkow, Biochemistry, 2007, 46, 10016. [58] B. Bulic, M. Pickhardt, I. Khlistunova, J. Biernat, E.-M. Mandelkow, E. Mandelkow and H. Waldmann, Angew. Chem. Int. Ed. Engl., 2007, 46, 9215. [59] M. L. Waters, Curr. Opin. Chem. Biol., 2002, 6, 736. [60] E. E. Congdon, M. Necula, R. D. Blackstone and J. Kuret, Arch. Biochem. Biophys., 2007, 465, 127. [61] N. S. Honson, J. R. Jensen, M. V. Darby and J. Kuret, Biochem. Biophys. Res. Commun., 2007, 363, 229. [62] M. Pickhardt, J. Biernat, I. Khlistunova, Y. P. Wang, Z. Gazova, E. M. Mandelkow and E. Mandelkow, Curr. Alzheimer Res., 2007, 4, 397. [63] S. Taniguchi, N. Suzuki, M. Masuda, S.-I. Hisanaga, T. Iwatsubo, M. Goedert and M. Hasegawa, J. Biol. Chem., 2005, 280, 7614. [64] C. M. Wischik, P. C. Edwards, R. Y. Lai, M. Roth and C. R. Harrington, Proc. Natl. Acad. Sci. U.S.A., 1996, 93, 11213. [65] C. M. Wischik, J. E. Rickard, C. R. Harrington, D. Horsley, J. M. D. Storey, C. Marshall and J. P. Sinclair, Patent Application WO2007110630, 2007. [66] D. Howlett, P. Cutler, S. Heales and P. Camilleri, FEBS Lett., 1997, 417, 249. [67] M. J. Volles and P. T. Lansbury Jr., Biochemistry, 2003, 42, 7871. [68] V. N. Uversky, J. Neurochem., 2007, 103, 17. [69] S. Krishnan, E. Y. Chi, S. J. Wood, B. S. Kendrick, C. Li, W. Garzon-Rodriguez, J. Wypych, T. W. Randolph, L. O. Narhi, A. L. Biere, M. Citron and J. F. Carpenter, Biochemistry, 2003, 42, 829. [70] V. N. Uversky, J. Biomol. Struct. Dyn., 2003, 21, 211. [71] M. Zhu, S. Rajamani, J. Kaylor, S. Han, F. Zhou and A. L. Fink, J. Biol. Chem., 2004, 279, 26846. [72] S. Mandel, G. Maor and M. B. Youdim, J. Mol. Neurosci., 2004, 24, 401. [73] M. Hirohata, K. Ono, A. Morinaga and M. Yamada, Neuropharmacol, 2008, 54, 620. [74] J. N. Rao, V. Dua and T. S. Ulmer, Biochemistry, 2008, 47, 4751. [75] A. M. Bodles, O. M. El-Agnaf, B. Greer, D. J. Guthrie and G. B. Irvine, Neurosci. Lett., 2004, 359, 89. [76] K. Ono and M. Yamada, J. Neurochem., 2006, 97, 105. [77] K. A. Conway, J. C. Rochet, R. M. Bieganski and P. T. Lansbury Jr., Science, 2001, 294, 1346. [78] J. Li, M. Zhu, A. B. Manning-Bog, D. A. Monte and A. L. Fink, FASEB J., 2004, 18, 962. [79] J. Li, M. Zhu, S. Rajamani, V. N. Uversky and A. L. Fink, Chem. Biol., 2004, 11, 1513. [80] Y. Porat, A. Abramowitz and E. Gazit, Chem. Biol. Drug Des., 2006, 67, 27. [81] D. P. Hong, A. L. Fink and V. N. Uversky, J. Mol. Biol., 2008, 383, 214.
Therapeutic Agents for Neurodegenerative Diseases
69
[82] T. Tomiyama, A. Shoji, K.-i. Kataoka, Y. Suwa, S. Asano, H. Kaneko and N. Endo, J. Biol. Chem., 1996, 271, 6839. [83] H. T. Li, D. H. Lin, X. Y. Luo, F. Zhang, L. N. Ji, H. N. Du, G. Q. Song, J. Hu, J. W. Zhou and H. Y. Hu, FEBS J., 2005, 272, 3661. [84] C. L. Kragh, L. B. Lund, F. Febbraro, H. D. Hansen, W. P. Gai, O. El-Agnaf, C. RichterLandsberg and P. H. Jensen, J. Biol. Chem., 2009, 284, 10211.
CHAPT ER
4 Case History: ChantixTM/ChampixTM (Varenicline Tartrate), a Nicotinic Acetylcholine Receptor Partial Agonist as a Smoking Cessation Aid Jotham W. Coe, Hans Rollema and Brian T. O’Neill
Contents
1. Introduction 2. Partial Agonists at Nicotinic ACh Receptors 3. The Search for Partial Agonists: Cytisine as a Key Starting Point 4. Semi-Synthetic Analogs of Cytisine 5. Cytisine Synthesis and Early Template Expansion 6. Discovery of the Bicyclic Benzazepine Core 7. Fused Bicyclic Benzazepines 8. In vivo Efficacy of Partial Agonists 9. Properties of Varenicline 9.1 Pharmacology 9.2 Absorption, distribution, metabolism, excretion (ADME) 10. Clinical Studies 11. Conclusions References
71 74 77 80 84 87 91 93 95 95 97 98 98 99
Pfizer Global Research and Development, Groton Laboratories, Eastern Point Road, Groton, CT 06340, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04404-2
r 2009 Elsevier Inc. All rights reserved.
71
72
JOTHAM W. COE et al.
1. INTRODUCTION Serious attention to the public health impact of smoking emerged in the United Kingdom in the late 1950s after a prospective study, begun in 1951, focused on the frequency of lung cancer among doctors with known smoking habits [1]. Early data demonstrated a link between smoking tobacco and lung cancer, which was born-out over the 50-year span of the study and grew to include many other causes of death. The overall conclusions were that tobacco use will kill at least half of all smokers and that smokers lose on average 10 years of life compared with non-smokers. Smokers who were able to quit smoking when they were between 30 and 40 years of age decreased their risk of premature mortality to that of a non-smoker [2]. In the United States, the Surgeon General’s 1964 report concluded that ‘‘lung cancer and chronic bronchitis are causally related to cigarette smoking’’ [3]. The report also noted that there was suggestive evidence, if not definite proof, for a causative role of smoking in other illnesses. Surprisingly, it was not until the 11th Surgeon General’s report in 1979 that smoking was defined as a ‘‘nicotine addiction’’ [4]. In 2007, the U.S. Senate proposed removing the pejorative term ‘‘abuse’’ from the name of the National Institute of Drug Abuse (NIDA) and the National Institute on Alcohol Abuse and Alcoholism (NIAAA) replacing it with the words ‘‘diseases’’ and ‘‘addiction’’ [5]. Evidence that profound neurological changes result from smoking has led experts to characterize nicotine addiction as a disease, a clear shift in perception from the view that this behavior was a habit one simply stops. Awareness of smoking’s health toll has been beneficial in the United States; cigarette smoking rates in U.S. men have dropped steadily in the past 50 years, from more than 42% to the current rate of B21%. Despite this trend, cigarette smoking, still the leading cause of preventable illness and mortality in the United States, contributes to the death of 443,000 people each year. More than 5 million people die annually in the developed world from smoking-related illness, a statistic on course to double by 2030 [6]. As health care costs have risen, worldwide government awareness of the problem has intensified. The World Health Organization (WHO) effort to define targets for smoking rate reductions highlights this international shift [7]. Still, smoking remains a ritual ingrained in the societal fabric throughout the developed world. Recognition that nicotine addiction is initiated by a pharmacological substance that profoundly impacts the physiology of the patient was a large step forward. This tenet stimulated research toward therapies that could address nicotine addiction by acting directly at the receptors that mediate the addictive effects. In the late 1980s, the only available treatment was nicotine replacement therapy (NRT), introduced in the late 1970s, which
Case History: Chantixt/Champixt (Varenicline Tartrate)
73
increased quit rates in clinical trials [8]. Later, in 1997, after our entry into smoking cessation research, bupropion (Zybans), first marketed as the anti-depressant (Wellbutrins), was found to be efficacious for smoking cessation [9]. Behavioral therapy and counseling have been offered in various ‘‘quit smoking’’ settings, and although common, they have been generally underutilized, their success has been limited by commitment and finances of the smoker [10]. All of these treatments approximately doubled the chance of continuous abstinence compared with placebo controls, highlighting the need for more efficacious therapies. NH N N
varenicline (1)
Varenicline (1) is the first medicine targeting nicotinic acetylcholine receptors (nAChRs), other than nicotine (2) as replacement therapy (NRT), approved as an aid to smoking cessation by both the Food and Drug Administration (FDA) and the European Medicines Agency (EMEA) in 2006. The effort to discover and develop varenicline began at Pfizer in 1993, with the objective from the outset to find a partial agonist to treat nicotine dependence. A partial agonist was theorized to provide significant but reduced nicotinic reward that an abstinent smoker craves while simultaneously blocking the reward from relapse smoking. Prior pharmacological approaches to treat addiction were limited to opioid treatment for heroin addiction. Our nAChR partial agonist program was seeded in the growing appreciation of the effectiveness of buprenorphine as an aid to opioid addiction [11]. Buprenorphine, available today as Buprenexs (Suboxones, Subutexs), a semisynthetic analgesic, was shown in 1976 to be a m-opioid receptor partial agonist and was studied in the 1970s and 1980s as a treatment for heroin (and cocaine) addiction. Compared to methadone, a m-opioid receptor full agonist and naloxone, a m-opioid receptor competitive antagonist, buprenorphine is more effective, produces less physical dependence, has a reduced risk of respiratory suppression resulting from overdose, and fewer side effects. Rose and Levin [12] proposed the concept of a combined agonist and antagonist for smoking cessation in a series of publications in 1989–1992. They originally suggested that NRT (nicotine (2) as the nAChR agonist) when combined with mecamylamine (3) (nAChR antagonist) would theoretically improve quit rates. The effects of this combination were studied in clinical trials that demonstrated the dual agent approach provided higher quit rates than NRT alone. This was an attractive concept,
74
JOTHAM W. COE et al.
but suffered from the challenges of co-administration of two agents with different pharmacokinetic (PK) profiles, and the unfavorable side effect profile of mecamylamine (3). The dual agonist/antagonist approach could be better achieved with a single molecule, a partial agonist. In 1993, it was agreed at Pfizer to pursue a partial agonist as a novel smoking cessation agent, in part because one of the key challenges, how to measure partial agonist responses in vitro, was in principle achievable using cloned nAChRs for in vitro electrophysiological studies of ligand effects [13]. However, at that time, it was still much less evident how to identify or design partial agonist chemical structures. Herein we describe our efforts as they unfolded over time from the beginning of this project in 1993. Interestingly, the considerable synthetic and pharmacologic effort in the opioid field in the 1970s played a pivotal role in the discovery of varenicline.
2. PARTIAL AGONISTS AT NICOTINIC ACh RECEPTORS ‘‘Partial agonists are ligands that produce a smaller than maximal response compared with the natural ligand at full receptor occupation’’ [14]. Acetylcholine (4) is the natural full agonist at two known receptor subclasses important for neurotransmission: the muscarinic (mAChRs) and nicotinic (nAChRs) receptors, which are distributed throughout the central nervous system (CNS) and the periphery. The nicotinic receptor subtypes are ligand-gated ion channels that mediate fast synaptic transmission upon acetylcholine binding and also mediate the full agonist effect of nicotine. These ion channels are formed from pentameric assemblies of subunits creating a central aqueous pore. There are 12 neuronal nAChR subunits divided into a and b subtypes (a2a10, b2b4) differentiated by the presence or absence of a key ‘‘cys-cys loop’’ near the acetylcholine-binding site respectively. The known combinations of subunits form a broad range of nAChR subtypes with significantly different characteristics of ligand pharmacology (Figure 1) and cation permeability [15]. H N H N
N HN
N
N
O O
nicotine, 2
Figure 1
O
mecamylamine, 3
Nicotinic ligands.
acetylcholine, 4
cytisine, 5
Case History: Chantixt/Champixt (Varenicline Tartrate)
N
O O
N
acetylcholine, 4
A • O N
O O
5.9 Å
HN
75
• C • B
cation center (-)-cytisine, 5 H-bond acceptor
polarizing group
Figure 2 Nicotinic pharmacophore (Beers and Reich [17]).
Nicotine (2) binds tightly to the a4b2 heteromeric nAChR, which is the most abundant subtype in the CNS [16], and it acts as a full agonist at this nAChR subtype. The pharmacophore and binding requirements of the nAChR site were first explored and described by Beers and Reich in 1970 [17]. Their model is still useful today as it identifies a spatial relationship for key interactions within the nicotinic-binding site (Figure 2) and differentiates those interactions from the mAChrR-binding site (not shown). The defined distance between the cation/amine functionality and the hydrogen bond acceptor region has been verified with dozens of known receptor ligands, including nicotine (2), acetylcholine (4), and cytisine (5), but the connection between structural and functional activity has been less well understood. However, the level of polarization of the hydrogen bond and the electron density of the polarizable group make important but not well-defined contributions to functional activity. The complex functional characteristics of the receptor, as defined by Changeux [18], are a consequence of at least three distinguishable states: activated, desensitized, and inactivated during a response to drug binding (Figure 3) [19]. The functional outcome due to ligand binding over the dynamic course of all three receptor states is not well defined: it is likely that the scope of behavioral responses can be linked to the dynamic relationship among multiple nicotinic states whether activated or desensitized. Receptor binding is measured at the high affinity desensitized state, whereas functional activity is of limited duration at high concentrations during channel opening and ion flow in the activated state [20]. The 1980–1990s witnessed the development of in vitro functional assays of nicotinic receptors in Xenopus oocytes, which made it possible to assess the inherent agonist activity of compounds, as well as their antagonism of nicotine [13]. Despite this, we felt that in vivo models of behavior or specific biochemical endpoints could give us a more meaningful picture of the consequences of ligand binding.
76
JOTHAM W. COE et al.
Figure 3 Schematic representation of the transitions among the functional states of nAChRs, indicating the effects of agonists (open squares) and antagonists (black squares) on the equilibrium between the resting, active, and desensitized state of the nAChR (J. Med. Chem., 2005, 48, 4705, reprinted with permission).
At the outset of our program, stable human cell lines expressing nicotinic receptor subtypes were not readily available to screen our sample collection nor were high throughput functional screens (such as Fluorescent Imaging Plate Reader (FLIPR) platforms). The most informative biological tools available were animal models of the rewarding effects of nicotine [21]. These consisted of quantitative measurements of effects on dopamine (DA) release in brain or in brain slices, and nicotine selfadministration and drug discrimination in rats, mirroring subjective effects experienced by smokers. Nicotine discrimination was pioneered by Morrison and Stephenson [22] in an experiment that demonstrated that rats could detect and respond to nicotine’s presence. The exact receptor that mediated this effect was not known at the time; however, discrimination could be blocked by a central nicotinic antagonist such as mecamylamine but not by chlorisondamine, a peripheral antagonist. Later studies by Romano [23], Rosecrans and coworkers [24], and Stolerman and coworkers [25] identified central nAChR stimulation as the likely source of the nicotine cue rather than mAChRs. Self-administration of nicotine in rats had been achieved by Corrigall and Coen [26] mimicking many of the behavioral responses displayed by humans. As with the discrimination model, the response was dose-dependently reduced by pre-treatment with either nicotine or mecamylamine but not with hexamethonium (10), a peripheral nicotinic antagonist. The dependence-producing effects of nicotine are believed to be mediated through its action as a full agonist at a4b2 nAChRs [27,28]. Activation of a4b2 receptors in the ventral tegmental area by nicotine
Case History: Chantixt/Champixt (Varenicline Tartrate)
77
increases the release of DA in the nucleus accumbens and prefrontal cortex [29], an effect shared by most substances of abuse, although each through distinct neurochemical pathways. We measured the change in DA metabolites in response to activation by partial agonist ligands and after co-administration with nicotine. We sought compounds that would provide a reduced reward relative to that of nicotine while simultaneously blocking the full effect of nicotine itself. These models were assembled for our program and validated with known nicotinic receptor ligands.
3. THE SEARCH FOR PARTIAL AGONISTS: CYTISINE AS A KEY STARTING POINT Early in our drug discovery program, we sought information on the level of partial agonist efficacy that would be effective for smoking cessation and any available chemical tools to distinguish a partial agonist from the effects of the full agonist nicotine. We required a substance that would displace nicotine from its neuronal binding site and reduce the subjective experience of nicotine administration by simultaneously attenuating the dopaminergic response to nicotine. Evidence for partial agonism would include lower efficacy than nicotine and antagonism of nicotine’s effects in vitro and in vivo. The de novo design of a single agent nAChR partial agonist was and still is hampered by the lack of a clear understanding of structural requirements needed for the appropriate functionality and selectivity. Fortunately, nature’s portfolio in the nicotinic area is substantial, and we directed our focus there as a source of medicinal leads, rather than to an artificial compound sample collection [30]. Many of these natural structural classes are presumed to posses a favorable pharmacokinetic profile, but CNS penetration and potency remained the key issues to be resolved. We first profiled nicotinic agonists with lower functional efficacy than nicotine (o70%) such as anabaseine (6), GTS-21 (7) [31], anabasine (8) [32], lobeline (9), and cytisine (5) (Figure 4) in nicotinic assays. These included drug discrimination and self-administration behavioral models and ex vivo biochemical endpoints such as DA release. Early debate in the team focused on the perceived advantages of anabasine and derivatives, which had been shown to display partial agonist activity in rat neuronal a4b2 nAChRs expressed in Xenopus oocytes (EC50 ¼ 30 mM vs ACh EC50 ¼ 2 mM), but full agonist activity at muscle receptors [33]. Reports by Stolerman and coworkers [34] had indicated that the nicotinic agent cytisine (5) was 5-fold more potent than nicotine (2) in binding to CNS receptors but 10% less potent in producing the nicotine cue. The latter response was determined to be centrally mediated since mecamylamine (3, Figure 1) blocked the effect, but the non-CNS
78
JOTHAM W. COE et al.
H N
OMe
MeO
OH N
N
N
HN
N N
Anabaseine, 6 Anabasine, 8
O
O
GTS-21, 7
cytisine, 5
Lobeline, 9
N N Hexamethonium, 10
Figure 4
Early nAChR lead ligands and the peripheral antagonist hexamethonium.
penetrant antagonist hexamethonium (10) did not. Anabasine (8), which was 30- to 60-fold less potent than cytisine in discrimination assays, as well as cytisine, appeared to be reasonable partial agonist templates. They both produced a maximal B60% response to a nicotine cue at their highest tolerated dose without reducing the control response rate. An in vivo nicotine-like effect was considered crucial as the drug must be ‘‘experienced’’ as nicotine-like to the smoker, and so the weak agonist responses of anabasine and cytisine pointed to platforms from which a partial agonist therapeutic could be built. Lobeline (9), despite displacing [3H]-nicotine in cells and achieving high brain penetration, did not produce nicotinic-like behavioral effects, possibly due to PK issues such as free fraction concentrations below the IC50, and so we moved away from this lead. In vivo, anabasine and cytisine were reported initially to be full agonists in DA release in the striatum; however, our work and that reported later by Sibia confirmed partial agonist activity in the striatum and on DA turnover (DATO) in rat nucleus accumbens [35]. Although anabasine (8) was shown to have partial agonist activity, its weak effect in drug discrimination models required doses that were close to those that impaired control responding for food, suggesting a limited therapeutic index. We were also concerned by anabasine’s full agonist profile at the muscle receptor. Anabaseine (6) possessed less attractive structural features such as potentially unstable imine functionality and was not pursued further. (–)-Cytisine (5), a nicotinic agent isolated from natural sources more than 100 years ago [36], had garnered little attention from a medicinal chemistry standpoint despite its use as a natural smoking cessation aid in Eastern Europe [37]. Cytisine’s unusual properties as a high affinity, but low efficacy, ligand for the neuronal nAChR may have restricted interest in this compound as a therapeutic agent [38]. We hypothesized that cytisine’s rigid structure and polarizable pyridone ring could potentially
Case History: Chantixt/Champixt (Varenicline Tartrate)
79
contribute to its partial agonist profile at the receptor level, and we targeted derivatives of this compound as a starting point. At about the same time, Papke and Heinemann [39] demonstrated that cytisine behaves as a partial agonist relative to acetylcholine at a4b2 nAChRs in vitro with an EC50 of 9 mM. In the discussion that follows, we describe structure–activity relationships (SARs) based on cytisine and analogs that were aimed to maintain or improve its partial agonist activity in vitro and to substantially improve its in vivo pharmacokinetic profile, as part of our efforts to discover novel partial agonist platforms. Commercial supplies of (–)-cytisine were limited and expensive, but we made the most of small quantities while pursuing local sources of the natural product. Commercial material had been derived from extracts of natural plant sources and we engaged our suppliers to locate a source of bulk material to support analog studies. We were not able to readily obtain the natural product from Sopharma, the Bulgarian firm that sells cytisine under the trade name Tabex. Material was eventually secured by working through the Austin Chemical Co., USA, and the Specs-Biospecs Company in the Netherlands. They contracted with Chinese and Russian farms that could produce plant material from Thermopsis lupinoides (L Link), a legume species from which (–)-cytisine would be readily extractable. Precise timing was critical to intercept the bean pods at peak production of the natural product in mid-July before the pod fell from the plants. From these sources, we obtained 2 kg of W98% pure (–)-cytisine; an early batch was procured for $30,000/kg. This legume species was outside of the normal medicinal herb target population, but reasonable cost estimates put the eventual price of production quantities of (–)-cytisine at $150/kg. We were also able to estimate that quantities of 13–18 kg of purified cytisine could be produced per acre. While we were awaiting harvest, we initiated a parallel collaboration with Smith College Botanical Garden in Northampton Massachusetts. Cytisine was known to be a constituent of the Maackia amurensis (M. amurensis) species and material from tree pruning was made available by the college. Tree material, including bark, leaves, roots, and stems, was milled and extracted with methanol for up to 20 h followed by concentration and chromatographic separation from which we obtained 5.5 g of W95% pure (–)-cytisine from 64 kg of tree material. These small quantities allowed us to prepare several close analogs that we hoped would provide access to a library of molecules with good brain penetration and hopefully better potency. Cytisine was profiled [40] in neurochemical models, ex-vivo DATO in rat nucleus accumbens and [3H]-DA release in rat striatum. We found that cytisine functions as a partial agonist, producing lower than the maximal nicotine-induced increases in [3H]-DA release in striatal slices and in DATO. Cytisine also antagonized the effect of nicotine in these
80
JOTHAM W. COE et al.
assays, displaying the dual effects expected of a partial agonist. At 1 mg/ kg subcutaneous administration (s.c.), nicotine maximally increased DATO in nucleus accumbens to 170% of control (ED50 ¼ 0.156 mg/kg s.c.), whereas the maximal cytisine-induced increase was 130% of controls (ED50 ¼ 1 mg/kg). In addition, cytisine reversed the effect of 1 mg/kg s.c. nicotine on DATO in rat nucleus accumbens with an ID50 of 1.77 mg/kg. These results met our criteria for a partial agonist profile, except for the rather low in vivo potency, and validated our thoughts around a suitable structural framework for developing a partial agonist portfolio. The high cytisine dose required for inhibiting nicotineinduced DATO was eventually linked to low central exposure. Cytisine has low molecular weight and low polar surface area with moderate basicity, but it is hydrophilic, as shown by its partition coefficients in octanol/water (log P ¼ 0.01; log D7.4 ¼ 0.2), which may partially explain its limited brain penetration. The measured brain-to-plasma ratio of cytisine, which does not bind to plasma or brain proteins, was 0.1 and the cerebral spinal fluid (CSF)-to-free-plasma ratio was 0.27 resulting in low observed brain concentrations, ranging from 10 to 50 ng/g over the course of 6 h after a 1 mg/kg p.o. dose.
4. SEMI-SYNTHETIC ANALOGS OF CYTISINE Semi-synthetic analogs of cytisine were targeted once supplies of the natural product were secured [40]. Early on we established significant decreases in nAChR binding activity with N-group substitution (e.g., Me, Bn, Allyl, COR, SO2CF3), a result later confirmed in related templates. Medicinal chemistry enabling synthetic studies targeted aromatic ring modifications as well as total synthesis options, which would require the synthesis of novel templates. Although cytisine was the foundation for further nicotinic partial agonists, we were aware that improvements in cytisine’s profile might not succeed or be thwarted by issues of toxicity, CNS penetration, or other PK issues. Furthermore, novel rigid templates that offered receptor interactions complementary to cytisine would diversify the approach into novel matter. Cytisine binds with high affinity at the a4b2 nAChR with an in vitro agonist EC50 of 9 mM, but its poor brain penetration contributes to its reduced in vivo potency. Therefore, sufficient in vivo efficacy would either require a bold potency improvement within the cytisine structural class, or a physical property improvement that allowed CNS penetration from a novel template outside of the cytisine structural domain. Since we did not want to overplay the benefits of a particular series such that high doses would be required to achieve efficacy, it seemed a better strategy to look to novel series based on the cytisine framework for partial agonist activity to
Case History: Chantixt/Champixt (Varenicline Tartrate)
81
maintain and improve the profile we had from nature. We planned to move in parallel from cytisine to novel rigid templates that possessed similar receptor interactions, while monitoring partial agonist properties using electrophysiology, neurochemistry, drug discrimination, and selfadministration models. Early cytisine derivatives were intended primarily as intermediates for further analog synthesis. (–)-Cytisine (5), obtained from natural sources, was protected as the N-t-Boc derivative under standard conditions. Halogenation gave access to novel cytisine intermediates, with bromination of N-protected cytisine affording a mixture of monobromo and dibromo adducts. Although mono-halogenation of 11 using N-bromosuccinimide (NBS) or iodine monochloride (ICl) suffered from poor selectivity, reaction of 11 with N-chlorosuccinimde (NCS) gave a 1:9 mixture of regioisomeric compounds 15 and 18, respectively. The opposite regioisomer predominated when trifluoroacetamide 12 was iodinated with ICl in buffered methanol (Scheme 1, Table 1). Binding affinities of our analogs, including synthetic intermediates such as the halogenated cytisine derivatives, were routinely evaluated in nAChR-binding screens. We were gratified to see a large potency shift in moving from cytisine to 3-bromocytisine 14. This derivative was more potent than either the 5-bromo-derivative 17 or 3,5-dibromocytisine with a B100-fold shift in agonist efficacy (EC50 ¼ 95 nM compared to cytisine EC50 ¼ 9 mM). Thus, with a small change in molecular weight, the inherent potency of cytisine could be improved. Comparison of physical properties suggested that 3-bromocytisine’s enhanced potency could be partially ascribed to increased lipophilicity (3-bromocytisine: log P ¼ 0.99, log D ¼ 0.35 vs cytisine: log P ¼ 0.01, log D ¼ 0.2), without decreasing either ligand efficiency (the binding energy per atom) [41], which remained at nearly 0.90, or lipophilic lipid efficiency [42] (LLE ¼ pIC50log P) which is B12. These high values can be partially attributed to the small volume of the nAChR binding pocket. We also postulated that the electronic effect of halogens either improved the X O
O
O N
halogenate
N
N
+
deprotect N P 11. P = t-Boc 12. P = CF3CO−
Scheme 1
N H 13. X = I 14. X = Br 15. X = Cl
Halogenated cytisine derivatives.
N H 16. Y = I 17. Y = Br 18. Y = Cl
Y
82
JOTHAM W. COE et al.
Table 1
Halogenation of Cytisine Derivatives.
Condition
Protection
Product
Ratio
ICl CaCO3 ICl CaCO3 NBS CH2Cl2 NCS CH2Cl2
t-Boc TFA t-Boc t-Boc
13/16 13/16 14/17 15/18
1:1 9:1 3:2 1:9
H N
H N
H N 3-Br-cyt
(-)-cytisine
O 5
5
N C 3
Figure 5
13
C-NMR 163.6 ppm
H N
O 14
N C
3-Cl-cyt
13
C-NMR 160.1 ppm
Br
O 15
N C
N
13
C-NMR 159.0 ppm
Cl
O 19
X
Halide effects on N-C(O) polarization by 13C-NMR.
hydrogen bond acceptor properties of the pyridone carbonyl or favorably contributed to the aromatic polarizability by increasing the double-bond character of the pyridine C–N bond, as depicted in pyridinium oxide 19 in Figure 5. Pharmacokinetic measurements indicated that the increase in lipophilicity of 3-bromocytisine 14 improved CNS penetration (B/P ¼ 1.4) with a symmetric CSF/free plasma ratio of 1 and virtually no protein binding (Cytisine (5), B/P ¼ 0.1 and CSF/free plasma ratio 0.27). Parallel in vivo studies demonstrated that 3-bromocytisine 14 was a potent partial agonist in DATO with a B70-fold shift in agonist activity (ED50 ¼ 0.032 mg/kg) relative to cytisine and potent antagonist activity (ID50 ¼ 0.179 mg/kg). Compared to cytisine, 3-bromocytisine fully substituted for nicotine in a drug discrimination model showing a B70-fold leftward shift of the dose–response curve. Importantly, 3bromocytisine 14 also potently inhibited nicotine self-administration in rats with an ED50 of 0.05 mg/kg. Close in analogs of cytisine demonstrated additional examples of potency enhancement but none exceeded that of 3-bromocytisine 14. Synthetic chemistry efforts gave access to multiple derivatives beginning from 14 as shown in Scheme 2. The 3-chlorocytisine derivative 15 was also a potent partial agonist as were the 3-cyano and 3-methyl derivatives, 23 and 26, respectively. It has been theorized that electronic effects play a crucial and complementary role in binding and functional activity at the nicotinic receptor, leading to the proposal of a three-point nicotinic pharmacophore
83
Case History: Chantixt/Champixt (Varenicline Tartrate)
O Br N
N
O
O
N
N
HN
N
O HN
N
t-BOC
t-BOC
(-)-cytisine (5)
11
Boc-14
20 Acylation
plus regioisomer
Me
B(OH)2 N
N
O N
CN
N t-BOC N
21
N
O
O
26 Stille
25 Buchwald Hartwig CO2Me
23 Cyanation
X O N
HN
O
HN
22 Suzuki
Scheme 2
O
HN
HN
HN
N
O
24 Carbonylation
Cytisine analog semi-synthesis through Pd catalysis.
providing two contacts with the protein and one locus of polarizability [17]. We therefore explored isomeric changes that maintain the physical separation between the carbonyl and basic amine within a rigid framework directly related to cytisine. Nitrogen was repositioned across the ring to evaluate the effect of this structural change on biological activity (Figure 6) [43]. The isomeric cytisine analog, racemic N-methyl isocytisine (27), displayed a B100-fold drop in binding potency and was a functional antagonist. The alkyl substituent on nitrogen was critical for maintaining any nicotinic binding activity in this series. Although isosteric with H N
O
H N
N
O
27
N
H N
H N
N
N
O
O
28 isocytisine
(-)-3-methylcytisine, 26
Figure 6 Methyl-cytisine/isocytisine resonance forms.
84
JOTHAM W. COE et al.
3-methylcytisine (26), the subtle structural change and corresponding shift in efficacy point to a delicate balance between electronic structure and functional activity. Perhaps, building positive charge on nitrogen (the oxo-pyridinium tautomeric form, 28) in that region of the binding pocket is detrimental to the binding and functional activity, despite maintaining carbonyl H-bond acceptor capacity presumed to be similar to cytisine. The unsubstituted isocytisine pyridone (N-H) derivative was inactive, likely due to repulsion of the hydrogen bond donor in that binding pocket. The cytisine template served to initiate an essential SAR study for the nicotinic partial agonist program. Our early work demonstrated that CNS penetration could be improved compared with the natural product without sacrificing the desired pharmacodyamic parameters. We further showed that cytisine-derived partial agonists could be discriminated as nicotinic agents and would provide a reduced nicotinic-like reward while blocking the effect of nicotine. We uncovered a range of functional activities by modifying ring substitutions or by subtle changes in the electronic character of the polarizable heteroaromatic ring. Unfortunately, the overall profile of these ligands was marred by genetic toxicology findings within the substituted cytisine class. Our original plan to create novel scaffolds using the chemistry that follows opened up a broad and entirely novel range of partial agonist templates.
5. CYTISINE SYNTHESIS AND EARLY TEMPLATE EXPANSION With the advances provided by studies of cytisine and derivatives, access to compounds not available by direct analog generation became more important. The cost of cytisine as a raw material had originally been a concern (B$1.7 M/kg, Aldrich, B1995), but had been addressed through potential agricultural harvest. To support further derivatization and SAR development, synthetic approaches became essential, as they offered the ability to make not only cytisine but also novel cytisine-like derivatives inaccessible from cytisine itself. These synthetic efforts began in earnest within our group in 1995. Of the strategies we considered for cytisine’s construction, two novel synthetic designs were ultimately demonstrated and later published. The most concise strategy is exemplified by biaryl coupling chemistry of pyridine precursors 29 and 30 to access both the pyridone and the piperidine ring atoms of cytisine (5) (Scheme 3). This step-efficient approach requires a nucleophilic nitrogen component in 31 to reveal the cytisine pyridone in the bicycle formation step [44] and allows the introduction of a range of substituents on the final pyridone ring [45].
Case History: Chantixt/Champixt (Varenicline Tartrate)
H
N
Z
O
O
O
29
Scheme 3
N
N
N
X
N
X
Y
N
85
30
(+/-)-cytisine 5
31
Biaryl coupling approach to cytisine [44].
A second strategy employed a palladium-mediated Heck cyclization of a glutarimide derived enol triflate intermediate (33) to access the bicyclic ring core (34), ultimately providing racemic cytisine in six steps from cyclopent-3-enyl-methanol (32) (Scheme 4) [46]. In this approach, the N-C bond of the pyridone ring was established first, with the piperidine synthesis finalized after bicycle formation (34-5). The construction of alternative templates not containing a nitrogen ring fusion atom was possible utilizing the same strategic approach (32-37, Scheme 5). Soon after the successful syntheses of racemic cytisine were established, we targeted non-pyridone-containing derivatives and the efforts rapidly yielded active compounds [43,47]. An early anisole derivative (38) had high affinity (Ki ¼ 1.4 nM) and displayed weak partial agonist activity at the receptor in vitro in oocytes (30% relative to nicotine, Figure 7). In vivo this analog displayed weak activity in the DATO assay, in part due to rapid demethylation in vivo to the less active phenol (40, Ki ¼ 90 nM, partial agonist). These promising early results launched an 18-month effort of intensive chemical synthesis, producing B100 novel carbon analogs of cytisine. Despite the targeted chemical exploration, derivatives from this series generally displayed either reduced affinity or antagonist activity in oocytes (e.g., 39 and 41, o20% efficacy relative to nicotine). Compounds worthy of further pursuit as partial agonists did not emerge from the effort. The most potent non-pyridone cytisine
H N X N
N HO 32
Scheme 4
N
O
O 33
O 34
(+/-)-cytisine 5
Palladium mediated heck approach to cytisine [46].
86
JOTHAM W. COE et al.
H N X R2
R2
HO
32
Scheme 5
R2
R1
R1 35
R1 non-pyridone derivatives 37
36
Approach to non-pyridone cytisine derivatives.
derivatives generally were compounds with antagonist action at the receptor, such as the difluoroderivative 42 [48]. One aspect of the pharmacophore brought into question by SAR developed in this series was the role of the cytisine carbonyl group. Most published binding motifs suggested the existence of a hydrogen bond between the cytisine carbonyl and the receptor [17,49]. In contrast, we saw enhanced potency of compounds possessing electron withdrawing functionality, the most potent being the difluoro derivative 42, which possesses H-bond acceptor interactions that are arguably weak [50]. These data suggested that dipole or other electronic interactions are important for binding and functional efficacy at the a4b2 receptor. These results were consistent with the profound changes we observed between cytisine (5) and isocytisine (27) activity, wherein H-bonding alone is insufficient to explain the pharmacophore interactions. Furthermore, 42 was almost a full antagonist (3% agonist), clearly demonstrating
H N
H In vitro N Ki α 4β2 1.4 nM 30% ag. / 51% antag. vs. nicotine In vivo active short half-life H3CO
OCH3 38 H N
In vitro Ki α4β2 0.4 nM 3% ag. / 85% antag. vs. nicotine In vivo active antagonist
F
39
antagonists
In vitro Ki α4β2 2.9 nM
3
R O
2
H
Figure 7
40
41
1 C
H
HO OH
42
F
HN
H N In vitro Ki α4β2 90 nM partial agonist
H N
In vitro Ki α4β2 >500 nM
H
H
O H
H
binding site
Selected SAR and H-bond pharmacophore probes.
untolerated
43
87
Case History: Chantixt/Champixt (Varenicline Tartrate)
H N
HN
O 44
H N
H N
O
N
N O
H N
H N
O 45
O 46
N 27
N R 47
48
Figure 8 Additional templates.
non-parallel responses of ligand binding and functional receptor responses. With no rational explanation regarding control of functional efficacy, structural guidance from SAR was for the most part empirically derived. A brief exploration of heterocyclic derivatives (e.g., 48, Figure 8) [43], accessed by the same synthetic route design, provided compounds with either weak functional activity or minimal selectivity over the nicotinic muscle receptor, another undesirable property.
6. DISCOVERY OF THE BICYCLIC BENZAZEPINE CORE The weak in vitro and in vivo responses of non-pyridone cytisine derivatives (Figure 7) drove our expanded search for alternative structural scaffolds targeting improved potency and partial agonist activity (Figure 8). We synthesized various reasonable assemblies of the critical functional groups found in many natural nicotinic agents, probing alternative locations and orientations of necessary functionality. Many of these approaches proved disappointing, since they did not provide improvements in nAChR activity. The avenue that defined much of our future work came from the opioid literature, as we recognized that morphine and cytisine both possess embedded [3.3.1]-bicyclic core structures. Paul Mazzocchi’s work from the 1970s explored various simplified versions of morphine 49, the complex alkaloid from the opium poppy, Papaver somniferum. Bicyclic substructures of morphine were evaluated, including the [3.3.1]-bicyclic benzomorphan 50, and Mazzocchi and coworkers found that N-alkyl derivatives of 50 possessed morphine-like anti-nociceptive effects mediated through m-opioid receptors. Additional publications compared modified bicyclic frameworks of 50, including 51 and 52, all of which produced analgesia with particular nitrogen substituents (Figure 9) [51]. In 1979 Mazzocchi published the synthesis and anti-nociceptive activity of derivatives of an N-positional isomer of 52, namely [3.2.1]-bicyclic benzazepine 53 [51]. Analogs of 53 with N-alkyl substitution were found to have greatly reduced anti-nociceptive activity. Importantly, Mazzocchi
88
JOTHAM W. COE et al.
antinociceptive
OH
OH
N
HN O N
H
OH morphine
49
?
HN
H N
HN
O OH
50
51
52
53
Figure 9 Mazzocchi’s benzomorphane modifications [52]. (See Color Plate 4.9 in Color Plate Section.)
generalized and emphasized his key finding in the last sentence of the discussion: ‘‘Clearly, the change in nitrogen position in proceeding from ‘52 to 53’ manifests itself by an almost total loss of anti-nociceptive activity and a marked increase in toxicity.’’ The toxic nature of nicotinic agents derived from plants – for example nicotine (2), anatoxin a (54), cytisine (5), and epibatidine (55) – is to protect the host from microbial, insect, and animal predators. We wondered whether the non-pyridone cytisine scaffold 37 and the bicyclic benzazepine 53, both 3,5-disubstituted piperidines, possessed nicotinic pharmacology that could explain the increased toxicity of derivatives of 53 reported by Mazzocchi [52]. Truncated analogs 53 and 52 were prepared [53] (Figure 9), tested in opioid and nicotinic-binding assays and found to be highly selective, 53 for the a4b2 nAChR, and 52 for opioid receptors [51] (53 Ki m-opioid W2 mM; 52 Ki a4b2 W5 mM); no cross-reactivity was observed. Furthermore, benzazepine 53 was equipotent at the a4b2 nAChR to the unsubstituted parent non-pyridone cytisine derivative 37 R1 and R2 ¼ H (Ki ¼ 20 vs 34 nM, 53 and 37 respectively; Scheme 5). Both unsubstituted parent compounds were antagonists; however, the bridged benzazepine 53 provided an achiral symmetric template, greatly simplifying SAR development. Our studies expanded into this fertile area and yielded promising results. Electron-deficient and sp2-hybridized groups are common functionalities in nicotinic natural products, for example, acetylcholine (4), anatoxin a (54), cytisine (5), nicotine (2), and epibatidine (55), all contain acyl or pyridyl groups (Figure 10). We initially targeted similar structures to explore bridged benzazepine SAR. Our initial attempts to functionalize the bicyclic benzazepine by electrophilic substitution chemistry were unsuccessful, as the parent structure and typical N-protected versions proved inert to substitution. Usually nitration readily introduces functionality for subsequent SAR development, but early attempts at nitration reactions with both the N-carbamate-protected and the free
89
Case History: Chantixt/Champixt (Varenicline Tartrate)
HN
N
HN
O O acetylcholine, 4
anatoxin a, 54
N
N
N
O
H
Cl
HN
O
N epibatidine, 55
(-)-cytisine, 5
CH3
(-)-nicotine, 2
Figure 10 Naturally occurring nicotinic ligands.
benzazepine 53 did not progress, even under more forcing conditions. We were surprised to find that even nitronium triflate [54], a powerful nitrating agent, failed to derivatize the carbamate-protected benzazpine. We suspected that interactions of electrophilic reagents with the amine or protected amines would generate non-productive cationic piperidinium intermediates that served to insulate the aryl ring against electrophilic chemistry. To test this hypothesis, the trifluoroacetamide (TFA) group was chosen to mask the nitrogen functionality from competing for electrophilic attack. With the TFA group, aromatic substitution of 56 progressed smoothly (Scheme 6), with a breakthrough achieved while employing the combination of N-TFA protection and nitronium triflate in methylene chloride to afford the mononitrated derivative 57 in 78% yield. After deprotection, 58 was found to exhibit potent partial agonist activity in vitro and in vivo in DATO after both subcutaneous and oral administration in rats. This result established that nicotinic agents derived from 53 were excellent partial agonist targets. We then returned to the original non-pyridone cytisine [3.3.1]-bicyclic structure to introduce nitro functionality. Three isomers were generated (59, 60, and 61) under the nitration conditions, and each displayed weaker binding affinity than the nitrobenzazepine, but more importantly, all displayed weaker functional agonist activity than the nitro-benzazepine derivative 58, in line with findings of greater antagonist behavior observed with earlier non-pyridone-based cytisine analogs [47] (Scheme 5, 38). Although these templates differ by only a single core carbon atom, the consequent effect on potency and efficacy is remarkable (Figure 11) from 58. Given the relative ease of preparation of bicyclic
CF3 O
N
H N
CF3 2.6 equiv CF3SO3H 1.3 equiv HNO3
O
N
OH-
78% O
56
Scheme 6
NO2 N
O
SO3CF3
57
58 NO2
Nitro benzazepine: potent partial agonist revealed.
In vitro Ki α4β2 0.75 nM 64% ag. / 36% antag. vs. nicotine In vivo active 50% partial agonist in DATO s.c. and p.o.
90
JOTHAM W. COE et al.
H N
H N In vitro Ki α4β2 4.5 nM 20% ag. / 40% antag. vs. nicotine
59
NO2
Figure 11
H N
In vitro Ki α4β2 6.5 nM 2% ag. /50% antag. vs. nicotine NO2
60
In vitro NO2 Ki α4β2 14 nM 0% ag. /100% antag. vs. nicotine
61
Nitrated non-pyridone cytisines.
benzazepine analogs from a common intermediate, the partial agonist activity of suitably substituted compounds at a4b2 nAChRs, and the promising early in vivo efficacy signals, our active pursuit moved to this series. During the scale up of the nitration reaction with a slight excess of nitronium triflate (56-57), we discovered that a dinitrated by-product was formed. This reaction gave the desired mono-nitrated product 57 after crystallization in high yield (Scheme 6), but we re-examined the residues for a doubly nitrated by-product that had been observed in low yield by gas chromatography with mass spectrometry (GCMS) in the crude material, now enriched in the mother liquors. After isolation, a 9:1 mixture of two dinitro derivatives were identified, with the major isomer assigned the structure resulting from unexpected vicinal dinitration, compound 62, the minor was the expected meta isomer 63 (Scheme 7). Upon exposure to greater than two equivalents of nitronium triflate, the conversion to the dinitrated benzazepine 62 proceeded in 77% yield. This discovery fueled much of our future chemistry effort by facilitating rapid access to fused rings [55]. As SAR developed within this series (Figure 12), we found that monofunctionalized products (65, 66) were uniformly active at the a4b2 nAChR, with electron-deficient functional groups displaying subnanomolar affinity. Electron-donating groups (R1) conferred weaker binding affinity, and functional activity and peri-substitution further reduced activity (R2 position in 67), consistent with the SAR
O
O
CF3 N
CF3 N
> 2 eq NO2.OSO2CF3
9:1 NO2
O 2N
56
Dinitration of benzazepine 56.
CF3 N
-78 - 20 °C, 24 h
Scheme 7
O
NO2
62
O2N
63
Case History: Chantixt/Champixt (Varenicline Tartrate)
H N
H N
H N
H N
N O R Cytisine Derivatives 64
91
R2 R1
R1
Benzazapines less active enantiomer
Benzazapines more active enantiomer
Benzazapines less active regioisomer
65
66
67
Figure 12 Benzazepine SAR trends.
observed within the non-pyridone analog series. Electrophysiological measurements at a4b2 nAChRs expressed in oocytes showed that the partial agonist activity within mono-substituted benzazepines in racemic form ranged from 0 to 86%, with most electron withdrawing groups (R1) imparting partial agonist profiles worthy of further evaluation in optically pure form [56]. Most pharmacophore models for high-affinity nAChR ligands identify two critical ligand components: an ammonium ion and a Hbond acceptor [17,49]. The resolved enantiomers revealed an unexpected SAR trend. We expected that substituents on mono-functionalized compounds (R1) would align with cytisine’s carbonyl to establish Hbonding interactions with the receptor. We were surprised to find these isomers were less active i.e. 65 (Figure 12). The more potent enantiomers 66 positioned the electron withdrawing group away from the cytisine carbonyl orientation (typically the more potent isomers were B10-fold higher binding affinity with greater agonist functional efficacy, Figure 12). In addition, electron-withdrawing groups of widely varying H-bond acceptor ability were generally equipotent and efficacious, although each presents dissimilar H-bonding orientations into the receptor pocket (R1 ¼ NO2, Ac, CN, halo, CF3, etc.). These findings have led us to question the validity of assertions that H-bonding is a predominant feature within the a4b2 nAChR pharmacophore. Instead, these results suggest that a key feature governing affinity and possibly functional efficacy may be an interface between the ligand sp2-hybridized component and key receptor residues through productive p-electron interactions. This notion may extend our understanding of the welldocumented p-cation interaction of the ammonium head group common to all nicotinic agents [57] by including productive interactions of the ligand p-system and receptor aryl groups in addition to the ammonium p-cation interaction.
92
JOTHAM W. COE et al.
7. FUSED BICYCLIC BENZAZEPINES As mentioned earlier, the dinitration reaction of TFA-protected benzazepine (56-62) led to a number of important fused heterocycles. SAR development in the ‘‘6,6-fused’’ derivatives revealed the importance of the ‘‘quinoline nitrogen’’ (5-position) to affinity and functional activity (Figure 13). Quinoxaline derivatives (68), directly accessed from dinitro intermediates in two steps, displayed optimal partial agonist activity with hydrogen substitution (e.g., varenicline (1)). Small groups were welltolerated at C-6 and 7 (H, CH3, OH, OCH3, etc.) but reduced the functional potency of these derivatives. Aryl appendages generally decreased binding affinities and reduced partial agonist efficacy. Both quinoline (69) and quinazoline (70) derivatives with small groups displayed high affinity, but were uniformly full agonists, being more efficacious at 10 mM compared with the effect of 10 mM nicotine in oocytes. 3-Substitutied quinolines with groups larger than methyl (7-position of 69, Figure 13) displayed decreased agonist activity and affinity. Isoquinolines (71) were 10-fold less potent compared with similar quinoline (69) and quinazoline (70) derivatives. Electronic and steric changes at the ‘‘5-position,’’ which occupies a position similar to the ‘‘cytisine carbonyl’’ in molecular modeling overlays, considerably reduced activity and revealed the importance of this site. 6,5-Fused heterocyclic analogs also exhibited a broad range of functional activities in oocytes from full agonist to antagonist at 10 mM relative to 10 mM nicotine (Figure 14). With the exception of larger groups in the 6-position, most compounds based on this general construction (72–75) had Ki values o1 nM. The functional efficacy of analogs within this 6,5-fused heterocyclic series was sensitive to structural changes at the 6-position, as illustrated by benzimidazole (72) and benzisoxazole (73) derivatives. All C-6 hydrogen-substituted analogs are partial agonists with high affinity, whereas C-6 methyl substituted analogs were consistently found to be agonists with greater efficacy in oocytes than nicotine itself when measured at 10 mM. The 2-methyl benzothiazole (74) and benzisoxazole (75) derivatives were also particularly efficacious agonists, but increases in C-2 substituent size beyond methyl (in 72 and 73) appeared to inversely affect functional efficacy and binding affinity of
8 N
R′
R′
HN
1
3
68
N 5
N
HN
HN R
N
N
R
69
Figure 13 6,6-Fused heterocyclic benzazepines.
70
HN
N
R
71
93
Case History: Chantixt/Champixt (Varenicline Tartrate)
R′ N 7 HN N
3
1
72
S
O R
R
HN
73
HN
N
N
N
5
O R
74
75
Figure 14 6,5-Fused heterocyclic benzazepines.
analogs. Benzimidazoles (72) with R7-substituents were uniformly potent presumably by accessing a pocket not found to be productive within the 6,6-fused heterocycles of Figure 13. An available binding pocket for substitution appended to the aryl group is an SAR point observed with other known nicotinic agents, but the orientational distinctions between the effect of 6,6- and the effect of 6,5-fused heterocycle substitution on activity is striking [49].
8. IN VIVO EFFICACY OF PARTIAL AGONISTS The in vivo efficacy of analogs was determined by measuring effects on mesolimbic DATO, which reflects a change in postmortem tissue concentration ratios of DA and its metabolites in the rat nucleus accumbens. Compounds were evaluated subcutaneously (s.c.) and, if active, were re-examined after oral administration. In this assay, nicotine (2) produces a maximal DATO response at 1 mg/kg s.c. of B180% of control levels. The partial agonist activity of an agent alone was determined by comparison of the effect of maximum well-tolerated doses with that of 1 mg/kg s.c. nicotine. These partial agonist efficacies at the maximal well-tolerated dose of the compounds are represented as black bars in Figure 15, showing a wide range of efficacies relative to nicotine. Compounds with sufficient potency at well-tolerated doses were then examined for another hallmark of partial agonist activity, that is, the ability to inhibit nicotine-induced increases in DATO. This property was uncovered in the ‘‘antagonist mode’’ of DATO, in which attenuation of the maximal nicotine response was evaluated after co-administration of the compound and nicotine, and is represented by the shaded bars in Figure 15. Although many compounds act as partial agonists alone and produce sub-maximal DATO increases relative to nicotine, few compounds fully antagonized nicotine’s effect at their maximal welltolerated dose. Only compounds that effectively blocked nicotine’s effect with sufficient tolerability displayed the desired partial agonist dual action to warrant further pursuit. Figure 15 shows that many cytisine analogs have an optimal profile as efficacious partial agonists that can completely block nicotine-induced DATO increases. 3-Bromocytisine is particularly efficacious, displaying
94
JOTHAM W. COE et al.
110
agent alone
100
agent + 1 mg/kg s.c. nicotine
Dopamine turnover
% nicotine response + SEM
90 80 +
70
*
+
+
60
+
++
**
* *
50
*
*
*
40
*
*
**
*
**
+
++
30 20 10 0 -10 0
2
5.6
5
0.178
14
5.6
20
8.0
38
5.0
42
5.6
72
1.78
73
3.2
74
5.6
75
5.6 mg/kg sc
1
Figure 15 Effects of (–)-nicotine, (–)-cytisine, and selected agents on DATO in rat nucleus accumbens 1 h post-dose. All values are expressed as percentages of the effect of 1.0 mg/kg s.c. nicotine (100%)7SEM (N ¼ 510). Each compound was administered at the indicated dose (mg/kg s.c.) alone (black bars) and together with 1 mg/kg s.c. nicotine (shaded bars). *p o.05 agent alone vs vehicle; **p o.01 agent alone vs vehicle; +p o.05 and ++p o.01: agent with nicotine vs nicotine alone (oneway ANOVA with post hoc Dunnett’s test).
high partial agonist efficacy and fully antagonizing the effect of nicotine when co-administered at a very low dose of 0.178 mg/kg. This in vivo potency is consistent with 3-bromocytisine’s potent functional agonist activity in vitro in oocytes (EC50 ¼ 90 nM). However, this dose also represents the maximum tolerated dose in animals, which translates to a very narrow therapeutic index. The non-pyridone cytisine analogs failed to potently elevate DATO or to block nicotine’s effect, consistent with their weaker binding and presumably poor brain exposure. The bicyclic aryl piperidines (benzazepines) showed a wide range of agonist activities, with some analogs being highly active as antagonists of nicotine’s effect. Potent benzazepine analogs that were particularly efficacious were also well-tolerated at the high doses used in this assay. Comparing the in vivo effects of compounds on DATO greatly aided our triage for suitable clinical candidates and limited the selection to cytisine derivatives (e.g., 5, 14, 20), mono-substituted bicyclic benzazepine derivatives (e.g., 72, 73) and fused bicyclic benzazepine derivatives (e.g., 74, 75, varenicline (1)). Fused heterocyclic compounds were eliminated from further evaluation if they displayed functional efficacy
Case History: Chantixt/Champixt (Varenicline Tartrate)
95
either too low or too high and fell outside the in vitro window of acceptable partial agonism. We had targeted in vitro functional efficacy of 25–80% of nicotine (10 mM in oocytes), but many compounds that were progressed and evaluated in DATO lacked adequate partial agonist activity in vivo, which we attributed to insufficient free brain exposure necessary to exhibit robust agonist and antagonist efficacy. For example, benzimidazole derivatives of benzazepine (6-propyl and 6-butyl derivatives 74 and 75, respectively) are partial agonists when given alone, but failed to fully antagonize nicotine at tolerated doses, whereas the ‘‘nonpyridone’’ cytisine analogs (e.g., 38, 42) were less efficacious in vivo and poor antagonists of the nicotine response. Unlike cytisine derivatives, which were uniformly potent in vivo as agonists and as nicotine antagonists (e.g., 5, 14, 20), the other series displayed a broader range of in vivo efficacies and potencies.
9. PROPERTIES OF VARENICLINE Varenicline (1) was selected from the compounds described in this study as the primary development candidate. It is a low molecular weight (221.27 g/mol) achiral alkaloid with high CNS penetration in rat brain (B/P ¼ 3.5), a symmetric CSF/free plasma ratio of 1 and low protein binding in blood (B80% free) and brain (B67% free). Varenicline’s properties are highlighted by a ligand efficiency of 0.82 and LLE (pIC50log P) of 11.
9.1 Pharmacology Receptor binding studies [55,58] demonstrated that varenicline (1) has high affinity only for the a4b2 neuronal nicotinic receptor subtype in rat and human cortex (Ki B0.1–0.4 nM). Varenicline did not bind with significant affinity to various other neurotransmitter receptors and transporters, enzymes, modulatory binding sites, and ion channels, in membranes derived from relevant tissues and cell lines (Ki values W1,000 nM). In vitro functional patch clamp studies in Xenopus oocytes and HEK cells expressing human nAChRs showed that varenicline (1) is a partial agonist with 45% of nicotine’s maximal effect at the a4b2 nAChR that can fully antagonize the effect of simultaneously applied nicotine [55,58]. In neurochemical models, varenicline (1) displayed significantly lower efficacy (40–60%) than nicotine in stimulating [3H]-DA release from rat brain slices in vitro and in increasing DATO in rat nucleus accumbens, while it inhibited nicotine-induced DATO, as shown earlier [55,58]. In vivo microdialysis was used to examine the effects of varenicline (1) on the concentration of extracellular DA levels in the nucleus accumbens
96
JOTHAM W. COE et al.
225 nicotine Dopamine release in n. accumbens % of basal ± SEM
200
175 varenicline 150
125 nicotine + varenicline 100 Nicotine 0.32 mg/kg
Varenicline 1 mg/kg
75 -120
-60
0
60
120
180
240
300
360
Time (minutes)
Figure 16 Effects of 0.32 mg/kg s.c. nicotine alone (curve without symbols), 1 mg/kg p.o. varenicline alone (gray curve with triangles) and combined administration of 1 mg/kg p.o. varenicline with 0.32 mg/kg s.c. nicotine (black curve with open squares) on DA release in rat n. accumbens. Data are expressed as percentage of basal levels7SEM (N ¼ 35) (adapted from Coe et al., 2005) [55].
of conscious rats, to directly assess varenicline’s effects on the mesolimbic DA system. Dose–response curves indicated that varenicline maximally increased DA release to 153% of baseline with an ED50 of 0.032 mg/kg p.o. The maximal response of varenicline is about 63% of the full agonist nicotine (2), which produces a maximal increase in DA release at 0.32 mg/kg s.c. to 184% of basal levels. When administered together, varenicline reduced the peak effect of nicotine to the level of its own maximal effect on DA release, that is, about 60% of the maximal nicotine increase (Figure 16) [55,58]. In behavioral animal models, varenicline (1) reduces nicotine selfadministration in rats and supports lower self-administration break points than nicotine. In a progressive ratio self-administration paradigm, animals must work at increasingly harder levels (more bar presses) to receive their next drug infusion [58]. The lower breakpoint found for varenicline than for nicotine is consistent with the notion that a partial agonist is less reinforcing and less dependence producing than nicotine. Neurochemical support for the lack of abuse liability is provided by the
Case History: Chantixt/Champixt (Varenicline Tartrate)
97
time courses for the effects on DA release that demonstrate the slow onset and long-lasting effect of varenicline compared with nicotine’s rapid steep rise and fall in accumbens DA (Figure 16), thought to be characteristic of drugs of abuse. These animal studies, as well as data from a later abuse liability study in human subjects confirm that varenicline has no abuse potential [58]. Taken together, all data support the hypothesis that the partial agonist varenicline (1) will have utility as a smoking cessation aid. Through submaximal activation of the mesolimbic dopaminergic system, it can reproduce to some extent the subjective effects of smoking to reduce cravings and withdrawal symptoms without nicotine’s abuse liability. In addition, since varenicline effectively reduces the rapid and robust mesolimbic DA response to nicotine, the antagonist action will prevent the reinforcing and rewarding effects of nicotine to protect against smoking relapse. On the basis of these findings, varenicline (1) was advanced into clinical development.
9.2 Absorption, distribution, metabolism, excretion (ADME) As a small hydrophilic, weak base (log P ¼ 1.1; pKa ¼ 9.9; measured Elog D7.4 0.28, MW ¼ 211), varenicline (1) is well absorbed, as indicated by preclinical disposition studies that found 89% to be recovered in the urine of carbon-14-labeled material after oral administration (B1% in feces). Protein binding of varenicline is low (human blood free unbound B80%) and it has a moderate volume of distribution (1.9 L/kg). Varenicline is virtually not metabolized, comprising 90% of circulating drug-related material and is excreted mostly unchanged in the urine [59]. Four minor metabolites of varenicline were observed, with two minor urinary metabolites comprising less than 5% of the dose. Metabolites observed in excreta arose through N-carbamoyl glucuronidation and oxidation. These metabolites were also observed in the circulation, in addition to metabolites that arose through N-formylation and formation of a novel hexose conjugate. Unbound renal clearance is in slight excess of glomerular filtration rate. Varenicline is neither a substrate nor an inhibitor of cytochrome P450 enzymes. It is expected to be neither the cause of nor subject to drug interactions through alterations of P450 activities [59]. As such, drug–drug interactions are generally limited to those induced by decreased nicotine intake upon quitting smoking, as nicotine is a cytochrome P450 2A6 substrate [60]. In phase 1 clinical trials, varenicline (1) was found to have an excellent PK profile, with a half-life of 24 h, steady-state levels achieved in 4 days, no food effects, and with the drug essentially completely absorbed and not metabolized (W89% of C-14-labeled material accounted for in urine) [61,62]. The long half-life is a particularly
98
JOTHAM W. COE et al.
beneficial property for protecting smokers from smoking relapse, as nicotine is rapidly absorbed and displays a short 1–2 h half-life in human subjects. Toleration at higher doses was limited by nausea and vomiting. The maximum well-tolerated dose of 1 mg BID p.o. (twice daily, orally) was chosen to minimize the nausea signal and to maximize efficacy in smoking cessation.
10. CLINICAL STUDIES Results of phase 2 and phase 3 studies have been described in detail and have been reviewed elsewhere [63]. Efficacy was established in phase 2 studies for three different doses [1 mg QD (once daily), 0.5 mg BID and 1 mg BID] given for 6 or 12 weeks [64] One milligram BID for 12 weeks was chosen as the most efficacious treatment and used for all further clinical trials. In two identically designed phase 3 clinical trials, varenicline (1) was compared to placebo and the active comparator, bupropion SR (Zybans sustained release). In these studies smokers were treated for 12 weeks pharmacotherapy with 40 weeks post-treatment follow-up of smoking status. At the end of the 12 week treatment period, the CO-confirmed 4-week abstinence rates were 44.0% for smokers who received varenicline (N ¼ 696) compared with 29.7% in the bupropion SR group (N ¼ 671) and 17.7% in the placebo group (N ¼ 685). At the end of 1 year, 22.4% in the varenicline group, 15.4% in the bupropion group, and 9.3% in the placebo group remained completely abstinent from smoking. In pooled analyses, varenicline was statistically superior to both bupropion SR and placebo at the end of the treatment period and at the 1-year follow-up [65–68]
11. CONCLUSIONS When smokers quit tobacco, they typically experience considerable side effects [69] and are highly susceptible to relapse from re-exposure to inhaled tobacco smoke, especially during quit attempts. Properties of nicotine make this particularly true, as it is completely absorbed in the lungs and delivered to the brain within 7–10 s, making nicotine addiction one of the most challenging addictions to overcome. We sought a medicine that not only decreased nicotine craving and withdrawal symptoms but effectively and safely reduced the neurochemical reinforcement produced by smoking. Partial agonists theoretically address both of these primary physiologic aspects of nicotine dependence, but required the identification of compounds with in vivo features that effectively competed with inhaled nicotine. These objectives helped
Case History: Chantixt/Champixt (Varenicline Tartrate)
99
define the testing parameters necessary to identify an effective partial agonist as a smoking cessation aid, with long duration to reduce craving and withdrawal symptoms while effectively shielding smokers in the event they smoke during quit attempts. Varenicline met these stringent criteria, not only in preclinical models, but later in clinical trials. Many important medicinal agents originated from work on smallmolecule natural product progenitors. Cytisine helped us establish preclinical parameters for a discovery program and to elucidate a starting point for a chemical and medicinal journey. Natural products not only provided a starting point, they inspired total synthesis strategies that helped access unnatural analogs in the search for novel templates with the desired activity. This ultimately led us to studies of structurally related opioid natural products. In retrospect, it is fitting that a partial agonist approach for opioid addiction sparked the conceptual foundation of this approach and that opioid research should later provide new inspiration leading to varenicline. Disclosure: Jotham W. Coe, Hans Rollema, and Brian T. O’Neill are employees of Pfizer Inc.
REFERENCES [1] R. Doll and A. B. Hill, Br. Med. J., 1954, 1451. [2] R. Doll, R. Peto, J. Boreham and I. Sutherland, Br. Med. J., 2004, 328, 1519. [3] ‘‘Smoking and Health: Report of the Advisory Committee to the Surgeon General of the Public Health Service,’’ 1964. [4] Smoking and Health: A Report of the Surgeon General, 1979. [5] The ‘‘Recognizing Addiction as a Disease Act of 2007,’’ SB-1011. [6] B. Adhikari, J. Kahende, A. Malarcher, T. Pechacek and V. Tong, Morb. Mortal. Wkly. Rep., 2008, 57, 1226. [7] WHO Report on the Global Tobacco Epidemic, 2008 – The MPOWER package. http:// www.who.int/tobacco/mpower/en/index.html [8] K.-O. Fagerstro¨m, J. Behav. Med., 1982, 5, 343. [9] J. R. Hughes et al., Antidepressants for smoking cessation. Cochrane Database of Systemic Reviews 1, CD000031. DOI: 10.1002/14651858.CD000031.pub3, 2007. [10] (a) C. A. Hughes, Jr., Cancer J. Clin., 2000, 50, 143; (b) M. C. Fiore, et al., J. Am. Med. Assoc., 2000, 283, 3244.(c) M. C. Fiore et. al., U.S. Public Health Service, U.S. DHHS, Rockville, MD, May 2008. [11] S. Wakhlu, J. Opioid Manag., 2009, 5, 59. [12] (a) J. E. Rose and E. D. Levin, Pharmacol. Biochem. Behav., 1991, 41, 219; (b) J. E. Rose, F. M. Behm, E. C. Westerman, E. D. Levin, R. M. Stein and G. V. Ripka, Clin. Pharmacol. Ther., 1994, 56, 86. [13] J. Boulter, J. Connolly, E. Deneris, D. Goldman, S. Heinemann and J. Patrick, Proc. Nat. Acad. Sci. U.S.A., 1987, 84, 7763. [14] R. Lape, D. Colquhoun and L. G. Sivilotti, Nature, 2008, 454, 722. [15] A. A. Jensen, B. Frolund, T. Liljefors and P. Krogsgaard-Larsen, J. Med. Chem., 2005, 48, 4705. [16] D. Paterson and A. Nordberg, Neurobiology, 2000, 61, 75.
100
JOTHAM W. COE et al.
[17] (a) W. H. Beers and E. Reich, Nature, 1970, 228, 917; (b) R. P. Sheridan, J. Med. Chem., 1986, 29, 899. [18] J. P. Changeux, A. Devillers-Thiery and P. Chemouilli, Science, 1984, 225, 1335. [19] M. R. Picciotto, N. A. Addy, Y. S. Mineur and D. H. Brunzell, Prog. Neurobiol., 2008, 84, 329. [20] R. C. Hogg and D. Bertrand, Curr. Drug. Targets. CNS Neurol. Disord., 2004, 3, 123. [21] M. E. M. Benwell and D. J. K. Balfour, Br. J. Pharamacol., 1992, 105, 849. [22] C. F. Morrison and J. A. Stephenson, Psychopharmacologia (Berl.), 1969, 15, 351. [23] C. Romano, A. Goldstein and N. P. Jewell, Psychopharmacology, 1981, 74, 310. [24] W. T. Chance, M. D. Kallman, J. A. Rosecrans and R. M. Spencer, Br. J. Pharmac., 1978, 63, 609. [25] J. A. Pratt, I. P. Stolerman, H. S. Garcha, V. Giardini and C. Feyerabend, Psychopharmacology, 1983, 81, 54. [26] W. A. Corrigall and K. M. Coen, Psychopharmacology, 1989, 99, 473. [27] (a) M. R. Picciotto, M. Zoli, R. Rimondini, C. Lena, L. M. Marubio, E. M. Pich, K. Fuxe and J. P. Changeux, Nature, 1998, 391, 173; (b) S. S. Watkins, M. P. Epping-Jordan, G. F. Koob and A. Markou, Pharmacol. Biochem. Behav., 1999, 62, 743. [28] A. R. Tapper, S. L. McKinney, R. Nashmi, J. Schwarz, P. Deshpande, C. Labarca, P. Whiteaker, M. J. Marks, A. C. Collins and H. A. Lester, Science, 2004, 5, 1029. [29] (a) G. Di Chiara, Eur. J. Pharmacol., 2000, 393, 295; (b) J. A. Dani and M. De Biasi, Pharmacol. Biochem. Behav., 2001, 70, 439. [30] R. M. Wilson and S. J. Danishefsky, J. Org. Chem., 2006, 71, 8329. [31] F. van Haaran, K. G. Anderson, S. C. Haworth and W. R. Kem, Pharmacol. Biochem. Behav., 1999, 64, 439. [32] W. R. Kem, V. M. Mahnir, R. L. Papke and J. L. Lingel, J. Pharmacol. Exp. Ther., 1997, 283, 979. [33] W. R Kem and R. L. Papke, Soc. Neurosci., 1992, 18, 1358. [34] C. Reavill, B. Walther, I. P. Stolerman and B. Testa, Neuropharmacology, 1990, 29, 619. [35] A. I. Sacaan, J. L. Dunlop and G. K. Lloyd, J. Pharmacol. Exp. Ther., 1995, 274, 224. [36] A. Partheil, Arch. Pharm., 1894, 232, 161. [37] G. Scharfenberg, S. Benndorf and G. Kempe, Dtsch Gesundheitsw., 1971, 26, 463. [38] C. W. Luetje and J. Patrick, J. Neurosci., 1991, 11, 837. [39] R. L. Papke and S. F. Heinemann, Mol. Pharmacol., 1994, 45, 142. [40] (a) R. Mansbach, F. D. Tingley, C. Rovetti, T. Davis, L. Chambers, C. Fox, S. Sands, E. Arnold, A. Elder, J. Huang, D. Schulz and B. T. O’Neill, Soc. Neurosci., 1998, 294, 10. (b) B. T. O’Neill, PCT Int. Appl. WO 98 18,798, Abstr. 1998, 119, 4774k, 1998. [41] A. L. Hopkins, C. R. Groom and A. Alex, Drug Discov. Today, 2004, 9, 430. [42] P. D. Leeson and B. Springthorpe, Nat. Rev. Drug Discov., 2007, 6, 881. [43] S. Demers, H. Stephenson, J. Candler, C. G. Bashore, E. P. Arnold, B. T. O’Neill and J. W. Coe, Tetrahedron Lett., 2008, 49, 3368. [44] (a) B. T. O’Neill, D. Yohannes, M. W. Bundesmann and E. P. Arnold, Org. Lett., 2000, 2, 4201; (b) E. Marriere, J. Rouden, V. Tadino and M.-C. Lasne, Org. Lett., 2000, 2, 1121; (c) O. Nicolotti, C. Canu Boido, F. Sparatore and A. Carotti, Farmaco, 2002, 57, 469; (d) C. Canu Boido, A. Carotti and F. Sparatore, Farmaco, 2003, 58, 265. [45] (a) A. P. Kozikowski, S. K. Chellappan, Y. Xiao, K. M. Bajjuri, H. Yuan, K. J. Kellar and P. l. A. Petukhov, Chem. Med. Chem., 2007, 2, 1157; (b) S. K. Chellappan, Y. Xiao, W. Tueckmantel, K. J. Kellar and A. P. Kozikowski, J. Med. Chem., 2006, 49, 2673. [46] J. W. Coe, Org. Lett., 2000, 2, 4205. [47] J. W. Coe, M. G. Vetelino, C. G. Bashore, M. C. Wirtz, P. R. Brooks, E. P. Arnold, L. A. Lebel, C. B. Fox, S. B. Sands, T. I. Davis, D. W. Schulz, H. Rollema, F. D. Tingley, III and B. T. O’Neill, Bioorg. Med. Chem. Lett., 2005, 15, 2974.
Case History: Chantixt/Champixt (Varenicline Tartrate)
101
[48] C. G. Bashore, M. G. Vetelino, M. C. Wirtz, P. R. Brooks, H. N. Frost, R. E. McDermott, D. C. Whritenour, J. A. Ragan, J. L. Rutherford, T. W. Makowski, S. J. Brenek and J. W. Coe, Org. Lett., 2006, 8, 5947. [49] R. A. Glennon and M. Ducat, Pharm. Acta. Helv., 2000, 74, 103. [50] E. Carosati, S. Sciabola and G. Cruciani, J. Med. Chem., 2004, 47, 5114. [51] (a) P. H. Mazzocchi and B. C. Stahly, J. Med. Chem., 1979, 22, 455; (b) M. Mokotoff and A. E. Jacobson, J. Het. Chem., 1970, 7, 773; (c) P. H. Mazzocchi and A. M. Harrison, J. Med. Chem., 1978, 21, 238. [52] T. Eisner and J. Meinwald (eds), Chemical Ecology: The Chemistry of Biotic Interaction, National Academy of Sciences, Washington, DC, 1995. [53] (a) P. R. Brooks, S. Caron, J. W. Coe, K. K. Ng, R. A. Singer, E. Vazquez, M. G. Vetelino, H. H. Watson, Jr., D. C. Whritenour and M. C. Wirtz, Synthesis, 2004, 11, 1755; (b) R. A. Singer, J. D. McKinley, G. Barbe and R. A. Farlow, Org. Lett., 2004, 6, 2357; (c) C. J. O’Donnell, R. A. Singer, J. D. Brubaker and J. D. McKinley, J. Org. Chem., 2004, 69, 5756. [54] C. L. Coon, W. G. Blucher and M. E. Hill, J. Org. Chem., 1973, 38, 4243. [55] J. W. Coe, P. R. Brooks, M. G. Vetelino, M. C. Wirtz, E. P. Arnold, J. Huang, S. B. Sands, T. I. Davis, L. A. Lebel, C. B. Fox, A. Shrikhande, J. H. Heym, E. Schaeffer, H. Rollema, Y. Lu, R. S. Mansbach, L. K. Chambers, C. C. Rovetti, D. W. Schulz, F. D. Tingley, III and B. T. O’Neill, J. Med. Chem., 2005, 48, 3474. [56] J. W. Coe, P. R. Brooks, M. C. Wirtz, C. G. Bashore, K. E. Bianco, M. G. Vetelino, E. P. Arnold, L. A. Lebel, C. B. Fox, F. D. Tingley, III., D. W. Schulz, T. I. Davis, S. B. Sands, R. S. Mansbach, H. Rollema and B. T. O’Neill, Bioorg. Med. Chem. Lett., 2005, 15, 4889. [57] J. P. Gallivan and D. A. Dougherty, Proc. Nat. Acad. Sci. U.S.A., 1999, 96, 9459. [58] H. Rollema, L. K. Chambers, J. W. Coe, J. Glowa, l. R. S. Hurst, L. A. Lebel, Y. Lu, R. S. Mansbach, R. J. Mather, C. C. Rovetti, S. B. Sands, E. Schaeffer, D. W. Schulz, F. D. Tingley, III and K. E. Williams, Neuropharmacol, 2007, 52, 985. [59] R. S. Obach, A. E. Reed-Hagen, S. S. Krueger, B. J. Obach, T. N. O’Connell, K. S. Zandi, S. A. Miller and J. W. Coe, Drug Metab. Dispos., 2006, 34, 121. [60] J. Hukkanen, P. Jacob, III and N. L. Benowitz, Pharmacol. Rev., 2005, 57, 79. [61] H. M. Faessel, B. J. Smith, M. A. Gibbs, J. S. Gobey, D. J. Clark and A. H. Burstein, J. Clin. Pharmacol., 2006, 46, 991. [62] H. M. Faessel, M. A. Gibbs, D. J. Clark, K. Rohrbacher, M. Stolar and A. H. Burstein, J. Clin. Pharmacol., 2006, 46, 1439. [63] K. I. Cahill, L. F. Stead, T. Lancaster, Cochrane Database of Systematic Reviews 1, 2007, Art. No.: CD006103, in The Cochrane Library, Issue 1, Wiley, Chichester, UK. doi:101002/ 14651858.CDC006103.pub2, 2007. [64] M. Nides, C. Oncken, D. Gonzales, S. Rennard, E. J. Watsky, R. Anziano and K. R. Reeves, Arch. Intern. Med., 2006, 166, 1547. [65] D. Gonzales, S. I. Rennard, M. Nides, C. Oncken, S. Azoulay, C. B. Billing, E. J. Watsky, J. Gong, K. E. Williams and K. R. Reeves, J. Am. Med. Assoc., 2006, 296, 47. [66] D. E. Jorenby, J. T. Hays, N. A. Rigotti, S. Azoulay, E. J. Watsky and K. E. Williams, J. Am. Med. Assoc., 2006, 296, 56. [67] S. Tonstad, P. Tonnesen, P. Hajek, K. E. Williams, C. B. Billing and K. R. Reeves, J. Am. Med. Assoc., 2006, 296, 64. [68] M. Nides, E. D. Glover, V. I. Reus, A. G. Christen, B. J. Make, C. B. Billing and K. E. Williams, Am. J. Health. Behav., 2008, 32, 664. [69] M. J. Jarvis, Bri. Med. J., 2004, 32, 277.
CHAPT ER
5 Case History on Tekturnas/ Rasilezs (Aliskiren), a Highly Efficacious Direct Oral Renin Inhibitor as a New Therapy for Hypertension Ju¨rgen Maibaum* and David L. Feldman**
Contents
1. 2. 3. 4.
Introduction Rationale for the Use of Direct Renin Inhibitors Pre-Clinical Models to Study Direct Renin Inhibitors Medicinal Chemistry Evolution — The Early Renin Inhibitor Program at Ciba-Geigy 4.1 The emerging novel topology design concept 4.2 Macrocyclic renin inhibitors 4.3 (P3-P1)-Tethered hydroxyethylene transition-state mimetics 4.4 SAR in the THQ series — early pre-clinical leads 4.5 Modification of the transition-state mimetic portion 4.6 SAR in the ‘Phenoxy’ series 4.7 Challenges of a multiple chemotype approach 5. First Convergent and Scalable Synthesis Development 6. Pre-Clinical Properties of Aliskiren 7. Effects of Aliskiren in Disease Models 8. Clinical Studies with Aliskiren 9. Conclusions References
105 106 107 108 110 111 112 113 115 116 119 120 122 122 123 124 124
Novartis Institutes for BioMedical Research, Novartis Campus, CH-4056 Basel, Switzerland Novartis Institutes for BioMedical Research, East Hanover, NJ, USA
Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04405-4
r 2009 Elsevier Inc. All rights reserved.
105
106
Ju¨rgen Maibaum and David L. Feldman
1. INTRODUCTION Hypertension is a major risk factor for cardiovascular diseases, affecting more than 25% of adults worldwide [1]. The high percentage of patients with insufficiently controlled blood pressure (BP) levels suggests a continued need for improved antihypertensive drug therapies. Direct renin inhibition is a promising new type of treatment for high BP. The renin-angiotensin-aldosterone system (RAAS) is a key physiological regulator of BP and fluid homeostasis. The first and rate-limiting step (i.e., point of activation) of this cascade is catalyzed by the highly specific aspartic protease renin [2]. Direct blockade of this step has long been recognized as most attractive for therapeutic intervention. This triggered extensive research activities initiated in the late 1970s in the quest for identifying clinically efficacious orally active direct renin inhibitors. However, various challenges persisted in this area for more than two decades despite all the efforts made in the pharmaceutical industry. Recently, an unprecedented topological structure-based drug design strategy at Ciba-Geigy (now Novartis) enabled the discovery of novel, highly potent and selective non-peptide transition-state mimetic (TSM) renin inhibitors. These efforts culminated in the discovery of TEKTURNAs/RASILEZs (aliskiren, 1, Figure 1) recently approved as a new therapy for the treatment of hypertension. As a first-in-class direct renin inhibitor, aliskiren has demonstrated orally active anti-hypertensive efficacy and end-organ protection. Thus, more than 10 years after the introduction of the angiotensin (Ang) AT1 receptor blockers (ARBs), a therapy directed at renin is available to patients.
2. RATIONALE FOR THE USE OF DIRECT RENIN INHIBITORS Blockade of the RAAS is a primary treatment in many cardiovascularrenal diseases including hypertension, congestive heart failure and renal disease [3]. The ultimate goal of such therapy is to block the formation or action of Ang II, the principal effector of this pathway. Until the approval of aliskiren, the RAAS could be inhibited either at the step of conversion of Ang I to Ang II with Ang I converting enzyme inhibitors (ACEi) or at CH3O
OH
H N
H2N O CH3O
Figure 1
TEKTURNAs/RASILEZs (aliskiren).
O 1
CONH2
Case History on Tekturnas/Rasilezs (Aliskiren)
107
the Ang II receptor level with ARBs. Such interventions have yet to halt the progression of cardiovascular diseases. That is, tissue damage may continue in the heart, blood vessels and kidney, possibly because the RAAS is not comprehensively blocked with ACEi or ARB. In contrast, inhibiting the RAAS at the point of its activation (i.e., renin) may lead to superior downstream RAAS inhibition as compared to existing therapies. Moreover, formation of Ang II by alternative pathways (e.g., chymase) is limited with renin inhibition because Ang I formation is blocked. A unique feature of renin inhibition is the ability to block the increase in plasma renin activity (PRA) that accompanies the rise in the plasma levels of renin during RAAS blockade. Thus, although ACEi and ARB actually raise PRA, renin inhibitors reduce PRA despite the high levels of circulating renin. This action leads to effective inhibition of Ang II formation and potentially greater organ protection.
3. PRE-CLINICAL MODELS TO STUDY DIRECT RENIN INHIBITORS The road towards developing human renin inhibitors has been challenging with respect to establishing appropriate in vivo models in which to test the efficacy of these compounds. This difficulty relates, at least in part, to the prominent species specificity displayed by renin and the high primate specificity of potent human renin inhibitors. Thus, most inhibitors of human renin are only weakly effective in rats. Consequently, although rats are the most commonly used species in hypertension efficacy studies, such models have been of limited use to study human renin inhibitors [4]. Potent inhibitors of rat renin have allowed in vivo studies in this species; however, the design of dual human–rat renin inhibitors has been extremely challenging [5,6]. Normotensive nonhuman primates such as cynomolgus, squirrel, rhesus and marmosets [7–9], and dogs [10,11], in which the RAAS is stimulated by a low salt diet and/or a diuretic, have been used to demonstrate the efficacy of renin inhibition on BP, cardiac function and plasma RAAS components. We initially established a model of sodium-depletion in restrained conscious marmosets (Callithrix jacchus). In this model, BP and heart rate (HR) measurements were performed invasively and thereby allowed to determine only short-term hemodynamic effects [12]. Subsequently, the application of radio-telemetry for continuous measurement of BP, HR and changes in the electrocardiogram in freely moving animals permitted the assessment of the effects of drug treatment under less stressful conditions over a longer time range [13]. We also measured the 24 h timecourse of PRA ex vivo as a surrogate marker to indicate that BP-lowering effects were specifically related to renin inhibition.
108
Ju¨rgen Maibaum and David L. Feldman
The ability to manipulate the genome of rodents has permitted the development of several transgenic rat and mouse models, in which renin is expressed. This eliminated the issue of species specificity of renin inhibitors in vivo. Indeed, beneficial effects of renin inhibition, which are considered to be of important clinical relevance, have been demonstrated in these models [14]. Double transgenic rats (dTGR) express human genes for renin and angiotensinogen [15]. The high expression of this ‘human RAAS’ leads to high levels of Ang II and consequently to renal and cardiac damage [16]. These animals are highly responsive to inhibition of their RAAS with human renin inhibitors. Aliskiren, 1, has been intensively studied in dTGRs and other transgenic rodents for its effects on BP and organ protection.
4. MEDICINAL CHEMISTRY EVOLUTION — THE EARLY RENIN INHIBITOR PROGRAM AT CIBA-GEIGY Tremendous resources have been devoted to the design of orally efficacious renin inhibitors, starting in the late 1970s [17–20], and a vast number of potent and selective peptide-like TSM inhibitors have been reported. Several candidates progressed into the clinic but suffered from poor intestinal absorption or high liver first-pass elimination, high synthetic complexity and hence high cost of goods, and/or insufficient clinical efficacy. The renin drug discovery project at Ciba-Geigy started in 1980 and transitioned through the evolution over three generations of renin inhibitors, ranging from modified linear peptides initially, to more drug-like peptidomimetics and ultimately to completely novel chemical structures that still function as TSM inhibitors. Computer-aided molecular modeling along with X-ray crystallography of human renin with and without bound inhibitors enabled us to adopt a systematic direct targetstructure based strategy. This eventually culminated in the discovery of aliskiren, 1, representing a unique class of topological peptidomimetic renin inhibitors, as described in more detail by this review. The large peptide inhibitor CGP29287, 2 (Figure 2), derived from the N-terminal sequence of angiotensinogen by incorporation of the statine TSM at the renin cleavage site and with modification of both N- and Cterminal sites, was the first proteolytically stable, long-acting renin inhibitor shown to be active after intravenous and oral administration to monkeys [12]. Although not considered ‘drug-like’ due to its peptide characteristics, this first-generation inhibitor proved its importance as a pharmacological tool for studying the endocrine, hemodynamic, renal and cardiac responses to tissue and systemic renin inhibition in vivo in primates [21]. Subsequently, reduction of molecular size by truncation at both the N- and the C-terminus and incorporation of a hydroxyethylene
Case History on Tekturnas/Rasilezs (Aliskiren)
109 O
O
O
H N
N
H N O
NH
N
NH
HN
O
H N
NH2
HN
O
H N
N H
O
OH
O N H
O
O
H N
O
N H
O
O
N
O N H
NH HN
NH2
Figure 2 First generation peptide renin inhibitor — CGP29287 (2).
N
NH
O S O O
N H
H N
N OH
O
H N
O
S
H N
N H
O O
O
3 N
O N
S
O O
N H
NH
H N O
OH
O
OH
4
N
N
S
O N
OH
OH
S
O O
5
N H
H N O
OH OH
6
Figure 3 Second-generation peptide-based renin inhibitors.
(HE) isostere provided inhibitors spanning the S4-S2u enzyme recognition sites. Further structure refinement identified CGP38560, 3 (Figure 3) [22], a renin inhibitor of the second generation and the first selected at CibaGeigy for clinical investigations. This potent and selective inhibitor of human and marmoset plasma renin in vitro (IC50 ¼ 0.7 nM) reduced mean arterial pressure (MAP) and suppressed PRA in Na-depleted restrained marmosets (10 mg/kg oral dose) [23]. Owing to its short duration of action in hypertensive patients and low (o1%) oral bioavailability in humans, 3 was not considered to be a clinically viable drug [24,25]. Remikiren (RO-42-5892, 4), with a plasma human renin IC50 ¼ 0.8 nM, combined the N-terminal tert-butylsulfonyl of 3 with a shorter C-terminal diol TSM [8] and was investigated by Hoffmann–La Roche in phase II trials. RO-42-5892 caused significant BP lowering in hypertensive
110
Ju¨rgen Maibaum and David L. Feldman
patients, but antihypertensive responses were only short-lasting at the maximal tolerated dose [26]. Oral bioavailability of 4 was low in monkeys, rats, dogs and humans (B1%) due to rapid liver first-pass metabolism and biliary excretion of unchanged drug [27,28]. It is noteworthy that 4 was proposed to act primarily by inhibiting an extra-plasma renin pool [29]. RO-42-5892 is the first renin inhibitor for which extensive partitioning to, and retention by, the kidney has been reported [30]. More water soluble second generation inhibitors with superior pharmacokinetics in animal species were subsequently disclosed. SC56525, 5 (IC50 ¼ 1.2 nM), was reported by Searle to exert potent oral BP lowering effects in both Na-depleted and renal hypertensive dogs [11]. Oral bioavailability of 5 was 66% in dogs (30 mg/kg po); however, nonlinear pharmacokinetics and a short elimination half-life (1.3 h) were observed [31]. Zankiren (A-72517, 6, IC50 ¼ 1.1 nM) culminated from a structure–oral absorption optimization strategy at Abbott [7,32–34]. Oral bioavailability in cynomolgus monkey, dog and rat was 8, 53 and 24%, respectively, with the dog predicting best the human bioavailability [33]. Remarkably, distribution studies in animals showed selective uptake of 6 into the kidneys, which could explain its favorable effects on renal hemodynamics [34]. In patients, 6 showed rapid oral absorption and significant antihypertensive efficacy [35]. Zankiren has been considered a hallmark second-generation renin inhibitor, although the drug was not developed further [33].
4.1 The emerging novel topology design concept During the late 1980s, the need became apparent to identify a structurally novel concept that could lead to the development of orally active renin inhibitors with improved properties. Such improvements included reduced molecular weight, increased aqueous solubility and other factors considered to impact oral efficacy. In view of the experience with CGP38560, 3, it was deliberately decided to move away in a radical fashion from the classical peptidomimetic approach. Re-direction of the design strategy and medicinal chemistry resources towards ‘conceptually different inhibitors’, the title of the new project, gave birth to the topological (P3-P1-P0) working hypothesis during 1987–1988 (Figure 4). No validated hits resulted from a small ‘random screening’ campaign that was based on a collection of B20,000 compounds delivered by our newly established New Lead Discovery Unit. Our new design concept was entirely target structure-based and evolved through several major steps: i) using a human renin homology model, exploration of aryl and bulky alkyl P3 fragments linked to diverse cycloalkyl P1 templates bearing polar head groups as putative truncated TSMs ; ii) starting from P1u-P2u extended HE isostere TSMs and growing
Case History on Tekturnas/Rasilezs (Aliskiren)
P4
P3
P2 N
O S O O
N H
P1
P1′
P2′
NH H N
OH
P0 Truncated transition-state mimetic OH *H2N
H N
P1′
111 P2′
OH* *optional
O
O
3
P3
P1
Figure 4 The initial (P3-P1-P0) topology approach.
towards P3 from a rigid P1 residue, providing sub-micromolar leads; iii) introduction of H-bond acceptors/donors to optimized (P3-P1)pharmacophores, leading to a significant increase in potency and in vivo activity; iv) discovery of a distinct non-substrate pocket S3sp by X-ray crystallography as a canonical binding pocket for different sub-series and experimental confirmation of the topology design concept; v) structurebased lead optimization of key series and the search for novel (P3-P1)chemotypes. The conceptual innovation was based on the hypothesis that both the S3 and the S1 renin recognition sites constitute a contiguous ‘hydrophobic super-pocket’ — or ‘hydrophobic hot spot’ — similar to fungal aspartyl proteases with strong preference for lipophilic ligands positioned in spatial proximity. This notion provoked the search for hydrophobic scaffolds composed of directly linked P1 and P3 motifs. Initially, a diverse array of novel conformationally more or less constrained mono-, bi- and tricyclic core structures was intensively investigated, which were modeled on optimized SAR for P1 and P3 of peptide-derived renin inhibitors. These compounds were further characterized by one or more OH and NH2 groups of an acyclic or cyclic terminal scaffold envisaged to target the catalytic aspartates as simplified versions of a TSM truncated at its prime-site (P3-P1-P0 hypothesis, Figure 4). Despite the tremendous efforts and difficult chemical syntheses involved, none of these molecules was found to inhibit renin up to 100 mM in the enzymatic assay.
4.2 Macrocyclic renin inhibitors The lack of success in identifying even very low-affinity inhibitors initially was disappointing and raised major concerns about the validity of the design concept. Up to this point, our modeling depended on overlaps with the crystal structures of pepstatin or related peptides complexed with non-mammalian aspartyl proteases such as penicillopepsin and rhizopuspepsin [36]. In the course of our work, a more refined homology model of human renin derived from the endothiapepsin X-ray structure [37] was used to predict the active site conformation of 3 [38]. The first apo-crystal structure of human renin
112
Ju¨rgen Maibaum and David L. Feldman
N O O
N H O O
Figure 5
NH H N
OH
H N O
7
(P3-P1)-linked macrocyclic peptidomimetic renin inhibitor.
was reported only in 1989 [39]. In our group, the human renin/3 complex was resolved shortly thereafter, confirming the accuracy of the computational model [40]. At this point, interest arose in the design of P3-P1 macrocyclic inhibitors for probing the feasibility of covalently linking the P1 and P3 residues of an open-chain congener inhibitor. It was furthermore envisaged that freezing such constrained peptidomimetics in their enzyme-bound conformation would lead to a gain in entropy thereby possibly allowing elimination of peripheral elements (e.g., P4 and/or P2u residues), and hence diminishing molecular size. The in vitro activity of the non-optimized (P3-P1)-macrocycle 7 (Figure 5) against human renin (IC50 ¼ 2 mM), although moderate, supported the notion that the renin S1/S3 specificity sites are not separated from each other but form a large hydrophobic cavity. More potent macrocyclic inhibitors incorporating directly linked P1 and P3 side chains were reported after we had concluded our limited efforts [41]. The proximal topology of other recognition sites of renin inspired the design of various classes of potent macrocyclic inhibitors [9].
4.3 (P3-P1)-Tethered hydroxyethylene transition-state mimetics As the original approach targeting truncated transition-state surrogates had failed to deliver any starting points for further lead optimization, the emphasis of the exploratory design was redirected. Additional binding interactions to the P1u and/or P2u sites of renin were conceived to be likely more important than was initially acknowledged for such putative (P3P1)-pharmacophores lacking interactions to the P2-P4 sites. The cyclohexyl-substituted HE isostere 8 (IC50 ¼ 30 mM, Figure 6) was selected as a starting ‘fragment’ based on the well-established preference of human renin for bulky hydrophobic P1 residues, providing selectivity against other human aspartyl proteases [42]. Inhibitor 9 was one of the most active (IC50 ¼ 0.3 mM) of several tethered 1,3- or 1,4-cis and transdisubstituted cyclohexyl analogues [43]. The 100-fold potency increase for 9 vs. 8 strengthened the evidence for the feasibility of the tethered P3-P1 topological inhibitor design.
Case History on Tekturnas/Rasilezs (Aliskiren)
113
10: R = H H2N
OH CH3 H N
R
O
H2N
R
OH CH3 H N
11: R =
O
trans (rac)
12: R =
8: R = H Cyclohexyl- 9: R = ‘optimized’ P1
Gem-dimethyl ‘symmetrical’ P1
N O
13: R = CH2CO2H
Figure 6 Growing the TSM ‘fragment’ — growing the evidence.
The synthetic complexity of cycloalkyl-derived inhibitors such as 9 restricted further optimization. Symmetrization and hence avoiding any stereocenter at P1 by tethering the geminal-dimethyl position of the TSM 10 (IC50 W100 mM; Figure 6) afforded 11 (IC50 ¼ 0.7 mM). This and subsequently the incorporation of a functionalized spacer were major breakthrough steps towards more simplified early leads with impressively improved in vitro potencies. The tetrahydroquinoline (THQ) derivative 12 (IC50 ¼ 0.05 mM) was W2,000-fold more potent than 10, which further increased our confidence in the design concept. The advanced carboxylic acid 13, appropriately N,O-protected [44], enabled a systematic and rapid SAR exploration of the THQ moiety (vide infra). Variation of the P3-P1 motif was explored intensively with the aim to optimize the interactions to the spacious S1/S3 sites. Hydrophobic van der Waals contacts to this hydrophobic ‘hot-spot’ were considered as key contributing factors for strong ligand binding. We therefore envisaged to maximize the contacts of an optimally designed lipophilic and conformationally rigid P3-P1 pharmacophore to the large surface of the S3/S1 cavity, as exemplified by inhibitors 14 (IC50 ¼ 2.7 mM; Figure 7) and 15 (IC50 ¼ 2.6 mM). Less constrained analogues of 15 with improved in vitro potencies resulted from P1 mono-substitution, and hence reintroduction of a stereocenter with a preferred configuration [45]. Enlarging the size of P1 from methyl (16, IC50 ¼ 2 mM) to isopropyl afforded inhibitor 17 (IC50 ¼ 0.1 mM). Sterically demanding residues such as tert-butyl (18, IC50 ¼ 1.5 mM) and phenyl (19, IC50 ¼ 39 mM) resulted in decreased enzyme affinities. During this work, advances in synthetic chemistry facilitated the practical access to HE isosteres by starting from the corresponding enantiomeric amino acids [45–47]. Yet, early SAR optimization often remained cumbersome due to the non-convergent, lengthy and only partially stereo-controlled reaction sequences required for each single structural modification of the P3-P1 scaffolds. This was particularly evident for inhibitors 16–19, which raised some controversy about the chances to progress this series with reasonable efforts. Incorporation of functionalized spacers into the (P3-P1)-motif of 17, allowing more rapid
114
Ju¨rgen Maibaum and David L. Feldman
H2N
OH CH3 H N
H2N
OH CH3 H N
O
O
14
15 H 2N
OH CH3 H N O
R
16: R=CH3 17: R=CH(CH3)2 18: R=tert-Butyl 19: R=Phenyl
Figure 7
Morphing the (P3–P1)-pharmacophore as a hydrophobic ‘hot spot’.
SAR studies similar to those for 12, were evaluated later in the fully elaborated ‘phenoxy’ series; however, these efforts resulted only in inferior inhibitors.
4.4 SAR in the THQ series — early pre-clinical leads Despite all efforts to optimize the (P3-P1)-scaffold by the ‘hydrophobic approach’, a plateau for in vitro activity apparently had been reached, exemplified by the moderate activities of 12 and 17. Next, we focused our attention on additional design strategies to achieve a further substantial gain in potency. Close examination of SAR data for second-generation renin inhibitors, as well as X-ray crystal structures of peptide inhibitors bound to fungal aspartyl proteases, emphasized the importance of a canonical H-bond between the P3/P2 amide carbonyl and the backbone NH of Ser219 for strong binding affinity. Intriguingly, computational docking of 11 and related analogues into the renin homology model suggested the ‘upper’ portion of the P3 moiety to be positioned in proximity to Ser219 [48,49]. Introduction of H-bond acceptors/donors at the 3- or 4-position of the naphthalene in 11 was extensively probed; however, these early attempts did not result in any affinity increase [43]. In contrast, substitution of the THQ heterocycle of 12 with a carboxylic ester at C-3 afforded inhibitor 20 (IC50 ¼ 0.8 nM, Figure 8) with a dramatic boost in potency. The 3(R)-configured 20 was 50-fold more potent than its (S)-isomer and 12, which indicated a highly specific interaction to the enzyme requiring the proper spatial orientation of the ester group [48]. The sequential discovery of 12 and 20 occurred within a few weeks of each other and represented a hallmark event at about the same time when the renin project was under close scrutiny. The ester in 20 could be replaced with amides, hydroxylated alkyl and short-chained alkyl ethers without a significant drop in the IC50s. THQ modifications were investigated by substitution, incorporation of
Case History on Tekturnas/Rasilezs (Aliskiren)
O
O H2N N
HN
OH CH3 H N
OH
N
O O
20
O O
NH
OH
21 (2′ R or S)
H N
H2N N
H N
H 2N
S
O O
O
115
O O 22
Figure 8 SAR of the THQ series — early pre-clinical lead candidates.
heteroatoms and changing the ring size. The 1,4-benzthiazine 21 (Figure 8; absolute stereochemistry not assigned), bearing a P1u isopropyl at the TSM portion (vide infra), was found to be a potent inhibitor of human and marmoset plasma renin (IC50 ¼ 3 and 16 nM) with selectivity towards bovine cathepsin D and porcine pepsin (IC50s W50 mM). It is noteworthy that 21 and similarly analogue 22 showed excellent aqueous solubility (1.74 and 3.44 g/L, pH 7.4) and low lipophilicity (logPoct 1.71 and 1.70). In restrained Na-depleted marmosets, oral administration of 21 (10 mg/kg) lowered BP significantly by 30 mmHg at peak, while PRA was completely blocked over 6 h. This first early pre-clinical lead was not progressed further, as more attractive analogues emerged from both this and other sub-series. THQ analogue 20 was the first from a total of six novel renin inhibitors from different classes, for which X-ray structures in complex with human renin were resolved [50]. Importantly, this crystal structure confirmed both the S1 and the S3 sites to be fully occupied by the geminaldimethyl group and the aryl portion of the THQ, providing the first structure proof for the computational-based topographical design. An additional H-bond interaction of the spacer carbonyl of 20 to the flexible enzyme flap domain was observed. The ester carbonyl of 20 formed H-bonds with the Ser219 NH, as was predicted, and in addition with the side chain OH. Most surprisingly, the X-ray revealed the methoxy group pointing into the direction of a solvent-shielded non-substrate binding site. This Ssp 3 pocket extends towards the center of renin and is flanked partly by hydrophobic amino acids. Given these results, focused lead optimization then targeted substitutions that would penetrate more deeply into the Ssp 3 pocket [50]. Late-stage SAR advanced towards in vivo potent THQ analogues while pre-clinical development of inhibitor 30 from the ‘phenoxy’ class was well under way (vide infra). Inhibitor 22 showed a remarkable improvement in potency compared to 21 (plasma
116
Ju¨rgen Maibaum and David L. Feldman
renin IC50 ¼ 0.5 nM vs. 3 nM) and excellent dose-dependent oral efficacy over 24 h in telemetered Na-depleted marmosets (peak DMAP of 25 mmHg at 3 mg/kg), accompanied by suppression of PRA for up to 24 h. However, 22 did not demonstrate a superior profile as compared to 30 in exploratory in vivo cardiovascular studies.
4.5 Modification of the transition-state mimetic portion Replacements of the classical HE isostere by shortening the C-terminal portion and eliminating stereocenters have been investigated intensively as a strategy towards simpler peptide-derived inhibitors with improved oral activity and bioavailability [17]. We too explored a small set of such TSM analogues early in view of the synthetic challenges we encountered during initial SAR work. The norstatine ester (23, IC50 ¼ 2.2 mM), the Nethyl oxazolidinone (24, IC50 ¼ 0.64 mM) [51], and the ‘Roche’ erythro-1,2diol (25, IC50 ¼ 4.2 mM) were at least three orders of magnitude less potent than parent 20 (Figure 9). Remarkably, the truncated amino alcohol 26 was almost equipotent (IC50 ¼ 1.0 mM) compared to 23–25, which indicated only minor binding contributions by the S1u extended TSM portion relative to that of the (P3-P1)-scaffold. Similar findings were subsequently observed for related analogues of 30. SAR by substitution of the TSM primary amine revealed, in general, a significant affinity loss even for small N-alkyl or N-acyl residues in all sub-series. These efforts, however, were limited since retaining the peptide backbone was not part of our strategy, although this approach was being used by others [20]. These results clearly emphasized that for this class of topological renin inhibitors interactions to the extended prime site of human renin are required to induce strong binding affinities. As a consequence, we restricted all our subsequent SAR optimization to the classical HE isostere, as this was thought to provide the optimal trajectory into the S2u pocket. Retrospectively, the weak potency of 26 bearing an ‘optimized’ (P3-P1)-scaffold is noteworthy in view of the expectations for hit identification at the outset of our conceptual design program (Figure 4). At the time, sensitive biophysical methods such as protein NMR to detect very low-affinity ligands were not available [52].
OH H2N N
MeO2C
O O 26: R = Me, Q = H
H2N
O
Q 24: R = Et, Q =
O
Figure 9
O
23: R = Et, Q =
RO2C
N O
Modifying the transition-state mimetic (TSM).
N
OH OH
O 25 (2′ R/S)
Case History on Tekturnas/Rasilezs (Aliskiren)
117
4.6 SAR in the ‘Phenoxy’ series The concept of incorporating suitable H-bond acceptors/donors into the (P3-P1)-scaffold was successfully applied across different series. Starting from 17 (Figure 7) and tethering a terminal ethyl carboxylate to the ortho position relative to the tert-butyl group of the phenyl spacer resulted in a 20-fold potency increase (27, purified human renin IC50 ¼ 6 nM, Figure 10) [45]. Modelling had suggested the ester carbonyl to be in binding distance to Ser219 with the methoxy group pointing towards the P4 recognition site of renin. Carboxamide 28 (IC50 ¼ 20 nM) showed only a slight drop in affinity against purified renin. However, inhibitor 28 (plasma IC50 ¼ 460 nM) and most other tert-butyl analogues were very poor inhibitors in vitro in the human plasma renin assay with angiotensinogen as the substrate. The significantly reduced potency in plasma generally observed in this series was attributed to the high intrinsic local lipophilicity of the tert-butyl group, yet the actual underlying mechanism for the potency loss under the physiological assay conditions remained unclear [53]. An important milestone was the replacement of tert-butyl by the smaller and more polar methoxy group without compromising binding affinity. This afforded inhibitors such as 29 (purified renin IC50 ¼ 42 nM, plasma IC50 ¼ 95 nM) that were equipotent under the different assay conditions. This ‘rule breaking’ finding merits emphasis in the light of the perceived preference of human renin for hydrophobic aryl or bulky alkyl residues at P3, as was suggested by ample previous SAR data. Extensive SAR optimization of 29 then focussed on modifications of the phenoxy side chain, starting from a common advanced intermediate [53]. Shortening or lengthening of the flexible linear alkoxy substituent and placing one or two oxygen atoms at different positions afforded a
O O H2N
R
MeHN
OH CH3 H N
O
O 29: Q = CH3O MeO
O 27: OMe 28: NH2
30: Q =
O
MeO MeO
OH
H N
H2N O MeO
O 31
Figure 10 Milestone SAR of the ‘Phenoxy’ series.
H2N Q
OH CH3 H N O
118
Ju¨rgen Maibaum and David L. Feldman
library of potent inhibitors, the most prominent analogue of which was 30 (purified renin/plasma IC50 ¼ 1 nM, Figure 10). The X-ray structure of human renin-bound 30, resolved shortly after that of the THQ 12, confirmed the close hydrophobic interactions of the P1 isopropyl, the phenyl spacer and the P3 methoxy to the S3-S1 ‘superpocket’ [50]. Again unpredicted by modeling, and most striking, was the finding that the methoxypropoxy residue occupied the non-substrate Ssp 3 pocket by forming a H-bond to Tyr14 at the bottom of the cavity. It is quite remarkable that optimization of the phenoxy side chain in 30, permitting a perfect fit into Ssp 3 of renin, was completed before its true binding interactions to this site were uncovered by X-ray. In telemetered Nadepleted marmosets, 30 induced pronounced BP reductions (peak DMAP 20–25 mmHg), which were still significant after 24 h (at 30 mg/kg) [53]. No changes in HR were observed over the 1–30 mg/kg oral dose range. Bioavailability in non-depleted marmosets was moderate (16%) with a short terminal half-life (2 h). Pre-clinical cardiovascular safety pharmacology, however, revealed 30 to induce a dose-dependent transient reduction in HR and changes in cardiac conduction (intravenous dose range 0.3–10 mg/kg) to anesthetized sub-primates, which were unrelated to in vivo renin inhibition. Owing to the insufficient therapeutic margin, pre-clinical investigation of 30 was terminated. Ongoing lead optimization activities aimed to identify analogues that were more efficacious in vivo with longer duration of action. An increasing body of evidence emerged within this and other series that P1u isopropyl substitution of the HE isostere gave inhibitors with similar in vitro, but markedly superior in vivo potency as compared to their P1u methyl congeners. This increased potency may have been due to as yet unknown beneficial pharmacokinetic properties. For example, 31 (purified/plasma renin IC50 ¼ 1/4 nM) showed significantly stronger depressor effects (peak DMAP of 20 mmHg) as compared to 30 over 4–5 h (3 mg/kg) [54]. Considerable efforts targeting P2u modifications identified several N-terminal carboxamides, which demonstrated tight binding in vitro to plasma renin and markedly increased oral efficacy with prolonged duration of action in Na-depleted marmosets. From this sub-series, 1 (plasma renin IC50 ¼ 0.6 nM) emerged as one of the most potent inhibitors of human renin that we had identified at that time. In vitro potency was less against plasma renin from marmoset (IC50 ¼ 2 nM) and non-primates (mouse, rat, dog, rabbit and guinea pig IC50s ¼ 4, 80, 7, 11 and 63 nM, respectively). The promising drug candidate 1 was intensively studied in vivo, first in Na-depleted marmosets [54–56]. The dose-dependent and sustained decrease in BP following oral administration (0.3–30 mg/kg) was accompanied by a decrease in PRA and an increase in plasma renin levels. Maximal efficacy of 30 mmHg DMAP was achieved at 3 mg/kg, which corresponded to the maximal BP
Case History on Tekturnas/Rasilezs (Aliskiren)
119
depressor effect induced by ACEi and ARB in this model [56]. Inhibitor 1 was markedly more effective than equivalent oral doses of remikiren, 4, and zankiren, 6. Repeated once-daily oral dosing (10 mg/kg/day) over 8 days caused effective BP lowering over 24 h without altering HR and without rebound increase in BP after termination of treatment [55]. On the basis of its promising pharmacology and safety profile, 1, which came to be known later as SPP100 and then aliskiren, was taken forward to clinical development.
4.7 Challenges of a multiple chemotype approach In the course of our novel chemotype renin inhibitor program, at least four structurally diverse major lead series emerged from the initial topological design approach providing sub-micromolar inhibitors [49]. Principal conceptual considerations such as the introduction of suitable H-bond acceptors/donors to the distinct P3-P1 scaffolds could be successfully transferred between sub-series, leading to breakthroughs in in vitro potency in each case. Although key in vitro SAR were consistent between the various sub-series, lead optimization for in vivo efficacy and refining of other key inhibitor profiles proceeded in a less parallel fashion, hence involving different sets of lead SAR. An additional complication, particularly during the late stage of the project, was the need to develop separate efficient synthetic paradigms for candidate inhibitors from different sub-series. As an example, the potent indole inhibitor 32 (purified/plasma renin IC50 ¼ 1/4 nM, Figure 11) was derived from weakly active 11 by incorporation of a synthetically flexible spacer bridging a mono-alkyl P1 residue and a P3 aryl residue [48]. This lead demonstrated a more pronounced BP reduction (DMAP 20 mmHg) but was significantly shorter acting than 30 (10 mg/kg po) in Na-depleted marmosets. Unexpectedly, X-ray analysis of an analogue of 32 in complex with human OH O N
OH
H N
H2N
O
O
N H
N
H N
H2N
O
O
N H 33
32 MeO
MeO OH O
O
N H
Figure 11 Multiple
OH
H N
H2N O
NH2
O
O
N H
34
(P3-Psp 3 )-pharmacophore
H N
O H 2N
approach.
O 35
N O
120
Ju¨rgen Maibaum and David L. Feldman
renin showed that the indole moiety was posed in a flipped binding conformation with the N-benzyl substituent positioned within the same Ssp 3 pocket, that had been identified shortly before by X-ray of 20 and 30 [50]. As a consequence, the indole was in close hydrophobic contact with the S3 pocket instead of mimicking the peptide backbone as proposed by modeling. This suggested a promiscuous character of the rigid Ssp 3 pocket accommodating both hydrophobic aryl groups and more polar heteroaliphatic side chains. Psp 3 phenyl-substituted analogues in both the THQ [48] and the ‘phenoxy’ series [53] were found to be highly potent, although IC50s dropped markedly against plasma renin due to the local lipophilicity of the Psp 3 aryl. Vice versa, polar N-substitution in the ‘indole series’ resulted in the potent inhibitor 33 (purified/plasma renin IC50 ¼ 9/16 nM). Remarkably, the preferred configuration of the P1 isopropyl of 32 and 33 was opposite to that seen in two other lead series exemplified by 30 and 35, requiring elaboration of, at least in part, independent synthesis routes. In view of these hurdles and the lack of apparent benefits, further SAR in the indole series was discontinued. Work was not resumed even when the potential of P1u isopropyl substitution for improved in vivo potency [54] as well as simplified synthetic accessibility (vide infra) became fully recognized for more advanced candidate series. In the quest for more readily accessible inhibitors of less synthetic complexity, a promising series of O-alkylated salicylamides (34, 35; Figure 11) evolved from a computational fragment-based approach starting from the high-affinity compound 30 [57]. Independent docking of the components of 30 generated by disconnecting the spacer bond between P1 and the phenyl moiety suggested a three-atom carboxamide linker as being optimal to re-connect the P3 aryl fragment through its ortho position to P1 of the TSM. Initial SAR resulted in potent and selective inhibitors [20,57], thus supporting the binding mode predicted by modelling. Subsequently, this prediction was confirmed by X-ray [50]. Since the phenyl ring is positioned more deeply into S3, extension of the ortho-alkoxy side chain by one carbon was required for inducing an Hbond with Tyr14 of the Ssp 3 cavity. Strikingly, the in vivo SAR of P2u optimization did not parallel that of other sub-series. For example, 34 (plasma IC50 ¼ 0.5 nM) bearing the same P2u residue as 1 showed an inferior pharmacological profile despite comparable in vitro potencies. The most attractive salicylamide, 35, was highly potent towards primate plasma renin (IC50 ¼ 0.3, 2, 40 and 2,000 nM for human, marmoset, dog and rat) and selective against other human aspartyl proteases. Excellent oral potency was observed in Na-depleted marmosets with a pronounced dose-dependent (1–10 mg/kg) reduction in MAP similar to the effects seen with 1, and complete blockade of PRA over 24 h (3 mg/kg). No effects on the action potential measures in isolated Langendorff rabbit heart were seen at 30 mM, the highest concentration tested, suggesting an attractive
Case History on Tekturnas/Rasilezs (Aliskiren)
121
cardiac safety potential. The failure to identify a crystalline salt of 35 suitable for drug formulation terminated its pre-clinical development.
5. FIRST CONVERGENT AND SCALABLE SYNTHESIS DEVELOPMENT The continued interest in the clinical candidate 1 required a practical stereoselective synthesis for scale-up to support comprehensive evaluation in pharmacology and animal safety studies, as well as subsequent early clinical investigations. The initial linear research route was of high complexity (W25 steps, multiple chromatography purifications) and hence was not suitable to produce larger quantities [54]. Therefore, the drug discovery team in research took the initiative to elaborate a highly efficient, convergent and stereoselective synthesis of 1. Early and tight collaborative integration of the medicinal chemistry and the process chemistry groups was a key factor for effective knowledge transfer, further optimization and eventually successful transition into the first process synthesis. The structure of 1 is unique due to the four acyclic stereocenters, which rendered cost-effective synthesis extremely challenging. Retrosynthetic analysis (Figure 12) revealed the pseudo-symmetric features of the center portion of the all-carbon skeleton, comprising a geminal synamino alcohol as well as a 1,4-diisopropyl substitution motif of identical absolute configuration. We also took advantage of knowledge previously established for CGP38560, 3, enabling the stereoselective preparation of HE isosteres [58]. Intermediate 36 displayed all stereocenters of desired configuration and had the proper functionalities for further elaboration in place. On the basis of a highly stereo-controlled bromo-lactonization/ azide displacement sequence, 36 could be prepared in four steps with each step affording a crystalline intermediate [59]. Further key reactions involved Grignard-coupling of the 3,4-dialkoxy-aryl fragment to the aldehyde derived from 36 to afford 37, followed by direct lactoneopening with the P2u amine, and finally concomitant reduction of the activated benzylic alcohol and the azide through hydrogenation. This convergent route afforded 1 in nine linear steps overall and was MeO
MeO O
O OH
H N
O MeO
NH2
O
Pseudo-symmetric core structure (aliskiren, 1)
OH
O CONH2
O
O
HO2C N3
36
N3
MeO
37
Figure 12 Retrosynthetic concept towards aliskiren, 1, and key intermediates.
122
Ju¨rgen Maibaum and David L. Feldman
amenable to a multi-kilogram scale-up for phase I clinical development. An alternative and elegant process synthesis, again involving a bromolactonization protocol, at acceptable manufacturing cost-of-goods was subsequently developed [60,61].
6. PRE-CLINICAL PROPERTIES OF ALISKIREN Comprehensive reviews on aliskiren, 1, describing its pre-clinical pharmacology and safety profile, its pre-clinical in vitro/in vivo ADME properties, as well as clinical pharmacokinetics have appeared [54–56, 61–65]. Oral bioavailability of 1 was low to moderate in rats, dogs and marmosets (2.4%, 32% and 16–30%) [54,56]. Elimination of 1 in all species involves predominantly biliary/fecal excretion of unchanged parent drug [65]. In vitro studies showed poor substrate-affinity for 1 to CYP3A4 (Km ¼ 24.3 mM), no significant interactions with a large panel of human cytochrome P450 isozymes (e.g., IC50s W200 mM for CYPs 1A2, 2C8, 2C9, 2C19, 2D6, 2E1) [65,66] and low hepatic metabolic clearance in human liver microsomes. Aliskiren is a relatively high-affinity substrate for P-glycoprotein (MDR-1)-mediated transport, which may play a role in the hepatobiliary/intestinal excretion of the drug. The drug is not an inhibitor of MDR-1 and only weakly inhibited other transporters (BCRP, OATP2B1, hOCT1) at high concentrations. Further in vivo studies supported a low likelihood of drug–drug interactions through interference of 1 with CYP enzymes or efflux transporters [62,63]. Aliskiren showed an excellent cardiac safety profile in vitro, indicating a very low risk of impacting cardiac repolarization and conduction time and a low risk of torsade de pointes (TdP). The human hERG channel was very weakly inhibited in a cellular patch-clamp assay (IC25 ¼ 671 mM). No effects were observed on the action potential measures in the isolated rabbit heart up to 100 mM, suggesting a lack of interaction of 1 with cardiac ion channels in whole organ tissues at clinically relevant concentrations [64]. In normal rats, aliskiren distributed to the kidney, possibly to tissue structures that are involved in BP control [67]. Moreover, the compound showed a markedly prolonged renal residence time, being present in the kidneys of dTGR even after a 3-week wash-out period [68]. This finding suggested the potential for sustained inhibition of the intra-renal RAAS, as confirmed in preliminary studies in TG(mRen-2)27 rats [69]. Interestingly, 1 appears to incorporate into renin granules in cultured cells, which, when stimulated appropriately, secrete renin that is already inhibited [70]. Taken together, the earlier findings indicate that aliskiren may be able to access intra-renal renin, resulting in enzyme inhibition even before secretion to the extracellular compartment.
Case History on Tekturnas/Rasilezs (Aliskiren)
123
7. EFFECTS OF ALISKIREN IN DISEASE MODELS As indicated previously, dTGR, which express the human genes for renin and angiotensinogen [15], are very responsive to inhibition with a human renin inhibitor. Single oral dosing of 1 to chronically catheterized dTGR resulted in a dose-dependent, rapid and marked lowering of MAP with a long duration of action [71]. The first evidence that 1 exhibits potent cardio-renoprotective effects was determined in dTGR after chronic treatment [16]. Left ventricular hypertrophy associated with hypertension in dTGR was prevented by 1, and this was accompanied by reductions in various markers of inflammation in cardiac tissue. In a mouse model of myocardial infarction following coronary artery ligation, cardiac function was improved and remodeling was inhibited by a dose of 1 that did not lower BP [72]. These findings suggested a benefit in hypertensive diseases in humans. Hypertensive TG(mRen-2)27 rats expressing mouse renin, which efficiently cleaves rat angiotensinogen, develop type 1 diabetes and nephropathy when pre-treated with streptozotocin. In this model of hypertensive diabetic renal damage, 1 showed renoprotective effects as determined by attenuated albuminuria and renal structural changes [67,73]. The reduction of tubulo-interstitial damage to a greater extent in aliskiren vs. ACEi treated diabetic TG(mRen-2)27 rats is of potential importance, as it may indicate tissue protective effects beyond BP control [73]. The RAAS has been recognized as a major contributor to hypercholesterolaemia-induced atherosclerosis. In two models of atherosclerotic mice [74,75], and in Watanabe heritable hyperlipemic rabbits [76], aliskiren greatly attenuated the development of atherosclerosis without affecting plasma total cholesterol levels. Moreover, evidence for the ability of 1 to improve endothelial function was obtained in rabbits [76]. Finally, in a 2-week study in spontaneously hypertensive rats (SHR), co-administration of sub-maximal effective doses of 1 with either valsartan (ARB) or benazepril (ACEi) demonstrated a significant synergistic BP-lowering effect. This effect was attributed to RAAS inhibition at multiple sites [56].
8. CLINICAL STUDIES WITH ALISKIREN Results from an extensive clinical development plan for aliskiren continue to become available [61,77]. This renin inhibitor has a very long half-life (B40 h) and consequently provides 24 h BP control [78], despite its low absolute oral bioavailability of 2.6% in human [62]. Aliskiren consistently shows strong anti-hypertensive activity, at least as effective as ACEi or ARB, in patients with mild-moderate hypertension. The side effect profile is comparable to placebo. It is noteworthy that this
124
Ju¨rgen Maibaum and David L. Feldman
BP-lowering effect is long-lasting after stopping treatment [79], an effect that may have a basis in the renal retention of aliskiren observed in preclinical studies. Moreover, evidence that aliskiren attenuates albuminuria, independently of BP control, in patients with diabetic nephropathy [80], and lowers brain natriuretic peptide (BNP) in patients with congestive heart failure [81] indicates renal and cardio-protective effects of the drug. When combined with valsartan [82], the diuretic hydrochlorothiazide [83] or amlodipine [84], aliskiren showed additional BP lowering effects vs. either monotherapy. Long renal residence time may explain the prolonged anti-hypertensive action of aliskiren in patients.
9. CONCLUSIONS Novel non-peptide topological renin inhibitors have been designed and further optimized based on computational modeling and X-ray crystallography. The unprecedented binding interaction to a non-substrate binding pocket was crucial for the tight enzyme affinity of these TSMs. Elaboration of several major lead series culminated in aliskiren, a remarkably potent and highly selective renin inhibitor. Aliskiren is orally efficacious, safe and well-tolerated in pre-clinical and clinical trials. TEKTURNAs/RASILEZs (aliskiren) received marketing approval in 2007 by the FDA and in Europe as the first direct renin inhibitor for the treatment of hypertension. The extent to which this new mechanism of action will provide improved organ protection will be determined by ongoing clinical trials with this direct renin inhibitor.
REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9]
B. W. Van Tassell and M. Munger, Ann. Pharmacother., 2007, 41, 456. V. J. Dzau, D. W. Burt and R. E. Pratt, Am. J. Physiol., 1988, 255, F563. U. C. Brewster, J. F. Setaro and M. A. Perazella, Am. J. Med. Sci., 2003, 326, 15. D. T. Pals, J. A. Lawson and S. J. Couch, J. Pharmacol. Meth., 1990, 23, 239. J. M. Wood, S. C. Mah, H. P. Baum, M. De Gasparo, F. Cumin, H. Ru¨eger and J. Nussbaumer, J. Pharmacol. Exp. Ther., 1990, 253, 513. M. L. Mangiapane, J. T. MacAndrew, S. S. Ellery, A. H. Smith, I. M. Purcell, E. F. Kleinman, W. F. Holt and W. R. Murphy, Clin. Exper. Hypertens., 1994, 16, 507. H. D. Kleinert, S. H. Rosenberg, W. R. Baker, H. H. Stein, V. Klinghofer, J. Barlow, K. Spina, J. Polakowski, P. Kovar, J. Cohen and J. Denissen, Science, 1992, 257, 1940. J.-P. Clozel and W. Fischli, Arzneim.-Forsch./Drug Res., 1993, 43, 260. A. E. Weber, A. G. Steiner, P. A. Krieter, A. E. Colletti, J. R. Tata, T. A. Halgren, R. G. Ball, J. J. Doyle, T. W. Schorn, R. A. Stearns, R. R. Miller, P. K. S. Siegl, W. J. Greenlee and A. A. Patchett, J. Med. Chem., 1992, 35, 3755.
Case History on Tekturnas/Rasilezs (Aliskiren)
125
[10] J. L. Wessale, H. D. Kleinert, S. V. Calzadilla, P. J. Kovar and S. H. Rosenberg, Am. J. Hypertens., 1993, 6, 514. [11] E. G. McMahon, P.-C. Yang, M. A. Babler, S. E. Bittner, O. D. Suleymanov, K. J. CainJanicki, L. J. Bedell, G. J. Hanson and C. S. Cook, Hypertension, 1995, 26, 95. [12] J. M. Wood, N. Gulati, P. Forgiarini, W. Fuhrer and K. G. Hofbauer, Hypertension, 1985, 7, 797. [13] C. R. Schnell and J. M. Wood, Am. J. Physiol., 1993, 33, H1509. [14] D. N. Mu¨ller, W. Derer and R. Dechend, J. Mol. Med., 2008, 86, 659. [15] J. Bohlender, A. Fukamizu, A. Lippoldt, T. Nomura, R. Dietz, J. Me´nard, K. Murakami, F. C. Luft and D. Ganten, Hypertension, 1997, 29, 428. [16] B. Pilz, E. Shagdarsuren, M. Wellner, A. Fiebeler, R. Dechend, P. Gratze, S. Meiners, D. L. Feldman, R. L. Webb, I. M. Garrelds, A. H. J. Danser, F. C. Luft and D. N. Muller, Hypertension, 2005, 46, 569. [17] W. J. Greenlee, Med. Res. Rev., 1990, 10, 173. [18] S. H. Rosenberg and S. A. Boyd, in Antihypertensive Drugs (eds P. A. Van Zwieten and W. J. Greenlee), Harwood, Amsterdam, The Netherlands, 1997, pp. 77–111. [19] S. H. Rosenberg, Prog. Med. Chem., 1995, 32, 37. [20] C. M. Tice, Ann. Rep. Med. Chem., 2006, 41, 155. [21] J. M. Wood, F. Cumin and J. Maibaum, Pharmacol. Ther., 1994, 61, 325. [22] P. Bu¨hlmayer, A. Caselli, W. Fuhrer, R. Go¨schke, V. Rasetti, H. Ru¨eger, J. L. Stanton, L. Criscione and J. M. Wood, J. Med. Chem., 1988, 31, 1839. [23] J. M. Wood, L. Criscione, M. De Gasparo, P. Bu¨hlmayer, H. Ru¨eger, J. L. Stanton, R. A. Jupp and J. Kay, J. Cardiovasc. Pharmacol., 1989, 14, 221. [24] M. De Gasparo, F. Cumin, J. Nussberger, T. T. Guyenne, J. M. Wood and J. Me´nard, Br. J. Clin. Pharmac., 1989, 27, 587. [25] X. Jeunemaıˆtre, J. Me´nard, J. Nussberger, T. T. Guyene, H. R. Brunner and P. Corvol, Am. J. Hypertens., 1989, 2, 819. [26] A. H. van den Meiracker, P. J. J. Admiraal, F. H. M. Derkx, C. Kleinbloesem, A. J. Man in ‘t Veld, P. Van Brummelen, P. Mulder and M. A. D. H. Schalekamp, J. Hypertens., 1993, 11, 831. [27] P. Coassolo, W. Fischli, J.-P. Clozel and R. C. Chou, Xenobiotica, 1996, 26, 333. [28] C. Weber, H. Birnbo¨ck, J. Leube, I. Kobrin, C. H. Kleinbloesem and P. Van Brummelen, Br. J. Clin. Pharmac., 1993, 36, 547. [29] W. Fischli, J.-P. Clozel, K. E. Amrani, W. Wostl, W. Neidhart, H. Stadler and Q. Branca, Hypertension, 1991, 18, 22. [30] W. F. Richter, B. R. Whitby and R. C. Chou, Xenobiotica, 1996, 26, 243. [31] P.-C. Yang, M. Babler, S. Bittner, O. Suleymanov, M. Perez, K. Cain-Janicki, L. Bedell, G. Hanson, C. Cook, J. Ottinger, J. Baran and E. McMahon, FASEB J., 1994, 8, A882. [32] S. H. Rosenberg, K. P. Spina, S. L. Condon, J. Polakowski, Z. Yao, P. Kovar, H. H. Stein, J. Coehen, J. L. Batlow, V. Klinghofer, D. A. Egan, K. A. Tricarico, T. J. Perun, W. R. Baker and H. D. Kleinert, J. Med. Chem., 1993, 36, 460. [33] S. H. Rosenberg and H. D. Kleinert, Pharm. Biotech., 1998, 11, 7. [34] H. D. Kleinert, Cardiovasc. Drugs Ther., 1995, 9, 645. [35] R. S. Boger, H. N. Glassman, R. Thys, S. K. Gupta, R. L. Hippensteel and H. D. Kleinert, Am. J. Hypertens., 1993, 6, 103A. [36] K. Suguna, E. A. Padlan, A. Eduardo, C. W. Smith, W. D. Carlson and D. R. Davies, Proc. Natl. Acad. Sci. U.S.A., 1987, 84, 7009. [37] B. L. Sibanda, T. Blundell, P. M. Hobart, M. Fogliano, J. S. Bindra, B. W. Dominy and J. M. Chirgwin, FEBS Lett., 1984, 174, 102. [38] N. C. Cohen, in Trends in Medicinal Chemistry’88 (eds H. van der Goot, G. Doma´ny, L. Pallos, and H. Timmerman), Elsevier, Amsterdam, The Netherlands, 1989, pp. 13–28.
126
Ju¨rgen Maibaum and David L. Feldman
[39] A. R. Sielecki, K. Hayakawa, M. Fujinaga, M. E. P. Murphy, M. Fraser, A. K. Muir, C. T. Carilli, J. A. Lewicki, J. D. Baxter and M. N. G. James, Science, 1889, 243, 1346. [40] J. Rahuel, J. P. Priestle and M. G. Gru¨tter, J. Struct. Biol., 1991, 107, 227. [41] C.E. Brotherton-Pleiss, S.R. Newman, L.D. Waterbury and M.S. Schwartzberg, Pept.: Chem. Biol., Proc. Am. Pept. Symp., 12th (1992), Meeting Date 1991, pp. 816–817. [42] M.G. Bock, R.M. DiPardo, B.E. Evans, R.M. Freidinger, W.L. Whitter, L.S. Payne, J. Boger, E.H. Ulm, E.H. Blaine, et al., Pept.: Struct. Funct., Proc. Am. Pept. Symp., 9th, 1985, p751. [43] V. Rasetti, N. C. Cohen, H. Ru¨eger, R. Go¨schke, J. Maibaum, F. Cumin, W. Fuhrer and J. M. Wood, Bioorg. Med. Chem. Lett., 1996, 6, 1589. [44] V. Rasetti, H. Ru¨eger, J. K. Maibaum, R. Mah, M. Gru¨tter and N. C. Cohen, Eur. Pat. EP 702004-A2, 1996, 2. [45] R. Go¨schke, N. C. Cohen, J. M. Wood and J. Maibaum, Bioorg. Med. Chem. Lett., 1997, 7, 2735. [46] M. W. Holladay, F. G. Salituro and D. H. Rich, J. Med. Chem., 1987, 30, 374. [47] D. J. Kempf, J. Org. Chem., 1986, 51, 3921. [48] J. Maibaum, V. Rasetti, H. Ru¨eger, N. C. Cohen, R. Go¨schke, R. Mah, J. Rahuel, M. Gru¨tter, F. Cumin and J. M. Wood, in Medicinal Chemistry: Today and Tomorrow, Proceedings of the AFMC International Medicinal Chemistry Symposium, Tokyo, September 3–8, 1995 (ed. M. Yamazaki), Blackwell Science, United Kingdom, 1997, pp. 155–162. [49] N. C. Cohen, Chem. Biol. Drug Des., 2007, 70, 557. [50] J. Rahuel, V. Rasetti, J. Maibaum, H. Ru¨eger, R. Go¨schke, N. C. Cohen, S. Stutz, F. Cumin, W. Fuhrer, J. M. Wood and M. G. Gru¨tter, Chem. Biol., 2000, 7, 493. [51] S. H. Rosenberg, J. F. Dellaria, D. J. Kempf, C. W. Hutchins, K. W. Woods, R. G. Maki, E. de Lara, K. P. Spina, H. H. Stein, J. Cohen, W. R. Baker, J. J. Plattner, H. D. Kleinert and T. J. Perun, J. Med. Chem., 1990, 33, 1582. [52] S. B. Shuker, P. J. Hajduk, R. B. Meadows and S. W. Fesik, Science, 1996, 274, 1531. [53] R. Go¨schke, S. Stutz, V. Rasetti, N. C. Cohen, J. Rahuel, P. Rigollier, H.-P. Baum, P. Forgiarini, C. R. Schnell, T. Wagner, M. G. Gru¨tter, W. Fuhrer, W. Schilling, F. Cumin, J. M. Wood and J. Maibaum, J. Med. Chem., 2007, 50, 4818. [54] J. Maibaum, S. Stutz, R. Go¨schke, P. Rigollier, Y. Yamaguchi, F. Cumin, J. Rahuel, H.-P. Baum, N. C. Cohen, C. R. Schnell, W. Fuhrer, M. G. Gru¨tter, W. Schilling and J. M. Wood, J. Med. Chem., 2007, 50, 4832. [55] J. M. Wood, J. Maibaum, J. Rahuel, M. G. Gru¨tter, N. C. Cohen, V. Rasetti, H. Ru¨eger, R. Go¨schke, S. Stutz, W. Fuhrer, W. Schilling, P. Rigollier, Y. Yamaguchi, F. Cumin, H.-P. Baum, C. R. Schnell, P. Herold, R. Mah, C. Jensen, E. O’Brien, A. Stanton and M. P. Bedigian, Biochem. Biophys. Res. Commun., 2003, 308, 698. [56] J. M. Wood, C. R. Schnell, F. Cumin, J. Menard and R. L. Webb, J. Hypertens., 2005, 23, 417. [57] J. Maibaum, N.C. Cohen, J. Rahuel, C. Schnell, H.-P. Baum, P. Rigollier, W. Schilling and J.M. Wood, XVth EFMC International Symposium on Medicinal Chemistry, Edinburgh, United Kingdom, 6–10 September 1998, Abstract Book, P229. [58] P. Herold, R. Duthaler, G. Rihs and C. Angst, J. Org. Chem., 1989, 54, 1178. [59] R. Go¨schke, J.K. Maibaum, W. Schilling, S. Stutz, P. Rigollier, Y. Yamaguchi, N.C. Cohen, P. Herold, Eur Patent EP 678503-A1, 1995. [60] J. Maibaum and D. L. Feldman, Expert. Opin. Ther. Patents, 2003, 13, 589. [61] C. Jensen, P. Herold and H. R. Brunner, Nat. Rev. Drug Discov., 2008, 7, 399. [62] S. Vaidyanathan, V. Jarugula, H. A. Dieterich, D. Howard and W. P. Dole, Clin. Pharmacokinet., 2008, 47, 515. [63] F. Waldmeier, U. Glaenzel, B. Wirz, L. Oberer, D. Schmid, M. Seiberling, J. Valencia, G. J. Riviere, P. End and S. Vaidyanathan, Drug Metab. Dispos., 2007, 35, 1414. [64] S. Ayalasomayajula, C.-M. Yeh, S. Vaidyanathan, B. Flannery, H. A. Dieterich, D. Howard, M. P. Bedigian and W. P. Dole, J. Clin. Pharmacol., 2008, 48, 799.
Case History on Tekturnas/Rasilezs (Aliskiren)
127
[65] S. Vaidyanathan, G. Camenisch, H. Schuetz, C. Reynolds, C.-M. Yeh, M.-N. Bizot, H. A. Dieterich, D. Howard and W. P. Dole, J. Clin. Pharmacol., 2008, 48, 1323. [66] S. Vaidyanathan, Y. Jin, H. Schiller and C. Jensen, Basic Res. Pharmacol. Toxicol., 2005, 97(Suppl 1), 230. [67] D. L. Feldman, L. Jin, H. Xuan, A. Contrepas, Y. Zhou, R. L. Webb, D. N. Mueller, S. Feldt, F. Cumin, W. Maniara, E. Persohn, H. Schuetz, A. H. J. Danser and G. Nguyen, Hypertension, 2008, 52, 130. [68] D. L. Feldman, E. Persohn, H. Schuetz, L. Jin, R. Miserindino-Moltini, H. Xuan, S. Zhuang and W. Zhou, J. Clin. Hypertens., 2006, 8, A80–A81, P-178. ˇ ervenka, L. Kopkan, Z. Vanˇourkova´ and Z. Huskova´, J. Hypertens., [69] I. Vaneˇcˇkova´, L. C 2008, 26, S521. [70] M. Krop, I. M. Garrelds, R. J. de Bruin, J. M. van Gool, N. D. Fisher, N. K. Hollenberg and A. H. J. Danser, Hypertension, 2008, 52, 1076. [71] D. F. Rigel, F. Fu, S. Li, W. Maniara, D. N. Mu¨ller and F. C. Luft, J. Am. Coll. Cardiol., 2004, 43(Suppl. A), 483A. [72] D. Westermann, A. Riad, O. Lettau, A. Roks, K. Savvatis, P. M. Becher, F. Escher, A. H. J. Danser, H. P. Schultheiss and C. Tschoepe, Hypertension, 2008, 52, 1068. [73] D. J. Kelly, Y. Zhang, G. Moe, G. Naik and R. E. Gilbert, Diabetologia, 2007, 50, 2398. [74] J. Nussberger, J. F. Aubert, K. Bouzourene, M. Pellegrin, D. Hayoz and L. Mazzolai, Hypertension, 2008, 51, 1306. [75] H. Lu, D. L. Rateri, D. L. Feldman, R. J. Jr., A. Fukamizu, J. Ishida, E. G. Oesterling, L. A. Cassis and A. Daugherty, J. Clin. Invest., 2008, 118, 984. [76] T. Imanishi, H. Tsujioka, H. Ikejima, A. Kuroi, S. Takarada, H. Kitabata, T. Tanimoto, Y. Muragaki, S. Mochizuki, M. Goto, K. Yoshida and T. Akasaka, Hypertension, 2008, 52, 563. [77] H. Siragy, J. Huang and D. C. Lieb, Expert Opin. Emerg. Drugs., 2008, 13, 417. [78] A. H. Gradman and R. Kad, J. Am. Coll. Cardiol., 2008, 51, 519. [79] B. H. Oh, J. Mitchell, J. R. Herron, J. Chung, M. Khan and D. L. Keefe, J. Am. Coll. Cardiol., 2007, 49, 1157. [80] H. H. Parving, F. Persson, J. B. Lewis, E. J. Lewis and N. K. Hollenberg, N. Engl. J. Med., 2008, 358, 2433. [81] A. Seed, R. Gardner, J. McMurray, C. Hillier, D. Murdoch, R. MacFadyen, A. Bobillier, J. Mann and T. McDonagh, Eur. J. Heart Fail., 2007, 9, 1120. [82] S. Oparil, S. A. Yarows, S. Patel, H. Fang, J. Zhang and A. Satlin, Lancet, 2007, 370, 221. [83] A. Villamil, S. G. Chrysant, D. Calhoun, B. Schober, H. Hsu, L. Matrisciano-Dimichino and J. Zhang, J. Hypertens., 2007, 25, 217. [84] M. A. Munger, W. Drummond, M. R. Essop, M. Maboudian, M. Khan and D. L. Keefe, Eur. Heart J., 2006, 27(Suppl), 117P–784.
CHAPT ER
6 Advances in Vasopressin Receptor Agonists and Antagonists Thomas Ryckmans
Contents
1. Introduction 1.1 Emerging roles of V1a and V1b receptors in anxiety, depression and sociality 1.2 Challenges in the development of small-molecule peptidergic GPCR modulators. 2. V1a Receptor Antagonists 2.1 Azepines 2.2 Lactams 2.3 Spirocyclic piperidines 3. V1b Receptor Antagonists 3.1 Indolones and benzimidazolones 3.2 Lactams 3.3 Quinazolinones and isoquinolinones 3.4 V1b receptors and pain 4. V2 Receptor Agonists 4.1 Azepines and diazepines 4.2 Emerging chemotypes 5. V2 Receptor Antagonists 6. Dual V1a-V2 Receptor Antagonists 6.1 Azepines 6.2 Triazolones 7. Summary References
129 130 131 131 132 132 133 134 134 136 137 138 139 139 140 140 141 142 143 143 144
Pfizer PGRD, Sandwich, Kent, UK Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04406-6
r 2009 Elsevier Inc. All rights reserved.
129
130
Thomas Ryckmans
1. INTRODUCTION Arginine vasopressin (AVP) is a 9–amino-acid cyclic peptide that exerts its effect through the V1a, V1b, V2 and oxytocin (OT) family of G-proteincoupled receptors (GPCRs). V1a receptors are expressed mainly in the liver (glycogenolysis regulation), the vascular smooth muscle cells (vasoconstriction and contractility regulation) and the brain (regulation of memory formation, stress adaptation, temperature and circadian rhythm). These receptors are also involved in platelet aggregation [1]. Accordingly, V1a receptor antagonists have been progressed to the clinic for a range of indications, including Raynaud’s syndrome (vasoconstriction) and dysmenorrhea (uterine blood flow and contraction). V1b receptors (also known as V3 receptors) are expressed in the limbic system and the pituitary gland, where they regulate the hypothalamus-pituitaryadrenal (HPA) axis, which is responsible for response to stress. The V1b receptor antagonist SSR149415 has been progressed to phase II clinical trials for depressive disorders. V2 receptors are mainly expressed in the kidney and are involved in the regulation of water and sodium excretion [2–5], and several V2 receptor antagonists and mixed V1a-V2 receptor antagonists have been progressed to the clinic. It has been suggested that there is a distinct advantage in blocking both V1a and V2 receptors for indications such as congestive heart failure [6]. This review covers recent advances in the development of non-peptidic vasopressin receptor ligands, focusing on new series, structure–activity relationship (SAR) and preclinical data.
1.1 Emerging roles of V1a and V1b receptors in anxiety, depression and sociality There is growing evidence that AVP, through its action at V1a and V1b receptors, is involved in the modulation of several higher brain functions such as response to stress, mood, memory formation, aggressivity and sociality. AVP and corticotropin-releasing factor (CRF) synergistically modulate the release of adrenocorticotropic hormone (ACTH) by the pituitary gland, a key step in the HPA axis cascade, which controls response to stress. Although a detailed description of the HPA axis is beyond the scope of this review, the reader is referred to recent publications on the subject [3,7,8]. HPA axis dysfunction has been linked to anxiety, posttraumatic stress disorder (PTSD) and depression [7,9,10], and both preclinical and clinical evidence point at a key role of AVP in these conditions [10]. For example, rats bred for high anxiety-related behaviour have higher levels of AVP mRNA than rats bred for low anxiety-related behaviour. The Brattleboro rat strain, in which AVP secretion is impaired,
Advances in Vasopressin Receptor Agonists and Antagonists
131
displays reduced stress response at both the behavioural and the endocrine levels [3]. Clinical evidence has shown that both depressed and PTSD patients have higher levels of systemic AVP and that single-nucleotide polymorphism of the V1b receptor is correlated with a lower probability of developing major depression [11]. Both V1a and V1b receptors have been implicated in the modulation of aggression in rodents [12–14]. AVP has emerged as a key regulator of complex social behaviours [13,15–17]. Pair bonding in voles is modulated by the action of AVP on the V1a receptor [18–20], and a recent study demonstrated the association of genetic variation of the V1a receptor with pair bonding in humans [21]. OT has also been implicated in such processes, but this is beyond the scope of this review [22–25]. The possible implication of AVP dysfunction in conditions where sociality is impaired, such as autism and schizophrenia, has been put forward, but more evidence is clearly needed [2,26,27]. A recent review of the patent literature covering the anxiolytic and antidepressant potential of AVP antagonists is available [28], and we have therefore focused on the more recent data in the field.
1.2 Challenges in the development of small-molecule peptidergic GPCR modulators. The development of small-molecule ligands of peptidergic receptors (with the exception of opioid receptor ligands) is still considered a challenge by medicinal chemists. The average molecular weight, total polar surface area (TPSA) and cLogP of peptidergic ligands are significantly higher than those of ligands of other targets such aminergic GPCRs, ion channels and transporters [29]. Since high molecular weight, high TPSA and high cLogP are associated with poor drug-like properties [30], increased risk of offtarget toxicity [31] and poor CNS penetration, small-molecule peptidergic GPCR ligands have suffered from high attrition between the bench and the clinic. It is therefore not surprising that despite significant efforts from the pharmaceutical industry, a relatively small number of AVP antagonists have been approved for marketing.
2. V1A RECEPTOR ANTAGONISTS Relcovaptan (SR49059, 1) has shown clinical efficacy for indications modulated by the peripheral V1a receptor including Raynaud’s syndrome, dysmenorrhea and preterm labour [5], but its development has been interrupted as a result of possible interference with cytochrome P450 [32]. The development of OPC-21268 (2), another V1a receptor antagonist, has been discontinued in Europe and in the United States
132
Thomas Ryckmans
[33]. Nevertheless, the positive proof-of-concept studies with relcovaptan have triggered the interest of the pharmaceutical industry, resulting in the discovery of several classes of V1a receptor antagonists. Critically, a biomarker for V1a receptor antagonism in healthy volunteers has been recently developed [34], which could facilitate clinical development.
Cl
O OH O
O
N
O
O NH2
N
S
N N
O
O
O
O N H O
2
1
O
2.1 Azepines The activity of JNJ-17308616 (3) at V1a, V1b, V2 and OT human, rat and mouse receptors has been reported. Although JNJ-17308616 is a potent and selective antagonist of human V1a receptors (hV1a Ki ¼ 5 nM, hV1b KiW10 mM, hV2 Ki ¼ 421 nM, hOT Ki ¼ 6.5 mM), its affinity and selectivity for rat V1a receptors is lower (rV1a Ki ¼ 216 nM, rV1b KiW10 mM, rV2 Ki ¼ 276 nM, rOT Ki ¼ 1.8 mM). The compound was found to be active in several rat models of anxiety [35]. Since the kidney-localized rat V2 receptors are not believed to be associated with anxiety, the anxiolytic effect of JNJ-17308616 appears to be mediated by the V1a receptor. O
H N
O
O
O
N
O N
N O O
N
N N
O
N
N
O
F
N O
F
O
N H
F
O F
N
N
N H
O
3
4
5
133
Advances in Vasopressin Receptor Agonists and Antagonists
2.2 Lactams SRX251 (4) is a potent and selective V1a receptor antagonist (hV1a Ki ¼ 0.7 nM, hV1b KiW1 mM). The compound is orally bioavailable in rats (t1/2 ¼ 4 h), is claimed to be CNS penetrant and inhibited aggression in a hamster model when administered orally at a dose of 20 mg/kg [28,36,37]. Interestingly, non-aggressive behaviours such as locomotor activity, olfactory communication and sexual motivation were not affected. The ability of SRX251 to selectively block aggression in rats was confirmed by functional magnetic resonance imaging [12]. SRX251 has completed phase I trials and is scheduled for a phase II study targeting dysmenorrhea [38]. SRX246 (5) has a profile similar to that of SRX251 (hV1a Ki ¼ 0.3 nM, hV1b KiW10 mM, rat t1/2 ¼ 2 h) and an Investigational New Drug filing for stress-related disorders is pending [38].
2.3 Spirocyclic piperidines A range of small, low lipophilicity, selective and potent triazoles were disclosed as V1a receptor antagonists, exemplified by 6 (hV1a Ki ¼ 7 nM) [39,40]. In a related chemotype in which the triazole is replaced by an amide (as in OPC-21268), piperidines such as 7 and 8 were shown to be moderately potent V1a antagonists (hV1a Ki ¼ 25 and 31 nM respectively), but no selectivity data was disclosed.
N N
O
O
N
H N
N
N
N O
O
S
N
O
Cl Cl
6
7
8
Several patents describe related amides, with variations of the spirocyclic and heteroaryl moieties, and combinations thereof [41–47]. Spirocyclic groups and aryl substituents are preferred at the piperidine 4-position, whereas glycine amides and 2- or 3-indolecarboxamides are preferred derivatives of the piperidine nitrogen. Compounds 9–13 (hV1a Kio10 nM) are examples of such combinations.
134
Thomas Ryckmans O
O
H N
N
Ph
O
N
O
O
N HN
HN
Cl
Cl
O N
Cl
N
N
HN
O
O
O
O
O
O
O
11
10
9
Cl
N
N
O N
Cl
N
O
HN
O N 12
13
F O
3. V1B RECEPTOR ANTAGONISTS 3.1 Indolones and benzimidazolones SSR149415 (14), a potent, selective V1b receptor antagonist was progressed to phase II trials for major depression [48,49] but appears to have been discontinued [50]. The compound was extensively studied in a range of anxiety and depression models [3]. For example, SSR149415 was recently demonstrated to have efficacy similar to that of imipramine in the rat olfactory bulbectomy-induced (OBX) hyperactivity depression model [35]. The compound (at 10 and 30 mg/kg i.p.) completely inhibited OBXinduced hyperactivity upon chronic and subchronic but not acute dosing. Although the mechanism of these effects of SSR149415 is unclear, it does increase norepinephrine (but not serotonin or dopamine) levels in the prefrontal cortex and appears to act as an antidepressant rather than an anxiolytic [51]. The V1b vs. OT receptor selectivity of SSR149415 has been discussed [52], but a recent report on the ex vivo binding of radiolabelled SSR149415 appears to confirm the selectivity of this compound for V1b receptors [53]. Pituitary V1b receptors have been
135
Advances in Vasopressin Receptor Agonists and Antagonists
suggested to play a role in the anxiolytic effect of SSR149415, while the antidepressant effects appear to be mediated through central V1b receptors [7]. OH
R3
O R1
N
Cl
R3
N
N O
N O
S
R1 O
O N
R4
R4
N
S R2
O
S
O
O
R2
O
O
O
15
14
O
16
R1 = F, Cl, CN R2 = 4-CN or 4-OMe or 2-OMe, 4-OMe-phenyl or 8-quinolyl R3 = aromatic or heteroaromatic R4 = most variable group
O
Since the disclosure of SSR149415, a limited number of V1b receptor antagonists that can be distributed into three classes have appeared in the patent literature. Although indolones 15 and benzimidazolones 16 are both related to SSR149415 with clear structural overlap, new classes of compounds with beta-lactam and quinazolinone cores have also been reported. In a series of indolones related to SSR149415, 17–19 exemplify the substitution pattern with variations of R3, R2 and R4. Although urea 17 is reported to be a potent and selective V1b receptor antagonist (hV1b Kio10 nM, W50- and W100-fold selectivity over V1a and OT receptors, respectively), no selectivity data has been published for urea 18 (hV1b Kio10 nM) and carbamate 19 (hV1b Kio50 nM) [54–57].
N
O
O O
H N
N
N O
N
N N
O O
O N O O N
N
O N
17
18
N N
S
O O
N
O
N N
S
O
O N
O N
S
O O
H N
19
N
136
Thomas Ryckmans
Related high-affinity analogues (hV1b Kio1 nM) in which the R4 group is a cyclic amine have also been disclosed, such as 20 and 21 [56,58,59].
N
O
N O
O
( )n N
N
R1
O
N
O N S O O
O N S O O O
O
N
N
O
O
N
N
20
Cl
N
N
O
N
N S
21
O
(R1=F, Cl, n=1-2)
O
22
O
Compound 22, illustrating the benzimidazolone chemotype, has been reported as a V1b receptor antagonist (hV1b Kio100 nM), but no selectivity data was reported [60].
3.2 Lactams Beta-lactams 23–24 represent a new class of V1b receptor antagonists (23: rV1b Ki ¼ 70 rM, 24: rV1b Ki ¼ 30 nM; no selectivity data against the V1a receptor reported). Compound 24 was found to restore both biochemical (testosterone/cortisol levels) and behavioural (seed finding) markers in hamster at the dose of 1 mg/kg i.p. with activity comparable to that of fluoxetine, buspirone and chlordiazepoxide [61]. Interestingly, this chemotype is related to a previously discussed class of selective V1a receptor antagonists represented by 4 and 5.
O
O
O
O N
N
O
O
N
N O
O
O
N H
N H O N N
23
N N
24
137
Advances in Vasopressin Receptor Agonists and Antagonists
3.3 Quinazolinones and isoquinolinones Several V1b receptor antagonists from the chemotype 25 have been disclosed. Binding of selected compounds in CHO cells expressing hV1b receptors or in an ex vivo rat anterior pituitary cell assay was reported. Quinazolinones 26 and 27 are potent antagonists of the human receptor (hV1b Kio10 nM). Compound 28 (Y ¼ CH) is described as a rat V1b receptor antagonist (V1b Ki ¼ 10–100 nM) [62]. A very similar series of azaanalogues exemplified by 28 (Y ¼ N) has been reported (hV1b IC50 1 nM) [63]. R3
H N
H N
O
O
O
O R
2
O
N
N N X
R1
N N X=N, CH R2=OR, Het R3=iPr, tBu, nPr
25
H N
O
26
H N
O
O O
O
O
O N
N
N
N N
N
Cl
27
O
28
Y Y=CH Y=N
A potent and selective V1b receptor antagonist ORG 52186 (hV1b Ki ¼ 4 nM; hV1a, V2 and OTW10 mM) is reported to be active when administered at a dose of 5 mg/kg p.o. in a rat model of CRF-mediated release of ACTH [64]. The structure of this compound has not been disclosed, but it may belong to the quinazolinone class [8,33,62]. Compounds related to 25 (X ¼ N) with an R2 nitrogen substituent have been disclosed, exemplified by 29 (hV1b Kio10 nM) and 30 (hV1b Ki ¼ 10–100 nM), which demonstrates that an aromatic group at the 6-position is not required to achieve potency. Quinazolinone 31 (rV1b Ki ¼ 10–100 nM, ex vivo anterior pituitary cells) is disclosed as a rat V1b receptor antagonist [65]. Isoquinolinone analogues 25 (X ¼ CH) have been disclosed with a similar substitution pattern; 32 is reported to be an antagonist of human V1b receptors with a Kio10 nM.
138
Thomas Ryckmans
H N
H N
O O
O O
NH N
N
N
N
29
Cl
H N
NH N
N
30
H N
O
O O
O
N
O
N
N
N
N N
31
O
32
3.4 V1b receptors and pain Compounds 33 and 34 were shown to be active in the Chung model of neuropathic pain in rats at a dose of 10 mg/kg p.o. In the same model, SSR149415 was active at a comparable dose (6 mg/kg). Compound 34 was also active in a model of chronic inflammatory pain in rats (CFA) at 10 mg/kg p.o. [66]. Neither 33 nor 34 was active in the mouse tail-flick model of acute thermal pain. Interestingly, V1b receptor knockout mice (but not V1a receptor knockout mice) showed a hyponociceptive response in the hot plate model [67]. SSR149415 (14) was found to be efficacious in a rat model of chronic stress-induced visceral hyperalgesia, but not in an acute model. Taken together, these results indicate the potential of V1b receptor antagonists for the treatment of pain and the effect of stress on pain response.
H N
H N
O
O O
O
O N
N
N
N
N
N F
33 O
Cl
34
Advances in Vasopressin Receptor Agonists and Antagonists
139
4. V2 RECEPTOR AGONISTS The V2 receptor is mainly localized in the kidney and is responsible for fluid homeostasis. V2 receptor agonists increase the expression of aquaporin-2 water channels, which increase water reabsorption by the kidney, thus concentrating the urine (increased osmolality) and decreasing its volume. Inactivating mutations of the V2 receptor are linked to nephrogenic diabetes insipidus, in which polyuria and excessive thirst is observed. Desmopressin, a V2 receptor agonist, is indicated for the treatment of polyuria, but since the compound is also a V1b receptor agonist, and has low bioavailability, more selective agents with improved oral bioavailability have been actively sought [68].
4.1 Azepines and diazepines OPC-51803 (35), a partial V2 receptor agonist, and VNA-932 (WAY151932, 36) are both in phase II clinical trials for enuresia [33]. O N H N
N
N O
O
Cl
Cl N 35
N
N
36
Diazepine 37 was active (10 mg/kg p.o.) in a rat model of polyuria, decreasing urine volume by nearly 90%. The compound was selective for V2 receptors (rV2 binding 490 nM, rV2 EC50 1.7 nM) over V1a receptors (rV1a binding 6 mM, ex vivo blockade of AVP-induced contractions of rat tail artery IC50W30 mM) and was found to be a weak antagonist of rOT receptors (ex vivo blockade of OT-induced contractions of rat uterine strips IC50W1260 nM) [69]. The structurally related azepine 38 (hV2 Ki ¼ 5 nM, rV2 Ki ¼ 19 nM, rEC50 2 nM) was also active in a similar rat model when administered at the dose of 0.26 mg/kg p.o. [70]. Diazepine 39 (hV2 EC50 4 nM, intrinsic efficacy relative to AVP 0.85) and its analogues are also potent V2 receptor agonists [71]. Compound 40 was
140
Thomas Ryckmans
recently reported to be a potent full V2 receptor agonist (hV2 EC50 24 nM, intrinsic efficacy relative to AVP 1.0; V1a, V1b EC50W10 mM; OT KiW10 mM) and to show robust efficacy in the Brattleboro rat model (3 mg/kg p.o.), with almost full inhibition of urinary output for 2 h [72]. N HN H N
O
N
N
F F
O N
O N
N
O
N O
O
H N
O
F
N
Cl
F
O
Cl
N
F
O
N
N
37
38
O
N
39
N
40
4.2 Emerging chemotypes Phthalamide 41 (pEC50 9.4 intrinsic efficacy relative to AVP 0.27) and tetrazole 42 (pEC50 5.1 intrinsic efficacy relative to AVP 0.68) were identified as selective, partial agonists of the human V2 receptor. Compound 41 (10 mg/kg, p.o.) showed efficacy in the Brattleboro rat model, reducing urine output by 60% [68].
F HO
F F
O
N
O
N N N
N N
41
42
5. V2 RECEPTOR ANTAGONISTS Several V2 receptor antagonists, including tolvaptan 43, lixivaptan 44 and satavaptan 45, have entered clinical trials for the treatment of hyponatremia, liver cirrhosis, chronic heart failure and polycystic kidney disease and have been extensively reviewed elsewhere [5,33].
141
Advances in Vasopressin Receptor Agonists and Antagonists
Mozavaptan 46 was approved in Japan in 2006 for hyponatremia and tolvaptan was recommended for approval by an FDA advisory board for hyponatremia in June 2008 [5]. In a clinical trial aimed at evaluating the efficacy of tolvaptan in preventing heart failure, tolvaptan modestly relieved the congestive symptoms of heart failure, but failed to reduce morbidity or mortality when added to standard therapy [73]. O HO
O
N
N
N O N
O
N O
O S O
O N H
N H
O
F
Cl
O
N
N
O
O
NH
O O
HN
43
44
45
46
Compounds 47 and 48 related to satavaptan have been reported recently (47 hV2 Ki ¼ 1 nM, 48 hV2 Ki ¼ 2 nM; selectivity over OT, V1a and V1b W100 for both) but no in vivo data was disclosed [74]. In yet another variation of the beta-lactam chemotype, compounds such as 49 were reported as potent V2 receptor antagonists (hV2 Ki ¼ 0.4 nM). Activity of such compounds at the OT and neurokinin receptors was also claimed in this disclosure [75]. O Cl
N H
Cl
O
O NH
O
Cl
N
O
O
O N
O N
O
S
O O O
HN
HN O
47
N H
O
S
O
N O
N N
O
48
F
F F
49
6. DUAL V1A-V2 RECEPTOR ANTAGONISTS Conivaptan (YM-087, 50) is currently the only approved V1a-V2 receptor antagonist, and is used to treat hyponatremia. Chronic blockade of both the V1a and the V2 receptors may be advantageous in the treatment of
142
Thomas Ryckmans
congestive heart failure [6] and has prompted vigorous research in this area.
6.1 Azepines
O HN
H N
O
OH
N
N
N
N
N
O
F
O
O
Cl N H
N H
O
50
O
N H
O
51
O
52
Spirobenzazepine RWJ-339489 (51) was identified as a potent, balanced functional V1a-V2 receptor antagonist (hV1a IC50 45 nM, hV2 IC50 36 nM) with adequate rat PK (F ¼ 22%, t1/2 ¼ 6.5 h) and produced robust aquaresis when administered at the dose of 10 mg/kg p.o. [76]. Because of unexpected toxicology findings with 51, RWJ-676070 (52) was identified as a back-up candidate. Compound 52 has a functional antagonist profile at V1a and V2 receptors similar to that of 51 (hV1a IC50 14 nM, hV2 IC50 13 nM), has a favourable rat pharmacokinetic profile (F ¼ 68%, t1/2 ¼ 8.7 h) and did not display the toxicity of 51 [1,77]. The compound entered phase I clinical trials and was well-tolerated in healthy volunteers [34]. O H N
N O
HN
N
N
O
N
O
O F
N H
53
N H
O
54
N H
O
55
O
Advances in Vasopressin Receptor Agonists and Antagonists
143
Another spirobenzazepine (53) was reported to be a potent dual functional antagonist of V1a and V2 receptors (hV1a IC50 29 nM, hV2 IC50 64 nM) [78]. The subtle SAR of the balance between V1a and V2 receptor antagonism is illustrated by lactams 54 (hV1a IC50 50 nM, hV2 IC50 60 nM) and 55 (hV1a IC50 50 nM, hV2 IC50 W 3 mM) [79].
6.2 Triazolones A new series of triazolones exemplified by 56, 57 and 58 has been reported. These analogues range from selective V1a or V2 receptor antagonists or are mixed V1a-V2 receptor antagonists. For example, although 56 (hV1a IC50 42 nM, hV2 IC50 12 nM) is a balanced dual antagonist, the fluoro analogue 57 (hV1a IC50 960 nM, hV2 IC50 6 nM) is a V2 receptor antagonist with 160-fold selectivity over V1a receptors. Compound 58 (hV1a IC50 40 nM, hV2 IC50 W10 mM) is a selective V1a receptor antagonist [80]. O
R
O
O
O N
N
N
HN
N
N
HN
F F F
N
F F S
F
56 R=OMe 57 R=F
58
Cl
Cl
7. SUMMARY The historical development and the complexity of AVP pharmacology is clearly captured by the title of a recent review ‘‘The vasopressin system – from antidiuresis to psychopathology’’ [2]. V2 receptor antagonists and agonists are being evaluated in the clinic for the modulation of renal function and cardiovascular conditions. V1a receptor antagonists are being actively investigated for indications such as dysmenorrhea, given the unmet medical need and the large patient population. On the contrary, the involvement of the V1a and V1b receptors in depression, anxiety and stress disorders has been clearly demonstrated in preclinical models; however, the discovery of an orally bioavailable, well-tolerated
144
Thomas Ryckmans
and CNS-penetrant antagonist of these receptors is still a formidable challenge. To date, no V1a or V1b receptor antagonists have been approved, but the growing evidence of the key role of AVP in mood disorders is leading to renewed interest in this venue of research. The fascinating role of AVP in higher brain function and behaviour, such as sociality and bonding, has emerged in recent years, suggesting potential new indications for V1a and V1b receptor antagonists.
REFERENCES [1] J. W. Gunnet, P. Wines, M. Xiang, P. Rybczynski, P. Andrade-Gordon, L. de Garavilla, T. J. Parry, W.-M. Cheung, L. Minor, K. T. Demarest, B. E. Maryanoff and B. P. Damiano, Eur. J. Pharmacol., 2008, 590, 333. [2] E. Frank and R. Landgraf, Eur. J. Pharmacol., 2008, 583, 226. [3] A. Surget and C. Belzung, Eur. J. Pharmacol., 2008, 583, 340. [4] F. Ali, M. Guglin, P. Vaitkevicius and J. K. Ghali, Drugs, 2007, 67, 847. [5] G. Decaux, A. Soupart and G. Vassart, Lancet, 2008, 371, 1624. [6] S. R. Goldsmith, Am. J. Med., 2006, 119, S93. [7] F. Thomson and M. Craighead, Neurochem. Res., 2008, 33, 691. [8] J. Roper, A.-M. O’Carroll1, E. J. Grant and S. J. Lolait, Proc. Physiol. Soc., 2007, 5, C3. [9] B. E. Leonard, Drugs Today, 2007, 43, 705. [10] L. Q. Stewart, J. A. Roper, W. Scott Young, A.-M. O’Carroll and S. J. Lolait, Psychoneuroendocrinology, 2008, 33, 405. [11] D. van West, J. Del-Favero, Y. Aulchenko, P. Oswald, D. Souery, T. Forsgren, S. Sluijs, S. Bel-Kacem, R. Adolfsson, J. Mendlewicz, C. Van Duijn, D. Deboutte, C. Van Broeckhoven and S. Claes, Mol. Psychiatry, 2004, 9, 287. [12] C. F. Ferris, T. Stolberg, P. Kulkarni, M. Murugavel, R. Blanchard, D. C. Blanchard, M. Febo, M. Brevard and N. G. Simon, BMC Neurosci., 2008, 9, 111. [13] S. R. Wersinger, H. K. Caldwell, M. Christiansen and W. S. Young, Jr., Genes Brain Behav., 2007, 6, 653. [14] S. R. Wersinger, J. L. Temple, H. K. Caldwell and W. S. Young, Jr., Endocrinology, 2008, 149, 116. [15] A. H. Veenema and I. D. Neumann, Prog. Brain Res., 2008, 170, 261. [16] O. J. Bosch and I. D. Neumann, Proc. Natl. Acad. Sci. U.S.A., 2008, 105, 17139. [17] B. C. Nephew and R. S. Bridges, Pharmacol. Biochem. Behav., 2008, 91, 77. [18] J. T. Winslow, N. Hastings, C. S. Carter, C. R. Harbaugh and T. R. Insel, Nature, 1993, 365, 545. [19] M. M. Lim, Z. Wang, D. E. Olazabal, X. Ren, E. F. Terwilliger and L. J. Young, Nature, 2004, 429, 754. [20] M. M. Lim, E. A. D. Hammock and L. J. Young, J. Neuroendocrinol., 2004, 16, 325. [21] H. Walum, L. Westberg, S. Henningsson, J. M. Neiderhiser, D. Reiss, W. Igl, J. M. Ganiban, E. L. Spotts, N. L. Pedersen, E. Eriksson and P. Lichtenstein, Proc. Natl. Acad. Sci. U.S.A., 2008, 105, 14153. [22] T. Baumgartner, M. Heinrichs, A. Vonlanthen, U. Fischbacher and E. Fehr, Neuron, 2008, 58, 639. [23] M. R. Delgado, Neuron, 2008, 58, 470. [24] A. J. Guastella, P. B. Mitchell and M. R. Dadds, Biol. Psychiatry, 2008, 63, 3. [25] P. Kirsch, C. Esslinger, Q. Chen, D. Mier, S. Lis, S. Siddhanti, H. Gruppe, V. S. Mattay, B. Gallhofer and A. Meyer-Lindenberg, J. Neurosci., 2005, 25, 11489.
Advances in Vasopressin Receptor Agonists and Antagonists
145
[26] H. K. Caldwell, H.-J. Lee, A. H. Macbeth and W. S. Young, Prog. Neurobiol., 2008, 84, 1. [27] E. A. D. Hammock and L. J. Young, Philos. Trans. R. Soc., B, 2006, 361, 2187. [28] N. G. Simon, C. Guillon, K. Fabio, N. D. Heindel, S.-f. Lu, M. Miller, C. F. Ferris, M. J. Brownstein, C. Garripa and G. A. Koppel, Recent Pat. CNS Drug Discov., 2008, 3, 77. [29] R. Morphy, J. Med. Chem., 2006, 49, 2969. [30] P. D. Leeson and B. Springthorpe, Nat. Rev. Drug Discov., 2007, 6, 881. [31] J. D. Hughes, J. Blagg, D. A. Price, S. Bailey, G. A. DeCrescenzo, R. V. Devraj, E. Ellsworth, Y. M. Fobian, M. E. Gibbs, R. W. Gilles, N. Greene, E. Huang, T. Krieger-Burke, J. Loesel, T. Wager, L. Whiteley and Y. Zhang, Bioorg. Med. Chem. Lett., 2008, 18, 4872. [32] V. Bernier, J.-P. Morello, A. Zarruk, N. Debrand, A. Salahpour, M. Lonergan, M.-F. Arthus, A. Laperrier, R. Brouard, M. Bouvier and D. G. Bichet, J. Am. Soc. Nephrol., 2005, 17, 232. [33] M. Manning, S. Stoev, B. Chini, T. Durroux, B. Mouillac and G. Guillon, Prog. Brain Res., 2008, 170, 473. [34] L. Coltamai, M. Bucher, M. P. Maillard, U. Shukla, N. Bohidar, L. Haskell, K. Bertelsen, M. Fedgchin, B. Vogt and M. Burnier, Clin. Pharmacol. Ther., 2009, 85, 145. [35] C. J. Bleickardt, D. E. Mullins, C. P. MacSweeney, B. J. Werner, A. J. Pond, M. F. Guzzi, F. D. C. Martin, G. B. Varty and R. A. Hodgson, Psychopharmacology, 2009, 202, 711. [36] C. F. Ferris, S.-F. Lu, T. Messenger, C. D. Guillon, N. Heindel, M. Miller, G. Koppel, F. R. Bruns and N. G. Simon, Pharmacol. Biochem. Behav., 2006, 83, 169. [37] C. D. Guillon, G. A. Koppel, M. J. Brownstein, M. O. Chaney, C. F. Ferris, S.-F. Lu, K. M. Fabio, M. J. Miller, N. D. Heindel, D. C. Hunden, R. D. G. Cooper, S. W. Kaldor, J. J. Skelton, B. A. Dressman, M. P. Clay, M. I. Steinberg, R. F. Bruns and N. G. Simon, Bioorg. Med. Chem., 2007, 15, 2054. [38] http://www.azevan.com/pipeline/default.asp, 2009. [39] J. S. Bryans, M. E. Bunnage, P. S. Johnson, H. J. Mason, L. R. Roberts, T. Ryckmans, A. Stobie and T. J. Underwood, WO Patent Application 2006114706, 2006. [40] J. S. Bryans, P. S. Johnson, L. R. Roberts and T. Ryckmans, WO Patent Application 2006123242, 2006. [41] C. Bissantz, E. Goetschi, C. Grundschober, R. Masciadri, H. Ratni, M. Rogers-Evans and P. Schnider, WO Patent Application 2008084005, 2008. [42] C. Bissantz, C. Grundschober, R. Masciadri, H. Ratni, M. Rogers-Evans and P. Schnider, WO Patent Application 2008068185, 2008. [43] C. Bissantz, C. Grundschober, R. Masciadri, H. Ratni, M. Rogers-Evans and P. Schnider, WO Patent Application 2008068183, 2008. [44] C. Bissantz, C. Grundschober, R. Masciadri, H. Ratni, M. Rogers-Evans and P. Schnider, WO Patent Application 2008068184, 2008. [45] C. Bissantz, C. Grundschober, R. Masciadri, H. Ratni, M. Rogers-Evans and P. Schnider, WO Patent Application 2008068159, 2008. [46] C. Bissantz, C. Grundschober, H. Ratni, M. Rogers-Evans and P. Schnider, US Patent Application 2007155761, 2007. [47] C. Bissantz, C. Grundschober, H. Ratni, M. Rogers-Evans and P. Schnider, WO Patent Application 2007006688, 2007. [48] http://clinicaltrials.gov/ct2/show/NCT00361491, 2006. [49] http://clinicaltrials.gov/ct2/show/NCT00358631, 2006. [50] Sanofi-Aventis, half-year financial report, 2008. [51] C. Louis, C. Cohen, R. Depoortere and G. Griebel, Neuropsychopharmacology, 2006, 31, 2180. [52] C. Griffante, A. Green, O. Curcuruto, C. P. Haslam, B. A. Dickinson and R. Arban, Br. J. Pharmacol., 2005, 146, 744.
146
Thomas Ryckmans
[53] C. Serradeil-Le Gal, D. Raufaste, S. Derick, J. Blankenstein, J. Allen, B. Pouzet, M. Pascal, J. Wagnon and M. A. Ventura, Am. J. Physiol., 2007, 293, R938. [54] W. Lubisch, T. Oost, W. Wernet, W. Hornberger, L. Unger and H. Geneste, WO Patent Application 2006100081, 2006. [55] W. Lubisch, T. Oost, W. Wernet, W. Hornberger, L. Unger and H. Geneste, WO Patent Application 2006100080, 2006. [56] T. Oost, W. Lubisch, W. Wernet, W. Hornberger and L. Unger, WO Patent Application 2006100082, 2006. [57] T. Oost, W. Lubisch, W. Wernet, W. Hornberger, L. Unger, H. Geneste and A. Netz, WO Patent Application 2007063123, 2007. [58] H. Geneste, T. Oost, C. W. Hutchins, W. Wernet, L. Unger, W. Hornberger, W. Lubisch and A. Netz, WO Patent Application 2008107399, 2008. [59] T. Kuwata, H. Nozawa and S. Hayashi, JP Patent Application 2008050354, 2008. [60] T. Arndt, T. Oost, W. Lubisch, W. Wernet, W. Hornberger, L. Unger and J. Ruiz Caro, WO Patent Application 2008025736, 2008. [61] G. A. Koppel, WO Patent Application 2006102308, 2006. [62] J. Letourneau, C. Riviello, K.-K. Ho, J.-H. Chan, M. Ohlmeyer, P. Jokiel, I. Neagu, J. R. Morphy and S. E. Napier, WO Patent Application 2006095014, 2006. [63] T. Kuwada, D. Nozawa, T. Ishizaka, M. Yoshinaga and K. Yoshikawa, WO Patent Application 2009017236, 2009. [64] M. Craighead, R. Milne, L. Campbell-Wan, L. Watson, J. Presland, F. J. Thomson, H. M. Marston and C. P. MacSweeney, Prog. Brain Res., 2008, 170, 527. [65] J. Letourneau, P. Jokiel, E. Napier, K.-K. Ho, M. Ohlmeyer, D. R. McArthur, J. Fiona, P. D. Ratcliffe and S. Jurgen, WO Patent Application 2008033764, 2008. [66] I. Neumann, WO Patent Application 2008071779, 2008. [67] K. Honda and Y. Takano, J. Pharmacol. Sci. (Tokyo, Jpn.), 2009, 109, 38. [68] A. L. Del Tredici, K. E. Vanover, A. E. Knapp, S. M. Bertozzi, N. R. Nash, E. S. Burstein, J. Lameh, E. A. Currier, R. E. Davis, M. R. Brann, N. Mohell, R. Olsson and F. Piu, Biochem. Pharmacol., 2008, 76, 1134. [69] A. A. Failli, J. S. Shumsky, R. J. Steffan, T. J. Caggiano, D. K. Williams, E. J. Trybulski, X. Ning, Y. Lock, T. Tanikella, D. Hartmann, P. S. Chan and C. H. Park, Bioorg. Med. Chem. Lett., 2006, 16, 954. [70] I. Tsukamoto, H. Koshio, S. Akamatsu, T. Kuramochi, C. Saitoh, T. Yatsu, H. YanaiInamura, C. Kitada, E. Yamamoto, S. Sakamoto and S.-i. Tsukamoto, Bioorg. Med. Chem., 2008, 16, 9524. [71] M. Kanbara, T. Suzuki, T. Kondo, K. Yokoyama and H. Kobayashi, JP Patent 2007308391, 2007. [72] C. M. Yea, C. E. Allan, D. M. Ashworth, J. Barnett, A. J. Baxter, J. D. Broadbridge, R. J. Franklin, S. L. Hampton, P. Hudson, J. A. Horton, P. D. Jenkins, A. M. Penson, G. R. W. Pitt, P. Riviere, P. A. Robson, D. P. Rooker, G. Semple, A. Sheppard, R. M. Haigh and M. B. Roe, J. Med. Chem., 2008, 51, 8124. [73] J. L. Cavalcante, S. Khan and M. Gheorghiade, Expert Rev. Cardiovasc. Ther., 2008, 6, 1331. [74] L. Foulon, P. Rochard, C. Serradeil Le Gal and G. Valette, FR Patent Application 2909668, 2008. [75] G. A. Koppel and N. D. Heindel, WO Patent Application 2007109615, 2007. [76] M. A. Xiang, R. H. Chen, K. T. Demarest, J. Gunnet, R. Look, W. Hageman, W. V. Murray, D. W. Combs, P. J. Rybczynski and M. Patel, Bioorg. Med. Chem. Lett., 2004, 14, 3143. [77] M. A. Xiang, P. J. Rybczynski, M. Patel, R. H. Chen, D. F. McComsey, H.-C. Zhang, J. W. Gunnet, R. Look, Y. Wang, L. K. Minor, H. M. Zhong, F. J. Villani, K. T. Demarest, B. P. Damiano and B. E. Maryanoff, Bioorg. Med. Chem. Lett., 2007, 17, 6623.
Advances in Vasopressin Receptor Agonists and Antagonists
147
[78] M. A. Xiang, M. Patel, P. Rybczynski, J. Gunnet, K. T. Demarest, R. Look, B. Maryanoff, M. J. Costanzo and S. C. Yabut, WO Patent Application 2008036755, 2008. [79] M. A. Xiang, M. Patel, P. Rybczynski, J. Gunnet, K. T. Demarest, R. Look, B. Maryanoff, M. J. Costanzo and S. C. Yabut, WO Patent Application 2008036759, 2008. [80] H. Meier, E. Bender, U. Brueggemeier, I. Flamme, D. Karthaus, P. Kolkhof, D. Meibom, D. Schneider, V. Voehringer, C. Fuerstner, J. Keldenich, D. Lang, E. Pook and C. Schmeck, WO Patent Application 2007134862, 2007.
CHAPT ER
7 The Emergence of GPR119 Agonists as Anti-Diabetic Agents Robert M. Jones and James N. Leonard
Contents
1. Introduction 2. Discovery and Characteristics of GPR119 3. The Biology of GPR119 3.1 Proposed endogenous regulators of GPR119 3.2 GPR119 regulation of insulin release 3.3 GPR119 regulation of GLP-1 and GIP release 4. GPR119 Agonists: Medicinal Chemistry 4.1 Six-membered heterocyclic ring–based agonists 4.2 Five-membered heterocyclic ring–based agonists 4.3 Bicyclic core–based agonists 4.4 Linear core–based agonists 5. Clinical Trial Status and Future Prospects References
149 150 150 150 152 153 154 154 159 161 164 166 167
1. INTRODUCTION Type 2 diabetes mellitus (T2DM) is emerging as a disease of staggering proportions in the twenty-first century, with an estimated 300 million cases worldwide projected by 2020 [1]. The physiological hallmarks of T2DM are severe insulin resistance, inappropriate hepatic glucose production in the hyperglycemic state, and insufficient insulin production Arena Pharmaceuticals, 6166 Nancy Ridge Drive, San Diego, CA 92121, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04407-8
r 2009 Elsevier Inc. All rights reserved.
149
150
Robert M. Jones and James N. Leonard
from pancreatic b cells. Historically, treatment regimens for T2DM have shown significant effectiveness for improving glucose homeostasis; however, after about 2 years of therapy, increasing failure to maintain glycemic control is observed [2]. Recent approaches based on the biology of glucagon-like peptide-1 (GLP-1) have offered new hope for T2DM patients, due to the ability of this gut hormone to (1) elicit insulin release in a glucose-dependent fashion, (2) maintain and perhaps even enhance b-cell mass, and (3) promote satiety and thus weight loss [3]. These features suggest that GLP-1 could promote improved glycemic control acutely and, furthermore, might have a chronic, disease-modifying effect by virtue of prevention or reversal of b-cell failure. GLP-1-based therapies have so far consisted of stabilized peptide agonists of the GLP-1 receptor and small molecule inhibitors of dipeptidyl peptidase-IV (DPP-IV), the enzyme primarily responsible for inactivation of endogenously produced GLP-1 [4]. DPP-IV inhibitors also prevent degradation of glucose-dependent insulinotropic peptide (GIP), the other major gutderived insulinotropic hormone, or ‘‘incretin.’’ Agonists of GPR119 have emerged from pharmaceutical discovery efforts to identify an improved GLP-1 therapeutic by combining the convenience offered by oral dosage of DPP-IV inhibitors and the pharmacological robustness of GLP-1 receptor agonists. Mounting data support this hypothesis and moreover indicate that GPR119 agonists may also spearhead a new class of anti-diabetic therapies through modulation of intestinal endocrine cells that sequester key regulators of energy homeostasis.
2. DISCOVERY AND CHARACTERISTICS OF GPR119 After the discovery of GPR119 in 1999 using data afforded by the Human Genome Project, it was subsequently described in the peer-reviewed literature as a Class A receptor with no close relatives [5]. The receptor possesses distant similarity to biogenic amine and cannabinoid receptors (B40% identity in the transmembrane regions). The complete understanding of GPR119 expression is rather recent due to the complexity of its distribution in rare endocrine cell types. Several laboratories eventually showed that GPR119 is highly expressed in pancreatic islets and in some regions of the gut, particularly, the colon [6–8]. There is some discrepancy regarding the GPR119-expressing cell types within the islet [9], but the most convincing data firmly demonstrate that GPR119 is present in the great majority of pancreatic b cells [7]. Heterologous expression of GPR119 causes substantial increases in 3’,5’-cyclic adenosine monophosphate (cAMP) levels, which are further elevated
151
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
by agonists of the receptor [6,7,10]. Therefore, like the GLP-1 receptor, GPR119 is both b-cell-expressed and Gas-coupled.
3. THE BIOLOGY OF GPR119 3.1 Proposed endogenous regulators of GPR119 Lysophosphatidylcholine (LPC, 1) was the first proposed endogenous ligand for GPR119, based on its ability to stimulate glucose-dependent insulin release and increase cAMP in GPR119-transfected cells [6]. However, subsequent studies showed that the efficacy of 1 is poor, relative to other GPR119 modulators [10–11]. 1 also mediates insulin release in insulinoma cell lines that lack GPR119, shedding further doubt on the likelihood that this lipid is a true endogenous regulator of GPR119 [11].
Me3N HO O
O P H3C
O
HO
O
O O
1
OH
2
NH
NH
O
O
3
In contrast, the satiety factor oleoyl ethanolamide (OEA, 2) is a significantly more efficacious agonist of GPR119 [10]. The in vitro efficacy of 2 is similar to that of synthetic GPR119 agonists [11]. 2 is also capable of eliciting biological responses in GLUTag cells that endogenously express GPR119, but this response is greatly attenuated when receptor levels are reduced by siRNA treatment [12]. Furthermore, levels of 2 in the gut are nutrient-regulated, consistent with the proposed role of GPR119 as a mediator of energy homeostasis [13]. On the contrary, some data cast doubt on the likelihood that 2 is a true endogenous regulator of GPR119. Nutrient regulation of 2 levels is seen primarily in the jejunem, which is not a site of significant GPR119 expression [13]. Additionally, 2 is known to modulate other molecular targets governing energy homeostasis, including the nuclear receptor PPARa [14]. It is therefore challenging to assess the relative contribution of GPR119 activation to the
152
Robert M. Jones and James N. Leonard
known functions of 2. Data from GPR119 knockout (KO) mice shed at least some light on this issue. Weight loss mediated by 2 is not seen in mice lacking PPARa, but remains fully intact in mice lacking GPR119 [14,15]. Moreover, the ability of 2 to trigger insulin release is at least partly GPR119-independent [16]. Therefore, at least some aspects of the function of 2 in vivo are not mediated through GPR119.
OH
HO
O NH
HN OH
O
4
O
5
Regardless of its putative role as an endogenous ligand, the identification of 2 as a GPR119 agonist suggests that fatty acyl amides warrant further exploration as possible ligands for the receptor. Neither endocannabinoids nor shorter chain fatty acyl amides (e.g., capsaicin) are active in this regard. However, the endovallinoids N-oleoyl dopamine (OLDA, 3) and olvanil 4 have emerged as a third class of lipidic signaling molecules proposed to be endogenous modulators of GPR119 [11]. Interestingly, close relatives of these endovallinoids, such as (R)-Noleoyltyrosinol 5, are the most potent fatty acyl amide agonists of GPR119 identified to date [11]. 3 is essentially indistinguishable from 2 in terms of its ability to stimulate cAMP in GPR119-transfected cells. 3 also stimulates insulin release in vitro, elevates the incretin hormone GIP in vivo, and improves glucose tolerance, all in a largely GPR119-dependent manner and to an extent similar to that of synthetic GPR119 agonists [11]. Synthesis of 3 in the gut and pancreas has not been assessed in the literature. However, it is noteworthy that fatty acyldopamines may be produced by direct conjugation of fatty acids and dopamine [17], both of which are highly regulated by nutrient challenge in the gut [18–19].
3.2 GPR119 regulation of insulin release Potent, selective GPR119 agonists have been used to demonstrate that the receptor mediates robust glucose-dependent insulin release in rodent islets, insulinoma cell lines, and in vivo [7]. By using the agonist AR231453
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
153
6, these effects were shown to be completely GPR119-dependent in GPR119-deficient mice and in a mouse model using siRNA to depress GPR119 levels. The hamster insulinoma cell line HIT-T15 expresses GPR119 at levels similar to those seen in pancreatic islets, which likely accounts for its particularly robust response to GPR119 agonists [7]. It is therefore a convenient in vitro model of GPR119 function. When measuring both cAMP and insulin release, the efficacy of 6 is similar to that achieved with forskolin. These data suggest that the insulinotropic effects of GPR119 agonists are highly significant. The GPR119 agonist tools, PSN375963 7, PSN632408 8, and 2, have been assessed in MIN6c4 cells, where a comparatively poor cAMP response was observed [16]. This is most likely due to relatively low levels of GPR119 in MIN6 cells when compared with HIT-T15 cells, although recent reports cast doubt on the GPR119 specificity of 7 and 8 (see Section 4.2).
N
O
N O
N
O NO2
N N
O
N
O N
NH
N
O N
N
F N
N
SO2Me AR231453 6
PSN375963 7
PSN632408 8
In rat and mouse islets, 6 stimulates insulin release at glucose concentrations of 8 mM or higher [7]. Thus, the actions of GPR119 are glucose-dependent, as would be expected from a Gas-coupled b-cell receptor. The agonist effects are robust, achieving efficacy similar to that of GLP-1. They are also GPR119-dependent, since 6 is inactive when incubated with islets from GPR119-deficient mice. These data suggest that oral GPR119 agonists should improve insulin release and glucose tolerance in vivo, and indeed, this has been shown to be the case in normoglycemic and diabetic rodent models [7]. Collectively, these data demonstrate the feasibility of developing orally active GPR119 modulators with a GLP-1-like ability to improve b-cell function.
154
Robert M. Jones and James N. Leonard
3.3 GPR119 regulation of GLP-1 and GIP release Aside from pancreatic islets, sub-regions of the gut are the only other sites of robust GPR119 expression identified to date. Cellular expression studies have extended these observations by showing that most GLP-1producing L cells in the ileum and colon also contain GPR119 [8]. This is consistent with data showing high GPR119 expression in most in vitro L-cell models [8,12]. GIP, the other major insulinotropic hormone of the gut, is produced primarily in the duodenal K cells. Here, the cellular expression of GPR119 is less clear. In one study, detectable levels of GPR119 were not seen in GIP-positive cells [8], but others have reported that primary K-cell cultures contain significant amounts of GPR119 [20]. Nevertheless, it is clear that GPR119 agonists such as 6 increase plasma levels of both GLP-1 and GIP in vivo [8]. In concordance with these data, GPR119 agonists stimulate GLP-1 release from primary colonic crypt cultures [21], fetal rat intestinal cultures [12], and GLUTag cells [8], an immortalized mouse L-cell line. GPR119 therefore enhances incretin release, whereas DPP-IV inhibitors mitigate incretin inactivation. Such joint therapeutic approaches will function synergistically with regard to GLP-1 secretion and glycemic control. In support of this hypothesis, combined administration of 6 and a DPP-IV inhibitor acutely increases plasma GLP-1 levels and improves glucose tolerance to a significantly greater degree than either agent alone [8]. Combination treatment also elevates GLP-1 release in a nutrientindependent fashion and thus may offer a more effective means to maintain elevated GLP-1 levels between meals in diabetic patients.
4. GPR119 AGONISTS: MEDICINAL CHEMISTRY 4.1 Six-membered heterocyclic ring–based agonists Following a cyclase-based high throughput screening (HTS) campaign, pyrimidine 9 was identified as an inverse agonist of hGPR119 (IC50 ¼ 84 nM) [22]. In structure activity relationship (SAR) studies related to 9, the trifluoromethyl pyrazole motif was replaced with a series of aryl ethers leading to the identification of agonists 10 (EC50 ¼ 1 mM) and 11 (EC50 ¼ 1.7 mM). Further extensive parallel SAR work led to the identification of multiple additional agonist trigger moieties, including sulfones, sulfonamides, and various five-membered heterocyclic substituents. The best activity was seen when such a group was attached to the 4-position of the phenyl ring. In addition, with the appropriate trigger in place, agonist activity was maintained with either an ether or aniline linker to the pyrimidinyl core.
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
155
Me
Me O
O
O
O
N
N
N NO2
N N
O
N NO2
N N
N O
O
NO2
N N
NH
N N F F F
X R
9
SO2Me
O
10. X = CO, R = OMe 11. X = CH2, R = Me
12
In further SAR work, the aniline sulfone group was fixed, and conservative ester replacements afforded potent GPR119 agonists. The SAR of the alkyl substituent on the oxadiazole ring was particularly remarkable. Increasing the size of this substituent from methyl to ethyl provided a more than 25-fold improvement in agonist potency. The inclusion of a-branching in this group again gave a significant improvement in potency, as exemplified by 12 (hGPR119 EC50 ¼ 5.8 nM, melanophore dispersion assay). Fluoro substituents on the aryl ring further improved potency, with 6 being the first agonist with subnanomolar potency (EC50 ¼ 0.68 nM) observed. Standard profiling of 6 (CEREP) and exhaustive counter screens against other receptors indicated that 6 is a highly selective GPR119 agonist [22–24]. In addition, 6 had no activity against DPP-IV in a fluorescent-based assay using diprotin. Importantly for a pharmacological tool, 6 had GPR119 agonist activity across species, albeit with somewhat lower potency for rodent GPR119 (cynomolgus monkey GPR119 EC50 ¼ 0.4 nM; mouse GPR119 EC50 ¼ 12 nM; rat GPR119 EC50 ¼ 14 nM; dog GPR119 EC50 ¼ 1.6 nM: melanophore dispersion assay). Notably, across all species, 6 displayed a significantly greater level of efficacy in these assays than the putative endogenous ligand 2. Importantly, 6 stimulated cAMP production through endogenously expressed GPR119 in the b-cell-like hamster insulinoma cell line HIT-T15, with an EC50 of 4.7 nM, a value comparable to the EC50 observed in cell lines over-expressing the cloned rodent receptors [24]. Notably, the EC50 for insulin release from this cell line was 3.5 nM.
156
Robert M. Jones and James N. Leonard
Oxadiazole 6 did not significantly inhibit any of the major cytochrome P450 enzymes and had no activity up to 10 mM at the human Ether-ago-go Related Gene (hERG) channel in a patch clamp assay. Furthermore, excellent exposure was observed after oral administration to C57/bl6J mice (10 mg/kg po: tmax ¼ 0.5 h, Cmax ¼ 9.84 mM). The in vivo effect of 6 with respect to % inhibition of glucose area under the curve (AUC) in an oral glucose tolerance test (oGTT) was similar in C57/bl6J mice and Sprague–Dawley (SD) rats at a dose of 3 mg/kg ip despite a clear species difference in the magnitude of the glucose excursion. This similarity may be expected in light of the similar potency and efficacy of 6 at both mouse and rat GPR119. In addition, 6 improved oral glucose tolerance in wildtype C57/bl6J mice after oral administration. Importantly, 6 also enhanced glucose-dependent insulin release and improved oral glucose tolerance in wild-type C57/bl6J mice but not in GPR119-deficient mice at a dose of 20 mg/kg po, clearly demonstrating the receptor dependence of the observed effects [23,24]. However, a statistically significant effect on glucose excursion in an oGTT following oral administration of 6 was not observed in rats, although a trend toward efficacy was seen at a dose of 30 mg/kg. The lack of a consistent effect was believed to result from the combination of a narrower window for measuring the decrease in glucose excursion in rats as well as the significantly poorer exposure of 6 in rats compared to mice following oral administration (rat, 10 mg/kg, po: tmax ¼ 1 h, Cmax ¼ 0.250 mM, F ¼ 12%). A key aspect of b-cell Gas-coupled receptors is their ability to enhance insulin release in a glucose-dependent fashion, and 6 has been used to show this is characteristic of GPR119. In both rat and mouse islets, 6 had no effect on insulin release at basal (5 mM) glucose concentrations but had GLP-1-like efficacy when the glucose concentration was W8 mM. Also, unlike the sulfonyl urea glyburide, 6 did not stimulate insulin release or cause hypoglycemia during a drug challenge to fasted mice. As noted earlier, 6 (10 mg/kg po) also stimulates both GLP-1 and GIP release during OGTTs in C57BL/6J mice [8,12,21]. No effect was seen on GIP release in GPR119 KO mice. Moreover, co-administration of 6 with a DPP-IV inhibitor leads to synergistic increases in GLP-1 levels, which results in inhibition of the glucose excursion during an OGTT by the 6/DPP-IV inhibitor combination that is greater than that achieved by either agent alone [8]. The potential of GPR119 agonists to exert b-cell protective effects through increased cAMP levels has also been demonstrated. In MIN6 pancreatic b cells expressing GPR119, 6 induced Akt phosphorylation and IRS-2 expression, key measures of islet mass protection [24,25].
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
157
O
O O
O
N
N
O
O N
N
R′
Me N
N
NH
O
Me
Me N
N R
13
14. R′ = Me, R = SO2Me 15. R′ = MeO, R = 1,2,4-triazolo
Several patent applications published during the past 24 months disclosed additional GPR119 agonists bearing six-membered ring heterocyclic core motifs. Picoline 13 was described as a selective agonist of human, dog, cynomolgus monkey, mouse, and rat GPR119 expressed in melanophores with EC50s of 2, 1, 35, 41, and 44 nM respectively [26]. Furthermore, 13 possessed significant aqueous solubility at pH 7, with no appreciable inhibition of five cytochrome P450 enzymes. Picoline 13 also elicited a dose-responsive mean inhibition of glucose excursion of 22, 24, and 70% when orally administered to normoglycemic male SD rats at doses of 0.3, 3, and 30 mg/kg respectively. Related pyrimidines including sulfone 14 and 1,2,4-triazole 15 have also been described [27,28]. Sulfone 14 displayed agonist activity at GPR119 from different species in melanophores (EC50 ¼ 2, 8, 43, and 42 nM; human, dog, cynomolgus monkey, mouse, and rat respectively). Upon dose escalation in rats, 14 afforded linearly increasing plasma exposure (AUC ¼ 4.73, 55.9, and 515.32 mg h/mL at per oral doses of 3, 30, and 300 mg/kg respectively). The ability of 14 to affect glucose homeostasis in male SD rats was assessed using an oGTT. 14 exhibited a 38% mean inhibition of glucose excursion following an oral 10 mg/kg dose. Similarly, triazole 15 showed agonist activity in GPR119-transfected melanophores (EC50 ¼ 2, 4, 57, and 81 nM at human, dog, cynomolgus monkey, mouse, and rat GPR119 respectively). It also exhibited linear plasma exposure upon dose escalation in rats (AUC ¼ 3.59, 79.82, and 285.99 mg.h/mL at 3, 30, and 300 mg/kg po), in addition to displaying oral efficacy in an oGTT.
158
Robert M. Jones and James N. Leonard
O
O
N
N
N O
O
N
O
O N
O
N
O N N
N F
F
SO2Me
SO2Me
SO2Me
16
17
18
A series of 3,6-substituted pyridines exemplified by 16 and 17 have been described as GPR119 agonists in a patent application [29]. The compounds, their salts, and their use for the treatment of a condition mediated through GPR119 such as metabolic disorders, diabetes, and obesity are claimed. Test compounds from this patent application were assayed for their effects on GLP-1 levels in vivo. Treatment of male C57/ B16 mice with 16 (30 mg/kg po) caused an increase in the total GLP-1 level to 7.572.0 pg/mL, compared to 2.371.2 pg/mL for vehicle-treated mice. A related pyridazine series has been independently disclosed exemplified by 18 [30]. Pyridazine 18 was reported to have a hGPR119 EC50 value of 0.046 mM. Assays to determine the effect of the compounds on weight gain and glucose-stimulated insulin release in mice are described, but no specific data are given. Additional GPR119 agonists bearing six-membered ring cores exemplified by pyridine 19 and bispiperidine 20 have recently been disclosed [31,32]. F O O
N
N
N N
O
N
O
N N
MeO2S SO2Me 19
20
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
159
4.2 Five-membered heterocyclic ring–based agonists In a seminal paper that described deorphanization of GPR119, the pharmacological profile of the 1,2,4-oxadiazole carbamate 8, identified by optimization of the HTS hit 7, was also disclosed as a GPR119 agonist [10]. Like the putative endogenous ligand 2, oxadiazole 8 produced a concentration-dependent increase in intracellular cAMP levels in a human embryonic kidney (HEK) cell line expressing GPR119 with an EC50 of 1.9 mM. Oral administration of 8 to rats at a dose of 100 mg/kg po reduced 24 h cumulative food intake. The reduction in food intake was not associated with drug-induced malaise, as no effects were seen on locomotor activity or in conditioned taste aversion and kaolin consumption tests. The acute anorectic effect translated into chronic effects on body weight. In dietinduced obese mice and in growing, high-fat diet-fed SD rats, attenuation of body weight gain produced by 8 (100 mg/kg/day po) was comparable to that of the prescribed anorectic drug sibutramine hydrochloride hydrate (5 mg/kg/day po) [33]. More recently, it was reported that 8 possessed similar potency and efficacy at hGPR119 (EC50 ¼ 5.6 mM, Emax ¼ 110%) relative to the putative endogenous ligand 2 (EC50 ¼ 3.2 mM, Emax ¼ 100%) in a yeast/b-galactosidase fluorescence reporter assay [34]. However, in MIN6c4 cells expressing endogenous GPR119, the synthetic agonists 7 and 8 differ from 2 in their effects on intracellular cAMP and calcium levels and insulin secretion [16]. While the endogenous GPR119 ligand 2 signals through GPR119 in a manner similar to GLP-1 and its receptor to elicit increased insulin secretion and increased intracellular Ca2+ and cAMP levels, oxadiazoles 7 and 8 produced divergent effects on these parameters. These studies suggest that 7 and 8, although they do weakly activate GPR119, may also modulate GPR119-independent pathways and thus may be unsuitable as GPR119-specific pharmacological tools [16]. These observations are somewhat corroborated by a recent report in which 8 was shown to possess an EC50 W20 mM in a CHO6CRE luciferase reporter assay [35].
O
O
O
O
O
R
O N
N
SO2Me 21
F
O O N
N
SO2Me 22 (PSN119-2)
O N
N
N
N
O N
N
O
F
N
SO2Me 23
160
Robert M. Jones and James N. Leonard
Exchange of the 4-pyridyl motif present in 7 and 8 with a 3-fluoro-4(methanesulfonyl)phenyl moiety led to agonist 21 (R ¼ F) possessing EC50 and Emax values of 0.9 mM and 375% respectively in a yeast reporter assay expressing recombinant human GPR119. Focused SAR modifications of the aromatic moiety in which the R-substituent was either replaced by other substituents (21, R ¼ –H, –Cl, –Me, –OMe) or moved to the adjacent 2-position led to reduced potency or to inactivity. Subsequently, installation of an (R)-methyl group to the methyleneoxy linker of 21 afforded PSN119-2 22 (EC50 ¼ 0.4 mM, Emax ¼ 358%). The antipode of 22 was substantially less active (10-fold) and less efficacious (twofold), as were the enantiomeric ethyl homologues. 22 was shown to stimulate insulin secretion from the hamster-derived insulinoma HITT15 cell line (EC50 ¼ 18 nM) and GLP-1 release from the immortalized entero-endocrine murine GLUTag cell line (EC50 ¼ 8 nM). In male SD rats, 22 orally administered at doses of 10 and 30 mg/kg achieved a Cmax of B2 and B5.1 mM, respectively, and significantly attenuated the glucose excursion in an oGTT. At the same doses, 22 significantly reduced cumulative 24 h food intake levels in SD rats compared to vehicle-treated control rats, but these effects were not as robust as those observed for sibutramine (5 mg/kg, po). Analogs of 22, resulting from replacement of the terminal tert-butyl carbamate with an N-linked 5-isopropyl-1,2,4-oxadiazole, 23, are the subject of a recent patent application [36]. Bioisosteric replacement of a carbamate by an oxadiazole is a successful GPR119 agonist SAR theme previously disclosed in earlier patent applications [37–41]. Additional five-membered ring core agonists have been independently reported in several recent patent applications [42–44]. In one disclosure, compounds were evaluated in cAMP stimulation assays, and thiazole 24 (10 mM) demonstrated a 69% change in Forster resonance energy transfer (FRET) signals relative to control [43]. Derivative 24 was also shown to stimulate insulin secretion by 1.66-fold in a SD rat–derived islet perfusion assay following a 16 mM glucose treatment [43]. Thiazole 24 (30 mg/kg, po) also facilitated a 52% reduction in glucose AUC in 8- to 10-week-old C57/6J fasted mice following a 2 g/kg oral glucose bolus challenge. In a more recent disclosure, a closely related series of triazolyl piperidine–based GPR119 agonists was described [43]. In vitro GPR119 agonism was again determined by measurement of the decrease in FRET signaling in response to changes in intracellular cAMP levels. Many of these compounds were reported to cause a significant reduction in FRET signaling relative to positive control (e.g., 25, 87.61% reduction). Additionally, triazole 25 (30 mg/kg po) was shown to reduce mean glucose excursion in male C57/6J mice by 40.8%.
161
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
F3C N
O O
N
N
S N
O
O
N
N
F N N N
O
N O N
O F
N F
N N 24
N
N N N
SO2Me
25
26
The characterization of the potent, selective, and orally bioavailable GPR119 agonist, MBX-3152 (structure not disclosed), has been reported [45]. In CHO cells stably expressing hGPR119, MBX-3152 afforded an EC50 value of 1.8 nM, and in immortalized murine GLUTag cells, it increased GLP-1 secretion 1.5-fold at concentrations of 200 nM and above with an EC50 value of 82 nM. When administered chronically to high fat– fed female Zucker diabetic fatty (ZDF) rats, MBX-3152 (30 mg/kg, po, 35 days) was shown to concomitantly delay disease onset and reduce food intake, body weight gain, and plasma triglycerides, thereby mitigating the profound increase in insulin resistance that develops in this animal model. In wild-type mice, but not in GPR119 KO mice, the compound [30 mg/kg, po] also improved glucose tolerance and increased incretin release in the presence of the DPP-IV inhibitor sitagliptin (1 mg/kg, po]. In rats, MBX-3152 (30 mg/kg, po) increased insulin secretion and glucose infusion rate under hyperglycemic clamp conditions implying a direct action of the agonist at pancreatic b cells. Pyrrolidinyl 1,3,4-oxadiazole 26 is representative of another chemical series of GPR119 modulators [44]. These analogs are similarly potent at stimulating cAMP in Flp-In-CHO-hGPR119 cells and are claimed to be useful for the treatment or prevention of disorders associated with GPR119.
4.3 Bicyclic core–based agonists The structure of GSK252A 27a has recently been disclosed [21]. This GPR119 agonist possesses a novel pyrrolopyrimidine-based bicyclic core and bears a striking resemblance to GPR119 agonists in which a pyrazolo
162
Robert M. Jones and James N. Leonard
[3,4-d]pyrimidine was utilized as the core scaffold, exemplified by 28, disclosed in an earlier application by an independent group [41]. Compound 27a was reported to have good pharmacokinetic properties in SD rats. At a dose of 30 mg/kg po, 27a achieved a Cmax ¼ 14 mM, a half life of 3.1 h, and absolute oral bioavailability of 60%. In a CHO6CRE reporter assay, 27a afforded an EC50 ¼ 40 nM compared to an EC50 ¼ 200 nM for 28 [21,41]. R N O
O O
O
O
N
N
N
N O O
O
N
N
N N N
N
N
N
N
N
F
F
SO2Me 27a. R = H = GSK252A 27b. R = Me
F MeO2S 28
NC 29
The effects of 27a on incretin, Peptide YY (PYY), glucagon, and insulin secretion both in vitro and in vivo were disclosed. The compound stimulated GLP-1 release from both GLUTag (EC50 ¼ 22 nM) and primary mouse colonic crypt cell cultures in vitro. In addition, 27a augmented glucose-stimulated insulin secretion from rat islet cells in the presence of 12 mM glucose but did not affect insulin output in the presence of 3 mM glucose, consistent with similar glucose-dependent effects of 6. Oral administration of 27a (10 mg/kg, po) to mice fasted overnight resulted in a fourfold increase in GLP-1 levels and a twofold increase in GIP levels when compared to vehicle control, with no increase seen in GPR119 KO mice. In addition, there was also a threefold increase in PYY levels and a twofold increase in glucagon levels in wild-type, but not in GPR119 KO mice. Augmentation of insulin secretion by 27a (10 mg/kg, po) was assessed in rats using the hyperglycemic clamp model. In the hyperglycemic state targeting a glucose concentration of 200 mg/dL, both glucose infusion rate and C-peptide levels were significantly increased indicating a robust enhancement of insulin release. Consistent with these data, 27a decreased the glucose AUC by 43% and increased the peak insulin levels by 1.4-fold during the intravenous glucose tolerance test (IVGTT) and improved glucose tolerance during the oGTT by 38% with no significant effect on the
163
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
insulin response. This effect was not significantly different from that of the DPP-IV inhibitor LAF237. A hyperglycemic euglycemic clamp was also used to show that GPR119 activation may also impact glucose utilization independent of insulin secretion. The glucose infusion rate needed to maintain euglycemia was approximately threefold higher in rats treated with 27a compared to vehicle-treated rats, suggesting that 27a promoted significantly enhanced glucose utilization [21]. Tert butyl carbamate 27b and related bicyclic pyrrolo[2,3-d]pyrimidines, including 1,2,4-oxadiazole 29 together with indoline core congeners, are claimed in two recent patent applications [46,47]. To test the effect of 27b on insulin sensitization in male SD rats, a hyperinsulinemic euglycemic clamp was initiated 90 min after oral administration of 27b (10 mg/kg po) or vehicle. As observed for 27a, the glucose infusion rate required to maintain euglycemia in rats given 27b was three times that of vehicle-treated rats. Br O
O O O
O N
N
O N N O
N N
N N N
N N O
N
N
NH
NH
F
Cl MeO2S 30
31
MeO2S 32
More recent patent filings describe selective GPR119 modulators comprised of a fused pyrimidinone core 30, specifically, 6-substituted 5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4(3H)-ones and 7-substituted 5,6,7,8-tetrahydropyrido[3,4-d]pyrimidin-4(3H)-ones [48,49]. Both series of GPR119 modulators are described as being useful for treating or preventing obesity, diabetes, metabolic disease, cardiovascular disease, or a disorder related to the activity of GPR119 in a patient. Undisclosed compounds from these applications are purported to activate GPR119 and stimulate cAMP production in transfected HEK293 cells, exhibiting EC50 values ranging from about 50 nM to 14,000 nM. A mouse oGTT protocol is described for assessing the activity of these compounds in vivo, and various undisclosed fused pyrimidinone examples were reported to be effective in lowering blood glucose levels after glucose challenge.
164
Robert M. Jones and James N. Leonard
Furthermore, illustrative examples were found to produce a sustained decrease in blood glucose in a non-genetic mouse model of T2DM. Building on the existing knowledge base of the SAR of bicyclic-based GPR119 agonists, two series of azabicyclic agonists have been described, exemplified by 31 and 32 [50,51]. An in vitro luciferase assay using HEK 293 cells expressing hGPR119 was conducted to assess the efficacy of the compounds. Bicyclic triazole 31 exhibited a hGPR119 EC50 value of 19.49 nM, and the pyrimidine fused oxazine 32 exhibited an EC50 value of 2.78 nM.
4.4 Linear core–based agonists Removal of the oxadiazolo core unit of 8 has been the focus of recent SAR optimization work [52]. The linear ether derivative 33a was equipotent with 8 but was a partial agonist in a yeast reporter assay expressing recombinant human or mouse GPR119. Repositioning the ether linkage as in pyridyl ether, 33b produced a marginal improvement in potency (EC50 ¼ 0.5 mM) without increasing in vitro efficacy. Exchange of the pyridine ring with a 3-fluoro-4-(methanesulfonyl)phenyl moiety produced PSN119-1M 35, which afforded a dramatic increase in efficacy (EC50 ¼ 0.2 mM, Emax ¼ 392%), implicating the hydrogen bond acceptor feature as a critical driver of potency and efficacy at human GPR119.
O
O
O
O
O
N
N
O N
Y O
X
N
F
F SOMe
33a. Y = O, X = CH2 33b. Y = CH2, X = O
O
34
SO2Me 35
Extensive pharmacological characterization of racemic sulfoxide PSN119-1 34 (EC50 ¼ 0.5 mM; Emax ¼ 407% yeast reporter assay) has been reported [52–54]. Unlike the glucose-independent insulin secretagogue repaglinide (1 mg/kg, po), 34 (30 mg/kg, po) did not cause hypoglycemia when evaluated in an oGTT in high fat diet–fed (10 days) SD rats. The anti-hyperglycemic effects of 34 were confirmed in 13-week-old diabetic ZDF rats, where its effect on glucose tolerance was similar to that
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
165
of the DPP-IV inhibitor P32/98. In contrast, its glucose-lowering effect was drastically reduced during an IVGTT, pointing to a significant contribution from gastrointestinal tract (GI)-derived incretins. Like the prescribed anorectic agent sibutramine (5 mg/kg, po), 34 (50 mg/kg, po) facilitated a statistically significant reduction in cumulative food intake at both the 6 and 24 h time points post dose. Following oral administration to SD rats, sulfoxide 34 was reported to be extensively metabolized to sulfone 35 [52]. Further pharmacological characterization demonstrated that sulfoxide 34 elicited a dose-responsive effect on GLP-1 secretion in murine GLUTag cells (EC50 ¼ 153770 nM). This was confirmed in a meal tolerance test using P32/98 (50 mg/kg, po) in combination with 34 (100 mg/kg) in rats. In addition, 34 (30 mg/kg, po) inhibited gastric emptying in rats in a manner similar to the antiobesic agent HMR1426 (50 mg/kg, po). Subchronic daily dosing of 34 to high fat diet–fed rats for 21 days (100 mg/kg/qd po) significantly lowered plasma glucose and insulin levels during an oGTT test performed on day 21 and also decreased fat pad mass and plasma leptin levels [53,54]. In addition, subchronic once daily oral administration (100 mg/kg/day) slowed the progression of diabetes in a study in young db/db mice, as highlighted by the ability of 34 to maintain normal fed blood glucose levels and glucose tolerance following an oGTT on day 21 of the experiment. Further optimization of 34 afforded development candidate PSN821 [54,55]. Although the structure of PSN821 has not been disclosed, it may be encompassed by generic structure 36, which captures the essential features of a series of GPR119 agonists disclosed in five recent patent applications [56–60]. Little biological and pharmacological characterization of these analogs is provided. However, the methods associated with the biological assays used to characterize the examples are given; a yeast-based reporter assay expressing human and mouse GPR119, a cAMP AlphascreenTM, an in vivo feeding study, HIT-T15 cAMP and insulin secretion assays, and an oGTT performed in either SD rat, male C57Bl/6, or male ob/ob mice.
N
N O
N
Me O E R1
Q R2
36
36a. E = CH, Q = CH, R2 = F, R1 = SO2CH3 36b. E = CH, Q = CH, R2 = F, Me, R1 = CONHR 36c. E = CH, Q = CH, R2 = F, R1 = CH2SO2CH3 36d. E = CH, Q = N, R2 = Me, R1 = CONHR 36e. E = N, Q = CH, R2 = Me, R1 = NHCMeCH2OH
166
Robert M. Jones and James N. Leonard
Other linear core–based agonists include a recently disclosed series of [[1,2,3,4-tetrahydro-2-(methylsulfonyl)-6-isoquinolinyl]oxy]propyl derivatives exemplified by 37 [61]. The compounds of this genus were characterized using Flp-In-CHO-hGPR119 cells wherein compound 37 showed a concentration-dependent increase in cAMP levels with an EC50 o100 nM. In addition, it was reported that the simple linear thioester 38 attenuated fed blood glucose levels in diabetic db/db mice [62]. Separately, it was reported that 38, which raised intracellular cAMP levels with an EC50 of 3.2 mM in 293-EBNA cells, stimulated insulin secretion and improved glucose tolerance in an oGTT when administered at the dose of 100 mg/kg ip to SD rats [63]. GPR119 agonist 38 also demonstrated antihyperglycemic effects in diabetic Goto–Kakizaki rats following oral administration at a dose of 100 mg/kg. O N
O
O
N SO2Me 37
S
O
N 38
5. CLINICAL TRIAL STATUS AND FUTURE PROSPECTS To date, four GPR119 agonists have been reported to be in the early stages of clinical development. The structures of these compounds have not been revealed. The first compound advanced to the clinic, APD-668, was co-developed by Arena Pharmaceuticals and Ortho McNeil [64–67]. Following two initial phase I trials, it was reported in January 2008 that APD-668 may improve glucose control in diabetic patients [68]. Subsequently, further development of APD-668 was halted and APD597 was advanced into a phase I trial in December 2008 [69]. Metabolex initiated a phase I trial with MBX-2982 in March 2008 [70], and in November 2008, phase 1a data were presented [71]. The placebocontrolled trial evaluated 10–1,000 mg of MBX-2982 in healthy volunteers. The drug was rapidly absorbed, had good exposure, and
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
167
demonstrated a half-life consistent with once-daily dosing. MBX-2982 also caused dose-dependent reductions in glucose and increases in GLP1 in these individuals following a mixed meal. All doses were well tolerated with no drug-related adverse events reported; at that time, a multiple-dose, pharmacokinetics, and pharmacodynamics phase Ib trial was under way [72]. In September 2008, OSI Pharmaceuticals began a phase I trial of PSN-821 [55]. To date, no results have been reported for this compound. It is clear from these reports that there are presently insufficient data to confirm the promise of GPR119 agonists suggested by pre-clinical studies. Nevertheless, the early reports are encouraging. Based on the robust activity around this target in the patent literature, GPR119 agonists are very likely to undergo vigorous clinical testing over the next few years, and the therapeutic value of this approach should clarify significantly. Clearly, the paramount consideration will be to determine the robustness of this mechanism with regard to the lowering of fasting plasma glucose and HbA1c in diabetic patients. The effects of GPR119 agonists on incretin and insulin levels will also be extremely valuable information to clinicians, who ultimately need to determine where GPR119 agonists belong in the treatment paradigm for T2DM.
REFERENCES [1] [2] [3] [4] [5] [6]
[7]
[8]
[9] [10]
[11]
J. C. Seidell, Br. J. Nutr., 2000, 83(Suppl. 1), S5. UK Prospective Diabetes Study Group., Lancet, 1998, 352, 854. D. J. Drucker, Cell Metab, 2006, 3, 153. D. J. Drucker and M. A. Nauck, Lancet, 2006, 368, 1696. R. Fredriksson, P. J. Ho¨glund, D. E. I. Gloriam, M. C. Lagerstrom and H. B. Schio¨th, FEBS Lett, 2003, 554, 381. T. Soga, T. Ohishi, T. Matsui, T. Saito, M. Matsumoto, J. Takasaki, S. Matsumoto, M. Kamohara, H. Hiyama, S. Yoshida, K. Momose, Y. Ueda, H. Matsushime, M. Kobori and K. Furuichi, Biochem. Biophys. Res. Commun., 2005, 326, 744. Z.-L. Chu, R. M. Jones, H. He, C. Carroll, V. Gutierrez, A. Lucman, M. Moloney, H. Gao, H. Mondala, D. Bagnol, D. Unett, Y. Liang, K. Demarest, G. Semple, D. P. Behan and J. Leonard, Endocrinology, 2007, 148, 2601. Z.-L. Chu, C. Carroll, J. Alfonso, V. Gutierrez, H. He, A. Lucman, M. Pedraza, H. Mondala, H. Gao, D. Bagnol, R. Chen, R. M. Jones, D. P. Behan and J. Leonard, Endocrinology, 2008, 149, 2038. Y. Sakamoto, H. Inoue, S. Kawakami, K. Miyawaki, T. Miyamoto, K. Mizuta and M. Itakura, Biochem. Biophys. Res. Commun., 2006, 351, 474. H. A. Overton, A. J. Babbs, S. M. Doel, M. C. T. Fyfe, L. S. Gardner, G. Griffin, H. C. Jackson, M. J. Proctor, C. M. Rasamison, M. Tang-Christensen, P. S. Widdowson, G. M. Williams and C. Reynet, Cell Metab, 2006, 3, 167. Z. Chu, R. Jones, R. Chen, C. Carroll, V. Gutierrez, A. Lucman, D. P. Behan and J. Leonard, Keystone Symposium. Diabetes: Molecular Genetics, Signalling Pathways and Integrated Physiology, 2007, Abstract 230.
168
Robert M. Jones and James N. Leonard
[12] L. Lauffer, R. Iakoubov and P. L. Brubaker, Diabetes, 2009 Feb 17. [Epub ahead of print] PMID: 19208912. [13] J. Fu, G. Astarita, S. Gaetani, J. Kim, B. F. Cravatt, K. Mackie and D. Piomelli, J. Biol. Chem., 2007, 282, 1518. [14] J. Fu, S. Gaetani, F. Oveisi, J. Lo Verme, A. Serrani, F. Rodriguez de Fonseca, A. Rosengarth, H. Luecke, B. Di Giacomo, G. Tarzia and D. Piomelli, Nature, 2003, 425, 90. [15] H. Lan, G. Vassileva, A. Corona, L. Liu, H. Baker, A. Golovko et al., Keystone Symposium. Diabetes: Molecular Genetics, Signalling Pathways and Integrated Physiology, 2007, Abstract 253. [16] Y. Ning, K. O’Neill, H. Lan, L. Pang, L. X. Shan, B. E. Hawes and J. A. Hedrick, Br. J. Pharmacol., 2008, 155, 1056. [17] S. M. Huang, T. Bisogno, M. Trevisani, A. Al-Hayani, L. De Petrocellis, F. Fezza, M. Tognetto, T. J. Petros, J. F. Krey, C. J. Chu, J. D. Miller, S. N. Davies, P. Geppetti, J. M. Walker and V. Di Marzo, Proc. Natl. Acad. Sci. U.S.A., 2002, 99, 8400. [18] G. Eisenhofer, A. Aneman, P. Friberg, D. Hooper, L. Fa˚ndriks, H. Lonroth, B. Hunyady and E. Mezey, J. Clin. Endocrinol. Metab., 1997, 82, 3864. [19] E. Eldrup and E. A. Richter, Am. J. Physiol. Endocrinol. Metab., 2000, 279, E815. [20] H. E. Parker, A. M. Habib, G. J. Rogers, F. M. Gribble and F. Reimann, Diabetologia, 2009, 52, 289. [21] C. Ammala, S. Bullard, J. Kashatus, S. Katamreddy, J. Way and A. Carpenter, Keystone Symposium, Islet and b-Cell Biology, 2008, Abstract #102. [22] G. Semple, B. Fioravanti, G. Pereira, I. Calderon, J. Uy, K. Choi, Y. Xiong, A. Ren, M. Morgan, V. Dave, W. Thomsen, D. J. Unett, C. Xing, S. Bossie, C. Carroll, Z. L. Chu, A. J. Grottick, E. K. Hauser, J. N. Leonard and R. M. Jones, J. Med. Chem., 2008, 51(17), 5172. [23] G. Semple, Abstracts of Papers, 41st Western Regional Meeting of the American Chemical Society, San Diego, CA, United States, October 9–13 2007, GEN-033. [24] R. M. Jones, Abstracts of Papers, 232nd ACS National Meeting, San Francisco, CA, United States, Sept. 10–14, 2006, MEDI-275. [25] Z. Chu, C. Carroll, V. Gutierrez, A. Lucman, M. Moloney, H. Gao et al., Keystone Symposium. Diabetes: Molecular Genetics, Signalling Pathways and Integrated Physiology, Keystone, Colorado, USA, 14–19 January 2007, Abstract 117. [26] R. M. Jones and J. Lehmann, WO Patent Application 2007035355 A2, 2007. [27] R. M. Jones and J. Lehmann, WO Patent Application 2008005576 A1, 2008. [28] R. M. Jones, J. Lehmann and A. Siu-Ting Wong, WO Patent Application 2008005569 A2, 2008. [29] J. Fang, J. Tang, A. J. Carpenter, G. Peckham, C. R. Conlee, S. K. Du and S. R. Katamreddy, WO Patent Application 2008070692, 2008. [30] P. Brandt, G. Johansson, L. Johansson, T. Koolmeister, B. M. Nilsson and T. Sandvall, WO Patent Application 2008025799, 2008. [31] D. A. Wacker, K. A. Rossi and Y. Wang, WO Patent Application 2009012277 A1, 2009. [32] H. B. Wood, A. D. Adams, S. Freeman, J. W. Szewczyk, C. Santini and Y. Huang, WO Patent Application 2008085316 A1, 2008. [33] M. C. T. Fyfe, H. Overton, J. White, R. Jones, R. V. Sorensen and C. Reynet, Diabetes, 2006, 55(Suppl. 1), 346-3OR. [34] M. C. T. Fyfe, A. J. Babbs, L. S. Bertram, S. E. Bradley, S. M. Doel, S. Gadher, W. T. Gattrell, R. P. Jeevaratnam, J. F. Keily, J. G. McCormack, H. A. Overton, C. M. Rasamison, C. Reynet, C. P. Sambrook Smith, V. K. Shah, D. F. Stonehouse, S. A. Swain, J. R. White, P. S. Widdowson, G. M. Williams and M. J. Procter, Abstracts of Papers, 236th ACS National Meeting, Philadelphia, PA, United States, August 17–21, MEDI-197, 2008.
The Emergence of GPR119 Agonists as Anti-Diabetic Agents
169
[35] A. Carpenter, C. Ammala, C. Briscoe, S. Bullard, J. Kashatus, S. Katamreddy, R. Mertz and S. Ross, Symposium, Islet and b-cell biology, 2008, Abstract #102. [36] L. S. Bertram, M. C. T. Fyfe, M. J. Procter and G. M. Williams, WO Patent Application 2007116229 A1, 2007. [37] R. M. Jones, G. Semple, Y. Xiong, Y-J Shin, A. S. Ren, I. Calderon, K. Choi, B. Fioravanti, J. Lehmann and M. A. Bruce, WO Patent Application 2005007647A1, 2005. [38] R. M. Jones, G. Semple, Y. Xiong, Y-J Shin, A. S. Ren, I. Calderon and K. Choi, WO Patent Application 2005121121A2, 2005. [39] R. M. Jones, J. Lehmann, A. Wong, D. Hirst and Y-J Shin, US Patent Application 2006155128 A1, 2006. [40] R. M. Jones, J. Lehmann, A. Wong, D. Hirst and Y-J Shin, US Patent Application 2007167473 A1, 2007. [41] R. M. Jones, G. Semple, Y. Xiong, Y-J Shin, A. S. Ren, I. Calderon, K. Choi, B. Fioravanti and C. R. Sage, WO Patent Application 2005007658 A2, 2005. [42] X. Chen, P. Cheng, E.L. Clemens, J.D. Johnson, J. Ma, A. Murphy, I. Nashashibi, C.J. Rabbat, J. Song, M.E. Wilson, Y. Zhu and Z. Zhao, WO Patent Application 2008083238 A2, 2008. [43] J. Ma, C. J. Rabbat, J. Song, X. Chen, I. Nashashibi, Z. Zhao, A. Novack, D.-F. Shi, P. Cheng, Y. Zhu and A. Murphy, WO Patent Application 2009014910 A2, 2009. [44] P. B. Alper, G. Lelais, R. Epple and D. Mutnick, WO Patent Application 2008109702 A1, 2008. [45] M. Wilson, F. Gregoire, B. Pandey, A. Chandalia, P. Zhang, O. Abdel-aleem, K. Marlen, E. Clemens, J. Johnson, X. Chen, J. Ma, I. Nashashibi, C. Rabbat, J. Song, A. Novack, D-F. Shi, S. Zhao and B. Lavan, Keystone Symposium, Type II Diabetes and Insulin Resistance, Banff, Canada, Abstract #369, January 2009. [46] S. R. Katamreddy, R. D. Caldwell, D. Heyer, V. Samano, J. B. Thompson, A. J. Carpenter, C. R. Conlee, E. E. Boros and B. D. Thompson, WO Patent Application 2008008887 A2, 2008. [47] C. Ammala and C. Briscoe, WO Patent Application 2008008895 A1, 2008. [48] C. D. Boyle and B. R. Neustadt, WO Patent Application 2008130615 A1, 2008. [49] C. D. Boyle, S. Chackalamannil, C. M. Lankin, U. Shah, B. R. Neustadt, H. Liu and A. W. Stamford, WO Patent Application 2008130584 A1, 2008. [50] J. M. Fevig and D. A. Wacker, WO Patent Application 2008137436 A1, 2008. [51] J. M. Fevig and D. A. Wacker, WO Patent Application 2008137435 A1, 2008. [52] M. C. T. Fyfe, A. Babbs, L. S. Bertram, S. E. Bradley, S. M. Doel, S. Gadher, W. T. Gattrel, J. G. Horswill, R. P. Jeevaratnam, J. F. Keily, J. G. McCormack, H. A. Overton, C. M. Rasamison, C. Reynet, P. J. Rushworth, C. P. Sambrook Smith, V. K. Shah, D. F. Stonehouse, S. A. Swain, J. R. White, P. S. Widdowson, G. M. Williams and M. J. Proctor, Abstracts of Papers, 234th ACS National Meeting, Boston, MA, United States, August 19–23, 2007, MEDI-062. [53] M. C. T. Fyfe, J. White, P. Widdowson, H. Overton and C. Reynet, Diabetes, 2007, 56(Suppl. 1), 532-P. [54] M. C. T. Fyfe, J. White, P. Widdowson, H. Overton and C. Reynet, 67th Sessions ADA Meeting, Chicago, Illinois, USA, 2007. [55] OSI Pharmaceuticals Initiates Clinical Development Program for Anti-Diabetes Candidate, PSN821, Company Press Release, September 3, 2008. [56] M. C. T. Fyfe, J. Keily and S. A. Swain, WO Patent Application 2008081204 A1, 2008. [57] L. S. Bertram, M. C. T. Fyfe, R. P. Jeevaratnam, J. Keily and S. A. Swain, WO Patent Application 2008081205 A1, 2008. [58] M. C. T. Fyfe, J. Keily, M. Proctor, D. F. Stonehouse and S. A. Swain, WO Patent Application 2008081206 A1, 2008.
170
Robert M. Jones and James N. Leonard
[59] M. C. T. Fyfe, R. P. Jeevaratnam, J. Keily and S. A. Swain, WO Patent Application 2008081207 A1, 2008. [60] M. C. T. Fyfe, R. P. Jeevaratnam, J. Keily and S. A. Swain, WO Patent Application 2008081208 A1, 2008. [61] P. Alper, M. Azimioara, C. Cow, R. Epple, S. Jiang, G. Leleais, P. Michellys, T. N. Nguyen, L. Westcott-Baker and B. Wu, WO Patent Application 2008097428 A2, 2008. [62] T. Ohishi, J. Takasaki, M. Matsumoto, T. Saito, M. Kamohara, T. Soga, S. Yoshida and Y. Ueda, WO Patent Application 2002044362 A1, 2002. [63] T. Ohishi, J. Takasaki, M. Matsumoto, T. Saito, M. Kamohara, T. Soga, S. Yoshida and Y. Ueda, EP Patent Application 1338651 A1. [64] Arena Pharmaceuticals Announces Global Diabetes Collaboration with Ortho-McNeil Pharmaceutical, December 21, 2004, http://invest.arenapharm.com/releasedetail.cfm? ReleaseID ¼ 320780 [65] Arena Pharmaceuticals Announces Selection of Two Arena-Discovered Compounds for Preclinical Development by Ortho-McNeil, December 23, 2004, http://invest. arenapharm.com/releasedetail.cfm?ReleaseID ¼ 320778 [66] Arena Pharmaceuticals, Inc. Announces Initiation of Phase 1 Clinical Trial of Arena Type 2 Diabetes Drug Candidate in Collaboration With Ortho-McNeil, February 7, 2006, http://invest.arenapharm.com/releasedetail.cfm?ReleaseID ¼ 320321 [67] Arena Pharmaceuticals Announces That Ortho-McNeil Extends Research Term Under Partnership to Develop Drugs to Treat Type 2 Diabetes, September 27, 2006, http:// invest.arenapharm.com/releasedetail.cfm?ReleaseID ¼ 320291. [68] Arena Pharmaceuticals Announces APD668 Initial Clinical Study Results Suggest Glucose-Dependent Insulinotropic Receptors May Improve Glucose Control in Patients With Type 2 Diabetes, January 7, 2008, http://invest.arenapharm.com/releasedetail. cfm?ReleaseID ¼ 320208 [69] Arena Pharmaceuticals Announces Initiation of Phase 1 Clinical Trial of Type 2 Diabetes Drug Candidate in Partnership with Ortho-McNeil-Janssen Pharmaceuticals, December 15, 2008, http://invest.arenapharm.com/releasedetail.cfm?ReleaseID ¼ 354391 [70] Metabolex initiates Phase 1 Trial of MBX-2982, March 26, 2008, http://www.metabolex. com/news/mar262008.html [71] S. Zhao, F. Gregoire, E. Clemens, D. Karpf, X. Chen, B. Lavan, J. Johnson and M. Wilson, The World Congress on Controversies to Consensus in Diabetes, Obesity and Hypertension (CODHy). October 30–November 2, 2008, Barcelona, Spain. [72] Metabolex Announces Positive Results from Phase 1a Clinical Trial of MBX-2982, November 12, 2008, http://www.metabolex.com/news/nov122008.html
CHAPT ER
8 Non-Peptide Ligands for the Gonadotropin Receptors Nicole van Straten and Marco Timmers
Contents
1. Introduction 2. Non-Peptide Small Molecule Gonadotropin Receptor Ligands 2.1 LHR agonists 2.2 LHR antagonists 2.3 FSHR agonists 2.4 FSHR antagonists 3. Therapeutic Indications and Clinical Findings 3.1 Infertility 3.2 Contraception and disorders of the reproductive tract 4. Conclusions and Future Prospects References
171 173 173 175 175 179 182 182 183 184 185
1. INTRODUCTION The gonadotropins luteinizing hormone (LH) and follicle-stimulating hormone (FSH) play fundamental roles in the complex process of human reproduction. LH and FSH belong to a family of glycoprotein hormones, also including choriogonadotropin (CG) and thyroid-stimulating hormone (TSH). These glycoproteins are large, non-covalently linked heterodimers consisting of a common a-subunit and a hormone-specific Schering-Plough Research Institute, Department of Medicinal Chemistry, PO Box 20, 5340BH Oss, The Netherlands Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04408-X
r 2009 Elsevier Inc. All rights reserved.
171
172
Nicole van Straten and Marco Timmers
b-subunit. The a-subunit contains 92 amino acids, whereas the b-subunit incorporates 121 (LH) and 111 (FSH) amino acids [1]. LH induces ovulation in females and controls testosterone production in males. FSH is responsible for ovarian follicle growth in females and is involved in spermatogenesis in males. LH and FSH are secreted from the anterior pituitary gland and transported to the gonads to activate the G-proteincoupled LH receptor (LHR) and FSH receptor (FSHR) respectively. LHRs are present on theca cells (women) and Leydig cells (men), whereas FSHRs can be found on granulosa cells (women) and Sertoli cells (men). Binding of LH stimulates the theca and Leydig cells to produce testosterone. Subsequently, testosterone is converted into estradiol in the female granulosa cell by the cytochrome P450 enzyme aromatase, which is generated after activation of the FSHR. CG is produced during pregnancy and is involved in maintaining early pregnancy. Interestingly, CG and LH both bind and activate the LHR with similar potency. H2N
LH
Testosterone
HOOC LH receptor GnRH (Theca cells) H2N
FSH
Aromatase
Estradiol
Follicle growth and ovulation
HOOC FSH receptor (granulosa cells)
FSH and CG are applied clinically for the treatment of infertility during assisted reproductive therapy (ART). In in vitro fertilization (IVF) protocols, multiple oocytes are generated by controlled ovarian stimulation (COS) with FSH. After downregulation of the hypothalamicpituitary-gonadal axis with a gonadotropin-releasing hormone (GnRH) receptor agonist or antagonist, FSH is used during the first half of the cycle to induce follicle growth. CG is then administered to induce the final oocyte maturation, after which the oocytes can be retrieved for fertilization. In ovulation induction protocols, FSH is used to induce
Non-Peptide Ligands for the Gonadotropin Receptors
173
monofollicular growth and CG is applied to effect oocyte maturation and ovulation. The currently marketed hormones are obtained either from urinary sources or produced using recombinant biotechnology and are administered parenterally through multiple injections. The development of orally administered non-peptide LH and FSH mimics would be a more convenient approach to infertility treatment and may improve patient compliance. Moreover, synthetically prepared low-molecular-weight (LMW) gonadotropin mimics should exhibit improved stability, consistency and homogeneity compared to their glycoprotein counterparts. Since chemical synthesis does not involve the use of human- or animalderived materials (from urinary origin or serum required for recombinant protein expression), the danger of viral contamination or transmissible spongiform encephalopathy would be eliminated. The risk for ovarian hyperstimulation syndrome that is associated with the long half-life of CG may also be reduced. Conversely, antagonists of the LHR and/or FSHR may also prove to be valuable as additions to the currently available palette of contraceptive methods. The discovery and development of non-peptide, orally available LHR or FSHR antagonists would open the way to innovative non-steroidal and gonad-specific contraceptive methods. In a review by Guo in 2005 [2], an excellent assessment of non-peptide ligands for the gonadotropin receptors was presented. In 2008, Heitman and IJzerman [3] reviewed the literature on ligands for the hypothalamicpituitary-gonadal axis with selective coverage of the patent literature related to FSHR ligands. The current review will briefly summarize the patent literature on gonadotropin ligands up to 2005 and new literature on compound classes and biological data that has appeared during the past 3–4 years.
2. NON-PEPTIDE SMALL MOLECULE GONADOTROPIN RECEPTOR LIGANDS The discovery of LMW agonists for G-protein-coupled receptors (GPCRs) that are normally activated by large proteins has met with limited success. Attempts to mimic the action of FSH with small peptide-like molecules have turned out to be unsuccessful [4]. Non-peptide antagonists for peptide-binding GPCRs are more often documented in literature. Since the late 1990s, several breakthrough starting points for both non-peptide agonists and antagonists of the gonadotropin receptors have emerged as a result of high-throughput screening. However, optimization of these initial hits into suitable drug candidates has proven to be a challenging task. Several factors are responsible for this, the most
174
Nicole van Straten and Marco Timmers
important of which is the difficulty in reducing lipophilicity while retaining potency. The relatively high lipophilicity reflects the assumed binding mode of these non-peptide ligands. Whereas the endogenous ligand is known to bind to the large extracellular domain of the receptor, which is hydrophilic in nature, LMW compounds are believed to bind to the lipophilic transmembrane domain of the receptor [5,6].
2.1 LHR agonists To date, two distinct classes of LHR agonists have been reported in the literature. This low number can partly be explained by the difficulty in finding non-peptide agonists for protein-binding GPCRs. On the contrary, LHR ligands have applications in limited market segments that are currently explored only by a relatively small number of laboratories. 1 R = OMe (Org 41841) R 2R=
H N N
NH2
O
O
N S
S
N
H N
HN
3R=
N O
O
N
(Org 43553) O
OH
N
NH N NH2
O
N O
N
N H
O
NH2
OH
N
4
5
The first compound class is characterized by a bicyclic thieno-[2,3-d]pyrimidine scaffold in 2000 [7]. Org 41841 (1, human LHR: EC50 ¼ 20 nM)
Non-Peptide Ligands for the Gonadotropin Receptors
175
was reported as the first orally active LMW agonist of the LHR. A model where immature mice were primed with FSH and then treated with Org 41841 resulted in ovulation in 40% of the animals at a dose of 50 mg/kg [8]. It was established that replacement of the pyrimidine scaffold by a pyridine scaffold was tolerated, as was substitution of the five-membered thiophene ring by a six-membered ring (4) [9]. Moreover, the thiophene ring could be replaced by a furan or pyrrole moiety. Methylation of the t-butylamide functionality was reported to yield a combined LHR/TSH receptor (TSHR) agonist [10]. Using a model of the Org 41841/TSHR complex, Org 41841 was later used as a starting point to identify TSHR antagonists in a rational design approach [11]. Lead optimization of Org 41841 yielded Org 43553 (2, human LHR: EC50 ¼ 3.7 nM), in which the –NH2 substituent of the 4-phenyl group is derivatized as a morphilinoacetamide [12]. 3H-labelled Org 43553 was used to assess the relationship between LHR binding affinity (Ki) and functional activity (EC50) of a series of related analogues, which was found to be linear [13]. Org 43553 was active in an ex vivo proof-ofprinciple ovulation induction assay and showed promising pharmacokinetic properties (rat, p.o.: Cmax ¼ 3.8 mg/L, T1/2 ¼ 4.5 h, F ¼ 79%). It is the first compound reported to induce complete ovulation in cyclic rats after single-dose oral administration. Fertilization of the Org 43553induced oocytes resulted in normal implantation of healthy embryos [14]. Based on its in vivo activity and bioavailability, Org 43553 was selected as a development candidate and progressed into phase 1 clinical studies. Interestingly, Org 43553 was reported to have some FSHR agonistic activity (human FSHR: EC50 ¼ 110 nM) in addition to its LHR agonistic activity. This phenomenon was further exemplified in so-called open chain analogues such as 3 (human LHR: EC50 ¼ 2.7 nM, human FSHR: EC50 ¼ 5.7 nM) [15]. A series of 1-phenylpyrazoles was disclosed in 2001 [16]. Solid-phase synthesis of focused libraries around the original hit and subsequent lead optimization yielded compound 5 (human LHR: EC50 ¼ 20 nM). When administered intraperitoneally, 5 induced increased testosterone levels in male rats in a dose-dependent manner [17]. As with the thieno[2,3-d] pyrimidine series, the pyrazoles also showed activity at the FSHR (human FSHR: EC50 ¼ 62 nM).
2.2 LHR antagonists A set of previously disclosed compounds, some of which are also phosphodiesterase inhibitors, were claimed as LHR antagonists for the treatment of estrogen deficiencies. Contraceptive applications were also claimed, but no pharmacological data were presented to support this [18].
176
Nicole van Straten and Marco Timmers
2.3 FSHR agonists In the field of non-peptide FSHR agonists, significantly more literature on discrete compound classes has appeared when compared to the nonpeptide LHR agonists. Piperidinecarboxamide 6, which was claimed for infertility treatment, was the first FSH mimetic described (human FSHR: EC50 ¼ 3.9 nM). Compound 6 induced estradiol production in a primary rat granulosa cell assay (EC50 ¼ 1.2 mM). No pharmacokinetic or in vivo efficacy data were revealed [19]. A series of 5-alkylated thiazolidinones displayed FSHR agonistic activity [20]. The thiazolidinone derivatives were prepared using solid-phase combinatorial chemistry methods [21]. A major drawback of structure 7 (human FSHR: EC50 ¼ 14 nM) proved to be the lack of control over the relative stereochemistry at the 2- and 5-positions of the thiazolidinone ring during synthesis. Furthermore, this compound was prone to isomerization after chiral purification, possibly caused by the activated hydrogen at position 5. To overcome the chemical instability, a methyl group was introduced at position 5 [22], furnishing compound 8 (human FSHR: EC50 ¼ 51 nM). In a second stabilization approach, the thiazolidinone scaffold was replaced by a g-lactam unit [23], exemplified by compound 9 (human FSHR: EC50 ¼ 25 nM). These compounds were shown to interact within the seven-transmembrane region of the FSHR in studies using FSH/TSH receptor chimeras [24]. O H N
N O O
R H N
N
H2N
X O
O
O
O O
N
O
O
6
7 X = S, R = H 8 X = S, R = CH3 9 X = CH2, R = CH3
R1 R2
O
N R3
N N
H N
N O
O
S
N
OO
N
S
10 R1 = R2 = R3 = OMe
12 N
11 R1 = R2 = H, R3 = O
N H
N
177
Non-Peptide Ligands for the Gonadotropin Receptors
Optimization of a micromolar hit obtained from high-throughput screening of an encoded combinatorial library of diketopiperazines yielded biaryl derivatives 10 and 11 as leads [25,26]. Trimethoxy derivative 10 and aminomethylurea 11 were reported to activate the FSHR with EC50 values of 13 and 1.2 nM respectively. The compounds were claimed for use in controlling fertility, for contraception or for treatment of hormone-dependent disorders [27].
O N O CF3 H 2N H N
N O
O
N
13
O
14
O N
S N S
N N O
N N
NH
15
N-Alkylated sulfonylpiperazines such as compound 12 were claimed as FSHR agonists (human FSHR: EC50 ¼ 13 nM), useful for the treatment of infertility [28]. In 2003, tetrahydroquinoline 13 (human FSHR: EC50 ¼ 4.4 mM) was identified in a screening campaign [29,30]. It is noteworthy to mention that hit optimization resulted in an agonist-toantagonist switch. Derivatization of the 6-amino substituent with lipophilic groups resulted in FSHR antagonists with IC50 values as low
178
Nicole van Straten and Marco Timmers
as 5 nM [30]. Depending on the nature of the 6-amino substituent and on the substituent on the phenyl ring at position 4 (see, e.g., compound 14), analogues showed partial agonistic or partial antagonistic activity with EC50 or IC50 values of less than 100 nM [31,32]. Since the various representatives in this series appear to be able to activate and/or inactivate the receptor, they were claimed for use in fertility regulation, including both infertility treatment and contraception. A series of thiazolyl-isoxazoles such as 15 (human FSHR: EC50 ¼ 1.1 mM) were claimed in a patent application for the treatment of infertility and related disorders [33]. In 2005 and 2006, a series of richly decorated hexahydroquinolines was claimed for the treatment of infertility disorders including COS and in vitro fertilization procedures. Compounds of this series such as 16 were described to have FSHR IC50s of less than 1 nM [34]. A Hantzschtype cyclocondensation reaction was used for the preparation of analogues in this series resulting in a mixture of four diastereoisomers that could be separated using a chiral auxiliary [35]. For example, compound 17 (human FSHR: EC50o1 nM) was obtained in 99.2% diastereomeric excess.
F OMe
F O S N H
O MeO
O OMe
O
Br
O
O CN
CN
N H 16
O
N H 17
In an attempt to reduce the lipophilicity of compounds within this series, pyridyl- and sulfonamide-substituted hexahydroquinolines such as 18 (human FSHR: EC50o10 nM) were prepared [36]. Substitution of a phenyl ring with a less lipophilic aliphatic sulfonamide was also tolerated (19, human FSHR: EC50o10 nM) [37].
179
Non-Peptide Ligands for the Gonadotropin Receptors
O N
HN
O S NH
O
H N
Br
Br
S O
O
O
O
O CN
CN
N H
N H
18
19
2.4 FSHR antagonists In 1992, suramine was reported to inhibit binding of FSH to its receptor [38]. Since suramin is a large heavily sulfonated compound and a notoriously promiscuous binder to a plethora of (glyco)proteins, it is not suitable for use as a non-hormonal gonad-specific contraceptive method. Still, it was the starting point for a more selective FSHR antagonistic sulfonic acid series [39,40]. Although stilbene bis-sulfonic acid 20 was only active in the micromolar range in an in vitro cAMP (cyclic AMP, adenosine 3’, 5’-cyclic monophosphate) assay (human FSHR: IC50 ¼ 1.3 mM), it turned out to be equipotent in an in vitro proof-of-principle assay using human granulosa cells (IC50 ¼ 2.7 mM) [41]. In contrast to other non-peptide ligands of the gonadotropin receptors, stilbene 20 was able to inhibit binding of 125 I-labelled human FSH to the truncated extracellular domain of the human FSHR [42], indicating that this relatively hydrophilic compound may exert its biological action by binding to the extracellular part of the receptor. SMe HO3S
O
H N
O S O O
N
N
O O
S O
N H
SO3H
O
MeS 20
It is known that within a single structural class, both FSHR agonists and FSHR antagonists can be identified, depending on the nature of their
180
Nicole van Straten and Marco Timmers
pharmacophores. This phenomenon can be seen in the thiazolidinone series [43], where compound 21 (human FSHR: IC50 o11 mM) showed antagonistic properties, as opposed to the earlier mentioned analogues 7–9 that showed agonist activity. This can also be seen in the tetrahydroquinoline series (see FSHR agonist 13 and 14). Derivative 22 was reported to be a FSHR antagonist (human FSHR: IC50 ¼ 28 nM), active in an ex vivo mouse follicle culture model, in which it significantly inhibited follicle growth, resulting in 78% inhibition of ovulation [30]. O H N
N
H2N
S
H N
O
H N
O
O
S
N O
21
22
The biphenyl FSHR pharmacophore present in the tetrahydroquinolines (22), diketopiperazines (10) and thiazolyl-isoxazoles (15) is also an essential element of aminoalkylamide FSHR antagonists exemplified by compound 23 (human FSHR: IC50 ¼ 40 nM) [44]. This compound inhibited cAMP production in primary rat granolusa cells (IC50 ¼ 20 nM), and its in vivo effects on progesterone levels in female cyclic rats (p.o. dosing, 20 mg/kg BID) and on the number of sperm cells in the testes of male rats were measured (p.o. dosing, 20 mg/kg BID for 25 days). The series was claimed for contraceptive use and reproductive disorders such as endometriosis, uterine fibroids, polycystic ovary syndrome and breast and ovarian cancers. N H N
O NH2
O N
H N
O S
N Cl N
NH
O O
23
24
N H
OH
181
Non-Peptide Ligands for the Gonadotropin Receptors
Pyrrolobenzodiazepines have also been claimed as FSHR antagonists for contraception [45–48]. Here, again, the biphenyl unit is present as an important pharmacophore element. Compound 24 was found to inhibit cAMP production (human FSHR: IC50 ¼ 70 nM), but in vitro or in vivo efficacy data were not provided. It is of interest to note that the same structural core is present in a series of oxytocin receptor antagonists earlier claimed by the same research group [49]. Acyltryptophanols such as compound 25 were claimed as FSHR antagonists for control of fertility in both men and women or for the prevention and/or treatment of osteoporosis [50]. Inhibition of cAMP production by 25 was reported to be in the submicromolar range (25, human FSHR: IC50 ¼ 100 nM).
OMe OH
25 R1 = nBu-CN, R2 = H, X =
F
R1 N NH X O
26 R1 = Et, R2 = O-iPr, X = R2
OMe
It is tempting to classify the biphenyl substituent in the acyltryptophanol series as another example of this frequently observed substructure in FSHR antagonists. However, it was shown [51] that separation of the two phenyl units by an acetylene linkage provided nanomolar active compounds, demonstrated by 26 (human FSHR: IC50 ¼ 20 nM). In the acetylene series, non-aromatic substituents were also tolerated [52]. Furthermore, it was established that the hydroxymethyl group could be replaced by a cyanomethyl group [53] and that a-substitution of the tryptophanol entity was tolerated [54]. Given the lipophilicity of the compounds, it is likely that representatives from this series bind to the seven helical transmembrane domain of the receptor. The plasticity of the transmembrane part of the receptor is illustrated by structures 27–29. Incorporation of a tetrahydroisoquinoline substituent (27, human FSHR: IC50 ¼ 6 nM) [55], a chromene core (28, human FSHR: IC50 ¼ 14 nM) [56,57] or a tetrahydrocarbazole group (29, human FSHR: IC50 ¼ 70 nM) [58,59] are well tolerated. Interestingly, the indole moiety seems to be the optimal FSHR pharmacophore [60]. No in vivo efficacy data has been reported for these acyltryptophanols.
182
Nicole van Straten and Marco Timmers
H N
F
H N
NH
NH
HO
NH
HO
O
O
O
HO O
O
NH O
F
HN O
HN O
27
28
HN
Cl O 29
3. THERAPEUTIC INDICATIONS AND CLINICAL FINDINGS 3.1 Infertility Infertility is generally defined as ‘the inability to conceive after 12 months of having regular unprotected sexual intercourse’. Women reach their maximum ability to conceive around the age of 24, while the average age of women seeking medical help to conceive is 32. Today 1 out of 6 couples of reproductive age experiences fertility problems [61]. A substantial percentage (10–15%) of infertility is unexplained, but known causes in women are of hormonal (oligo- or amennorhoea) and tubal (blocked fallopian tubes) origin and are related to age. Female infertility is often the result of a disorder of the reproductive tract such as endometriosis, polycystic ovarian syndrome or uterine fibroids. Tubal and pelvic pathology is responsible for 30% of female infertility, and ovulatory dysfunction affects 20% of the patient population. Although male infertility accounts for the remaining 35% of all infertility cases, treatment is focused on the female. This may be explained by the development of novel fertilization techniques, such as intracytosolic sperm injection, allowing in vitro fertilization by male partners with low sperm count and/or mobility. Infertility treatment does not aim at solving the cause of infertility (if known) but rather focuses on ART to become pregnant. Several forms of ART are available
Non-Peptide Ligands for the Gonadotropin Receptors
183
to help couples. In the first instance, anti-estrogens, such as clomiphene citrate, may be used to accomplish ovulation induction. A second-line option is the use of gonadotropins for ovulation induction. This can be followed by intrauterine insemination to instigate a natural, monofollicular conception. To establish multifollicular fertilization in the laboratory, gonadotropins are used for COS and subsequent in vitro fertilization, after which one or more embryos are transferred. The fertility market is growing, and the recombinant and urinary gonadotropins account for the majority of sales. It is envisaged that replacement of these agents by orally active LMW mimics will improve patient convenience and even compliance, reduce the risk of side effects, improve the ‘time-to-pregnancy’ and ultimately may lead to higher ‘takehome-baby-rates’. The current success rates for IVF treatment are about 30%, which represents a major need for improvement. On the contrary, scientific developments in human reproductive biology and medicine have the potential to positively affect pregnancy outcomes [62]. Potential additional indications for LMW LHR and FSHR agonists include male indications such as male infertility, late onset hypogonadism in elderly men, and delayed puberty and mal-descendent testes in boys. Since the majority of the LMW non-peptide agonists for the gonadotropin receptors have only been identified in the past decade, almost no clinical data are available. Moreover, for those in development, clinical proof-of-concept has yet to be reported. For LHR agonist Org 43553 (2), phase I clinical data have been published [63]. Tolerability and pharmacokinetic effects of Org 43553 were assessed in sterilized women of reproductive age following single oral administration of seven escalating doses (5–2,700 mg). Mean elimination half-life varied between 30 h and 47 h, and exposure was dose proportional up to 1,800 mg. Pharmacodynamic evaluation included progesterone measurements and ultrasound to confirm ovulation. Ovulation was observed in 83 and 75% of the women in the 300 and 900 mg groups respectively. No serious adverse events occurred and those reported were mild in severity.
3.2 Contraception and disorders of the reproductive tract In principle, all women of reproductive age are potential users of contraceptive agents. In the western world, there is high penetration of contraceptives in the target population but a declining number of women of childbearing age. Special attention is needed for the developing countries, where thousands of maternal deaths are related to unintended pregnancies every year [64]. The significant population growth in the developing world further emphasizes the importance of broadly available and effective contraceptive methods.
184
Nicole van Straten and Marco Timmers
Female contraception includes core benefits related to pregnancy prevention (the need for which is well met), menstruation disorders and menstruation management. Current contraception methods rely on the prevention of follicle maturation, ovulation and implantation by the action of estrogenic and/or progestagenic steroids [65]. The active pharmaceutical ingredients can be delivered orally, as in the contraceptive pill, or through alternative routes. Most recent advances are developments of new delivery systems such as intrauterine devices [66], vaginal ring [67] or implantable polymeric rod [68,69] containing known steroidal compounds. Several non-steroidal progestins have been reported in the literature [70], and one non-steroidal progestin has thus far progressed into clinical development [71]. The only effective contraceptive in the marketplace still relies on the administration of steroidal drugs. The impact of genomics and proteomics for the discovery of new, validated non-steroidal targets for female contraception is still in the discovery stage [72]. A second drawback is the fact that currently available contraception methods are solely available for women. A safe, 100% effective male contraceptive is regarded as desirable [73]; yet, no male contraceptive pill is on the market. As more and more knowledge of the genes involved in male fertility is generated, new opportunities for male contraception may arise. The currently available palette of male fertility-associated genes may contain good target candidates for discovery of male contraceptives [74]. Thus far, the most obvious and well-validated target in this respect is the FSHR, since mutations may lead to subinfertility or complete infertility in men [75]. Vaccination against FSH-b chain leads to infertility in male primates [76], supporting the concept that the FSHR is a validated target for male contraception. In this light, the discovery of orally available non-peptide antagonists for the gonadotropin receptors may be a novel alternative to contraceptive steroids. Such a discovery would open the way for a new, non-steroidal, gonad-specific approach for contraception in both women and men. Furthermore, numerous disorders of the female reproductive tract are estrogen-related. Since the action of FSH at its receptor influences the levels of estrogens, it is suggested that FSHR antagonists may become valuable drugs to treat, for example, endometriosis, uterine fibroids, polycystic ovarian syndrome and even hormone-dependent breast cancer and ovarian cancer. These assumptions remain speculative since none of the published gonadotropin receptor antagonists has progressed into clinical development.
4. CONCLUSIONS AND FUTURE PROSPECTS The pioneering work in the late 1990s to find non-peptide ligands for protein-binding gonadotropin receptors using high-throughput
Non-Peptide Ligands for the Gonadotropin Receptors
185
screening technologies has induced a paradigm shift in the science of reproduction. Small heterocyclic molecules were identified that completely mimic the action of the endogenous glycoproteins but effect cellular signaling through allosteric receptor-binding sites. At the same time, turning these interesting starting points into bioavailable drug candidates presents a challenging task that requires years of intensive optimization. The past 4–5 years have yielded several new interesting series of LMW ligands for both the LHR and the FSHR that show promise as drug candidates. For LHR agonists, clinical results with Org 43553 indicate that it is possible to mimic the action of the endogenous hormone with an orally available small molecule drug candidate. Although ovulation induction can be regarded as proof of principle, it remains to be established whether this will indeed result in the ultimate proof of concept, for example, induction of pregnancy and live birth. FSHR agonists and antagonists are still early in the clinical development process. Whereas for the LHR agonists in vivo activity has been confirmed in both animal models and humans, translation of the available in vitro and ex vivo results for the FSHR ligands to more relevant animal and clinical models is still a subject of discovery research. In summary, strategies employing screening of large chemical libraries and subsequent hit and lead optimization have resulted in the identification of new chemical entities that mimic or modulate the activity of endogenous gonadotropins. The availability of orally active drugs for the gonadotropin receptors will certainly be of great added value to the current portfolio of parenteral infertility drugs and hormonal, non-gonad-specific contraceptives. The coming decade will reveal whether these scientific promises can be realized.
REFERENCES [1] [2] [3] [4] [5]
[6] [7] [8]
[9]
J. G. Pierce and T. F. Parsons, Ann. Rev. Biochem., 1981, 50, 465. T. Guo, Expert Opin. Ther. Patents, 2005, 15, 1555. L. H. Heitman and A. P. IJzerman, Med. Res. Reviews, 2008, 28, 975. M. Hage-van Noort, W. C. Puijk and R. H. Meloen, WO Patent Application 1992/02542, 1992. C. J. van Koppen, G. J. R. Zaman, C. M. Timmers, J. Kelder, S. Mosselman, R. v. d. Lagemaat, M. J. Smit and R. G. J. M. Hanssen, Naunyn Schmiedeberg’s Arch. Pharmacol., 2008, 378, 503. B. J. Arey, Endocrine, 2008, 34, 1. G. G. Gerritsma, N. C. R. van Straten and A. E. P. Adang, WO Patent Application 2000/ 61586, 2000. N. C. R. van Straten, G. G. Schoonus-Gerritsma, R. G. van Someren, J. Draaijer, A. E. P. Adang, C. M. Timmers, R. G. J. M. Hanssen and C. A. A. van Boeckel, ChemBioChem, 2002, 210, 1023. C. M. Timmers and W. F. J. Karstens, WO Patent Application 2002/24703, 2002.
186
Nicole van Straten and Marco Timmers
[10] M. C. Gershengorn, S. Neumann, C. Thomas, H. Jaeschke, S. Moore, G. Krause, B. Raaka and R. Paschke, WO Patent Application 2007/136776, 2007. [11] S. Neumann, G. Kleinau, S. Costanzi, S. Moore, J. K. Jiang, B. M. Raaka, C. J. Thomas, G. Krause and M. C. Gershengorn, Endocrinology, 2008, 149, 5945. [12] R. G. J. M. Hanssen and C. M. Timmers, WO Patent Application 2003/020726, 2003. [13] L. H. Heitman, J. Oosterom, K. M. Bonger, C. M. Timmers, P. H. G. Wiegerinck and A. P. IJzerman, Mol. Pharm., 2008, 73, 518. [14] R. v. d. Lagemaat, C. M. Timmers, J. Kelder, C. van Koppen, S. Mosselman and R. G. J. M. Hanssen, Hum. Reprod., 2009, 24, 640. [15] R. G. J. M. Hanssen, C. M. Timmers and J. Kelder, WO Patent Application 2003/020727, 2003. [16] H. Shroff, A. P. Reddy, N. El Tayer, N. Brugger and C. Jorand-Lebrun, WO Patent Application 2001/87287, 2001. [17] C. Jorand-Lebrun, B. Brondyk, J. Ling, S. Magar, R. Murray, A. Reddy, H. Shroff, G. Wands, W. Weiser, Q. Xu, S. McKenna and N. Brugger, Bioorg. Med. Chem. Lett., 2007, 17, 2080. [18] C. Grøndahl, WO Patent Application 1999/20223, 1999. [19] N. El tayer, A. Reddy, D. Buckler and S. Magar, WO Patent Application 2000/08015, 2000. [20] R. A. Scheuerman, S. D. Yanofsky, C. P. Holmes, D. Maclean, B. Ruhland, R. W. Barrett, J. E. Wrobel, W. Kao, A. Gopalsamy and F. Sum, WO Patent Application 2002/09706, 2002. [21] D. MacLean, F. Holden, A. M. Davis, R. A. Scheuerman, S. Yanofsky, C. P. Holmes, W. L. Fitch, K. Tsutsui, R. W. Barrett and M. A. Gallop, J. Comb. Chem., 2004, 6, 196. [22] J. Wrobel, J. Jetter, W. Kao, J. Rogers, L. Di, J. Chi, M. C. Pere´z, G.-C. Chen and E. S. Shen, Bioorg. Med. Chem., 2006, 14, 5729. [23] J. C. Pelletier, J. Rogers, J. Wrobel, M. C. Pere´z and E. S. Chen, Bioorg. Med. Chem., 2005, 13, 5986. [24] S. D. Yanofsky, E. S. Shen, F. Holden, E. Whitehorn, B. Aguilar, E. Tate, C. P. Holmes, R. Scheuerman, D. MacLean, M. M. Wu, D. E. Frail, F. J. Lo´pez, R. Winneker, B. J. Arey and R. W. Barrett, J. Biol. Chem., 2006, 19, 13226. [25] T. Guo, A. E. P. Adang, R. E. Dolle, G. Dong, D. Fitzpatrick, P. Geng, K.-K. Ho, S. G. Kultgen, R. Liu, E. McDonald, B. F. McGuinness, K. W. Saionz, K. J. Valenzano, N. C. R. van Straten, D. Xie and M. L. Webb, Bioorg. Med. Chem. Lett., 2004, 14, 1713. [26] T. Guo, A. E. P. Adang, G. Dong, D. Fitzpatrick, P. Geng, K.-K. Ho, C. H. Jilibian, S. G. Kultgen, R. Liu, E. McDonald, K. W. Saionz, K. J. Valenzano, N. C. R. van Straten and M. L. Webb, Bioorg. Med. Chem. Lett., 2004, 14, 1717. [27] T. Guo, K.-K. Ho, E. McDonald, R. E. Dolle, K.W. Saionz, S. G. Kultgen, R. Liu, G. Dong, P. Geng, A. E. P. Adang and N. C. R. van Straten, WO Patent Application 2002/070493, 2002. [28] S. Magar, A. Goutopoulos, M. Schwarz and T. J. Russell, WO Patent Application 2004/ 031182, 2004. [29] N. C. R. van Straten, R. G. van Someren and J. Schulz, WO Patent Application 2003/ 004028, 2003. [30] N. C. R. van Straten, T. H. J. van Berkel, D. R. Demont, W. F. J. Karstens, R. Merkx, J. Oosterom, J. Schulz, R. G. van Someren, C. M. Timmers and P. M. van Zandvoort, J. Med. Chem., 2005, 48, 1697. [31] C. M. Timmers and W. F. J. Karstens, WO Patent Application 2004/056779, 2004. [32] C. M. Timmers and W. F. J. Karstens, WO Patent Application 2004/056780, 2004. [33] V. J. Santora, J. A. Covel, R. Hayashi and R. R. Webb, WO Patent Application 2005/ 087765, 2005.
Non-Peptide Ligands for the Gonadotropin Receptors
187
[34] N. C. R. van Straten, G. G. Gerritsma and L. A. van der Veen, WO Patent Application 2006/117023, 2006. [35] P. M. Grima Poveda, W. F. J. Karstens and C. M. Timmers, WO Patent Application 2006/ 117368, 2006. [36] C. M. Timmers, W. F. J. Karstens and P. M. Grima Poveda, WO Patent Application 2006/ 117370, 2006. [37] W. F. J. Karstens and C. M. Timmers, WO Patent Application 2006/117371, 2006. [38] R. L. Daugherty, A. T. K. Cockett, S. R. Schoen and P. M. Sluss, J. Urol., 1992, 147, 727. [39] J. E. Wrobel, J. F. Rogers and W. Kao, WO Patent Application 2000/58276, 2000. [40] J. E. Wrobel, J. F. Rogers, D. M. Green, W. Kao and J. W. Winfield, WO Patent Application 2000/58277, 2000. [41] J. Wrobel, D. Green, J. jetter, W. Kao, J. Rogers, M. C. Pe´rez, J. Hardenburg, D. C. Deecher, F. J. Lo´pez, B. J. Arey and E. S. Shen, Bioorg. Med. Chem., 2002, 10, 639. [42] B. J. Arey, D. C. Deecher, E. S. Shen, P. E. Stevis, E. H. Meade, Jr., J. Wrobel, D. E. Frail and F. J. Lo´pez, Endocrinology, 2002, 143, 3822. [43] R. A. Scheuerman, S. D. Yanofsky, C. P. Holmes, D. Maclean, B. Ruhland, R. W. Barrett, J. E. Wrobel and A. Gopalsamy, WO Patent Application 2002/09705, 2002. [44] S. J. Coats, D. J. Hlasta, C. L. Carolina, M. J. Macielag, R. Rivero, L. J. Fitzpatrick and K. Pan, WO Patent Application 2001/47875, 2001. [45] A. Failli, G. D. Heffernan, A. A. Santilli, D. A. Quagliato, R. D. Coghlan, P. M. Andrae, S. C. Croce, E. S. Shen and E. J. Trybulski, WO Patent Application 2006/135687, 2006. [46] A. A. Failli, D. Quagliato, P. Andrae, G. D. Heffernan, R. D. Coghlan and E. S. Shen, US Patent Application 2006/0199806, 2006. [47] A. A. Failli, A. A. Santilli, D. A. Quagliato and E. S. Shen, US Patent Application 2006/ 0258644, 2006. [48] A. A. Failli, D. A. Quagliato, G. Heffernan, R. D. Coghlan and E. S. Shen, US Patent Application 2006/0258645, 2006. [49] A. A. Failli, J. S. Shumsky, T. J. Caggiano, J. P. Sabatucci, K. A. Memoli and E. J. Trybulski, WO Patent Application 2002/083683, 2002. [50] L. Wortmann, A. Cleve, H. P. Muhn, G. Langer, A. Schrey, R. Ku¨hne, B. Menzenbach, M. Koppitz and D. Kosemund, WO Patent Application 2007/017289, 2007. [51] L. Wortmann, H. P. Muhn, B. Menzenbach, A. Schrey, R. Ku¨hne, D. Kosemund and M. Koppitz, EP Patent Application 1932831A1, 2008. [52] L. Wortmann, M. Koppitz, H.-P. Muhn, T. Frenzel, F. P. Liesener, A. K. Schrey and R. Ku¨hne, WO Patent Application 2009/013333, 2009. [53] L. Wortmann, M. Koppitz, H.-P. Muhn, T. Frenzel, F. P. Liesener, A. K. Schrey and R. Ku¨hne, WO Patent Application 2009/016253, 2009. [54] L. Wortmann, M. Koppitz, H.-P. Muhn, A. K. Schrey and R. Kuehne, WO Patent Application 2009/021980, 2009. [55] L. Wortmann, H. P. Muhn, B. Menzenbach, A. Schrey, R. Ku¨hne, D. Kosemund and M. Koppitz, WO Patent Application 2008/071453, 2008. [56] L. Wortmann, B. Menzenbach, M. Koppitz, D. Kosemund, H. P. Muhn, A. Schrey, R. Ku¨hne, T. Frenzel and F. P. Liesener, WO Patent Application 2008/071455, 2008. [57] L. Wortmann, B. Menzenbach, M. Koppitz, D. Kosemund, H. P. Muhn, A. Schrey, R. Ku¨hne, T. Frenzel and F. P. Liesener, EP Patent Application 1956016A1, 2008. [58] L. Wortmann, M. Koppitz, H. P. Muhn, T. Frenzel, F. P. Liesener, A. Schrey and R. Ku¨hne, WO Patent Application 2008/0116671, 2008. [59] L. Wortmann, M. Koppitz, H. P. Muhn, T. Frenzel, F. P. Liesener, A. Schrey and R. Ku¨hne, EP Patent Application 1975159A1, 2008. [60] L. Wortmann, M. Koppitz, B. Menzenbach, D. Kosemund, N. Schmees, H.-P. Muhn, T. Frenzel, F. P. Liesener, A. K. Schrey and R. Ku¨hne, WO Patent Application 2009/ 013354, 2009.
188 [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76]
Nicole van Straten and Marco Timmers
Datamonitor, Commercial and pipeline insight: Infertility, October, 2007. M. M. Matzuk and D. J. Lamb, Nat. Med., 2008, 14, 1197. B. Mannaerts, in Proc. 4th World Congress on Ovulation (ed. M. Filicori), p. 157. Global Health Council, 2008, http://www.globalhealth.org/view_top.php3?id ¼ 225. M. A. Economidis and D. R. Mishell, Jr., Exp. Opin. Invest. Drugs, 2005, 14, 449. ESHRE Capri Workshop Group., Hum. Reprod. Update, 2008, 14, 197. F. J. M. E. Roumen, D. Apter, T. M. T. Mulders and T. O. M. Dieben, Human Reprod, 2001, 16, 469. H. B. Croxatto, Contraception, 2002, 65, 15. O. Graesslin and T. Korver, Eur. J. Contracept. Reprod. Health Care, 2008, 13(S1), 4. P. H. H. Hermkens, S. Kamp, S. Lusher and G. H. Veeneman, IDrugs, 2006, 9, 488. J. Bapst, G. Orczyk and J. Ermer, Clin. Pharm. Ther., 2005, 77(Suppl. S(44)). R. J. Aitken, M. A. Baker, G. F. Doncel, M. M. Matzuk, C. K. Mauck and M. J. K. Harper, J. Clin. Invest., 2008, 118, 1330. S. T. Page, J. K. Amory and W. J. Bremmer, Endocrine Rev, 2008, 29, 465. M. M. Matzuk and D. J. Lamb, Nat. Med, 2002, 8, S1–S49. I. Huhtaniemi, J. Reprod. Fert., 2000, 119, 173. N. R. Moughdal, M. Jeyakumar, H. N. Krishnamurthy, S. Sridhar, H. Krishnamurthy and F. Martin, Hum. Reprod. Update, 1997, 3, 335.
CHAPT ER
9 Recent Advances in Coagulation Serine Protease Inhibitors Joanne M. Smallheer and Mimi L. Quan
Contents
1. Introduction 2. Factor Xa Inhibitors 2.1 Compounds in late-stage clinical trials 2.2 Compounds in early clinical trials 2.3 Preclinical compounds 3. Thrombin Inhibitors 3.1 Compounds in clinical trials 3.2 Preclinical compounds 3.3 Dual thrombin/factor Xa inhibitors 4. Factor VIIa/TF Inhibitors 5. Factor IXa and XIa Inhibitors 6. Conclusions References
189 190 190 192 193 195 195 196 198 199 201 203 203
1. INTRODUCTION As orally bioavailable thrombin and factor Xa inhibitors advance through late-stage clinical trials and begin to gain regulatory approvals, a new era in treatment options for thromboembolic disorders may be on the horizon [1–3]. It is thought that these new highly selective inhibitors may eliminate food and drug–drug interactions and obviate the need Bristol-Myers Squibb Company, P.O. Box 5400, Princeton, NJ 08543 Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04409-1
r 2009 Elsevier Inc. All rights reserved.
189
190
Joanne M. Smallheer and Mimi L. Quan
for therapeutic monitoring and dose titration that go hand-in-hand with current oral anticoagulant therapy with warfarin [4]. In addition to thrombin and factor Xa inhibitors, research continues to identify potential new therapeutic agents that target other key serine proteases in the coagulation cascade. This review will summarize recent results for several compounds in clinical trials and survey new small-molecule inhibitors reported in the literature over the past 2–3 years.
2. FACTOR Xa INHIBITORS Factor Xa has been a major focus of pharmaceutical research directed at novel antithrombotics during the past decade because of its central and unique position in the coagulation cascade. Factor Xa is a serine protease located at the junction of the intrinsic and extrinsic pathways, and its inhibition is considered to be an ideal approach to achieve strong efficacy and an improved therapeutic index compared to current therapies. Significant progress has been made in the discovery and development of orally bioavailable small-molecule factor Xa inhibitors. A few inhibitors have been advanced to phase III clinical studies, and the first factor Xa inhibitor was recently approved in Europe and Canada for the prophylaxis of venous thromboembolism (VTE) in patients after elective hip and knee replacement. Several comprehensive reviews have been published on Factor Xa inhibitors [5–7].
2.1 Compounds in late-stage clinical trials Rivaroxaban (BAY 59-7939, 1) is an oxazolidinone derivative with a Xa Ki of 0.4 nM and W10,000-fold selectivity against other related serine proteases [8]. The oral bioavailability of rivaroxaban was found to be 57–66% in rats and 60–86% in dogs with a relatively short half-life (2.3 and 0.9 h, respectively) in both species [9]. In rabbit models of thrombosis, rivaroxaban was effective in prevention and treatment of venous thrombosis [10]. In clinical studies, rivaroxaban was rapidly absorbed and displayed an oral bioavailability of 80% and a half-life of 5–9 h [11]. In a series of phase IIb studies, a range of rivaroxaban doses were found to have efficacy similar to that of standard treatment [12]. The efficacy and safety data from four phase III studies, including two randomized doubleblind trials in patients undergoing total knee arthroplasty, and two trials in patients after hip arthroplasty, have been published [13–16]. Rivaroxaban has been approved in the EU and Canada for the above indications. Phase III trials in chronic indications are ongoing, including treatment of VTE, stroke prevention in patients with atrial fibrillation (AF), and secondary prevention of acute coronary syndrome (ACS).
191
Recent Advances in Coagulation Serine Protease Inhibitors O Cl
S
HN
N O
O
N N
N
N
O
N
O HN
O NH HN
O
O
1
CONMe2
Cl
H2NOC
N O
O
S N NMe
OMe
2
3
Apixaban (BMS-562247, 2) is a second-generation pyrazole-based Xa inhibitor derived from an earlier clinical candidate, razaxaban [17]. The core structures differ in that the pyrazole ring of razaxaban is cyclized onto the amide nitrogen to form the apixaban bicyclic core [18]. Apixaban is a potent and selective Xa inhibitor with a Xa Ki of 0.08 nM and W30,000-fold selectivity over other relevant proteases. It has an excellent pharmacokinetic (PK) profile in dogs with low clearance (0.02 L/kg/h) and a low volume of distribution (0.2 L/kg). Apixaban is effective in rabbit models of thrombosis alone or in combination with aspirin, or aspirin and clopidogrel, without excessive increases in bleeding time [19,20]. Initial studies in humans suggest that apixaban is safe and well tolerated [21] and is rapidly absorbed, with a mean elimination half-life of 8–15 h and an oral bioavailability of 66% [22]. In phase II studies in patients with total knee replacement or for treatment of proximal deep vein thrombosis (DVT), apixaban had efficacy and safety similar to that of standard therapy [23,24]. Both indications are being evaluated in phase III studies, and data from one phase III trial in patients undergoing knee replacement surgery has been presented [25]. Phase II studies have also been completed in patients with recent ACS [26], and for prevention of thromboembolic events in patients with advanced cancer. Additional phase III trials are ongoing for treatment of VTE, stroke prevention in AF, and secondary prevention of ACS. The structure of edoxaban (DU-176b, 3) contains an N,N-dimethylcyclohexane carboxamide scaffold and has a chloropyridine moiety at the P1 position. Edoxaban is a potent and selective Xa inhibitor with a Ki of 0.56 nM and W10,000-fold selectivity over other relevant serine proteases [27]. Edoxaban was shown to produce a dose-dependent inhibition of thrombus formation in rat and rabbit models of thrombosis and did not significantly prolong bleeding time at an antithrombotic dose. Preliminary results from a phase II study show that edoxaban was superior to dalteparin in patients with total hip replacement in terms of incidence of VTE, with little bleeding, and could be dosed once daily [28]. In patients with total knee replacement, edoxaban produced a significant, dosedependent reduction in VTE [29]. A phase II trial of edoxaban in patients with AF showed a similar safety profile with once daily dosing compared
192
Joanne M. Smallheer and Mimi L. Quan
to standard treatment with warfarin [30]. A phase III trial in AF patients was initiated in early 2009.
2.2 Compounds in early clinical trials Eribaxaban (PD 0348292, 4) has a Xa IC50 of 0.32 nM with W1,000-fold selectivity over relevant serine proteases [31]. The oral bioavailability and half-life were 82% and 2.0 h in rats, and 41% and 4.9 h in dogs. Clearance in rats and dogs was 12 and 2.6 mL/kg/min, respectively. Eribaxaban showed a dose-dependent reduction in thrombus weight with an EC50 of 40 ng/mL in the rabbit arteriovenous (A-V) shunt model. A dose-ranging study to determine a dose of eribaxaban equivalent to enoxaparin 30 mg b.i.d. for prevention of VTE in patients undergoing total knee replacement has been completed [32]. Betrixaban (PRT54021, 5) has a Xa Ki of 0.12 nM with W86,000-fold selectivity against related serine proteases [33]. Oral bioavailability and half-life, respectively, were 24% and 8.8 h in rats, 52% and 21.2 h in dogs, and 59% and 9.6 h in monkeys. The antithrombotic efficacy of betrixaban was studied in rabbit, rat, and baboon thrombosis models to determine the target concentration required for clinical trials [34]. Early clinical studies showed that betrixaban was well tolerated across a wide range of doses. A phase II open-label VTE prevention trial showed that betrixaban was effective in preventing VTE after total knee replacement surgery [35]. A phase II trial for prevention of thromboembolic complications in patients with AF has been initiated. LY517717 (6, Xa KiB6 nM; 1,000-fold selectivity over related serine proteases) was non-inferior to enoxaparin in a phase II trial for prevention of VTE after hip or knee replacement surgery and showed similar rates of bleeding [36,37]. YM150 (Xa Ki ¼ 31 nM; low micromolar affinity for other relevant serine proteases), whose structure was not disclosed, is also in multiple phase II trials [38]. In a dose escalation study for VTE in patients undergoing hip replacement surgery, YM150 was effective, safe, and well-tolerated [39]. TAK-442 is another orally bioavailable Xa inhibitor currently in phase II clinical trials, but neither the structure nor any trial results have been disclosed.
MeO
H N N
HN
O O
MeO
F N
N H HN
NH
O N
N
4
5 Cl
Cl
N
O
O
H N
N
O
N N H
O
6
193
Recent Advances in Coagulation Serine Protease Inhibitors
2.3 Preclinical compounds Several series of pyrazole-based inhibitors have been reported. Compound 7 was derived from razaxaban [17] and was designed to prevent the potential formation of a primary aniline by replacing the amide linker with a ketone moiety [40]. Compound 7 has a Xa Ki of 0.11 nM with EC2x values of 5.2 and 5.1 mM in the prothrombin clotting time (PT) and activated partial thromboplastin time (aPTT) assays, respectively. It was found to have lower clearance and higher oral bioavailability in dogs than razaxaban. Compound 8 has a bicyclic core similar to that of apixaban with the P4 group replaced with an a-substituted phenylcyclopropyl moiety [41]. This analog has a Xa Ki of 0.035 nM and PT EC2x of 1.3 mM. Replacement of the fused piperidinone ring of apixaban with fluorophenyl afforded compounds with a 7-fluoroindazole core exemplified by 9, which has a Xa Ki of 4.4 nM and a twofold aPTT EC50 of 4.4 mM [42]. F
H2NOC N
O N
N
O
F3C N
F
H2NOC N N
N
NMe2 O
N
NMe2 N
F NH2
OMe
OMe
7
O N
8
9
Additional Xa inhibitors related to betrixaban possessing the anthranilamide scaffold and a chloropyridine P1 moiety have been reported. This is exemplified by anthranilamide 10, which also employed the P4 piperidinone moiety of apixaban [43]. Compound 10 has a Xa Ki of 0.057 nM and displayed efficacy comparable to that of apixaban in a rabbit A-V shunt thrombosis model. In compound 11 (Xa Ki ¼ 0.005 nM), the P4 phenylpiperidinone of 10 was replaced with a chlorothiophene substituted by a (methylamino(imidazol-1-yl)methyl) moiety [44]. Although this resulted in about a 10-fold improvement in Ki, the PT EC2x values of the two compounds were similar (1.5 mM for 10 and 1.2 mM for 11). Compound 11 has low clearance (4.4 mL/kg/min) and high oral bioavailability (98%) in dogs. The estimated EC50 of 11 in an intravenous (iv) rat vena cava stasis model was 0.36 mg/kg.
N H HN
O
OMe O
Cl
OMe O
Cl
N
S
N H
O HN
O
NHMe Cl
N
N 10 Cl
Cl
11
N
N
194
Joanne M. Smallheer and Mimi L. Quan
Several series of Xa inhibitors have been reported that contain a pyrrolidine core structure and a 5-chlorothiophene P1 moiety similar to that of rivaroxaban. For example, compound 12 has a Xa Ki of 3.0 nM and PT EC2x of 1.7 mM [45]. Compound 13, where a longer linker to the P1 thiophene residue was employed, is comparable in affinity (Xa IC50 ¼ 5.5 nM) [46]. Factor Xa affinity was restored in compound 14 (Kio1 nM) by incorporation of a pyrrolidinone scaffold and the basic biaryl P4 moiety present in razaxaban [47]. This analog showed clearance of 13 mL/kg/min and good oral bioavailability (52%) in rats.
H N Cl
S
H N
N
F O
O O
O
N
S
N
OMe
O
O S N H
N
Cl
12
N
13
Cl S H S N O O
N N
O
O
CH2OH S
O
NMe2
15
N
16
O N
N Me
14
Cl
N O
Cl
N
O
O O S N
F
O O S N
N
F
N
O O S
N
N
N
N N
Cl
O
17
Xa inhibitors containing piperidine and piperazine cores have also been reported. Compound 15 has a chloronaphthyl P1 group appended through a sulfonamide linkage to a 3-aminopiperidinone core to provide a potent inhibitor of Xa (Ki ¼ 0.02 nM) with PT EC2x of 1.7 mM [48]. Substitution on the nitrogen of the sulfonamide improved Xa affinity. The 6-chloronaphthylsulfonyl P1 residue of 15 in combination with a fused spirocyclic core afforded 16 that displayed a Xa IC50 of 1.2 nM [49]. Compound 16 was effective in a rat venous thrombosis model, but was
195
Recent Advances in Coagulation Serine Protease Inhibitors
less efficacious than warfarin. Another series of Xa inhibitors containing a chloronaphthylsulfonyl P1 moiety is exemplified by 17 that has a Xa IC50 of 4.8 nM with PT EC2x value of 1.0 mM and was reported to be orally bioavailable in rats and monkeys [50]. Several edoxaban analogs have been reported as Xa inhibitors. Replacement of the 5-chloropyridine P1 moiety of edoxaban with a 5-chloroindole gave 18 (Xa IC50 ¼ 2.3 nM) resulting in 10-fold loss of Xa affinity [51]. As illustrated by 19 (Xa IC50 ¼ 8.4 nM), it was possible to remove a chiral center and retain most of the Xa affinity by replacement of the cyclohexyl ring with a piperidine [52]. Cyclopentyl compound 20 includes a 4-chlorothiophene at the P1 position and a pyridone as the P4 moiety and has a Xa Ki of 0.43 nM and PT EC2x of 1.7 mM [53]. CONMe2 X O
Cl
N H
N NH H O
O
Cl S N
N
NH
S O
N H
O N 20
18, X = CH (S) 19, X = N
3. THROMBIN INHIBITORS 3.1 Compounds in clinical trials Direct inhibitors of thrombin have been the target of anticoagulant research for many years, based on the key role played by thrombin in the formation of fibrin clots [54,55]. Ximelagatran, 21, was the first orally bioavailable, active site-specific serine protease inhibitor to be approved for human use. A double prodrug, 21 is rapidly converted after oral absorption to melagatran, 22, a potent and selective direct thrombin inhibitor [56]. The data from clinical trials have been extensively reviewed [57,58]. Ximelagatran was approved in Europe in 2003 for the prevention of DVT, but failed to receive approval in the United States due to concerns relating to potential liver toxicity. This toxicity ultimately led to the withdrawal of the drug from all markets and discontinuation of ongoing clinical trials in February 2006 [59,60]. A structurally related compound, 23 (AZD0837), which is also a benzamidine prodrug, is currently in phase II clinical trials [55]. At a 150 mg b.i.d. dose in AF patients, the safety and tolerability of 23 was shown to be comparable to that of warfarin [61]. A phase III trial with an extended release
196
Joanne M. Smallheer and Mimi L. Quan
formulation for once daily dosing in patients with AF is projected to start later this year [62]. Dabigatran etexilate, 24, is an orally bioavailable double prodrug, which is rapidly metabolized by non-specific esterases in the blood to dabigatran, 25, a potent, reversible, and highly selective direct thrombin inhibitor [63]. Several recent reviews of the clinical findings with dabigatran etexilate are available [64–67]. The half-life in humans was 7–9 h for a single dose and 14–17 h after t.i.d. dosing [68]. Oral bioavailability of dabigatran after administration of a single dose of the prodrug was B6% [69]. The results of three phase III clinical trials for the prevention of VTE after knee or hip surgery have been published [70–72]. Dabigatran etexilate has been approved in Europe and Canada for the prevention of VTE in patients undergoing total hip or total knee replacement surgeries [63]. A phase II dose-finding study in patients with ACS is ongoing, as are phase III studies for prevention of stroke in patients with AF and in treatment of VTE.
O O
O R1 NH2
O
HN
OMe
NH2
O
OH
N
N H
N
N
N H
N
R2 O
OCHF2 23
Cl
21 R1 = OH, R2 = Et 22 R1 = R2 = H
O O R1 N H2N
H N
OR2 N
N N Me
N
24 R1 = CO2-n-hex, R2 = Et 25 R1 = R2 = H
3.2 Preclinical compounds Much of the recent research to identify orally active direct thrombin inhibitors has focused on the replacement of the highly basic amidinecontaining P1 groups with less basic aminoheterocycles, as has been successfully achieved with the Xa inhibitors described earlier. P1
197
Recent Advances in Coagulation Serine Protease Inhibitors
imidazole and 2-aminothiazole groups were introduced as amidine replacements into a 3-aminopyridone series of thrombin inhibitors [73]. Single-digit nanomolar thrombin affinity was maintained with 26 (Ki ¼ 8 nM) and 27 (Ki ¼ 10 nM), but these compounds were not orally bioavailable. The introduction of the aminomethylimidazole P1 group into a 3-aminopyrazinone core series resulted in 28 which, although less potent (Ki ¼ 53 nM), was the first compound with this P1 moiety to show improved PK parameters [73]. When 28 was administered to dogs at an oral dose of 4 mg/kg, a 2-h half-life and Cmax of 1.4 mM were observed. Further optimization of the methylimidazole P1 group by introduction of an acetamide substituent on the imidazole nitrogen provided 29, with a thrombin Ki of 1.2 nM and aPTT EC2x of 0.7 mM [74]. Compound 29 was efficacious in the rat ferric chloride efficacy assay at an iv dose of 10 mg/kg/min, and some exposure was achieved after a 5 mg/kg oral dose to both dogs (Cmax ¼ 1.46 mM; t½ ¼ 1.6 h) and rhesus monkeys (Cmax ¼ 0.36 mM; t½ ¼ 1.1 h). Thrombin inhibitors with a 2-(2-chloro-6-fluorophenyl)acetamide central core exemplified by 30 (Ki ¼ 0.7 nM) and 31 (Ki ¼ 47 nM) have been reported [75]. A crystal structure of 31 bound to thrombin shows the aromatic F substituent in proximity to the backbone NH of Gly 216, suggesting that this F forms a F–H bond that contributes to the affinity of this compound. Further optimization of compounds retaining the oxyguanidine P1 moiety of 31 provided 2-cyano-6-fluoroacetamide analog 32, which had a thrombin Ki of 1.2 nM and aPTT EC2x of 0.36 mM and good oral bioavailability in dogs, but not in rats (49% and 2%, respectively) [76]. Various substitutions on the P3 pyridine moiety were made without loss of thrombin affinity; however, methyl and chloro substitution resulted in increased plasma protein binding as compared to 32. Efficacy was observed with this compound in a dog A-V shunt model when administered at a dose of 3 mg/kg p.o. O O S N H
O N
R
N H
O
H N N
N H
N O
26 R =
N H
S N
O N H
N
O
N N
28
H2N
27 R =
N
HN
N
O
H N
N
N H
O
29
N H
198
Joanne M. Smallheer and Mimi L. Quan
O
Cl
N H H2N
Cl N H
F
N
N
NH H2N
F F
O N H
O
R
N H
30
F
N H F F
X
31 R = Cl, X = CH 32 R = CN, X = N
3.3 Dual thrombin/factor Xa inhibitors It has been proposed by several groups that a dual-acting Xa/thrombin inhibitor might provide a better overall profile than either mechanism alone by the prevention of thrombin formation through the inhibition of Xa and simultaneous direct inhibition of thrombin activity [77–79]. A dual thrombin/Xa inhibitor, tanogitran (BIBT 986, 33), which has Xa and thrombin Ki values of 26 nM and 2.7 nM, respectively, was evaluated in a phase II clinical trial by iv administration in a human model of endotoxin-induced coagulation [80]. Tanogitran prolonged plasma aPTT and reduced in vivo thrombin generation in a dose-dependent manner and was safe and well tolerated. OH O NH
O N
N H2N NH
O
N
S
S
H N
O N
O
O
Cl
HN
N O
33
34
Vinyl sulfonamide 34 is a dual inhibitor of thrombin and Xa that has equal affinity for both enzymes (Ki ¼ 2 nM), whereas the analog lacking the vinyl methyl group is a selective Xa inhibitor. Compound 34 displayed aPTT and PT EC1.5x of 32 mM and 0.54 mM, respectively, and oral bioavailability in both rats and dogs [81]. Modification of a series of selective Xa inhibitors by the introduction of a proline residue in place of a glycine moiety at the P2 position provided compounds such as 35 which were potent dual inhibitors of Xa (Ki ¼ 1.5 nM) and thrombin (Ki ¼ 3.5 nM) [82]. The ratio of Xa to thrombin affinity could be reversed by the modification of the P3 pyridyl to a 4-hydroxyphenyl to provide a compound with Xa Ki of 8.3 nM and thrombin Ki of 3.2 nM. Compound 35 displayed good in vitro anticoagulant activity and moderate Caco-2 permeability (Papp ¼ 50 nm/s). Bioavailability in rats after an oral dose of 5 mg/kg was o5%.
Recent Advances in Coagulation Serine Protease Inhibitors
N
Cl
H2N
O N HN Cl
O
O
199
N N
N H
CO2H
N
N
HN
HN
N 35
36
In a series of oxazolopyridine-based thrombin inhibitors, replacement of the P3 pyridine moiety with a piperidine provided dual thrombin/Xa inhibitors exemplified by 36 with good in vitro and in vivo anticoagulant efficacy [83]. Racemic compound 36 has thrombin and Xa Ki values of 0.04 and 3.9 nM, respectively, and aPTT EC2x of 0.07 mM. In an iv PK study in dogs, 36 had a 4.2 h half-life, high clearance, and moderate volume of distribution. The compound completely blocked occlusion in a rat FeCl3 arterial thrombosis model but was not orally bioavailable.
4. FACTOR VIIa/TF INHIBITORS The high-affinity complex formed between the serine protease factor VIIa and the cell-associated glycoprotein tissue factor (VIIa/TF) has long been considered an ideal point of intervention by virtue of its unique role in the initiation of the coagulation cascade in response to an injury at the vessel wall [84]. Despite much research to identify orally bioavailable inhibitors of VIIa/TF, no orally bioavailable small-molecule inhibitors have progressed into clinical trials to date. A phase I clinical trial with PCI-27483, a parenteral VIIa inhibitor of undisclosed structure, was completed in late 2008. This compound will be further evaluated in a phase Ib/II trial in pancreatic cancer patients [85]. The fluorinated phenylglycine derivative 37 was reported to be an inhibitor of VIIa (Ki ¼ 81 nM), displayed good selectivity over thrombin, Xa, and trypsin, and had a PT EC2x of 2.0 mM [86]. In PK studies in rats, the compound had a half-life of B6 h with low clearance and volume of distribution. Oral administration of a 3 mg/kg dose of the ethyl ester, amidoxime double prodrug of 37 to rats resulted in 20% bioavailability. In a guinea pig model of arterial thrombosis, oral administration of the prodrug resulted in dose-dependent plasma concentrations of 37. Antithrombotic efficacy was achieved in this model with minimal effect on bleeding time, in contrast to a dual VIIa/Xa inhibitor, where bleeding time was significantly increased at higher doses.
200
Joanne M. Smallheer and Mimi L. Quan
Further optimization of previously reported amidinobenzimidazole VIIa inhibitors provided potent inhibitor 38 (VIIa Ki ¼ 4 nM), wherein the pendant fluorophenol moiety imparted good selectivity over Xa, and incorporation of the succinate side chain led to improvement in PK parameters in rats (clearance ¼ 0.3 mL/min/kg; mean residence time ¼ 9 h) [87,88]. Urea analog 39 was also a potent and highly selective VIIa inhibitor (Ki ¼ 2 nM) with much-improved ex vivo coagulation efficacy (PT EC2x ¼ 1.9 mM) [89]. Other acid, amide, and amine functionalities were also well tolerated in place of the urea at this position. With the additional interaction afforded by the urea substituent, it was found that the succinic acid side chain could be replaced by methyl, fluoro, or hydrogen without loss of VIIa affinity, anticoagulant efficacy, or selectivity versus other proteases [90]. HO2C
CO2H
HO
HO
OEt HO2C H2N
F
HN
NH
O HN
37
OH
N
NH
O
H2N
OH
R
NH C5H11
NH
O
HN
38 R = F 39 R = CH2NHCONH2
NH2
40
Additional modification of this scaffold led to indole 40 (VIIa Ki ¼ 13 nM), which has a hexanamide functionality in place of the urea of 39 [90]. Some loss of selectivity, particularly versus Xa, was observed on replacement of the urea with amide substituents, but a desirable improvement in PK parameters after iv dosing to rats was obtained with 40 (mean residence time ¼ 89 min). To assess the potential to achieve oral bioavailability, 40 was converted to its hydroxyamidine prodrug. After oral administration to rats at a dose of 10 mg/kg, the prodrug was orally absorbed (F ¼ 11%), but the parent compound 40 was not detected in vivo. A related analog 41 was reported wherein the 5-amidinobenzimidazole and indole P1 groups of 39 and 40 were successfully replaced with the less basic 5-amino-4-azaindole, maintaining good VIIa affinity (Ki ¼ 20 nM) [91]. The extended P3 urea substitution conferred B15-fold improvement in affinity compared to a nitro substituent at the same position of the biphenyl core. Introduction of 2,6-difluoro substitution on the urea phenyl substituent of 5-azaindole 42 (VIIa Ki ¼ 33 nM) provided a twofold improvement in affinity compared to its unsubstituted phenylurea counterpart [92]. This moiety was also effective in
201
Recent Advances in Coagulation Serine Protease Inhibitors
combination with a 4-amino-5-azaindole moiety as the benzamidine mimic, as exemplified by 43 (Ki ¼ 2.6 nM, PT EC2x ¼ 2.4 mM), which had improved VIIa affinity and ex vivo anticoagulant activity [93]. All three of these latter compounds have good selectivity for VIIa over other relevant serine proteases, but no PK data was reported. R2
CO2H
R1 O
N NH OH
H2N
N H
O F
N
NH HO
N H
N H
N H
F 42 = =H 43 R1 = NH2; R2 = C(CH3)2CO2H R1
41
R2
5. FACTOR IXa AND XIa INHIBITORS Both factor IXa and factor XIa are primarily associated with the intrinsic or propagation phase of coagulation and offer an alternative strategy for therapeutic intervention [94,95]. TTP889 is an orally bioavailable, selective, small-molecule partial inhibitor of factor IXa whose structure has not been disclosed. Antithrombotic efficacy was demonstrated with TTP889 in rat and porcine AV shunt models with no effect on bleeding times [96]. In phase I clinical trials, TTP889 was safe and well tolerated with a half-life suitable for once daily dosing [97]. Phase II clinical trial results for extended prevention of VTE in patients undergoing hip surgery were disappointing. At a 300-mg dose administered daily over 3 weeks following 1 week of standard treatment with low-molecularweight heparin, TTP889 was ineffective in reducing biomarkers for thrombin generation or the incidence of VTE compared to placebo [98,99]. A second phase II trial in patients with vascular assist devices is planned.
NH H2N
N H HO
44
O
NH
N H2N
N N H
45
B
O
N NH HO
H2N
H N NH
O O
46
N
202
Joanne M. Smallheer and Mimi L. Quan
HN
NH2 HN
NH
N H
NH
R
O
H N
N H
O
Br
NH2
O
S N H
N O
OH H N
Cl
N H
O 47 R = 4-hydroxyphenyl 48 R = 3-pyridyl
S
O
N O
49
Biarylbenzimidazole 44 has a IXa Ki of 99 nM and was identified through screening of small-molecule libraries [100]. This compound showed good selectivity with respect to thrombin, but lacked selectivity over Xa and VIIa and had poor efficacy in ex vivo clotting assays. Replacement of the central phenol with a 4-hydroxypyrazole moiety provided 45 (IXa Ki ¼ 50 nM) with B2-fold improved IXa affinity and also imparted improved selectivity over thrombin, VIIa, and Xa. Compound 45 also had improved potency in the aPTT assay, with an EC2x of 2.64 mM compared to W10 mM for 44. Several small-molecule inhibitors of factor XIa have also been reported. Racemic boronate 46 has an XIa IC50 of 1.4 mM with 30- and 8-fold selectivity over Xa and thrombin, respectively [101]. A series of potent keto-arginine-based peptidomimetics, exemplified by 47, are irreversible inhibitors of XIa and form a covalent bond to the catalytic serine of the enzyme [102]. Compound 47 has a XIa IC50 of 6 nM and aPTT EC2x of 2.4 mM and shows good selectivity over VIIa, Xa, and thrombin. In an iv PK study in rats, 47 had high clearance (32 mL/kg/ min), a short half-life (45 min), and a small volume of distribution (B236 mL/kg). In an iv rat vena cava model of venous thrombosis, 47 showed efficacy comparable to that of heparin. Pyridyl analog 48 (XIa IC50 ¼ 12 nM) was evaluated in a rat mesenteric bleeding model, and at fourfold the efficacious dose (1 mg/kg, continuous iv infusion) did not alter bleeding time. Modifications to reduce the peptidic character and molecular weight of these inhibitors led to 49 with XIa IC50 of 0.116 mM [103]. Br NH H2N
O
CO2H N
N H
N O 50
N
NH2 N H
HN
O
O NH
N H
OH Br HO2C
O 51
Recent Advances in Coagulation Serine Protease Inhibitors
203
The in vitro and in vivo antithrombotic profile of BMS-262084, 50, a potent irreversible inhibitor of XIa with IC50 of 2.8 nM and aPTT EC2x of 2.2 mM was reported [104]. Efficacy in FeCl2 models of arterial and venous thrombosis was observed in rats dosed iv with 50 (12 mg/ kg+12 mg/kg/h), whereas cuticle, mesenteric, or renal bleeding times were unchanged compared to vehicle controls. Clavatadine A, 51, a natural product isolated from a marine sponge was also reported to be an irreversible inhibitor of XIa with an IC50 of 1.3 mM [105].
6. CONCLUSIONS Potent and selective inhibitors of factor Xa and thrombin are well established. Several of these are undergoing advanced clinical trials, and rivaroxaban and dabigatran recently gained regulatory approval. In addition, prototype, selective inhibitors of factors VIIa, IXa, and XIa have been identified. From the perspective of the medicinal chemist, the biggest challenge inherent to these serine proteases as drug targets lies in the design of compounds that have good oral bioavailability and are suitable for b.i.d. or q.d. dosing. In the final analysis, it is likely that differentiation between these serine protease targets may ultimately require human clinical trials in multiple thromboembolic disorders.
REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
S. Haas, J. Thromb. Thrombolysis, 2008, 25, 52. B. I. Eriksson, D. J. Quinlan and J. I. Weitz, Clin. Pharmacokinet., 2009, 48, 1. M. Hammwo¨hner and A. Goette, J. Cardiovasc. Pharmacol., 2008, 52, 18. A. C. Spyropoulos, Thromb. Res., 2008, 123, S29. J. Harenberg, Therapy, 2008, 5, 177. J. P. Piccini, M. R. Patel, K. W. Mahaffey, K. A. A. Fox and R. M. Califf, Expert Opin. Investig. Drugs, 2008, 17, 925. J. Carreiro and J. Ansell, Expert Opin. Investig. Drugs, 2008, 17, 1937. E. Perzborn, J. Strassburger, A. Wilmen, J. Pohlmann, S. Roehrig, K. H. Schlemmer and A. Straub, J. Thromb. Haemost., 2005, 3, 514. C. Weinz, U. Buetehorn, H. P. Daehler, C. Kohlsdorfer, U. Pleiss, S. Sandmann, K. H. Schlemmer, T. Schwarz and W. Steinke, Xenobiotica, 2005, 35, 891. B. J. Biemond, E. Perzborn, P. W. Friederich, M. Levi, U. Buetehorn and H. R. Bu¨ller, Thromb. Haemost., 2007, 97, 471. D. Kubitza, M. Becka, B. Voith, M. Zuehlsdorf and G. Wensing, Clin. Pharmacol. Ther., 2005, 78, 412. H. R. Buller, A. W. Lensing, M. H. Prins, G. Agnelli, A. Cohen, A. S. Gallus, F. Misselwitz, G. Raskob, S. Schellong and A. Segers, Blood, 2008, 112, 2242. M. R. Lassen, W. Ageno, L. C. Borris, J. R. Lieberman, N. Rosencher, T. J. Bandel, F. Misselwitz and A. G. G. Turpie, N. Engl. J. Med., 2008, 358, 2776.
204
Joanne M. Smallheer and Mimi L. Quan
[14] A. G. G. Turpie, K. A. Bauer, B. Davidson, M. Gent, L. Kwong, M. R. Lassen, F. Cushner, P. Lotke, S. D. Berkowitz, T. J. Bandel, F. Misselwitz and W. Fisher, Blood (ASH Annual Meeting Abstracts), 2008, 112, Abstract 35. [15] B. I. Eriksson, L. C. Borris, R. J. Friedman, S. Haas, M. V. Huisman, A. K. Kakkar, T. J. Bandel, H. Beckmann, E. Muehlhofer, F. Misselwitz, W. Geerts and M. Levine, N. Engl. J. Med., 2008, 358, 2765. [16] A. K. Kakkar, B. Brenner, O. E. Dahl, B. I. Eriksson, P. Mouret, J. Muntz, A. G. Soglian, ´ . F. Pap, F. Misselwitz and S. Haas, Lancet, 2008, 372, 31. A [17] M. L. Quan, P. Y. S. Lam, Q. Han, D. J. P. Pinto, M. Y. He, R. Li, C. D. Ellis, C. G. Clark, C. A. Teleha, J. H. Sun, R. S. Alexander, S. Bai, J. M. Luettgen, R. M. Knabb, P. C. Wong and R. R. Wexler, J. Med. Chem., 2005, 48, 1729. [18] D. J. P. Pinto, M. J. Orwat, S. Koch, K. A. Rossi, R. S. Alexander, A. Smallwood, P. C. Wong, A. R. Rendina, J. M. Luettgen, R. M. Knabb, K. He, B. Xin, R. R. Wexler and P. Y. S. Lam, J. Med. Chem., 2007, 50, 5339. [19] P. C. Wong, C. A. Watson and E. J. Crain, J. Thromb. Haemost., 2008, 6, 1736. [20] P. C. Wong, E. J. Crain, B. Xin, R. R. Wexler, P. Y. S. Lam, D. J. Pinto, J. M. Luettgen and R. M. Knabb, J. Thromb. Haemost., 2008, 6, 820. [21] C. Frost, Z. Yu, S. Nepal, R. Mosqueda-Garcia and A. Shenker, J. Thromb. Haemost., 2007, 5(Suppl. 2), P-M-665. [22] C. Frost, Z. Yu, S. Nepal, A. Bragat, K. Moore, A. Shenker, Y. Barrett and F. Lacreta, J. Clin. Pharmacol., 2008, 48, 1099, Abstract 142. [23] H. Buller, D. Deitchman, M. Prins and A. Segers, J. Thromb. Haemost., 2008, 6, 1313. [24] M. R. Lassen, B. L. Davidson, A. Gallus, G. Pineo, J. Ansell and D. Deitchman, J. Thromb. Haemost., 2007, 5, 2368. [25] M. R. Lassen, A. S. Gallus, G. F. Pineo and G. E. Raskob, Blood (ASH Annual Meeting Abstracts), 2008, 112, Abstract 31. [26] J. Alexander, Annual Congress European Society of Cardiology, Munich, Germany, August 30 to September 3, 2008, Abstract 3208. [27] T. Furugohri, K. Isobe, Y. Honda, C. Kamisato-Matsumoto, N. Sugiyama, T. Nagahara, Y. Morishima and T. Shibano, J. Thromb. Haemost., 2008, 6, 1542. [28] G. Raskob, Annual Congress European Cardiology Society, Munich, Germany, August 30 to September 3, 2008, Abstract 3712. [29] T. Fuji, S. Fujita, S. Tachibana and Y. Kawai, Blood (ASH Annual Meeting Abstracts), 2008, 112, Abstract 34. [30] J. I. Weitz, S. J. Connolly, S. Kunitada, J. Jin and I. Patel, Blood (ASH Annual Meeting Abstracts), 2008, 112, Abstract 33. [31] J. T. Kohrt, C. F. Bigge, J. W. Bryant, A. Casimiro-Garcia, L. Chi, W. L. Cody, T. Dahring, D. A. Dudley, K. J. Filipski, S. Haarer, R. Heemstra, N. Janiczek, L. Narasimhan, T. McClanahan, J. T. Peterson, V. Sahasrabudhe, R. Schaum, C. A. Van Huis, K. M. Welch, E. Zhang, R. J. Leadley and J. J. Edmunds, Chem. Biol. Drug Des., 2007, 70, 100. [32] A. T. Cohen, D. Armstrong, T. Gazdzik, C. Ryge, R. Pak, J. Mandema, R. Boyd, S. McBride and L. A. DiCarlo, Blood (ASH Annual Meeting Abstracts), 2008, 112, Abstract 980. [33] P. Zhang, H. Wenrong, L. Wang, L. Bao, Z. J. Jia, S. M. Bauer, E. A. Goldman, G. D. Probst, Y. Song, T. Su, J. Fan, Y. Wu, W. Li, J. Woolfrey, U. Sinha, P. W. Wong, S. T. Edwards, A. E. Arfsten, L. A. Clizbe, J. Kanter, A. Pandey, G. Park, A. Hutchaleelaha, J. L. Lambing, S. J. Hollenbach, R. M. Scarborough and B. Y. Zhu, Bioorg. Med. Chem. Lett., 2009, 19, 2179. [34] K. Abe, G. Siu, S. Edwards, P. H. Lin, B. Y. Zhu, U. Marzec, S. Hanson, Y. Pak, S. Hollenbach and U. Sinha, Blood (ASH Annual Meeting Abstracts), 2006, 108, Abstract 901.
Recent Advances in Coagulation Serine Protease Inhibitors
205
[35] A. G. G. Turpie, K. A. Bauer, B. L. Davidson, W. D. Fisher, M. Gent, M. H. Huo, U. Sinha and D. D. Gretler, Thromb. Haemost., 2009, 101, 68. [36] G. Agnelli, S. Haas, J. S. Ginsberg, K. A. Krueger, A. Dmitrienko and J. T. Brandt, J. Thromb. Haemost., 2007, 5, 746. [37] F. Kuo, D. K. Clodfelter, T. R. Priest and D. L. K. Kau, J. Label. Compd. Radiopharm., 2004, 47, 599. [38] Y. Iwatsuki, T. Shigenaga, Y. Moritani, M. Suzuki, T. Ishihara, F. Hirayama and T. Kawasaki, Blood (ASH Annual Meeting Abstracts), 2006, 108, Abstract 911. [39] B. I. Eriksson, A. G. G. Turpie, M. R. Lassen, M. H. Prins, G. Agnelli, P. Ka¨lebo, M. L. Gaillard and L. Meems, J. Thromb. Haemost., 2007, 5, 1660. [40] J. G. Varnes, D. A. Wacker, D. J. P. Pinto, M. J. Orwat, J. P. Theroff, B. Wells, R. A. Galemo, J. M. Luettgen, R. M. Knabb, S. Bai, K. He, P. Y. S. Lam and R. R. Wexler, Bioorg. Med. Chem. Lett., 2008, 18, 749. [41] J. X. Qiao, D. L. Cheney, R. S. Alexander, A. M. Smallwood, S. R. King, K. He, A. R. Rendina, J. M. Luettgen, R. M. Knabb, R. R. Wexler and P. Y. S. Lam, Bioorg. Med. Chem. Lett., 2008, 18, 4118. [42] Y.-K. Lee, D. J. Parks, T. Lu, T. V. Thieu, T. Markotan, W. Pan, D. F. McComsey, K. L. Milkiewicz, C. S. Crysler, N. Ninan, M. C. Abad, E. C. Giardino, B. E. Maryanoff, B. P. Damiano and M. R. Player, J. Med. Chem., 2008, 51, 282. [43] J. R. Corte, T. Fang, D. J. P. Pinto, W. Han, Z. Hu, X.-J. Jiang, Y.-L. Li, J. F. Gauuan, M. Hadden, D. Orton, A. R. Rendina, J. M. Luettgen, P. C. Wong, K. He, P. E. Morin, C.-H. Chang, D. L. Cheney, R. M. Knabb, R. R. Wexler and P. Y. S. Lam, Bioorg. Med. Chem. Lett., 2008, 18, 2845. [44] B. Ye, D. O. Arnaiz, Y.-L. Chou, B. D. Griedel, R. Karanjawala, W. Lee, M. M. Morrissey, K. L. Sacchi, S. T. Sakata, K. J. Shaw, S. C. Wu, Z. Zhao, M. Adler, S. Cheeseman, W. P. Dole, J. Ewing, R. Fitch, D. Lentz, A. Liang, D. Light, J. Morser, J. Post, G. Rumennik, B. Subramanyam, M. E. Sullivan, R. Vergona, J. Walters, Y.-X. Wang, K. A. White, M. Whitlow and M. Kochanny, J. Med. Chem., 2007, 50, 2967. [45] K. G. Zbinden, L. Anselm, D. W. Banner, J. Benz, F. Blasco, G. De´coret, J. Himber, B. Kuhn, N. Panday, F. Ricklin, P. Risch, D. Schlatter, M. Stahl, S. Thomi, R. Unger and W. Haap, Eur. J. Med. Chem., 2009, 44, 2787. [46] Y. Shi, D. Sitkoff, J. Zhang, W. Han, Z. Hu, P. D. Stein, Y. Wang, L. J. Kennedy, S. P. O’Connor, S. Ahmad, E. C. K. Liu, S. M. Seiler, P. Y. S. Lam, J. A. Robl, J. E. Macor, K. S. Atwal and R. Zahler, Bioorg. Med. Chem. Lett., 2007, 17, 5952. [47] R. J. Young, A. D. Borthwick, D. Brown, C. L. Burns-Kurtis, M. Campbell, C. Chan, M. Charbaut, M. A. Convery, H. Diallo, E. Hortense, W. R. Irving, H. A. Kelly, N. P. King, S. Kleanthous, A. M. Mason, A. J. Pateman, A. N. Patikis, I. L. Pinto, D. R. Pollard, S. Senger, G. P. Shah, J. R. Toomey, N. S. Watson, H. E. Weston and P. Zhou, Bioorg. Med. Chem. Lett., 2008, 18, 28. [48] J. M. Smallheer, S. Wang, M. L. Laws, S. Nakajima, Z. Hu, W. Han, I. Jacobson, J. M. Luettgen, K. A. Rossi, A. R. Rendina, R. M. Knabb, R. R. Wexler, P. Y. S. Lam and M. L. Quan, Bioorg. Med. Chem. Lett., 2008, 18, 2428. [49] F. Saitoh, H. Nishida, T. Mukaihira, N. Kosuga, M. Ohkouchi, T. Matsusue, I. Shiromizu, Y. Hosaka, M. Matsumoto and I. Yamamoto, Chem. Pharm. Bull., 2007, 55, 317. [50] Y. Imaeda, T. Kuroita, H. Sakamoto, T. Kawamoto, M. Tobisu, N. Konishi, K. Hiroe, M. Kawamura, T. Tanaka and K. Kubo, J. Med. Chem., 2008, 51, 3422. [51] T. Nagata, T. Yoshino, N. Haginoya, K. Yoshikawa, M. Nagamochi, S. Kobayashi, S. Komoriya, A. Yokomizo, R. Muto, M. Yamaguchi, K. Osanai, M. Suzuki and H. Kanno, Bioorg. Med. Chem., 2009, 17, 1193. [52] A. Mochizuki, Y. Nakamoto, H. Naito, K. Uoto and T. Ohta, Bioorg. Med. Chem. Lett., 2008, 18, 782.
206
Joanne M. Smallheer and Mimi L. Quan
[53] J. X. Qiao, C.-H. Chang, D. L. Cheney, P. E. Morin, G. Z. Wang, S. R. King, T. C. Wang, A. R. Rendina, J. M. Luettgen, R. M. Knabb, R. R. Wexler and P. Y. S. Lam, Bioorg. Med. Chem. Lett., 2007, 17, 4419. [54] H. Bounameaux, Seminars Thromb. Hemostasis., 2008, 34(Suppl. 1), 12. [55] A. Schwienhorst, Cell. Mol. Life Sci., 2006, 63, 2773. [56] D. Gustafsson and M. Elg, Thromb. Res., 2003, 109, S9. [57] A. Choudhury, D. Goyal and G. Y. H. Lip, Drugs Today, 2006, 42, 3. [58] S.-J. Ho and T. A. Brighton, Vasc. Health Risk Manag., 2006, 2, 49. [59] AstraZeneca PLC, Press Release, February 14, 2006, http://www.astrazeneca.com/ media/latest-press-releases/2006/5217?itemId ¼ 3891692 (last accessed 25/02/09). [60] G. Agnelli, B. I. Eriksson, A. T. Cohen, D. Bergqvist, O. E. Dahl, M. R. Lassen, P. Mouret, N. Rosencher, M. Andersson, A. Bylock, E. Jensen and B. Boberg, Thromb. Res., 2009, 123, 488. [61] B. Olsson, L. H. Rasmussen, A. Tveit, E. Jensen, P. Wessman, S. Panfilov, H. D. Ekdal and K. Wa˚hlander, J. Thromb. Haemost., 2007, 5(Suppl. 2), O-W-053. [62] AstraZeneca PLC, Press Release, January 29, 2009, http://www.astrazeneca.com/ _mshost3690701/content/resources/media/investors/AZN-Q4-2008/q4-results2008-narrative.pdf (last accessed 25/02/09). [63] B. I. Eriksson, H. Smith, U. Yasothan and P. Kirkpatrick, Nat. Rev. Drug Discov., 2008, 7, 557. [64] B. E. Baetz and S. A. Spinler, Pharmacotherapy, 2008, 28, 1354. [65] M. Ieko, Curr. Opin. Investig. Drugs, 2007, 8, 758. [66] H. Nishio, M. Ieko and T. Nakabayashi, Expert Opin. Pharmacother., 2008, 9, 2509. [67] M. Sanford and G. L. Plosker, Drugs, 2008, 68, 1699. [68] J. Stangier, K. Rathgen, H. Sta¨hle, D. Gansser and W. Roth, Br. J. Clin. Pharmacol., 2007, 64, 292. [69] S. Blech, T. Ebner, E. Ludwig-Schwellinger, J. Stangier and W. Roth, Drug Metab. Dispos., 2008, 36, 386. [70] B. I. Eriksson, O. E. Dahl, N. Rosencher, A. A. Kurth, C. N. van Dijk, S. P. Frostick, P. Ka¨lebo, A. V. Christiansen, S. Hantel, R. Hettiarachchi, J. Schnee and H. R. Bu¨ller, J. Thromb. Haemost., 2007, 5, 2178. [71] B. I. Eriksson, O. E. Dahl, N. Rosencher, A. A. Kurth, C. N. van Dijk, S. P. Frostick, M. H. Prins, R. Hettiarachchi, S. Hantel, J. Schnee and H. R. Bu¨ller, Lancet, 2007, 370, 949. [72] J. S. Ginsberg, B. L. Davidson, P. C. Comp, C. W. Francis, R. J. Friedman, M. H. Huo, J. R. Lieberman, J. E. Muntz, G. E. Raskob, M. L. Clements, S. Hantel, J. M. Schnee and J. A. Caprini, J. Arthroplasty, 2009, 24, 1. [73] R. C. A. Isaacs, M. G. Solinsky, K. J. Cutrona, C. L. Newton, A. M. Naylor-Olsen, J. A. Krueger, S. D. Lewis and B. J. Lucas, Bioorg. Med. Chem. Lett., 2006, 16, 338. [74] R. C. A. Isaacs, M. G. Solinsky, K. J. Cutrona, C. L. Newton, A. M. Naylor-Olsen, D. R. McMasters, J. A. Krueger, S. D. Lewis, B. J. Lucas, L. C. Kuo, Y. Yan, J. J. Lynch and E. A. Lyle, Bioorg. Med. Chem. Lett., 2008, 18, 2062. [75] L. Lee, K. D. Kreutter, W. Pan, C. Crysler, J. Spurlino, M. R. Player, B. Tomczuk and T. Lu, Bioorg. Med. Chem. Lett., 2007, 17, 6266. [76] K. D. Kreutter, T. Lu, L. Lee, E. C. Giardino, S. Patel, H. Huang, G. Xu, M. Fitzgerald, B. J. Haertlein, V. Mohan, C. Crysler, S. Eisennagel, M. Dasgupta, M. McMillan, J. C. Spurlino, N. D. Huebert, B. E. Maryanoff, B. E. Tomczuk, B. P. Damiano and M. R. Player, Bioorg. Med. Chem. Lett., 2008, 18, 2865. [77] E. U. Graefe-Mody, U. Schu¨hly, K. Rathgen, H. Sta¨hle, J. M. Leitner and B. Jilma, J. Thromb. Haemost., 2006, 4, 1502. [78] A. Kranjc and D. Kikelj, Curr. Med. Chem., 2004, 11, 2535.
Recent Advances in Coagulation Serine Protease Inhibitors
207
[79] H. Nar, M. Bauer, A. Schmid, J.-M. Stassen, W. Wienen, H. W. M. Priepke, I. K. Kauffmann, U. J. Ries and N. H. Hauel, Structure, 2001, 9, 29. [80] J. M. Leitner, B. Jilma, F. B. Mayr, F. Cardona, A. O. Spiel, C. Firbas, K. Rathgen, H. Sta¨hle, U. Schu¨hly and E. U. Graefe-Mody, Clin. Pharm.Ther., 2007, 81, 858. [81] R. J. Young, D. Brown, C. L. Burns-Kurtis, C. Chan, M. A. Convery, J. A. Hubbard, H. A. Kelly, A. J. Pateman, A. Patikis, S. Senger, G. P. Shah, J. R. Toomey, N. S. Watson and P. Zhou, Bioorg. Med. Chem. Lett., 2007, 17, 2927. [82] D. Do¨nnecke, A. Schweinitz, A. Stu¨rzebecher, P. Steinmetzer, M. Schuster, U. Stu¨rzebecher, S. Nicklisch, J. Stu¨rzebecher and T. Steinmetzer, Bioorg. Med. Chem. Lett., 2007, 17, 3322. [83] J. Z. Deng, D. R. McMasters, P. M. A. Rabbat, P. D. Williams, C. A. Coburn, Y. Yan, L. C. Kuo, S. D. Lewis, B. J. Lucas, J. A. Krueger, B. Strulovici, J. P. Vacca, T. A. Lyle and C. S. Burgey, Bioorg. Med. Chem. Lett., 2005, 15, 4411. [84] R. A. Shirk and G. P. Vlasuk, Arterioscler. Thromb. Vasc. Biol., 2007, 27, 1895. [85] Pharmacyclics, Company Website, 2009, http://www.pharmacyclics.com/wt/page/ pci_27483 (last accessed 05/03/09). [86] K. G. Zbinden, D. W. Banner, K. Hilpert, J. Himber, T. Lave´, M. A. Riederer, M. Stahl, T. B. Tschopp and U. Obst-Sander, Bioorg. Med. Chem., 2006, 14, 5357. [87] A. Kolesnikov, R. Rai, W. B. Young, J. Mordenti, L. Liu, S. Torkelson, W. D. Shrader, E. M. Leahy, H. Hu, E. Gjerstad, J. Janc, B. A. Katz and P. A. Sprengeler, Bioorg. Med. Chem. Lett., 2006, 16, 2243. [88] W. D. Shrader, A. Kolesnikov, J. Burgess-Henry, R. Rai, J. Hendrix, H. Hu, S. Torkelson, T. Ton, W. B. Young, B. A. Katz, C. Yu, J. Tang, R. Cabuslay, E. Sanford, J. W. Janc and P. A. Sprengeler, Bioorg. Med. Chem. Lett., 2006, 16, 1596. [89] W. B. Young, J. Mordenti, S. Torkelson, W. D. Shrader, A. Kolesnikov, R. Rai, L. Liu, H. Hu, E. M. Leahy, M. J. Green, P. A. Sprengeler, B. A. Katz, C. Yu, J. W. Janc, K. C. Elrod, U. M. Marzec and S. R. Hanson, Bioorg. Med. Chem. Lett., 2006, 16, 2037. [90] D. Vijaykumar, R. Rai, M. Shaghafi, T. Ton, S. Torkelson, E. M. Leahy, J. R. Riggs, H. Hu, P. A. Sprengeler, W. D. Shrader, C. O’Bryan, R. Cabuslay, E. Sanford, E. Gjerstadt, L. Liu, J. Sukbuntherng and W. B. Young, Bioorg. Med. Chem. Lett., 2006, 16, 3829. [91] R. Rai, A. Kolesnikov, P. A. Sprengeler, S. Torkelson, T. Ton, B. A. Katz, C. Yu, J. Hendrix, W. D. Shrader, R. Stephens, R. Cabuslay, E. Sanford and W. B. Young, Bioorg. Med. Chem. Lett., 2006, 16, 2270. [92] J. R. Riggs, H. Hu, A. Kolesnikov, E. M. Leahy, K. E. Wesson, W. D. Shrader, D. Vijaykumar, T. A. Wahl, Z. Tong, P. A. Sprengeler, M. J. Green, C. Yu, B. A. Katz, E. Sanford, M. Nguyen, R. Cabuslay and W. B. Young, Bioorg. Med. Chem. Lett., 2006, 16, 3197. [93] H. Hu, A. Kolesnikov, J. R. Riggs, K. E. Wesson, R. Stephens, E. M. Leahy, W. D. Shrader, P. A. Sprengeler, M. J. Green, E. Sanford, M. Nguyen, E. Gjerstad, R. Cabuslay and W. B. Young, Bioorg. Med. Chem. Lett., 2006, 16, 4567. [94] E. L. Howard, K. C. D. Becker, C. P. Rusconi and R. C. Becker, Arterioscler. Thromb. Vasc. Biol., 2007, 27, 722. [95] D. Gailani and T. Renne´, J. Thromb. Haemost., 2007, 5, 1106. [96] R. Rothlein, J. M. Shen, N. Naser, D. R. Gohimukkula, T. B. Caligan, R. C. Andrews, A. M. Schmidt, E. A. Rose and A. M. M. Mjalli, Blood (ASH Annual Meeting Abstracts), 2005, 106, Abstract 1886. [97] TransTech Pharma, Press Release, 2004, http://www.ttpharma.com/press_releases/ first_phase_i.html (last accessed 28/02/09). [98] B. I. Eriksson, O. E. Dahl, M. R. Lassen, D. P. Ward, R. Rothlein, G. Davis and A. G. G. Turpie, J. Thromb. Haemost., 2008, 6, 457.
208
Joanne M. Smallheer and Mimi L. Quan
[99] O. E. Dahl, B. I. Eriksson, M. R. Lassen, D. P. Ward, R. Rothlein, G. Davis and A. G. G. Turpie, J. Thromb. Haemost., 2007, 5(Suppl. 2), P-T-065. [100] D. Vijaykumar, P. A. Sprengeler, M. Shaghafi, J. R. Spencer, B. A. Katz, C. Yu, R. Rai, W. B. Young, B. Schultz and J. Janc, Bioorg. Med. Chem. Lett., 2006, 16, 2796. [101] T. I. Lazarova, L. Jin, M. Rynkiewicz, J. C. Gorga, F. Bibbins, H. V. Meyers, R. Babine and J. Strickler, Bioorg. Med. Chem. Lett., 2006, 16, 5022. [102] J. Lin, H. Deng, L. Jin, P. Pandey, J. Quinn, S. Cantin, M. J. Rynkiewicz, J. C. Gorga, F. Bibbins, C. A. Celatka, P. Nagafuji, T. D. Bannister, H. V. Meyers, R. E. Babine, N. J. Hayward, D. Weaver, H. Benjamin, F. Stassen, S. S. Abdel-Meguid and J. E. Strickler, J. Med. Chem., 2006, 49, 7781. [103] H. Deng, T. D. Bannister, L. Jin, R. E. Babine, J. Quinn, P. Nagafuji, C. A. Celatka, J. Lin, T. I. Lazarova, M. J. Rynkiewicz, F. Bibbins, P. Pandey, J. Gorga, H. V. Meyers, S. S. Abdel-Meguid and J. E. Strickler, Bioorg. Med. Chem. Lett., 2006, 16, 3049. [104] W. A. Schumacher, S. E. Seiler, T. E. Steinbacher, A. B. Stewart, J. S. Bostwick, K. S. Hartl, E. C. Liu and M. L. Ogletree, Eur. J. Pharmacol., 2007, 570, 167. [105] M. S. Buchanan, A. R. Carroll, D. Wessling, M. Jobling, V. M. Avery, R. A. Davis, ¨ ster, T. Fex, J. Deinum, J. N. A. Hooper and R. J. Quinn, J. Med. Y. Feng, Y. Xue, L. O Chem., 2008, 51, 3583.
CHAPT ER
10 Advances in the Discovery of Anti-Inflammatory FMS Inhibitors Carl L. Manthey and Mark R. Player
Contents
1. Introduction 1.1 Pivotal driver of the macrophage lineage 1.2 Role in tumor biology 1.3 Role in skeletal pathophysiology 1.4 Role in inflammation 2. Recent FMS Inhibitors 2.1 Aryl- and heteroarylamides 2.2 Quinazolines, quinolines, and quinolones 2.3 Other chemotypes 2.4 Multitargeted kinase inhibitors 3. Conclusion References
211 211 212 213 213 215 215 217 218 220 221 222
1. INTRODUCTION 1.1 Pivotal driver of the macrophage lineage The tyrosine kinase receptor, FMS, is emerging as an attractive drug target to control macrophage numbers in multiple disease settings. The biology of FMS and its ligand, colony stimulating factor-1 (CSF-1), have been the subject of recent reviews [1–3]. Robust expression of FMS is restricted primarily to the cells of the macrophage lineage including Johnson & Johnson Pharmaceutical Research and Development, L.L.C., Spring House, PA 19477, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04410-8
r 2009 Elsevier Inc. All rights reserved.
211
212
Carl L. Manthey and Mark R. Player
monocytes, tissue macrophages, dendritic cells, and osteoclasts. FMS dimerization and autophosphorylation of the cytoplasmic domain occurs when CSF-1 binds to the extracellular domains of FMS. Small molecule inhibitors of the FMS kinase domain block autophosphorylation and the signals that would otherwise ensure macrophage survival, prime expression of cytokines, activate proliferation of macrophage progenitors, and differentiate some macrophages into osteoclasts. Mice, nullizygous for either CSF-1 or FMS, exhibit severe deficits in osteoclasts and in several subpopulations of tissue macrophages including those found in the synovium, skin, kidney, and gastric mucosa [4,5]. Dendritic cells are also reduced. For the aforementioned reasons, FMS inhibitors are being optimized and may be useful in treating diseases such as cancer, rheumatoid arthritis (RA), and other autoimmune or remodeling diseases that are characterized by pathogenic macrophages, dendritic cells, and osteoclasts.
1.2 Role in tumor biology Treatment of malignancy by targeting of tumor-associated macrophages is an exciting and emerging concept. Direct correlations have been made between macrophage numbers, angiogenesis and tumor progression [6]. In some tumors the microenvironment appears to activate macrophage expression of soluble factors that: (i) promote angiogenesis, (ii) support tumor cell survival, proliferation, drug resistance, and motility, and (iii) suppress anti-tumor immunity (reviewed in Ref. [7]). CSF-1 is overexpressed by many cancers, where it is a negative prognostic factor [8]. Preclinical studies have identified a role for CSF-1dependent macrophages in tumor angiogenesis. Tumor angiogenesis is blunted in mice that are nullizygous for CSF-1, whereas forced overexpression of CSF-1, by either the host or the tumor, caused remarkable increases in tumor infiltrating macrophages and tumor vessel densities, as well as enhanced oncogenicity [9,10]. Pharmacological intervention studies corroborate a role for CSF-1 in tumor growth. CSF-1 antisense oligonucleotides, CSF-1 antibody, FMS siRNA, and a selective FMS kinase inhibitor have reduced the growth rates of diverse xenografts [11–13]. In these studies, tumor-associated macrophages and macrophage-derived pro-angiogenic factors were depleted in association with marked reductions in tumor microvasculature. Anti-CSF-1 reversed the chemoresistance of MCF-7 xenografts, which identified an unanticipated role for CSF-1 in chemoresistance. Additionally, the siRNA, Mi160 was determined to inhibit CSF-1 synthesis and was, itself, down-regulated in several drug-resistant cancer lines [14]. Macrophages also mediate hormone resistance in prostate cancers by a nuclear receptor derepression pathway [15]. CSF-1-dependent macrophages also mediate tumor
Advances in the Discovery of Anti-Inflammatory FMS Inhibitors
213
cell egress. In fact, spontaneous metastasis of mammary carcinomas of the lung is profoundly reduced in CSF-1-nullizygous mice [16,17]. High expression of FMS has been documented in several types of cancers [8]. The contribution of FMS to oncogenesis of these tumors remains speculative, but FMS inhibitors may, in some instances, have direct anticancer activity. Nonetheless, by targeting macrophages, FMS inhibitors may have broad utility in blocking critical tumor–stroma interactions.
1.3 Role in skeletal pathophysiology The skeletal microenvironment activates a program of macrophage fusion and differentiation whereby osteoclasts are formed and endowed with the ability to resorb bone. Osteoclast differentiation requires two factors, CSF-1 and RANK ligand, that act interdependently [18]. Metastatic bone disease and RA are characterized by increased local expression of CSF-1. Osteoclasts provide the resorbing edge of the tumor/pannus tissue that erodes bone and leads to deformity and fracture in the case of cancer. Bone protection is a salient feature of FMS inhibitor pharmacology. Anti-FMS antibody blocked completely the otherwise fulminate osteoclastogenesis and bone erosion in a murine KRN-serum transfer arthritis model [19]. Four structurally unrelated oral FMS inhibitors have provided dramatic protection to bone in murine models of collagen- or adjuvant-induced arthritis [20–23]. FMS inhibitors, tested in models of metastatic bone disease, reduced radiographic lesions [24]. These data provide the preclinical rationale for evaluating FMS inhibitors in RA and metastatic bone disease. Pfizer has advanced an anti-CSF-1 antibody into Phase I safety testing in RA [25]. Completion of this trial, anticipated in 2009, might provide the first human bone biomarker data for CSF-1/FMS inhibition. Preclinical studies suggest that FMS inhibitors may provide oral alternatives to zoledronate, currently, the first-line therapy for the prevention of skeletal events in cancer patients, and to the RANK ligand antibody, denosumab, an experimental agent with efficacy as a bone protective agent in RA [26].
1.4 Role in inflammation Macrophages amplify and resolve inflammation and remodel tissue in auto-immune or auto-inflammatory diseases [2]. Of cell populations present in the rheumatoid synovium, macrophage numbers provide the strongest correlation with clinical symptoms, and successful therapy is highly associated with declines in this cell population [27]. Immunohistochemical studies identified macrophages as the dominant
214
Carl L. Manthey and Mark R. Player
source of tumor necrosis factor-alpha (TNFa) and interleukin-1b (IL-1b) in rheumatoid pannus [28]. In addition to activating fibroplasia and vascular inflammation, TNFa and IL-1b induce CSF-1 expression by many cell types [29]. CSF-1 levels are elevated markedly in rheumatoid plasma, synovial fluid, and synovium [30,31], where, presumably, it contributes to monocyte recruitment and differentiation, as well as macrophage survival, proliferation, and further expression of IL-1b and TNFa. Inhibition of CSF-1/FMS provides a tractable intervention point in this apparent positive feedback loop. Indeed, CSF-1-nullizygous mice were resistant to collagen-induced arthritis, while recombinant CSF-1 exacerbated clinical signs in this classic mouse model of RA. Additionally, neutralizing anti-CSF-1 and two unrelated oral FMS kinase inhibitors inhibited clinical signs of disease [21,23,32]. In one study, FMS inhibition reduced disease-associated splenic CD11b macrophages to normal levels and reduced type II collagen-induced expression of TNFa and IL-1b. A FMS inhibitor was also effective at reducing ankle swelling in a streptococcal cell wall–induced model of arthritis in rats [22]. These data provide preclinical rationale for evaluating FMS inhibitors in RA. However, anti-FMS antibody and oral FMS inhibitors exerted less influence on inflammation in mouse KRN-serum transfer arthritis and in rat adjuvant arthritis, despite robust bone protection [19,20,22]. One FMS inhibitor provided anti-inflammatory activity in adjuvant arthritis only when combined with methotrexate [22]. It is not presently known why these latter models are less responsive to the anti-inflammatory activity of FMS inhibitors. Many forms of immune-mediated nephritis in humans, including lupus nephritis, are associated with an expanded population of renal macrophages and increased tubular and glomerular expression of CSF-1. A negative correlation exists between renal CSF-1 and creatinine clearance [33]. Plasma CSF-1 is elevated in patients with systemic lupus [34], and persistently high levels of CSF-1 in urine after initial remission, is a predictor of renal flares [35]. MRL-Faslpr mice exhibit spontaneous lupus-like disease including nephritis and cutaneous lesions together with increased local and systemic levels of CSF-1. Transgenic overexpression of CSF-1 exacerbates disease, and CSF-1-nullizygous mice demonstrate resistance [36,37]. Because UVB light triggers cutaneous lesions in wild-type but not in CSF-1-deficient MRL-Faslpr mice [37], FMS inhibitors might find additional application treating the subset of patients with cutaneous disease. Elevated CSF-1 or FMS expression, together with increased macrophage numbers, has been associated with coronary artery disease [38], inflammatory bowel diseases [39], sarcoidosis [40], and obesity [41]. CSF-1 deficiency conferred resistance to diet-induced atherogenesis in mice [42]. Anti-CSF-1 or CSF-1 deficiency reduced dextran sodium
Advances in the Discovery of Anti-Inflammatory FMS Inhibitors
215
sulfate colitis in mice [43]. CSF-1 deficiency also conferred resistance to peripheral demyelinating disease [44], and a FMS kinase inhibitor reduced markedly the loss of motor function in a MOG peptide–induced model of multiple sclerosis [45]. However, dinitrobenzene sulfonic acid–induced colitis was exacerbated in CSF-1-deficient mice [46]. In some settings, CSF-1-dependent macrophages may be a source of antiinflammatory molecules and may mediate tissue repair [2]. Although much remains to be learned regarding the long-term consequences of FMS inhibition in man, FMS inhibition in rats is well tolerated [20], and FMS provides a tractable and exciting target to control the macrophage lineage across a spectrum of human disease.
2. RECENT FMS INHIBITORS 2.1 Aryl- and heteroarylamides From a simple arylamide high-throughput screening (HTS) hit (IC50 ¼ 400 nM), a series of 2-cyano-5-carboxamide inhibitors were identified with FMS kinase IC50s of about 5 and 100 nM in a cellular assay of CSF-1stimulated bone marrow–derived macrophage (BMDM) proliferation [47–49]. A co-crystal structure of an early analog (1) was obtained with FMS. Interaction occurs in the hinge region of the adenosine triphosphate (ATP) pocket and involves a hydrogen bond between the arylamide carbonyl and the backbone NH of Cys666. The ring oxygen of the furan is not involved in direct binding to a specific FMS residue; instead, it shares an intramolecular hydrogen bond with the amide NH, thereby stabilizing a flat conformation of the arylamide core. The methylpiperidine occupies the ATP sugar pocket and the hydroxymethyl extends away from this region, making a hydrogen bond with the phenol of Tyr665 [50]. In a 20-kinase selectivity panel, 1 inhibited only TrkA greater than 50% at 1 mM. Further optimization afforded 2,4-disubstituted arylamides that demonstrated IC50s in the 1 nM range in enzyme and cellular assays [23,51]. Compound 2 had a FMS kinase IC50 of 800 pM but inhibited several other receptor tyrosine kinases (c-Kit, Axl, TrkA, Flt-3, and IRKb) at IC50s less than 100 nM. This compound, at doses of 30 and 5 mg/kg, p.o. BID, reduced clinical scores by 65 and 35%, respectively, in a murine collagen–induced arthritis model. Histological examination revealed that pannus growth and bone destruction were reduced by 80%, a level of efficacy comparable to anti-TNF strategies in collagen-induced arthritis models. Additional optimization efforts replaced the phenylenediamine moiety with carbon-based linkers, maintaining potency and mitigating the potential for idiosyncratic drug reactions as evidenced by the loss of the compounds’ ability to conjugate glutathione in vitro [52].
216
Carl L. Manthey and Mark R. Player
For example, 3 demonstrated IC50s of 400 and 360 pM in kinase and BMDM cellular assays respectively.
N
N
H N
O
CN
O N
1
CN N H O
N HO
N
H N
2
N
H N
CN N H O
HN
3
A series of 2-(a-methylbenzylamino)pyrazines exemplified by CYC10268 (4) have been recently disclosed [53]. Pyrazines, such as 4, have been proposed to bind with the 4-N hydrogen bonding to the Cys666 amide NH and the 1-N and 2-NH hydrogen bonding to the Thr663 hydroxyl (gatekeeper residue) [54]. There are also potential interactions with the DFG motif, which is in the ‘‘out’’ configuration. Compound 4 is a 15 nM FMS inhibitor with the ability to block CSF-1mediated survival of murine BMDM [55]. It is also a moderately potent PDGFRb and c-Kit inhibitor with IC50s of 89 and 210 nM respectively. Compound 4 also inhibited both CSF-1-induced ERK1/2 and Akt phosphorylation, and macrophage gene expression. In addition, 4 prevented CSF-1 from priming the macrophages for increased lipopolysaccharide (LPS)-induced production of TNFa, IL-6, and IL-12. In contrast to BMDM, thioglycollate-elicited peritoneal macrophages (TEPM) do not require CSF-1 for survival but behave as CSF-1-primed macrophages. After overnight treatment of TEPM with 4, LPS-induced TNFa, IL-6, and IL-12 production were markedly impaired (regardless
217
Advances in the Discovery of Anti-Inflammatory FMS Inhibitors
of CSF-1 pretreatment), and this effect was not a result of decreased survival. Pyridyl and thiazolyl bisamides have also been prepared; an example is compound 5, which possesses a FMS IC50 of 7 nM. In an assay of CSF1-driven proliferation of 3T3 cells engineered to express human fulllength FMS, 5 inhibited growth with an IC50 of 110 nM [56–58]. In rats, 5 had oral bioavailability of 31%, low clearance (2 mL/kg/min) and a t½ of over 7 h. A mouse pharmacodynamic (PD) model was constructed using 3T3 cells engineered to express human mutant full-length FMS in which the kinase activity was constitutively on. After implantation in nude mice and tumor growth to W250 mm3, phosphoFMS (pFMS) was measured by ELISA. Compound 5, when dosed at 50 mg/kg p.o., resulted in 100% inhibition of pFMS at 2 and 6 h after dosing. O HO N 1 4 N
O
H N
F3C
N H
2
Cl
O
H N
N H
O
N
4
F
S N
5
2.2 Quinazolines, quinolines, and quinolones Early FMS inhibitors were derived from the widely used 4-aminoaryl6,7-dimethoxyquinazoline chemotype [59,60]. Moderate potency was obtained by methylation of the 4-aminoaryl moiety (IC50 ¼ 500 nM) as well as by bioisosteric replacement of the 6,7-dimethoxyquinazoline with a pyrazolo[3,4-d]pyrimidine (IC50 ¼ 180 nM). A quinoline urea derivative, Ki20227 (6) with kinase IC50 of 2 nM as well as modest selectivity versus VEGFR2 and c-Kit (IC50s ¼ 12 and 451 nM respectively), has been shown to inhibit CSF-1-dependent growth of M-NFS-60 cells in vitro [24,61]. Compound 6 also inhibits osteoclast development both in vitro and in vivo and suppresses metastatic tumorinduced osteolysis. Compound 6 inhibited CSF-1-dependent LPSinduced TNF production in BMDM in vitro [26]. Administration of 0.02% 6 in food suppressed joint inflammatory cell infiltration as measured by F4/80 immunostaining and improved the clinical score from day 8 to 30 in a mouse collagen–induced arthritis model (p ¼ 0.0032) [21].
218
Carl L. Manthey and Mark R. Player
A series of 3,4,6-trisubstituted 2-quinolones such as 7 were developed from a quinolone 3-carboxylate ester HTS hit through isosteric replacement. Compound 7 was shown to have FMS kinase and cellular IC50s of 2.5 and 5.0 nM respectively [62]. CSF-1-induced c-fos mRNA elevation in a PD mouse model was reduced to control levels with a dose of 50 mg/kg p.o., demonstrating that FMS signaling was completely inhibited. A co-crystal structure of an analog bound to the inactive form of FMS was obtained, and the N-1 and the O-2 of the keto tautomer were shown to form hydrogen bonds to the backbone of hinge residues Cys666 and Glu664 [50]. A series of 3-amido-4-arylamino-7-arylquinolines has been identified with enzyme IC50s of 8–100 nM, along with 100 nM–3 mM IC50s for growth inhibition of FMS-expressing cell lines [63]. These compounds were generally cleared rapidly, and no kinase selectivity data was reported. A related series of 3-amido-4-arylaminoquinolines, with small 6- and 7-position alkoxy and alkylamino substitutions, has also been identified [64–67]. The series was optimized for reduced clearance and modest cellular potency to achieve 8, which attained an enzyme IC50 of 6 nM, and in an assay based on inhibiting the CSF-1-driven proliferation of huFMS-expressing 3T3 cells, possessed an IC50 of 230 nM. Compound 8 had good pharmacokinetic properties in rats (%F ¼ 100, Cl ¼ 12 mL/ kg/min, t1/2 ¼ 2.1 h). In addition, 8 showed excellent selectivity, only showing appreciable inhibitory activity versus Ark5 (51% at 1 mM) when evaluated in a panel of 85 kinases. In a mouse PD model [56,67], 8 resulted in 90 and 65% inhibition of pFMS at 2 and 6 h after dosing at 25 mg/kg p.o. Homologous 4-aminoarylcinnolines have also recently been described [68].
O
H N
H N O
O
S N
N NC
N H
O O
N H N
6
7
O
Advances in the Discovery of Anti-Inflammatory FMS Inhibitors
219
2.3 Other chemotypes Oxyheteroaryl benzthiazoles, such as 9, have been found to be potent inhibitors of FMS kinase activity (100% inhibition at 1 mM) while offering good selectivity versus PDGFRb and c-Kit [69]. Tyrosine phosphorylation of FMS in HEK293H cells, transfected to express full-length hFMS, was completely inhibited by 9. In addition, 9 completely inhibited the proliferation of CSF-1-dependent M-NFS-60 cells in vitro at a 1 mM concentration. F
F
N
NH
O
O
N
N S
N
N
O
NH2 7
O
N
8
S
O NH
N
N
9
Structure-based optimization of pyrido[2,3-d]pyrimidin-5-ones led to hydroxamate 10, which possessed an IC50 of 400 pM in an enzyme assay and 600 pM in a BMDM cellular assay [22,70–73]. The compound interacts with the hinge region of FMS through hydrogen bonding interactions mediated through the C-2 anilino NH and the pyrimidine N-3. A water-bridged interaction is also present between the C-5 carbonyl oxygen and the hydroxyl group of Thr663. Compound 10 effectively blocked CSF-1-induced c-fos mRNA elevation in a PD mouse model with an ED90 value of 3 mg/kg p.o. During the chronic phase of streptococcal cell wall–induced arthritis in rats, 10 (10, 3, and 1 mg/kg p.o. BID) was highly effective at reversing established joint swelling. In an adjuvant-induced arthritis model in rats, 10 partially prevented joint swelling at 10 mg/kg. In this model, osteoclastogenesis and bone erosion were prevented by low doses (1 or 0.33 mg/kg) that had minimal impact on inflammation. However, at a low dose (0.33 mg/kg), 10 demonstrated good anti-inflammatory activity when given in combination with a sub-maximally effective dose of methotrexate. The diaminopyrimidine GW2580 (11) has been reported to completely inhibit FMS at a concentration of 60 nM [74]. Compound 11 binds to the hinge region of the kinase where the pyrimidine 2-NH2 and 3-N hydrogen bond to the carbonyl and backbone NH of Cys666, while the pyrimidine 4-NH2 forms a bidentate hydrogen bond with the carbonyl of Glu664 and the side chain of Thr663 [75]. Contacts are also made with the backbone NH of Asp796 in the phosphate region with the ortho ether
220
Carl L. Manthey and Mark R. Player
oxygens of 11. The ability of CSF-1 to induce growth of mouse M-NFS-60 myeloid cells and human monocytes in vitro was completely blocked by 11. Compound 11 was selective versus 26 kinases, and this inactivity was confirmed in multiple cell-based assays relevant to those kinases. In vitro, 11 inhibited bone degradation by human osteoclasts and in rat calavaria and rat fetal long bone assays by 80–100% at 1 mM. Dosing of 11 at 40 mg/kg p.o. blocked the ability of a subsequent injection of recombinant CSF-1 to prime mice for enhanced LPS-induced IL-6 production. Compound 11 also inhibited LPS-induced TNFa production with or without CSF-1 priming. In an adjuvant-induced arthritis model, 11 dose-dependently inhibited joint connective tissue and bone destruction measured by radiological, histological, and bone mineral content criteria, while not affecting ankle swelling [20]. N
O N N H
HN
O O
N
NH2
N
N H2N
O
N O
10
O
11
Numerous patent filings also identify 3-substituted-7-azaindoles as selective FMS inhibitors, though many compounds are c-Kit inhibitors as well. Neither discrete IC50s nor in vivo PD or efficacy data have been reported [76–82].
2.4 Multitargeted kinase inhibitors Multitargeted kinase inhibitors have been reported in the literature to include FMS in their target profile. Examples include sunitinib (12), imatinib (13), and ABT-869 (14).
221
Advances in the Discovery of Anti-Inflammatory FMS Inhibitors
N HN
N
N
O N H
N HN
N H
F
O
N N
O N H 13
12
H N
NH2
H N
N HN
O
F
14
Compounds 12 and 14 potently inhibit VEGFR2, PDGFRb, c-Kit, FLT3, and FMS [83], and 12 is approved for the treatment of renal cell carcinoma and gastrointestinal stromal tumors (GIST). The phosphorylation of FMS in a 3T3-hFMS cell line was inhibited by 12 with an IC50 between 50 nM and 100 nM [84,85], and in vitro differentiation of BMDM into osteoclasts was inhibited half maximally at a concentration between 10 nM and 100 nM. A MDA-MB-435-HAL-luc breast cancer xenograft model was used to demonstrate that 12 at 80 mg/kg/day inhibited tumor growth in bone by 89% as measured by bioluminescence imaging. Compound 13 is approved for the treatment of chronic myelogenous leukemia and GIST. In cellular assays, 13 inhibits Abl (IC50 ¼ 250 nM), c-Kit (IC50 ¼ 100 nM), PDGFRb (IC50 ¼ 250 nM), and FMS (IC50 ¼ 1.4 mM). Therapeutic responses in chronic myelogenous leukemia and GIST are based on inhibition of Abl and Kit respectively. Nonetheless, longterm administration of 13 promotes bone formation, a possible consequence of either PDGFR or FMS inhibition, or both [86]. Furthermore, 13 is efficacious in mouse collagen–induced arthritis [87], and case reports indicate potential utility in RA [88]. At present, there is no published data on inhibition of FMS phosphorylation by 12, 13, or 14 in dosed individuals, and it is unclear
222
Carl L. Manthey and Mark R. Player
to what extent FMS is inhibited at prescribed doses. FMS inhibition may contribute to the ability of 12 to inhibit solid tumor growth and of 13 to increase bone formation and to relieve symptoms in RA. However, tolerance issues may limit the widespread use of these agents outside of oncology.
3. CONCLUSION A significant investment has been made in the discovery of a richly diverse set of oral small molecule FMS inhibitors. It can be anticipated that at least some of these compounds will advance to clinical trials and reveal the human pharmacology of selective FMS inhibition. The association of CSF-1 expression and macrophage numbers with diverse human diseases, and the preclinical efficacy of FMS inhibitors in models of cancer, arthritis, and multiple sclerosis, presage a broad range of potential therapeutic applications for these compounds.
REFERENCES [1] V. Chitu and E. R. Stanley, Curr. Opin. Immunol., 2006, 18, 39. [2] J. Hamilton, Nat. Rev. Immunol., 2008, 8, 533. [3] W. Yu, J. Chen, Y. Xiong, F. J. Pixley, X. M. Dai, Y. G. Yeung and E. R. Stanley, J. Leukoc. Biol., 2008, 84, 852. [4] M. G. Cecchini, M. G. Dominguez, S. Mocci, A. Wetterwald, R. Felix, H. Fleisch, O. Chisholm, W. Hofstetter, J. W. Pollard and E. R. Stanley, Development, 1994, 120, 1357. [5] X. M. Dai, G. R. Ryan, A. J. Hapel, M. G. Dominguez, R. G. Russell, S. Kapp, V. Sylvestre and E. R. Stanley, Blood, 2002, 99, 111. [6] L. Bingle, N. J. Brown and C. E. Lewis, J. Pathol., 2002, 196, 254. [7] S. K. Biswas, A. Sica and C. E. Lewis, J. Immunol., 2008, 180, 2011. [8] B. Kascinski, Cancer Treat. Res., 2002, 107, 285. [9] L. Wang, G.-G. Zheng, C.-H. Ma, Y.-M. Lin, H.-Y. Zhang, Y.-Y. Ma, J.-H. Chang and K.-F. Wu, Cancer Res., 2008, 68, 5639. [10] E. Y. Lin, J.-F. Li, L. Gnatovskiy, Y. Deng, L. Zhu, D. A. Grzesik, H. Qian, X. N. Xue and J. W. Pollard, Cancer Res., 2006, 66, 11238. [11] S. Aharinejad, P. Paulus, M. Sioud, M. Hofmann, K. Zins, R. Scha¨fer, E. R. Stanley and D. Abraham, Cancer Res., 2004, 64, 5378. [12] P. Paulus, E. R. Stanley, R. Schafer, D. Abraham and S. Aharinejad, Cancer Res., 2006, 66, 4349. [13] Y. Kubota, K. Takubo, T. Shimizu, H. Ohno, K. Kishi, M. Shibuya, H. Saya and T. Suda, J. Exp. Med., 2009, 206, 1089. [14] A. Sorrentino, C.-G. Liu, A. Addario, C. Peschle, G. Scambia and C. Ferlini, Gynecol. Oncol., 2008, 111, 478. [15] P. Zhu, S. H. Baek, E. M. Bourk, K. A. Ohgi, I. Garcia-Bassets, H. Sanjo, S. Akira, P. F. Kotol, C. K. Glass, M. G. Rosenfeld and D. W. Rose, Cell, 2006, 124, 615. [16] J. B. Wyckoff, Y. Wang, E. Y. Lin, J. F. Li, S. Goswami, E. R. Stanley, J. E. Segall, J. W. Pollard and J. Condeelis, Cancer Res., 2007, 67, 2649.
Advances in the Discovery of Anti-Inflammatory FMS Inhibitors
223
[17] E. Y. Lin, A. V. Nguyen, R. G. Russell and J. W. Pollard, J. Exp. Med., 2001, 193, 727. [18] S. L. Teitelbaum, Am. J. Pathol., 2007, 170, 427. [19] H. Kitarua, P. Zhow, H.-J. Kim, D. V. Novack, F. P. Ross and S. L. Teitelbaum, J. Clin. Invest., 2005, 12, 3418. [20] J. G. Conway, H. Pink, M. L. Bergquist, B. Han, S. Depee, S. Tadepalli, P. Lin, R. C. Crumrine, J. Binz, R. L. Clark, J. L. Selph, S. A. Stimpson, J. T. Hutchins, S. D. Chamberlain and T. A. Brodie, J. Pharmacol. Exp. Ther., 2008, 326, 41. [21] H. Ohno, Y. Uemura, H. Murooka, H. Takanashi, T. Tokieda, Y. Ohzeki, K. Kubo and I. Serizawa, Eur. J. Immunol., 2008, 38, 283. [22] H. Huang, D. A. Hutta, J. M. Rinker, H. Hu, W. H. Parsons, C. Schubert, R. L. DesJarlais, C. S. Crysler, M. A. Chaikin, R. R. Donatelli, Y. Chen, D. Cheng, Z. Zhou, E. Yurkow, C. L. Manthey and M. R. Player, J. Med. Chem., 2009, 52, 1081. [23] C. R. Illig, J. Chen, M. J. Wall, K. J. Wilson, S. K. Ballentine, J. M. Rudolf, R. J. DesJarlais, Y. Chen, C. Schubert, I. P. Petrounia, C. M. Crysler, C. J. Molloy, M. A. Chaikin, C. L. Manthey, M. R. Player, B. E. Tomczuk and S. K. Meegalla, Bioorg. Med. Chem. Lett., 2008, 18, 1642. [24] H. Ohno, K. Kubo, H. Murooka, Y. Kobayashi, T. Nishitoba, M. Shibuya, T. Yoneda and T. Isoe, Mol. Cancer Ther., 2006, 5, 2634. [25] http://www.clinicaltrials.gov/ct2/show/NCT00550355 [26] P. Durez, J. Malghem, A. N. Toukap, G. Depresseux, B. R. Lauwerys, R. Westhovens, F. P. Luyten, L. Corluy, F. A. Houssiau and P. Verschueren, Arth. & Rheum., 2008, 58, 1299. [27] J. J. Haringman, D. M. Gerlag, A. H. Zwinderman, T. J. M. Smeets, M. C. Kraan, D. Baeten, I. B. McInnes, B. Bresnihan and P. P. Tak, Ann. Rheum. Dis., 2005, 64, 834. [28] C. Q. Chu, M. Field, S. Allard, E. Abney, M. Feldmann and R. N. Maini, Br. J. Rheumatol., 1992, 31, 653. [29] I. K. Campbell, G. Ianches and J. A. Hamilton, Biochim. Biophy. Acta, 1993, 1182, 57. [30] K. Nakano, Y. Okada, K. Saito, R. Tanikawa, N. Sawamukai, Y. Sasaguri, T. Kohro, Y. Wada, T. Kodama and Y. Tanaka, Rheumatol. (Oxford), 2007, 46, 597. [31] I. Rioja, F. J. Hughes, C. H. Sharp, L. C. Warnock, D. S. Montgomery, M. Akil, A. G. Wilson, M. H. Binks and M. C. Dickson, Arth. & Rheum., 2008, 58, 2257. [32] I. K. Campbell, M. J. Rich, R. J. Bischof and J. A. Hamilton, J. Leukoc. Biol., 2000, 68, 144. [33] N. M. Isbel, D. J. Nikolic-Paterson, P. A. Hill, J. Dowling and R. C. Atkins, Nephrol. Dial. Transplant., 2001, 16, 1638. [34] P. T. Yang, W. G. Xiao, L. J. Zhao, J. Lu, L. M. He, H. Kasai and M. Ito, Ann. Rheum. Dis., 2008, 67, 429. [35] S. Tian, J. Li, L. Wang, T. Liu, H. Liu, G. Cheng, D. Liu, Y. Deng, R. Gou, Y. Wan, J. Jia and C. Chen, Inflamm. Res., 2007, 56, 304. [36] D. M. Lenda, E. R. Stanley and V. R. Kelley, J. Immunol., 2004, 173, 4744. [37] J. Menke, M.-Y. Hsu, K. T. Byrne, J. A. Lucus, W. A. Rabcal, B. P. Croker, X.-J. Zong, E. R. Stanley and V. R. Kelley, J. Immunol., 2008, 181, 7367. [38] L. S. Rallidis, M. G. Zolindaki, P. C. Pentzeridis, K. P. Poulopoulos, A. H. Velissaridou and T. S. Apostolou, Heart, 2004, 90, 25. [39] F. H. Klebl, J. E. Olsen, S. Jain and W. F. Doe, J. Pathol., 2001, 195, 609. [40] H. Kreipe, H. J. Radzun, K. Heidorn, J. Barth, J. Kiemle-Kallee, W. Petermann, J. Gerdes and M. R. Parwaresch, Lab. Invest., 1990, 62, 697. [41] I. Harman-Boehm, M. Bluher, H. Redel, N. Sion-Vardy, S. Ovadia, E. Avinoach, I. Shai, N. Kloting, M. Stumvoll, N. Bashan and A. Rudich, J. Clin. Endocrinol. & Metabol., 2007, 92, 2240. [42] T. Rajavashisth, J. H. Qiao, S. Tripathi, J. Tripathi, N. Mishra, M. Hua, X. P. Wang, A. Loussararian, S. Clinton, P. Libby and A. Lusis, J. Clin. Invest., 1998, 101, 2702.
224
Carl L. Manthey and Mark R. Player
[43] D. Marshall, J. Cameron, D. Lightwood and A. D. Lawson, Inflamm. Bowel Dis., 2007, 13, 219. [44] M. Mu¨ller, M. Berghoff, I. Kobsar, R. Kiefer and R. Martini, Exp. Neurol., 2007, 203, 55. [45] Y. Uemura, H. Ohno, Y. Ohzeki, H. Takanashi, H. Murooka, K. Kubo and I. Serizawa, J. Neuroimmunol., 2008, 195, 73. [46] J.-E. Ghia, F. Galeazzi, D. C. Ford, C. M. Hogaboam, B. A. Vallance and S. Collins, Am. J. Physiol. Gastrointest. Liver Physiol., 2008, 294, G770. [47] R. J. Patch, B. M. Brandt, D. Asgari, N. Baindur, N. K. Chadha, T. Georgiadis, W. S. Cheung, I. P. Petrounia, M. A. Chaikin and M. R. Player, Bioorg. Med. Chem. Lett., 2007, 17, 6070. [48] M. R. Player, N. Baindur, B. Brandt, N. Chadha, R. J. Patch, D. Asgari and T. Georgiadis, US Patent 7,427,683, 2008. [49] M. R. Player, N. Baindur, B. Brandt, N. Chadha, R. J. Patch, D. Asgari and T. Georgiadis, US Patent 7,429,603, 2008. [50] C. Schubert, C. Schalk-Hihi, G. T. Struble, H.-C. Ma, I. P. Petrounia, B. Brandt, I. C. Deckman, R. J. Patch, M. R. Player, J. C. Spurlino and B. A. Springer, J. Biol. Chem., 2007, 282, 4094. [51] C. R. Illig, S. K. Ballentine, J. Chen, R. L. DesJarlais, S. K. Meegalla, M. Wall and K. Wilson, US Patent 7,414,050, 2008. [52] S. K. Meegalla, J. Chen, M. J. Wall, K. J. Wilson, S. K. Ballentine, R. J. DesJarlais, C. Schubert, R. Donatelli, C. M. Crysler, Y. Chen, C. J. Molloy, M. A. Chaikin, C. L. Manthey, M. R. Player, B. E. Tomczuk and C. R. Illig, Bioorg. Med. Chem. Lett., 2008, 18, 3632. [53] C. J. Burns, M. F. Harte and J. T. Palmer, WO Patent Application 2008/058341-A1, 2008. [54] C. J. Burns, M. F. Harte, X. Bu, E. Fantino, M. Giarusso, M. Joffe, M. Kurek, F. S. Legge, P. Razzino, S. Su, H. Treutlein, S. S. Wan, J. Zeng and A. F. Wilks, Bioorg. Med. Chem. Lett., 2009, 19, 1206. [55] K. M. Irvine, C. J. Burns, A. F. Wilks, S. Su, D. A. Hume and M. J. Sweet, FASEB J., 2006, 20, 1921. [56] D. A. Scott, B. M. Aquila, G. A. Bebernitz, D. J. Cook, L. A. Dakin, T. L. Deegan, M. M. Hattersley, S. Ioannidis, P. D. Lyne, C. A. Omer, M. Ye and X.-K. Zheng, Bioorg. Med. Chem. Lett., 2008, 18, 4794. [57] B. Aquila, D. Cook L. Dakin, S, Ioannidis, P. Lyne, D. Scott and X. Zhang, WO Patent Application 2007/071955-A1, 2007. [58] L. Almeida, B. Aquila, D. Cook, S. Cowen, L. Daikan, J. Ezhuthachan, S. Ioannidis, S. Lee, P. Lyne, T. Pontz, D. Scott, M. Su and X. Zheng, WO Patent Application 2006/ 067445-A2, 2006. [59] M. R. Myers, N. N. Setzer, A. P. Spada, P. E. Persons, C. Q. Ly, M. P. Maguire, A. L. Zulli, D. L. Cheney, A. Zilberstein, S. E. Johnson, C. F. Franks and K. J. Mitchell, Bioorg. Med. Chem. Lett., 1997, 7, 421. [60] M. R. Myers, A. P. Spada, M. P. Maguire, P. E. Persons, A. Zilberstein, C.-Y. J. Hou and S. E. Johnson, US Patent 5,714,493, 1998. [61] K. Kubo, H. Ohno, T. Isoe and T. Nishitoba, WO Patent Application 2006/0235033-A1, 2006. [62] M. J. Wall, J. Chen, S. Meegalla, S. Ballentine, K. Wilson, R. J. DesJarlais, C. Schubert, M. A. Chaikin, C. Crysler, R. R. Donatelli, E. J. Yurkow, M. R. Player, R. J. Patch, C. L. Manthey, C. Molloy, B. Tomczuk and C. Illig, Bioorg. Med. Chem. Lett., 2008, 18, 2097. [63] T. L. Smalley, S. D. Chamberlain, W. Y. Mills, D. L. Musso, S. A. Randhawa, J. A. Ray, V. Samano and L. Frick, Bioorg. Med. Chem. Lett., 2007, 17, 6257. [64] D. Cook, L. Dakin, D. Delvalle, T. Gero, D. Scott and X. Zheng, WO Patent Application 2007/119046-A1, 2007.
Advances in the Discovery of Anti-Inflammatory FMS Inhibitors
225
[65] L. Dakin, K. Daly, D. Delvalle, T. Gero, C. A. Ogoe, D. Scott and X. Zheng, WO Patent Application 2008/056148-A1, 2008. [66] D. A. Scott, C. L. Balliet, D. J. Cook, A. M. Davies, T. W. Gero, C. A. Omer, S. Poondru, M.-E. Theoclitou, B. Tyurin and M. J. Zinda, Bioorg. Med. Chem. Lett., 2009, 19, 697. [67] D. A. Scott, K. J. Bell, C. T. Campbell, D. J. Cook, L. A. Dakin, D. J. Del Valle, L. Drew, T. W. Gero, M. M. Hattersley, C. A. Omer, B. Tyurin and X. Zheng, Bioorg. Med. Chem. Lett., 2009, 19, 701. [68] L. Dakin, C. A. Ogoe, D. Scott and X. Zheng, WO Patent Application 2009/0012084-A1, 2009. [69] J. C. Sutton, M. Wiesmann, W. Wang, M. Lindvall, J. Lan, S. Ramurthy, A. Sharma, E. J. Mieuli, L. M. Klivansky, W. Lenahan, S. Kaufman, H. Yang, S. C. Ng, K. Pfister, A. S. Wagman, V. Sung and M. Sendzik, WO Patent Application 2008/0045528-A1, 2008. [70] H. Huang, D. A. Hutta, H. Hu, R. L. DesJarlais, C. Schubert, I. P. Petrounia, M. A. Chaikin, C. L. Manthey and M. R. Player, Bioorg. Med. Chem. Lett., 2008, 18, 2355. [71] M. R. Player, H. Huang and D. A. Hutta, WO Patent Application 2007/0060577-A1, 2007. [72] M. R. Player, H. Huang and D. A. Hutta, WO Patent Application 2007/0060578-A1, 2007. [73] M. R. Player, W. H. Parsons, H. Huang, D. A. Hutta, H. Hu and J. Rinker, WO Patent Application 2008/0114007-A1, 2008. [74] J. G. Conway, B. McDonald, J. Parham, B. Keith, D. W. Rusnack, E. Shaw, M. Jansen, P. Lin, A. Payne, R. M. Crosby, J. H. Johnson, L. Frick, M. Jasmine Lin, S. Depee, S. Tadepalli, B. Votta, I. James, K. Fuller, T. J. Chambers, F. C. Kull, S. D. Chamberlain and J. T. Hutchins, Proc. Natl. Acad. Sci. U.S.A., 2005, 102, 16078. [75] L. M. Schewchuk, A. M. Hassell, W. D. Holmes, J. M. Veal, H. K. Emmerson, D. L. Musso, S. D. Chamberlain and G. E. Peckham, US Patent Application 2004/ 0002145-A1, 2004. [76] P. N. Ibrahim, D. R. Artis, R. Bremer, S. Mamo, M. Nespi, C. Zhang, J. Zhang, Y.-L. Zhu, J. Tsai, K.-P. Hirth, G. Bollag, W. Spevak, H. Cho, S. Gillette, G. Wu, H. Zhu and S. Shi, WO Patent Application 2007/002325-A1, 2007. [77] P. N. Ibrahim, D. R. Artis, R. Bremer, G. Habets, S. Mamo, M. Nespi, C. Zhang, J. Zhang, Y.-L. Zhu, R. Zuckerman, B. West, Y. Suzuki, J. Tsai, K.-P. Hirth, G. Bollag, W. Spevak, H. Cho, S. Gillette, G. Wu, H. Zhu and S. Shi, WO Patent Application 2007/ 002433-A1, 2007. [78] C. Zhang, J. Zhang, P. N. Ibrahim, C. R. Hurt, R. Zuckerman, D. R. Artis, R. Bremer, W. Spevak, G. Wu and H. Zhu, WO Patent Application 2007/013896-A2, 2007. [79] C. Zhang, J. Zhang, P. N. Ibrahim, D. R. Artis, R. Bremer, G. Wu, H. Zhu and M. Nespi, WO Patent Application 2008/063888-A2, 2008. [80] C. Zhang, J. Zhang, P. N. Ibrahim, D. R. Artis, R. Bremer, G. Wu, H. Zhu and M. Nespi, WO Patent Application 2008/064255-A2, 2008. [81] C. Zhang, J. Zhang, P. N. Ibrahim, D. R. Artis, R. Bremer, G. Wu, H. Zhu and M. Nespi, WO Patent Application 2008/064265-A2, 2008. [82] P. N. Ibrahim, R. Bremer, C. Zhang, J. Zhang, K.-P. Hirth, G. Wu and H. Zhu, WO Patent Application 2008/080001-A2, 2008. [83] J. Guo, P. A. Marcotte, J. O. McCall, Y. Dai, L. J. Pease, M. R. Michaelides, S. K. Davidsen and K. B. Glaser, Mol. Can. Ther., 2006, 5, 1007. [84] L. Murray, T. Abrams, K. R. Long, T. J. Ngai, L. M. Olson, W. Hong, P. K. Keast, J. A. Brassard, A. M. O’Farrell, J. M. Cherrington and N. K. Pryer, Clin. Exp. Met, 2003, 20, 757. [85] L. Murray, A.-M. O’Farrell and T. Abrams, US Patent Application 2004/0209937-A1, 2004. [86] S. Fitter, A. L. Dewar, P. Kostakis, L. B. To, T. P. Hughes, M. M. Roberts, K. Lynch, B. Vernon-Roberts and A. C. W. Zannettino, Blood, 2008, 111, 2538.
CHAPT ER
11 Recent Advances in the Discovery of CB2 Selective Agonists Sangdon Han, Jayant Thatte and Robert M. Jones
Contents
1. Introduction 2. Evaluation of CB2 Agonists In Preclinical Models 2.1 Pain 2.2 Experimental autoimmune encephalomyelitis 2.3 Other potential indications 3. Medicinal Chemistry 3.1 Monocyclic core-based CB2 agonists 3.2 Bicyclic core-based CB2 agonists 3.3 Tricyclic core-based CB2 agonists 4. Clinical Trials Status 5. Conclusions References
227 228 228 231 232 232 233 237 241 241 242 242
1. INTRODUCTION The CB2 receptor is a member of a family of G-protein-coupled receptors (GPCRs) that mediate the effects of several structurally related endocannabinoids such as anandamide and 2-arachidonoylglycerol (2-AG) [1]. The endocannabinoid system is important in an array of biological processes involving the central nervous system, the immune system, and metabolism. CB2 appears to be a promising therapeutic target for the treatment of pain and inflammation. However, the therapeutic utility of Arena Pharmaceuticals Inc., 6166 Nancy Ridge Dr., San Diego, CA 92121, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04411-X
r 2009 Elsevier Inc. All rights reserved.
227
228
Sangdon Han et al.
non-selective, brain penetrable cannabinoid agonists is limited by undesirable psychotropic effects associated with activation of the CB1 receptor. Such side effects are not apparent upon CB2 receptor activation. Selectivity for CB2 activation over CB1 activation is thus an essential feature needed for therapeutic uses directed at CB2. CB2 receptor activation results in the inhibition of adenylyl cyclase [2] and mitogenactivated protein (MAP) kinase activation [3] through coupling to the asubunit of Gi/o proteins. Although expression of CB2 was initially thought to be predominantly restricted to immune cells in the periphery [4], recent data suggest that this receptor is also expressed centrally in perivascular microglial cells [5] and in brainstem neurons [6]. CB2 is up-regulated in both dorsal root ganglia and peripheral neurons following injury [7,8]. Preclinical data using CB2 agonists and receptor knockout mice have validated the receptor as a potential target for the treatment of pain and neuro-inflammation. However, the challenge for this field has remained the identification of highly selective CB2 agonists that maintain functional activity across species. Despite these difficulties, several candidate molecules have advanced into clinical development. In this chapter, we describe pharmacological studies using novel CB2 agonists, which suggest utility for these agents in the treatment of pain and other disorders. We also describe the key medicinal chemistry features of this class of compounds, their current clinical trial status and future prospects as therapeutics.
2. EVALUATION OF CB2 AGONISTS IN PRECLINICAL MODELS 2.1 Pain CB2 agonists such as GW405833 (1) have been found to be efficacious in several preclinical models for inflammatory, neuropathic, and postoperative pain. GW405833 is a selective CB2 agonist with a binding affinity of 3.9271.58 nM for human CB2 versus 477271676 nM for CB1 (W1,000-fold selectivity) [9]. Selectivity for CB2 was diminished for rat receptors, with only a 78-fold selectivity ratio for CB2 over rat CB1. Functionally, this compound appeared to be a partial agonist at the human CB2 receptor, showing maximum inhibition of forskolinmediated cAMP production of only 44.673.4% [9]. Interestingly, 1 has also been reported to be a partial inverse agonist at both human and rat CB2 receptors [10], suggesting that it might be a protean agonist, properties that were previously reported for a structurally similar compound, AM1241 [11]. The i.v. pharmacokinetic profile of 1 in rats included a plasma half-life of 3.68 h; clearance rate of 1.9670.06 L/h/kg, volume of distribution 10.470.2 L/kg, and brain-to-plasma ratio of 5.1570.43.
Recent Advances in the Discovery of CB2 Selective Agonists
229
In a neuropathic pain model utilizing partial ligation of sciatic nerve, 1 showed 63.3% reversal of pain at 1 h following a 10 mg/kg i.p. dose with an ED50 ¼ 0.077 mg/kg. In the hind paw surgical incision model and the Freund’s complete adjuvant (FCA)-induced inflammatory pain model, 1 displayed a dose-dependent reversal of hyperalgesia with an ED50 ¼ 2.58 mg/kg and 0.17 mg/kg, respectively, at the 1 h time point. The on-target effect of this compound was confirmed in CB2 knockout mice where it did not reverse inflammatory pain. Central effects of 1 were evaluated in the catalepsy assay, the rotorod assay for motor function, and tail flick and hot plate tests for acute analgesia. The highest dose, 100 mg/kg, induced catalepsy and decreased rotorod performance, suggesting CB1-associated activity at this dose, but not at the lower doses [9]. Abbott has published data for two CB2 agonists, A-796260 (2) [10] and A-836339 (3) [12]. In the [3H]CP-55,940 competitive binding assay, A-796260 was 193-fold selective for human CB2 with Ki ¼ 845 nM for hCB1 versus 4.37 nM for hCB2, and 30-fold selective for rat CB2 with Ki ¼ 395 nM for rat CB1 compared to 13 nM for rat CB2. Functional selectivity was 1,380-fold for hCB2 and 175-fold for rat CB2 as determined using the adenylate cyclase assay. Although the pharmacokinetic properties for A-796260 have not been described, it was administered intraperitoneally in rats in specific pharmacological models. In the FCA-induced model of thermal hyperalgesia, A-796260 displayed dose-dependent efficacy (ED50 ¼ 2.8 mg/kg). In this model, specificity was confirmed using two CB2-receptor-selective antagonists: SR144528 [13] and AM630 [14], both of which blocked the analgesic effects of A-796260. In contrast, the CB1-receptor-selective antagonist SR141716A had no effect on A-796260-induced analgesia. In the skin incision model of postoperative pain, A-796260, administered i.p. 1.5 h post-surgery dose-dependently attenuated tactile allodynia with an ED50 ¼ 18 mg/kg and maximal efficacy of 6877%. Tolerance to A-796260 did not develop in this model after repeat administration, b.i.d. for 5 days. In the chronic constriction injury of sciatic nerve model of neuropathic pain, A-796260 dose-dependently attenuated tactile allodynia with an ED50 ¼ 15 mg/kg and 6679% efficacy. A-796260 is the only CB2 agonist reported to date to have an efficacy in mono-iodoacetate-induced osteoarthritis pain. Efficacy was comparable to celecoxib (56% efficacy at 35 mg/kg dose of A-796260 versus 62% efficacy at 38 mg/kg dose of celecoxib). This compound did not show significant effects in locomotor activity up to a dose of 35 mg/kg. A-836339 exhibited sub-nanomolar potencies in competition binding assays (Ki values of 0.64 nM for hCB2 and 0.76 nM for rat CB2), with 425- and 189-fold selectivity for CB2 receptor over the human and rat CB1 receptors, respectively. Functionally, the potency of A-836339
230
Sangdon Han et al.
(EC50 ¼ 1.6 nM) was similar to A-796260 (EC50 ¼ 0.71 nM). A-836339 demonstrated full agonist activity (Emax ¼ 102%) at the CB2 receptor and also showed weaker but full agonist activity at both human and rat CB1 receptors (EC50 ¼ 740 nM and 1,200 nM, respectively). A-796260, as well as A-836339, showed minimal off-target (o50% inhibition) activities in the CEREP binding platform containing a panel of 74 GPCRs and ion channels. The only activity observed with A-796260 was on the d-opioid receptor (67% displacement) and with 10 mM A-836339 on the A3 and 5-HT2C receptors (55% and 53%, respectively). In the FCA-induced inflammatory pain model, A-836339 dose-dependently reversed thermal hyperalgesia with an ED50 ¼ 1.96 mmol/kg. A-836339 was also found to be efficacious in the skin incision model, the capsaicin-induced mechanical hyperalgesia model and in the chronic constriction injury model of neuropathic pain with ED50 ¼ 12 mmol/kg, 10.4 mmol/kg, and 12.9 mmol/kg, respectively. The activities of both A-796260 and A-836339 were found to be m-opioid receptor-independent since naloxone did not antagonize their anti-hyperalgesic effects. Despite the high in vitro selectivity of A-836339 for the CB2 receptor over the CB1 receptor (189fold), it clearly showed a CB1-associated reduction in horizontal locomotor activity at the 15 and 45 mmol/kg doses, which was blocked by treatment with the CB1 receptor antagonist SR141716A.
O N O O
O
N N
S
N
N
O N
Cl
O
Cl 1
O
2
3
GlaxoSmithKline’s (GSK) clinical candidate GW842166X (4) was reported to be fully selective on human CB2 receptor with EC50 ¼ 63 nM and 91 nM for human and rat CB2 receptors, respectively [15]. It had no significant agonist activity at concentrations up to 30 mM on either human or rat CB1 receptors. Oral bioavailability was 58% with a 3 h half-life and a brain:plasma ratio of 0.8. In the FCA model of inflammatory pain, it showed high potency with oral ED50 ¼ 0.1 mg/kg and full reversal of
Recent Advances in the Discovery of CB2 Selective Agonists
231
hyperalgesia at the 0.3 mg/kg dose. In 4-day studies, development of tolerance was not observed. MDA7 (5), a compound developed at the MD Anderson Cancer Center, was shown to dose-dependently attenuate tactile allodynia produced by spinal nerve ligation or paclitaxel in rats [16]. In a [3H]CP55,840 radioligand displacement assay, 5 showed no affinity for the human CB1 receptor at concentrations up to 10 mM, with a 4227123 nM Ki on human CB2. On rat receptors, the mean Ki values were 25657695 nM and 2387143 nM for CB1 and CB2 receptors, respectively. In the GTPgS functional assay, this compound showed weak partial agonist activity on human CB1 receptor at concentration of W1 mM. A transient thermal anti-nociceptive effect was reported in naı¨ve rats at the 10 mg/kg dose administered intraperitoneally. PF-03550096 (6), another cross-species-selective agonist from Pfizer, was reported to significantly suppress visceral hypersensitivity in the 2,4,6-trinitrobenzene sulfonic acid (TNBS)-induced colonic pain model in rats [17,18]. SR144528 reversed the inhibitory effects of 6 indicating that the compound is functionally active on the rat CB2 receptor. O
H2N O
O
O
N
NH O
F3C N
N
N
O
O N
NH Cl
Cl 4
OH 5
6
2.2 Experimental autoimmune encephalomyelitis In this preclinical animal model for multiple sclerosis (MS), two key sets of data lend support for a potential role for CB2 agonists in treatment. First, experimental autoimmune encephalomyelitis (EAE) is exacerbated in CB2 knockout mice upon disease induction compared to wild-type littermates [19,20]. Second, therapeutic treatment of mice at the peak of disease with the CB2 agonist HU-308 (7, Ki human CB2 ¼ 22.7+3.9 nM; human CB1 KiW10 mM; Ki for rodent CB2 not reported) [21] show a significant reduction in disease score compared to the vehicle-treated
232
Sangdon Han et al.
animals [20]. This effect was attributed to decreased microglial and infiltrating myeloid cell proliferation. HO CF3 O O N Cl O 7
N H
N H N
N
8
2.3 Other potential indications CB2 agonists have shown promise in other indications at the preclinical level. GW833972A (8), a 1,000-fold CB2-selective compound, was shown to inhibit citric acid-induced cough in a conscious guinea-pig model [22]. The CB2 antagonist SR144528 but not the CB1 antagonist SR141716A, reversed the effect of 8, demonstrating involvement of the CB2 receptor. A potential role for CB2 agonists in the treatment of osteoporosis has been proposed based on in vitro suppression of trabecular osteoclastogenesis due to inhibition of proliferation of osteoclast precursors and receptor activator of NFkB ligand, RANKL, by the CB2 agonist 7 [23]. In the same report, 7 was tested in an ovarectomized mouse model of postmenopausal osteoporosis. The CB2 agonist HU-308 attenuated trabecular bone loss from 41% in the untreated mice to 27% in this model. A potential role for CB2 in treatment of atherosclerosis in apolipoprotein E knockout mice has been reported using the non-selective cannabinoid agonist D9-tetrahydrocannabinol (THC) [24]. Involvement of CB2 was demonstrated in this report using CB2 antagonist SR144528, which completely abolished efficacy of THC.
3. MEDICINAL CHEMISTRY This section is focused on advances that have occurred in the medicinal chemistry of CB2 agonists since the publication of an earlier review [25]. With the recent entry of several CB2-selective agonists into the clinic, there has been an increase in the number of reports of novel selective agonists with demonstrable efficacy in both acute and chronic pain models. Pharmacophore-based de novo virtual screening methods [26] and high-throughput hit-based optimization have been the two main
233
Recent Advances in the Discovery of CB2 Selective Agonists
strategies adopted for the discovery of new ligands. Broadly speaking, the majority of the reported selective CB2 agonists are characterized by an aromatic heterocyclic core connected to a bulky alkyl or an aryl amide motif. In general, these highly lipophilic molecules possess high CB2 receptor affinity with tandem selectivity over the CB1 subtype.
3.1 Monocyclic core-based CB2 agonists 3.1.1 Six-membered aromatic cores
O Cl
N H Cl
N H
N
Cl
N
O Cl
9
N H
O
N 10
Following the disclosure of the chemical structure of the GSK clinical candidate (4), a number of alternate six-membered ring aromatic corebased CB2 ligands have appeared in the literature. Motivated in part by the poor aqueous solubility of 4 (2.0 mg/mL at pH 7.4), replacement of the pyrimidine core with pyridine was scrutinized [27]. The nature of the aniline group appears to be important for aqueous solubility of the pyridine series. Replacement of the 2,4-dichloro aniline in 9 with 3-methoxy aniline afforded improved aqueous solubility (168 mg/mL at pH 7.4), although this modification reduced affinity for the CB2 receptor. Compound 9 was found to be selective over CB1 (hCB2/hCB1 EC50 ¼ 79 nM/W30 mM) in a yeast reporter assay. Although 9 had low aqueous solubility (7.0 mg/mL at pH 7.4), it exhibited good exposure and oral bioavailability in rat pharmacokinetic studies (AUC(0t)/dose ¼ 15 min g/L, %F ¼ 39, 3.0 mg/kg, p.o.). Compound 9 was evaluated in the FCA model of inflammatory pain (ED50 ¼ 0.07 mg/kg, p.o.) and demonstrated full reversal of hyperalgesia (0.3 mg/kg, p.o.) [27]. New reports describing a conformationally restricted morpholinyl motif, exemplified by 10, have appeared [28–30]. Compound 10 retained in vitro activity and selectivity for CB2 (hCB2 EC50 ¼ 10 nM, selectivity (hCB1) W2,000-fold); however, it was not competitive with [3H]-CP55,940 in a binding assay (hCB2 KiW10 mM) possibly indicating an interaction with a different binding pocket within the CB2 receptor. Interestingly, this (R)-phenyl morpholine analog was reported to function as an inverse agonist. Compound 10 demonstrated oral efficacy in a mouse zymosan-induced paw inflammation (ZIPI) model (85%
234
Sangdon Han et al.
inhibition at 100 mg/kg, p.o.). Its ZIPI model efficacy confirmed an antiinflammatory activity at CB2 [28,29].
R1
O S O
N
O S O
O N
R1 O
O HN
O S O
NH
N O
11 R1 = morpholine
12
13
Several sulfamoyl benzamide-based CB2-selective agonists have emerged, composed of a central phenyl ring with a bulky lipophilic cycloalkyl carboxamide [31–34]. Saturated cyclic amines such as morpholine, pyrrolidine, and piperidine at R1 displayed similar in vitro profiles, with 11 being slightly more selective for CB2 [31]. Amide 11 exhibited 120-fold selectivity over CB1 in the [35S]GTPgS functional assay (hCB2/hCB1 EC50 ¼ 4.6 nM/550 nM) and did not produce catalepsy at doses of 6 and 10 mg/kg, i.p. Compound 11 reversed the nerve injuryinduced tactile allodynia in the L5 SNL rat model of neuropathic pain (3 mg/kg, i.p.) and also produced a significant anti-allodynic effect (10 mg/kg, i.p.) comparable to the effect produced by morphine (3 mg/ kg, s.c.) in the hind paw incision model of post-operative pain. However, 11 had no oral efficacy in the incision pain model due to poor oral bioavailability [31]. Structure–activity relationship studies of a closely related genus of sulfamoyl benzamides have also been reported [32], wherein, 12 (R1 ¼ CH3) is described as the preferred compound in this series (hCB2/ hCB1 EC50 ¼ 4 nM/W8.9 mM in the adenylate cyclase assay) [33]. The desmethyl analog (R1 ¼ H), or derivatives, composed of a large alkyl group (R1 ¼ cyclohexyl) resulted in a significant loss of potency (hCB2 EC50 W20 mM). The anti-inflammatory effect of 12 (R1 ¼ cyclohexyl) was evaluated in a mouse zymosan-induced paw inflammation (ZIPI) model [33]. A retro-amide modification of 11 led to a new series of agonists exemplified by 13 (hCB2 EC50 ¼ 11 nM). These compounds possessed improved selectivity over CB1 while maintaining good binding affinity for CB2 [34]. Compound 13 demonstrated anti-allodynic activity in the hind paw incision model of post-operative pain when delivered intraperitoneally
235
Recent Advances in the Discovery of CB2 Selective Agonists
(30 mg/kg). However, 13 was not orally efficacious in the incision pain model due to metabolic instability (RLM ¼ 2% remaining at 30 min). Oral efficacy in the skin incision model from the aminobenzotriazole (ABT, the cytochrome P450 suicide inhibitor) pretreated animals also confirmed a rapid metabolism of 13 in the liver [34].
3.1.2 Five-membered heterocyclic cores N
N Cl
N H
Cl
Cl
N
Cl N
N F
O N
O N
14
F
15
1,2,4-Oxadiazole-based CB2 agonists have been identified following examination of a number of alternative five-membered heterocycles derived from an initial HTS hit [35]. The amino quinoline analog 14 (hCB2/hCB1 EC50 ¼ 2.2 nM/550 nM in the adenylate cyclase assay) displayed improved physicochemical properties [simulated intestinal fluid (SIF) solubility ¼ 87 mg/mL] and a good pharmacokinetic profile (AUC ¼ 43.1 mg.h/mL, Cmax ¼ 8.4 mg/mL, t1/2 ¼ 5.1 h, and %FW100 at 10 mg/kg, p.o.) in rats [35]. Structural modification of 14 led to the conformationally constrained N-arylpiperidine oxadiazole 15 with improved metabolic stability (Clint ¼ 38 mL/min/mg in HLM) and in vitro selectivity (hCB2/hCB1 EC50 ¼ 11 nM/W2.0 mM) in the cyclase assay. Oxadiazole 15 displayed reasonable exposure despite poor oral bioavailability (AUC024h ¼ 5240 ng h/mL, Cmax ¼ 682 ng/mL, t1/2 ¼ 8.9 h, and %F ¼ 2 at 10 mg/kg, p.o.) in rats [36,37]. O CF3
CF3
N N
N O
S
N N
N O
F O
N NH
S N
16
17
18
F F
236
Sangdon Han et al.
For thiazolylidene benzamide 16 [38,39], the cyclopropylmethyl group at the 3-position and a bulky lipophilic tert-butyl group at 5position of the thiazole were important for obtaining high affinity and selectivity for the CB2 receptor (16, hCB2 IC50 ¼ 13 nM, CB2 selectivity ¼ 270-fold in the [3H]-CP-55,940 binding assay). Amide 16 also possessed good stability in HLM (93% remaining at 15 min) and an acceptable pharmacokinetic profile (AUC024h ¼ 326 ng h/mL, Cmax ¼ 43.1 ng/mL, t1/2 ¼ 4.8 h, %F ¼ 52, 3 mg/kg) in rats [39]. A report describing the synthesis and SAR of a related pyrazolylidene derivative CBS0550, 17, has also appeared [40]. Although 16 was poorly soluble (o0.01 mg/100 mL in water), the pyrazole derivative 17 exhibited improved solubility in water (5.9 mg/100 mL) presumably due to its increased basicity. Compound 17 also displayed an improved in vitro profile in the [3H]-CP-55,940 binding assay (hCB2 IC50 ¼ 2.9 nM, CB2 selectivity ¼ 1400-fold). The pharmacokinetic properties of 17 were evaluated in rats (AUC08h ¼ 2160 ng h/mL, Cmax ¼ 545 ng/mL, t1/2 ¼ 2.5 h at 10 mg/kg, p.o.), where it demonstrated dose-dependent reversal of mechanical hyperalgesia in the Randall-Selitto model of inflammatory pain (10 mg/kg and 30 mg/kg, p.o.), whereby the antinociceptive effect lasted at least 3 h after administration [40]. A new series of CB2 receptor modulators based on a thiazole core have also been reported [41]. Analog 18 is representative (hCB2 EC50 ¼ 0.7 nM in the cAMP assay). The thiazole appears to be optimal, as replacement with an isoxazole or a phenyl ring led to a decrease in potency [41].
S O O
Cl
H N O 19
O N
S O O
H N O
N N
20
Two closely related a-amidosulfone series have been independently described by two different research groups [42,43]. Alkylsulfone 19 exhibited potent CB2 agonist activity (hCB2 EC50 ¼ 0.04 nM) in the adenylate cyclase assay, and cyclo-alkyl derivatives (cyclohexane, cycloheptane, tetrahydropyran, etc.) also retained sub-nanomolar CB2 activity, but no further biological data was given [42]. Arylsulfone 20 appears to be another promising chemotype as evidenced by the reported affinity for CB2 versus CB1 (hCB2/hCB1 EC50 ¼ 25 nM/W2.0 mM) in the adenylate cyclase assay [43]. SAR studies revealed that the five-membered heterocycles had strict size requirements for their substituents, with only the tert-butyl group maintaining good potency
Recent Advances in the Discovery of CB2 Selective Agonists
237
and selectivity for the CB2 receptor. Compound 20 had low intrinsic clearance in both human and rat liver microsomes (HLM CLint ¼ 14 mL/ min/mg and RLM CLint ¼ 20 mL/min/mg) and good pharmacokinetic properties (AUC0inf ¼ 3570 ng h/mL, Cmax ¼ 971 ng/mL, %F ¼ 43 at 10 mg/kg, p.o.) in rats and was therefore selected for further in vivo evaluation [43].
3.2 Bicyclic core-based CB2 agonists 3.2.1 Benzimidazole cores
X
N N R1
S O O
N
N
N O O
21 R1 = 3-ethoxypyridin-4-yl, X = O 22 R1 = Et, X = O 23 R1 = 2,6-dichloropyridin-4-yl, X = O 24 R1 = Cyclopentyl, X = N-COCH3
25
Two distinct classes of benzimidazole-derived selective CB2 modulators have also emerged recently. Compound 21 exhibited potent agonist activity (hCB2 EC50 ¼ 0.3 nM) and high selectivity over the CB1 receptor (W4,200-fold). Furthermore, this compound displayed good pharmacokinetic properties in male SD rats (AUC0inf ¼ 2501 ng h/mL, Cmax ¼ 731 ng/mL, t1/2 ¼ 3.4 h, %F ¼ 43 at 10 mg/kg, p.o.). SAR studies revealed that the size of the substituent at the 2-position of the benzimidazole was a key determining factor of in vitro efficacy (Emax) and agonist potency. Of note, ethyl analog 22 is potent and selective (hCB2 EC50 ¼ 1.2 nM, CB1/CB2 B500), but induced the prototypical CB1mediated psychotropic effects when dosed in rats [44]. By contrast, 21, circumvented the CB1-mediated side effects due to limited blood–brain barrier (BBB) penetration (brain/plasma ¼ 0.1 in rat) and good selectivity. The major reported liabilities associated with 21 include poor solubility and inhibition of both CYP2C9 and CYP2C19 [44]. A closely related analog 23, wherein the ethoxy group was replaced with two chlorine substituents, displayed improved in vitro selectivity, up to W13,000-fold, over CB1 in the cyclase assay (hCB2/hCB1 EC50 ¼ 0.74
238
Sangdon Han et al.
nM/W10 mM) [45]. Replacement of the pyran with the N-acetyl piperidine afforded compound 24 which retained single nanomolar agonistic CB2 activity in the cyclase assay (hCB2 EC50 ¼ 3.23 nM). CB1related side effects of 24 were examined in a hypothermia experiment [lowest acceptable dose (LAD) W40 mg/kg] [46]. In the second benzimidazole-derived series [47], bis-alkyl amides in place of the sulfone were well tolerated, suggesting the presence of a large hydrophobic binding pocket at the receptor site. The mono-ethyl amide or primary amide resulted in loss of activity at CB2. The N-1benzimidazole position can accommodate a wide variety of groups, with a preference for alkyl substituents. However, selectivity versus CB1 diminished with increasing steric bulk. A representative compound, 25, had good in vitro potency and selectivity in the [3H]-CP55,940 binding assay (hCB2/hCB1 Ki ¼ 4.5 nM/W5.0 mM) [47] but was not desirable for in vivo evaluation due to poor metabolic stability, oral bioavailability, and solubility [48].
3.2.2 Imidazo bicyclic cores O
F
NH
F
O
N HN
N
O N
N
N
HN O
N
N O
N
N
N n
N
N
O F3C 26
27
28
29 n = 1
Imidazopyridine-based CB2 agonists have also been disclosed and assessed in both the FCA model of inflammatory pain and iodoacetate model of osteoarthritis [49]. Compound 26 (hCB2/hCB1 EC50 ¼ 5.2 nM/ W17 mM in cAMP) is representative of this series. The structurally related 27 was found to retain good potency for CB2 but was somewhat less selective over CB1 (hCB2/hCB1 EC50 ¼ 2.2 nM/878 nM) [49]. In a subsequent PCT filing by the same group, removal of carbonyl group from the 1 or 3 substituent at the imidazopyridine core was described. Compound 28 (hCB2 EC50 ¼ 17 nM in cAMP) likely afforded an improvement in solubility due to its basic nitrogen [50]. Insertion of a five-membered heterocyclic spacer (pyrazole, imidazole, etc.) between
239
Recent Advances in the Discovery of CB2 Selective Agonists
the core and the phenyl was also explored as part of the SAR in this application; however, this modification did not improve potency for CB2 [50]. A presumed water soluble CB2 series based on imidazo diazepine, 29 (n ¼ 1) and its imidazo piperazine (n ¼ 0) congener, is the subject of a separate disclosure [51]. Compound 29 (hCB2/hCB1 EC50 ¼ 10 B 100 nM/W10 mM in cAMP) exhibited analgesic activity in the murine acetic acid-induced writhing of model visceral pain (3, 10, and 30 mg/kg, s.c.), in the carrageenan model of acute inflammation in rats (30 mg/kg, s.c.), and exhibited an anti-allodynic response in the spinal nerve ligation model of neuropathic pain in rats (30 mg/kg, p.o.) [51].
3.2.3 Miscellaneous bicyclic cores
O
HN N
O
N
O
N
N S O O
N
N Cl
Cl 30
Cl
31
32
Various 6,6-fused heterocycles were reported as selective CB2 modulators. SAR studies of a series of quinolone-3-carboxamides have been described [52]. The nature of the 3-carboxamide substituent seems to be important for both receptor affinity and selectivity, since the highly lipophilic adamantylamide is the key feature for governing excellent in vitro profiles. Compound 30 exhibited W190-fold selectivity over CB1 in the [3H]CP-55,940 binding assay (hCB2/hCB1 Ki ¼ 6.3 nM/1220 nM). Moreover, this compound displayed significant analgesic effects in the formalin test of acute peripheral and inflammatory pain in mice (3.0 mg/ kg i.p.). Specificity was confirmed by full reversal of the anti-nociceptive response by AM630 (3.0 mg/kg i.p.), a selective CB2 antagonist [52]. A new chemical series based on tetrahydronaphthyridinone scaffolds were reported. CB2-selective compound 31 is representative of this series (hCB2/hCB1 EC50 o100 nM/W10 mM in cAMP assay). Replacement of the methyl sulfone with acetyl led to a loss of selectivity [53]. A novel series of aminoquinazolines were also disclosed as selective CB2 receptor modulators and are exemplified by 32. The successful
240
Sangdon Han et al.
replacement of the piperidine group with pyrrolidine, morpholine, thiomorpholine, or piperazine was demonstrated in this disclosure [54].
R1 HN
CN
HN
O
O
N N
O
NH S
O H
F
NH N H
Cl 33 R1 = CH3 34 R1 = CH2OH
H
O
H 35
36
Glenmark has reported a series of CB2 agonists based on a bridged bicyclic pyrazole core motif [55] and has initiated phase I clinical trials with GRC-10693 (structure not disclosed) for inflammatory and neuropathic pain [56]. GRC-10693 showed W80-fold selectivity (hCB2/hCB1 Ki ¼ 11.8 nM/985.2 nM) over CB1 in [3H]CP-55,940 binding studies in human CHO cells, and B20-fold selectivity [rCB2 (spleen) Ki ¼ 12.8 nM and rCB1 (brain) Ki ¼ 253.5 nM] in rat tissues. GRC-10693 had a good pharmacokinetic properties in SD rats (t½ ¼ 5.78 h, %F ¼ 48, 10 mg/kg, p.o.). GRC-10693 demonstrated good oral efficacy in the FCA model of inflammatory pain (ED50 ¼ 1.67 mg/kg) and also significantly reversed chronic construction injury in male SD rats (ED50 ¼ 2.15 mg/kg, p.o.) [57]. In a subsequent PCT filing by Glenmark, ethanol-amide 34 displayed a significant improvement in pharmacokinetic parameters (Cmax ¼ 1,949 ng/mL, AUC0inf ¼ 11,970 ng h/mL, t1/2 ¼ 4.8 h) relative to compound 33 (Cmax ¼ 131.3 ng/mL, AUC0inf ¼ 1,553 ng h/mL, t1/2 ¼ 7.3 h) [58]. Recently, a new chemical series composed of a mono or a bicyclic aromatic core with vicinal bis-amides has appeared. Thieno pyran 35 was reported to be a potent and selective CB2 modulator (hCB2/hCB1 EC50 ¼ 21 nM/W1.5 mM). Intraperitoneal administration of 35 demonstrated analgesia in the rat FCA-induced mechanical hyperalgesia, skin incision, and spinal nerve ligation pain models [59,60]. Since the initial discovery of an aminoalkylindole cannabinoid ligand, WIN-55,212-2 [61], the indole has been repeatedly reported as a CB2-biased moiety. Indole 2-indane amides have been described as a new class of CB2 receptor modulators [62]. Unlike previous indole series, the indole 2-carboxamide chemotype recognizes the CB2 receptor
241
Recent Advances in the Discovery of CB2 Selective Agonists
without an N-1 alkyl substituent (e.g., compound 36, hCB2 binding Ki ¼ 0.24 nM) [62].
3.3 Tricyclic core-based CB2 agonists
O
O
O N N N
O O
N S
O
37
38
O
O OO S N H N H
O O 39
Several groups have disclosed CB2 ligands composed of a conformationally restricted tricyclic core [63–65]. Sch35966 37 was reported to be a potent CB2 agonist for both human and monkey receptors (human CB2 Ki ¼ 6.872.3 nM, monkey CB2 Ki ¼ 5.470.4 nM) and for rodent receptors (rat CB2 Ki ¼ 2.470.5 nM, mouse CB2 Ki ¼ 4.871.6 nM). Furthermore, this compound exhibited W450-fold selectivity over the CB1 receptor [63]. A new class of CB2 agonists based on the tetrahydropyrrolo indole (e.g., 38) moiety was also identified and assessed in vitro [64]. SAR of the tetrahydropyran greatly impacted the CB2 receptor affinity and selectivity. Pyrrolo tetrahydropyran 38 was described as a potent and full CB2 agonist (hCB2 EC50 ¼ 1.7 nM, Emax ¼ 107% relative to WIN55212-2) as determined in the GTPg[35S] assay and possessed the best selectivity ratio (hCB2/hCB1 Ki ¼ 17.6 nM/6183 nM) in this series. Further biological evaluation of several examples from within this series is currently underway [64]. A PCT application describing the synthesis and SAR of a series of octahydrophenanthridine core-based CB2 modulators, exemplified by 39 (hCB2/hCB1 EC50 ¼ 36.7 nM/W10 mM in cAMP), has been reported [65]. In the mouse acetic-acid-induced writhing assay, 39 demonstrated % maximum possible effect of 41% at 10 mg/kg, s.c., along with reduction of body temperature change of 0.21C [65].
4. CLINICAL TRIALS STATUS Three pharmaceutical companies have reported entering clinical trials with CB2 agonists for treatment of pain. Pharmos tested their candidate
242
Sangdon Han et al.
Cannabinor [PRS-211375 (40), a 37-fold selective CB2 agonist] i.v. in a third molar dental extraction phase IIa trial. Results of this trial were confounding. The lowest 12-mg dose showed significant effect, whereas two higher doses of 24 and 48 mg did not achieve significance. Pharmos also ran a separate phase IIa study for induced-pain in healthy volunteers. The 48-mg dose of Cannabinor was not effective in reducing capsaicin-induced pain, but did show activity in mechanical and thermal hyperalgesia in normal skin [66]. Pharmos has discontinued future development efforts with this compound [67]. O
O OH
O O HO 40
GSK has completed a phase II trial for dental pain (third molar tooth extraction) and two phase II trials for osteoarthritis pain with their clinical candidate GW842166. The results of these trials have not been disclosed thus far. Glenmark pharmaceutical’s recent press release indicated successful completion of a phase I trial in Europe for their clinical candidate GRC10693 for neuropathic pain, osteoarthritis, and inflammatory pain disorders [68]. The structure of this candidate has not been disclosed.
5. CONCLUSIONS Novel pain therapeutic alternatives with minimal adverse side effects and abuse potential are highly desired by patients and healthcare professionals. The preclinical data that have emerged so far with CB2 agonists has been promising and suggestive that therapies directed at this target could fulfill this unmet therapeutic need. Numerous CB2 agonists have now shown efficacy in multiple animal pain models without apparent CB1-associated psychotropic effects. Whether this promise will be fulfilled in the clinic will become apparent as clinical trial data emerges.
Recent Advances in the Discovery of CB2 Selective Agonists
243
REFERENCES [1] G. T. Whiteside, G. P. Lee and K. J. Valenzano, Curr. Med. Chem., 2007, 14, 917. [2] D. M. Slipetz, G. P. O’Neill, L. Favreau, C. Dufresne, M. Gallant, Y. Gareau, D. Guay, M. Labelle and K. M. Metters, Mol. Pharmacol., 1995, 48, 352. [3] (a) M. Bouaboula, C. Poinot-Chazel, J. Marchand, X. Canat, B. Bourrie´, M. RinaldiCarmona, B. Calandra, G. Le Fur and P. Casellas, Eur. J. Biochem., 1996, 237, 704; (b) V. Di Marzo and L. De Petrocellis, Ann. Rev. Med., 2006, 57, 553. [4] (a) S. Munro, K. L. Thomas and M. Abushaar, Nature, 1993, 365, 61; (b) S. Galiegue, S. Mary, J. Marchand, D. Dussossoy, D. Carriere, P. Carayon, M. Bouaboula, D. Shire, G. Lefur and P. Casellas, Eur. J. Biochem., 1995, 232, 54. [5] E. Nunez, C. Benito, M. R. Pazos, A. Barbachano, O. Fajardo, S. Gonzalez, R. M. Tolon and J. Romero, Synapse, 2004, 53, 208. [6] M. D. Van Sickle, M. Duncan, P. J. Kingsley, A. Mouihate, P. Urbani, K. Mackie, N. Stella, A. Makriyannis, D. Piomelli, J. S. Davison, L. J. Marnett, V. Di Marzo, Q. J. Pittman, K. D. Patel and K. A. Sharkey, Science, 2005, 310, 329. [7] U. Anand, W. R. Otto, D. Sanchez-Herrera, P. Facer, Y. Yiangou, Y. Korchev, R. Birch, C. Benham, C. Bountra, I. P. Chessell and P. Anand, Pain, 2008, 138, 667. [8] G. Wotherspoon, A. Fox, P. McIntyre, S. Colley, S. Bevan and J. Winter, Neuroscience, 2005, 135, 235. [9] K. T. Valenzano, L. Tafesse, G. Lee, J. E. Harrison, J. M. Boulet, S. L. Gottshall, L. Mark, M. S. Peason, W. Miller, S. Shan, L. Rabadi, Y. Rotshteyn, S. M. Chaffer, P. I. Turchin, D. A. Elsemore, M. Toth, L. Koetzner and G. T. Whiteside, Neuropharmacology, 2005, 48, 658. [10] B. B. Yao, G. C. Hsieh, J. M. Frost, Y. Fan, T. R. Garrison, A. V. Daza, G. K. Grayson, C. Z. Zhu, M. Pai, P. Chandran, A. K. Salyers, E. J. Wensink, P. Honore, J. P. Sullivan, M. J. Dart and M. D. Meyer, Br. J. Pharmacol., 2008, 153, 390. [11] B. Bingham, P. G. Jones, A. J. Uveges, S. Kotnis, P. Lu, V. A. Smith, S. C. Sun, L. Resnick, M. Chlenov, Y. He, B. W. Strassle, T. A. Cummons, M. J. Piesla, J. E. Harrison, G. T. Whiteside and J. D. Kennedy, Br. J. Pharmacol., 2007, 151, 1061. [12] B. B. Yao, G. Hsieh, A. V. Daza, Y. Fan, G. K. Grayson, T. F. Garrisob, O. E. Kouhen, B. A. Hooker, M. Pai, E. J. Wensink, A. K. Salyers, P. Chandran, C. Z. Zhu, C. Zhong, K. Ryther, M. E. Gallagher, C. L .Chin, A. E. Tovcimak, V. P. Hradil, G. B. Fox, M. J. Dart, P. Honore and M. D. Meyer, J. Pharmacol. Exp. Therap., 2009, 328, 141. [13] M. Rinaldi-Carmona, F. Barth, J. Millan, J. M. Derocq, P. Casellas, C. Congy, D. Oustric, M. Sarran, M. Bouaboula, B. Calandra, M. Portier, D. Shire, J. C. Brelie`re and G. L. Fur, J. Pharmacol. Exp. Therap., 1998, 284, 644. [14] Y. Hosohata, R. M. Quock, K. Hosohata, A. Makriyannis, P. Consroe, W. R. Roeske and H. I. Yamamura, Eur. J. Pharmacol., 1997, 321(1), R1. [15] G. M. P. Giblin, C. T. O’shaughnessy, A. Naylor, W. L. Mitchell, A. J. Eatherton, B. P. Slingsby, D. A. Rawlings, P. Goldsmith, A. J. Brown, C. P. Haslam, N. M. Clayton, A. W. Wilson, I. P. Chessell, A. R. Wittington and R. Green, J. Med. Chem., 2007, 31, 2597. [16] M. Naguib, P. Diaz, J. J. Xu, F. Astruc-Diaz, S. Craig, P. Vivas-Mejia and D. L. Brown, Br. J. Pharmacol., 2008, 155, 1104. [17] A. Kikuchi, K. Ohashi, Y. Sugie, H. Sugimoto and H. Omura, J. Pharmacol. Sci., 2008, 106, 219. [18] H. Omura, M. Kawai, A. Shima, Y. Iwata, F. Ito, T. Masuda, A. Ohta, N. Makita, K. Omoto, H. Sugimoto, A. Kikuchi, H. Iwata and K. Ando, Bioorg. Med. Chem. Lett., 2008, 18, 3310. [19] K. Maresz, G. Pryce, E. D. Ponomarev, G. Marsicano, J. L. Croxford, L. P. Shriver, C. Ledent, X. Cheng, E. J. Carrier, M. K. Mann, G. Giovannoni, R. G. Pertwee, T.
244
[20]
[21]
[22] [23]
[24] [25] [26]
[27]
[28]
[29] [30] [31]
[32]
[33]
[34]
[35]
[36]
[37]
Sangdon Han et al.
Yamamura, N. E. Buckley, C. J. Hillard, B. Lutz, D. Baker and B. N. Dittel, Nat. Med., 2007, 13, 492. J. Palazuelos, N. Davoust, B. Julien, E. Hatterer, T. Aguado, R. Mechoulam, C. Benito, J. Romero, A. Silva, M. Guzman, S. Nataf and I. Galve-Roperh, J. Biol. Chem., 2008, 283, 13320. L. Hanus, A. Breuer, S. Tchilibon, S. Shiloah, D. Goldenberg, M. Horowitz, R. G. Pertwee, R. A. Ross, R. Mechoulam and E. Fride, Proc. Natl. Acad. Sci. U.S.A., 1999, 96, 14228. M. G. Belvisi, H. J. Patel, V. Freund-Michel, D. J. Hele, N. Crispino and M. A. Birrell, Br. J. Pharmacol., 2008, 155, 547. O. Ofek, M. Karsak, N. Leclerc, M. Fogel, B. Frenkel, K. Wright, J. Tam, M. AttarNamdar, V. Kram, E. Shohami, R. Mechoulam, A. Zimmer and I. Bab, Proc. Natl. Acad. Sci. U.S.A., 2006, 103, 696. S. Steffens, N. R. Veillard, C. Arnaud, G. Pelli, F. Burger, C. Staub, A. Zimmer, J. Frossard and F. Mach, Nature, 2005, 434, 782. E. K. Ho¨genauer, Exp. Opin. Ther. Patents, 2007, 17(12), 1457. P. Markt, C. Feldmann, J. M. Rollinger, S. Raduner, D. Schuster, J. Kirchmair, S. Distinto, G. M. Spitzer, G. Wolber, C. Laggner, K.-H. Altmann, T. Langer and J. Gertsch, J. Med. Chem., 2009, 52, 369. W. L. Mitchell, G. M. P. Giblin, A. Naylor, A. J. Eatherton, B. P. Slingsby, A. D. Rawlings, K. S. Jandu, C. P. Haslam, A. J. Brown, P. Goldsmith, N. M. Clayton, A. W. Wilson, I. P. Chessell, R. H. Green, A. R. Whittington and I. D. Wall, Bioorg. Med. Chem. Lett., 2009, 19, 259. R. Zindell, D. Reither, T. Bosanac, A. Berry, M. J. Gemkow, A. Ebneth, S. Lo¨bbe, E. L. Raymond, D. Thome, D.-T. Shih and D. Thomson, Bioorg. Med. Chem. Lett., 2009, 19, 1604. D. Reither, D. Thomson and R. Zindell, WO Patent Application 2008/48914, 2008. C. Leung, M. Tomaszewski and S. Woo, WO Patent Application 2004/060882, 2004. K. Worm, Q. J. Zhou, C. T. Saeui, R. C. Green, J. A. Cassel, G. J. Stabley, R. N. DeHaven, N. Conway-James, C. J. LaBuda, M. Koblish, P. J. Little and R. E. Dolle, Bioorg. Med. Chem. Lett., 2008, 18, 2830. M. Ermann, D. Riether, E. R. Walker, I. F. Mushi, J. E. Jenkins, B. Noya-Marino, M. L. Brewer, M. G. Taylor, P. Amouzegh, S. P. East, B. W. Dymock, M. J. Gemkow, A. F. Kahrs, A. Ebneth, S. Lo¨bbe, K. O’Shea, D.-T. Shih and D. Thomson, Bioorg. Med. Chem. Lett., 2008, 18, 1725. M. Emann, D. Riether, I. Mushi, J. E. Jenkins, E. Walker, B. Noya-Marino, M. Brewer, M. Taylor, P. Amouzegh, S. P. East, B. Dymock, M. Gemkow, S. Lo¨bbe, A. Ebneth, A. F. Kahrs, D. Thome, K. O’Shea, R. Dinallo, E. Raymond, D.-T. Shih and D. Thomson, Molecular Medicine Tri-Conference, San Francisco, CA, March 2008. A. J. Goodman, C. W. Ajello, K. Worm, B. L. Bourdonnec, M. A. Savolaninen, H. O’Hare, J. A. Cassel, G. J. Stabley, R. N. DeHaven, C. J. LaBuda, M. Koblish, P. J. Little, B. L. Brogdon, S. A. Smith and R. E. Dolle, Bioorg. Med. Chem. Lett., 2009, 19, 309. Y. Cheng, B. K. Albecht, J. Brown, J. L. Buchanan, W. H. Buckner, E. F. DiMauro, R. Emkey, R. T. Fremeau, J.-C. Harmange, B. J. Hoffman, L. Huang, M. Huang, J. H. Lee, F.-F. Lin, M. W. Martin, H. Q. Nguyen, V. F. Patel, S. A. Tomlinson, R. D. White, X. Xia and S. A. Hitchcock, J. Med. Chem., 2008, 51, 5019. E. F. DiMauro, J. L. Buchanan, A. Cheng, R. Emkey, S. A. Hitchcock, L. Huang, M. Y. Huang, B. Janosky, J. H. Lee, X. Li, M. W. Martin, S. A. Tomlinson, R. D. White, X. M. Zheng, V. F. Patel and R. T. Fremeau,, Bioorg. Med. Chem. Lett., 2008, 18, 4267. E. F. DiMauro, J. L. Buchanan and A. Cheng, XXth International Medicinal Chemistry Symposium, Vienna, Austria, September 2008.
Recent Advances in the Discovery of CB2 Selective Agonists
245
[38] H. Ohta, T. Ishizaka, M. Yoshinaga, A. Morita, Y. Tomishima, Y. Toda and S. Saito, Bioorg. Med. Chem. Lett., 2007, 17, 5133. [39] H. Ohta, T. Ishizaka, M. Tatsuzuki, M. Yoshinaga, I. Iida, Y. Tomishima, Y. Toda and S. Saito, Bioorg. Med. Chem. Lett., 2007, 17, 6299. [40] H. Ohta, T. Ishizaka, M. Tatsuzuki, M. Yoshinaga, I. Iida, T. Yamaguchi, Y. Tomishima, N. Futaki, Y. Toda and S. Saito, Bioorg. Med. Chem., 2008, 16, 1111. [41] P. F. Cirillo, D. S. Thomson, D. Reither, A. Berry, L. Wu and E. R. Hickey, WO Patent Application 2008/064054, 2008. [42] A. Berry, P. F. Cirillo, E. R. Hickey, D. Reither, D. Thomson, R. M. Zindell, M. Ermann, J. E. Jenkins, I. Mushi, M. Taylor, C. Chowdhury, C. F. Palmer and N. Blumire, WO Patent Application 2008/039645, 2008. [43] I. E. Marx, E. F. DiMauro, A. Cheng, R. Emkey, S. A. Hitchcock, L. Huang, M. Y. Huang, J. Human, J. H. Lee, X. Li, M. W. Martin, R. D. White, R. T. Fremeau, Jr. and V. F. Patel, Bioorg. Med. Chem. Lett., 2009, 19, 31. [44] B. M. P. Verbist, M. A. Cleyn, M. Surkyn, E. Fraiponts, J. Aerssens, M. J. M. A. Nijsen and H. J. M. Gijsen, Bioorg. Med. Chem. Lett., 2008, 18, 2574. [45] H. J. M. Gijsen, M. A. J. De Cleyn, M. Surkyn and B. M. P. Verbist, WO Patent Application 2008/003665, 2008. [46] H. J. M. Gijsen, M. A. J. De Cleyn and M. Surkyn, WO Patent Application 2008/119694, 2008. [47] D. Page, E. Balaux, L. Boisvert, Z. Liu, C. Milburn, M. Tremblay, Z. Wei, S. Woo, X. Luo, Y.-X. Cheng, H. Yang, S. Srivastava, F. Zhou, W. Brown, M. Tomaszewski, C. Walpole, L. Hodzic, S. St-Onge, C. Godbout, D. Salois and K. Payza, Bioorg. Med. Chem. Lett., 2008, 18, 3695. [48] Z. Liu, H. Yang, J. Ducharme, D. Page, Z.-Y. Wei, M. Tremblay, S. Srivastava, R. Dolaine, C. J. Milburn, E. Lessard, P. E. Morin, S. St-Onge, K. Payza and C. Walpole, Abs 240, The 31st National Medicinal Chemistry Symposium, Pittsburgh, PA, June 2008. [49] M. T. Bilodeau, C. S. Burgey, Z. J. Deng, J. C. Hartnett, N. R. Kett, J. Melamed, P. M. Munson, K. K. Nanda, W. Thomson, B. W. Trotter and Z. Wu, WO Patent Application 2008/085302, 2008. [50] Z. Wu, A. I. Green and J. C. Hartnett, WO Patent Application 2009/025785, 2009. [51] R. P. Beckett, R. Foster, C. Henault, J. L. Ralbovsky, C. M. Gauss, G. R. Gustafson, Z. Luo, A.-M. Campbell and T. E. Shelekhin, WO Patent Application 2008/157751, 2008. [52] S. Pasquini, L. Botta, T. Semeraro, C. Mugnaini, A. Ligresti, E. Palazzo, S. Mainoe, V. Marzo and F. Corelli, J. Med. Chem., 2008, 51, 5075. [53] G. R. Gustafson and R. P. Beckett, WO Patent Application 2008/079316, 2008. [54] T. C. Gahman, D. J. Thomas, H. Lang and M. E. Massari, WO Patent Application 2008/ 157500, 2008. [55] M. Muthuppalanippan, G. Balasubramanian, S. Gullapalli, N. Khairatkar-Joshi and S. Narayanan, WO Patent Application 2006/129178, 2006. [56] http://www.glenmarkpharma.com/media/pdf/releases/glenmark_molecule_GRC_ 10693.pdf [57] S. Narayanan, M. Muthuppalanippan, A. Thomas, N. Khairatkar-Joshi, K. Varanasi, S. Gullapalli and S. K. V. S. Vakkalanka, Society for Neuroscience, Atlanta, Georgia, Oct. 14, 2006. [58] M. Muthuppalanippan, K. Sukeerthi, G. Balasubramanian, S. Gullapalli, N. KhairatkarJoshi, S. Narayanan and P. V. Karnik, WO Patent Application 2008/053341, 2008. [59] W. A. Carroll, D. W. Nelson and A. Perez-Medrano, WO Patent Application 2009/ 009550, 2009. [60] D. W. Nelson, M. E. Gallagher, K. Ryther, Abs 220, The 31st National Medicinal Chemistry symposium, Pittsburgh, PA, June 2008.
246
Sangdon Han et al.
[61] D. R. Compton, L. H. Gold, S. J. Ward, R. L. Balster and B. R. Martin, J. Pharmacol. Exp. Therap, 1992, 263(3), 1118. [62] C. Liu, S. T. Wrobleski, K. Leftheris, G. Wu, P. M. Sher and B. A. Ellsworth, WO Patent Application 2009/015169, 2009. [63] W. Gonsiorek, C. A. Lunn, X. Fan, G. Deno, J. Kozlowski and R. W. Hipkin, Br. J. Pharmacol., 2007, 151, 1261. [64] D. Page, H. Yang, W. Brown, C. Walpole, M. Fleurent, M. Fyfe, F. Gaudreault and S. St-Onge, Bioorg. Med. Chem. Lett., 2007, 17, 6183. [65] J. L. Ralbovsky and P. R. Beckett, WO Patent Application 2008/109007, 2008. [66] http://www.pharmoscorp.com/news/pr/pr042407.html: Company press release. [67] http://integrity.prous.com/integrity/servlet/xmlxsl/pkpipeline.xmlProductMielstone? pentryNumber ¼ 352376 [68] http://www.glenmarkpharma.com/media/pdf/releases/Glenmarks_molecule_ neuropathicpain_osteoarthritis_GRC10693_succes.pdf. Company press release.
CHAPT ER
12 Advances in the Discovery of Small Molecule JAK3 Inhibitors Stephen T. Wrobleski and William J. Pitts
Contents
1. Introduction 2. Rationale for Selective Targeting of JAK3 in Inflammatory Diseases 3. Clinical Trials and Supporting Preclinical Data 3.1 Clinical candidates 3.2 Organ transplantation 3.3 Rheumatoid arthritis 3.4 Psoriasis 4. Challenges in Designing Selective JAK3 Inhibitors 5. Recent Medicinal Chemistry Efforts 5.1 Pyrrolopyrimidines 5.2 Pyrrolopyridines 5.3 Purines and purinones 5.4 Monocyclic pyridines and pyrimidines 5.5 Other heterocycles 6. Conclusions References
247 248 249 250 251 252 253 253 255 255 257 258 259 260 261 261
1. INTRODUCTION JAnus Kinase 3 (JAK3) is a member of the JAK family of non-receptor protein tyrosine kinases (PTKs) that include the closely related isoforms JAK1, JAK2 and tyrosine kinase 2 (TYK2). In the early 1990s, TYK2 was identified as the prototypical member of this new kinase family [1]. Bristol-Myers Squibb R&D, Route 206 & Province Line Rd., Princeton, NJ 08543, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04412-1
r 2009 Elsevier Inc. All rights reserved.
247
248
Stephen T. Wrobleski and William J. Pitts
Subsequent to this initial discovery, a second group of researchers independently identified two additional family members, which they noted contained two adjacent kinase domains: the catalytically active JH1 domain and a non-typical, pseudokinase domain JH2 that appears to be catalytically inactive [2]. The presence of the dual kinase domains led the researchers to name these kinases as JAK1 and JAK2 after the two-faced Roman god, Janus. The function of these kinases was not known at the time of their discovery, and JAK was also referred to as ‘Just Another Kinase’. Ironically, it was subsequently discovered that these kinases played a critical role in the signal transduction of the cytokine receptor superfamily [3]. Once the importance of this kinase family became known, a third group of researchers used a polymerase chain reaction (PCR) strategy to identify JAK3 as the last family member [4]. Since inhibition of the JAK kinases would be expected to block cytokine signaling, members of this family have become attractive targets for small-molecule drug discovery [5]. The realization that human defects in JAK3 signaling result in the clinical manifestation of a severe combined immunodeficiency (SCID) phenotype has suggested that selective JAK3 inhibitors may be useful as therapeutic agents in the areas of organ transplantation and autoimmune diseases.
2. RATIONALE FOR SELECTIVE TARGETING OF JAK3 IN INFLAMMATORY DISEASES The JAK signal transducer and activator of transcription (STAT) signaling pathway has been extensively studied but specific details regarding cascade regulation remain unelucidated [6]. In cells, JAK3 associates with cytokine receptors, which homodimerize and heterodimerize upon binding of the ligand. JAK3 binds specifically to the gamma common chain (gc) that is shared by the cytokine receptors for interleukin-2 (IL-2), IL-4, IL-7, IL-9, IL-15 and IL-21 [5]. Upon ligand binding, JAK3 becomes activated in concert with JAK1 through transphosphorylation which in turn results in phosphorylation of the receptor. Receptor phosphorylation then results in binding and phosphorylation of an associated STAT protein (e.g., STAT5 in the case of IL-2 stimulation and STAT6 in the case of IL-4 stimulation). After STAT dimerization, the complex is thought to translocate to the nucleus, interact with DNA, and initiate transcription of target genes (Figure 1). Because JAK3 plays a specific role in regulating gc cytokine signaling and is primarily expressed in lymphoid tissues, it appears to be a selective regulator of lymphoid development and function within the immune system [7]. This is consistent with the SCID phenotype that has been identified in a subpopulation of patients that harbor abnormalities associated with JAK3 based on genetic analysis [8]. These abnormalities
Advances in the Discovery of Small Molecule JAK3 Inhibitors
249
IL-2, IL-4, IL-7, IL-9, IL-15, IL-21
γc
JAK1
JAK3
STAT
STAT STAT
STAT
P
P Cytoplasm P
Nucleus
STAT STAT STAT STAT P DNA
Gene transcription
Figure 1 Schematic of gamma common chain cytokine receptor and JAK3 signaling pathway.
include a ‘kinase dead’ mutant, a mutant incapable of binding the gc of the IL-2 family of receptors, or a lack of detectable JAK3 protein, each of which is associated with a human SCID phenotype [9]. This disorder is characterized by a significant decrease in the number of circulating T and NK cells with normal numbers of B cells (albeit with compromised B-cell function). Hematopoetic stem cell transplantation results in the normalization of the T-cell population in these patients [10]. Since patients with defects in JAK3 show symptoms that are restricted to the immune system, it follows that a selective inhibitor of JAK3 could function as a well-tolerated immunosuppressant for use in a number of autoimmune disorders [11]. In comparison with JAK3, the other JAK family members, JAK1, JAK2 and TYK2, are known to be more ubiquitous in their expression patterns. In addition to participating in concert with JAK3 in the signaling of the IL-2 family of cytokine receptors, JAK1 has been shown to be a regulator of IL-6 and gp130 cytokine signaling. It has therefore been suggested that dual inhibition of JAK1 and JAK3 might lead to enhanced cellular potency and broader immunosuppressive effects, but may also increase the risk of viral and bacterial infections [12]. Moreover, JAK1 has been linked to tumor surveillance [13], and JAK1 knockout mice do not thrive, as the pups fail to nurse because of presumed neurological defects [14]. Despite these potential concerns, no adverse clinical outcome has been conclusively linked to JAK1 inhibition
250
Stephen T. Wrobleski and William J. Pitts
(vide infra). These observations make it difficult to assess the risk to benefit ratio of dual inhibition of JAK1 and JAK3 a priori. JAK2 is classically associated with interferon-g (IFN-g) production through the IL-12 pathway; however, it also mediates the signaling of important hematopoietic growth factors such as erythropoietin (EPO), thrombopoietin (TPO) and granulocyte macrophage colony-stimulating factor (GM-CSF). As a result, JAK2 inhibitors are being examined for oncological applications, and this is beyond the scope of this review. It has been suggested that JAK2 inhibition may result in adverse hematopoietic effects such as anemia, thrombocytopenia and generalized leukopenia in the clinic [12].
3. CLINICAL TRIALS AND SUPPORTING PRECLINICAL DATA 3.1 Clinical candidates
N
N
CN O
N N
N H 1
The clinical candidate CP-690,550 (1) is one of the most extensively studied JAK3 inhibitor to date. A review on CP-690,550 has recently been published [15]. This compound was originally reported to be a selective JAK3 inhibitor within the JAK family (JAK3 IC50 ¼ 1 nM, JAK2 IC50 ¼ 20 nM, JAK1 IC50 ¼ 112 nM) with generally high selectivity against other tested kinases [16]. However, a recent disclosure based on Ki values has indicated that CP-690,550 is less selective within the JAK family than originally reported (JAK3 Ki ¼ 0.2 nM, JAK2 Ki ¼ 1.0 nM, JAK1 Ki ¼ 0.7 nM, Tyk2 Ki ¼ 4.4 nM) [17]. Other researchers have also reported a lower degree of JAK3 selectivity versus JAK2 and JAK1 [18,19]. However, a high degree of selectivity against greater than 300 non-JAK family kinases has been confirmed [20]. Despite its potent activity against the JAK2 enzyme, CP-690,550 shows modest selectivity for inhibition of JAK1/3 signaling pathways compared to JAK2 in cellular assays. CP-690,550 in human whole blood stimulated with anti-CD3/CD28 in the presence of IL-2 gave potent inhibition of the JAK1/3 pathway (IC50 ¼ 34 nM) and was B15-fold more selective relative to the inhibition
Advances in the Discovery of Small Molecule JAK3 Inhibitors
251
of the JAK2 pathway as measured by IL-12 stimulation in human whole blood (IC50 ¼ 501 nM) [17]. In separate assays using fluorescenceactivated cell sorting (FACS) analysis in human whole blood, the inhibition of IL-15 activation of pSTAT5 in CD8+ T cells (JAK1/3 pathway) relative to GM-CSF activation of pSTAT3 in CD14+ monocytes (JAK2 pathway) was assessed. In these assays, CP-690,550 was determined to be B24-fold selective for JAK1/3 relative to JAK2 (IC50 ¼ 56 and 1377 nM, respectively) [17]. To date, phase II clinical studies in rheumatoid arthritis with CP-690,550 have been completed, and patients for phase III studies are being actively recruited. Phase II studies in kidney transplant have also been completed with additional phase II studies recruiting patients in psoriasis (oral and topical administration), Crohn’s disease and dry-eye syndrome (topical) [21]. Two additional JAK3 inhibitors, R-348 and VX-509 (structures undisclosed), have been reported to have entered clinical trials. R-348 has entered phase I trials for rheumatoid arthritis, psoriasis and other immunerelated disorders [22]. A subsequent report disclosed that R-348 acts as a prodrug that is converted to an active metabolite, R-333. The latter inhibits both the JAK1/3 and Syk pathways in cellular assays (IC50 ¼ 180 and 140 nM, respectively) [23]. The JAK3 and Syk enzyme IC50 values for R-348 or R-333 have not been reported. The JAK3 inhibitor VX-509 has recently completed phase I studies in healthy volunteers with phase II studies in rheumatoid arthritis expected to begin in the second half of 2009 [24]. The potency and selectivity profile of VX-509 has not been reported to date. CN
N N
N N H
N 2
In addition to these JAK3 inhibitors, a reported JAK1/JAK2 inhibitor, INCB18424 (2), is being studied for oncological indications and in rheumatoid arthritis and psoriasis clinical trials [25]. This compound has been reported to give potent inhibition of both JAK1 and JAK2 with selectivity versus JAK3 and TYK2 (IC50 values for JAK1, JAK2, JAK3 and TYK2 are 2.7, 4.5, 332 and 19 nM, respectively) [26]. The paradox of a
252
Stephen T. Wrobleski and William J. Pitts
JAK1/2 inhibitor with selectivity versus JAK3 demonstrating efficacy in autoimmune disease may be explained by the fact that all receptors which use JAK3 also use JAK1 for signaling [27]. INCB28050 (structure undisclosed) has also been reported as a potent JAK1/JAK2 inhibitor in phase I clinical trails as a backup clinical compound to INCB18424 [28].
3.2 Organ transplantation As discussed earlier, patients with deletions or mutations in JAK3 are severely immunocompromised. As a result, the prevention of organ rejection is a logical indication for JAK3 inhibitors. CP-690,550 monotherapy has been reported to prevent kidney allograft rejection in nonhuman primates [29] and decrease T cell and NK cell populations. In combination with mycophenolate mofetil [30], significant improvement in allograph survival was noted. CP-690,550 has also been shown to prevent allograph vasculopathy in a rodent model of aortic transplantation [31]. CP-690,550 has been examined in a phase I trial (dosed at 5, 15 and 30 mg BID for 28 days) in a population of stable renal allograft recipients maintained on mycophenolate mofetil and, in the case of the 5 and 15 mg dose, a calcinurin inhibitor (cyclosporin or tacrolimus). In this study CP-690,550 demonstrated an acceptable safety and tolerability profile. The most frequent adverse events were related to infections and GI tolerability [32]. Slight decreases in hemoglobin and reticulocyte counts (reversible) at the higher doses were also noted. Examination of the immune system of these patients showed a decrease in the number of NK cells and T-reg cells (other T-cell subpopulations appeared unchanged), an increase in the number of B cells and a decrease in IL-2stimulated IFN-g production [33]. The JAK3/Syk inhibitor R-348 has also been reported to be effective in a rat cardiac allograft model [23].
3.3 Rheumatoid arthritis CP-690,550 has been shown to be effective at reducing clinical scores and related histological inflammatory changes when administered by minipump with an ED50 of B1.5 mg/kg/day in both a mouse collagen– induced arthritis model and a rat adjuvant–induced arthritis model [34]. Analysis of dendritic cells from the synovial tissue of patients with rheumatoid arthritis showed high levels of expression of JAK3, STAT4 and STAT6 [35]. CP-690,550 was evaluated in a randomized placebocontrolled phase IIa study of patients with moderate to severe active rheumatoid arthritis at doses of 5, 15 and 30 mg BID. At week 6, 13–28% of patients achieved an ACR70 score compared to placebo (3%), 33–54% of patents achieved an ACR50 score compared to placebo (6%), and 70–81% of patients achieved an ACR20 score compared to placebo (29%)
Advances in the Discovery of Small Molecule JAK3 Inhibitors
253
[36]. In a separate study, patients (n ¼ 509, 80% women) with active disease while maintained on a background of methotrexate were randomized to placebo, or CP,690-550 dosed at 1, 3, 5, 10, 15 mg, all BID, or 20 mg QD [37]. At week 12, doses of 3 mg BID or higher were efficacious as measured by ACR20, ACR50 and ACR70 scores. The most frequent adverse events were nausea (2.4%), headache (2.2%) and increase in alanine aminotransferase (ALT) levels (2.0%). Five serious infections were reported with no apparent dose-related pattern. Other observations included minor dose-related decreases in hemoglobin, and dose-dependent increases in low-density lipoprotein (LDL), high-density lipoprotein (HDL) and total cholesterol. The JAK1/JAK2 inhibitor, INCB18424, has been reported to demonstrate preliminary efficacy in a small clinical trial (n ¼ 41) of rheumatoid arthritis. ACR 20, 50 or 70 scores were achieved in up to 83, 50 or 30% of patients when dosed at 15, 25 or 50 mg BID for 28 days, respectively [38].
3.4 Psoriasis CP-690,550 has been reported to produce dose-dependent improvements in mPASI scores (28 at 5 mg BID, 52 at 30 mg BID) when dosed for 14 days in patients with psoriatic lesions [39]. A decrease in keratin 16 expression, as measured by immunohistochemistry, was observed in three of four skin biopsies in the 30 mg BID group. This was consistent with a reversal of hyperplasia and other disease-associated pathology. INCB18424 has been studied as a topical agent (1.5% cream applied BID) in patients with psoriatic lesions and was shown to produce a similar level of efficacy to topical steroids [40].
4. CHALLENGES IN DESIGNING SELECTIVE JAK3 INHIBITORS Selective inhibition of specific isoforms within a kinase family represents a formidable challenge in drug discovery since small-molecule kinase inhibitors most commonly target the highly conserved adenosine triphosphate (ATP)-binding domain. The JAK kinase family consists of four members, JAK1, JAK2, JAK3 and TYK2, which contain B1,100 amino acids that constitute the seven homology domains JH1–JH7. The JH1 domain (kinase domain) is highly conserved across the family, with JAK2 having the highest sequence identity compared to JAK3 (62% homologous) closely followed by JAK1 (52%) and TYK2 (50%). X-ray crystallographic structures of the kinase domain of JAK1, JAK2 and JAK3 in complex with inhibitors have been reported [18,41,42]. These structures have elucidated the JAK1 and JAK2 binding modes of the clinical candidate CP-690,550 (1), the JAK3 binding mode of the staurosporine analog AFN941 (3) and the JAK2 binding mode of the
254
Stephen T. Wrobleski and William J. Pitts
Glycine Loop
Glycine Loop H3C
CN
Glycine Loop
NH
O
Me Me
H3C
O N
CH3
O
N
H HN
Me
N
N N Me N
N
N H O
H
N
1 H
N Hinge region
3 O
H N Hinge region
F
N
O
O
H
4 O
H N
Hinge region
Figure 2 Schematic of compounds 1, 3 and 4 illustrating key hydrogen bonds to hinge region in a generic JAK family active site.
tetracyclic pyridone 4 (Figure 2). In all cases, the inhibitors bind within the ATP-binding site and form dual H-bonds to the hinge region of the protein. In addition, the N-terminal portion of the activation loop that contains the highly conserved Asp-Phe-Gly (DFG) motif adopts the ‘inward’ conformation characteristic of active kinases. A patent application disclosing a crystal structure of a JAK3 kinase domain complexed with adenylyl-imidodiphosphate (AMP-PNP) has also been published [43]. At present, there are no reported crystal structures for TYK2 or any fulllength JAK family member. On the basis of the JAK3 X-ray structure of 3, a sequence alignment analysis of the active site residues that are in proximity to the complexed inhibitor revealed only two differences that the authors believed could be exploited to provide selectivity between JAK3 and its most closely related isoform JAK2 [41]. They speculated that JAK2-selective compounds could be designed to exploit the extra space afforded by the difference between an active site glycine (JAK2) and Ala966 (JAK3). A subsequent report disclosing the JAK2 structure with 4 also revealed that the JAK2 glycine carbonyl is in a flipped conformation relative to that found in all other tyrosine kinase structures including the JAK3 structure with 3 [42]. In addition, exploiting the residue difference between Cys909 (JAK3) and an analogous serine residue in JAK2 has been suggested as a potential strategy for obtaining JAK3 selectivity over JAK2. Several JAK3 inhibitors have been reported, which might interact with Cys909.
255
Advances in the Discovery of Small Molecule JAK3 Inhibitors
Mannich base NC1153, and has been shown to be a selective inhibitor of JAK3 capable of preventing allograft rejection in a rodent transplantation model [44]. A patent has disclosed irreversible Bruton’s tyrosine kinase (BTK) inhibitors that react with the free thiol of an active site cysteine and are relatively selective [45]. Compound 6 inhibited both BTK and JAK3, whereas 7 did not inhibit JAK3. JAK2 inhibition data for these compounds were not disclosed. OPh HCl Me
Me
Me
N
N
NH2
Me
5
O
N
HCl
O
O
N H
N
N N
N
6 BTK IC50 = 0.5 nM JAK3 IC50 = 10.4 nM
7 BTK IC50 = 1.0 nM JAK3 IC50 > 10,000 nM
A recent comparison of the binding modes of 4 and CP-690,550 in JAK1 and JAK2 shows the protein structures to be very similar [18]. As a result the authors suggest that designing highly selective inhibitors for a specific JAK family member may represent a challenge. Three JAK1 residues Phe958, Arg879 and His885 differ when compared to JAK2, although these residues are in proximity to the active site. The authors also postulate that subtle variations in the electrostatic potential differences in the active site might also be useful to consider in the design of more selective compounds. Finally, a recent report using a JAK1 homology model has suggested that two residue differences between JAK3 and JAK1 [Glu(JAK1)-Asp(JAK3) and Phe(JAK1)-Tyr(JAK3)] may be important for obtaining selectivity between JAK3 and JAK1 [46].
5. RECENT MEDICINAL CHEMISTRY EFFORTS 5.1 Pyrrolopyrimidines The pyrrolopyrimidine ring system that is contained in the two clinical candidates, CP-690,550 and INCB18424, has been the most useful for generating potent JAK family inhibitors. Because of the potent activity and promising clinical efficacy reported for CP-690,550, knowledge of the SAR that was generated in its development would be of interest in medicinal chemistry efforts at optimizing JAK3 inhibition. Extensive disclosure of this SAR has not appeared to date, but some of the SAR of the chiral piperidine residue has been reported. This includes a recent patent application that provided JAK3 IC50 values for CP-690,550 and 30
256
Stephen T. Wrobleski and William J. Pitts
analogs using a glutathione S-transferase (GST)-tagged JH1 kinase domain [47]. CP-690,550 was the most potent analog with a JAK3 IC50 of 2.4 nM. Replacement of the cyanoacetyl group in CP-690,550 with the ethyl and propyl sulfonyl groups gave the less potent analogs 8 and 9 (JAK3 IC50 ¼ 14.3 and 17.3 nM, respectively). Importantly, preference for the 3R, 4R absolute configuration on the piperidine ring was illustrated by the loss in potency for the corresponding cis, racemic analogs 10–12 (JAK3 IC50 of 3.4, 137 and 59.5 nM, respectively). A further reduction in potency was observed for the C-4 des-methyl racemates 13 and 14 (IC50 ¼ 625 and 459 nM, respectively), and this suggests that the methyl group makes a significant contribution to JAK3 potency in this series. The des-methyl analog of CP-690,550 was not reported. The remaining analogs were significantly less potent than the enantiopure analogs 1, 8 and 9 and were mainly derivatives with modified piperidine N-substitutions. enantiopure
cis, racemates NH
NH 3R
R N
NH
N
N 4R
C-4 des-methyl, racemates
N
N
R N
1, R = -C(O)CH2CN 8, R = -SO2Et 9, R = -SO2CH2CH2CH3
N N
N
R N 4
10, R = -C(O)CH2CN 11, R = -SO2Et 12, R = -SO2CH2CH2CH3
N
N
13, R = -SO2Et 14, R = -SO2CH2CH2CH3
NH O NC
3
N
N 4
N
N
15, 3(S),4(S)-isomer 16, 3(R),4(S)-isomer 17, 3(S),4(R)-isomer
A recent report also confirmed the preference for the 3R, 4R configuration in 1 by independently synthesizing and testing the other three possible stereoisomers 15–17 (Ambit JAK3 Kd values were 0.7, 190, 180 and 150 nM, respectively) [48]. Additional Kd values for 1 indicated potent affinity for JAK2 (2 nM) and JAK1 (3 nM) and less for TYK2 (250 nM). These Kd values are more consistent with the Ki values that had been recently reported [17] and are significantly more potent than the JAK2 IC50 of 20 nM and the JAK1 IC50 of 112 nM that were originally
257
Advances in the Discovery of Small Molecule JAK3 Inhibitors
reported [16] for 1. Interestingly, despite its potent affinity for the JAK2 enzyme, 1 was shown to be ineffective at inhibiting JAK2 or TYK2mediated STAT4 phosphorylation in IL-12-stimulated human CD4+ T cells at concentrations up to 500 nM. In comparison, 1 was shown to be effective at significantly inhibiting JAK3/JAK1-mediated phosphorylation of STAT5 in IL-2-stimulated CD4+ cells at 10-fold lower concentrations (50 nM). On the basis of these results, the authors speculate that 1 may be capable of selectively inhibiting JAK3 without disrupting the functions of JAK2 or TYK2 in a cellular environment at the concentrations tested. These results are consistent with the previous studies of 1 in IL-2-stimulated versus IL-12-stimulated human whole blood, which showed IC50 of 34 and 501 nM, respectively, indicating B15-fold selectivity with respect to the inhibition of the JAK3/JAK1 relative to the JAK2/Tyk2 pathway [17]. As previously mentioned, the binding mode of 1 in the active site of JAK1 and JAK2 has been determined by recent X-ray crystallography studies [18]. Similar binding modes were observed, and this is consistent with the nearly equipotent activity of 1 that has been reported recently. A conformational analysis of 1 and molecular docking studies using the available crystal structures of JAK3 and JAK2 revealed similar binding modes for both JAK3 and JAK2 consistent with their similar Ki values against these enzymes. In the case of JAK3, the proposed model suggests the pyrrolopyrimidine core as the hinge-binding element forming key hydrogen bonding interactions between the N3 and N9 nitrogen atoms and residues Glu903 and Leu905. Furthermore, the chiral piperidine was proposed to bind within the phosphate-binding region with the cyano substituent forming a H-bond interaction with Arg953 of the activation loop. Other JAK3 inhibitors containing a pyrimidine-based 5,6-ring system have recently been reported in the patent literature. This includes substituted pyrrolopyrimidines such as 18 that have been claimed to be JAK and CDK inhibitors with particular embodiments around inhibiting JAK3 and CDK4 [49]. Pyrrolopyrimidines containing mainly saturated monocyclic and bicyclic amines linked to the pyrrolopyrimidine core as in 19 have also been claimed as JAK3 inhibitors [50].
O
N
CH3O N
N N H 18
N
O N
N H
F
N H 19
N
N H
258
Stephen T. Wrobleski and William J. Pitts
5.2 Pyrrolopyridines Analogs based on a pyrrolopyridine ring system have also been claimed as potent JAK3 inhibitors. Pyrrolopyridine 20 functionalized with a chiral 1-amino-2-methyl cyclohexane moiety was disclosed as having a JAK3 IC50 ¼ 3 nM [51]. A closely related tricyclic variant 21 with a dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2(1H)-one ring system was also reported to have similar potency (JAK3 IC50 ¼ 3 nM). Its ethyl analog, 22, still maintained respectable JAK3 potency (JAK3 IC50 ¼ 5.1 nM). Replacement of the chiral cyclohexane residue with other pharmacophores provided analogs with improved JAK3 inhibition, for example, the adamantyl analog 23 (JAK3 IC50 ¼ 0.65 nM) and the piperidine analog 24 (JAK3 IC50 ¼ 0.3 nM) [52,53]. R O HN
O
N HN
H2N N
N H
N H
N
21, R = Me 22, R = Et
20
CN F
N
O
O N
O
HN
CN N H
N 23
N
HN
H2N N
N H
24
5.3 Purines and purinones A purine ring system was utilized as a scaffold to generate JAK3 inhibitors such as compounds 25–27 [54]. They were claimed to have oral
259
Advances in the Discovery of Small Molecule JAK3 Inhibitors
activity in a mouse delayed-type hypersensitivity (DTH) model. A series of substituted 2-benzimidazoyl purine and purinone derivatives has also been recently disclosed as a new class of JAK3 inhibitors [55–58]. This structural class is unique among reported JAK3 inhibitors since the benzimidazoyl group can be recognized as the hinge binding motif [59] rather than the purine core. PS020613 (28) has been reported to be comparable in JAK3 potency and selectivity to CP-690,550 and has been shown to be orally active in a mouse DTH model [60]. Additional analogs closely related to 28, including 29–31, were also disclosed. In general, N-substitution at the 8-position of the purine ring, as in the case of 29, was found to improve JAK3 selectivity with respect to Aurora A (58-fold) [58]. Modification of the fluorine substitution pattern on the chromene moiety and replacement of the fluorine substituent on the benzimidazole ring with a cyano group gave 30, which displayed a further enhancement in JAK3 selectivity versus Aurora A (73-fold). An N,N-dimethylamino ethylene group at the 8-position of the purinone ring gave 31, which had a JAK3 selectivity versus Aurora A of 724-fold. R′
N
R′′ F
F
O
N
O
N
O N
N N H
N
N H
25, R′ = Me, R′′ = Pr 26, R′ = Et, R′′ = Pr 27, R′ = R′′ = Et
N N
N
N N
F
N
N
N
O N
N R
28, R = H 29, R = Me
NC
N
F O
N R
30, R = H 31, R = CH2CH2N(CH3)2
5.4 Monocyclic pyridines and pyrimidines In addition to bicyclic and tricyclic chemotypes, JAK3 inhibitors with monocyclic pyridine or pyrimidine cores such as 32 and 33 have been claimed [61,62]. They carry a potentially reactive acrylamide group like those used in irreversible, covalent inhibitors of BTK and HER2. JAK3 and these kinases contain an active site cysteine residue with a nucleophilic thiol functionality that can react with acrylamides to form a covalent attachment that irreversibly inhibits the enzyme. As previously mentioned, irreversible inhibition of JAK3 by targeting this cysteine residue has been postulated as a viable strategy for designing inhibitors that are selective against JAK3 since the JAK1 and JAK2 isoforms contain
260
Stephen T. Wrobleski and William J. Pitts
a much less nucleophilic serine residue at this location. Compounds 32 and 33 are claimed as selective JAK3 inhibitors that are at least 10-fold more potent against JAK3 relative to JAK1 and JAK2. A series of 2,4-diaminopyrimidines has also been claimed as selective inhibitors of the JAK pathway relative to the Syk pathway and as being effective in treating immune-related diseases [63]. Inhibition of the Syk pathway was determined using mast cell degranulation assay, whereas inhibition of the JAK pathway was determined using an IL-4-activated Ramos B-cell assay. For example, pyrimidine 34 was claimed to be W207-fold selective for the JAK pathway (IC50o0.240 nM) compared to the Syk pathway (IC50W50 uM) in these cellular assays [63]. As previously mentioned, the clinical candidate R-348 has been reported to inhibit both the JAK3 and the Syk pathways. H N H N
O
O
O
S
O N
N
N H N
N
N H 33
32
N HN F
N N
N N
N H
Cl 34
5.5 Other heterocycles Finally a report on the use of virtual screening to identify JAK3 inhibitors appeared [64]. Starting with the pan-JAK inhibitor 4 [41], iterative similarity searches identified 35 as a promising starting point for further
Advances in the Discovery of Small Molecule JAK3 Inhibitors
261
optimization. A pharmacophore search identified indazole 36. JAK family selectivity data for either lead were not presented. F
N NH
O NH
N
Ph N
O
N N H
HN N
H N
NH
N N
CF3
N N
4 JAK3 IC50 = 5 nM
35 JAK3 IC50 = 98 nM
36 JAK3 IC50 = 2,640 nM
6. CONCLUSIONS The pharmaceutical industry continues to invest in the discovery and development of orally active small molecules that may offer the potential to effectively treat significant inflammatory and immune-related diseases with minimal or no undesirable side effects. Targeted inhibition of the tyrosine kinase JAK3 has appeared as an attractive strategy in this regard primarily due to its localized tissue expression and specific effects within the immune system. The promising clinical efficacy reported for the JAK3 inhibitor CP-690,550 in rheumatoid arthritis patients is noteworthy and suggests that obtaining efficacy comparable to, or perhaps better than, the current marketed biologic therapies in this disease may be possible with a small molecule. Although highly selective inhibition of JAK3 for immunosuppression is particularly attractive from a safety perspective, it remains to be convincingly demonstrated in the clinic. While CP-690,550 does potently inhibit JAK3, it has been shown to inhibit to some extent other JAK family members, namely JAK1 and JAK2, which may contribute to enhanced efficacy in the clinic relative to purely selective JAK3 inhibition [27]. Subsequent medicinal chemistry efforts in generating inhibitors with alternative JAK family selectivity profiles, although challenging, may allow for the dissection of the pharmacology associated with the inhibition of JAK3 as well as the other members of this important family of enzymes.
REFERENCES [1] J. J. Krowelski, R. Lee, R. Eddy, T. B. Shows and R. Dalla-Favera, Oncogene, 1990, 5, 277. [2] A. F. Wilks, A. G. Harpur, R. R. Kurban, S. J. Ralph, G. Zurcher and A. Ziemiecki, Mol. Cell. Biol., 1991, 11, 2057.
262
Stephen T. Wrobleski and William J. Pitts
[3] J. N. Ihle, Trends Endocrinol. Metab., 1994, 5, 137. [4] S. G. Rane and E. P. Reddy, Oncogene, 1994, 9, 2415. [5] M. Pesu, A. Laurence, N. Kishore, S. H. Zwillich, G. Chan and J. J. O’Shea, Immunol. Rev., 2008, 223, 132. [6] P. J. Murray, J. Immunol., 2007, 178, 2623. [7] M. Aringer, S. R. Hofmann, D. M. Frucht, M. Chen, M. Centola, A. Morinobu, R. Visconti, D. L. Kastner, J. S. Smolen and J. J. O’Shea, J. Immunol., 2003, 170, 6057. [8] J. L. Roberts, A. Lengi, S. M. Brown, M. Chen, Y.-J. Zhou, J. J. O’Shea and R. H. Buckley, Blood, 2004, 103, 2009. [9] C. M. Mjannes, R. W. Hendershot, R. R. Quinones and E. W. Gelfand, J. Allergy Clin. Immunol., 2007, 119, 1542. [10] M. A. Slatter, K. Brigham, A. M. Dickinson, H. L. Harvey, D. Barge, A. Jackson, N. Bown, T. J. Flood, A. J. Cant, M. Abinun and A. R. Gennery, J. Allergy Clin. Immunol., 2008, 121, 316. [11] J. J. O’Shea, M. Husa, D. Li, S. R. Hofmann, W. Watford, J. L. Roberts, R. H. Buckley, P. Changelian and F. Candotti, Mol. Immunol., 2004, 41, 727. [12] L. L. Rokosz, J. R. Beasley, C. D. Carroll, T. Lin, J. Zhao, K. C. Appell and M. L. Webb, Expert Opin. Ther. Targets, 2008, 12, 883. [13] V. Sexl, B. Kovacic, R. Piekorz, R. Moriggl, D. Stoiber, A. Hoffmeyer, R. Liebminger, O. Kudlacek, E. Weisz, K. Rothammer and J. N. Ihle, Blood, 2003, 101, 4937. [14] S. J. Rodig, M. A. Meraz, J. M. White, P. A. Lampe, J. K. Riley, C. D. Arthur, K. L. King, K. C. F. Sheehan, L. Yin, D. Pennica, E. M. Johnson, Jr. and R. D. Schreiber, Cell, 1998, 93, 373. [15] L. A. Sorbera, N. Serradell, J. Bolos, E. Rosa and J. Bozzo, Drugs Future, 2007, 32, 674. [16] P. S. Changelian, M. E. Flanagan, D. J. Ball, C. R. Kent, K. S. Magnuson, W. H. Martin, B. J. Rizzuti, P. S. Sawyer, B. D. Perry, W. H. Brissette, S. P. McCurdy, E. M. Kudlacz, M. J. Conklyn, E. A. Elliott, E. R. Koslov, M. B. Fisher, T. J. Strelevitz, K. Yoon, D. A. Whipple, J. Sun, M. J. Munchhof, J. L. Doty, J. M. Casavant, T. A Blumenkopf, M. Hines, M. F. Brown, B. M. Lillie, C. Subramanyam, S.-P. Chang, A. J. Milici, G. E. Beckius, J. D. Moyer, C. Su, T. G. Woodworth, A. S. Gaweco, C. R. Beals, B. H. Littman, D. A. Fisher, J. F. Smith, P. Zagouras, H. A. Magna, M. J. Saltarelli, K. S. Johnson, L. F. Nelms, S. G. Des Etages, L. S. Hayes, T. T. Kawabata, D. Finco-Kent, D. L. Baker, M. Larson, M.-S. Si, R. Paniagua, J. Higgins, B. Holm, B. Reitz, Y.-J. Zhou, R. E. Morris, J. J. O’Shea and D. C. Borie, Science, 2003, 302, 875. [17] X. Li, M. Jesson, J. Lee, J. Hirsch, M. Saabye, S. Bonar, N. Venkatraman, J. Zhang, L. Kahn, S. Ghosh, C. Sommers, D. Meyer and N. Kishore, Poster A164, The 15th International Inflammation Research Association Conference, Chantilly, VA, September 2008. [18] N. K. Williams, R. S. Bamert, O. Patel, C. Wang, P. M. Walden, A. F. Wilks, E. Fantino, J. Rossjohn and I. S. Lucet, J. Mol. Biol., 2009, 387, 219. [19] M. P. Clark, K. M. George, R. G. Bookland, J. Chen, S. K. Laughlin, K. D. Thakur, W. Lee, J. R. Davis, E. J. Cabrera, T. A. Brugel, J. C. VanRens, M. J. Laufersweiler, J. A. Maier, M. P. Sabat, A. Golebiowski, V. Easwaran, M. E. Webster, B. De and G. Zhang, Bioorg. Med. Chem. Lett., 2007, 17, 1250. [20] M. W. Karaman, S. Herrgard, D. K. Treiber, P. Gallant, C. E. Atteridge, B. T. Campbell, K. W. Chan, P. Ciceri, M. I. Davis, P. T. Edeen, R. Faraoni, M. Floyd, J. P. Hunt, D. J. Lockhart, Z. V. Milanov, M. J. Morrison, G. Pallares, H. K. Patel, S. Pritchard, L. M. Wodicka and P. P. Zarrinkar, Nature Biotech 2008, 26, 127 (supplemental Info). [21] www.clinicaltrials.gov as of March 7, 2009. [22] www.rigel.com as of March 7, 2009. [23] T. Deuse, J. B. Velotta, G. Hoyt, J. A. Govaert, V. Taylor, E. Masuda, E. Herlaar, G. Park, D. Carroll, M. P. Pelletier, R. C. Robbins and S. Schrepfer, Transplantation, 2008, 85, 885.
Advances in the Discovery of Small Molecule JAK3 Inhibitors
263
[24] www.vpharm.com as of March 7, 2009. [25] Q. Lin, D. Meloni, Y. Pan, M. Xia, J. Rodgers, S. Shepard, M. Li, L. Galya, B. Metcalf, T.-Y. Yue, P. Liu and J. Zhou, Org. Lett., 2009, 11, 1999. [26] S. Verstovsek, H. Kantarjian, A. Pardanani, D. Thomas, J. Cortes, R. Mesa, J. Redman, C.-M. Staschen, J. Fridman, K. Vaddi and A. Tefferi, Presentation 558, American Society of Hematology Meeting, Atlanta, GA, December 2007. [27] K. Ghoreschi, A. Laurence and J. J. O’Shea, Nat. Immunol., 10, 356. [28] J. Fridman, P. Scherle, R. Collins, T. Burn, Y. Li, J. Li, M. Covington, B. Thomas, M. Favata, J. Shi, R. McGee, S. Shepard, J. Rodgers, S. Yeleswaram, G. Hollis, R. Newton, B. Metcalf, S. Friedman and K. Vaddi, American College of Rheumatology Scientific Meeting, San Francisco, CA, October 2008, Poster Abstract 352. [29] R. Paniagua, M.-S. Si, M. G. Flores, G. Rousvoal, S. Zhang, O. Aalami, A. Campbell, P. S. Changelian, B. A. Reitz and D. C. Borie, Transplantation, 2005, 80, 1283. [30] D. C. Borie, M. J. Larson, M. G. Flores, A. Campbell, G. Rousvoal, S. Zhang, J. P. Higgins, D. J. Ball, E. M. Kudlacz, W. H. Brissette, E. A. Elliott, B. A. Reitz and P. A. Changelian, Transplantation, 2005, 80, 1756. [31] G. Rousvoal, M.-S. Si, M. Lau, S. Zhang, G. J. Berry, M. G. Flores, P. S. Changelian, B. A. Reitz and D. C. Borie, Transpl. Int., 2006, 19, 1014. [32] E. Van Gurp, W. Weimar, R. Gaston, D. Brennan, R. Mendez, J. Pirsch, S. Swan, M. D. Pescovitz, G. Ni, C. Wang, S. Krishnaswami, V. Chow and G. Chan, Am. J. Transplant., 2008, 8, 1711. [33] E. A. F. J. Van Gurp, W. Schoordijk-Vershoor, M. Klepper, S. S. Korevaar, G. Chan, W. Weimar and C. C. Baan, Transplantation, 2009, 87, 79. [34] A. J. Milici, E. M. Kudlacz, L. Audoly, S. Zwillich and P. Changelian, Arthritis Res. Ther., 2008, 10, R14, (http://arthritis-research.com/content/10/1/R14). [35] J. G. Walker, M. J. Ahern, M. Coleman, H. Weedon, V. Papangelis, D. Beroukas, P. J. Roberts-Thomson and M. D. Smith, Ann. Rheum. Dis., 2007, 66, 992. [36] T. Hampton, JAMA, 2007, 297, 28. [37] J. Kremer, S. Cohen, B. Wilkinson, C. Connell, J. French, J. Gomez Reino, D. Gruben, K. Kanik, S. Krishnaswami, V. Pascual-Ramos, G. Wallenstein and S. Zwillich, American College of Rheumatology Meeting Abstract, San Francisco, CA, October 2008, Presentation L13. [38] W. Williams, P. Scherle, J. Shi, R. Newton, E. McKeever, J. Fridman, T. Burn, K. Vaddi, R. Levy and L. Moreland, American College of Rheumatology Meeting Abstract, San Francisco, CA, October 2008, Presentation 714. [39] B. Wilkinson, A. Gaweco, P. Changelian, M. Boy, C. Wang, V. Chow, G. Chan, J. Herron, S. Zwillich and J. Krueger, Ann. Rheum. Dis., 2007, 66 (Suppl. 2, EULAR meeting), Barcelona, Spain, June 2007, Abstract THU0099. [40] N. Punwani, W. Williams, P. Scherle, J. Shi, R. Newton, R. Flores, S. Friedman, K. Vaddi, R. Levy, A. VanVorhees and A. Gottlieb, Presentation, European Academy of Dermatology Venerology meeting, Paris, France, September 2008. [41] T. J. Boggon, Y. Li, P. W. Manley and M. J. Eck, Blood, 2005, 106, 996. [42] I. S. Lucet, E. Fantino, M. Styles, R. Bamert, O. Patel, S. E. Broughton, M. Walter, C. J. Burns, H. Treutlein, A. F. Wilks and J. Rossjohn, Blood, 2006, 107, 176. [43] H. Zuccola, M. Jacobs, L. Swenson and K. Saxena, WO Patent Application 2005/105988, 2005. [44] S. M. Stepkowski, J. Kao, M.-E. Wang, N. Tejpal, H. Podder, L. Furian, J. Dimmock, A. Jha, U. Das, B. D. Kahan and R. A Kirken, J. Immunol., 2005, 175, 4236. [45] L. Honigberg, E. Verner, J. J. Buggy, D. Loury and W. Chen, WO Patent Application 2008/121742, 2008. [46] X. Zhang, Y. Hu and Z. Yuan, Biochem. Biophys. Res. Comm., 2008, 370, 72. [47] P. S. Changelian and S. H. Zwillich, WO Patent Application 2008/029237, 2008.
264
Stephen T. Wrobleski and William J. Pitts
[48] J.-K. Jiang, K. Ghoreschi, F. Deflorian, Z. Chen, M. Perreira, M. Pesu, J. Smith, D.-T. Nguyen, E. H. Liu, W. Leister, S. Costanzi, J. J. O’Shea and C. J. Thomas, J. Med. Chem., 2008, 51, 8012. [49] C. T. Brain, G. Thoma, M. J. Sung and L. T. McNally, WO Patent Application 2007/ 140222, 2007. [50] J. Salas Solana, C. Almansa Rosales, R. Soliva, M. Fontes Ustrell and M. Vendrell Escobar, WO Patent Application 2008/119792, 2008. [51] T. Inoue, T. Tojo, M. Morita, Y. Nakajima, K. Hatanaka, S. Shirakami, H. Sasaki, A. Tanaka, F. Takahashi, K. Mukoyoshi, Y. Higashi, A. Okimoto, T. Hondo and H. Sawada, WO Patent Application 2007/007919, 2007. [52] T. Inoue, A. Tanaka, K. Nakai, H. Sasaki, F. Takahashi, S. Shirakami, K. Hatanaka, Y. Nakajima, K. Mukoyoshi, H. Hamaguchi, S. Kunikawa and Y. Higashi, WO Patent Application 2007/077949, 2007. [53] S. Shirakami, T. Inoue, K. Mukoyoshi, Y. Nakajima, H. Usuda, H. Hamaguchi, Y. Higashi and K. Hatanaka, WO Patent Application 2008/084861, 2008. [54] J. Salas Solana, C. Almansa Rosales, R. Soliva Soliva, M. Fontes Ustrell, M. Virgili Bernado, J. Comelles Espuga and J. J. Pastor Porras, WO Patent Application 2008/ 090181, 2008. [55] M. Ohlmeyer, A. Bohnstedt, C. Kingsbury, K.-K. Ho, J. Quintero, M. You, H. Park and Y. Lu, WO Patent Application 2006/108103, 2006. [56] Y. Lu, C. Kingbury, A. Bohnstedt, M. Ohlmeyer and V. Paradkar, WO Patent Application 2008/043019, 2008. [57] I. Neagu, D. Dillar, C. Kingsbury, A. Bohnstedt, M. Ohlmeyer, V. Paradkar and N. Ansari, WO Patent Application 2008/043031, 2008. [58] M. Ohlmeyer, A. Bohnstedt, C. Kingsbury, K.-K. Ho and J. Quintero, WO Patent Application 2008/060301, 2008. [59] M. Sabat, J. C. VanRens, M. J. Laufersweiler, T. A. Brugel, J. Maier, A. Golebiowski, B. De, V. Easwaran, L. C. Hsieh, R. L. Walter, M. J. Mekel, A. Evdokimov and M. J. Janusz, Bioorg. Med. Chem. Lett., 2006, 16, 5973. [60] M. Sills, K. Appell, A. Bohnstedt, et al., Inflamm. Res., 2006, 55, S118. [61] M. R. Kling and C. J. Burns, WO Patent Application 2007/062459, 2007. [62] D. G. Bourke, X. Bu, C. J. Burns, A. N. Cuzzupe, J. T. Trill, T. L. Nero, M. B. Blannin, J. Zeng and S. P. Gaynor, WO Patent Application 2008/092199, 2008. [63] B. Wong, US Patent Application 2006/0270694, 2006. [64] X. Chen, L. J. Wilson, R. Malaviya, R. L. Argentieri and S.-M. Yang, J. Med. Chem., 2008, 51, 7015.
CHAPT ER
13 Recent Advances in Adenosine Receptor (AR) Ligands in Pulmonary Diseases Rao Kalla and Jeff Zablocki
Contents
1. Introduction 2. A1 Adenosine Receptor Antagonists: L-97-1, BG-9928, FK-838, and WRC-0571 3. A2A Adenosine Receptor Agonists: CGS-21680, UK-371104, and GW-328276 4. A2B Adenosine Receptor Antagonists: CVT-6883, MRE 2029-F20, LAS-38096, and OSIP-339391 5. A3 Adenosine Receptor Antagonists: MRS-1523, KF-26777, and MRE-3008-F20 6. Summary References
265 266 268 270 272 274 275
1. INTRODUCTION Adenosine is an endogenous ligand that has a short half-life due to its rapid conversion to inosine by adenosine deaminase [1,2]. When generated locally in various tissues of the body, adenosine binds to one of its four P1 family of G-protein receptors: A1 and A3 that are coupled to Gi and lower cyclic adenosine monophosphate (cAMP) levels and A2A and A2B that are coupled to Gs and increase cAMP levels [1]. Adenosine has been implicated in both the pro-inflammatory and immunomodulatory Department of Medicinal Chemistry, CV Therapeutics, 3172 Porter Drive, Palo Alto, CA 94304, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04413-3
r 2009 Elsevier Inc. All rights reserved.
265
266
Rao Kalla and Jeff Zablocki
pathogenic mechanisms of asthma and chronic obstructive pulmonary disease (COPD) [3–5]. For instance, adenosine levels are elevated in the bronchoalveolar lavage fluid (BALF) of asthmatics relative to healthy volunteers [6], and adenosine is found in the exhaled breath condensate of asthmatics [7]. In addition, when adenosine monophosphate (AMP) is administered to asthmatics and healthy, normal individuals, it provides a source of adenosine that leads to bronchoconstriction in asthmatics, but not normals [8]. Cigarette smokers with COPD demonstrate a higher prevalence of hyper-responsiveness to AMP than non-smokers with COPD [9]. Furthermore, AMP induces bronchoconstriction in COPD patients [9]. In addition, an adenosine uptake blocker, dipyridamole, can precipitate asthma [10]. Thus, substantial evidence supports the role of adenosine inducing bronchoconstriction in asthma and COPD. The bronchoconstrictor effect is predominantly through mast cell–mediated histamine release [11–13] and the release of prostaglandins, leukotrienes, and interleukin (IL)-8 [14,15]. The role of adenosine in asthma and COPD will be further delineated in the description of therapeutic ligands by each receptor subtype.
2. A1 ADENOSINE RECEPTOR ANTAGONISTS: L-97-1, BG-9928, FK-838, AND WRC-0571 Activation of A1 adenosine receptors (ARs) may play a role in mucus production by human epithelial cells [16], human bronchial smooth muscle contraction [17], and neutrophil chemotaxis [18,19]. However, activation of A1 ARs on macrophages has the following effects: inhibits the production of several pro-inflammatory cytokines, including tumor necrosis factor-a (TNF-a), IL-6, and IL-8, and enhances the release of the anti-inflammatory cytokine, IL-10 [20–22]. Several compounds from different classes of A1 AR antagonists are noteworthy for either demonstrating efficacy in animal models of asthma or entering clinical trials for asthma or other indications. Xanthines are considered classical antagonists for ARs. For example, the non-selective xanthine AR antagonists, theophylline (1,3-dimethylxanthine) and caffeine (1,3,7-trimethylxanthine), display micromolar affinity at various AR subtypes, with some affinity at the A1 AR [23]. Over the past 20 years, multiple research groups contributed to the structure activity relationships (SAR) for substitution at the 1-, 3-, 7-, and 8-positions of the xanthine core to provide selective antagonists for the AR subtypes [24–26]. Compound 1 (L-97-1) is a result of such SAR optimization and was described as a water-soluble, small molecule, A1 AR antagonist with high affinity (580 nM) and W100-fold selectivity over the other AR subtypes
267
Recent Advances in Adenosine Receptor Ligands in Pulmonary Diseases
[27]. In an allergic rabbit model, compound 1 blocked early and late allergic responses and bronchial hyper-responsiveness to histamine [27]. Also, in the same rabbit model, it reduced the number of eosinophils, neutrophils, and lymphocytes in bronchoalveolar lavage. Following oral dosing in rats, L-97-1 displayed good plasma concentrations, potentially supporting its use as an oral anti-asthma treatment [27]. In addition, allergen-sensitized rabbits that were treated with A1 AR antisense required a W10-fold increase in adenosine to induce a 50% decrease in dynamic compliance of the lung, thus supporting the role of the A1 AR in the rabbit [28].
N
OH O
O N
N O
N
N
NH2
H N
N O
N
COOH
N
BG-9928 (2)
L-97-1 (1)
HO N
N NH N
N
N
N O
N
N
N
O HO FK-838 (3)
WRC-0571 (4)
Substitution of the xanthine core, at the 8-position, with a bicyclo[2.2.2]octyl group produced compound 2 (BG-9928) that has high affinity (7.4 nM) for the A1 AR and displays good selectivity (915-fold)
268
Rao Kalla and Jeff Zablocki
over the A2A AR [29]. BG-9928 has high oral bioavailability in rat (99%), dog (78%), and cynomolgus monkey (94%), and it demonstrated excellent oral efficacy (ED50 ¼ 0.01 mg/kg) in a rat diuresis model [29]. Compound 2 was well tolerated when dosed to rats and cynomolgus monkeys through either intravenous or oral administration for a period of 3 months. Compound 2 (BG-9928) is in phase III clinical trails for the treatment of the edema associated with congestive heart failure [30]. FK-838 (3) represents a structurally unique class of A1 AR antagonists, pyrazolo[1,5-a]pyridines [31]. Compound 3 displayed high affinity (120 nM) and selectivity (W50-fold over A2A AR) for A1 AR and had very good aqueous solubility (10 mg/mL) and high oral availability (78%). Compound 3 is in phase II clinical trials as a diuretic antihypertensive agent [31]. Compound 4 (WRC-0571) is an adenine-based A1 antagonist that has high affinity (1.7 nM) and W100-fold selectivity over other AR subtypes [32]. Compound 4 was demonstrated to have good oral bioavailability across species and is in preclinical development as both a diuretic agent and for the treatment of renal failure [30,32].
3. A2A ADENOSINE RECEPTOR AGONISTS: CGS-21680, UK-371104, AND GW-328276 Activation of A2A AR has the following anti-inflammatory effects on various cell types: inhibition of the release of histamine and tryptase from mast cells [33] and reduction of chemotaxis, in addition to activation/degranulation of neutrophils [19,34,35]. Activation of A2A AR also increases the production of anti-inflammatory IL-10 from monocytes and macrophages [36]. Therefore, various A2A AR agonists have been evaluated in animal models of asthma and have even advanced into clinical trials [37]. Early on, compound 5 was synthesized by exploring the SAR studies on N-ethyl carboxamide adenosine (NECA), a non-selective AR agonist [38]. Substitution at the 2-position of NECA with a phenethyl amine derivative provided compound 5 (CGS-21680) that displayed 144-fold selectivity for the A2A AR over the A1 AR [38]. When compound 5 was given intratracheally, it inhibited both the early and the late inflammatory reactions in allergen-challenged Brown Norway rats [39]. Compound 5 inhibited both inflammatory cell influx of BALF in asthmatic mice and neutrophil activation in a mouse model of COPD [40]. However, cardiovascular side effects were observed with similar doses of CGS-21680, thus demonstrating no apparent separation of the anti-inflammatory effects of CGS-21680 from its cardiovascular effects [39]. Since cardiovascular effects, in particular,
269
Recent Advances in Adenosine Receptor Ligands in Pulmonary Diseases
hypotension, were observed in laboratory animals when A2A agonists were administered systemically [39], inhaled administration of an A2A agonist was targeted. The compounds were designed to have inhaled local lung effects by optimizing the physiochemical properties of the compound by increasing the molecular weight and hydrophilicity for high systemic clearance. After extensive SAR, N6-(2,2-diphenylethyl)-2[N-[2-(1-piperidinyl)ethyl]carbamoyl] adenosine compound 6 (UK-371, 104) displayed good A2A AR affinity (65 nM), and it also inhibited the release of inflammatory mediators from isolated human neutrophils [41]. Compound 6 inhibited the capsaicin-induced bronchoconstriction in an anaesthetized guinea pig model without affecting blood pressure [37]. The cardiovascular side effects of compound 6, following the intratracheal administration at a dose of 1 mg/kg, are much less compared to CGS-21680 because of the low systemic exposure of the drug. UK432,097, an analog of UK-371,104, had a similar pharmacological profile in the anaesthetized guinea pig model, and it was moved into clinical trials [37]. In 2008, the clinical trials were discontinued for the development of UK-432,097 because of lack of efficacy [42]. Another A2A AR agonist, which targets the lung, compound 7 (GW-328267X), was found to have good A2A and A3 AR affinity (16 nM) with good selectivity over A1 and A2B AR subtypes (W60- and W300-fold respectively) [43]. Treatment of non-smoking, atopic asthmatics, who underwent an inhaled allergen challenge, did not provide significant protection against the allergen-induced early and late asthmatic reaction [44]. In the same study, compound 7 did not inhibit the accompanying inflammatory response, as measured by sputum total cell counts, number of EG2+ cells, and the concentrations of IL-8 and eosinophil cationic protein [44,45]. Compound 7, which was in phase II clinical trials for COPD, has been dropped from development because of lack of efficacy [30] that might be due to the low dose used (25 mg, twice daily), limited by prohibitive cardiovascular side effects [45].
NH2 N O
N O
N H
NH2 N
N O
N
N O
N N
N H
N H HO
OH
NECA
HO
OH
O
CGS-21680 (5)
OH
270
Rao Kalla and Jeff Zablocki Ph
HN N N O
N
Ph H N
N O
HO HO
NH2 N
N
N
N N O
N
UK-371, 104 (6)
N
N H
N HO
OH
OH
N
OH
GW-328276X (7)
4. A2B ADENOSINE RECEPTOR ANTAGONISTS: CVT-6883, MRE 2029-F20, LAS-38096, AND OSIP-339391 There is substantial evidence to indicate that A2B ARs contribute to airway inflammation in asthma and that a compound specifically inhibiting the A2B AR may have beneficial effects [46–49]. Activation of A2B ARs on human bronchial smooth muscle cells (BSMCs) has been shown to induce the release of the inflammatory cytokines IL-6 and monocytic chemotactic protein-1 [50]. A2B AR activation on human bronchial epithelial cells (HBECs) releases IL-19, which in turn activates human monocytes to induce the release of TNF-a, thereby upregulating A2B AR expression on human bronchial endothelial cells [51]. The above evidence supports the hypothesis that adenosine plays a role in asthma, and its effects may be at least, in part, mediated through the A2B AR. Several high-affinity and selective antagonists of A2B AR have been reported recently by several groups. These examples can be used to fully understand the therapeutic potential of A2B AR antagonists as anti-inflammatory and anti-angiogenic agents [24–26,52]. The structural approach taken by these groups can be divided into two classes of compounds, xanthines and non-xanthine derivatives. Exploration of 8-heteroaryl substitution on xanthines led to the identification of 8-(pyrazol-4-yl) xanthines as high-affinity and selective A2B AR antagonists [53]. Differential substitution at the N-1 and N-3 positions of the xanthine core with propyl and ethyl groups, respectively, led to compound 8 (CVT-6883), which displays high affinity (22 nM) for the A2B AR with good selectivity (88-, 148-, and 48-fold over A1, A2A, and A3 ARs respectively) [54]. Compound 8 antagonized the NECA-induced cAMP accumulation in HEK-A2B cells and NIH 3T3 cells (KB values of 6 and 2 nM respectively) [55] and has also been shown to completely abolish the NECA-induced cAMP accumulation in BSMCs [50]. When dosed orally to rats at 2 mg/kg, compound 8 had excellent systemic exposure with a t1/2 of 4 h, Cmax W1100 ng/mL, and a dose-adjusted area
271
Recent Advances in Adenosine Receptor Ligands in Pulmonary Diseases
under curve (dAUC) of 6500 ngh/mL [54]. In a mouse model of asthma, compound 8 demonstrated a dose-dependent (0.42.5 mg/kg) blocking effect on NECA-induced increases in airway reactivity [56]. Also, in this same model, compound 8 significantly reduced both the late allergic airway response and the recruitment of inflammatory cells in BALF [56]. In the adenosine deaminase–deficient mouse model, compound 8 attenuated pulmonary inflammation, fibrosis, airway enlargement, and lung injury [55]. CVT-6883 is currently being developed as an oral treatment for asthma and has completed phase I clinical trials [30,57].
O
O H N
N O
N
N
N
H N
N
N
N
N
O
O N
N
O
NH
CF3 O
MRE 2029-F20 (9)
CVT-6883 (8)
O
H N
N NH
H N
N
O
O
N
N N N
N
N
N H
O
N
LAS-38096 (10)
OSIP-339391 (11)
In another series of compounds, substitution at the 8-position of xanthine, with a 5-pyrazole group, led to compound 9 that displayed high affinity and selectivity for the A2B AR [58]. The 5-pyrazolyl derivative, 9 (MRE 2029-F20), displayed good A2B AR affinity (5.5 nM) and selectivity (W180-fold over all AR subtypes). Compound 9 blocks NECA-induced cAMP accumulation with IC50 values in the nanomolar range [58]. The tritium-labeled derivative of 9 displayed a Kd value of 1.6570.10 nM in Chinese Hamster Ovary (CHO) cells that express hA2B receptors, and it can be useful as a pharmacological tool in competitive binding studies [59]. Two series of compounds, 2-aminopyridines and 2aminopyrimidines, were explored as A2B AR antagonists. This work led
272
Rao Kalla and Jeff Zablocki
to the identification of N-heteroaryl 4u-furyl-4,5u-bipyrimidin-2u-amines as high-affinity and selective A2B AR antagonists [60]. The lead compound, 2u-amino(3-pyridyl) derivative 10 (LAS-38096) has a A2B AR affinity of 17 nM and very good selectivity (W58-, W147-, and W58-fold over A1, A2A, and A3 ARs respectively). Compound 10 inhibited the NECAinduced cAMP levels in HEK-293 expressing human A2B AR (IC50 of 321 nM) and CHO cells (IC50 of 349 nM) that were transfected with mouse A2B AR [60]. Compound 10 displayed good systemic exposure, with a Cmax of 11 mM and an AUC of 16 mM/h, when dosed to rats, and also displayed good exposure following oral dosing in mice and dogs [60]. After treatment of an allergic mouse (female Balb/c mice) with compound 10, the mouse showed decreased bronchial hyperresponsiveness, mucus production, and a slight decrease in eosinophil infiltration and Th2 cytokine levels [61]. Compound 11 (OSIP-339391) is a representative of the deaza-adenine class of A2B AR antagonists [62]. A 2-phenyl-7-deazaadenine analog, compound 11, demonstrated excellent A2B AR affinity (0.5 nM) and good selectivity (74-, 656-, and 900-fold over A1, A2A, and A3 ARs respectively). The tritium-labeled analog of 11 displayed a Kd value of 0.4170.06 nM for binding to human A2B AR expressed in HEK-293 cells [62]. This selective and high-affinity radioligand can be a useful tool in further characterization of the pharmacology of the A2B AR. There is now considerable evidence that A2B ARs are involved in both airway inflammation and the process of airway remodeling. The availability of high-affinity and selective A2B AR antagonists allows for the clinical evaluation of the role of the A2B AR in asthma and COPD.
5. A3 ADENOSINE RECEPTOR ANTAGONISTS: MRS-1523, KF-26777, AND MRE-3008-F20 There are major differences among species in the expression and function of the A3 AR subtype, which complicates the understanding of the functional significance of this receptor subtype in the pathogenesis of chronic inflammatory airway diseases [63]. It has been shown that the A3 AR mediates the adenosine-induced mast cell degranulation in rats, mice, and guinea pigs [64], whereas A3 receptors have not been identified on human mast cells [65]. A3 ARs have been shown to play an important role in eosinophilia and mucus production in animal models [65]. Also, A3 ARs are found on human eosinophils, and elevated levels of A3 receptors are observed in lung biopsies of patients with asthma or COPD [66].
273
Recent Advances in Adenosine Receptor Ligands in Pulmonary Diseases
O
O
S
N H N
N
O N
O
N
N
MRS-1523 (12)
KF-26777 (13)
O
CN
O N H
NH N
N
N
O
N NH
N
N N
MRE-3008-F20 (14)
N
S
N
N
N Compound (15)
The classical xanthine antagonists of ARs, theophylline and caffeine, typically have low binding affinities for the A3 AR, and therefore, the search for A3 AR antagonists has been focused on novel heterocyclic systems. Initially, a dihydropyridine derivative that has good affinity for the A3 AR was discovered by screening diverse chemical libraries. Further optimization led to the corresponding pyridine derivative, compound 12 (MRS-1523), that displayed high affinity (18.9 nM for human and 113 nM for rat) and selectivity (W130-fold for rA1 AR) for the A3 AR [67]. Treatment of adenosine deaminase (ADA)-deficient mice with selective A3 AR antagonist 12 reduced the number of airway eosinophils and decreased mucus production in the airways. Similar findings were also observed in the lungs of ADA/A3 knockout mice, supporting the important role of A3 AR mediation of lung eosinophilia and mucus hyperplasia in pulmonary disorders triggered by elevated adenosine levels [65]. A series of tricyclic imidazo[2,1-i]purinones have been prepared as A3 AR antagonists by modifications of the xanthine
274
Rao Kalla and Jeff Zablocki
core [68]. An analog of the series is compound 13 (KF-26777) that displays high affinity (0.20 nM) and selectivity (1,000-fold) for the A3 AR. Compound 13 inhibited the binding of [35S]GTPgS, which was stimulated by 2-chloro-N6-(3-iodobenzyl)adenosine-5-N-methyluronamide (Cl-IB-MECA) with an IC50 value of 270785 nM. It also antagonized the [Ca2+]I mobilization, induced by Cl-IB-MECA, with a KB value of 0.4270.14 nM, suggesting that it is a highly potent and selective A3 AR antagonist [68]. A tricyclic series, illustrated by lead compound 14 (MRE-3008-F20), displayed high affinity (0.28 nM) and selectivity (W30,000-fold) for the A3 AR [69]. Compound 14 blocked the IB-MECA-induced cAMP production in CHO cells with an IC50 of 4.5 nM, confirming that it is a functional antagonist [69]. The tritiumlabeled analog of 14 bound to hA3 ARs, expressed in CHO cells, with a Kd value of 0.82 nM, suggesting that it can be a useful tool for characterization of A3 ARs in both normal and pathological conditions [70]. In recent reports, a series of 5-heterocyclic-substituted aminothiazoles have been identified as dual antagonists for both the A2B and the A3 ARs [71]. Of these analogs, trisubstituted aminothiazole 15 displayed high affinity for both A2B (3 nM) and A3 (10 nM) ARs and gave selectivity against the A1 (W20-fold) and A2A (W160-fold) ARs. Compound 15 showed good oral bioavailability (30%) in Wistar rats and displayed good absorption distribution metabolism excretion (ADME) properties [71]. Some of the compounds in this series also inhibited the p38 mitogen-activated protein (MAP) kinases a and b and the phosphodiesterase 4D (PDE4D) isoenzyme with IC50 values in the low nanomolar range [72]. This compound represents a new series that inhibits both A2B and A3 ARs and can potentially be used as a new tool in testing the therapeutic potential of dual inhibition in allergic diseases.
6. SUMMARY Adenosine has been strongly implicated in asthma and COPD based on supporting studies in animal models, AMP (adenosine precursor) effects in humans, and elevated adenosine levels in BALF of asthmatic patients. The major biological effect of adenosine is thought to occur through histamine released from activated mast cells through A2B AR activation. However, additional effects are mediated through multiple ARs including inflammation (A1, A2A, and A2B), mucus production (A1), and inflammatory cell recruitment (A1, A2A, A2B, and A3). Selective AR antagonists have been discovered for each AR isoform through optimization of substituents on each core scaffold. This optimization has been driven by a mixture of classical medicinal chemistry, ligand-based modeling, and initially homology AR modeling based on the rhodopsin X-ray structure
Recent Advances in Adenosine Receptor Ligands in Pulmonary Diseases
275
[24–26]. Recently, the A2A AR X-ray has been solved with an A2A AR antagonist ZM-241385 that may prove useful in the design of subsequent subtype selective AR antagonists and AR agonists [73]. Although many selective A1 antagonists have been discovered, the principal focus of these compounds has been on treatment of the edema associated with CHF and not asthma or COPD. This may be due to an early bias in animal models (e.g., rabbit) where the A1 AR may play a more critical role in asthma than in humans. Selective A2A agonists are available for local delivery to the lung, but initial asthma clinical studies were not promising. Certainly, a number of A2B antagonists possess very high affinity and selectivity and demonstrate activity in animal models of asthma (8 and 10). Compound 8 is the most advanced A2B antagonist and is currently in phase I clinical trials for asthma. Some early animal data with respect to the role of ARs in asthma and COPD is under scrutiny. For instance, the studies with MRS1523 in rodents suggest that the activation of A3 AR may contribute to airway obstruction by inducing hypersecretion of the mucus or by migration of eosinophils to the airways. However, the distribution of A3 ARs in rodents does not match the distribution in humans; therefore, the clinical relevance of these animal models is unclear. The consequence of A3 AR activation in humans must be further investigated due to conflicting animal and human data with respect to the possible role of A3 receptors. The availability of selective antagonists of AR subtypes will help define the role of adenosine in asthma and COPD in future clinical trials.
REFERENCES [1] C. E. Muller, Farmaco, 2001, 56, 77. [2] B. Pavan and A. P. Ijzerman, Biochem. Pharmacol., 1998, 56, 1625. [3] D. Zeng, R. Polosa, I. Biaggioni and L. Belardinelli, in Asthma: Modern Therapeutic Targets (eds R. Polosa and S. T. Holgate), Clinical Publishing, Oxford, UK, 2007, pp. 1–13. [4] R. A. Brown, D. Spina and C. P. Page, Br. J. Pharmacol., 2008, 153, S446. [5] M. van der Berge, M. N. Hylkema, M. Versluis and D. S. Postma, Drugs RD, 2007, 8, 13. [6] A. G. Driver, C. A. Kukoly, S. Ali and S. J. Mustafa, Am. Rev. Respir. Dis., 1993, 148, 91. [7] E. Huszar, G. Vass and E. Vizi, Eur. Respir. J., 2002, 20, 1393. [8] M. J. Cushley, A. E. Tattersfield and S. T. Holgate, Am. Rev. Respir. Dis., 1984, 129, 380. [9] Y. Oosterhoff, J. W. de Jong and M. A. Jansen, Am. Rev. Respir. Dis., 1993, 147, 553. [10] J. R. Fozard and C. Carth, Curr. Opin. Investig. Drugs, 2002, 3, 69. [11] G. D. Phillips, P. Rafferty and S. T. Holgate, Thorax, 1987, 42, 939. [12] G. D. Phillips and S. T. Holgate, Am. Rev. Respir. Dis., 1989, 139, 463. [13] S. R. Rutgers, G. H. Koetger and T. W. van der Mark, Clin. Exp. Allergy, 1999, 29, 1287. [14] M. K. Church, S. T. Holgate and P. J. Hughes, Br. J. Pharmacol., 1983, 80, 719. [15] P. J. Hughes, S. T. Holgate and M. K. Church, Biochem. Pharmacol., 1984, 33, 3847. [16] N. McNamara, M. Gallup, A. Khong, A. Sucher, I. Maltseva, J. Fahy and C. Basbaum, FASEB J., 2004, 18, 1770. [17] M. F. Ethier and J. M. Madison, Am. J. Respir. Cell Mol. Biol., 2006, 35, 496.
276
Rao Kalla and Jeff Zablocki
[18] B. N. Cronstein, L. Daguma, D. Nicholas, A. J. Hutchison and M. Williams, J. Clin. Invest., 1990, 85, 1150. [19] B. N. Cronstein, R. I. Levin, M. Philips, R. Hirschhorn, S. B. Abramson and G. Weissmann, J. Immunol., 1992, 148, 2201. [20] F. G. Sajjadi, K. Takabayashi, A. C. Foster, R. C. Domingo and G. S. Firestein, J. Immunol., 1996, 156, 3435. [21] G. Hasko, C. Szabo, Z. H. Nemeth, V. Kvetan, S. M. Pastores and E. S. Vizi, J. Immunol., 1996, 157, 4634. [22] O. Monie, P. Stordeur, L. Schandene, A. Marchant, D. de Groote, M. Goldman and J. Deviere, J. Immunol., 1996, 156, 4408. [23] K. A. Jacobson, A. P. Ijzerman and J. Linden, Drug. Dev. Res, 1999, 47, 45. [24] R. Akkari, J. C. Burbiel, J. Hockemeyer and C. E. Muller, Curr. Top. Med. Chem., 2006, 6, 1375. [25] K. A. Jacobson and Z.-G. Gao, Nat. Rev. Drug Discov., 2006, 5, 247. [26] P. G. Baraldi, M. A. Tabirizi, S. Gessi and P. A. Borea, Chem. Rev., 2008, 108, 238. [27] P. C. M. Obiefuna, V. K. Batra, A. Nadeem, P. Borron, C. N. Wilson and S. J. Mustafa, J. Pharmacol. Exp. Therp., 2005, 315, 329. [28] J. W. Nyce and W. J. Metzger, Nature, 1997, 385, 721. [29] W. F. Kiesman, J. Zhao, P. R. Conlon, J. E. Dowling, R. C. Petter, F. Lutterodt, X. Jin, G. Smits, M. Fure, A. Jayaraj, J. Kim, G. Sullivan and J. Linden, J. Med. Chem., 2006, 49, 7119. [30] http://integrity.prous.com [31] A. Akahane, H. Katayama, T. Mitsunaga, T. Kato, T. Kinoshita, Y. Kita, T. Kusunoki, T. Terai, K. Yoshida and Y. Shiokawa, J. Med. Chem., 1999, 42, 779. [32] P. L. Martin, R. J. Wysocki, Jr., R. J. Barrett, J. M. May and J. Linden, J. Pharmacol. Exp. Therp., 1996, 276, 490. [33] H. Suzuki, M. Takei, T. Nakahata and H. Fukamachi, Biochem. Biophys. Res. Commun., 1998, 242, 697. [34] A. Wollner, S. Wollner and J. B. Smith, Am. J. Respir. Cell. Mol. Biol., 1993, 9, 179. [35] B. B. Fredholm, Y. Zhang and I. van der Ploeg, Naunyn Schmiedebergs Arch. Pharmacol., 1996, 354, 262. [36] S. Gessi, K. Varani, S. Merighi, E. Ongini and P. A. Borea, Br. J. Pharmacol., 2000, 129, 2. [37] M. A. Trevethick, S. J. Mantell, E. F. Stuart, A. Barnard, K. N. Wright and M. Yeadon, Br. J. Pharmacol., 2008, 155, 463. [38] A. J. Hutchison, R. L. Webb, H. H. Oel, G. R. Ghai, M. B. Zimmerman and M. Williams, J. Pharmacol. Exp. Ther., 1989, 251, 47. [39] J. R. Fozard, K. M. Ellis, M. F. V. Dantas, B. Tigani and L. Mazzoni, Eur. J. Pharmacol., 2002, 438, 183. [40] O. Bonneau, D. Wyss, S. Ferretti, C. Blaydon, C. S. Stevenson and A. Trifilieff, Am. J. Physiol. Lung Cell Mol. Physiol., 2006, 290, L1036. [41] S. J. Mantell, P. T. Stephenson, S. M. Monaghan, G. N. Maw, M. A. Trevethick, M. Yeadon, R. F. Keir, D. K. Walker, R. M. Jones, M. D. Selby, D. V. Batchelor, S. Rozze, H. Chavaroche, T. J. Hobson, P. G. Dodd, A. Lemaitre, K. N. Wright and E. F. Stuart, Bioorg. Med. Chem. Lett., 2008, 18, 1284. [42] http://www.clinicaltrials.gov. Safety and efficacy of UK-432,097 in chronic obstructive pulmonary disease, study was terminated as of March 17, 2009. [43] N. Bevan, P. R. Butchers, R. Cousins, J. Coates, E. V. Edgar, V. Morrison, M. J. Sheehan, J. Reeves and D. J. Wilson, Eur. J. Pharmacol., 2007, 564, 219. [44] B. Luijk, M. van der Berge, H. A. M. Kerstjens, D. S. Postma, L. Cass, A. Sabin and J.-W. J. Lammers, Allergy, 2008, 63, 75. [45] J. Rimmer, H. L. Peake, C. M. C. Santos, M. Lean, P. Bardin, R. Robson, B. Haumann, F. Loehrer and M. L. Handel, Clin. Exp. Allergy, 2007, 37, 8.
Recent Advances in Adenosine Receptor Ligands in Pulmonary Diseases
277
[46] I. Feoktistov and I. Biaggioni, Pharmacol. Rev., 1997, 49, 381. [47] I. Feoktistov, R. Polosa, S. T. Holgate and I. Biaggioni, Trends Pharmacol. Sci., 1998, 19, 148. [48] S. Ryzhov, A. E. Goldstein, A. Matafonov, D. Zeng, I. Biaggioni and I. Feoktistov, J. Immunol., 2004, 172, 7726. [49] S. T. Holgate, Br. J. Pharmacol., 2005, 145, 1009. [50] H. Zhong, L. Beladinelli, T. Maa, I. Feoktistov, I. Biaggioni and D. Zeng, Am. J. Respir. Cell Mol. Biol., 2004, 30, 118. [51] H. Zhong, Y. Wu, L. Belardinelli and D. Zeng, Am. J. Respir. Cell Mol. Biol., 2006, 35, 587. [52] J. Zablocki, E. Elzein and R. Kalla, Expert. Opin. Ther. Patents, 2006, 16, 1347. [53] R. V. Kalla, E. Elzein, T. Perry, X. Li, V Palle, V. Varkhedkar, A. Gimbel, T. Maa, D. Zeng and J. Zablocki, J. Med. Chem., 2006, 49, 3682. [54] E. Elzein, R. V. Kalla, X. Li, T. Perry, A. Gimbel, D. Zeng, D. Lustig, K. Leung and J. Zablocki, J. Med. Chem., 2008, 51, 2267. [55] C.-X. Sun, H. Zhong, A. Mohsenin, E. Morschi, J. L. Chunn, J. G. Molina, L. Belardinelli, D. Zeng and M. Blackburn, J. Clin. Invest., 2006, 116, 2173. [56] S. J. Mustafa, A. Nadeem, M. Fan, H. Zhong, L. Belardinelli and D. Zeng, J. Pharmacol. Exp. Ther., 2007, 320, 1246. [57] www.cvt.com press releases. [58] P. G. Baraldi, T. M. Aghazadeh, D. Preti, A. Bovero, R. Romagnoli, F. Fruttarolo, A. N. Zaid, A. R. Moorman, K. Varani, S. Gessi, S. Merighi and P. A. Borea, J. Med. Chem., 2004, 47, 1434. [59] P. G. Baraldi, T. M. Aghazadeh, D. Preti, A. Bovero, R. Romagnoli, F. Fruttarolo, A. R. Moorman, K. Varani and P. A. Borea, Bioorg. Med. Chem. Lett., 2004, 14, 3607. [60] A. Carotti, M. I. Cadavid, N. B. Centeno, C. Esteve, M. I. Loza, A. Martinez, R. Nieto, E. Ravina, F. Sanz, V. Segarra, E. Sotelo, A. Stefanachi and B. Vidal, J. Med. Chem., 2006, 49, 282. [61] M. Aparici, A. Nueda, J. Beleta, N. Prats, R. Fernandez and M. Miralpeix, Proceedings of the Symposium of the Collegium International Allergologicum, 2006, Poster 162. [62] M. Stewart, A. G. Steinig, C. Ma, J.-P. Song, B. McKibben, A. L. Castelhano and S. J. MacLennan, Biochem. Pharmacol., 2004, 68, 305. [63] J. Linden, Trends Pharmacol. Sci., 1994, 15, 298. [64] J. R. Fozard and J. P. Hanon, Clin. Exp. Allergy, 2000, 30, 1213. [65] H. W. J. Young, J. G. Molina, D. Dimina, H. Zhong, M. Jacobson, L.-N. L. Chan, T.-S. Chan and M. R. Blackburn, J. Immunol., 2004, 173, 1380. [66] B. A. Walker, M. A. Jacobson, D. A. Knight, C. A. Salvatore, T. Weir, D. Zhou and T. R. Bai, Am. J. Respir. Cell Mol. Biol., 1997, 16, 531. [67] A.-H. Li, S. Moro, N. Melman, X.-D. Ji and K. A. Jacobson, J. Med. Chem., 1998, 41, 3186. [68] M. Saki, H. Tsumuki, H. Nonaka, J. Shimada and M. Ichimura, Eur. J. Pharmacol., 2002, 444, 133. [69] P. G. Baraldi, B Cacciari, R. Romagnoli, G. Spalluto, K.-N. Koltz, E. Leung, K. Varani, S. Gessi, S. Merighi and P. A. Borea, J. Med. Chem., 1999, 42, 4473. [70] K. Varani, S. Merighi, S. Gessi, K.-N. Koltz, E. Leung, P. G. Baraldi, B. Cacciari, R. Romagnoli, G. Spalluto and P. A. Borea, Mol. Pharmacol., 2000, 57, 968. [71] N. J. Press, R. J. Taylor, J. D. Fulletron, P. Tranter, C. McCarthy, T. H. Keller, L. Brown, R. Cheung, J. Christie, S. Haberthuer, J. D. I. Hatto, M. Keenan, M. K. Mercer, N. E. Press, H. Sahri, A. R. Tuffnell, M. Tweed and J. R. Fozard, Bioorg. Med. Chem. Lett., 2005, 15, 3081. [72] A. Trifilieff, T. H. Keller, N. J. Press, T. Howe, P. Gedeck, D. Beer and C. Walker, Br. J. Pharmacol., 2005, 144, 1002. [73] V.-P. Jaakola, M. T. Griffith, M. A. Hanson, V. Cherezov, E. Y. T. Chien, J. R. Lane, A. P. Ijzerman and R. C. Stevens, Science, 2008, 322, 1211.
CHAPT ER
14 Recent Progress in the Development of Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase Mark D. Wittman, Upender Velaparthi and Dolatrai M. Vyas
Contents
1. Introduction 2. IGF-1R Signaling 3. ATP-Competitive Inhibitors 3.1 5,7-Disubstititued pyrrolopyrimidines 3.2 5-7-Disubstituted pyrrolopyrimidine isosteres 3.3 2,4-Disubstituted pyrimidines 3.4 2,4-Disubstituted pyrimidine isosteres 3.5 Miscellaneous heterocyclic systems 4. Non ATP-Competitive Inhibitors 4.1 Catechols 4.2 Natural lignans 5. Conclusions References
281 282 283 283 284 286 288 290 293 293 295 295 296
1. INTRODUCTION Cancer therapeutics that target growth factor receptors are gaining prominence in the successful treatment of a wide variety of malignancies. Bristol-Myers Squibb Research and Development, 5 Research Parkway, Wallingford, CT 06492-7660, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04414-5
r 2009 Elsevier Inc. All rights reserved.
281
282
Mark D. Wittman et al.
Kinase activation or overexpression is often pivotal in triggering the aberrant signaling pathways typical of the malignant cellular phenotype. The identification of appropriate kinase targets for inhibition with small molecules remains an active area of research for oncology drug discovery. Since the early observations of the mitogenic properties of insulin [1], interest in the insulin-like growth factor-1 receptor (IGF-1R) signaling pathway has been increasing. The convergence of efficacy data for an IGF-1R-specific antibody [2] and epidemiological findings [3] have prominently positioned IGF-1R among the emerging cell signaling pathways currently being explored for cancer therapy. Several approaches to suppress IGF-1R signaling are actively being investigated. These include the use of monoclonal antibodies (mAb) directed against the extracellular ligand-binding domain of the receptor, modulation of the circulating levels of ligand (IGF-1 and IGF-2) through the IGF-binding proteins (IGFBPs), and small molecule kinase inhibitors of IGF-1R. The mAb approach has yielded the first IGF-1R inhibitors to enter clinical development [4–6]. The advantage of the mAbs is their inherent selectivity for IGF-1R over the closely related insulin receptor (IR). However, the lack of cross-reactivity with IR may adversely affect antitumor efficacy since the IR-A isoform exhibits high affinity for IGF-2 and is expressed at high levels in some breast cancers [7]. In addition, the downregulation of IGF-1R by siRNA in breast tumor cell lines sensitizes cells to insulin activation of downstream signaling pathways [8]. The most advanced mAb, CP-751871, is currently in phase III trials. Responses have been reported in advanced adrenocortical cancer and various sarcomas [9]. CP-751871 has also shown responses in non-small cell lung cancer (NSCLC) in combination with paclitaxel and carboplatin [10]. These studies provide proof of concept for the importance of inhibiting IGF-1R signaling in cancer therapy. Several earlier reviews of IGF-1R inhibitors have appeared [11–15]. This chapter will focus on the most recent advances in small molecule inhibitor design and specifically highlight those inhibitors that are entering the early stages of clinical development. Small molecule inhibitors of IGF-1R fall into two sub-categories: those that target the ATP-binding pocket of IGF-1R kinase (ATP-competitive) and those that target the substrate-binding site (non ATP-competitive). The most advanced small molecules are ATP-competitive kinase inhibitors. The greatest challenge for ATP-competitive inhibitors is achieving selectivity versus other kinases, including the closely related IR which shares high overall sequence homology (84%) and complete homology among the residues that contact ATP in the kinase-binding domain [16].
Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase
283
2. IGF-1R SIGNALING IGF-1R is a heterotetramer composed of two extracellular a-subunits that contain the ligand-binding domain and two b-subunits that contain the cytoplasmic kinase domain. Binding of the ligands IGF-1 and IGF-2 to the extracellular domain of the receptor leads to autophosphorylation of the cytoplasmic b-subunit and activation of the intrinsic kinase activity of the receptor. Activation results in the phosphorylation of insulin receptor substrates (IRS-1-4) and Src-homology containing adapter protein (Shc). These in turn activate the PI-3K/Akt/mTOR survival pathway and the mitogenic RAS/Raf/MAPK pathway respectively [17,18]. IGF1R signaling has pleiotropic effects ranging from cell proliferation, differentiation, and migration to regulation of the apoptotic machinery. The crosstalk observed between epidermal growth factor receptor (EGFR) and IGF-1R signaling suggests wide potential for using IGF-1R inhibitors in combination therapy with other targeted agents, cytotoxics, and radiation therapy. IGF-1R activation has also been implicated in the development of resistance toward trastuzumab treatment in breast cancer [19] and lung cancer [20]. mAbs
Insulin
IGF-I
IGF-I IGF-II
pp p Insulin R
p p
pp p
IGF-1R/IR
IGF-1R
IRS PI3K Akt mTOR Survival
Shc Grb2/SOS Ras MAPK Proliferation
284
Mark D. Wittman et al.
3. ATP-COMPETITIVE INHIBITORS 3.1 5,7-Disubstititued pyrrolopyrimidines The 5,7-disubstituted pyrrolopyrimidines were among the first chemotypes described as inhibitors of IGF-1R. NVP-AEW-541, 1 [21], and NVPADW742, 2 [22], are the most studied members of this class. To date, no structure-activity relationship (SAR) studies have been published for this series, but additional analogs are exemplified in a patent application [23]. Pyrrolopyrimidine 1 is a 150 nM(IC50) inhibitor of IGF-1R kinase and is equipotent versus IR. Despite the lack of selectivity in the in vitro kinase assay, 1 is 27-fold selective over IR in a cellular context. The authors suggest that the cellular selectivity arises from conformational differences that exist between the native forms of the enzymes in the cellular context, which are not present in the recombinant enzymes [21]. It remains to be seen if the observed cellular selectivity will translate into a clinical benefit compared to other non-selective inhibitors of IGF-1R/IR signaling. This compound has been extensively studied in vitro and in preclinical animal models [24–26] and was reported to have entered phase I clinical trials in 2004, but nothing has been published to date on the clinical findings.
O NH2
R=
N 1
N N
N
R=
N 2
R
3.2 5-7-Disubstituted pyrrolopyrimidine isosteres 3.2.1 Imidazopyrazines Several isosteric replacements of the pyrrolopyrimidine core have been investigated by various groups. In general, these isosteric scaffolds possess the same cellular selectivity described for the pyrrolopyrimidines
Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase
285
(Section 3.1). Included among these scaffolds are the imidazopyrazines [27–29]. Initial lead optimization efforts focused on replacing the benzyl ether present in 1 and 2 to improve metabolic stability and potency. These studies identified an advanced lead, PQIP, 3 (IC50 ¼ 24 nM) [30]. In a Geo colon carcinoma model, 3 is able to inhibit 80% of IGF-1R phosphorylation within the tumor resulting in 70–80% inhibition of tumor growth (%TGI) after a daily oral dose of 25 mg/kg [31]. Further absorption, distribution, metabolism, and excretion (ADME) optimization led to the clinical candidate OSI-906, 4 (IC50 ¼ 35 nM) with a high degree of selectivity over other kinase targets. Imidazopyrazine 4 shows a wide range of anti-proliferative effects in colorectal, NSCLC, breast, pancreatic, and rhabdomyosarcoma with TGI ranging from 53% to W100% in preclinical models at oral doses of 30–60 mg/kg [32].
N
N NH2
NH2 N
N
N
N
N
N
3
4 OH
N
N
3.2.2 Pyrazolopyrimidines From a series of pyrazolopyrimidine inhibitors, A-928605, 5, was identified with both IGF-1R (IC50 ¼ 35 nM) and EGFR activity (IC50 ¼ 65 nM) [33]. The dual IGF-1R/EGF activity was further optimized to take advantage of the crosstalk between IGF-1R and EGFR, leading to compound 6, which represents the optimal balance of IGF-1R (IC50 ¼ 81 nM), EGF (IC50 ¼ 58 nM), and ErB-2 (IC50 ¼ 54 nM) activity, cellular activity, and pharmacokinetics [34].
286
Mark D. Wittman et al.
3.2.3 Pyrrolotriazines The pyrrolotriazine scaffold, 7, has also been used to optimize for IGF-1R potency. A patent application specifically claiming these compounds as IGF-1R inhibitors has appeared, although no IGF-1R inhibitory data is disclosed [35].
R1 HN
HN
HN N
NH2
NH2
NH2
N
N
N N N
N
N MeO
Cl
N
N
N
N
N
N R2
5
6 N
7 HN
O
OMe
3.3 2,4-Disubstituted pyrimidines A number of inhibitors utilize the pyrimidine scaffold with various substitutions at the 2 and 4 positions. Pyrimidine 8 is representative of one such series with IGF-1R activity (IC50o50 nM) [36,37]. From this series, XL-228 (structure not disclosed) has advanced into the clinic. XL-228 is a multi-targeted protein kinase inhibitor with singledigit nanomolar activity reported for IGF-1R, IR, Src, AurA/B, Fak, FGFR1,2,3 (fibroblast growth factor receptor 1,2,3), and BCR-Abl. Ph+CML and Ph+ALL patients were administered a 1-h intravenous infusion of XL-228 at a dose of 10.8 mg/kg on a once-weekly schedule. The dose-limiting toxicities observed included hyperglycemia and syncope [38]. A patent has been filed specifically claiming compound 9, which has an IC50 of 4.3 nM versus IGF-1R with eightfold selectivity over the IR [39]. The aminopyrazole element is common to both 8 and 9 and most likely forms a hydrogen bond triad with the hinge region of the kinase.
287
Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase
NH
N
N
NH N
HN
HN
N O
N HN
N
N
N O
N H
MeO
N
N
N
8
9
N
3-Aminoquinoline-containing pyrimidines are also claimed as IGF-1R inhibitors. Pyrimidine 10 is equipotent against IGF-1R and IR (IC50 ¼ 25 nM) and demonstrated 33% tumor growth inhibition at 100 mg/kg in a Calu6 tumor xenograft model. Despite the lack of in vitro selectivity over IR, no significant effects on blood glucose levels were observed following insulin and glucose challenge [40]. The closely related compound 11 is a 120 nM (IC50) inhibitor of IGF-1R [41]. N N
O
HN
H N
O R=
N N
N H
R
N H
10
11
The 2-aminoimidazole-substituted pyrimidine, 12, has modest IGF-1R activity (IC50 ¼ 150 nM) [42]. The related pyrimidine, TEA 226, 13, is described as a dual inhibitor of IGF-1R and FAK with inhibitory effects on mTOR signaling in esophageal cancer cells indicating a potential application in esophageal cancer [43]. O N
N H N H
HN
O H N
N N
HN
O
Cl
N N
N H
N 12
N H 13
O
288
Mark D. Wittman et al.
Imidazo[1,2-a]pyridine inhibitor 14 was identified as a screening hit with modest IGF-1R activity (IC50 ¼ 180 nM). C-2 optimization coupled with the reversal of the amide bond connectivity culminated in GSK1904529A, 15, which is equipotent against IGF-1R and IR in a receptor autophosphorylation assay. The compound has 35–124 nM (IC50) potency in cell lines representing multiple tumor types and is orally bioavailable in rat, dog, and monkey [44,45].
F
F O
N
N
H N
N N H F
F
N
H
N
O MeO
N H
H N
N
N
14
N OMe
15
N N
N
N O
S
O
3.4 2,4-Disubstituted pyrimidine isosteres The pyrrolotriazine BMS-754807, 16, has recently been presented as a 2 nM (IC50) inhibitor of IGF-1R with no selectivity over IR and is orally active in a transgenically derived, IGF-1R-driven, IGF-1R Sal tumor model at a dose of 3 mg/kg [46]. The compound is also orally active at 3 mg/kg in the IGF-1R-driven sarcoma model, Rh41, and the Geo colon carcinoma model at 12 mg/kg. The combination of 16 plus cetuximab (EGFR inhibitor) is therapeutically synergistic. Initial single ascending dose studies in normal healthy volunteers demonstrated good bioavailability and tolerability. Further clinical evaluation is ongoing [47]. A similarly substituted triazine, 17, is described in the patent literature which inhibits 96% of tumor growth in the IGF-Sal tumor model at a 3 mg/kg oral dose [48].
289
Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase
N
NH
N
HN
NH
HN O
H N
N N
O N N
N
N
H N
N
N F
N
N
N N
O
16
17
OMe
Several pyrrolopyrimidine-based inhibitors of IGF-1R have also been described. The 4,6-bis-anilino-1H-pyrrolo[2,3,-d]pyrimidine, 18, has an IC50 of 5 nM in an enzymatic assay and an IC50 of 109 nM in a cellular assay of IGF-1R phosphorylation [49]. Further optimization at C-5u provided compounds with single-digit nanomolar inhibition of IGF-1R in enzymatic and cellular assays. Pyrrolopyrimidine 19 is a potent IGF-1R inhibitor (enzyme IC50 ¼ 2 nM, cellular IGF-1R phosphorylation IC50 ¼ 85 nM), W1,000-fold selective over the JNK1 and JNK3 kinases, and has 98% oral bioavailability in rats [50]. The N,N,-dimethyl glycinamide forms a hydrogen bond with the backbone NH of Asp1056. Indoline 20 was designed to prevent the acid-mediated cyclization of the pyrimidine moiety onto the pendant carboxamide observed with 18 and 19. This compound maintains enzymatic and cellular potency while improving chemical stability (t1/2 ¼ W1,000 h at 231C in 0.1N HCl) [51].
NH2
F
NH2
O
O R= R
HN
HN
N
N
19 HN MeO
N H
N
18
HN
N H
N
MeO R=
O
N N N
N
NMe2
20
290
Mark D. Wittman et al.
3.5 Miscellaneous heterocyclic systems 3.5.1 Benzimidazoles Benzimidazole-pyridones were among the early small molecule chemotypes described as ATP-competitive inhibitors of IGF-1R. The initial screening hit was optimized for potency and Cytochrome p-450 (CYP) inhibition to provide an early lead structure, BMS-536924, 21 [52–60]. This compound inhibits both IGF-1R and IR with equal potency (IC50 ¼ 120 nM), is selective versus other kinases, inhibits the phosphorylation of Akt and MAP kinase (MAPK) in cells, and blocks proliferation in a wide variety of human cancer cell lines including colon, breast, lung, pancreas, prostate, sarcoma, and multiple myeloma (IC50’s of 110–460 nM). Tumor growth inhibition is observed in vivo when dosed orally in the IGF-1R Sal tumor model [46] and in a broad range of human tumor xenografts such as Colo205, Geo, and RD1 (50–100 mg/kg). Oral bioavailability is observed across all species, and a twofold window between antitumor efficacy and glucose elevation at the efficacious dose was reported [55]. Benzimidazole 21 reverses IGF-1R-induced transformation of mammary epithelial cells, blocks proliferation, and restores apical-basal polarity in MCF-7 cells [56]. The potential for CYP inhibition, time-dependent CYP inhibition, and pregnane X receptor (PXR) transactivation was reduced by replacing the morpholine ring in 21 with a C-linked piperidine. Combining this modification with the chloropyrazole side chain [57] led to the discovery of BMS-695735, 22, which demonstrates in vivo efficacy in the IGF-Sal, Colo205, Geo colon carcinoma, and JJN3 multiple myeloma models when administered orally at doses between 50 mg/kg and 100 mg/kg [59]. O N
H N
F
O
N
O
H N
NH
N
NH
N HN
HN
HO N N Cl
21
Cl
22
3.5.2 Bicyclic pyrazoles The patent literature also describes bicyclic pyrazole inhibitors of IGF-1R. These structures represent a significant departure from other reported
291
Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase
leads. Limited IGF-1R activity is presented in these patents with 23 being the most potent example reported (IC50 ¼ 49 nM) [61–63].
N N H N
N HN
O
HN
N H O
N
O S O
23
F
F
3.5.3 Ureas One of the most promising diarylurea (DAU) inhibitors is PQ401, 24, which inhibits the autophosphorylation of IGF-1R in human cultured MCF-7 cells with an IC50 of 12 mM. Treatment of MCNeuA cells implanted into mice with 24 reduced tumor growth when dosed three times per week intraperitoneally [64]. Lead optimization around a series of 3,5-disubstituted 1H-pyrrolo[2,3-b]pyridines led to the identification of compound 25, which shows potent in vitro kinase activity (IC50 ¼ 21 nM) and inhibits IGF-1R phosphorylation in cells (IC50 ¼ 68 nM) [65].
O H N
H N
N
O
O
H N
H N
N O
O
24
Cl
25 N
N H
292
Mark D. Wittman et al.
3.5.4 Pyrrolocarboxaldehydes Pyrrolocarboxaldehydes have been disclosed as monocyclic ATPcompetitive inhibitors of IGF-1R [66]. Aldehyde 26 is modestly selective versus IR in enzymatic and cell-based assays (IGF-1R, IC50 ¼ 490 nM; IR, IC50 ¼ 2 mM) and forms a reversible, covalent adduct with the kinase active site. CHO HN
O
OEt O O
26
3.5.5 Quinolines IGF-1R inhibitors have also been built around the quinoline core structure. Optimization of the cyanoquinoline template provided 27 with potent IR (IC50 ¼ 2 nM) and IGF-1R (IC50 ¼ 9 nM) activity as well as activity in a cellular myloid assay with an IC50 of 90 nM [67]. Cl S
N
H N
R2
N HN MeO
R1
O
CN NH
O
N
O R1
Br
28
N
R2
N
N
29 27
O
N
N
N H
The isoquinolinedione 28 was identified as an initial micromolar hit that binds at the ATP-binding site in a similar mode to the
293
Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase
benzimidazoles described above (Section 3.5.1). Optimization of R1 and R2 culminated in compound 29 (IGF-1R, IC50 ¼ 319 nM), which is equipotent against IR and has improved selectivity over cyclindependent kinase (CDK)-4 [68].
4. NON ATP-COMPETITIVE INHIBITORS The discovery and development of IGF-1R selective, non ATP-competitive inhibitors has been slow relative to ATP-competitive inhibitors due to the inherent medicinal chemistry challenges involved in optimizing potency for the more open substrate-binding site, requiring the use of peptide or peptidomimetic elements. To date, the two main classes of non ATP-competitive inhibitors are catechols and naturally occurring lignans.
4.1 Catechols A systematic discovery effort to design more potent and selective IGF-1R substrate inhibitors commenced with the screening of the ‘‘tyrphostins’’ (Tyrosine Phosphorylation Inhibitors) that are synthetic catechols previously shown to inhibit EGFR [69–71]. These efforts identified biscatechols 30 [72,73] and 31 [74]. Tyrphostin 30 has activity against IGF-1R (IC50 ¼ 61 nM), IR (IC50 ¼ 113 nM), and EGFR (IC50 ¼ 370 nM) receptor kinases. Based on the published X-ray structure of the kinase domain of IR, the catechol rings in 30 function as phosphate bioisosteres of phosphotyrosines 1158 and 1162 [71]. The novel tertiary amine catechol 31 inhibited the IGF-1R receptor with an IC50 of 170 nM in a cell-free kinase assay (inhibition of polyTyrGlu (pGT) phosphorylation catalyzed by IGF-1 receptor) and exhibited inhibitory IC50 values in the range 4–6 mM in a colony formation assay in soft agar for three cell lines (MCF-7, LNCap, PC-3). HO
O
N
OH
HO
HO
OH OH
CN OH
HO
O O
30
31
SAR and lead optimization efforts to mitigate metabolic liabilities due to the catechol rings led to some moderately potent benzoxazolone
294
Mark D. Wittman et al.
analogs 32a–c [73]. No selectivity over IR was observed. Replacement of both catechols with benzoxazolone rings led to inactive compounds.
R1
32a
R2
32b
CN 32
IGF-1R IC50
O
OH
N H
OH
~370 nM
O
O R1
R2
H N
OH
O
OH
~430 nM
O H N
HO 32c
~600 nM O
O
HO
To date, none of the synthetic catechols or their close analogs discussed above have progressed into clinical development. The naturally occurring bis-catechol, nordihydroguaiaretic acid, 33 (NDGA, INSM-18), has advanced to phase II clinical trials using continuous oral dosing for the treatment of prostate cancer [75–77]. NDGA inhibits IGF1R phosphorylation of a synthetic nonspecific tyrosine kinase substrate and proliferation of MCF-7 breast cell line with an IC50 of 0.9 and 24.6 mM respectively. Some early positive clinical data is now emerging from the prostate trial with respect to reduction in prostate-specific antigen (PSA) levels and delay in PSA doubling time in patients. Doses up to 2,500 mg/ day are tolerated with minimal toxic effects. NDGA has Her-2 receptor kinase and 15-lipoxygenase inhibitory activity. Although there is no direct mention of 33 being a non ATP-competitive IGF-1R inhibitor, it is a non ATP-competitive inhibitor of FGFR3 tyrosine kinase [78].
OH HO OH HO 33
Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase
295
4.2 Natural lignans The naturally occurring lignan picropodophyllotoxin (PPP) or AXL-1717, 34, has recently advanced into human clinical trials (oral dosing) [79,80]. The epimeric podophyllotoxin 35 (PPT), a cytotoxic agent known for its potent antimitotic activity, is also reported to be an IGF-1R selective substrate inhibitor. Both 34 and 35 are reported to inhibit IGF-1R catalyzed substrate phosphorylation of pGT with an IC50 value of 6 nM. Antiproliferative IC50 values against 11 cell lines for both compounds are between 20 nM and 25 nM. In preclinical models, 34 is efficacious when dosed intraperitoneally in a wide range of tumor models such as breast, prostate, malignant melanoma, and multiple myeloma. Results from cell culture studies using IGF-1R-deficient cell lines, mouse embryonic fibroblasts (MEFs), and HepG2 cells have called into question whether the observed antitumor activity is due to IGF-1R inhibition [81,82]. OH
OH
O
O O
O
O O
O
H3CO
OCH3
O
H3CO
OCH3
OCH3
OCH3
34
35
5. CONCLUSIONS Significant progress has been made in developing small molecule inhibitors of IGF-1R for use in the clinic. Coupled with the advances being made toward exploiting and validating the IGF-1R target using mAbs, the stage is set to determine the clinical potential of IGF-1R inhibitors. It will be interesting to follow the clinical development of molecules such as AXL-1717, BMS-754807, INSM-18, OSI-906, and XL-228 with respect to efficacy, safety, and tolerability. The hope remains that small molecule inhibitors will provide a complementary approach to mAb therapeutics in terms of efficacy and dosing flexibility, particularly, in combination studies with cytotoxics and EGFR antagonists.
296
Mark D. Wittman et al.
REFERENCES [1] J. J. Elias, Proc. Soc. Exp. Biol. Med., 1959, 101, 500. [2] C. L. Arteaga, L. J. Kitten, E. B. Coronado, S. Jacobs, F. C. Kull, Jr., D. C. Allred, C. K. Osborne and C. Kent, J. Clin. Invest, 1989, 84, 1418. [3] J. M. Chan, M. J. Stempfer, E. Giovannucci, P. H. Gann, J. Ma, P. Wilkinson, C. H. Hennekens and M. Pollak, Science, 1998, 279, 563. [4] J. Rodon, V. De Santos, R. J. Ferry, Jr. and R. Kurzrock, Mol. Cancer Ther., 2008, 7, 2575. [5] M. Pollak, Nat. Rev. Cancer, 2008, 8, 915. [6] M. Pollak, Curr. Opin. Pharm., 2008, 8, 384. [7] F. Frasca, G. Pandini, P. Scalia, L. Sciacca, R. Mineo, A. Costantino, I. D. Goldfine, A. Belfiore and R. Vigneri, Mol. Cell Biol., 1999, 19, 3278. [8] H. Zhang, A. M. Pelzer, D. T. Kiang and D. Yee, Cancer Res., 2007, 67, 391. [9] D. Olmos, R. Molife, S. Okuno, F. Worden, G. Hammer, T. Yap, H. Shaw, S. Schuetze, L. Roberts, A. Gualberto, J. de-Bono and P. Haluska, AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics, San Francisco, CA, October 22–26, 2007, Abstract A63. [10] D. D. Karp, L. G. Paz-Ares, L. J. Blakely, H. Kreisman, P. D. Eisenberg, R. B. Cohen, L. Garland, C. J. Langer, C. L. Melvin and A. Gualberto, 43rd American Society of Clinical Oncology, June, 2007, Abstract 7506. [11] M. Hewish, I. Chau and D. Cunningham, Recent Pat. Anticancer Drug Discov., 2009, 4, 54. [12] P. K. S. Sarma, R. Tandon, G. Praful, S. G. Dastidar, A. Ray, B. Das and I. A. Cliffe, Expert Opin. Ther. Patents, 2007, 17, 25. [13] R. D. Hubbard and J. L. Wilsbacher, Chem. Med. Chem., 2007, 2, 41. [14] C. Garcia-Echeverria, Idrugs, 2006, 9, 415. [15] F. Hofmann and C. Garcia-Echeverria, Drug Discov. Today, 2005, 10, 1041. [16] S. Favelyukis, J. H. Till, S. R. Hubbard and W. T. Miller, Nat. Struct. Biol., 2001, 8, 1058. [17] E. Y. Skolnik, C. H. Lee, A. Batzer, L. M. Vicentini, M. Zhoa, R. Daly, M. J. Myers, Jr., J. M. Becker, A. Ulrich, et al., EMBO J., 1993, 12, 1929. [18] B. Vanhaesebroeck and D. R. Alessi, J. Biochem., 2000, 346, 561. [19] A. Camirand, M. Zakikhani, F. Young and M. Pollak, Breast Cancer Res., 2005, 7, R570. [20] F. Morgillo, J. K. Woo, E. S. Kim, W. K. Hong and H. Y. Lee, Cancer Res., 2006, 66, 10100. [21] C. Garcia-Echeverria, M. A. Pearson, A. Marti, T. Meyer, J. Mestan, J. Zimmermann, J. Gao, J. Brueggen, H.-G. Capraro, R. Cozens, D. B. Evans, D. Fabbro, P. Furet, D. G. Porta, J. Liebetanz, G. Martiny-Baron, S. Ruetz and F. Hofmann, Cancer Cell, 2004, 5, 231. [22] C. S. Mitsiades, N. S. Mitsiades, C. J. McMullan, V. Poulaki, R. Shringarpure, M. Akiyama, T. Hideshima, D. Chauhan, M. Joseph, T. A. Libermann, C. Garcia-Echeverria, M. A. Pearson, F. Hofmann, K. C. Anderson and A. L. Kung, Cancer Cell, 2004, 5, 221. [23] H.-G. Capraro, P. Fuuret and C. Garcia-Echeverria, WO Patent Application 2004/043962, 2004. [24] K. Scotlandi, M. C. Manara, G. Nicoletti, P. L. Lollini, S. Lukas, S. Benini, S. Croci, S. Perdichizzi, D. Zambelli, M. Serra, C. Garcia-Echeverria, F. Hofmann and P. Picci, Cancer Res., 2005, 65, 3868. [25] B. Tanno, C. Mancini, R. Vitali, M. Mancuso, H. P. McDowell, C. Dominici and G. Raschella, Clin. Cancer Res., 2006, 12, 6772. [26] M. C. Manara, L. Landuzzi, P. Nanni, G. Nicoletti, D. Zambelli, P. L. Lollini, C. Nanni, F. Hofmann, C. Garcia-Echeverria, P. Picci and K. Scotlandi, Clin. Cancer Res., 2007, 13, 1322. [27] M. J. Mulvihill, Q.-S. Ji, D. Werner, P. Beck, C. Cesario, A. Cooke, M. Cox, A. Crew, H. Dong, L. Feng, K. W. Foreman, G. Mak, A. Nigro, M. O’Connor, L. Saroglou,
Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase
[28]
[29] [30]
[31]
[32] [33]
[34]
[35]
[36]
[37] [38]
[39] [40] [41] [42] [43]
[44]
[45]
297
K. M. Stolz, I. Sujka, B. Volk, Q. Weng and R. Wilkes, Bioorg. Med. Chem. Lett., 2007, 17, 1091. L. D. Arnold, C. Cesario, H. Coate, A. P. Crew, H. Dong, K. Foreman, A. Honda, R. Laufer, A.-H. Li, K. M. Mulvihill, M. J. Mulvihill, A. Nigro, B. Panicker, A. G. Steinig, Y. Sun, Q. Weng, D. S. Werner, M. J. Wyle and T. Zhang, WO Patent Application 2005/ 097800, 2005. A. P. Crew, M. J. Mulvihill and D. S. Werner, US Patent Application 2006/019957, 2006. M. J. Mulvihill, Q.-S. Ji, H. R. Coate, A. Cooke, H. Dong, L. Feng, K. Foreman, M. Rosenfeld-Franklin, A. Honda, G. Mak, K. M. Mulvihill, A. I. Nigro, M. O’Connor, C. Pirrit, A. G. Steinig, K. Siu, K. M. Stolz, Y. Sun, P. A. R. Tavares, Y. Yao and N. W. Gibson, Bioorg. Med. Chem., 2008, 16, 1359. Q.-S. Ji, M. J. Mulvihill, M. Rosenfeld-Franklin, A. Cooke, L. Feng, G. Mak, M. O’Connor, Y. Yao, C. Pirritt, E. Buck, A. Eyzaguirre, L. D. Arnold, N. W. Gibson and J. A. Pachter, Mol. Cancer Ther., 2007, 6, 2158. M. Mulvihill, ACS Prospectives Conference PK/PD for Medicinal Chemists, Boston, MA, September 8, 2008. R. D. Hubbard, N. Y. Bamaung, S. D. Fidanze, S. A. Erickson, R. A. Manteil, P. Kovar, G. T. Wang, J. Wang, G. S. Sheppard and R. L. Bell, Abstract 566, Eur. J. Cancer Suppl., 2008, 6(October), 176. R. D. Hubbard, N. Y. Bamaung, S. D. Fidanze, S. A. Erickson, F. Palazzo, J. L. Wilbacher, Q. Zhang, L. A. Tucker, X. Hu, P. Kovar, D. J. Osterling, E. F. Johnson, J. Bouska, J. Wang, S. K. Davidsen, R. L. Bell and G. S. Sheppard, Bioorg. Med. Chem. Lett., 2009, 19, 1718. S. J. O’Connor, J. Dumas, W. Lee, J. Dixon, D. Cantin, D. Gunn, J. Burke, B. Phillips, D. Lowe, T. Shelekhin, G. Wang, X. Ma, S. Ying, A. Mcclure, F. Achebe, M. Lobell, F. Ehrgott, C. Iwuagwu and K. Parcella, WO Patent Application 2007/056170, 2007. J. Chen, L. E. Dalrymple, S. Epshteyn, T. P. Forsyth, T. P. Huynh, M. A. Ibrahim, J. W. Leahy, G. L. Lewis, G. Mann, L. W. Mann, R. T. Noguchi, B. H. Ridgway, J. C. Sangalang, K. L. Schnepp, X. Shi, C. S. Takeuchi, M. A. Williams, J. Nuss and A. K. Cheung, WO Patent Application 2006/074057, 2006. W. Zhang, WO Patent Application 2008/005538, 2008. J. Cortes, R. Paquette, M. Talpaz, J. Pinilla-Ibarz, K. El-Shami, M. Wetzler, J. Lipton, C. Kasap, L. A. Bui, P. Woodard, O. O Clary and N. Shah, 50th Annual Meeting of the American Society of Hematology, December 6–9, 2008, Abstract 3232. T. Nowak, S. C. Purkiss and A. P. Thomas, WO Patent Application 2008/117051, 2008. J.-C. Harmange, J. L. Buchanan, S. Chaffee, P. M. Novak, S. Van Der Plas and X. Zhu, US Patent 2005/6939874, 2005. T. Heinrich, A. Blaukat, and M. Kordowicz, WO Patent Application 2006/108487, 2006. T. Heinrich, A. Blaukat and M. Kordowicz, US Patent Application 2008/0194605, 2008. Z. G. Wang, T. Fukazawa, T. Nishikawa, N. Watanabe, K. Sakurama, T. Motoki, M. Takaoka, S. Hatakeyama, O. Omori, T. Ohara, S. Tanabe, Y. Fujiwara, Y. Shirakawa, T. Yamatsuji, N. Tanaka and Y. Naomoto, Oncol. Rep., 2008, 20, 1473. K. A. Emmitte, B. J. Wilson, E. W. Baum, H. K. Emerson, K. W. Kuntz, K. E. Nailor, J. M. Salovich, S. C. Smith, M. Cheung, R. M. Gerding, K. L. Stevens, D. E. Uehling, R. A. Mook, Jr., G. S. Moorthy, S. H. Dickerson, A. M. Hassell, M. A. Leesnitzer, L. M. Shewchuk, A. R. Groy, J. L. Rowand, K. Anderson, C. L. Atkins, J. Yang, P. Sabbatini and R. Kumar, Bioorg. Med. Chem. Lett., 2009, 19, 1004. P. Sabbatini, J. L. Rowand, A. Groy, S. Korenchuk, Q. Liu, C. Atkins, M. Dumble, J. Yang, K. Anderson, B. J. Wilson, K. A. Emmitte, S. K. Rabindran and R. Kumar, Clin. Cancer Res., 2009, 15, 3058.
298
Mark D. Wittman et al.
[46] J. M. Carboni, A. V. Lee, D. L. Hadsell, B. R. Rowley, F. Y. Lee, D. K. Bol, A. E. Camuso, M. Gottardis, A. F. Greer, C. P. Ho, W. Hurlburt, A. Li, M. Saulnier, U. Velaparthi, C. Wang, M.-L. Wen, R. A. Westhouse, M. Wittman, K. Zimmermann, B. A. Rupnow and T. W. Wong, Cancer Res., 2005, 65, 3781. [47] M. Wittman, J. Carboni, Z. Yang, F. Lee, G. Cantor, M. Antman, R. Attar, P. Balimane, C. Chen, S. Cheng, L. Discenza, C. Fairchild, F. G. Finckenstein, D. Frennesson, M. Gottardis, A. Greer, X. Gu, W. Hurlburt, A. Li, J. Li, P. Liu, W. Johnson, D. Langley, H. Mastalarz, A. Mathur, K. Menard, K. Patel, J. Sack, X. Sang, M. Saulnier, K. Stefanski, S. Traeger, G. Trainor, U. Velaparthi, S. Yeola, G. Zhang, K. Zimmerman and D. Vyas, MEDI-163, 235th ACS National Meeting, Salt Lake City, UT, March, 2009. [48] U. Velaparthi, P. Liu, M. D. Wittman and D. R. Langley, WO Patent Application 2009/ 015254, 2009. [49] S. D. Chamberlain, J. W. Wilson, F. Deanda, S. Patnaik, A. M. Redman, B. Yang, L. Shewchuk, P. Sabbatini, M. A. Leesnitzer, A. Groy, C. Atkins, R. Gerding, M. Hassell, H. Lei, R. A. Mook, Jr., G. Moorthy, J. L. Rowand, K. L. Stevens, R. Kumar and J. B. Shotwell, Bioorg. Med. Chem. Lett., 2009, 19, 469. [50] S. D. Chamberlain, A. M. Redman, J. W. Wilson, F. Deanda, J. B. Shotwell, R. Gerding, H. Lei, B. Yang, K. L. Stevens, A. M. Hassell, L. M. Shewchuk, M. A. Leesnitzer, J. L. Smith, P. Sabbatini, C. Atkins, A. Groy, J. L. Rowand, R. Kumar, R. A. Mook, Jr., G. Moorthy and S. Patnaik, Bioorg. Med. Chem. Lett., 2009, 19, 360. [51] S. D. Chamberlain, A. M. Redman, S. Patnaik, K. Brickhouse, Y. C. Chew, F. Deanda, R. Gerding, H. Lei, G. Moorthy, M. Patrick, K. L. Stevens, J. W. Wilson and J. B. Shotwell, Bioorg. Med. Chem. Lett., 2009, 19, 373. [52] M. D. Wittman, B. Balasubramanian, K. Stoffan, U. Velaparthi, P. Liu, S. Krishnananthan, J. Carboni, A. Li, A. Greer, R. Attar, M. Gottardis, C. Chang, B. Jacobson, Y. Sun, S. Hansel, M. Zoeckler and D. Vyas, Bioorg. Med. Chem. Lett., 2007, 17, 974. [53] U. Velaparthi, M. Wittman, P. Liu, K. Stoffan, K. Zimmermann, X. Sang, J. Carboni, A. Li, R. Attar, M. Gottardis, A. Greer, C. Y. Chang, B. L. Jacobsen, J. S. Sack, Y. Sun, D. R. Langley, B. Balasubramanian and D. Vyas, Bioorg. Med. Chem. Lett., 2007, 17, 2317. [54] U. Velaparthi, P. Liu, B. Balasubramanian, J. Carboni, R. Attar, M. Gottardis, A. Li, A. Greer, M. Zoeckler, M. D. Wittman and D. Vyas, Bioorg. Med. Chem. Lett., 2007, 17, 3072. [55] M. Wittman, J. Carboni, R. Attar, B. Balasubramanian, P. Balimane, P. Brassil, F. Beaulieu, C. Chang, W. Clarke, J. Dell, J. Eummer, D. Frennesson, M. Gottardis, A. Greer, S. Hansel, W. Hurlburt, B. Jacobson, S. Krishnanathan, F. Y. Lee, A. Li, T. A. Lin, P. Liu, C. Ouellet, X. Sang, M. G. Saulnier, K. Stoffan, Z. Yang, K. Zimmermann, M. Zoeckler and D. Vyas, J. Med. Chem., 2005, 48, 5639. [56] B. C. Litzenburger, H. J. Kim, I. Kuiaste, J. M. Carboni, R. M. Attar, M. M. Gottardis, C. R. Fairchild and A. V. Lee, Clin. Cancer Res., 2009, 15, 226. [57] M. G. Saulnier, D. B. Frennesson, M. D. Wittman, K. Zimmermann, U. Velaparthi, D. R. Langley, C. Struzynski, X. Sang, J. Carboni, A. Li, A. Greer, Z. Yang, P. Balimane, M. Gottardis, R. Attar and D. Vyas, Bioorg. Med. Chem. Lett., 2007, 18, 1702. [58] K. Zimmermann, M. D. Wittman, M. G. Saulnier, U. Velaparthi, D. R. Langley, X. Sang, D. Frennesson, J. Carboni, A. Li, A. Greer, M. Gottardis, R. M. Attar, Z. Yang, P. Balimane, L. N. Discenza and D. Vyas, Bioorg. Med. Chem. Lett., 2008, 18, 4075. [59] U. Velaparthi, M. Wittman, P. Liu, J. M. Carboni, F. Y. Lee, R. Attar, P. Balimane, W. Clarke, M. W. Sinz, W. Hurlburt, K. Patel, L. Discenza, S. Kim, M. Gottardis, A. Greer, A. Li, M. Saulnier, Z. Yang, K. Zimmermann, G. Trianor and D. Vyas, J. Med. Chem., 2008, 51, 5897. [60] P. Haluska, J. M. Carboni, D. A. Loegering, F. Y. Lee, M. Wittman, M. G. Saulnier, D. B. Frennesson, K. R. Kalli, C. A. Conover, R. M. Attar, S. H. Kaufmann, M. Gottardis and C. Erlichman, Cancer Res., 2006, 66, 362.
Small Molecule Inhibitors of Insulin-Like Growth Factor-1 Receptor Kinase
299
[61] T. Bandiera, A. L. Borgia, P. Polucci, M. Villa, M. Nesi, M. Angiolini and M. Varsi, WO Patent Application 2007/068619, 2007. [62] T. Bandiera, A. L. Borgia, S. C. Orrenius, E. Perrone, I. Beria, D. Fancelli and A. Galvani, WO Patent Application 2007/068637, 2007. [63] T. Bandiera, E. Perrone, A. L. Borgia and M. Varasi, WO Patent Application 2007/099166, 2007. [64] K. L. Gable, B. A. Maddux, C. Penaranda, M. Zavodovskaya, M. J. Campbell, M. Lobo, L. Robinson, S. Schow, J. A. Kerner, I. D. Goldfine and J. F. Youngren, Mol. Cancer Ther., 2006, 5, 1079. [65] S. Patnaik, K. L. Stevens, R. Gerding, F. Deanda, J. B. Shotwell, J. Tang, T. Hamajima, H. Nakamura, M. A. Leesnitzer, A. M. Hassell, L. M. Shewchuck, R. Kumar, H. Lei and S. D. Chamberlain, Bioorg. Med. Chem. Lett., 2009, 19, 3136. [66] I. M. Bell, S. M. Stirdivant, J. Ahern, J. C. Culberson, P. L. Darke, C. J. Dinsmore, R. A. Drakas, S. N. Gallicchio, S. L. Graham, D. C. Heimbrook, D. L. Hall, J. Hua, N. R. Kett, A. S. Kim, M. Kornienko, L. C. Kuo, S. K. Munshi, A. G. Quigley, J. C. Reid, B. W. Trotter, L. H. Waxman, T. M. Williams and C. B. Zartman, Biochemistry, 2005, 44, 9430. [67] L. M. Miller, S. C. Mayer, D. M. Berger, D. H. Boschelli, F. Boschelli, L. Di, X. Du, M. Dutia, M. B. Floyd, M. Johnson, C. H. Kenny, G. Krishnamurthy, F. Moy, S. Petusky, D. Tkach, N. Torres, B. Wu and W. Xu, Bioorg. Med. Chem. Lett., 2009, 19, 62. [68] S. C. Mayer, A. L. Banker, F. Boschelli, L. Di, M. Johnson, C. H. Kenny, G. Krishnamurthy, K. Kutterer, F. Moy, S. Petusky, M. Ravi, D. Tkach, H. Tsou and W. Xu, Bioorg. Med. Chem. Lett., 2008, 18, 3641. [69] A. Gazit, P. Yaish, C. Gilon and A. Levitzki, J. Med. Chem., 1989, 32, 2344. [70] A. Gazit, N. Osherov, I. Posner, P. Yaish, E. Poradosu, C. Gilon and A. Levitzki, J. Med. Chem., 1991, 34, 1896. [71] G. Blum, A. Gazi and A. Levitzki, Biochemistry, 2000, 39, 15705. [72] A. Levitzki and E. Mishani, Ann. Rev. Biochem., 2006, 75, 93. [73] G. Blum, A. Gazit and A. Levitzki, J. Biol. Chem., 2003, 42, 40442. [74] L. Steiner, G. Blum, Y. Friedmann and A. Levitzki, Eur. J. Pharmacol., 2007, 562, 1. [75] J. F. Youngren, K. Gable, C. Penaranda, B. A. Maddux, M. Zavodovskaya, M. Lobo, M. Campbell, J. Kerner and I. D. Goldfine, Breast Cancer Res. Treat., 2005, 94, 37. [76] J. E. Blecha, M. O. Anderson, J. M. Chow, C. C. Guevarra, C. Pender, C. Penaranda, M. Zavodovskaya, J. F. Youngren and C. E. Berkman, Bioorg. Med. Chem. Lett., 2007, 17, 4026. [77] C. J. Ryan, A. H. Harzstark, J. Rosenberg, A. Lin, C. Claros, I. D. Goldfine, J. F. Kerner and E. J. Small, Brit. J. Urol. Int., 2008, 101, 436. [78] A. N. Meyer, C. W. McAndrew and D. J. Donoghue, Cancer Res., 2008, 68, 7362. [79] O. Larsson and M. Axelson, WO PCT Publication 2002/102804. [80] A. Girnita, L. Girnita, F. Del Prete, A. Bartolazzi, O. Larsson and M. Axelson, Cancer Res., 2004, 64, 236. [81] S. Linder, M. C. Shoshan and R. S. Gupta, Cancer Res., 2007, 67, 2899. [82] O. Larsson and M. Axelson, Cancer Res., 2007, 67, 2899.
CHAPT ER
15 Case History: Discovery of Ixabepilone (IXEMPRATM), a First-in-Class Epothilone Analog for Treatment of Metastatic Breast Cancer Robert M. Borzilleri and Gregory D. Vite
Contents
1. Introduction 2. Epothilone Natural Products 2.1 Fermentation and isolation 2.2 Biosynthesis 2.3 Biological properties and mechanism of action 2.4 Evaluation of epothilones as drug leads 3. Approaches to Identify Drug Candidates 3.1 General strategy 3.2 Total synthesis 3.3 Semisynthesis 4. Preclinical Pharmacology 4.1 Cytotoxicity 4.2 In vivo efficacy 4.3 Profiling 5. Clinical Results 5.1 Phase I/II highlights 5.2 Phase II/III registrational trials 5.3 Pharmacogenomics 6. Future Directions 7. Conclusions References
301 302 302 303 303 304 304 304 306 307 310 311 311 314 314 314 315 316 317 318 319
Bristol-Myers Squibb R&D, P.O. Box 4000, Princeton, NJ 08543-4000 Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04415-7
r 2009 Elsevier Inc. All rights reserved.
301
302
Robert M. Borzilleri and Gregory D. Vite
1. INTRODUCTION Since Food and Drug Administration (FDA) approval in 1991, Taxols (paclitaxel) has been a mainstay of chemotherapy for various cancers including breast, ovarian, and lung cancers [1,2]. Despite the clinical success, shortcomings of this treatment were noted early on, and even today, our understanding of factors that determine whether patients will respond to the drug is still evolving. For these reasons, intense research has been directed toward discovering other drug modalities that take advantage of and improve upon paclitaxel’s unique mechanism of action (i.e., microtubule stabilization). These research efforts include development of new paclitaxel formulations and delivery systems, synthesis of taxane analogs, and exploration of new classes of natural products that kill tumor cells by the same mechanism [3–5]. At Bristol-Myers Squibb (BMS), the initial discovery of promising novel taxane analogs eventually gave way to an in-depth study of two newly discovered natural products, eleutherobin and epothilone [6]. Comparisons of early preclinical activities and the feasibility of re-supply for these two agents led to a decision to pursue the epothilone class. The main objective of this effort was to discover an agent with a broader spectrum of antitumor activity, especially against paclitaxel-resistant tumors and preferably with a reduction of undesired side effects. Accordingly, this account describes the drug discovery path from epothilone natural product to semisynthetic analog ixabepilone (IXEMPRATM), currently FDAapproved for the treatment of metastatic and locally advanced breast cancer.
2. EPOTHILONE NATURAL PRODUCTS 2.1 Fermentation and isolation German chemist Gerhard Ho¨fle and microbiologist Hans Reichenbach of Gesellschaft fu¨r Biotechnologische Forschung (GBF, now the Helmholtz Center for Infection Research) spent much of their careers studying myxobacteria, a unique class of microorganisms. They discovered that a particular strain of the species Sorangium cellulosum (So ce90), cultured from a soil sample collected in South Africa, produced the epothilone natural products as secondary metabolites [7,8]. In general, S. cellulosum bacteria can be readily found in soil and decaying plant material, but culturing these bacteria in the laboratory is not trivial. Epothilone production was found to be highly dependent on strain selection and fermentation media. To facilitate recovery of the epothilones, the fermentation was carried out with XAD-16 resin. Extraction with
Discovery of Ixabepilone (IXEMPRATM)
S
R
R
O
S
12 OH
15
N
303
OH
N
17 O 1 O
O OH
1 (R = H) 2 (R = Me)
O
O
OH
O
3 (R = H) 4 (R = Me)
Figure 1 Epothilones AD.
methanol followed by purification using reverse-phase chromatography and recrystallization afforded the major metabolites epothilones A (1) and B (2) in a 2:1 ratio. A typical fermentation provided approximately 120 mg/L of epothilone A. Strain and process improvements resulted in higher yield and enhanced recovery of epothilone B, which was required for semi-synthesis and manufacture of ixabepilone (vide infra). The molecular structures of epothilones A and B were confirmed through single crystal X-ray analysis by Ho¨fle and co-workers (Figure 1) [8]. In addition, Ho¨fle’s careful analysis of the So ce90 fermentation broth revealed the presence of more than 30 other epothilone-related, minor metabolites [9].
2.2 Biosynthesis The German group established early on that the epothilone biosynthesis occurs through the action of a distinct polyketide synthase (PKS), because a radiolabel (13C) could be incorporated into the natural product through addition of isotope-enriched acetate or propionate to the fermentation medium [10,11]. Furthermore, the thiazolyl starter unit required for initiation of the biosynthesis was reported to be derived from cysteine. Shortly thereafter, the epothilone PKS gene cluster was independently cloned by researchers at Novartis and Kosan Biosciences [12,13]. The latter group demonstrated successful heterologous expression of the PKS in Streptomyces coelicolor and validated production of epothilones in this organism. This work demonstrated that chain extension by the PKS is followed by macrocyclization and release from the PKS to give the 12,13-olefinic epothilones C (3) and D (4) (Figure 1). Subsequently, post-PKS oxidation by a cytochrome P450 affords the corresponding 12,13-oxiranes [14]. Discovery of the P450 gene epoK by Kosan Biosciences led to the development of an epoK-mutant strain of Myxococcus xanthus that could be used for improved recovery of olefinic epothilones [15]. Interestingly, the entire genome of S. cellulosum (closely
304
Robert M. Borzilleri and Gregory D. Vite
related strain So ce56) was sequenced recently and consequently was claimed to be the largest bacterial genome sequencing to date [16].
2.3 Biological properties and mechanism of action At the time of the initial discovery of the epothilones, it was determined that these natural products possess potent antifungal activity. This biological activity and implications for agricultural applications were briefly explored by GBF and collaborators. However, Bollag and co-workers later reported that the epothilones were potent cytotoxic agents that induce G2-M cell-cycle arrest [17]. He further demonstrated that this activity against cancer cells was maintained in paclitaxelresistant cell lines, likely due to less susceptibility to drug efflux pumps. The epothilones were found to inhibit the binding of radiolabeled paclitaxel to b-tubulin in a competitive binding assay, suggesting both a common mechanism of action and a common binding site for the two molecules. This was further supported by structural studies by Nettles and co-workers [18]. Accordingly, electron diffraction studies using zincinduced, two-dimensional sheets of a,b-tubulin suggested a common binding site for epothilone A and paclitaxel, although the epothilone was shown to make unique interactions with the protein. There is controversy regarding the accuracy of this model because this work does not recapitulate the three-dimensional characteristics of a- and b-tubulin protein interactions that are present in intact microtubules [19]. Still, a fair amount of the structure-activity relationships (SAR) for the epothilones can be rationalized by the Nettles model, suggesting that it is not far from the true binding mode.
2.4 Evaluation of epothilones as drug leads Initial studies at BMS confirmed the potent cytotoxicity of the epothilone natural products. Therefore, there was considerable enthusiasm to assess the antitumor activity of these agents in preclinical models of cancer. In a xenograft model (Pat-7) derived from a breast cancer patient, epothilones A and B were not efficacious when dosed intravenously (iv) at their respective maximum tolerated doses (MTD) [6,20]. Similarly, iv bolus administration of epothilone B in a murine allograft model (M5076) was inactive. However, slow iv infusion of the compound was efficacious, suggesting that epothilone B was metabolically unstable in mice. This hypothesis was confirmed by in vitro studies in both murine plasma and liver microsomal fractions. Higher stability was observed in the corresponding human in vitro assays (vide infra). Additional in vitro and in vivo studies in the presence of a general esterase inhibitor, bisdinitrophenyl phosphate, revealed that the macrolactone is susceptible to
Discovery of Ixabepilone (IXEMPRATM)
305
esterase-mediated cleavage. Demonstration of activity in murine xenograft models was an arbitrary prerequisite to advancement of a drug candidate at BMS. Therefore, it was deemed necessary to identify an epothilone analog that demonstrated activity in rodent xenograft models which might be predictive of efficacy in humans.
3. APPROACHES TO IDENTIFY DRUG CANDIDATES 3.1 General strategy The initial disclosure of the absolute stereochemistry of epothilones A and B in 1996 [8] spawned a highly competitive research endeavor to discover biologically active drug analogs. The pioneering total synthesis routes developed by the Danishefsky [21,22], Nicolaou [23,24], and Schinzer [25] groups provided the foundation for the preparation and evaluation of a myriad of diverse epothilone analogs [26–29]. Complementary to these efforts, several additional academic laboratories and industrial research organizations obtained epothilones through fermentation, biosynthesis, biotransformation, total synthesis, semisynthesis, and combinatorial synthesis methods [26,30–34]. Medicinal chemistry efforts at BMS initially focused on utilizing total synthesis approaches to access epothilone analogs. However, semisynthesis approaches were quickly adopted to drive the SAR, once multi-gram quantities of several natural epothilones became available through the collaboration with GBF. Structural modifications were readily accessible at three regions of the molecule, as outlined in Figure 2. The primary objective of the BMS drug discovery program was to identify epothilone analogs with improved: (1) metabolic stability, (2) chemical stability (for oral administration), (3) physicochemical/ thiazole derivatization and C-21 methyl substitution tolerated S
O
isosteric replacements of epoxide allowed
R
OH
N
modifications to lactone and C2−C3 positions allowed
O O
OH
5
Figure 2 Summary of BMS medicinal chemistry efforts.
O
306
Robert M. Borzilleri and Gregory D. Vite
pharmaceutical properties, and (4) pharmacokinetic parameters relative to the natural products [30–31]. Moreover, selection of a suitable development candidate from this unique structural class would require robust in vivo efficacy in multiple preclinical human tumor xenograft models, particularly those resistant to the taxanes, and an acceptable safety profile (therapeutic index). The intrinsic activity of the epothilone analogs or their ability to induce the formation of hyperstable tubulin polymers was measured spectrophotometrically as a function of changes in turbidity. The rate of change in the proportion of polymerized tubulin was expressed as the effective concentration of drug capable of inducing an initial slope of 0.01 when plotting absorbance (A280 nm/min rate) versus time (EC0.01). Preliminary in vitro antitumor activity was assessed using the taxanesensitive human colon carcinoma cell line HCT-116 (non-P-gp expressing line) and expressed as the concentration required for 50% growth inhibition (IC50).
3.2 Total synthesis To address the metabolic instability associated with the epothilones and evaluate the practicality of a total synthesis strategy, researchers at BMS initiated a synthesis of the corresponding lactam analogs using a ring closing metathesis (RCM) route based on published syntheses of epothilones A and C (Figure 3) [27,28]. The key olefin metathesis precursor 8, derived from coupling of allylic amine 6 and polypropionate
OH
S
EDCI, HOBt DMF, 77%
+ N
HO2C NH2
OTBS O
6
7
S 7 OH
N 3
HN O
RuBnCl2(PCy3)2
O
N
benzene 41% (~5:1 E/Z)
OTBS O
HN O
8 TFA, CH2Cl2 0 °C, 68%
Figure 3
OH
S
OP
9a (E-isomer, P = TBS) 9b (Z-isomer, P = TBS) 10 (P = H)
Initial work on the total synthesis of lactam analogs of Epo C.
Discovery of Ixabepilone (IXEMPRATM)
307
acid 7, underwent a ruthenium-catalyzed RCM reaction to afford the C12,13-E-isomeric macrolactam 9a, predominantly [35]. Subsequent deprotection of the silyl ether 9a gave E-epothilone C-lactam 10, which was found to be inactive in the tubulin polymerization assay (EC0.01 W 1,100 mM). The stereochemical outcome of the RCM process was shown to be highly substrate dependent in the context of epothilone research [27–29], and subsequent studies revealed that an additional silyl protecting group on the C7-alcohol produced 1:1 mixtures of the E-and Z-olefinic macrocycles [36]. While the initial total synthesis work demonstrated that the critical RCM reaction was operative, the yields of the desired cis-olefinic macrocycle 9b were unsatisfactory. In addition, the lengthy synthetic routes subsequently developed by others to prepare the lactam analogs [37] were deemed limiting in terms of the quantities of material necessary to sustain a competitive medicinal chemistry effort. To circumvent these issues, efficient and reliable semisynthetic methods were explored to access epothilone derivatives, including the desired macrolactam analogs.
3.3 Semisynthesis The natural olefin-containing epothilones C (3) and D (4) (Figure 1) were found to retain potent activity in the tubulin polymerization assay with EC0.01 values of 3.7 and 0.6 mM respectively [20]. These data suggest that the epoxide oxygen does not engage in critical non-bonded interactions with the tubulin protein. To explore bioisosteric replacements of the C12,13-epoxide moiety and identify more stable epothilone derivatives, larger quantities of the minor metabolites 3 and 4 were required. BMS scientists developed stereoselective titanium- and tungsten-promoted deoxygenation reactions of the more abundant epothilones A and B to gain access to substantial supplies of 3 and 4, respectively [38]. The Z-12,13-cyclopropyl analogs 11 and 12 obtained from direct stereospecific cyclopropanation of the olefinic epothilone derivatives were found to be equipotent in the tubulin polymerization (EC0.01 ¼ 1.42.1 mM) and HCT-116 cell line cytotoxicity (IC50 ¼ 0.701.4 nM) assays (Figure 4) [38]. Unfortunately, high plasma protein binding (W99%) precluded further development of these analogs [20]. Initial attempts to prepare 12a,13a-aziridine analogs by direct aziridination of 3 or 4 using nitrene cycloaddition chemistry were unsuccessful. However, the aziridine analog of epothilone A (13) was prepared using a short synthetic sequence involving double inversion of stereochemistry at C12 and C13 of 1 (Figure 4) [39]. The parent aziridine 13 was approximately sevenfold less potent than epothilones A (1) and B (2) (Figure 1) in the tubulin polymerization assay.
308
Robert M. Borzilleri and Gregory D. Vite
R
R
N
S
S OH
N O
O O
OH
O
O
O
O S OH
N O O
OH
17 (R = OH) 18 (R = NH2)
Figure 5
O
Epothilone cyclopropane and aziridine analogs.
S R
OH
13 (R = H) 14 (R = Me) 15 (R = Ac) 16 (R = SO2NMe2)
11 (R = H) 12 (R = Me)
Figure 4
OH
N
O
N+ O-
OH O O
OH
O
19
Thiazole ring SAR.
However, the N-methyl, N-acyl, N-sulfonylureido derivatives 1416 demonstrated comparable in vitro potency relative to parent macrolides (EC0.01 ¼ 1.02.6 mM) [39]. More importantly, 14 was found to be six times more potent than 2 in inhibiting the proliferation of HCT-116 cells (IC50 ¼ 0.13 nM). Structural permutations of the C15-thiazolylethenyl side chain of the epothilone backbone have been extensively investigated [26–34]. Numerous analogs were found to be active in vitro; however, only a few examples provided significant potential for further development. Because epothilone F (17) was initially isolated in relatively low abundance from fermentation, GBF and BMS pursued alternative routes such as semisynthesis [40,41] to secure sufficient amounts of material for in-depth preclinical testing (Figure 5). Epothilone F was found to be equipotent to 2 in the tubulin polymerization (EC0.01 ¼ 1.8 mM) and HCT-116 cytotoxicity (IC50B0.28 nM) assays [20]. The corresponding 21-amino analog 18 (BMS-310705) was approximately three- to fourfold less potent than 2 in the primary in vitro biochemical and cellular assays. However, the rate of metabolism of 18 in mouse S9 liver fraction was significantly (W15-fold) improved relative to
Discovery of Ixabepilone (IXEMPRATM)
309
epothilone B [20]. Based on its robust preclinical pharmacology and pharmacokinetic profile, the second-generation epothilone 18 was advanced into phase I trials [42]. The interesting N-oxide derivative 19, isolated as an intermediate in the semisynthesis of 17 and 18, also performed well in the tubulin polymerization assay (EC0.01 ¼ 3.90 mM) and inhibited the proliferation of colon (HCT-116), lung (A549), and cervical (KB-3.1) carcinoma cell lines with IC50s of 12 nM [20,40]. However, the compound was found to be considerably less potent against cell lines that express the multi-drug resistant (MDR) phenotype, such as the HCT116/VM46 variant [20]. Structural modifications at the C1C3 positions of the polypropionate-derived portion of the macrocycle have been relatively scarce with few analogs demonstrating significant in vitro activity. The transenoates 20 and 21 (Figure 6) were prepared through semisynthetic routes involving chemoselective dehydration of the 3-hydroxyl groups of epothilones A and B, respectively [43]. While 21 was equipotent to epothilone B in the tubulin polymerization assay, the enoate was approximately 10-fold less potent than the natural product in cell culture [20]. Efforts to address the metabolic stability of the lactone moiety remained a high priority within the research group at BMS. Although the total synthesis efforts to prepare the lactam analogs of the epothilones were met with challenges, a straightforward semisynthetic approach to directly convert the lactone oxygen to nitrogen was realized (Figure 7) [35]. Since the macrolactone moiety of the epothilones is allylic, it was postulated that it may be susceptible to a palladium-catalyzed ring opening to form a p-allylpalladium complex, which could then be trapped by a nitrogen nucleophile. Indeed, unprotected epothilones A and B undergo Pd(0)-catalyzed azidation in the presence of sodium
O
R
S OH
N O O 20 (R = H) 21 (R = Me)
Figure 6 Enoate analogs of 1 and 2.
O
310
Robert M. Borzilleri and Gregory D. Vite
O
R
O
S OH
N
Pd(PPh3)4, NaN3 O OH
HO2C N3
THF-H2O, 45°C 65−70% O
OH
N
O
OH
1 (R = H) 2 (R = Me)
O
O
22 (R = H) 23 (R = Me)
"one-pot" 20−25%
PMe3, THF-H2O 53−89% R
O
R
S
S OH
N HN O
OH
O
DPPA, DMF 4 °C, 24 h or
N
EDCI, HOBt MeCN-DMF 40−65%
26 (R = H) 27 (R = Me)
Figure 7
R
S
OH HO2C NH2 OH
O
24 (R = H) 25 (R = Me)
Semisynthesis of esterase-resistant lactam analogs.
azide to produce 22 and 23, respectively, as single diastereomers (Figure 7) [35]. These intermediates were formed with complete retention of configuration, presumably through stereospecific anti-attack of the palladium and subsequent regioselective reaction of azide from the opposite face of the p-allylpalladium intermediate. Chemoselective reduction of the azide derivatives with trimethylphosphine, followed by cyclization of the corresponding amino acid intermediates 24 and 25, gave the desired macrolactam analogs 26 and 27 (BMS-247550, ixabepilone). The entire three-step sequence can be telescoped into a remarkably efficient and practical ‘‘one pot’’ process that can be carried out over a 24-h period without isolation of intermediates or use of protecting groups. Epothilone F-lactam was synthesized in similar fashion, whereas the macrolactam analogs of epothilones C and D were obtained from 3 and 4, respectively, using the aforementioned stereoselective metal-mediated deoxygenation chemistry [35]. In general, the tubulin-polymerizing and cytotoxic potencies of the lactam analogs were reduced compared to their natural epothilone counterparts. One important exception to this trend, however, was 27. Although the lactam analog of epothilone B (EC0.01 ¼ 3.8 mM) is approximately 1.5-fold to 3-fold less potent than epothilones A and B at inducing tubulin polymerization, the semisynthetic analog is comparable to paclitaxel in both in vitro assays (HCT-116, IC50 ¼ 3.4 nM).
Discovery of Ixabepilone (IXEMPRATM)
311
As anticipated, the metabolic stability of 27 in mouse S9 liver fraction was found to be superior to that of the other natural epothilone analogs with a rate of hydrolysis 100-fold lower than that of epothilone B [20]. Furthermore, 27 demonstrated dramatically lower plasma protein binding in mice (79%) and only slightly higher MDR susceptibility relative to the natural epothilones [20]. Since several of these in vitro parameters serve as predictors of in vivo efficacy in mice, 27 was positioned as the lead development candidate.
4. PRECLINICAL PHARMACOLOGY 4.1 Cytotoxicity In preliminary experiments, 27 was found to possess broad spectrum cytotoxic activity (IC50s ¼ 1.435 nM) against a diverse panel of 21 tumor cell lines characterized as either paclitaxel-sensitive (e.g., ovarian: A2780/TAX; breast: MCF-7; prostate: LNCaP, PC3; colon: HCT-116, LS174T; lung: A549, LX-1) or paclitaxel-resistant (A2780/TAX-R, HCT116/VM46) [44]. Additional profiling across panels of specific tumor cell lines revealed that 23 human lung carcinoma cell lines were highly sensitive to 27 (IC50 ¼ 2.319.2 nM) [45]. Only 4 of 35 human breast and 2 of 20 human colon cell lines were found to be significantly resistant to this agent (IC50s W 100 nM). Collectively, the cytotoxicity data suggested that 27 addressed the known mechanisms associated with resistance to taxanes, that is, MDR-mediated efflux due to P-gp overexpression (HCT-116/VM46), expression of specific b-tubulin mutations (A2780/ TAX-R), and overexpression of the bIII-tubulin isoform (LX-1) [46]. Based on IC90 values from clonogenic cell survival (colony formation) assays, 27 was significantly more effective at killing HCT-116/VM46 and A2780/TAX-R cells relative to paclitaxel following a 16-h drug exposure [44,45]. Mechanistically, lactam 27 was as effective as epothilone B in its ability to arrest proliferating HCT-116 human colon carcinoma cells during the mitotic phase of the cell cycle (G2-M transition) as measured by flow cytometry [44]. Moreover, the concentration of 27 needed to arrest cells in mitosis corresponded to the concentration required to kill cells over the same treatment duration. Subsequent studies with the HCT-116 line revealed 27 affects multiple apoptotic pathways. The compound induced apoptosis of this BAX-positive cell line in a p53-dependent manner through upregulation of the pro-apoptotic protein PUMA [47], although a transcriptionindependent pathway may also be operative in the BAX activation response to 27.
312
Robert M. Borzilleri and Gregory D. Vite
4.2 In vivo efficacy Consistent with its in vitro cytotoxicity profile, 27 demonstrated robust in vivo efficacy in a wide array (W 90%) of the paclitaxel-sensitive and paclitaxel-resistant human tumor xenograft models (Figure 8) [44–46,48,49]. These models were derived from subcutaneous implants of tumor fragments in athymic mice. Antitumor activity was determined at the MTD or the dose level immediately below where excessive toxicity (W one death) was observed. Responses of Z1 tumor log cell kill (LCK), defined as the tumor growth delay in days (TC) divided by the time for a log increase in tumor volume (calculated as 3.32 times the tumor volume doubling time (TVDT)) were considered active. In general, compound 27 was administered iv as a solution in ethanol/water or Cremophort/ethanol/water to nude mice bearing staged tumors at its optimal dose of 616 mg/kg on an intermittent q4d 3 schedule. For example, significant antitumor effects (W four LCK) were observed with 27 in paclitaxel-sensitive tumor models such as HCT-116 and A2780. Similar to paclitaxel at its optimal dose and schedule, this agent was found to be curative (no detectable disease for W 10 times the TVDT) in W50% of the animals bearing HCT-116 tumors [44]. More importantly, 27 demonstrated robust efficacy against several paclitaxel-resistant tumor models. For instance, the MDR variant of HCT-116 (HCT-116/VM46) was equally sensitive to 27 (LCK ¼ 2.4) as compared to the parent tumor line. In addition, 27 demonstrated activity against the P-gp expressing Pat-7 xenografts. This model was established directly from an ovarian cancer patient who was treated with multiple chemotherapeutic agents, including Taxols, but ultimately, this patient’s cancer became resistant. >4 3.5 3 2.5 2 1.5
Active
1 0.5
BT 47 KP 4 L M M -4 M CF7 CF D /A 7 A M -M DR D B A- -2 M 31 B4 Pa 35 t Pa -14 t-2 A5 1 C 49 al u L2 -6 98 7 LX Pa -1 t-2 Pa 4 tPa 25 tPa 26 A2 A t-2 78 27 7 0/ 80 TA s X C -R D M 22 W 8 38 PA 7 35 Pa 4 Pa t-7 tPa 18 CW t-2 2 R M LuC -22 D A- ap PC 35 a2 PCb 3 G H H E C T- CT O 11 -1 6/ 16 VM 4 H 6 N T2 C 9 I-H 69 N 8 A4 7 31
0
Breast
Lung
Pancreatic
Ovarian
Prostate
Colon
SCLC Gastric Squamous
Figure 8 Antitumor efficacy of 27 in several human tumor xenograft models.
Discovery of Ixabepilone (IXEMPRATM)
313
Since 27 demonstrated good oral bioavailability in the presence of a phosphate buffer (pH 8.0), the compound was evaluated in Pat-7 and HCT-116 xenograft models, comparing oral (po) and iv dosing regimens [44]. In these tumor models, equivalent antitumor efficacy (Z2.4 LCK) was obtained following oral administration (6080 mg/kg) of the compound in the phosphate buffer on an every other day (five times) schedule or through the 10 mg/kg iv dose (q4d 3). In fact, 27 cured 7 of 8 mice bearing the HCT-116 tumors when administered orally at 90 mg/kg on the q2d 5 schedule. To better understand the ability of 27 to overcome paclitaxel resistance, additional tumor models were evaluated. For instance, 27 was equally effective against A2780/TAX-R, a non-MDR-related tumor model with acquired paclitaxel resistance attributed to b-tubulin mutations [44]. However, it has been suggested that specific tubulin point mutations may not be relevant to clinical resistance. Therefore, additional tumor models were developed from direct implantation of patient biopsies into immunocompromised mice. Gratifyingly, 27 elicited superior responses in the Pat-21 (2.3 LCK) and Pat-26 (1.2 LCK) human breast and pancreatic carcinoma xenograft models relative to optimally dosed paclitaxel (r0.4 LCK) [45]. Pat-21 tumor cells do not overexpress the efflux transporters (P-gp or MDR-1) or carry tubulin mutations but rather have elevated levels of bIII-tubulin [46]. Selective overexpression of the bIII-tubulin isotype has been reported in various advanced tumors (breast, lung, ovarian) of patients treated with taxanes, and in some cases, this mechanism for resistance has been associated with aggressive disease and lower probability of a response to chemotherapy [46]. The mechanism of resistance for Pat-26 xenograft model, which was derived from a metastatic pancreatic cancer patient innately resistant to paclitaxel, is still not well understood. Childhood malignancies, particularly advanced metastatic disease, represent another area of high unmet medical need and potential opportunities for new chemotherapeutic agents. The tubulin depolymerization agents vincristine and vinblastine (Vinca alkaloids) represent the mainstay of multimodality therapy for many pediatric cancers, whereas the taxanes are considerably less effective. Interestingly, 27 induced objective responses (Z50% volume regression) in several pediatric solid tumor models, including human rhabdomyosarcomas (3/3), neuroblastomas (3/5), osteosarcomas (2/6), and Wilms’ tumor models (6/7) when administered iv at 6.6 and 10 mg/kg on an every 4 day 3 schedule in mice [48]. Attempts to develop tumor models resistant to 27 have been challenging. In fact, resistance to 27 has not been observed following passage of the A2780 xenografts for more than 3 years in the presence of
314
Robert M. Borzilleri and Gregory D. Vite
the drug, whereas resistance to paclitaxel in this ovarian tumor model emerged in o6 months following continuous exposure to the taxane [20]. Therapeutic concentrations of drug are likely maintained in the tumor cells, since 27 exhibits reduced susceptibility to MDR-related efflux and the compound does not readily induce tumor cells to overexpress drug efflux pumps (e.g., ATP-binding cassette transporters P-gp or MRP-1). In addition to robust in vivo efficacy as a single agent, 27 demonstrated synergistic antitumor effects when combined with the targeted antiangiogenic agent bevacizumab in models derived from breast (Pat21, KPL4), colon (HCT-116/VM46, WiDr, GEO), lung (L2987), and kidney (151 b) cancers [49]. Statistically superior efficacy (synergy) as measured by growth delay (LCK) and tumor volume reduction was observed with the 27-bevacizumab combination regardless of the tumor model sensitivity (or resistance) to either of the single agents alone at their optimal doses and schedules. Even in the case of the 151b renal model, which was resistant to both single agents (r0.6 LCK), the combination resulted in significant tumor growth delay (1.5 LCK). Similar results were obtained with another antiangiogenic agent, sunitinib in the 151b model. Importantly, the 27-bevacizumab combination was statistically more effective at inhibiting the growth of the GEO and the MDRexpressing HCT-116/VM46 tumor models compared to the combination of paclitaxel and bevacizumab. It has been postulated that the enhanced in vivo efficacy of 27 relative to paclitaxel in combination of bevacizumab may be due to higher antiangiogenic activity (killing of tumor-associated endothelial cells) and/or reduced susceptibility to drug efflux proteins [49].
4.3 Profiling Pharmacokinetic analysis of 27 in female nude mice revealed rapid clearance (4.35.1 L/h/kg), extensive tissue distribution (Vss ¼ 2137 L/ kg), and a mean half-life of 1316 h following a single iv dose of either 10, 6, or 4 mg/kg [45]. Additional in vitro studies indicated that 27 is primarily metabolized by CYP3A4 to many inactive metabolites [20]. In human liver microsomes, 27 did not inhibit a representative panel of CYP450 isozymes (1A2, 3A4, 2B6, 2C8, 2C9, 2C19, 2D6) or induce the activity of CYP3A4, CYP1A2, CYP2B6, or CYP2C9 in cultured human hepatocytes at clinically relevant concentrations [20,50]. Due to its impressive in vivo efficacy profile across a range of taxane-resistant xenograft models and favorable ADME properties, 27 (ixabepilone) was selected for clinical development.
Discovery of Ixabepilone (IXEMPRATM)
315
5. CLINICAL RESULTS 5.1 Phase I/II highlights Based on the robust preclinical activity of ixabepilone, multiple clinical trials across a broad range of tumor types were initiated, including some in collaboration with the National Cancer Institute (NCI). Early phase I studies explored dose, schedule, method of administration, and tolerability [51,52]. Weekly iv infusions (25 mg/m2) did not appear to offer significant advantages over a standard 3-week dosing schedule. Dose escalation for the latter schedule reached 50 mg/m2, but the recommended phase II dose was set at 40 mg/m2 (3 h iv infusion) due to high incidence of peripheral neuropathy at the higher dose. Daily dosing (6 mg/m2, QD 5, every 3 weeks) demonstrated dose-limiting neutropenia, but neuropathies were reduced with this schedule [53]. This daily dosing regimen provided a promising 57% response rate (5.5 months time-to-progression) in a phase II breast cancer trial where patients had not received prior taxane therapy (i.e., taxane naı¨ve) (Table 1) [54]. Another chemotherapy-naı¨ve phase II trial, in hormone refractory prostate cancer, showed an ixabepilone response rate of 32%, which increased to 48% when ixabepilone was combined with estramustine (EMP) [55]. Prostate-specific antigen (PSA) declines of W50% were observed for 48% of patients in the single-agent arm, while 69% of patients had W50% PSA decline in the drug combination arm of the trial. Efficacy was also observed in patients with chemoresistant renal carcinomas, lymphomas, and drug-resistant lung cancers, whereas
Table 1 Select clinical data in taxane-naı¨ve patients and clinical data in heavily pre-treated patients that supported NDA filing N
Dose
Schedule
Response rate (%)
6 mg/m2 35 mg/m2 35 mg/m2 + 280 mg EMP
Daily 5, Q 3weeks 57 Q 3weeks 32 Q 3weeks 48 TID 5, po
Phase II Breast [54] Prostate [55]
23 45 47
Phase II/III Registrational Breast [50,57] 126 Breast [58] 752
Q 3weeks 40 mg/m2 40 mg/m2 Q 3weeks + 1000 mg/m2 CAPE BID 14, po
12 35
316
Robert M. Borzilleri and Gregory D. Vite
results obtained with other tumor types, such as colorectal cancer, were less impressive [51]. The pharmacokinetics in humans at the recommended phase II dose of 40 mg/m2 iv can be characterized as rapid and extensive tissue distribution (large Vdss) and a favorable half-life of 35 h [45,56]. Appreciable oral bioavailability (Fpo ¼ 54%) was achieved with ixabepilone at a 25 mg/m2 dose in a phase I dose finding study [30].
5.2 Phase II/III registrational trials Following on the promising phase I/II studies in breast cancer, ixabepilone was studied in a phase II trial with patients referred to as ‘‘triple-refractory’’ [57]. These patients received prior treatment with an anthracycline, a taxane, and capecitabine (CAPE), and all experienced disease progression before entering the ixabepilone trial. Patients were dosed every 3 weeks with 40 mg/m2 of ixabepilone (3-h iv infusion). In this highly drug-resistant population, the objective response rate as determined by independent radiological review was 12%, with an additional 50% of patients experiencing stable disease. Investigator-determined response rate was 18% with 44% of patients characterized as having stable disease (Table 1). A phase III trial enrolling W750 metastatic breast cancer patients was conducted to compare the combination of ixabepilone and CAPE versus CAPE alone in a setting where patients had progressive disease following treatment with an anthracycline or a taxane [58]. The dose of ixabepilone and schedule of administration were the same as noted above, along with orally dosed CAPE (1,000 mg/m2, bid 14) in the drug combination arm. An approximate twofold improvement in objective response rate and a progression-free survival benefit of 5.8 months (vs. 4.2 months for CAPE) was observed for the drug combination. Detailed retroanalysis suggested that certain subgroups showed preferential benefit, including those patients referred to as triple negative (i.e., no or low expression of human epidermal growth factor receptor-2 (HER2), estrogen receptor (ER), and progesterone receptor (PR)). Grade 3/4 peripheral neuropathy (23%) and neutropenia (68%) were notably higher in the combination arm of the trial. Taken together, these two studies demonstrated the effectiveness of ixabepilone against breast cancer and served as the basis for submitting a New Drug Application to the FDA.
5.3 Pharmacogenomics In the post-genomic era, identifying subsets of patients that are likely to be responders to chemotherapy as well as targeted treatments is a high priority. Accordingly, retrospective analysis of clinical data from ixabepilone trials has been used in attempt to identify genomic
Discovery of Ixabepilone (IXEMPRATM)
317
signatures that predict positive outcomes [59]. Candidate genes include those encoding microtubule-associated proteins (MAPs), HER2, and ER. Low expression of ER correlated with cytotoxicity of ixabepilone in vitro and with tumor response in patients, whereas expression level of tau protein (a MAP) was less predictive [60]. The relevance of tubulin isotype expression as an indicator of drug sensitivity is emerging. For instance, overexpression of the bIII-tubulin isotype is associated with resistance to taxanes. As noted above, preclinical studies in cancer cells and taxaneresistant xenograft models that overexpress bIII-tubulin suggest that ixabepilone’s clinical activity in a refractory setting may be related to the drug’s indifference to overexpression of this isoform [46]. Clearly, further studies and analyses are required to develop a better understanding of predictive biomarkers for ixabepilone.
6. FUTURE DIRECTIONS In addition to ongoing clinical trials across a wide range of malignancies (e.g., prostate, NSCLC, endometrial), ixabepilone is being investigated in combination with targeted agents such as bevacizumab and trastuzumab [61]. These pivotal phase II trials are supported by the impressive synergistic activity observed with the drug combinations in preclinical models and the manageable safety/tolerability profile of ixabepilone observed in patients (vide supra). There are several other epothilone analogs being evaluated in the clinic (Figure 9). Novartis continues to develop one of the initially isolated natural products, epothilone B (2, patupilone, EPO-906) [62]. Clinical activity has been observed with 2 in the drug-resistant disease setting, including patients with renal cell carcinoma (RCC) and relapsed/ refractory ovarian cancer. In contrast to ixabepilone, diarrhea is the doselimiting toxicity (DLT), which requires careful monitoring and management. Epothilone B is currently being investigated in phase III trials involving patients with platinum-resistant ovarian cancer [61,62]. Kosan, in collaboration with Roche, initiated clinical studies on epothilone D (4) but later discontinued development of the drug to pursue a secondgeneration epothilone D analog, KOS-1584 (28, dehydelone) [63]. Based on disease stabilization observed in various advanced malignancies and a confirmed partial response in NSCLC, KOS-1584 was advanced into phase II trials (NSCLC). BMS is conducting preclinical studies on thirdgeneration analogs, such as KOS-1803 (29), for oncology. Other groups are evaluating brain penetrant analogs for potential neurodegenerative disease indications (e.g., Alzheimer’s) [64]. The completely synthetic epothilone analog 30 (sagopilone, ZK-EPO) is being developed by Bayer Schering Pharma AG [65,66]. Confirmed partial responses have been
318
Robert M. Borzilleri and Gregory D. Vite
R1
R2
O S O
O
OH OH
N O OH
O O
28 (R1 = CH3, R2 = 2-methylthiazole) 29 (R1 = CF3, R2 = 5-methylisoxazole)
OH
O
30
HN
NH2 NH
CO2H
O N H
O N
HN H2N
N
N H
H N O
O N H O CO2H
H N
O
CO2H
N H CO2H
S
O
O
S O
N
N S
OH
N
31
O O
Figure 9
OH
O
Additional epothilone analogs of interest.
observed in taxane-pretreated or resistant breast cancer, uterine cancer, platinum-resistant ovarian cancer, NSCLC, and melanoma [66]. Similar to ixabepilone, peripheral neuropathy has been defined as the most common drug-related adverse event (principal DLT). In an alternative approach to selectively target tumors that overexpress the membrane-bound folate receptor alpha (FRa) (e.g., ovarian, endometrial, breast, renal, and lung cancers), BMS and collaborators at Endocyte Corporation designed the aziridine conjugate 31 (BMS-753493) [67–69]. The exquisitely potent hydroxyethyl-substituted aziridine (HCT-116, IC50 ¼ 0.34 nM) was appended to vitamin folic acid through a cleavable disulfide linker followed by a short ionizable peptide sequence (Figure 9) [68,69]. After delivery and subsequent internalization of the folate-drug conjugate into tumor cells through FR-mediated endocytosis, the cytotoxic aziridine can be released by reduction of the disulfide bond. As predicted, 31 demonstrated in vivo efficacy against the FR-positive human KB nasopharyngeal, IGROV ovarian, HeLa cervical, and murine 98M109 lung models but was found to be less active against FR-deficient tumors [70]. Moreover, the epothilone-folate conjugate demonstrated synergistic antitumor activity against FR-positive human tumor xenografts when combined with ixabepilone, bevacizumab, and cisplatin. Phase I studies are underway to assess the safety and tolerability of 31 in cancer patients [61].
Discovery of Ixabepilone (IXEMPRATM)
319
7. CONCLUSIONS The discovery of the epothilones by Ho¨fle and Reichenbach galvanized global research and development efforts to identify drug candidates from this class with potential to provide clinical benefit when patients fail to respond to existing chemotherapies. While these polyketide-derived macrolides provided potent antineoplastic activity in vitro, these effects were not observed in murine tumor xenograft models in our laboratories, presumably due to esterase-mediated inactivation of the compound in vivo. Consequently, a collaboration between BMS and GBF was established to identify metabolically stable epothilone derivatives. A highly efficient semisynthetic sequence was developed to access the macrolactam analogs of the natural macrolides. Ixabepilone, the lactam analog of epothilone B, demonstrated potent tubulin polymerization activity and robust cytotoxicity versus several paclitaxel-resistant cell lines. Moreover, improved metabolic stability and pharmacokinetic properties were achieved with this analog relative to the natural macrolides. In vivo, the novel microtubule-stabilizing agent demonstrated broad spectrum antitumor activity against various human tumor xenograft models, including those resistant to paclitaxel. The robust preclinical efficacy was recapitulated in early human clinical trials for several cancers, including metastatic breast and prostate cancer patients with multidrug-resistant disease. Ixabepilone (IXEMPRAt) was approved by the FDA in October 2007 as a first-in-class agent for treatment of drug-resistant/refractory metastatic or locally advanced breast cancers in combination with CAPE or as monotherapy following failure of an anthracycline, a taxane, and CAPE.
REFERENCES [1] M. C. Wani, H. L. Taylor, M. E. Wall, P. Coggon and A. T. McPhail, J. Am. Chem. Soc., 1971, 93, 2325. [2] P. B. Schiff, J. Fant and S. B. Horwitz, Nature, 1979, 277, 665. [3] C. Ferlini, D. Gallo and G. Scambia, Expert Opin. Investig. Drugs, 2008, 17, 335. [4] L. He, G. A. Orr and S. B. Horwitz, Drug Discov. Today, 2001, 6, 1153. [5] D. C. Myles, Annu. Rep. Med. Chem., 2002, 37, 125. [6] J. T. Hunt, Mol. Cancer Ther., 2009, 8, 275. [7] K. Gerth, N. Bedorf, G. Ho¨fle, H. Irschik and H. Reichenbach, J. Antibiot., 1996, 49, 560. [8] G. Hoefle, N. Bedorf, H. Steinmetz, D. Schomburg, K. Gerth and H. Reichenbach, Angew. Chem., Int. Ed. Engl., 1996, 35, 1567. [9] I. Hardt, H. Steinmetz, K. Gerth, F. Sasse, H. Reichenbach and G. Hofle, J. Nat. Prod., 2001, 64, 847. [10] K. Gerth, H. Steinmetz, G. Ho¨fle and H. Reichenbach, J. Antibiot., 2000, 53, 1373. [11] K. Gerth, H. Steinmetz, G. Ho¨fle and H. Reichenbach, J. Antibiot., 2001, 54, 144.
320
Robert M. Borzilleri and Gregory D. Vite
[12] I. Molna´r, T. Schupp, M. Ono, R. E. Zirkle, M. Milnamow, B. Nowak-Thompson, N. Engel, C. Toupet, A. Stratmann, D. D. Cyr, J. Gorlach, J. M. Mayo, A. Hu, S. Goff, J. Schmid and J. M. Ligon, Chem. Biol., 2000, 7, 97. [13] L. Tang, S. Shah, L. Chung, J. Carney, L. Katz and C. Khosla, Science, 2000, 287, 640. [14] B. Julien, S. Shah, R. Ziermann, R. Goldman, L. Katz and C. Khosla, Gene, 2000, 249, 153. [15] B. Julien and S. Shah, Antimicrob. Agents Chemother., 2002, 46, 2772. [16] S. Schneiker, O. Perlova, O. Kaiser, K. Gerth, A. Alici, M. O. Altmeyer, D. Bartels, T. Bekel, S. Beyer, E. Bode, H. B. Bode, C. J. Bolten, J. V. Choudhuri, S. Doss, Y. A. Elnakady, B. Frank, L. Gaigalat, A. Goesmann, C. Groeger, F. Gross, L. Jelsbak, J. Kalinowski, C. Kegler, T. Knauber, S. Konietzny, M. Kopp, L. Krause, D. Krug, B. Linke, T. Mahmud, R. Martinez-Arias, A. C. McHardy, M. Merai, F. Meyer, S. Mormann, J. Munoz-Dorado, J. Perez, S. Pradella, S. Rachid, G. Raddatz, F. Rosenau, C. Rueckert, F. Sasse, M. Scharfe, S. C. Schuster, G. Suen, A. Treuner-Lange, G. J. Velicer, F.-J. Vorhoelter, K. J. Weissman, R. D. Welch, S. C. Wenzel, D. E. Whitworth, S. Wilhelm, C. Wittmann, H. Bloecker, A. Puehler and R. Mueller, Nat. Biotechnol., 2007, 25, 1281. [17] D. M. Bollag, P. A. McQueney, J. Zhu, O. Hensens, L. Koupal, J. Liesch, M. Goetz, E. Lazarides and C. M. Woods, Cancer Res., 1995, 55, 2325. [18] J. H. Nettles, H. Li, B. Cornett, J. M. Krahn, J. P. Snyder and K. H. Downing, Science, 2004, 305, 866. [19] D. W. Heinz, W.-D. Schubert and G. Hoefle, Angew. Chem. Int. Ed. Engl., 2005, 44, 1298. [20] F. Y. F. Lee, R. Borzilleri, C. R. Fairchild, A. Kamath, R. Smykla, R. Kramer and G. Vite, Cancer Chemother. Pharmacol., 2008, 63, 157. [21] A. Balog, D. Meng, T. Kamenecka, P. Bertinato, D.-S. Su, E. J. Sorensen and S. J. Danishefsky, Angew. Chem. Int. Ed. Engl., 1996, 35, 2801. [22] D.-S. Su, D. Meng, P. Bertinato, A. Balog, E. J. Sorensen, S. J. Danishefsky, Y.-H. Zheng, T.-C. Chou, L. He and S. B. Horwitz, Angew. Chem. Int. Ed. Engl., 1997, 36, 757. [23] Z. Yang, Y. He, D. Vourloumis, H. Vallberg and K. C. Nicolaou, Angew. Chem. Int. Ed. Engl., 1997, 36, 166. [24] K. C. Nicolaou, F. Sarabia, S. Ninkovic and Z. Yang, Angew. Chem. Int. Ed. Engl., 1997, 36, 525. [25] D. Schinzer, A. Limberg, A. Bauer, O. M. Bohm and M. Cordes, Angew. Chem. Int. Ed. Engl., 1997, 36, 523. [26] K.-H. Altmann, B. Pfeiffer, S. Arseniyadis, B. A. Pratt and K. C. Nicolaou, Chem. Med. Chem., 2007, 2, 396. [27] K. C. Nicolaou, F. Roschangar and D. Vourloumis, Angew. Chem. Int. Ed. Engl., 1998, 37, 2014. [28] K. C. Nicolaou, A. Ritze´n and K. Namoto, Chem. Commun., 2001, 17, 1523. [29] C. R. Harris and S. J. Danishefsky, J. Org. Chem., 1999, 64, 8434. [30] R. M. Borzilleri and G. D. Vite, Drugs Fut., 2002, 27, 1149. [31] G. D. Vite, R. M. Borzilleri, S.-H. Kim, A. Regueiro-Ren, W. G. Humphreys and F. Y. F. Lee, in Anticancer Agents: Frontiers in Cancer Chemotherapy, ACS Symposium Series 796 (eds I. Ojima, G. D. Vite, and K.-H. Altmann), American Chemical Society, Washington, D.C., 2001, p. 97. [32] K.-H. Altmann, M. J. J. Blommers, G. Caravatti, A. Florsheimer, K. C. Nicolaou, T. O’Reilly, A. Schmidt, D. Schinzer and M. Wartmann, in Anticancer Agents: Frontiers in Cancer Chemotherapy, ACS Symposium Series 796 (eds I. Ojima, G. D. Vite, and K.-H. Altmann), American Chemical Society, Washington, D.C., 2001, p. 112. [33] U. Klar, W. Skuballa, B. Buchmann, W. Schwede, T. Bunte, J. Hoffmann and R. B. Lichtner, in Anticancer Agents: Frontiers in Cancer Chemotherapy, ACS Symposium Series 796 (eds I. Ojima, G. D. Vite, and K.-H. Altmann), American Chemical Society, Washington, D.C., 2001, p. 131.
Discovery of Ixabepilone (IXEMPRATM)
321
[34] K. C. Nicolaou, N. Winssinger, J. Pastor, S. Ninkovic, F. Sarabia, Y. He, D. Vourloumis, Z. Yang, T. Li, P. Giannakakou and E. Hamel, Nature, 1997, 387, 268. [35] R. M. Borzilleri, X. Zheng, R. J. Schmidt, J. A. Johnson, S.-H. Kim, J. D. DiMarco, C. R. Fairchild, J. Z. Gougoutas, F. Y. F. Lee, B. H. Long and G. D. Vite, J. Am. Chem. Soc., 2000, 122, 8890. [36] D. Schinzer, K.-H. Altmann, F. Stuhlmann, A. Bauer and M. Wartmann, Chem. Bio. Chem., 2000, 1, 67. [37] S. J. Stachel, M. D. Chappell, C. B. Lee, S. J. Danishefsky, T.-C. Chou and S. B. Horwitz, Org. Lett., 2000, 2, 1637. [38] J. A. Johnson, S.-H. Kim, M. Bifano, J. DiMarco, C. Fairchild, J. Gougoutas, F. Lee, B. Long, J. Tokarski and G. Vite, Org. Lett., 2000, 2, 1537. [39] A. Regueiro-Ren, R. M. Borzilleri, X. Zheng, S.-H. Kim, J. A. Johnson, C. R. Fairchild, F. Y. F. Lee, B. H. Long and G. D. Vite, Org. Lett., 2001, 3, 2693. [40] G. Ho¨fle, N. Glaser, M. Kiffe, H.-J. Hecht, F. Sasse and H. Reichenbach, Angew. Chem. Int. Ed. Engl., 1999, 38, 1971. [41] G. Ho¨fle, N. Glaser, T. Leibold and M. Sefkow, Pure Appl. Chem., 1999, 71, 2019. [42] A. V. Kamath, M. Chang, F. Y. Lee, Y. Zhang and P. H. Marathe, Cancer Chemother. Pharmacol., 2005, 56, 145. [43] A. Regueiro-Ren, K. Leavitt, S.-H. Kim, G. Ho¨fle, M. Kiffe, J. Z. Gougoutas, J. D. DiMarco, F. Y. F. Lee, C. R. Fairchild, B. H. Long and G. D. Vite, Org. Lett., 2002, 4, 3815. [44] F. Y. F. Lee, R. Borzilleri, C. R. Fairchild, S.-H. Kim, B. H. Long, C. Reventos-Suarez, G. D. Vite, W. C. Rose and R. A. Kramer, Clin. Cancer Res., 2001, 7, 1429. [45] F. Y. F. Lee, R. Smykla, K. Johnston, K. Menard, K. McGlinchey, R. W. Peterson, A. Wiebesiek, G. Vite, C. R. Fairchild and R. Kramer, Cancer Chemother. Pharmacol., 2009, 63, 201. [46] C. Dumontet, M. A. Jordan and F. F. Y. Lee, Mol. Cancer Ther., 2009, 8, 17. [47] H. Yamaguchi, J. Chen, K. Bhalla and H.-G. Wang, J. Biol. Chem., 2004, 279, 39431. [48] J. K. Peterson, C. Tucker, E. Favours, P. J. Cheshire, J. Creech, C. A. Billups, R. Smykla, F. Y. F. Lee and P. J. Houghton, Clin. Cancer Res., 2005, 11, 6950. [49] F. Y. F. Lee, K. L. Covello, S. Castaneda, D. R. Hawken, D. Kan, A. Lewin, M.-L. Wen, R.-P. Ryseck, C. R. Fairchild, J. Fargnoli and R. Kramer, Clin. Cancer Res., 2008, 14, 8123. [50] FDA-Approved Patient Label for IXEMPRAt (ixabepilone), Bristol-Myers Squibb Company, Princeton, NJ, October 2007. [51] N. Denduluri and S. M. Swain, Expert Opin. Investig. Drugs, 2008, 17, 423. [52] M. Harrison and C. Swanton, Expert Opin. Investig. Drugs, 2008, 17, 523. [53] J. Abraham, M. Agrawal, S. Bakke, A. Rutt, M. Edgerly, F. M. Balis, B. Widemann, L. Davis, B. Damle, D. Sonnichsen, D. Lebwohl, S. Bates, H. Kotz and T. Fojo, J. Clin. Oncol., 2003, 21, 1866. [54] N. Denduluri, J. A. Low, J. J. Lee, A. W. Berman, J. M. Walshe, U. Vatas, C. K. Chow, S. M. Steinberg, S. X. Yang and S. M. Swain, J. Clin. Oncol., 2007, 25, 3421. [55] J. J. Lee and W. K. Kelly, Nat. Clin. Pract. Oncol., 2009, 6, 85. [56] S. Mani, H. McDaid, A. Hamilton, H. Hochster, M. B. Cohen, D. Khabelle, T. Griffin, D. E. Lebwohl, L. Liebes, F. Muggia and S. B. Horwitz, Clin. Cancer Res., 2004, 10, 1289. [57] E. Thomas, J. Tabernero, M. Fornier, P. Conte, P. Fumoleau, A. Lluch, L. T. Vahdat, C. A. Bunnell, H. A. Burris, P. Viens, J. Baselga, E. Rivera, V. Guarneri, V. Poulart, J. Klimovsky, D. Lebwohl and M. Martin, J. Clin. Oncol., 2007, 25, 3399. [58] E. S. Thomas, H. L. Gomez, R. K. Li, H.-C. Chung, L. E. Fein, V. F. Chan, J. Jassem, X. B. Pivot, J. V. Klimovsky, F. Hurtado de Mendoza, B. Xu, M. Campone, G. L. Lerzo, R. A. Peck, P. Mukhopadhyay, L. T. Vahdat and H. H. Roche´, J. Clin. Oncol., 2007, 25, 5210.
322
Robert M. Borzilleri and Gregory D. Vite
[59] J. J. Lee and S. M. Swain, Clin. Cancer Res., 2008, 14, 1618. [60] H. Lee, L. Xu, S. Wu, B. Paul, J. Baselga, A. Llombart, G. G. Steger, S. Galbraith and E. Clark, Abstract # 3011, 42nd ASCO Annual Meeting, Atlanta, Georgia, June, 2006. [61] U.S. National Institutes of Health, ClinicalTrials.gov. [62] J. M. Larkin, Drugs Fut., 2007, 32, 323. [63] Y. S. Cho, K.-D. Wu, M. A. S. Moore, T.-C. Chou and S. J. Danishefsky, Drugs Fut., 2005, 30, 737. [64] D. Huryn, C. Ballatore, K. Brunden, E. Hyde, R. F. Deiches, V. M. Y. Lee, J. Trojanowski, J. Potuzak and A. B. Smith, MEDI-200, 235th ACS National Meeting, New Orleans, LA, April, 2008. [65] U. Klar, B. Buchmann, W. Schwede, W. Skuballa, J. Hoffmann and R. B. Lichtner, Angew. Chem. Int. Ed. Engl., 2006, 45, 7942. [66] U. Klar, J. Hoffmann and M. Giurescu, Expert Opin. Investig. Drugs, 2008, 17, 1735. [67] C. P. Leamon, Curr. Opin. Invest. Drugs, 2008, 9, 1277. [68] G. D. Vite, F. Y. F. Lee, C. P. Leamon and I. R. Vlahov, US Patent Application 2007/ 275904-A1, 2007. [69] G. D. Vite, ACS Princeton Fall Organic Chemistry Symposium, September 19, 2008. [70] K. Covello, C. Flefleh, K. Menard, A. Wiebesiek, K. McGlinchey, M.-L. Wen, R. Westhouse, J. Reddy, I. Vlahov, J. Hunt, W. Rose, C. Leamon, G. Vite and F. Lee, Abstract # 2326, 99th AACR Annual Meeting, San Diego, CA, April, 2008.
CHAPT ER
16 Hedgehog Signaling Pathway Inhibitors as Cancer Therapeutics Stefan Peukert and Karen Miller-Moslin
Contents
1. Introduction 2. Mechanism of Hedgehog Pathway Signaling 3. Inhibitors of the Hedgehog Pathway 3.1 Smoothened inhibitors 3.2 Non-Smoothened inhibitors 4. Hedgehog Inhibitors in Clinical Trials 5. Conclusions References
323 324 325 325 331 332 333 334
1. INTRODUCTION Targeting fundamental molecular signaling pathways that control growth and cell death have been suggested as a promising new paradigm for the discovery of drugs [1]. Of particular interest is the Hedgehog (Hh) pathway, initially discovered in Drosophila by Wieschaus and Nu¨ssleinVollhard [2], which directs the development of multiple tissues during embryonic development and contributes to tissue homeostasis in adults [3–5]. Inactivation of the pathway during embryonic development causes birth defects [6,7], whereas abnormal activation is linked to tumorigenesis in several cancers [3]. Genetic validation of this pathway in human tumors comes from observations that patients with a germline mutation in Patched (Ptch1), a component of the Hh pathway, have activated Hh signaling and develop Gorlin syndrome, also known as nevoid basal cell Novartis Institutes for Biomedical Research Inc., Cambridge, MA, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04416-9
r 2009 Elsevier Inc. All rights reserved.
323
324
Stefan Peukert and Karen Miller-Moslin
carcinoma (BCC) [8]. Gorlin syndrome patients present with skeletal and dental abnormalities due to developmental patterning defects, show high incidence of sporadic BCC, and an elevated incidence of medulloblastoma, ovarian cysts, and ovarian carcinoma. Moreover, somatic mutations in Patched and Smoothened (Smo), another Hh pathway component leading to constitutive pathway activation, have been found in 20% of pediatric medulloblastomas [9,10] and in more than 70% of sporadic BCCs [11,12]. Furthermore, aberrant activation of the Hh pathway without known mutation is implicated in a number of additional tumor types, including small cell lung cancer, gut-related tumors, pancreatic and prostate cancer [3]. Inhibition of the Hh pathway in either tumor cells directly or nonmalignant stromal cells, which, as part of the tumor microenvironment, support tumor growth [13], has emerged as an attractive target in anticancer therapy [14–16]. Recent positive results from the topical and systemic treatment of BCC patients with Hh antagonists provide the first evidence for the therapeutic benefit resulting from inhibition of this signaling pathway [17,18].
2. MECHANISM OF HEDGEHOG PATHWAY SIGNALING Although a detailed description of the components of the Hh pathway and their interactions is described in recent reviews [19,20], some information is necessary to understand the mechanism of action of the Hh inhibitors described herein. The Hh signaling pathway (Figure 1) received its name from a small family of secreted proteins which comprises Sonic hedgehog (Shh), Inhibitors of Hh protein
Smo inhibitors PTCH1 HH
SMO SUFU Gli
Inhibitors of Gli-Transcription
Gli
Target genes
Figure 1 Components of the Hh signaling pathway and molecular sites targeted by Hh pathway inhibitors. Source: From Ref. 16 with permission of the publisher. (See Color Plate 16.1 in Color Plate Section.)
325
Hedgehog Signaling Pathway Inhibitors as Cancer Therapeutics
Indian hedgehog (Ihh), and Desert hedgehog (Dhh) in mammals. The Shh gene was discovered in 1993 and was named after Sonic the Hedgehog, a popular video game hero at this time, since mutations of this gene in fruit flies gave rise to spiky hairs. The other related Hh genes are named after species of living hedgehogs [21]. In the absence of Hh ligands, the 12-pass transmembrane receptor Ptch1 inhibits activity of the downstream receptor Smoothened, which resembles G-protein-coupled receptors (GPCR) in general topology. The interactions of the components of the Hh pathway can occur in the cilium of cells [22]. In the absence of Hh protein, Ptch1 is localized in the primary cilium and seems to exclude ciliary localization of Smo. Binding of Shh to Ptch1 causes Smo, stored in intracellular vesicles, to move to the cilium and activate signal transduction. Active Smo signals through a cytosolic complex of proteins including Suppressor of Fused (SuFu). This leads to activation of the glioma (Gli) family of transcription factors and their translocation to the nucleus, where they trigger the expression of specific genes that promote cell proliferation and differentiation. On the basis of the current understanding of the pathway, several druggable nodes have been identified, and assays have been developed that can detect small molecules able to alter the activity of these targets.
3. INHIBITORS OF THE HEDGEHOG PATHWAY Small-molecule modulators of the Hh signaling pathway have been the subject of recent reviews [15,23–26], and the last few years have brought a tremendous increase in reports of novel inhibitors. Although the majority of such inhibitors target the Smoothened receptor, a few reports of small molecules targeting other members of this pathway have also appeared.
3.1 Smoothened inhibitors The natural product alkaloid cyclopamine (1) was among the first smallmolecule inhibitors of the Hh pathway to be reported in the literature [7]. It was later established that cyclopamine achieves this inhibition through direct binding to the heptahelical bundle of the Smoothened receptor [27]. HN H
H
O
H N
H O
H H
H H
HO
H
H
O
1: cyclopamine
2: IPI-269609/IPI-609
H
326
Stefan Peukert and Karen Miller-Moslin
H
H N
O
H
O
H
H O
O
H
H
H H
S N H
H N
H
H
HN N
H
H
3: IPI-926
4
A recent report described the semi-synthesis of the D-homocyclopamine analog 2 (IPI-269609/IPI-609) from cyclopamine through an oxidation/cyclopropanation/ring expansion sequence [28]. Compound 2 was then further functionalized through acylation, reductive amination, or direct alkylation. Hh pathway inhibition was assessed in murine progenitor C3H10T1/2 cells. When the Hh pathway is active, these cells differentiate into osteoblasts and produce high levels of alkaline phosphatase (AP). A reduction in AP production in the presence of compound is indicative of Hh pathway inhibition. Several of these analogs showed an EC50 o1 mM in this assay. Relative to cyclopamine (EC50 ¼ 0.17 mM), 2 (EC50 ¼ 0.20 mM) showed improved aqueous solubility at pH 7.4 (W20-fold), enhanced stability in simulated gastric fluid, and good oral exposure (80% oral bioavailability in mice). Compound 2 was also shown to block Hh signaling in pancreatic cancer cells in vitro and in vivo and to abolish distant metastases in orthotopic murine xenografts of human pancreatic cancer [29]. Further modifications to the D-homocyclopamine skeleton have been exemplified in recent patent applications [30,31]. For example, methyl sulfonamide 3 (IPI-926) and fused pyrazole 4 were each reported to have EC50 o20 nM in the AP assay. Both compounds were evaluated in in vivo models of pancreatic cancer and medulloblastoma. IPI-926 showed regression of medulloblastoma and no tumor regrowth upon dosing at 40 mg/kg for 50 days [32]. Proline derivatives such as 5 (Cur61414) have been identified as Hh pathway inhibitors that act through binding to Smo [33,34]. Compound 5 was able to suppress proliferation and induce apoptosis in in vitro models of BCC. Phenyl quinazolinone ureas such as 6 are also reported to be potent inhibitors of Hh signaling [35,36]. Compounds were evaluated in a cellbased reporter gene assay (to be abbreviated Gli-Luc RGA), which quantified Hh pathway activity based on downstream activation of GliLuciferase. Some structure-activity relationship (SAR) within this series has been reported. Replacement of the 4-fluorophenyl ring of 6 (IC50 ¼ 70 nM)
327
Hedgehog Signaling Pathway Inhibitors as Cancer Therapeutics
with an isopropyl group led to a significant loss in activity (B6-fold), and replacement of the quinazolinone by a pyrimidinone ring was not tolerated. Replacement of the para-chloro substituent on the aniline ring with a hydrogen led to a decrease in potency (B10-fold). Structurally related quinazolines (7) and pyridopyrimidines (8) have also been reported as potent Hh inhibitors in the patent literature (IC50 of 2.8 and 3.6 nM, respectively, in a Gli-Luc RGA) [37].
F O
O N
N
O
NH F
N
N N
F
H N
HN
F
O
O
O Cl O
5: Cur61414
6
O N H
N N
O
N H
N
7
N H
N N
N H
8
Various Smoothened inhibitors have been reported in the patent literature that contain either an N-(4-chloro-phenyl)-benzamide or an N-(4-chloro-phenyl)-nicotinamide core. This core has been substituted at the 3-position of the aniline with benzimidazoles to provide compounds such as 9 (Hh-Antag691) and 10 [38,39]. Compound 9 is reported to have IC50 o0.001 mM in a Gli-Luc RGA and has been shown to eliminate medulloblastoma in Ptch1+/ p53/ mice [40]. Substitution at the 3-position with quinoxalines provided compounds such as 11 (IC50 o1 mM in Gli-Luc RGA) [41]. Pyrazines (e.g., 12, IC50 ¼ 0.01 mM in an AP assay) [42] and pyridines (e.g., 13, IC50 ¼ 3 nM in a Gli-Luc RGA) [43,44] have also been exemplified at the 3-position. Amide substitution at the
328
Stefan Peukert and Karen Miller-Moslin
3-position has also been reported, affording inhibitors such as 14 [45]. N-phenyl-benzamides that lack the 4-chloro substituent, such as 15 (IC50 o0.003 mM in an AP assay), have also been reported [46]. Cyclization of the amide to afford 1-aminoisoquinolines such as 16 has been described in a recent patent application. Such compounds afforded EC50o 500 nM in a Gli-Luc RGA [47]. Cl O O
N
N H NH O
9: Hh-Antag691
N
Cl
O
Cl O N
O N
N H
N H
N N F
NH
F
10
Cl
Cl
O N
O
N
N H
H2N
Cl
N H
F
N
O
11
F
O
N
S
F F
12
O
13: GDC-0449
Cl O HN
N H
O N
N
N
N
O
N H O
NH
14
15
N
Hedgehog Signaling Pathway Inhibitors as Cancer Therapeutics
329
Cl N H N
N H N
N 16
O
A related series of biphenylcarboxamides, represented by 17, has also appeared in the recent patent literature [48,49]. When dosed orally for 10 days at 50 mg/kg/day, 17 completely blocks the expansion of luciferased lymphoma cells in vivo. Migration of the carboxamide substituent in these compounds to the ortho position is tolerated, as demonstrated by ortho-biphenyl carboxamides such as 18, which have been reported as potent antagonists of Smoothened [50,51]. Compound 18 showed excellent activity in a Gli-Luc RGA (IC50 ¼ 10 nM), as well as direct binding assays using either mouse or human Smo membranes (IC50 ¼ 9 and 7 nM, respectively). SAR showed that a -CH2- linkage between the aminoindane moiety, and the heteroaryl ring was superior to an amide or sulfonamide linkage, and that the (S)-aminoindane enantiomer afforded a 50-fold improvement in binding relative to the (R)-enantiomer. Various aryl and heteroaryl substituents were tolerated on the methylene unit. F O
F F F
F
F
H N
O H N O
N H
N
18
N 17
S
O
Benzamides substituted with a saturated ring also are potent Smo inhibitors, as demonstrated by the appearance of compounds such as 19 in the patent literature [52]. Compound 19 afforded 114.1% inhibition at 2 mM in a Gli-Luc RGA. The cyclohexane core can also be successfully replaced with nitrogen-containing heterocycles, for example, compound 20 (114.0% inhibition at 2 mM) [53].
330
Stefan Peukert and Karen Miller-Moslin F
F F
N
N
N
N H
O
N H
N H
19
O N H
20
N H
O
O
Another distinct class of Smo inhibitors that appeared in the recent patent literature are heteroaryl-piperazines such as 21 [54] and 22 [55,56]. Compound 21 showed dose-related anti-tumor activity and Gli1 inhibition when administered orally in a Ptch+/ p53/ mouse medulloblastoma allograft model. SAR revealed that the phthalazine was superior to several other heterocyclic systems (e.g., isoquinoline, indole) and that substitu;tion at the 5-position of the pyridine ring had a significant impact on binding affinity [57]. Oral dosing of 22 at 10 mg/kg/day for 6 days in a Ptch+/ p53/ mouse medulloblastoma allograft model afforded a W99% reduction in tumor size relative to vehicle control. Triazole derivatives such as 23 have also been reported as Smoothened antagonists [58]. An unspecified compound from this patent application led to tumor shrinkage in a Ptch+/ mouse medulloblastoma allograft model when dosed orally at 80 mg/kg bid. OH
F F
O O
N N
N
N
N
N
N
N
N N
N N N
N F F F
F F
21
22
F 23
Hedgehog Signaling Pathway Inhibitors as Cancer Therapeutics
331
3.2 Non-Smoothened inhibitors 3.2.1 Hedgehog protein The extracellular protein Shh that binds to the transmembrane receptor Ptch, reversing its inhibitory effect on Smo, is the target of the macrocycle robotnikinin (24) [59]. A small-molecule microarraybased screen of a bacterially expressed biologically active Shh N-terminal fragment (ShhN) provided a macrocyclic hit. Optimization of this hit resulted in the identification of robotnikinin. The compound binds to ShhN with a Kd of 3.1 mM and inhibits Hh signaling in a Gli-luciferase reporter gene cell line, in human primary keratinocytes, and in a synthetic model of the human skin in a dose-dependent fashion. The authors suggest a mechanism involving inhibition of the actions of Shh, either directly or indirectly, by interfering with a precursor complex.
3.2.2 Gli-mediated transcription Two low-molecular weight compounds, 25 (GANT58) and 26 (GANT61), were identified, which inhibit Hh signaling downstream of Smo and SuFu at the level of Gli with an IC50 of B5 mM in a Gli-luciferase cellular assay [60]. Mechanistically, both inhibitors act at the nucleus to block Gli function, and one of them, 26, interferes with DNA binding of Gli1. Both compounds display selectivity for the Hh pathway over several unrelated signal transduction pathways such as the TNF/NFkB signaling, glucocorticoid receptor gene transactivation, and the Ras-Raf-MekMapk cascade. Subcutaneous application of 26, dosed at 25 mg/kg/day for 18 days, in a human prostate cancer xenograft mouse model induced growth regression with concurrent strongly reduced mRNA levels of Ptch. Screening of natural products in a cell-based reporter assay of Gli1-mediated transcription provided several hits. The physalins F (27) and B (28) were the most potent inhibitors of Gli1-mediated activity (IC50 values of 0.66 mM and 0.62 mM, respectively) [61]. Additionally, these two compounds also reduced Gli2-mediated transcription in a separate cell assay (IC50 values of 1.5 mM and 1.4 mM, respectively). Compounds 27 and 28 were cytotoxic in a PANC1 human pancreatic cancer cell line (IC50 values of 2.6 and 5.3 mM) and decreased the mRNA expression of Gli1, 2 and Ptch genes. The authors did not identify the molecular targets but point out that physalins are known modulators of the NFkB cascade, and the mechanism of their Gli-mediated transcriptional inhibition might include a path of PKC inhibition.
332
Stefan Peukert and Karen Miller-Moslin O O
N
O
N
N H
N
S
N
N
O N
N
HN N
24: robotnikinin
N
25: GANT58
Cl
N
O
O O HO H
N
H
H
O
26: GANT61
H
O O
O
O
HO
O
H
H
O
OEt
N
O
H
H
S
O
N H
O
27: physalin F
28: physalin B
29: JK184
3.2.3 Microtubule depolymerization JK184 (29) potently inhibited Hh signaling in a Gli-luciferase cellular assay (IC50 ¼ 30 nM); this inhibition was confirmed by measuring the mRNA levels of Gli1 and Ptch1 by quantitative reverse transcription–polymerase chain reaction (RT–PCR) in this cell line [62]. Compound 29 was not found to bind to Smo, but rather binds directly to alcohol dehydrogenase 7 (Adh 7) with a Kd of 348 nM, and was shown to inhibit the oxidation of retinol by this enzyme with similar potency. The oxidation of retinol by Adh7 is part of the retinoic acid (RA) signaling pathway, which in mouse embryos affects Hh signaling. However, recent studies demonstrated that compound 29 exerts its effects on the Hh pathway by destabilizing microtubules, and its inhibition of Adh7 is an ancillary activity. In this model, 29 disrupts microtubule-dependent processes that are required for the conversion of endogenous full-length Gli proteins into functional transcriptional activators [63].
4. HEDGEHOG INHIBITORS IN CLINICAL TRIALS The first Hh inhibitor studied in humans was the natural product cyclopamine (1) for the topical treatment of BCC. Tas- and Avci reported
Hedgehog Signaling Pathway Inhibitors as Cancer Therapeutics
333
rapid regression of facial BCC in four out of four patients without adverse effects [17]. In 2005, a phase 1 clinical trial was initiated with Cur61414 (5) for the topical treatment of BCC [64]. The trial was halted a year later, as the compound did neither produce significant clinical changes nor downregulate the pharmacodynamic marker Gli1, leading to speculations that the drug candidate may not have adequately penetrated the human skin [65]. A second compound, GDC-0449 (13), provided proof-of-concept in 2008 in a first-in-human study in patients with metastatic or locally advanced BCC. In eight of nine patients, the compound caused either partial responses or stable disease upon oral administration, and mRNA levels of Gli1 were reduced in skin biopsies. The compound seemed to have minimal toxicity and was well tolerated at doses of up to 540 mg/ day. The long half-life of the drug caused marked accumulation and provided a steady-state plasma concentration of around 20–30 mM at all three doses tested (150, 270 and 540 mg) [18,66]. A pivotal phase 2 study is underway to evaluate the efficacy of this compound in patients with either metastatic BCC or locally advanced, inoperable BCC [67]. Additional phase 2 clinical trials will also investigate the anti-tumor effects of 13 in patients with metastatic colorectal and ovarian cancers [68,69]. Furthermore, the safety and pharmacokinetics of this compound are being evaluated in children with medulloblastoma in a clinical trial sponsored by the National Cancer Institute [70]. IPI-926 (3), a cyclopamine derivative with better biophysical and pharmacokinetic properties than the natural product, entered a phase 1 study in 2008 in patients with solid tumor malignancies [71]. Around the same time, BMS-833923 (XL-139, structure not disclosed) entered phase 1 clinical trials in patients with solid tumors, including sporadic BCC [72]. The latest Hh/Smo inhibitor to enter clinical trials is LDE225 (structure not disclosed), which started phase 1 studies in early 2009, again in patients with solid tumors [73].
5. CONCLUSIONS The search for Hh inhibitors of therapeutic benefit has resulted in Smoothened antagonist 13, which provided proof-of-concept in a phase 1 study in advanced and metastatic BCC patients and was well tolerated at efficacious doses [18]. At least three additional and structurally distinct compounds have entered clinical trials. The clinical results in BCC patients and the impressive tumor regression observed in murine medulloblastoma models with various compounds provide promise that Smoothened inhibitors will become useful monotherapy agents against
334
Stefan Peukert and Karen Miller-Moslin
these two genetically driven tumor types. In other tumor types, which either show a paracrine requirement for Hh signaling in tumorigenesis [13] or where cancer stem cells play an important role [74], Hh inhibitors may be more appropriate as an adjunct therapy. Combination treatment might also be a suitable therapeutic strategy for cases in which Hh inhibitors show efficacy against metastatic spread but only modest growth inhibition against the primary tumor [29]. Because the Hh signaling pathway is essential in the development and tissue homeostasis, great care should be taken to evaluate the safety profile of these drugs. Phase 1 clinical results for 13 indicate that the compound is well tolerated [75], but a study looking at the effects of 9 in young mice (10–14 days) showed permanent defects in bone growth [76]. This effect was not seen in adult mice [40]. Although it is unclear whether and how these bone toxicities in young mice may translate into effects in children, the cost versus benefit needs to be carefully determined for a treatment in a pediatric population such as medulloblastoma cancer patients.
REFERENCES [1] [2] [3] [4] [5] [6] [7] [8]
[9] [10]
[11]
[12]
[13]
[14] [15] [16]
M. C. Fishman and J. A. Porter, Nature, 2005, 437, 491. C. Nu¨sslein-Vollhard and E. Wieschaus, Nature, 1980, 287, 795. M. Pasca di Magliano and M. Hebrok, Nat. Rev. Cancer, 2003, 3, 903. P. A. Beachy, S. S. Karhadkar and D. M. Berman, Nature, 2004, 432, 324. L. Lum and P. A. Beachy, Science, 2004, 304, 1755. J. P. Incardona, W. Gaffield, R. P. Kapur and H. Roelink, Development, 1998, 125, 3553. M. K. Cooper, J. A. Porter, K. E. Young and P. A. Beachy, Science, 1998, 280, 1603. R. L. Johnson, A. L. Rothman, J. Xie, L. V. Goodrich, J. W. Bare, J. M. Bonifas, A. G. Quinn, R. M. Myers, D. R. Cox, E. H. Epstein, Jr. and M. P. Scott, Science, 1996, 272, 1668. Y. Lee, H. L. Miller, P. Jensen, R. Hernan, M. Connelly, C. Wetmore, F. Zindy, M. F. Roussel, T. Curran, R. J. Gilbertson and P. J. McKinnon, Cancer Res., 2003, 63, 5428. M. C. Thompson, C. Fuller, T. L. Hogg, J. Dalton, D. Finkelstein, C. C. Lau, M. Chintagumpala, A. Adesina, D. M. Ashley, S. J. Kellie, M. D. Taylor, T. Curran, A. Gajjar and R. J. Gilbertson, J. Clin. Oncol., 2006, 24, 1924. M. R. Gailani, M. Stahle-Backdahl, D. J. Leffell, M. Glynn, P. G. Zaphiropoulos, C. Pressman, A. B. Unden, M. Dean, D. E. Brash, A. E. Bale and R. Toftga˚rd, Nat. Genet., 1996, 14, 78. J. Xie, M. Murone, S.-M. Luoh, A. Ryan, Q. Gu, C. Zhang, J. M. Bonifas, C.-W. Lam, M. Hynes, A. Goddard, A. Rosenthal, H. E. Epstein, Jr. and F. J. de Savage, Nature, 1998, 391, 6662. R. L. Yauch, S. E. Gould, S. J. Scales, T. Tang, H. Tian, C. P. Ahn, D. Marshall, L. Fu, T. Januario, D. Kallop, M. Nannini-Pepe, K. Kotkow, J. C. Marsters, Jr., L. L Rubin and F. J. de Sauvage, Nature, 2008, 455, 405. J. Xie, Curr. Oncol. Rep., 2008, 10, 107. L. L. Rubin and F. J. de Savage, Nat. Rev. Drug Disc., 2006, 5, 1026. E. H. Epstein, Jr., Nat. Rev. Cancer, 2008, 8, 743.
Hedgehog Signaling Pathway Inhibitors as Cancer Therapeutics
[17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28]
[29]
[30] [31] [32]
[33]
[34] [35] [36] [37]
[38] [39] [40] [41] [42] [43]
335
S. Tas- and O. Avci, Eur. J. Dermatol., 2004, 14, 96. D. D. Van Hoff, Proc. 99th Annu. Meeting Am. Assoc. Cancer Res., 2008, Abstract LB-138. R. Rohatgi and M. P. Scott, Nat. Cell Biol., 2007, 9, 1005. M. Varjosalo and J. Taipale, J. Cell. Sci., 2007, 120, 3. J. Rennie, Sci. Am., 1994, 270, 4. R. Rohatgi, L. Milnekovic and M. P. Scott, Science, 2007, 317, 372. N. Mahindroo, C. Punchihewa and N. Fujii, J. Med. Chem., 2009, 52, 3829. A. S. Kiselyov, Anti-Canc. Agents Med. Chem., 2006, 6, 445. G. V. Borzillo and B. Lippa, Curr. Top. Med. Chem., 2005, 5, 147. R. W. King, J. Biol., 2002, 1, 8. J. K. Chen, J. Taipale, M. K. Cooper and P. A. Beachy, Genes Dev., 2002, 16, 2743. M. R. Tremblay, M. Nevalainen, S. J. Nair, J. R. Porter, A. C. Castro, M. L. Behnke, L.-C. Yu, M. Hagel, K. White, K. Faia, L. Grenier, M. J. Campbell, J. Cushing, C. N. Woodward, J. Hoyt, M. A. Foley, M. A. Read, J. R. Sydor, J. K. Tong, V. J. Palombella, K. McGovern and J. Adams, J. Med. Chem., 2008, 51, 6646. G. Feldmann, V. Fendrich, K. McGovern, D. Bedja, S. Bisht, H. Alvarez, J.-B. M. Koorstra, N. Habbe, C. Karikari, M. Mullendore, K. L. Gabrielson, R. Sharma, W. Matsui and A. Maitra, Mol. Cancer Ther., 2008, 7, 2725. M. J. Grogan and M. Tremblay, WO Patent Application 2008/109184 A1, 2008. B. Austad, S. Janardanannair, A. Lescarbeau, M. Tremblay, D. Grayzel and W. Matsui, WO Patent Application 2008/083252 A2, 2008. M. R. Tremblay, A. Lescarbeau, M. J. Grogan, E. Tan, G. Lin, B. C. Austad, L.-C. Yu, M. L. Behnke, S. J. Nair, M. Hagel, K. White, J. Conley, J. D. Manna, T. M. Alvarez-Diez, J. Hoyt, C. N. Woodward, J. R. Sydor, M. Pink, J. MacDougall, M. J. Campbell, J. Cushing, J. Ferguson, M. S. Curtis, K. McGovern, M. A. Read, V. J. Palombella, J. Adams and A. C. Castro, J. Med. Chem., 2009, 52, 4400. J. A. Williams, O. M. Guichert, B. I. Zaharian, Y. Xu, L. Chai, H. Wichterle, C. Kon, C. Gatchalian, J. A. Porter, L. L. Rubin and F. Y. Wang, Proc. Nat. Acad. Sci. U.S.A., 2003, 100, 4616. A. D. Baxter, E. A. Boyd, O. M. Guichert, S. Price and L. Rubin, Patent Application WO 01/26644 A2, 2001. S. A. Brunton, J. H. A. Stibbard, L. L. Rubin, L. I. Kruse, O. M. Guicherit, E. A. Boyd and S. Price, J. Med. Chem., 2008, 51, 1108. A. D. Baxter, E. A. Boyd, O. M. Guichert, S. Price and L. D. Rubin, WO Patent Application 01/19800 A2, 2001. S. Bahceci, W. Bajjalieh, J. Chen, S. Epshteyn, T. P. Forsyth, T. P. Huynh, B. G. Kim, J. W. Leahy, M. S. Lee, G. L. Lewis, M. B. Mac, G. Mann, C. K. Marlowe, B. H. Ridgway, J. C. Sangalang, X. Shi, C. S. Takeuchi and Y. Wang, WO Patent Application 2008/112913 A1, 2008. L. Rubin, O. M. Guichert, S. Price and E. A. Boyd, WO Patent Application 03/011219 A2, 2003. O. M. Guichert, E. A. Boyd, S. A. Brunton, S. Price, J. H. A. Stibbard and C. H. MacKinnon, Patent Application WO 2006/050506 A1, 2006. J. T. Romer, H. Kimura, S. Magdaleno, K. Sasai, C. Fuller, H. Baines, M. Connelly, C. F. Stewart, S. Gould, L. L. Rubin and T. Curran, Cancer Cell, 2004, 6, 229. M. F. T. Koehler, R. Goldsmith and D. P. Sutherlin, WO Patent Application 2006/078283 A2, 2006. B. Fauber, A. Hird, J. Janetka, D. J. Russell and B. Yang, WO Patent Application 2009/ 030952 A2, 2009. J. Gunzner, D. Sutherlin, M. Stanley, L. Bao, G. Castanedo, R. Lalonde, S. Wang, M. Reynolds, S. Savage, K. Malesky and M. Dina, WO Patent Application 2006/028958, 2006.
336
Stefan Peukert and Karen Miller-Moslin
[44] D. P. Sutherlin, Abstracts of Papers, 237th ACS National Meeting, Salt Lake City, UT, March 22–26, 2009, MEDI-177. [45] R. A. Goldsmith, D. P. Sutherlin, K. D. Robarge and A. G. Olivero, WO Patent Application 2007/059157 A1, 2007. [46] L. Dakin, B. Fauber, A. Hird, J. Janetka, D. J. Russell, Q. Su, B. Yang and X. Zheng, WO Patent Application 2009/027746 A1, 2009. [47] D. Cheng, D. Han, W. Gao, J. Jiang, S. Pan and Y. Wan, WO Patent Application 2008/ 014291 A2, 2008. [48] W. Gao, J. Jiang, Y. Wan, D. Cheng, D. Han, X. Wu and S. Pan, WO Patent Application 2007/131201 A2, 2007. [49] C. Dierks, M. Warmuth and X. Wu, WO Patent Application 2008/154259 A1, 2008. [50] S. Peukert, R. K. Jain, A. Geisser, Y. Sun, R. Zhang, A. Bourret, A. Carlson, J. DaSilva, A. Ramamurthy and J. F. Kelleher, Bioorg. Med. Chem. Lett., 2009, 19, 328. [51] R. K. Jain, J. Kelleher, S. Peukert and Y. Sun, WO Patent Application 2007/120827 A2, 2007. [52] M. J. Munchhof, L. A. Reiter, A. Shavnya, C. S. Jones, Q. Li and R. G. Linde, WO Patent Application 2008/075196 A1, 2008. [53] M. J. Munchhof, L. A. Reiter, S. D. La Greca, C. S. Jones and Q. Li, US Patent Application 2009/0005416-A1, 2009. [54] M. Dai, F. He, R. K. Jain, R. Karki, J. Kelleher, J. Lei, L. Llamas, M. A. McEwan, K. Miller-Moslin, L. B. Perez, S. Peukert and N. Yusuff, WO Patent Application 2008/ 110611 A1, 2008. [55] R. J. Austin, J. Kaizerman, B. Lucas, D. L. McMinn and J. Powers, WO Patent Application 2009/002469 A1, 2009. [56] J. Kaizerman, B. Lucas, D. L. McMinn and R. Zamboni, WO Patent Application 2009/ 035568 A1, 2009. [57] K. Miller-Moslin, S. Peukert, R. K. Jain, M. A. McEwan, R. Karki, L. Llamas, N. Yusuff, F. He, Y. Li, Y. Sun, M. Dai, L. Perez, W. Michael, T. Sheng, H. Lei, R. Zhang, J. Williams, A. Bourret, A. Ramamurthy, J. Yuan, R. Guo, M. Matsumoto, A. Vattay, W. Maniara, A. Amaral, M. Dorsch and J. F. Kelleher, J. Med. Chem., 2009, 52, 3954. [58] J. M. Balkovec, R. Thieringer and S. T. Waddell, US Patent Application 2008/262051 A1, 2008. [59] B. Z. Stanton, L. F. Peng, N. Maloof, K. Nakai, X. Wang, J. L. Duffner, K. M. Taveras, J. M. Hyman, S. W. Lee, A. N. Koehler, J. K. Chen, J. L. Fox, A. Mandinova and S. L. Schreiber, Nat. Chem. Biol., 2009, 5, 154. ˚ . Bergstro¨m, T. Shimokawa and R. Toftga˚rd, Proc. Natl. Acad. Sci. U.S.A., [60] M. L. Lauth, A 2007, 104, 8455. [61] T. Hosoya, M. A. Arai, T. Koyano, T. Kowithayakorn and M. Ishibashi, Chem. Bio. Chem., 2008, 9, 1082. [62] J. Lee, X. Wu, M. P. di Magliano, E. C. Peters, Y. Wang, J. Hong, M. Hebrok, S. Ding, C. Y. Cho and P. G. Shultz, Chem. Bio. Chem., 2007, 8, 1916. [63] T. Cupido, P. G. Rack, A. J. Firestone, J. M. Hyman, K. Han, S. Sinha, C. A. Ocasio and J. K. Chen, Angew. Chem. Int. Ed., 2009, 48, 2321. [64] Curis press release, June 8, 2005, http://phx.corporate-ir.net/phoenix.zhtml?c ¼ 123198&p¼ irol-newsArticle_print&ID ¼ 718277&highlight ¼ [65] Curis press release, July 11, 2006, http://phx.corporate-ir.net/phoenix.zhtml?c ¼ 123198&p¼ irol-newsArticle&ID ¼ 881050&highlight ¼ [66] F. de Sauvage, Targeting the Hedgehog pathway in cancer: From bench to clinic, DanaFarber Cancer Institute, October 28, 2008, Boston. [67] A study evaluating the efficacy and safety of GDC-0449 (Hedgehog pathway inhibitor) in patients with advanced basal cell carcinoma, http://clinicaltrials.gov/ct2/show/ NCT00833417?term ¼ GDC-0449&rank ¼ 1
Hedgehog Signaling Pathway Inhibitors as Cancer Therapeutics
337
[68] A study of GDC-0449 (Hedgehog pathway inhibitor) with concurrent chemotherapy and bevacizumab as first-line therapy for metastatic colorectal cancer, http:// clinicaltrials.gov/ct2/show/NCT00636610?term ¼ GDC-0449&rank ¼ 4 [69] A study of GDC-0449 (Hedgehog pathway inhibitor) as maintenance therapy in patients with ovarian cancer in a second or third complete remission, http:// clinicaltrials.gov/ct2/show/NCT00739661?term ¼ GDC-0449&rank ¼ 5 [70] GDC-0449 in treating young patients with medulloblastoma that is recurrent or did not respond to previous treatment, http://clinicaltrials.gov/ct2/show/NCT00822458?term¼ GDC-0449&rank ¼ 2 [71] A phase 1 study of IPI-926 in patients with advanced and/or metastatic solid tumor malignancies, http://clinicaltrials.gov/ct2/show/NCT00761696?term¼ IPI-926 &rank¼ 1 [72] A phase 1 study of BMS-833923 (XL139) in subjects with advanced or metastatic cancer, http://clinicaltrials.gov/ct2/show/NCT00670189?term ¼ XL-139&rank ¼ 1 [73] Novartis new molecules presentation, AACR 100th Annual Meeting, Denver, CO, April 18–22, 2009. [74] I. Ischenko, H. Seeliger, M. Schaffer, K.-W. Jauch and C. J. Bruns, Curr. Med. Chem, 2008, 15, 3171. [75] P. M LoRusso, C. M. Rudin, M. J. Borad, L. Vernillet, W. C. Darbonne, H. Mackey, J. F. DiMartino, F. de Savage, J. A. Low and D. D. Von Hoff, J. Clin. Oncol. (Meet. Abstr.), 2008, 26, 3516. [76] H. Kimura, J. M. Y. Ng and T. Curran, Cancer Cell, 2008, 13, 249.
CHAPT ER
17 Emerging Therapies Based on Inhibitors of PhosphatidylInositol-3-Kinases John M. Nuss, Amy Lew Tsuhako and Neel K. Anand
Contents
1. Introduction 2. Inhibitors of PI3Ks: Early Studies 3. Clinically Investigated Pan-Active Class I PI3K Inhibitors 4. Additional Pan-Active Class I PI3K Inhibitors 5. Clinically Investigated Isozyme-Selective PI3K Inhibitors 6. Pre-Clinical Isozyme-Selective PI3K Inhibitors 7. Conclusions References
339 341 343 347 347 349 350 351
1. INTRODUCTION The superfamily of lipid kinases collectively referred to as phosphoinositide kinases (PIKs, including PI3K, PI4K, and others) has been an area of intense investigation due to the crucial role of members of this class in various signal transduction–mediated events [1–5]. As the basic biology of the PIKs has become better understood, their roles in various pathophysiologies such as cancer, inflammation, and cardiovascular disease are also becoming clearer, leading to the development of agents targeting the function of these kinases as new modalities for the treatment of a number of disease states. The PIKs catalyze the conversion of phosphatidyl inositol (PI) to various phosphoinositides [1–5]. Of particular importance is the Exelixis, Inc., South San Francisco, CA 94080, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04417-0
r 2009 Published by Elsevier Inc.
339
340
John M. Nuss et al.
phosphoinositide 3-kinase (PI3K) catalyzed formation of phosphatidylinositol-3,4,5-triphosphate (PIP3) by phosphorylation of phosphatidylinositol-4,5-bisphosphate (PIP2); this second messenger is a critical regulator of cellular functions such as metabolism, cell growth, differentiation, and chemotaxis. A simplified view of PI3K signaling is shown in Figure 1. Activation of PI3Ks, either by growth factor (e.g., insulin, insulin-like growth factor (IGF), epidermal growth factor (EGF)) receptor binding or by G-proteincoupled receptor (GPCR) activation, results in the formation of PIP3, which recruits the kinases 3-phosphoinositide dependent kinase (PDK) and protein kinase B (PKB, aka AKT) to the cell membrane; AKT is activated through phosphorylation by the membrane-bound PDK. Activation of AKT effects a myriad of downstream processes, including modulation of numerous proteins such as mTOR, GSK3, forkhead, NFkB transcription factors, and eNOS, which are involved in processes such as metabolism, cell growth, survival, and angiogenesis [6]. Dysregulation of these processes is crucial for the pathophysiology of several diseases, as attenuated signaling of the insulin receptor is a major contributor to type-2 diabetes and mutations that lead to amplification of PI3K signaling are among the most common mutations in human cancers [7,8]. PI3K signaling is also negatively regulated by the tumor suppressor PTEN (phosphatase and tensin homolog deleted on chromosome 10); this phosphatase is responsible for the dephosphorylation of PIP3 to PIP2 [9]. PTEN has been shown to be mutated or absent in many human cancers, further implicating aberrant PI3K signaling in cancer.
Figure 1
Schematic of PI3K signaling. (See Color Plate 17.1 in Color Plate Section.)
Emerging Therapies Based on Inhibitors of Phosphatidyl-Inositol-3-Kinases
341
PI3Ks can be categorized into three classes, I, II, and III [1–5,10]. The most thoroughly studied class, the class I PI3Ks, are heterodimeric proteins, with each member of this family containing a smaller regulatory domain and a larger 110-kD catalytic domain in each of the four isoforms, p110a, p110b, p110g, and p110d. These have been further divided into two sub-classes, the class Ia (p110a, p110b, p110d) and the class Ib (p110g). The class Ia catalytic domains combine with SH2-domain containing 85-kD regulatory subunits and are predominantly activated by growth factor receptor tyrosine kinases such as the insulin receptor tyrosine kinase, whereas the class Ib isoforms generally combine with a 101-kD regulatory domain and are activated as a result of GPCR signaling. Structurally, the catalytic domains of the PI3Ks display the highly conserved domain structure of the protein kinases, having two lobes, an N-terminal lobe composed of five b-sheets and three a-helices and a C-terminal lobe consisting of mainly a-helices with a flexible linker joining the two lobes [11]. The class II and III PI3Ks have not garnered much attention from the perspective of inhibitor design and pharmacological study and will not be discussed in detail here [12]. Additionally, a number of related kinases, including the serine/threonine kinases mTOR, DNA dependent protein kinase (DNA-PK), and ataxia telangiectasia mutated kinase (ATM), are classified as PI3K-related kinases, or class IV PI3Ks, due to gross structural similarities to the PI3Ks. This similarity often impacts inhibitor design, as cross-reactivity with these kinases is often observed [4–6]. In fact, the intentional design of mTOR/PI3K dual inhibitors is an area of intense investigation, and several compounds having this profile are currently under clinical evaluation as cancer therapeutics (vide infra). Functional elucidation of the class I PI3K isozymes is an extremely active area of investigation. Studies have linked p110g and p110d to chronic inflammation [13], p110b to thrombosis [14], and p110a to many tumor types including ovarian, breast, gastric, and colon cancers [15]. PI3Ka and -b are distributed in many tissues, in contrast to the more narrowly distributed PI3Kg and -d, which are mainly found in leukocytes, endothelial cells, and smooth muscle cells [16]. Importantly, genetic studies suggest that mutation in the p110a gene is oncogenic [17,18]. Consequently, p110a is considered to be a highly validated target for human cancers and has been an extremely active area of drug discovery.
2. INHIBITORS OF PI3Ks: EARLY STUDIES Given the crucial role that the dysregulation of PI3Ks, and in particular, the amplification of PI3K-derived signaling, plays in various pathophysiologies, it is not surprising that considerable effort has gone into the
342
John M. Nuss et al.
Figure 2 Interaction of LY294002 with ATP-binding site of PI3Kg. (See Color Plate 17.2 in Color Plate Section.)
design and pharmacological characterization of both pan-active and isozyme-selective inhibitors of the PI3Ks [3–6]. As with a majority of protein kinase inhibitors, most reported inhibitors of the PI3Ks target the ATP-binding pocket in the catalytic domain. Some of the early inhibitors that have been reported include LY294002, 1 [19], and wortmannin, 2 [20]. Many inhibitors of the PI3Ks have their genesis in the quercetin derivative 1, a promiscuous kinase inhibitor that was found to be an ATP-competitive, pan-active inhibitor of class I PI3Ks [19]. The morpholine oxygen forms the critical H-bonding interaction with the Val882 (PI3Kg notation) residue on the flexible hinge region of the kinase in the ATP-binding pocket; the carbonyl group stabilizes the binding further by forming two hydrogen bonds, one with the backbone NH of Asp964 in the activation loop and one with the Lys833 of the salt bridge (Figure 2) [21]. Many groups have attempted to optimize these interactions to design ‘‘morpholine’’-based inhibitors of PI3Ks having enhanced potency and selectivity [22]. O
O O MeO
O
O
17
N O
O
H O
O 20 1
O 2
The steroidal furan 2 and its derivatives have been found to be PI3K inhibitors [20]. It potently inhibits all of the class I PI3Ks with IC50o10 nM; this natural product also inhibits the PI3K-related kinases
Emerging Therapies Based on Inhibitors of Phosphatidyl-Inositol-3-Kinases
343
mTOR and DNA-PK, inter alia. Structural and mechanistic studies have shown that 2 binds to the ATP-catalytic site, the critical interaction being the formation of a H-bond between the hinge residue Val882 and the C-17 carbonyl in the D-ring (steroid notation) [23,24]. In contrast to 1, 2 binds irreversibly to PI3Kg by forming a covalent adduct between Lys833 in the catalytic site and the electrophilic site C-20 in 2, in turn leading to destruction of the catalytic ability of the kinase. Despite the general reluctance of the pharmaceutical industry to develop irreversible enzyme inhibitors, derivatives of 2 have been investigated clinically (vide infra). Structural studies to provide insight into ligand–enzyme interactions to facilitate inhibitor design are just emerging. Of the B20 PI3K structures that have been reported in the Protein Data Base Repository (PDB) as of April 2009 [25], the only ligand-bound structures disclosed involve PI3Kg structures. The PI3Ka apo-enzyme structure has been solved [26]; however, no ligand-bound structures have been reported. Structures of the other isozymes have not been disclosed. It is expected that intensified structural efforts in this area will greatly facilitate the design of both pan-active and isozyme-selective inhibitors. As a result of extensive efforts in the design of kinase inhibitors and screening, a diverse array of scaffolds have also been reported to inhibit the lipid PI3Ks. The number and types of PI3K inhibitors continues to increase rapidly. Both pan-active and isozyme-selective inhibitors have been described and are discussed in the next section.
3. CLINICALLY INVESTIGATED PAN-ACTIVE CLASS I PI3K INHIBITORS SF1126 (3) – Although 1 emerged as an invaluable tool in establishing the biological role of PI3K in human cancer, poor physicochemical properties and short plasma half-life made it unsuitable for further development. Attempts to improve the pharmacokinetic profile of 1 have led to the development of a water soluble prodrug 3 (Figure 3), currently in phase I clinical trials for cancer. Prodrug 3 is a peptide conjugate of 1 that specifically targets cell-surface integrins within the tumor. Although 3 demonstrates only B100–500 nM activity against various PI3K isoforms, significant inhibition of tumor growth was observed in various xenograft models (U87MG, PC-3, U251, and U251vIII) due to the high levels of accumulation of the drug in the tumor tissue [27]. It should also be noted that not only does 3 inhibit all forms of PI3Ks, but it also inhibits mTOR, DNA-PK, PLK-1, PIM-1, and CK2. This drug is reported to be well tolerated in patients with solid tumors treated twice a week for 4 weeks by i.v. infusion. In this study, 3 of the 12 patients experienced stable disease [28].
344
John M. Nuss et al.
O
RGDS
O MeO
O O
O
H O
O
N+
O
O N
O O
N
O
N
N N
OH O
N
RGDS=Ar g-Gly -Glu-Ser
3
4
O
5 O
SO2Me N
N N HN
N N
S
N
O N S
N HO
N
N
6
N N
7 N
O
N HO
8
N
F
OMe N
O
F
S N
EtHN N
N
O
N
N
O
9
OMe
N
HN S
N N H
N
N
N H
OMe
O
10
11 O
O HN
N
S N
R1HN
O
N
N H
X
12 HN N
R1
R2
N
R1
R2
N
15
Figure 3
N N
Pan-active PI3K inhibitors.
N
R3
N
N
R2
13
R1HN
N
14
OR2
Emerging Therapies Based on Inhibitors of Phosphatidyl-Inositol-3-Kinases
345
PX-866 (4) – As previously mentioned, the irreversible inhibitor 2 is one of the earliest discovered inhibitors of the PI3K pathway [20,29,30]; multiple groups have attempted to make derivatives with an improved pharmacokinetic profile and therapeutic index. The wortmannin derivative 4 (Figure 3) is currently in phase I clinical trials [31,32]. This broad spectrum PI3K inhibitor demonstrated improved stability and hepatotoxicity compared to 2 [33,34]. In addition, 4 showed tumor regression in human ovarian and colon carcinoma xenograft models, as well as nonsmall-cell lung cancer resistant to gefitinib. However, in mice treated with 4, increases in both blood insulin and glucose are observed. This hyperglycemic effect can be moderated with pioglitazone, a PPARg agonist approved for the treatment of diabetes [35]. In July 2008, clinical trials were initiated in patients with advanced solid tumors [36]. This is currently the only irreversible PI3K inhibitor in clinical trials. NVP-BEZ235 (5) and BGT226 (structure undisclosed) – BGT226 and 5 (Figure 3) are two dual PI3K/mTOR inhibitors currently undergoing trials for breast and other solid tumors [37]. Both are reported to be reversible, ATP competitive inhibitors of, inter alia, the class I PI3Ks as well as the PIK-related kinase, mTOR. 5 inhibits PI3K a, b, g, d at 4, 76, 7, and 5 nM respectively, as well as mTOR (21 nM), but not DNA-PK [38,39]. In addition, 5 potently inhibits cellular proliferation for both wild-type and oncogenic p110-a mutated cells and in multiple myeloma [40]. This potent antiproliferative effect has been shown to translate into tumor regression in xenograft models, where the regressions are wellcorrelated with the inhibition of 437S-p-AKT and other PI3K pathway markers [41–43]. In addition, recent data suggest that hyperactivation of the PI3K pathway leads to lapatinib resistance, which can be reversed by 5 [44]. Of concern for anti-PI3K therapy is the potential induction of insulin resistance. However, although 5 is a pan PI3K/mTOR inhibitor, no effect on insulin or glucose levels were reported in vivo in rodents given efficacious dosages for 13 weeks, indicating that attaining isozymeselective PI3K inhibitors may not be necessary for developing PI3K inhibitors with manageable effects on glucose regulation [45]. GDC-0941 (6) – This compound (Figure 3) is currently being investigated in multiple phase I trials. It is a pan-selective PI3K inhibitor (3, 33, 75, 3 nM against a, b, g, d) that does not potently inhibit mTOR (580 nM) or DNA-PK (1230 nM) [46,47]. The compound was developed from modifications of previously reported pyrimidine derivatives (Figure 3, 7, 8-aka PI-103) that had PI3Ka inhibition IC50o10 nM, but poor pharmacokinetic profiles [48,49]. Attempts to improve the pharmacokinetic profile took advantage of the available crystal structure of p110g and a p110a homology model [11]. 6 shows broad antiproliferative effects associated with inhibition of pAKT, the downstream signal of PI3K
346
John M. Nuss et al.
inhibition. In animal models (U87MG human glioblastoma xenografts), oral doses of 6 reduces tumor growth by 83%. Of particular note, reduction of pAKT was also observed in patients with solid tumors dosed daily for 3 weeks with 6 [50,51]. Thus far, no reports have emerged regarding tumor regression in humans, but in general, the drug has been reported to be well tolerated. Several patent applications have published describing thienopyrimidines structurally related to 6 [52–55]. XL147 (structure undisclosed) – This compound is currently in phase I clinical trials as a class I PI3K inhibitor (IC50: 39, 383, 23, 36 nM for p110a, b, g, d, respectively) and importantly has no mTOR or DNA-PK activity [56–58]. In pre-clinical studies, XL147 is reported to block PI3K signaling in cultured tumor cells vascular endothelial growth factor (VEGF)induced tubule formation in cultured endothelial cells, and hepatocyte growth factor (HGF)-stimulated migration of B16 melanoma cells. In mouse xenograft models, XL147 showed strong tumor growth inhibition. This antitumor efficacy correlates in a dose-dependent manner with increased apoptosis and inhibition of angiogenesis. In clinical trials, XL147 at doses of up to 600 mg has generally been well tolerated in patients with metastatic or unresectable solid tumors. At the 600 mg daily oral dose, reductions of 70–80% in the phosphorylation of the PI3K pathway markers such as pAKT, pPRAS40, and pS6 have been observed in the patient tumor tissues. To date, 8 of the 23 patients tested in this phase I trial showed prolonged stable disease (W3 months). In addition, glucose levels were stable in the 23 patients, although insulin levels did increase in some patients [59]. XL765 (structure undisclosed) – In contrast to XL147, XL765 inhibits both class I PI3Ks and mTOR (IC50: 39, 113, 43, 9,157 nM for p110a, b, g, d, and mTOR, respectively). In preclinical studies, XL765 potently inhibits PI3K pathway signaling in cultured cells as well as various xenograft tumor models [60]. Data reported from the phase I dose escalation trial of XL765 showed that in 28 patients with metastatic or unresectable solid tumors dosed orally b.i.d., XL765 was well tolerated at 30 mg [61,62]. As with XL147, dosing of XL765 in patients led to an increase in plasma insulin levels, but no change in blood glucose levels. Pharmacodynamic analyses showed significant reductions of 80–90% in pAKT, p4EBP1, and pS6 levels in patient tumor tissue [63], with 5 of the 28 patients enrolled showing stable disease (W3 months) [64]. GSK-1059615 or GSK-615 (9) – The PI3K inhibitor 9 (Figure 3) inhibits AKT phosphorylation in various tumor cell lines, and especially in T47-D breast carcinoma cell (IC50: 34 nM)[65]. In in vivo studies with xenograft models such as BT474 breast tumor, and HCC1954 breast cancer cell, 9 shows dose-dependent tumor growth inhibition without significant
Emerging Therapies Based on Inhibitors of Phosphatidyl-Inositol-3-Kinases
347
body weight loss or toxicity. GSK-615 is currently in phase I clinical trials for lymphoma, solid tumors, metastatic breast cancer, and endometrial cancer [66].
4. ADDITIONAL PAN-ACTIVE CLASS I PI3K INHIBITORS ZSTK474 (10) – A potent, pan-active, ATP competitive PI3K inhibitor 10 (Figure 3) has been identified (IC50: 16, 44, 46, 49 nM for a, b, g, d, respectively) with considerably lower activity for mTOR and DNA-PK [67,68]. Although 10 is a long way from proving that isoform nonspecific inhibitors can achieve acceptable therapeutic indices, this triazine analog shows strong antiproliferative effects in cells and antitumor activity in in vivo xenograft models without body weight loss [69–72]. Disclosures of compounds structurally related to 10 have recently appeared [73–80]. AEZS-126 (11) – This compound (Figure 3) has been reported to be a potent, orally bioavailable pan-class I PI3K inhibitor [81,82]. It inhibits the class I PI3Ks (51, 3000, 177, 139 nM for a, b, d, g, respectively), but does not inhibit mTOR (12.9 uM) or DNA-PK (W31 uM). In pre-clinical xenograft studies, 11 demonstrated up to 30–50% tumor growth inhibition in HCT116 and PC3 tumor models. In addition to these compounds, many other class I pan-active compound have recently been reported. Although there is a paucity of data associated with many of these reports, a sampling of these structures are given in Figure 3, 12–15, and the reader is referred to the original publications for further information [83–86].
5. CLINICALLY INVESTIGATED ISOZYME-SELECTIVE PI3K INHIBITORS The development of isozyme-selective inhibitors of PI3K has started to attract great interest as this may enable the possibility of targeting specific therapeutic areas while reducing off-target effects [6]. This vision has the potential to become reality due to recent research advances culminating in availability of data regarding tissue distribution, functional elucidation, and structural variation of the individual isozymes [3,6,87]. CAL-101(structure undisclosed) – This compound is a PI3Kd-selective inhibitor with an IC50 of 1–10 nM with a reported W30-fold selectivity over other isoforms, mTOR and DNA-PK [88]. It is likely that the structure of CAL-101 derives from a series of quinazolinone derivatives that yielded the inhibitor IC-87114 (aka D-030) (Figure 4, 16) [89–91]. The d-isozyme
348
John M. Nuss et al.
O NH2
N
N
N
N
N N
H2N
N
OH OH
N
N
N NH2 16
Figure 4
17
Isozyme-selective PI3K inhibitors.
selectivity of these compounds (and also the PIK-39, PIK-293, PIK-294 analogs) is postulated to arise due to a conformational rearrangement of Met752 (Met804 in PI3Kg), which is distal to the highly homologous adenine binding pocket [22]. CAL-101 is being evaluated as an oral therapeutic in two phase I clinical trials: in a recently completed trial as a treatment for allergic rhinitis (presumably as an anti-inflammatory agent) [92] and in an on-going trial as a treatment for cancer, focused on patients with hematologic malignancies [88,93–95]. CAL-101 inhibits p110dmediated basophil activation in whole blood with an IC50 of 30–70 nM and has demonstrated well-tolerated, sustained plasma concentrations of 500–5000 nM in a 7-day multidose healthy human volunteer study, indicating a viable therapeutic index [88,94]. Although data on preclinical PI3Kd-selective inhibitors other than IC-87114 is limited, recent patent disclosures suggest that this area remains active [96–98]. TG100-115 (17) – The dual d/g-selective ATP-competitive inhibitor 17 (Figure 4) (PI3K IC50: 1300, 1200, 83, 235 nM for a, b, g, d, respectively) has been the subject of clinical evaluation in a number of cardiovascular indications [99,100]. The design of the pteridine 17 involved pharmacophore-based modeling starting from known PI3K inhibitors, 2, and quercetin [99]. Initial SAR was developed using an in vivo Miles assay evaluating vascular permeability, with in vitro assays being subsequently introduced to determine the molecular target and the cellular signaling pathway for the demonstrated inhibition of vascular permeability [99]. In contrast to pan-active PI3K inhibitors, 17 showed no effect on mitogenesis [99]. Pre-clinically, 17 exhibited well-behaved pharmacokinetic properties when administered intravenously and demonstrated cardioprotective attributes in animal models [99,101,102]. A phase I/II clinical trial to evaluate intravenously administered 17 in terms of safety in patients who suffer a heart attack and then undergo angioplasty to restore blood flow and also to evaluate its potential efficacy to reduce heart muscle damage related to this ischemia/reperfusion injury, was completed in 2008, with no further clinical development reported [103].
349
Emerging Therapies Based on Inhibitors of Phosphatidyl-Inositol-3-Kinases
Compound 17 is also being investigated pre-clinically for the treatment of inflammatory diseases [104]. AZD6482 (structure undisclosed) – No pre-clinical data has been disclosed for AZD6482, a PI3Kb-selective inhibitor that is being investigated as an antithrombotic agent in multiple phase I clinical trials [105,106]. The first clinical trial, completed in May 2008, was a 5-month trial that investigated the safety and tolerability of a single escalating intravenously administered dose of AZD6482 alone, and co-administered with acetylsalicylic acid (ASA) [105]. In February 2009, a second phase I clinical trial was opened for patient recruitment to evaluate the effect of AZD6482, compared with clopidogrel on bleeding times in healthy volunteers receiving low dose ASA (75 mg aspirin) [106].
6. PRE-CLINICAL ISOZYME-SELECTIVE PI3K INHIBITORS A number of compounds that exhibit PI3K isozyme selectivity have been disclosed, which have yet to enter the clinic. Generally, these pre-clinical compounds are selectively potent for a sub-set of the PI3K isozymes (i.e., they are not specific inhibitors), but for clarity in this review, they will be classified by the isozyme which is most potently inhibited. TGX-221 (18) – Compound 18 (Figure 5) (PI3K IC50: W5000, 5–10, W3500, 100–200 nM for a, b, g, d, respectively) is the lead compound among a series reported to show potent, selective inhibition of PI3Kb and is also being investigated as an antithrombotic agent [14,107,108]. 18 utilizes key structural characteristics of the core scaffold of 1; specifically the 2-morpholine moiety, the 4-carbonyl, and the 8-aryl substituents. From docking studies in a homology model of PI3Kb, it is postulated that 18 resides in the ATP-binding pocket of PI3Kb where it forms a critical hydrogen bond interaction with Val854 in the hinge through the morpholine oxygen and forms additional contacts with Lys805 O
N NH
O
NO2
O
O
N
S
N
N O
O
NH
S
NH
O
S
N N
Br
O
O
NH S
O
HO
N
N
F O N
F O
F
18
Figure 5
19
20
Additional isozyme-selective PI3K inhibitors.
21
22
O
350
John M. Nuss et al.
(salt bridge) and Asp937 (DFG loop) through the carbonyl group and exocyclic aniline, respectively [109]. In the absence of X-ray crystal structure of PI3Kb, the origin of the isozyme selectivity has also been investigated by mutagenesis, resulting in a number of interesting hypotheses regarding binding modes and ways to improve potency against PI3Kb [110]. Owing to their selective PI3Kb inhibitory profile, 18 and related analogs may serve as very important tools for elucidating the role of PI3Kb signaling in thrombosis as well as cancer [14,111–116]. PIK-75 (19) – Compound 19 (Figure 5) (PI3K IC50: 3–6, 850–1300, 40–76, 510 nM for a, b, g, d, respectively, mTOR IC50: W1000 nM, and DNA-PK IC50: 2 nM) was identified as a leading compound from a series of imidazopyridines that exhibit potent inhibition of PI3Ka [22,114,117–119]. This analog has been used as a reference compound for further exploring SAR for selective PI3Ka inhibition [120]. Compound 19 potently inhibits tumor cell proliferation in vitro (IC50o100 nM) in various PI3K-driven cell lines and has demonstrated tumor growth inhibition in a xenograft model [121]. Owing to its PI3Ka inhibitory profile, 19 has served as a tool compound for elucidating the role of PI3Ka signaling [22,111]. However, as 19 has demonstrated generalized cytotoxicity likely unrelated to PI3K inhibition [114,122], it is unlikely to be developed further. As aberrant PI3Ka signaling has been strongly implicated in various cancers, particularly through activating mutations [114], interest in PI3Ka-selective inhibitors remains very strong as evidenced by recently disclosed inhibitors in the open and patent literature [48,49,123–125]. AS-252424 (20)/AS-605240 (21)/AS-604850 (22) – The PI3Kg-selective inhibitors (Figure 5) are exemplified by the thiazolidinedione derivatives 20 (PI3K IC50: 940, W20,000, 30, W20,000 nM for a, b, g, d, respectively) [113,126], 21 (PI3K IC50: 60, 270, 8, 300 nM for a, b, g, d, respectively) [127– 130] and 22 (PI3K IC50: 4,500, W20,000, 250, W20,000 nM for a, b, g, d, respectively) [130]. The crystal structures in PI3Kg of 20 [126] and 22 [130] have been disclosed, and initial docking studies to elucidate differences in key interactions between the thiazolidinedione derivatives and various isozymes have been conducted [109]. Furthermore, these derivatives have been used to investigate the role of PI3Kg signaling [13,113,131] and are reported to demonstrate anti-inflammatory effects [130–133]. There is a lot of interest in PI3Kg-selective inhibitors as evidenced by recently disclosed inhibitors in the literature and patents [134–138].
7. CONCLUSIONS Recent research has clearly shown that PI3K signaling is involved in many pathways and that the isozymes do not have identical roles.
Emerging Therapies Based on Inhibitors of Phosphatidyl-Inositol-3-Kinases
351
Currently, the optimal PI3K isoform inhibitory profile (with or without mTOR or DNA-PK inhibition) for a successful oncology, thrombosis, inflammation, or cardiac therapeutic is undetermined as there is insufficient clinical data to form hard conclusions. Early concerns that all panactive PI3K inhibitors may suffer from an unacceptable therapeutic index due to issues such as metabolic toxicity may be overstated. Additionally, it may be hypothesized that drug resistance due to mutations in certain PI3K isozymes may be more easily overcome by inhibiting multiple PI3K isozymes along with PIK-related kinases. Conversely, inhibiting a specific isozyme (or group of isozymes) may lead to a potent indication-focused therapy with an improved therapeutic index due to minimized off-critical isozyme effects. It is still to be determined whether intervention in these pathways with PI3K inhibitors will have therapeutic benefit in humans, although admittedly the story is just beginning. Only with more time and more pre-clinical and clinical results will we be able to determine whether the incredible therapeutic promise of PI3K inhibitors will come to fruition.
REFERENCES [1] J. A. Engelman, J. Luo and L. C. Cantley, Nat. Rev. Genet., 2006, 7, 606. [2] B. T. Hennessy, D. L. Smith, P. T. Ram, Y. Lu and G. B. Mills, Nat. Rev. Drug Disc., 2005, 4, 988. [3] R. Marone, V. Cmiljanovic, B. Giese and M. P. Wymann, Biochim. Biophys. Acta, 2008, 1784, 159. [4] T. J. Sundstrom, A. C. Anderson and D. L. Wright, Org. Biomol. Chem., 2009, 7, 840. [5] C. Garcia-Echeverria and W. R. Sellers, Oncogene, 2008, 27, 5511. [6] N. T. Ihle and G. Powis, Mol. Cancer Ther., 2009, 8, 1. [7] I. Sansal and W. R. Sellers, J. Clin. Oncol., 2004, 22, 2954. [8] T. F. Franke, Oncogene, 2008, 27, 6473. [9] S. Kang, A. G. Bader and P. K. Vogt, Proc. Natl. Acad. Sci. U.S.A., 2005, 102, 802. [10] M. P. Wymann and L. Pirola, Biochim. Biophys. Acta, 1998, 1436, 127. [11] E. H. Walker, O. Perisic, C. Reid, L. Stephens and R. Williams, Nature, 1999, 402, 313. [12] M. Falasca and T. Maffucci, Biochem. Soc. Trans., 2007, 35, 211. [13] T. Ruckle, M. K. Schwarz and C. Rommel, Nat. Rev. Drug Discov., 2006, 5, 903. [14] S. P. Jackson, S. M. Schoenwaelder, I. Goncalves, W. S. Nesbitt, C. L. Yap, C. E. Wright, V. Kenche, K. E. Anderson, S. M. Dopheide, Y. Yuan, S. A. Sturgeon, H. Prabaharan, P. E. Thompson, G. D. Smith, P. R. Shepherd, N. Daniele, S. Kulkarni, B. Abbott, D. Saylik, C. Jones, L. Lu, S. Giuliano, S. C. Hughan, J. A. Angus, A. D. Robertson and H. H. Salem, Nat. Med., 2005, 11, 507. [15] L. Zhao and P. K. Vogt, Oncogene, 2008, 27, 5486. [16] B. Vanhaesbroeck, S. J. Leevers, K. Ahmadi, J. Timms, R. Katso, P. C. Driscoll, R. Woscholski, P. J. Parker and M. D. Waterfield, Annu. Rev. Biochem., 2001, 70, 535. [17] Y. Samuels, Z. Wang and A. Bardelli, Cell, 2007, 129, 957. [18] J. J. Zhao, Z. Liu, L. Wang, E. Shin, M. F. Loda and T. M. Roberts, Proc. Natl. Acad. Sci. U.S.A., 2005, 102, 18443. [19] C. J. Vlahos, W. F. Matter, K. Y. Hui and R. F. Brown, J. Biol. Chem., 1994, 269, 5241. [20] A. Arcaro and M. P. Wymann, Biochem. J., 1993, 296, 297.
352
John M. Nuss et al.
[21] E. H. Walker, M. E. Pacold, O. Perisic, L. Stephens, P. T. Hawkins, M. P. Wymann and R. L. Williams, Mol. Cell, 2000, 6, 909. [22] Z. A. Knight, B. Gonzalez, M. E. Feldman, E. R. Zunder, D. D. Goldenberg, O. Williams, R. Loewith, D. Stokoe, A. Balla, B. Toth, T. Balla, W. A. Weiss, R. L. Williams and K. M. Shokat, Cell, 2006, 125, 733. [23] B. H. Norman, C. Shih, J. E. Toth, J. E. Ray, D. W. Dodge, D. W. Johnson, P. G. Rutherford, R. M. Schultz, J. F. Worzall and C. J. Vlahos, J. Med. Chem., 1996, 39, 1106. [24] S. Stoyanovea, G. Bulgarelli-Leva, C. Kirsch, T. Hanck, R. Klinger, R. Wetzker and M. P. Wymann, Biochem. J., 1997, 324, 489. [25] RCSB Protein Structure Database, http://www.rcsb.org/pdb/home/home.do [26] C. H. Huang, D. Mandelker, O. Schmidt-Kittler, Y. Samuels, V. E. Velculescu, K. W. Kinzler, B. Vogelstein, S. B. Gabelli and L. M. Amzel, Science, 2007, 318, 1744. [27] J. R. Garlich, P. De, N. Dey, N. Dey, J. D. Su, X. Peng, A. Miller, R. Murali, Y. Lu, G. B. Mills, V. Kundra, H. K. Shu, Q. Peng and D. L. Durden, Cancer Res., 2008, 68, 206. [28] http://www.semaforepharma.com/pdfs/news/Semafore_ASCO_June-04-2008.pdf [29] M. P. Wymann, G. Bulgarelli-Leva, M. J. Zvelebil, L. Pirola, B. Vanhaesebroeck, M. D. Waterfield and G. Panayotou, Mol. Cell Biol., 1996, 16, 1722. [30] H. Yano, S. Nakanishi, K. Kimura, N. Hanai, Y. Saitoh, Y. Fukui, Y. Nonomura and Y. Matsuda, J. Biol. Chem., 1993, 268, 25846. [31] J. W. Millard, L. D. Kirkpatrick, L. A. Pestano and G. Powis, 19th Rocky Mountain Regional Meeting of the American Chem. Soc., Tucson, AZ, October 2006. [32] P. Wipf, L. Kirkpatrick and G. Powis, US Patent US20090087441, 2009. [33] N. T. Ihle, R. Williams, S. Chow, W. Chew, M. I. Berggren, G. Paine-Murrieta, D. J. Minion, R. J. Halter, P. Wipf, R. Abraham, L. Kirkpatrick and G. Powis, Mol. Cancer Ther., 2004, 3, 763. [34] A. L. Howes, G. G. Chiang, E. S. Lang, C. B. Ho, G. Powis, K. Vuori and R. T. Abraham, Mol. Cancer Ther., 2007, 6, 2505. [35] N. T. Ihle, R. Lemos, D. Schwartz, J. Oh, R. J. Halter, P. Wipf, L. Kirkpatrick and G. Powis, Mol. Cancer Ther., 2009, 8, 94. [36] www.clinicaltrials.gov/ct2/show/NCT00726583 [37] http://www.novartisoncology.com/research-innovation/pipeline/bgt226.jsp [38] F. Stauffer, S.-M. Maira, P. Furet and C. Garcia-Echeverria, Bioorg. Med. Chem. Lett., 2008, 18, 1027. [39] S.-M. Maira, F. Stauffer, J. Brueggen, P. Furet, C. Schnell, C. Fritsch, S. Brachmann, P. Chene, A. De Pover, K. Schoemaker, D. Fabbro, D. Gabriel, M. Simonen, L. Murphy, P. Finan, W. Sellers and C. Garcia-Echeverria, Mol. Cancer Ther., 2008, 7, 1851. [40] P. Baumann, S. Mandl-Weber, F. Oduncu and R. Schmidmaier, Exp. Cell Res., 2009, 315, 485. [41] V. Serra, B. Markman, M. Scaltriti, P. J. A. Eichhorn, V. Valero, M. Guzman, M. L. Botero, E. Llonch, F. Atzori, S. Di Cosimo, M. Maira, C. Garcia-Echeverria, J. L. Parra, J. Arribas and J. Baselga, Cancer Res., 2008, 68, 8022. [42] P. Cao, S.-M. Maira, C. Garcia-Echeverria and D. W. Hedley, Br. J. Cancer, 2009, 100, 1267. [43] O. Dorigo, C. Santiskulvong, M. Fekete, A. Karam, D. Mulholland, C. Engl and H. Wu, 17, 40th Annual Meeting Women’s Cancer, San Antonio, TX, February, 2009. [44] P. J. A. Eichhorn, M. Gili, M. Scaltriti, V. Serra, M. Guzman, W. Nijkamp, R. L. Beijersbergen, V. Valero, J. Seoane, R. Bernards and J. Baselga, Cancer Res., 2008, 68, 9221. [45] E. Marrer, et al., 215, Proceedings of the American Association for Cancer Research (AACR), San Diego, CA, April, 2008. [46] A. J. Folkes, K. Ahmadi, W. K. Alderton, S. Alix, S. J. Baker, G. Box, I. S. Chuckowree, P. A. Clark, P. Depledge, S. A. Eccles, L. Friedman, A. Hayes, T. C. Hancox,
Emerging Therapies Based on Inhibitors of Phosphatidyl-Inositol-3-Kinases
[47]
[48] [49]
[50]
[51]
[52] [53]
[54]
[55]
[56] [57] [58] [59] [60] [61]
[62] [63]
[64] [65]
353
A. Kugendradas, L. Lensun, P. Moore, A. Olivero, J. Pang, S. Patel, G. Pergl-Wilson, F. Raynaud, A. Robson, N. Saghir, L. Salphati, S. Sohal, M. Ultsch, M. Valenti, H. Wallweber, N. Wan, C. Weisman, P. Workman, A. Zhyvoloup, M. Zvelebil and S. Shuttleworth, J. Med. Chem., 2008, 51, 5522. T. P. Heffron, M. Berry, G. Castanedo, C. Chang, I. Chuckowree, J. Dotson, A. Folkes, J. Gunzner, J. Lesnick, C. Lewis, K. Malesky, S. Mathieu, J. Nonomiya, A. Olivero, J. Pang, D. Peterson, L. Salphati, D. Sampath, D. Sutherlin, V. Tsui, M. Ultsch, N. Wan, S. Wang, C. Weismann, S. Wong and B. Zhu, MEDI-220, American Chemical Society National Meeting (ACS), Salt Lake City, UT, March 2009. M. Hayakawa, H. Kaizawa, H. Moritomo, T. Koizumi, T. Ohishi, M. Okada, M. Ohta, S. Tsukamoto, P. Parker, P. Workman and M. Waterfield, Bioorg. Med. Chem., 2006, 14, 6847. M. Hayakawa, H. Kaizawa, H. Moritomo, T. Koizumi, T. Ohishi, M. Yamano, M. Okada, M. Ohta, S. Tsukamoto, F. Raynaud, P. Workman, M. Waterfield and P. Parker, Bioorg. Med. Chem. Lett., 2007, 17, 2438. S. Patel, D. Sampath, M. Belvin, A. Brown, K. Edgar, L. Lensun, P. Moore, L. Salphati, H. Stern and L. Friedman, 217, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, Geneva, October 2008. P. LoRusso, D. Sarker, D. Von Hoff, R. Tibes, M. K. Derynck, J. A. Ware, Y. Yan, G. D. Demetri, J. S. deBono and A. J. Wagner, 223, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, Geneva, October 2008. T. C. Hancox, A. N. Pegg, C. M. Beswick, J. T. Blench, A. E. Dechaux, J. J. Kulagowski, J. A. Nadin and S. Price, WO Patent WO2008152390, 2008. G. Castanedo, R. Goldsmith, J. Gunzner, T. Heffron, S. Mathieu, A. Olivero, S. Staben, D. Sutherlin, V. Tsui, S. Wang, B. Zhu, T. Bayliss, I. Chuckowree, A. Folkes and N. Wan, WO Patent WO2008073785, 2008. T. Bayliss, I. Chuckowree, A. Folkes, S. Oxenford, N. Wan, G. Castanedo, R. Goldsmith, J. Gunzner, T. Heffron, S. Mathieu, A. Olivero, S. Staben, D. Sutherlin and B. Zhu, WO Patent WO2008070740, 2008. A. Folkes, S. Shuttleworth, I. Chuckowree, S. Oxenford, N. Wan, G. Castanedo, R. Goldsmith, J. Gunzner, T. Heffron, S. Mathieu, A. Olivero, S. Staben, D. Sutherlin and B. Zhu, WO Patent WO2007127175, 2007. P. Foster, C199, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, San Francisco, CA, October 2007. G. Shapiro, G. Edelman and E. Calvo, C205, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, San Francisco, CA, October 2007. W. Bajjalieh, L. C. Bannen, S. D. Brown, P. Kearney, M. Mac, C. K. Marlowe, J. M. Nuss, Z. Tesfai, Y. Wang and W. Xu, WO Patent WO2007044729, 2007. http://www.exelixis.com/pipeline_xl147.shtml D. Laird, B250, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, San Francisco, CA, October 2007. B. Markman, P. M. LoRusso, A. Patnaik, E. Heath, A. D. Laird, B. Van Leeuwen, K. P. Papadopoulos and J. Baselga, 216, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, Geneva, October 2008. A. Patnaik, P. M. LoRusso and J. Tabernero, B265, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, San Francisco, CA, October 2007. K. P. Papadopoulos, B. Markman, J. Tabernero, A. Patnaik, E. I. Heath, A. DeCillis, D. Laird, S. K. Aggarwal, L. Nguyen and P. M. LoRusso, 3510, American Society of Clinical Oncology (ASCO), Chicago, IL, May 2008. http://www.exelixis.com/pipeline_xl765.shtml K. R. Auger, L. Luo, S. Knight, G. Van Aller, P. J. Tummino, R. A. Copeland, M. Diamond, D. Sutton, J. R. Jackson and D. Dhanak, 221, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, Geneva, October 2008.
354 [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81]
[82]
[83] [84] [85] [86] [87] [88]
[89] [90] [91] [92] [93] [94]
[95]
[96]
John M. Nuss et al.
http://www.clinicaltrials.gov/ct2/show/NCT00695448 D. Kong and T. Yamori, Cancer Sci., 2007, 98, 1638. D. Kong, S. Yaguchi and T. Yamori, Biol. Pharm. Bull., 2009, 32, 297. D. Kong, M. Okamura, H. Yoshimi and T. Yamori, Eur. J. Cancer, 2009, 45, 857. S. Dan, H. Yoshimi, M. Okamura, Y. Mukai and T. Yamori, Biochem. Biophys. Res. Comm., 2009, 379, 104. T. Yamori and M. Okamura, 225, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, Geneva, October 2008. D. Kong, H. Yoshimi and T. Yamori, 226, EORTC-NCI-AACR International Conference Molecular Targets Cancer Therapy, Geneva, October 2008. S. Butterworth, J.E. Griffen, B.G. Hill and M. Pass, WO Patent WO2008032036, 2008. S. Butterworth, J.E. Griffen, B.G. Hill and M. Pass, WO Patent WO2008032027, 2008. S. Butterworth, J.E. Griffen and M. Pass, WO Patent WO2008032077, 2008. S. Butterworth, J.E. Griffen and M. Pass, WO Patent WO2008032064, 2008. S. Butterworth, J.E. Griffen and M. Pass, WO Patent WO2008032091, 2008. S. Butterworth, J.E. Griffen, B.G. Hill and M. Pass, WO Patent WO2008032033, 2008. S. Butterworth, J.E. Griffen and M. Pass, WO Patent WO2008032028, 2008. S. Butterworth, J.E. Griffen and M. Pass, WO Patent WO2008032086, 2008. I. Seipelt, M. Gerlach, L. Blumenstein, G. Mueller, M. Teifel, E. Polymeropoulos and E. Guenther, 3706, Proceedings of the American Association for Cancer Research (AACR), Denver, CO, April 2009. I. Seipelt, S. Baasner, M. Gerlach, M. Teifel, J. Fensterle, L. Blumenstein, G. Mueller and E. Guenther, 3705 Proceedings of the American Association for Cancer Research (AACR), Denver, CO, April 2009. M. G. Buckley, T. Morgan and M.V. Sabin, WO Patent WO2008044022, 2008. P. Imbach, G. Stauffer, P. Furet and H-G. Capraro, WO Patent WO2008037477, 2008. S. Pecchi, Z. Ni, M. Burger, A. Wagman, G. Atallah, S. Bartulis, S. Ng, K. B. Pfister, A. Smith, Y, Zhang, H. Merritt and C. Voliva, WO Patent WO2008098058, 2008. S. D. Knight, C. A. Parrish, L. H. Ridgers, M. A. Sarpong and A. M. Chaudhari, WO Patent WO2009021083, 2009. L. M. Amzel, C.-H. Huang, D. Mandelker, C. Lengauer, S. B. Gabelli and B. Vogelstein, Nat. Rev. Cancer, 2008, 8, 665. B. J. Lannutti, S. A. Meadows, A. Kashishian, B. Steiner, S. May, A. J. Johnson, R. G. Ulrich, A. Yu, M. l. W. Gallatin, J. C. Byrd, K. D. Puri and N. A. Giese, Blood, 2008, 112(ASH Annual Meeting Abstracts), 16. C. Sadhu, K. Dick, J. Treiberg, C. G. Sowell, E. A. Kesicki and A. Oliver, US Patent, US6,518,277, 2003. C. Sadhu, K. Dick, J. Treiberg, C. G. Sowell, E. A. Kesicki and A. Oliver, US Patent US6,667,300, 2003. C. Billottet, V. L. Grandage, R. E. Gale, A. Quattropani, C. Rommel, B. Vanhaesebroeck and A. Khwaja, Oncogene, 2006, 25, 6648. http://www.clinicaltrials.gov/ct2/show/NCT00836914 http://www.clinicaltrials.gov/ct2/show/study/NCT00710528 S. E. May, A. Kashishian, T. S. Lin, J. A. Jones, J. M. Flynn, R. G. Ulrich, H. Chen, A. S. Yu, K. D. Puri, B. J. Lannutti, N. A. Giese, J. C. Byrd and A. J. Johnson, Blood, 2008, 112(ASH Annual Meeting Abstracts), 3165. H. Ikeda, T. Hideshima, Y. Okawa, S. Vallet, S. Pozzi, L. Santo, G. Gorgun, M. Fulciniti, N. S. Raje, G. Perrone, N. Munshi, P. Richardson, B. J. Lannutti, K. D. Puri, N. A. Giese and K. C. Anderson, Blood, 2008, 112(ASH Annual Meeting Abstracts), 2753. Y. Chen, T. D. Cushing, X. Hao, X. He, A. Reichelt, R. M. Rzasa, J. Seganish, Y. Shin and D. Zhang, WO Patent WO2008118454, 2008.
Emerging Therapies Based on Inhibitors of Phosphatidyl-Inositol-3-Kinases
355
[97] Y. Chen, T. D. Cushing, X. Hao, X. He, A. Reichelt, R. M. Rzasa, J. Seganish, Y. Shin and D. Zhang, WO Patent WO2008118455, 2008. [98] Y. Chen, T. D. Cushing, J. A. Duquette, F. Gonzalez Lopez de Turiso, X. Hao, X. He, B. Lucas, L. R. McGee, A. Reichelt, R. M. Rzasa, J. Seganish, Y. Shin and D. Zhang, WO Patent WO2008118468, 2008. [99] M. S. S. Palanki, E. Dneprovskaia, J. Doukas, R. M. Fine, J. Hood, X. Kang, D. Lohse, M. Martin, G. Noronha, R. M. Soll, W. Wrasidlo, S. Yee and H. Zhou, J. Med. Chem., 2007, 50, 4279. [100] W. Wrasidlo, J. Doukas, I. Royston, G. Noronha, J. D. Hood, E. Dneproyskaia, X. Gong, U. Splittgerber and N. Zhao, US Patent, US7,208,493, 2007. [101] J. Doukas, W. Wrasidlo, G. Noronha, E. Dneprovskaia, J. Hood and R. Soll, Biochem. Soc. Trans., 2007, 35, 204. [102] J. Doukas, W. Wrasidlo, G. Noronha, E. Dneprovskaia, R. Fine, S. Wels, J. Hood, A. DeMaria, R. Soll and D. Cheresh, Proc. Natl. Acad. Sci. U.S.A., 2006, 103, 19866. [103] http://www.clinicaltrials.gov/ct2/show/NCT00103350 [104] J. Doukas, L. Eide, K. Stebbins, A. Racaneli-Layton, L. Dellamary, M. Martin, E. Dneprovskaia, G. Noronha, R. Soll, W. Wrasidlo, L. M. Acevedo and D. A. Cheresh, J. Pharmacol. Exp. Ther., 2009, 328, 758. [105] http://clinicaltrials.gov/show/NCT00688714 [106] http://clinicaltrials.gov/show/NCT00853450 [107] S. P. Jackson, A. D. Robertson, V. Kenche, P. Thompson, H. Prabaharan, K. Anderson, B. Abbott, I. Goncalves, W. Nesbitt, S. Schoenwaelder, D. Saylik, WO Patent WO2004/ 016607, 2004. [108] S. A. Sturgeon, C. Jones, J. A. Angus and C. E. Wright, Eur. J. Pharmacol., 2008, 587, 208. [109] M. J. Zvelebil, M. D. Waterfield and S. J. Shuttleworth, Archiv. Biochem. Biophys., 2008, 477, 404. [110] M. Frazzetto, C. Suphiolu, J. Zhu, O. Schmidt-Kittler, I. G. Jennings, S. L. Cranmer, S. P. Jackson, K. W. Kinzler, B. Vogelstein and P. E. Thompson, Biochem. J., 2008, 414, 383. [111] C. Chaussade, G. W. Rewcastle, J. D. Kendall, W. A. Denny, K. Cho, L. M. Gronning, M. L. Chong, S. H. Anagnostou, S. P. Jackson, N. Daniele and P. R. Shepherd, Biochem. J., 2007, 404, 449. [112] J. Guillermet-Guibert, K. Bjorklof, A. Salpekar, C. Gonella, F. Ramadani, A. Bilancio, S. Meek, A. J. H. Smith, K. Okkenhaug and B. Vanhaesebroeck, Proc. Natl. Acad. Sci. U.S.A., 2008, 105, 8292. [113] R. Heller, Q. Chang, G. Ehrlich, S. N. Hsieh, S. M. Schoenwaelder, P. J. Kuhlencordt, K. T. Preissner, E. Hirsch and R. Wetzker, Cardiovasc. Res., 2008, 80, 96. [114] N. E. Torbett, A. Luna-Moran, Z. A. Knight, A. Houk, M. Moasser, W. Weiss, K. M. Shokat and D. Stokoe, Biochem. J., 2008, 415, 97. [115] P. K. Vogt, M. Gymnopoulos and J. R. Hart, Curr. Opin. Genet. Dev., 2009, 19, 12. [116] C. Benistant, H. Chapuis and S. Roche, Oncogene, 2000, 19, 5083. [117] M. Hayakawa, H. Kaizawa, K.-I. Kawaguchi, N. Ishikawa, T. Koizumi, T. Ohishi, M. Yamano, M. Okada, M. Ohta, S.-I. Tsukamoto, F. I. Raynaud, M. D. Waterfield, P. Parker and P. Workman, Bioorg. Med. Chem., 2007, 15, 403. [118] M. Hayakawa, K.-I. Kawaguchi, H. Kaizawa, T. Koizumi, T. Ohishi, M. Yamano, M. Okada, M. Ohta, S.-I. Tsukamoto, F. I. Raynaud, P. Parker, P. Workman and M. D. Waterfield, Bioorg. Med. Chem., 2007, 15, 5837. [119] M. Hayakawa, H. Kaizawa, K.-I. Kawaguchi, K. Matsuda, N. Ishikawa, T. Koizumi, M. Yamano, M. Okada and M. Ohta, US Patent US6,403,588, 2001. [120] J. D. Kendall, G. W. Rewcastle, R. Frederick, C. Mawson, W. A. Denny, E. S. Marshall, B. C. Baguley, C. Chaussade, S. P. Jackson and P. R. Shepherd, Bioorg. Med. Chem., 2007, 15, 7677.
356
John M. Nuss et al.
[121] M. Hayakawa, K.-I. Kawaguchi, H. Kaizawa, T. Koizumi, T. Ohishi, M. Yamano, M. Okada, M. Ohta, S.-I. Tsukamoto, F. I. Raynaud, P. Parker, P. Workman and M. D. Waterfield, Bioorg. Med. Chem., 2007, 15, 5837. [122] J. S. Chen, L. J. Zhou, M. Entin-Meer, X. Yang, M. Donker, Z. A. Knight, W. Weiss, K. M. Shokat, D. Haas-Kogan and D. Stokoe, Mol. Cancer Ther., 2008, 7, 841. [123] N. D. Adams, J. L. Burgess, A. M. Chaudhari, S. D. Knight and C. A. Parrish, WO Patent WO2008150827, 2008. [124] N. D. Adams, J. L. Burgess, M. G. Darcy, S. D. Knight, K. A. Newlander, L. H. Ridgers and S. J. Schmidt, WO Patent WO2008157191, 2008. [125] N. D. Adams, D. Dhanak, S. D. Knight, L. Schaller and J. Tang, WO Patent WO2008014219, 2008. [126] V. Pomel, J. Klicic, D. Covini, D. D. Church, J. P. Shaw, K. Roulin, F. Burgat-Charvillon, D. Valognes, M. Camps, C. Chabert, C. Gillieron, B. Franon, D. Perrin, D. Leroy, D. Gretener, A. Nichols, P. A. Vitte, S. Carboni, C. Rommel, M. K. Schwarz and T. Ru¨ckle, J. Med. Chem., 2006, 49, 3857. [127] Z. A. Knight and K. M. Shokat, Biochem. Soc. Trans., 2007, 35, 245. [128] T. Rueckle, X. Jiang, P. Gaillard, D. Church and T. Vallottton, WO Patent WO04007491, 2004. [129] G. De Luca, WO Patent WO04006916, 2004. [130] M. Camps, T. Ru¨ckle, H. Ji, V. Ardissone, F. Rintelen, J. Shaw, C. Ferrandi, C. Chabert, C. Gillieron, B. Franc- on, T. Martin, D. Gretener, D. Perrin, D. Leroy, P.-A. Vitte, E. Hirsch, M. P. Wymann, R. Cirillo, M. K. Schwarz and C. Rommel, Nat. Med., 2005, 11, 936. [131] A. M. Condliffe, K. Davidson, K. E. Anderson, C. D. Ellson, T. Crabbe, K. Okkenhaug, B. Vanhaesebroeck, M. Turner, L. Webb, M. P. Wymann, E. Hirsch, T. Ru¨ckle, M. Camps, C. Rommel, S. P. Jackson, E. R. Chilvers, L. R. Stephens and P. T. Hawkins, Blood, 2005, 106, 1432. [132] K. Ito, G. Caramori and I. M. Adcock, J. Pharmacol. Exp. Ther., 2007, 321, 1. [133] D. F. Barber, A. Bartolome, C. Hernandez, J. M. Flores, C. Redondo, C. FernandezAria, M. Camps, T. Ru¨ckle, M. K. Schwarz, S. Rodrı´guez, C. Martinez-A, D. Balomenos, C. Rommel and A. C. Carrera, Nat. Med., 2005, 11, 933. [134] T. B. Lanni, Jr., K. L. Greene, C. N. Kolz, K. S. Para, M. Visnick, J. L. Mobley, D. T. Dudley, T. J. Baginski and M. B. Liimatta, Bioorg. Med. Chem. Lett., 2007, 17, 756. [135] T. Crabbe, Biochem. Soc. Trans., 2007, 35, 253. [136] N. C. Barvian, C. N. Kolz, K. S. Para, W. C. Patt and M. Visnick, WO Patent WO2004052373, 2004. [137] R. D. Gogliotti, K. L. Muccioli, K. S. Para and M. Visnick, WO Patent WO2004056820, 2004. [138] M. Shimada, T. Murata, K. Fuchikami, H. Tsujishita, N. Omori, I. Kato, M. Miura, K. Urbahns, F. Gantner and K. Bacon, WO Patent WO04029055, 2004.
CHAPT ER
18 The Anti-Infective and Anti-Cancer Properties of Artemisinin and its Derivatives Christopher Paul Hencken*, Alvin Solomon Kalinda* and John Gaetano D’Angelo**
Contents
1. Introduction 2. Antiparasitic Uses of Artemisinin 2.1 Antimalarial activity 2.2 Anti-toxoplasmosis activity 2.3 Anti-leishmaniasis activity 2.4 Anti-shistosomiasis activity 2.5 Anti-trypanosomiasis activity 3. Other Therapeutic Uses of Artemisinin 3.1 Antiviral activity 3.2 Antifungal activity 3.3 Anticancer activity 4. Toxicity 4.1 Evidence in favor of safety 4.2 Evidence of neurotoxicity 5. Conclusion References
359 360 360 365 367 368 369 370 370 371 372 373 374 374 375 375
1. INTRODUCTION Artemisinin, a 1,2,4-trioxane sesquiterpene lactone with an atypical endoperoxide moiety, was isolated as the active component from the * Johns Hopkins University, 3400N. Charles Street, Baltimore, MD 21218, USA ** Alfred University, 1 Saxon Drive, Alfred, NY 14802, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04418-2
r 2009 Elsevier Inc. All rights reserved.
359
360
Christopher Paul Hencken et al.
Chinese medicinal plant commonly known as Qinghaosu (Artemisia annua) [1,2] and was found to be extremely active against the deadly cerebral form of malaria. While it has become the new cornerstone of antimalarial treatment, it has biological applications that reach far beyond this one disease. These applications include activity against toxoplasmosis, leishmaniasis, viral infections, certain bacterial infections, shistosomiasis, and cancer. The use of artemisinin and its derivatives as a broad-spectrum anti-infective agents was recently patented [3]. It has been shown that the active pharmacophore is the 1,2,4-trioxane, specifically (and perhaps strangely), the peroxide unit [4]. In this chapter, we have paid particular attention to the activity of artemisinin-based compounds beyond the well-documented area of malaria, while providing a cursory overview of their antimalarial properties. For the interested reader, we suggest some recent reviews on the antimalarial activity of artemisinin and its derivatives [5–7]. For the purposes of saving space, we will use the following short hand for the structural core of artemisinin (1) throughout the chapter. H O H
O O O C-10
O C-10
O
O
1
1
2. ANTIPARASITIC USES OF ARTEMISININ The use of herbal teas brewed from A. annua against fever and chills dates to AD 340; however, the first mention of its use dates as far back as 168 BC [8]. Since that time, artemisinin has been identified as the active constituent of A. annua in 1972 [4]. More recently, the uses of artemisinin have been shown to be quite vast. The wide therapeutic applications of artemisinin make it difficult to attribute artemisinin’s effectiveness to a common mechanism. Since intra-cellular/parasitic metabolic degradation of artemisinin is known to induce the formation of oxygen and carbon-centered radicals, radical-induced cell death is certainly one attractive possibility. Despite the increasing known breadth of activity, the most popular uses
Anti-Infective and Anti-Cancer Properties of Artemisinin
361
remain the antiparasitic uses. These uses, though, are not necessarily limited to antimalarial (Section 2.1 for which artemisinin has been made famous), anti-toxoplasmosis (Section 2.2), anti-leishmaniasis (Section 2.3) anti-shistosomiasis (Section 2.4), and anti-trypanosomiasis (Section 2.5).
2.1 Antimalarial activity Malaria (from Italian origin, ‘‘aria male,’’ meaning bad air) is caused by an erythrocytic protozoan parasite first identified in 1880 by Alfonse Laveran [7]. Every year, between 300 and 500 million people are infected with malaria in the endemic areas (Africa, India, southeast Asia, the Middle East, Oceania, and Central/South America) with 1–2 million of these infected people dying each year and a child dying of malaria approximately every 30 s. Human malaria is caused by four species of the genus Plasmodium; vivax, ovale, malariae, and falciparum. The falciparum species is responsible for the majority of human deaths from malaria. Humans contract malaria when bitten by the female of any one of the 60 species of Anopheles mosquito [9]. The life cycle of the parasite from mosquito to human blood, to the human liver, back to the blood, and back to another mosquito is well-known [9,10]. The antimalarial mechanism of action for the artemisinin class of compounds is the most extensively studied of its mechanisms of action. It is possible, though unproven, that all of the potential mechanisms, two prevailing hypotheses presented below, are at work. It is now widely believed that the 1,2,4-trioxane class of antimalarials, including artemisinin, exert their activity on the erythrocytic stage of the parasite life cycle. Additionally, the fact that the endoperoxide is vital for artemisinin’s antimalarial activity has been well-established [4]. The most widely agreed-upon view is that liberated heme acts as the source of FeII responsible for the activation of the endoperoxide bridge of artemisinin (1) to produce cytotoxic radical species. The liberated heme results from the breakdown of hemoglobin that Plasmodium uses as a food source. The hemoglobin is transported to the parasitic food vacuole where it is digested becoming amino acids essential to parasite life and the aforementioned free heme that is toxic to the parasite. The parasite then detoxifies the free heme by converting it to hemazoin, a polymeric form of heme. It is believed that this potentiated heme in the form of hemazoin is what activates the artemisinin family of compounds. The activation cascade is initiated by reduction of the peroxide bond with heme FeII to give an oxy radical and oxidized heme FeIII that can lead to a number of subsequent radical entities, all of which may be lethal to the parasite. This mechanism has been called into question by others who believe artemisinins disrupt parasitic calcium homeostasis by targeting sarco/endoplasmic reticulum Ca+2-ATPases (SERCAs) [11,12]. Artemisinin
362
Christopher Paul Hencken et al.
and thapsigargin, a sesquiterpene lactone with structural similarities to artemisinin and known SERCA inhibitor, were shown to be similarly potent as inhibitors of a P. falciparum SERCA ortholog (PfATP6) in Xenopus oocytes. However, this has been contested by the Posner lab with their demonstration that enantiomers of a fully synthetic trioxane had the same level of activity [13]. Were the SERCA enzymes the primary target, one would expect only one of the enantiomers to be active as is seen in other enzyme-based mechanisms. Additionally, the epimers of artemether (3) showed nearly equal in vivo activity (1.02 mg/kg of body weight for the a epimer vs. 1.42 mg/kg of body weight for the b epimer) [14] furthering the heme-based mechanism over the SERCA-based mechanism where a greater disparity in the activities would be expected. Some final points regarding the mechanism bear mentioning here. First, it was recently shown [15] that there exists a strong correlation between a set of fully synthetic trioxolanes’ ability to alkylate heme in the presence of FeII and their respective in vitro antimalarial activity. Curiously, artemisinin-derived derivatives were less potent heme-alkylators, suggesting possible alternative modes of action for the more structurally complex artemisinin derivatives. Another study [16] found that endoperoxides tagged with a fluorescent label localized within the parasitic digestive vacuole. It was proposed that heme-iron activation then followed, allowing parasite membrane damage to occur. Despite this promising progress, more work must be done to fully elucidate the mechanism of these compounds. For a more detailed discussion, we direct the interested reader to work done by the Posner lab and others [13,16], which supports this accepted mechanism, especially evidence of the alkylation of heme by artemisinin [17], even in malaria-infected mice [18], and those that argue against it, especially the work by the Haynes lab [5,6,19–21]. Despite the debate that still surrounds the mechanism of action, this family of compounds has displayed and continues to display an exceptional safety profile and very rapid, high levels of activity. However, artemisinin and dihydroartemisinin (DHA) (2) do suffer drawbacks such as low bioavailability and recrudescence of the disease. Even with these drawbacks, they are especially effective against severe cerebral malaria, one of the most lethal forms of the disease. Arteether (4), the ethyl ether of DHA, was designed with the intent on optimizing the lipophilicity of artemisinin to improve passive permeability through cell membranes and improve bioavailability. This is especially important in the treatment of cerebral malaria where the treatment needs to permeate the bloodbrain barrier (BBB). The BBB requires compounds to be fairly lipophilic with an octanol-water partition coefficient, or logP, value of W2. That is to say that the compound in question is two times more soluble in octanol than it is in water. Arteether, logP of 3.99, is more lipophilic than the parent artemisinin, logP of 2.94, and therefore is expected to be more effective against cerebral malaria. Arteether is also fast-acting and effective
Anti-Infective and Anti-Cancer Properties of Artemisinin
363
against drug-resistant strains of Plasmodium. Notably, the World Health Organization selected arteether as an emergency treatment for individuals infected with malaria. Sodium artesunate, the sodium salt of artesunic acid (5), is well-tolerated and less toxic than (1) and along with (5) represents an example of one of the few water-soluble derivatives [22].
O C-10 O
O C-10
O OR 2 R=H 3 R = CH3 4 R = CH2CH3
HO O 5
To combat the shortcomings of first generation endoperoxides such as metabolic instability, short plasma half-life, poor bioavailability, and chemical instability, a new generation of molecules has been designed. One new generation of molecules, the C-10 carba analog class, replaces the exocyclic C-10 carbon–oxygen bond with a carbon–carbon bond, resulting in derivatives with no exocyclic oxygen group at C-10. The range of C-10 analogs is large and varies from fluorinated substituents [23] to amino sulfones [24] to differing-length linked dimers [25]. Begue’s lab sought to improve the activity of artemisinin by the addition of fluorinated alkyl groups to a variety of artemsinins. The most active of these derivatives, (6), substitutes a trifluoromethyl group for the C-10 hydrogen of (2) creating a trifluoromethyl alcohol moiety. When tested in vitro against the D6 and W2 strains of falciparum malaria, (6) was active at very low levels, 2.6 and 0.9 nM, respectively. When (6) was given subcutaneously to P. bergei-infected mice, ED50 was 0.7 mg/kg and ED90 was 1.8 mg/kg, but when given orally, the ED50 and ED90 values were 4.3 and 13.0 mg/kg respectively. The oral values were not significantly better than those for sodium artesunate (5.4 mg/kg for ED50 and 15.3 mg/kg for ED90) while those for the subcutaneous route of administration were appreciably better than those for sodium artesunate (2.8 mg/kg for ED50 and 10.4 mg/kg for ED90) [23].
O C-10
OH
F3C 6
The Haynes lab prepared and tested several C-10 nitrogen derivatives against P. bergei of which (7) demonstrated the best mix of lipophilicity
364
Christopher Paul Hencken et al.
(logP of 2.49), ease of preparation (crystalline compound that is easily purified), safety parameters (considered ‘‘nontoxic’’ by the author), and activity. The activity of (7) was tested against chloroquine-sensitive P. bergei-infected and chloroquine-resistant P. yoelii-infected mice. When administered subcutaneously, the ED90 values were 1.5 and 3.9 mg/kg, respectively. When given orally to the mice the values were 3.1 and 5.0 mg/kg, respectively [24].
O
C-10
N
S O
O 7
The in vivo mouse activity results for several C-10 isobutylene-linked dimers prepared by the Posner lab showed these analogs to be orally ‘‘curative’’ 30 days post-infection at doses of 3 30 mg/kg. Eight derivatives were shown to double the lifetime of the animals relative to the control animals at 3 10-mg/kg doses. Shown below are the most active derivatives prepared and tested by the Posner lab to date. Both (8, 9) were curative at 3 30 mg/kg, while 3 10 mg/kg dosing of (8) displayed 17.7 days average survival post-infection and 3 10 mg/kg dosing of (9) displayed 16.3 days average survival post-infection. Control animals (drug-delivery vehicle only) survived 6–7 days in both dosing groups. Animals given (1) survived 7.2 days at 30 mg/kg and 6.5 days at 10 mg/kg [25].
O C-10
O C-10
O C-10
O C-10
O O O
N O
N H N
8
9
365
Anti-Infective and Anti-Cancer Properties of Artemisinin
Curiously, a family of hybrid analogs related to (10), which incorporate a traditional antimalarial drug into a hybrid molecule with (1), showed moderate increases in activity compared to (1), but less activity than (3) [26]. This is somewhat surprising when it is considered that these drugs should have the potential for multiple mechanisms of action. It is possible that the parasite was able to pump the compound out of the cell due to the presence of the chloroquine unit based on the accepted mechanism of malaria resistance to chloroquine.
O C-10 O O N O C-10
Cl
HN
H N
H N
HO
N 3
3
N
O N
O
10
Cl
N chloroquine
11
In another recent study, an artemisinin-quinine hybrid was found to have potent antimalarial activity with IC50 values ranging from 8.95 nm to 10.4 nM against various strains of P. falciparum [27]. Notably, hybrid analog (11) was found to have more potent activity than quinine or artemisinin alone and, even more importantly, more potent activity than a 1:1 molar concentration of artemisinin and quinine combination treatment. Together, these studies make the important point that artemisininbased compounds can be chemically combined with other drugs leading to increased biological activity compared to combined treatment with two separate drugs. Due to the fact that combination therapy has become the recommended form of treatment for malaria, HIV, cancer, and many other diseases, artemisinin’s tolerance of such incorporation is highly important. Furthermore, with the fact that combination therapy often involves a complicated pill-taking regimen, a molecule that incorporates both drugs is an important simplification of treatment that will lead to less drug failure due to poor patient compliance. As an added benefit, as with (11), these hybrid analogs have been shown to possess better activity than even a 1:1 molar concentration treatment of the two drugs alone.
366
Christopher Paul Hencken et al.
2.2 Anti-toxoplasmosis activity Toxoplasma gondii is an apicomplexan protozoan parasite that infects humans and has been linked to chronic neuropsychiatric diseases and behavioral abnormalities [28]. Toxoplasmosis is most commonly transmitted by consumption of undercooked, contaminated meats; contaminated water; or contact with feces from an infected cat. In most immuno-competent humans, the disease is asymptomatic; however, there are serious fetal complications if acquisition of the disease occurs near conception. In immunocompromised individuals, such as HIV patients or organ transplant patients, systemic infection and widespread organ damage can occur [29]. The first reported activity of artemisinin against T. gondii was in the early 1990s by Ou-Yang and co-workers [30]. More recently, new C-10 carba derivatives such as (12) have been prepared by the Posner lab demonstrating an IC50 ¼ 1.2 mM [31]. An additional study by the same lab revealed multiple potential mechanisms of action. In this work [32], derivatives of artemisinin (13, 14) were found to not only inhibit growth (IC50 ¼ 1.0 and 1.7 mM, respectively) of the parasite but also result in parasite death and prevent entry of the parasite into the cells. Of particular note, a derivative lacking the endoperoxide (15) was found to be more effective at inhibiting entry into the cell by the parasite than its peroxide-containing analog. This result suggests, especially when it is considered that this same nonperoxide derivative showed no growth inhibition and no parasite death, that entry inhibition may not be dependent on the peroxide while growth inhibition clearly is. H no peroxide O
O C-10
H
O O
Br
O C-10
O C-10
O C-10 O S
N
S
N
O
12
13
14
15 inactive
At this time, no mechanism of action for the observed T. gondii activity has been proposed. However, recent studies have implicated calcium homeostasis as a possible mechanism of action of artemisinin against apicomplexa through its interaction with SERCA-type Ca2+
Anti-Infective and Anti-Cancer Properties of Artemisinin
367
ATPase [33], similar to what has been proposed for malaria. However, it is also believed that activation of the peroxide by ferrous iron is essential and that the likely source of this ferrous iron is heme. Given the close proximity of the molecule to heme upon activation, it has also been argued that heme alkylation may be the mechanism of action, especially with respect to malarial activity, as described earlier. This is a less likely mechanism in the case of T. gondii, which perhaps is explained by the SERCA-based mechanism. To date, no study has been done that investigates the relative effectiveness of two enantiomers for T. gondii like was done for malaria disallowing for the same counterargument made with malaria. More work must be done to fully elucidate the mechanism of action against this target. It is possible, though thus far unproven, that these two similar parasites share a biological target for artemisinin. It is appropriate to point out here that artemisinin is not the only peroxide-containing molecule that has shown activity against T. gondii or malaria. For example, see the work by Chang and co-workers [34] and the work of Vennerstrom [15] who demonstrated that fully synthetic peroxides (16, 17) also possess activity against Toxoplasma and malaria, respectively. The results of Chang’s study [34] suggest that 1,2,4-trioxanes were able to block nucleotide synthesis of intracellular parasites. While it is possible that artemisinin-derived analogs have the same mode of action; there is currently no proof this is the case. O
Ph
Ph
O
O
Ph
O
O
Ph
O
O
16
17
IC50 = 5.98 µM
IC50 = 71.7 µM
2.3 Anti-leishmaniasis activity Leishmaniasis is a widespread disease that takes three major forms in humans: cutaneous leishmaniasis, mucocutaneous leishmaniasis, and the potentially lethal visceral leishmaniasis. All of these forms are caused by various protozoan parasites of the genus Leishmania and are transmitted by the female sand fly. Most of the visceral leishmanaiasis cases have been related to HIV infections [35]. Similar to the antimalarial activity, a wide variety of derivatives of artemisinin have been prepared to target
368
Christopher Paul Hencken et al.
leishmaniasis, with substitutions on different regions of the artemisinin parent molecule. All of the derivatives presented here are known to also possess antimalarial activity. CF3 Cl
O C-10
O C-10
CF3
Cl O
19
18
CF3
O C-10
O 20
Several derivatives (18–20) have been reported to have quite potent activity against Leishmania, notably (19), whose IC50 is a very impressive 0.3 mM [36].
2.4 Anti-shistosomiasis activity Schistosoma mansoni and other schistosoma flatworms are the causative agents for schistosomiasis. The schistosomas are ordinarily located in the blood vessels of the human host. Although S. mansoni is the most widespread schistosoma, others such as Schistosoma japonicum and Schistosoma haematobia are also known. The parasite is contracted by humans through contact with infected water by way of direct penetration of the human skin. After in vivo maturation, severe cases give rise to fibrosis of the liver and hepatosplenomegaly. Artemisinin derivatives, especially (3), are known to have activity against all of the schistosomas [37]. A recent clinical study in Sudan, where artesunate–sulfamethoxypyrazine–pyrimethamine or artemether-lumefantrine combinations were administered to patients co-infected with P. falciparum and S. mansoni, showed complete clearance of both parasites [38]. Curiously,
Anti-Infective and Anti-Cancer Properties of Artemisinin
369
a separate study found a lower cure rate for artesunate–sulfadoxide– pyrimethamine combinations compared to praziquantel alone [39]. In another study where schoolchildren infected with Schistosoma haematobium were treated with artesunate or praziquantel, both drugs demonstrated reduced egg counts in patients at a single dose of 40 mg/kg of praziquantel and 20 mg/kg artesunate. However, praziquantel was found to perform better [40], consistent with the aforementioned results [39].
2.5 Anti-trypanosomiasis activity The two most common human-afflicting trypanosomal infections are American and African trypanosomiasis. The American type, Trypanosoma cruzi, more commonly known as Chagas disease (named for Carlos Chagas, its discoverer in 1909), is passed from the feces of triatomine bugs to humans and affects 8–11 million people in South and Central America and Mexico. Many infected individuals are unaware of their infection and if left untreated will be lifelong and possibly fatal. The triatomine bugs are commonly found in earthen-made structures common in the rural regions of the affected countries. The African type, more commonly known as African sleeping sickness, is further distinguished based on the region of Africa where the infection was contracted. Trypanosoma brucei rhodesiense is responsible for East African sleeping sickness and Trypanosoma brucei gambiense is responsible for the West African form. Both types of African sleeping sickness are contracted from the painful bite of the honeybee-sized tsetse fly with 50,000–70,000 cases reported annually. T. b. gambiense is responsible for the majority of the African cases [41,42]. A recent report demonstrated the efficacy of (1, 2, 7, 21) against T. b. rhodesiense, T. cruzi, and L. donovani in vitro.
O
C-10
O
C-10
N
S O
O
7
F
21
370
Christopher Paul Hencken et al.
Both (7) and (21) exhibited IC50 values similar to that of (1) and (2) against both trypanosomes tested. The demonstrated IC50 values for (1), (2), (7), and (21) against T. cruzi were 13.4, 12.8, 23.3, and 17.9 mM, respectively. Likewise, the values against T. b. rhodesiense were 20.4, 24.6, 22.5, and 15.7 mM, respectively [43]. Recently, hybrid derivatives have been reported that incorporate either a moiety that permits targeted delivery of the drug (22, 23) or a second active unit with an ideally different mechanism of biological activity (24, 10). MeO NH2 N
O
N
O C-10
O N
N H2N
N
N H
O
X
22 X = O 23 X = NH
O
N
C-10 O
24 Cl
Chollet et al. recently demonstrated the superior activity of (22) and (23), compared to artesunate against all evaluated strains of Trypanosoma brucei [44]. T. brucei is known to take up diamidine, thus the superior biological activity of the diamidine-containing (22) and (23) was speculated to be due to targeted delivery of the artemisinin-based drug. However, no direct evidence of such an increase in delivery was provided.
3. OTHER THERAPEUTIC USES OF ARTEMISININ Although the antiparasitic uses of artemisinin are much more common, there are other uses that are becoming more prevalent. For example, antiviral (Section 3.1), antifungal (Section 3.2), and anticancer (Section 3.3) properties are discussed here. Although the anticancer activity is mentioned rather briefly here, the anticancer properties of this family of compounds are at least as well-documented as each of the antiinfective properties, with the only exception being malaria.
3.1 Antiviral activity One of the earliest reports of artemisinin’s antiviral activity came in the early 1980s [45]. Sodium artesunate has been shown to inhibit the human
Anti-Infective and Anti-Cancer Properties of Artemisinin
371
cytomegalovirus (HCMV) (IC50 ¼ 5.9 mM against the AD169 strain with 91% inhibition at 15 mM) [34] and the Herpes simplex virus type 1 (93% inhibition at 15 mM for a clinical isolate, no IC50 reported) [46]. It was found that sodium artesunate inhibited central regulatory processes of HCMV-infected cells such as activation pathways dependent on NF-kB or Sp1 [46]. This was suggested to interfere with critical host-cell-type interactions and metabolism requirements for viral replication. However, since other sesquiterpene lactones have shown similar anti-HCMV activity [47,48], this suggests it may not be the endoperoxide that is responsible for the activity. It should be stated, however, that no mechanism of action has been elucidated to date and that although artemisinin does contain a lactone, sodium artesunate does not. Sodium artesunate has also been found to inhibit the Epstein–Barr virus [49]. Furthermore, sodium artesunate was found to be active at 600 nM concentration against both the M-tropic and the T-tropic HIV-1 strains [46]. Although this concentration is quite high, given the relative safety of the artemisinin family of drugs, even treatment at this high concentration may be well-tolerated. Considering that this is thus far an unused method of HIV treatment, the possibility becomes more attractive. Curiously, a separate study investigating patients with HIV/malaria co-infection not only showed delayed clearance of P. falciparum but no report of anti-HIV activity for artemisinin [50]. It was also found that hepatitis B virus (HBV) DNA release was inhibited by sodium artesunate at an IC50 of 0.5 mM with host cell viability being reduced at 20 mM, and the compound also showed activity against HBV at W10 mM. Meanwhile, artemisinin was found to inhibit hepatitis C virus (HCV) replicon replication in a dose-dependent manner in two HCV subgenomic replicon constructs at concentrations that did not affect the host cells. The reader is referred to recent broad reviews [47–49] and a recent patent [51], describing the antiviral activity of artemisinin and its semisynthetic derivatives, for more information and additional references.
3.2 Antifungal activity There have been several accounts of artemisinin derivatives possessing antifungal activity. For example [52], it has been shown that a-arteether (4) inhibits various genotypes of the EG-1-103 and F-400 strains of Saccharomyces cerevisiae at minimal inhibitory concentrations of 2.0 and 1.0 mg/mL respectively. Furthermore in another study [53], a wide variety of artemisinin-derived compounds (25–27) displayed activity against the yeasts Canadida albicans and Cryptococcus neoformans at IC50 values ranging from 0.045 mg/mL to 30 mg/mL. Curiously, a deoxo derivative (27) related to 15 also displayed marginal activity, suggesting
372
Christopher Paul Hencken et al.
the peroxide may not be the active pharmacophore in this application. Generally, the activity against C. albicans was far weaker than C. neoformans and in some cases was completely absent.
O
H
O
O
O
O
H
O H
O
HO HO 25
26
27
It was also recently discovered [54] that artemisinin displayed growth inhibitory properties against a set of isogenic S. cerevisiae strains that carried disruptions of the major multidrug ABC transporter genes in the multidrug-resistant PDR1-3 background. In this study, a synergistic effect was observed between artemisinin and ketoconazole where genotypes (PDR1-3 and PDR1) that displayed minimum inhibitory concentrations (MICs) W200 mg/mL with artemisinin alone displayed MICs of 25 and 3 mg/mL, respectively, when artemisinin was employed to potentiate ketoconazole. Against two other strains (D5 and D1D2D5), artemisinin alone possessed an MIC of 50 mg/mL. To date, there is no clear explanation that accounts for the antifungal activity of the artemisinin family of compounds.
3.3 Anticancer activity Nearly 1.5 million Americans were diagnosed with cancer in 2008 with more than one-third of them dying from the disease. Cancer (sarcoma, carcinoma, leukemia, lymphoma/myeloma, and central nervous system cancer) treatments have come to include radiation therapy, chemotherapy, surgery, and other treatment methods including anti-angiogenesis therapy, gene therapy, and hyperthermia [55]. The anticancer properties of artemisinin have been under in vivo investigation since the 1980s. Anfosso et al. demonstrated a correlation between angiogenesis-related genes and the cellular response to artemisinins [56]. They found many angiogenesisregulating factors among their panel such as, vascular endothelial growth factor-C (VEGFC), fibroblast growth factor-1 (FGF1), matrix metalloproteinase-9 (MMP9), thrombospondin-1 (THBS1), hypoxia-inducing factor-a (HIF1A), and angiogenin (ANG) [57–59]. Artemisinin has also been found
Anti-Infective and Anti-Cancer Properties of Artemisinin
373
to inhibit proliferation, migration, and tube formation in human umbilical vein endothelial cells (HUVEC); inhibit VEGF binding to surface receptors on HUVEC; and reduce expression of VEGF receptors on HUVECs [60– 62]. Artesunate has been shown to play other roles in the control of cancer such as inducing apoptosis and oncogene deactivation/tumor suppressor gene activation [63]. The reader is encouraged to read the review by Krishna for further discussion of this topic [64].
O C-10
O C-10 O O
O P OR
28 R = Me 29 R = Ph
Two of a series of bis-trioxane dimers (28, 29) were shown to be more potent against cancer, in vitro, than doxorubicin, a currently used cancer chemotherapy agent. It was noted that all the cell lines sensitive to (28) and (29) all overexpressed transferrin receptors [65]. Based on that information, nuclear iron-dependent activation to free radical species that damage DNA was proposed to be involved in the tumoricidal mechanism of action [66]. To further the idea that transferrin receptors are a likely target, a series of artemisinin–transferrin conjugates were prepared and tested against the prostate cancer cell line DU 145. These conjugates were found to be cytotoxic to the cell line through a transferrin receptor–dependent induction of apoptosis [67]. Currently, it is unknown whether the anticancer properties of artemisinin derivatives are due to cytotoxicity against the cancer cells, antiangiogenesis properties, or some combination of both. Other recent work has shown that artemisinin-acridine hybrids such as the aforementioned (23) were found to have a cytotoxicity profile against several cancer cell lines, with IC50 values four times lower than DHA in several derivatives [68], demonstrating the widespread applicability of this concept of hybrid drug analogs.
4. TOXICITY Great debate has raged on the topic of toxicity for this compound class. While generally characterized as having excellent tolerability and
374
Christopher Paul Hencken et al.
safety [69], some adverse affects, particularly in non-human mammals, have been observed.
4.1 Evidence in favor of safety A review of clinical trials [70] showed that together, 9% of patients showed adverse drug reactions. These reactions included neutropenia, reduced reticulocyte count, anemia, eosinophilia, acute haemolysis, elevated aspartate aminotransferase, ECG abnormalities (w/o clinical effect), transient bradycardia, prolongation of the QTc interval, prolonged PR interval (first-degree atrioventricular block), atrial extrasystoles, and non-specific T-wave changes. In all cases, the effects were independent of the artemisinin derivative and route of administration. An additional study [71] showed that when treating uncomplicated falciparum malaria with artemisinin derivatives alone, substantially fewer side effects were observed than with mefloquine-containing combination therapies. Notably, there was significantly less nausea, vomiting, anorexia, and dizziness. Furthermore, studies have shown no adverse effects on mother or fetus if artesunate or artemether are used to treat acute falciparum at various stages of gestation [72] or during breast feeding [69]. However, this matter can hardly be considered closed. Although several studies have been done and no evidence of adverse effects have been observed, the sample size has thus far been small [73], and complications (most notably, a high percentage of post-implantation losses) have been observed at doses of 35 and 75 mg/kg in Wistar rats, especially when administered early in the pregnancy [74]. Meanwhile, another study using Wistar rats and considerably lower doses of artemether (3.5 and 7 mg/kg) during blastogenesis, organogenesis and fetal period had no adverse effects other than reduction in fetal body weight and pre-term delivery in 3/10 rats at the 7-mg/kg dose. Some fetal growth retardation without incidence of malformations was also noted in the study [75].
4.2 Evidence of neurotoxicity It is important to note that dogs [76], mice, rats, and Rhesus monkeys [77–79] have all demonstrated evidence of neurotoxicity apparently caused by artemisinin derivatives. Specifically, gait disturbances, loss of spinal and pain response reflexes, cardio respiratory depression, and even death have been documented. In mice, balance was damaged irreversibly and death also occurred. Importantly, it was found that intramuscular administration was found to be more toxic than oral [69]. In humans, even when adverse neurological effects have been documented, the effects resolved with time after the treatment was
Anti-Infective and Anti-Cancer Properties of Artemisinin
375
stopped [71]. These effects included problems with co-ordination, fine finger dexterity, hearing, nystagmus, and balance. For more information concerning the toxicity of this important class of compounds, we direct the interested reader to a relatively recent review [80]. An explanation for the apparent lack of a correlation between animal toxicity and human toxicity is still lacking and is urgently needed.
5. CONCLUSION The artemisinin family of compounds has become more important in recent years. The very diverse biological activity and relative safety of this family of drugs make it important in the global fight against many diseases. In the coming years, it is anticipated that this exotic natural product will become even more important to the chemotherapy of various diseases, perhaps even above and beyond those mentioned here.
REFERENCES [1] J. M. Liu, M. Y. Ni, J. F. Fen, Y. Y. Tu, Z. H. Wu, Y. L. Wu and W. S. Zhou, Huaxue Xueabo, 1979, 37, 129. [2] D. L. Klayman, A. J. Lin, W. Acton, J. P. Scovill, J. M. Hoch, W. K. Milhous and A. D. Theoharides, J. Nat. Prod., 1984, 47, 715. [3] M. A. Avery and K. M. Muraleedharan, International patent 03/095444A1 2003. [4] D. L. Klayman, Science, 1985, 228, 1049. [5] R. K. Haynes and S. Krishna, Microbes Infect., 2004, 6, 1339. [6] S. Krishna, L. Bustamante, R. K. Haynes and H. M. Staines, Trends Pharmacol. Sci., 2008, 29, 520. [7] N. J. White, Science, 2008, 320, 330. [8] T. Efferth, Planta Med., 2007, 73, 299. [9] www3.niaid.nih.gov/topics/Malaria/understandingMalaria/facts.htm [10] www.cdc.gov/malaria/faq.htm [11] R. K. Haynes, D. Monti, D. Taramelli, N. Basilico, S. Parapini and P. Olliaro, Antimicrob. Ag. Chemother, 2003, 47, 1175. [12] U. Eckstein-Ludwig, R. J. Webb, I. D. A. vanGoethem, J. M. East, A. G. Lee, M. Kimura, P. M. O’Neill, P. G. Bray, S. A. Ward and S. Krishna, Nature, 2003, 424, 957. [13] P. M. O’Neill, S. L. Rawe, K. Borstnik, A. Miller, S. A. Ward, P. G. Bray, J. Davies, C. H. Oh and G. H. Posner, ChemBioChem, 2005, 6, 2048. [14] Chinese-Cooperative-Research-Group, J. Trad. Chin. Med., 1982, 2, 31. [15] D. J. Creek, W. M. Charman, F. C. K. Chiu, R. J. Prankerd, Y. Dong, J. L. Vennerstrom and S. A. Charman, Antimicrob. Ag. Chem., 2008, 52, 1291. [16] C. L. Hartwig, A. S. Rosenthal, J. D’Angelo, C. F. Griffin, G. H. Posner and R. A. Cooper, Biochem. Pharmacol., 2009, 77, 322. [17] P. L. Olliaro, R. K. Haynes, B. Meunier and Y. Yuthavong, Trends Parasitol., 2001, 17, 122. [18] F. B.-E. Garah, C. Claparols, F. Benoit-Vical, B. Meunier and A. Robert, Antimicrob. Ag. Chem., 2008, 52, 2966. [19] R. K. Haynes, W. C. Chan, C.-M. Lung, A.-C. Uhlemann, U. Eckstein, D. Taramelli, S. Parapini, D. Monti and S. Krishna, ChemMedChem, 2007, 2, 1480.
376
Christopher Paul Hencken et al.
[20] S. Krishna, A.-C. Uhlemann and R. K. Haynes, Drug Resist. Updat., 2004, 7, 233. [21] A.-C. Uhlemann, A. Cameron, U. Eckstein-Ludwig, J. Fischbarg, P. Iserovich, F. A. Zuniga, M. East, A. Lee, L. Brady, R. K. Haynes and S. Krishna, Nat. Struct. Mol. Biol., 2005, 12, 628. [22] A. Ryde´n and O. Kayser, Top. Heterocycl. Chem., 2007, 9, 1. [23] J. Be´gue´ and D. Bonnet-Delpon, ChemMedChem, 2007, 2, 608. [24] R. K. Haynes, B. Fugmann, J. Stetter, K. Rieckmann, H.-D. Heilmann, H.-W. Chan, M.-K. Cheung, W.-L. Lam, H.-N. Wong, S. L. Croft, L. Vivas, L. Rattray, L. Stewart, W. Peters, B. L. Robinson, M. D. Edstein, B. Kotecka, D. E. Kyle, B. Beckermann, M. Gerisch, M. Radtke, G. Schmuch, W. Steinke, U. Wollborn, K. Schmeer and A. Ro¨mer, Angew. Chem. Int. Ed., 2006, 45, 2082. [25] G. H. Posner, W. Chang, L. Hess, L. Woodard, S. Sinishtaj, A. R. Usera, W. Maio, A. S. Rosenthal, A. S. Kalinda, J. G. D’Angelo, K. S. Peterson, R. Stohler, J. Chollet, J. Santo-Tomas, C. Snyder, M. Rottmann, S. Wittlin, R. Brun and T. A. Shapiro, J. Med. Chem., 2008, 51, 1035. [26] N. C. P. Arauju, V. Barton, M. Jones, P. A. Stocks, S. A. Ward, J. Davies, P. G. Brat, A. E. Shone, M. L. S. Cristiano and P. M. O’Neill, Bioorg. Med. Chem. Lett., 2009, 19, 2038. [27] J. J. Walsh, D. Coughlan, N. Hehghan, G. Gaynor and A. Bell, Bioorg. Med. Chem. Lett., 2007, 17, 3599. [28] S. Bachmann, J. Schro¨der, C. Bottmer, E. F. Torrey and R. H. Yolken, Psychopathology, 2005, 38, 87. [29] A. M. Tenter, A. R. Keckeroth and L. M. Weiss, Int. J. Parasitol., 2000, 30, 1217. [30] K. Ou-Yang, E. C. Krug, J. J. Marr and R. L. Berens, Antimicrob. Ag. Chem., 1990, 34, 1961. [31] L. Jones-Brando, J. D’Angelo, G. H. Posner and R. Yolken, Antimicrob. Ag. Chem., 2006, 50, 4206. [32] J. G. D’Angelo, C. Bordo´n, G. H. Posner, R. Yolken and L. Jones-Brondo, J. Antimicrob. Chem., 2009, 63, 146. [33] K. Nagamune, W. L. Beatty and L. D. Sibley, Eukaryot. Cell, 2007, 6, 2147. [34] H. R. Chang, C. W. Jefford and J.-C. Peche`re, Antimicrob. Ag. Chem., 1989, 33, 1748. [35] J. Moreno, C. Caravate, C. Chamizo, F. Laguna and J. Alvar, Tran. R. Soc. Trop. Med. Hyg., 2000, 94, 328. [36] M. A. Avery, K. M. Muraleedharan, P. V. Desai, A. K. Bandyopadhyaya, M. M. Furtado and B. L. Tekwani, J. Med. Chem., 2003, 46, 4244. [37] X. Shuhua, M. Tanner, E. K. N’Goran, J. Utzinger, J. Chollet, R. Benquist, C. Minggang and Z. Jiang, Acta Trop., 2002, 82, 175. [38] I. Adam, O. A. Elhardello, M. O. Elhadi, E. Abdalla, K. A. Elmandi and F. H. Jansen, Ann. Trop. Med. Parasitol., 2008, 102, 39. [39] A. A. Mohamed, H. M. Mahgoub, M. Magzoub, G. I. Gasim, W. N. Eldein, A. Ahmed and I. Adam, Trans. Roy. Soc. Trop. Med. Hyg., 2009, in press. [40] D. DeClercq, J. Vercruysse, A. Kongs, P. Verlei, J. P. Dompnier and P. C. Faye, Acta Trop., 2002, 82, 61. [41] http://wwwn.cdc.gov/travel/yellowbookCh4-AfricanSleepingSickness.aspx [42] http://wwwn.cdc.gov/travel/yellowBookCh4-Chagas.aspx [43] Y. V. Mishina, S. Krishna, R. K. Haynes and J. C. Meade, Antimicrob. Ag. Chem., 2007, 51, 1852. [44] C. Chollet, A. Baliani, P. E. Wong, M. P. Barrett and I. H. Gilbert, Bioorg. Med. Chem., 2009, 17, 2512. [45] R. S. Qian, Z. L. Li, J. L. Yu and D. J. Ma, J. Tradit. Chin. Med., 1982, 2, 271. [46] T. Efferth, M. Marschall, X. Wang, S.-M. Huong, I. Hauber, A. Olbrich, M. Kronschnabl, S. Stamminger and E.-S. Huang, J. Mol. Med., 2002, 80, 233. [47] P. M. Bork, M. L. Schmitz, M. Kuhnt, C. Escher and M. Heinrich, FEBS Lett., 1997, 402, 85.
Anti-Infective and Anti-Cancer Properties of Artemisinin
377
[48] B. Siedle, A. J. Garcia-Pineres, R. Murillo, J. Schulte-Mo¨nting, V. Castro, P. Ru¨ngeler, C. A. Klaas, F. B. DaCosta, W. Kisiel and I. Merfort, J. Med. Chem., 2004, 47, 6042. [49] T. Efferth, M. R. Romero, D. G. Wolf, T. Stamminger, J. J. G. Marin and M. Marschall, Clin. Infect. Dis., 2008, 47, 804. [50] Y. Binku, E. Mekonnen, A. Bjorkman and D. Wolday, Ethiop. Med. J., 2002, 40, 17. [51] J. Vandenkerckhov, E. B. Sas, E. Pets and J. VanHemel, WO International Patent 2004/ 041176A2, 2004. [52] S. Kumar, S. P. S. Khanuja, T. R. S. Kumar, D. C. Jain, S. Srivastava, A. K. Bhattacharya, D. Saikia, A. K. Shasany, M. P. Darokar and R. P. Sharma, US Patent 6,127,405, 2000. [53] A. M. Galal, S. A. Ross, M. Jacob and M. A. ElSohly, J. Nat. Prod., 2005, 68, 1274. [54] M. Kolaczkowski, A. Kolaczkowska and F. R. Stermitz, Microb. Drug Resist., 2009, 15, 11. [55] http://seer.cancer.gov/statfacts/html/all.html [56] L. Anfosso, T. Efferth, A. Albini and U. Pfeffer, Pharmacogenomics J., 2006, 6, 269. [57] J. Folkmann, J. Natl. Cancer Inst., 1990, 82, 4. [58] A. F. Karamysheva, Biochemistry (Moscow), 2008, 73, 751. [59] M. Shibuya, J. Biochem. Mol. Biol., 2006, 39, 469. [60] H.-H. Chen, H.-J. Zhou, W.-Q. Wang and G.-D. Wu, Cancer Chemother. Pharmacol., 2004, 53, 423. [61] H.-H. Chen, H.-J. Zhou, G.-D. Wu and X.-E. Lou, Pharmacology, 2004, 71, 1. [62] H.-H. Chen, H.-J. Zhou and X. Fang, Pharmacol. Res., 2003, 48, 231. [63] T. Efferth, G. Ruecker, M. Falkenberg, D. Manns, A. Olbrich and U. Fabry, Arzneimittelforschung, 1996, 46, 196. [64] S. Krishna, L. Bustamante, R. K. Haynes and H. M. Staines, Trends Pharmacol. Sci., 2008, 29, 520. [65] G. H. Posner, J. D’Angelo, P. M. O’Neill and A. Mercer, Expert Opin. Ther. Patents, 2006, 16, 1665. [66] J. P. Jeyadevan, P. G. Bray, J. Chadwick, A. E. Mercer, A. Byrne, S. A. Ward, B. K. Park, D. P. Williams, R. Cosstick, J. Davies, A. P. Higson, E. Irving, G. H. Posner and P. M. O’Neill, J. Med. Chem., 2004, 47, 1290. [67] I. Nakase, B. Gallis, T. Takatani-Nakase, S. Oh, E. Lacoste, N. P. Singh, D. R. Goodlett, S. Tanaka, S. Futaki, H. Lai and T. Sasaki, Cancer Lett., 2009, 274, 290. [68] M. Jones, A. E. Mercer, P. A. Stocks, L. J. I. LaPense´e, R. Cosstick, B. K. Park, M. E. Kennedy, I. Piantanida, S. A. Ward, J. Davis, P. G. Bray, S. L. Rawe, J. Baird, T. Charidza, O. Janneh and P. M. O’Neill, Bioorg. Med. Chem. Lett., 2009, 19, 2033. [69] W. R. J. Taylor and N. J. White, Drug Saf., 2004, 27, 25. [70] I. R. Ribeiro and P. Olliaro, Med. Trop. (Mars), 1998, 58(3 suppl.), 50. [71] R. Price, M. vanVugt, L. Phaipun, C. Luxemburger, J. Simpson, R. McGready, F. TerKuile, A. Kham, T. Chongsuphajaisiddhi, N. J. White and F. Nosten, Am. J. Trop. Med. Hyg., 1999, 60, 547. [72] R. McGready, T. Cho, N. K. Keo, K. L. Thwai, L. Villegas, S. Looareesuwan, N. J. White and F. Nosten, Clin. Infect. Dis., 2001, 33, 2009. [73] S. Dellicour, S. Hall, D. Chandramohan and B. Greenwood, Malar. J., 2007, 6, 15. [74] A. C. Boareto, J. C. Muller, A. C. Bufalo, G. G. K. Botelho, S. L. deAraujo, M. A. Foglio, R. N. deMorais and P. R. Dalsenter, Reprod. Toxicol., 2008, 25, 239. [75] M. H. El-Dakdoky, Food Chem. Toxicol., 2009, 47, 1437. [76] China-Cooperative-Research-Group-On-Qinghaosu-and-its-Derivatives-as-Antimalarials, J. Tradit. Chin. Med., 1982, 2, 45. [77] R. F. Genovese, D. B. Newman, J. M. Petras and T. G. Brewer, Pharmacol. Biochem. Behav., 1998, 60, 449.
378
Christopher Paul Hencken et al.
[78] J. M. Petras, G. D. Young, R. A. Bauman, D. E. Kyle, M. Gettayacamin, H. K. Webster, K. D. Corcoran, J. O. Peggins, M. A. Vane and T. G. Brewer, Anat. Embryol. (Berl), 2000, 201, 383. [79] A. Nontprasert, S. Pukrittayakamee, A. M. Dondorp, R. Clemens, S. Looareesuwan and N. J. White, Am. J. Trop. Med. Hyg., 2002, 67, 423. [80] T. Gordi and E.-I. Lepist, Toxicol. Lett., 2004, 147, 99.
CHAPT ER
19 Recent Advances in the Inhibition of Bacterial Type II Topoisomerases Gregory S. Bisacchi and Jacques Dumas
Contents
1. Introduction 2. Inhibition at the ATP-Binding Site 2.1 Novobiocin/clorobiocin analogs 2.2 Cyclothialidine analogs 2.3 Ethyl-ureas 2.4 Lead compounds derived from fragment-based approaches 2.5 Miscellaneous compounds or inhibitors 3. Inhibition Outside of the ATP-Binding Site 3.1 Classical quinolone inhibitors 3.2 Quinolone-like structures 3.3 Quinolines 3.4 Quinoline pyrimidine triones 4. Conclusion References
379 381 381 382 383 385 387 388 388 390 391 392 393 393
1. INTRODUCTION The natural product antibiotic novobiocin (1) was reported in 1955 [1], and in 1960, synthetic antibacterial compounds having a 3-carboxyquinolone core (e.g., 2) were first disclosed [2]. These discoveries were made AstraZeneca Pharmaceuticals LP, 35 Gatehouse Drive, Waltham, MA 02451, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04419-4
r 2009 Elsevier Inc. All rights reserved.
379
380
Gregory S. Bisacchi and Jacques Dumas
during the so-called golden age of antibiotics [3] and were based on inhibition of bacterial growth in culture. Target-based discovery of antibacterials was still several decades in the future. By the mid-to-late 1970s, the molecular mechanism of novobiocin and the quinolones was elucidated to be the inhibition of bacterial topoisomerase, specifically DNA gyrase [4–5]. In 1994, a target-based screen of fermentation broths led to the discovery of cyclothialidine (7) as an inhibitor of DNA gyrase [6]. From the 1990s to the present day, bacterial topoisomerase has steadily gained prominence as a target of choice for the discovery of antibacterial agents [7–10].
OH O H2N
O
O
O
O
O OH
OH
O
H N O
CO2H O
Me
X
N
Me
1
2 X = CH 3 X=N
Bacterial topoisomerases, like their eukaryotic counterparts, are divided into two main groups: Type I, which involves DNA singlestrand processing and is typically ATP-independent, and Type II, which processes double-stranded DNA and requires ATP. Topoisomerase function is absolutely required by both prokaryotic and eukaryotic cells to allow, among other processes, proper unwinding and topological display of the DNA strands before replication [11]. To date, no advanced inhibitors of bacterial Type I topoisomerases have been reported [12]. In contrast, useful inhibitors of Type II topoisomerases are abundant, from the well-established fluoroquinolone class to a variety of emerging classes in various stages of clinical or preclinical evaluation. The two principal subclasses of bacterial Type II topoisomerase are DNA gyrase and topoisomerase IV. DNA gyrase exists as a tetramer, consisting of two subunits of GyrA and two subunits of GyrB [13]. The GyrB subunit contains the ATP-binding pocket. Tetrameric topoisomerase IV is closely related to DNA gyrase and consists of two subunits of ParC and two subunits of ParE, with ParE containing the ATP-binding pocket (Figure 1). Novobiocin and cyclothialidine inhibit DNA gyrase and topoisomerase IV by binding to the ATP sites in GyrB and ParE, respectively, while the quinolone class inhibits these enzymes by binding to a site in GyrA and ParC near the intersection of the subunits and the associated DNA strand. Medicinal chemistry programs targeting inhibitors of the ATP-binding pocket of GyrB/ParE have the benefit of
Recent Advances in the Inhibition of Bacterial Type II Topoisomerases
381
ATP binding sites novobiocin cyclothialidine ethyl-ureas X-ray structure-guided 2GyrB
2ParE Other binding sites quinolones isothiazoloquinolone 3-aminoquinazolinedione quinolines quinoline pyrimidine triones emerging structural information
2GyrA
DNA Gyrase
2ParC
Topoisomerase IV
Figure 1 Schematic diagram of bacterial type II topoisomerases.
crystallographic structural information for analog design [14]. However, outside the ATP sites, details of how inhibitors bind are only emerging [15]. This chapter broadly organizes small-molecule inhibitors of bacterial Type II topoisomerase into those that bind at the ATP sites and those that bind at other sites within the enzyme tetramers. Because the quinolone class of GyrA/ParC inhibitors has been employed therapeutically for decades, many bacteria species have developed substantial, and ever-increasing, resistance to this class, as they have to varying degrees for all other classes of antibacterials. The pharmaceutical industry and medical community response to the problem of increasing resistance to marketed quinolone antibacterial drugs has been, in part, both to investigate further variants of this established class, which may possess incremental advantages against resistant pathogens, and to discover and develop entirely different classes of inhibitors of bacterial topoisomerase II, which are not susceptible to pre-established resistance mechanisms. Both of these approaches are described in this chapter.
2. INHIBITION AT THE ATP-BINDING SITE 2.1 Novobiocin/clorobiocin analogs Antibiotics of the coumarin class, such as novobiocin (1) or clorobiocin (4), bind to the ATP pocket of GyrB. These analogs are produced by Streptomyces strains; for the most part, the gene clusters responsible for their biosynthesis have been elucidated and characterized [16]. A large body of work has been devoted to the generation of novel analogs through directed mutagenesis of Streptomyces strains. Deletion of the cloQ
382
Gregory S. Bisacchi and Jacques Dumas
and cloL genes in Streptomyces coelicolor M512 suppresses the biosynthesis of clorobiocin [17]. As a second step, introduction of the couL sequence (an amide synthase with broad specificity, involved in the biosynthesis of coumermycin) resulted in the identification of three novel analogs with a modified substitution pattern on the coumarin core, for example, ferulobiocin (5). A similar two-stage mutagenesis approach produced analogs with significant anti-bacterial potencies against Gram-positive organisms [18,19]. For example, coumarin 6 inhibits the growth of Staphylococcus aureus ATCC 29213 with a minimum inhibitory concentration (MIC) of o0.06 mg/mL. Despite this, activity against wild-type Gram-negative organisms remains elusive for this compound class. OH OH O O
H N
O
H N
O O
O
H3C
O
O
Cl
OH
4
OH OH H N
O O H N
O
OCH3 O
O
O
H3C
O
OH
O
Cl 5
OH OH H N
O O H N H3C
O O O
O
O
OH
CH3 6
O
383
Recent Advances in the Inhibition of Bacterial Type II Topoisomerases
2.2 Cyclothialidine analogs Cyclothialidine (7), a 12-membered ring lactone isolated from Streptomyces filipinensis, is a potent inhibitor of GyrB and, like novobiocin, binds to the ATP site [20]. Its structure, albeit complex, allows the preparation of analogs through total synthesis. The binding mode of cyclothialidine with GyrB is known and suggests that one of the phenolic groups is required for activity, while other functional groups in the molecule can tolerate derivatization. A medicinal chemistry approach has focused on simplified macrolactones and modifications of the C-terminal chain [21]. Macrolactone 8 is a potent inhibitor of Escherichia coli DNA gyrase in a supercoiling assay, exhibiting a maximum non-effective concentration (MNEC) of 2 ng/mL. This analog shows a potent MIC against S. aureus Smith (0.25 mg/mL) and in vivo efficacy in a murine septicaemia model using the same pathogen (ED50 ¼ 8.5 mg/kg). The introduction of a thioamide functionality and the enlargement of the macrolactone ring significantly raised potency. Higher plasma free fraction and improved pharmacokinetics have been attributed to the aminomethyl group on the oxadiazole ring.
HN OH
S
O
COOH OH
O HN
O
O
O
N
N H
HO O
O
HN OH
NH2
N
S
NH2
N
S
O MeO O
HO
7
8
OH
Cl OH
N N H
N MeO Br
Cl
O
9
O 10
Simplified analogs conserve the key phenolic residue of cyclothialidine as a minimum pharmacophore. The same team used 9, an high throughput screen (HTS) hit, as a starting point for further modification [22].
384
Gregory S. Bisacchi and Jacques Dumas
Quinoline 10 shows broad Gram-positive activity and comes close to novobiocin in terms of potency. The introduction of the dehydro-indolone core was a critical step in improving anti-bacterial activity. In addition, the authors suggest that the acidity of the phenol group, and/or the overall lipophilicity of the molecule seem to play an important role in penetration of the bacterial membrane.
2.3 Ethyl-ureas A new class of GyrB inhibitors, built around an ethyl-urea pharmacophore, has recently emerged in the literature. Carbamate 11 was first identified in a high-throughput assay targeting the ATP-ase activity of GyrB [23]. The optimization of 11 into VRT-752586 (12) was aided by docking into a crystal structure of novobiocin complexed with S. aureus enzyme. Three major steps led to 12: The carbamate oxygen of 11 is replaced by an NH, improving an interaction with Asp-73, an amino acid residue bound to the adenine of ATP. The 3-pyridyl substitution in 12 takes advantage of a conserved arginine residue, Arg-136 (E. coli numbering), creating an interaction already present in the complex between novobiocin and GyrB. The 3-fluoro-pyridine-2-yl substituent fills available space in the carbohydrate-binding pocket of the enzyme and makes favorable lipophilic interactions, while locking the inhibitor in its most productive conformation with the use of an internal hydrogen bond. N OH
O
O O
HN N
N HN
HN
N H
N H
12 F N
11
N
N HN
N
O
O HN
N
N
N HN
HN N
N
N N
14
13 N
N
N
385
Recent Advances in the Inhibition of Bacterial Type II Topoisomerases
N
N
O HN
O N
HN N
HN N 15 S
N HN
N
S
16 S
N
Two publications devoted to 12 report a robust antibacterial profile. VRT-752586 acts as a dual inhibitor of GyrB and ParE, and each of these enzymes may be responsible for its anti-bacterial activity, depending on the bacterial species [24]. In addition, its dual targeting leads to very low resistance frequencies (o5.2 1010 at 4 MIC measured in Enterococcus faecalis). This analog shows broad activity against Gram-positive organisms (MIC90 equal of better than 0.12 mg/mL in all organisms tested) [25]. VRT-752586 is, however, less potent against fastidious Gramnegative pathogens, Mycobacterium tuberculosis, two atypical pathogens (Legionella pneumophila and Mycoplasma pneumoniae), and a panel of anaerobes. Evidence of Gram-negative efflux was reported against E. coli (no wild-type activity, but MIC of 0.13 mg/mL against a tolC mutant). Finally, VRT-752586 shows in vivo activity in two models of infection in mice (S. aureus/thigh and Streptococcus pneumoniae/lung), at doses ranging from 25 mg/kg to 50 mg/kg i.v. [23]. The potent activity of this class generated much interest and led several pharmaceutical companies into the field. Scaffold-hopping initiatives have conserved the urea moiety, the adjacent heteroatom, and both heteroaryl substituents, resulting in triazolo-pyridines such as 13. While this analog defines new proprietary space and shows useful anti-bacterial potencies against Gram-positive organisms, it appears less potent than its benzimidazole counterpart [26]. Additional triazolo-pyridines [27], as well as the related imidazo-pyridines [28] and benzothiazoles [29], have been described in the patent literature (compounds 14–16 are representative examples). Overall, the chemical space around the ethyl-ureas is now becoming increasingly crowded.
2.4 Lead compounds derived from fragment-based approaches The ATP-binding site of GyrB appears well-suited for fragment-based approaches. The adenine of ATP binds to a conserved aspartate residue (Asp-73 in S. aureus) and an active-site water molecule in a bi-dentate fashion. These key interactions can be mimicked with a variety of adenine-like cores, some of them not dissimilar to donor-acceptor kinase
386
Gregory S. Bisacchi and Jacques Dumas
inhibitor fragments. In addition, the X-ray structures of novobiocin and clorobiocin complexed with DNA gyrase provide a hint that these fragments could be grown toward the solvent interface and point at two conserved arginine residues (Arg-76 and Arg-136) to deliver additional binding energy. A first example of such an approach makes use of a triazine core [30]. Triazine 17 was identified as a screening hit in a DNA gyrase assay (IC50 ¼ 0.75 mM), and its binding mode was elucidated by X-ray crystallography. Fine-tuning of the triazine substitution, and incorporation of a coumarin fragment to mimic novobiocin, led to the identification of 18, a potent DNA gyrase inhibitor (IC50 ¼ 70 nM). H N
H N
N N
F
H N
Cl
N
N
F
HN
O
N OH
NH2
Cl
17
O
N
18 F
H N N
H N N
O
S
O2N O
19
20 O
NH2
OH
O H N
N
O
N N
N
O
O
O 22
21
Cl Cl
Br
O
387
Recent Advances in the Inhibition of Bacterial Type II Topoisomerases H2N
H N
O
H N
O
OH
O S
HN
Br
N
HN
Cl
N
Cl
N
23
O
N
24
Cl
In a similar way, indazole 19, identified in a ‘‘needle screening’’ effort [31], effectively binds to DNA gyrase in the high micromolar range, through a contact at Asp-73. As seen previously, addition of a moiety targeted at Arg-76 and Arg-136 led to 20, a much more potent analog (MNEC ¼ 0.03 mg/mL). A similar approach from the same team [32] culminated in pyrazolo-pyrimidine 21, also a potent inhibitor of DNA gyrase (MNEC ¼ 0.06 mg/mL). Recently, an NMR screen directed at the ATP pocket of GyrB led to pyrrole 22, with a Kd of 160 mM [33]. Iterative, structure-based design [34,35] optimized the pyrrole substitution and introduced additional interactions with both conserved arginines, as in pyridine 23, which shows an IC50 of 10 nM in an E. coli ATPase assay. Productive interactions of 23 with a 24-kDa subunit of GyrB were confirmed by X-ray crystallography. The pyrrolamide series demonstrates promising activity across a broad range of pathogens; its mode of action is bactericidal and consistent with DNA synthesis inhibition. Thiazole-carboxylic acid 24, a representative example, shows dose-dependent, oral activity in a murine S. pneumoniae lung infection [36,37].
2.5 Miscellaneous compounds or inhibitors In the past few years, several GyrB inhibitor structures have been published in the literature. All three compounds below bind in the ATP pocket of DNA gyrase, albeit with modest potencies. OH
H N O
HO
O OH
H N OH 25
N
OH
O
26
388
Gregory S. Bisacchi and Jacques Dumas
OH OH HO
O OH O OH
OH O 27 OH OH
The indolinone core 25 binds to GyrB through its donor–acceptor pair, in a tyrosine-kinase-like fashion [38]. The natural flavinoid quercetin (26) and its analog ()-epigallocathechin gallate (27) also bind in the ATP pocket of GyrB with affinities in the double-digit micromolar range [39,40].
3. INHIBITION OUTSIDE OF THE ATP-BINDING SITE 3.1 Classical quinolone inhibitors Following the launch of nalidixic acid (3) in 1962, research into variants of the classic quinolone/napthyridone structures has continued unabated to this day [15,41]. With only a few notable exceptions (covered in Section 3.2), replacement of the critical 3-carboxy substituent has resulted in inactive or less useful agents. The vast majority of useful antibacterial agents in this class rely upon variation of peripheral substituents, leaving the heterocyclic cores largely intact. First-generation analogs such as 2 and 3 had narrow activity against Gram-negative pathogens, mainly E. coli. Second- and third-generation analogs culminated in the commercialization of a number of 6-fluoroquinolones, notably, ciprofloxacin (1987), levofloxacin (1993), and moxifloxacin (1999), whose spectrum and potencies encompassed most Gram-positive and Gramnegative species, as well as anaerobes and atypical bacterial pathogens. The structure of moxifloxacin (28) is shown to exemplify this later, broadspectrum class. Yet, many other quinolones launched during this more recent period were beset by unexpected toxicological issues of varying severity and were withdrawn from a number of markets. Notable examples include trovafloxacin (hepatoxicity), temafloxacin (hemolytic reactions), grepafloxacin (cardiotoxicity), and gatifloxacin (glucose
389
Recent Advances in the Inhibition of Bacterial Type II Topoisomerases
homeostasis abnormalities) [15,42]. Other side effects associated with fluoroquinolones have included phototoxicity, central nervous system (CNS) effects, and tendonitis. To date, the fluoroquinolone field is challenged with providing drugs that are predictably safe, maintain or further expand pathogen spectrum, and show therapeutically useful potency against current quinolone-resistant strains. Delafloxacin (29) and nemonoxacin (30), two of several quinolones now in clinical development, seem to fit this profile. Delafloxacin is at least 32-fold more potent than comparators such as levofloxacin and moxifloxacin against many quinolone-resistant Gram-positive pathogens including methicillinresistant s. aureus (MRSA) while maintaining good Gram-negative potency. [43]. Likewise, nemonoxacin, which lacks the traditional 6-fluoro substituent, exceeds comparators (ciprofloxacin, levofloxacin, moxifloxacin) with regard to in vitro potency against quinolone-resistant MRSA and retains a good Gram-negative spectrum [44]. Additionally, it appears to be less prone to bacterial resistance development, as assessed in vitro using clinical isolates of S. pneumoniae, compared to standard quinolone comparators [45]. Both 29 and 30 have recently completed phase II clinical trials for community-acquired pneumonia. O
O
O F
F
CO2H
H N
N
N
H3C
N Cl
OCH3
N
N OCH3
F N
HO
NH H
CO2H
CO2H
NH2
H2N F
28
29
30
O F O N
O
O CO2H
F
F
CO2H
CO2H
HN N
N
N
N
N
N O
OCH3
F
F
NH2
31
32
N
F
NH2
33
Analogs 31–33 are representative of recent preclinical efforts. Compound 31 was evaluated in vitro against clinical strains of S. aureus having defined mutations in the quinolone resistance-determining regions (QRDRs) of its GyrA and ParC [46]. Double and triple mutant
390
Gregory S. Bisacchi and Jacques Dumas
strains are resistant to ciprofloxacin and other standard quinolone comparators, but still susceptible to 31 (MICs ¼ 0.25–0.5 mg/mL). Analog 32 has shown potent in vitro activity against quinolone-susceptible and quinolone-resistant S. aureus and S. pneumoniae clinical isolates, showing superiority in all cases to moxifloxacin [47]. This compound also shows superiority to moxifloxacin in an in vivo model of quinolone-resistant S. aureus infection. Analog 33, similar to 31 and 32, displays a moxifloxacin-like spectrum of in vitro antibacterial activity, but with enhanced Gram-positive potency. In particular, 33 shows an MIC90 of 0.5 mg/mL versus MRSA, which is eightfold more potent than moxifloxacin [48]. Its MIC90 versus quinolone-resistant S. pneumoniae (0.25 mg/mL) is 32-fold more potent than moxifloxacin [49]. Considering the relative diversity of substitution patterns at N1, C6, C7, and C8 within this set of analogs (29–33), it is remarkable that all achieve a rather similar antibacterial profile, that is, enhanced Gram-positive spectrum, with distinct advantages against quinolone-resistant strains. Although the medicinal chemistry of quinolone antibacterials has over the past 50 years generated vast SAR guidance with which to frame current and future analog work, it still apparently has room to surprise. The recent clinical need for more effective Gram-positive agents, especially those with excellent potency against MRSA, has undoubtedly led to a focussed interest in, and bias toward, recent agents such as 29–33, having those qualities. However, one major challenge for future quinolone research remains the discovery of analogs with enhanced potency against multidrug-resistant Gram-negative pathogens, especially Klebsiella, Acinetobacter, and Pseudomonas aeruginosa species [50].
3.2 Quinolone-like structures Two separate medicinal chemistry efforts have developed scaffold series containing viable replacements for the otherwise canonic quinolone 3-carboxy group. The 3-aminoquinazolinediones, represented by lead structure 34, evolved from an early tricyclic hit series [51]. Based on analysis of resistant mutants in S. pneumoniae, 34 and related analogs apparently target primarily GyrB and ParE, unlike quinolones [52]. This series is not cross-resistant to established fluoroquinolone-resistant pathogenic strains. Sequencing of resistant mutants grown in Neisseria gonorrhoeae identified mutations in or near the QRDR suggesting that the binding site is near or overlapping that of the fluoroquinolones. Against a levofloxacin-resistant strain of S. pneumoniae, 34 exhibits an MIC90 of 0.06 mg/mL, with fluoroquinolone comparators in the 2- to 16-mg/mL range. In general, activity against Gram-positive pathogens is the strength of this series, with potencies against Gram-negatives either equal to or less than current fluoroquinolone comparators.
391
Recent Advances in the Inhibition of Bacterial Type II Topoisomerases
The isothiazolo-quinolone series originated in the 1980s as part of a quinolone carboxylic acid replacement program. Subsequently, the compounds synthesized at that time were found to carry significant eukaryotic topoisomerase inhibitory liability [53,54] and further interest in the series waned. O
O
F N
NH2
O
F
H
NH
H N
N
H2N
O
N
N
H2N
34
S
OCH3 35
Recently, however, the series was revived and progress has been made to enhance inhibition of prokaryotic versus eukaryotic topoisomerase Type II [55]. Compound 35, a representative of the series, shows MIC90s of 0.5, 0.5, and 2 mg/mL against MRSA, E. faecalis, and Enterococcus faecium, respectively [56]. In contrast, levofloxacin has MIC90s of W 16 mg/mL against all of these Gram-positive pathogens. Furthermore, Gram-negative activity is on a par with levofloxacin. Mutation analysis indicated that 35 targets both GyrA and ParC, with first-step mutations occurring in GyrA only [57].
3.3 Quinolines Nearly 150 years following Pasteur’s chemical fragmentation of quinine [58,59] to provide d-quinotoxine (36), certain synthetic derivatives of that scaffold were found to possess broad-spectrum antibacterial activity [60], which provided a fresh starting point for a potential new class of broadspectrum antibacterial agents. Analogs 37–39 are variations within this class, which acts by inhibition of bacterial gyrase and topoisomerase IV. Compound 37 retains activity in multiple fluoroquinolone-resistant strains of S. aureus having a variety of mutations in the QRDR, suggesting that it targets topoisomerases in a manner distinct to that of fluoroquinolones [61,62]. Quinoline 37 is efficacious against a strain of fluoroquinolone-resistant S. pneumoniae in a mouse lung infection model. When dosed orally at 100 mg/kg b.i.d., 37 resulted in 70% survival, compared to 10% for moxifloxacin at the same dose [63]. This compound advanced to phase I clinical trials but was discontinued due to QTc (the electrocardiogram QT interval corrected for heart rate) prolongation [64].
392
Gregory S. Bisacchi and Jacques Dumas
Compound 38 and related analogs were shown to inhibit DNA gyrase and topoisomerase IV in S. aureus, S. pneumoniae, and E. coli and displays single-digit or sub-micromolar MICs against all three pathogens [65]. This series also retains excellent potency against a panel of fluoroquinolone-resistant S. aureus species. For example, 39 shows MICs ranging from 0.007 mg/mL to 0.5 mg/mL, compared to 4 to W128 mg/mL for three fluoroquinolone comparators. Yet another sub-series in this class, exemplified by 39, was reported to have potencies of o2 mg/mL against a panel of Gram-positive and Gram-negative pathogens [66]. Other research groups have also reported their own scaffold entrants to this general class [67–69]. HO2C NH
O MeO
MeO N
S
N
HO
S
F
36
37 N
OMe
N N
O H N
N
38
O
N H
N
H N
S MeO
N
F
N
O
O
39
N
3.4 Quinoline pyrimidine triones A new class of antibiotics, exemplified by PNU-286607 (40), has recently been reported [70,71]. PNU-286607 was identified by whole cell screening for MICs, followed by a reverse-genomics approach to elucidate its mode of action, and features an unusual quinoline pyrimidine trione scaffold. Its structure was fully elucidated by X-ray crystallography [70]. Quinoline pyrimidine trione 40 has significant antimicrobial activity against a wide range of Gram-positive and to a lesser extent against Gram-negative pathogens. This compound acts through inhibition of the
Recent Advances in the Inhibition of Bacterial Type II Topoisomerases
393
DNA gyrase complex [71], does not show cross-resistance with fluoroquinolones, and is orally active in lethal systemic infection model in mice, using an MRSA strain (ED50 ¼ 19.5 mg/kg). Overall, 40 represents a promising lead structure, targeting a new binding site of the GyrA/GyrB complex.
O
H N
O
O2N
NH H
O CH3
N O 40 CH3
4. CONCLUSION Over 50 years have passed since the first discovery of DNA gyrase inhibitors. What started as a classical, MIC-driven drug discovery approach (novobiocin and first-generation quinolones) has now become a multidisciplinary field expanding at the frontiers of medicinal chemistry, microbiology, genomics, and structural biology. The ‘‘classical’’ quinolone family still produces extremely potent, broad-spectrum analogs, and year after year proves to be extremely flexible with regard to structural modifications. In addition, in the past decade, a number of new lead structures targeting DNA gyrase have been reported in the literature. Many of these already show useful MICs against Grampositive pathogens and a complete lack of cross-resistance against the widely used fluoroquinolones. Based on this progress, it appears probable that DNA gyrase inhibition will continue to produce novel, life-saving antibiotics for many years to come.
REFERENCES [1] R. W. Fairbrother and B. L. Williams, Lancet, 1956, 1177. [2] N. Barton, A. F. Crowther, W. Hepworth, D. N. Richardson and G. W. Driver, BR Patent Application, 830,832, 1960. [3] P. B. Fernandes, R. Menzel, D. J. Hardy, Y.-C. Tse-Ding, A. Warren and D. A. Elsemore, Med. Res. Rev., 1999, 19, 559.
394
Gregory S. Bisacchi and Jacques Dumas
[4] M. Gellert, M. H. O’Dea, T. Itoh and J.-I. Tomizawa, Proc. Natl. Acad. Sci. U.S.A., 1976, 73, 4474. [5] A. Sugino, C. L. Peebles, K. N. Kreuzer and N. R. Cozzarelli, Proc. Natl. Acad. Sci. U.S.A., 1977, 74, 4767. [6] J. Watanabe, N. Nakada, S. Sawairi, H. Shimada, S. Oshima, T. Kamiyama and M. Arisawa, J. Antibiot., 1994, 47, 32. [7] B. J. Bradbury and M. J. Pucci, Curr. Opin. Pharmacol., 2008, 8, 574. [8] M. Oblak, M. Kotnik and T. Solmajer, Curr. Med. Chem., 2007, 14, 2033. [9] Y. C. Tse-Dinh, Infect. Disord. Drug Targets, 2007, 7, 3. [10] A. Maxwell and D. M. Lawson, Curr. Top. Med. Chem., 2003, 3, 283. [11] J. C. Wang, Annu. Rev. Biochem., 1996, 65, 635. [12] B. Cheng, I.-F. Liu and Y.-C. Tse-Dinh, J. Antimicrob. Chemother., 2007, 59, 640. [13] C. Levine, H. Hiasa and K. J. Marians, Biochim. Biophys. Acta., 1998, 1400, 29. [14] S. Bellon, J. D. Parsons, Y. Wei, K. Hayakawa, L. L. Swenson, P. S. Charifson, J. A. Lippke, R. Aldape and C. H. Gross, J. Antimicrob. Chemother., 2004, 48, 1856. [15] I. Laponogov, M. K. Sohi, D. A. Veselkov, X. S. Pan, R. Sawhney, A. W. Thompson, K. McAuley, L. M. Fisher and M. R. Sanderson, RCSB Protein Databank, 3fof. [16] L. Heide, B. Gust, C. Anderle and S.-M. Li, Curr. Top. Med. Chem., 2008, 8, 667. [17] C. Anderle, S.-M. Li, B. Kammerer, B. Gust and L. Heide, J. Antibiot., 2007, 60, 504. [18] C. Anderle, M. Stieger, M. Burrell, S. Reinelt, A. Maxwell, M. Page and L. Heide, Antimicrob. Agents Chemother., 2008, 52, 1982. [19] C. Anderle, S. Hennig, B. Kammerer, S.-M. Li, L. Wessjohann, B. Gust and L. Heide, Chem. Biol., 2007, 14, 955. [20] N. Nakada, H. Shimada, T. Hirata, Y. Aoki, T. Kamiyama, J. Watanabe and M. Arisawa, Antimicrob. Agents Chemother., 1993, 37, 2656. [21] P. Angehrn, S. Buchmann, C. Funk, E. Goetschi, H. Gmuender, P. Hebeisen, D. Kostrewa, H. Link, T. Luebbers, R. Masciadri, J. Nielsen, P. Reindl, F. Ricklin, A. Schmitt-Hoffmann and F.-P. Theil, J. Med. Chem., 2004, 47, 1487. [22] T. Luebbers, P. Angehrn, H. Gmuender and S. Herzig, Bioorg. Med. Chem. Lett., 2007, 17, 4708. [23] P. S. Charifson, A.-L. Grillot, T. H. Grossman, J. D. Parsons, M. Badia, S. Bellon, D. D. Deininger, J. E. Drumm, C. H. Gross, A. LeTiran, Y. Liao, N. Mani, D. P. Nicolau, E. Perola, S. Ronkin, D. Shannon, L. L. Swenson, Q. Tang, P. R. Tessier, S.-K. Tian, M. Trudeau, T. Wang, Y. Wei, H. Zhang and D. Stamos, J. Med. Chem., 2008, 51, 5243. [24] T. H. Grossman, D. J. Bartels, S. Mullin, C. H. Gross, J. D. Parsons, Y. Liao, A.-L. Grillot, D. Stamos, E. R. Olson, P. S. Charifson and N. Mani, Antimicrob. Agents Chemother., 2007, 51, 657. [25] N. Mani, C. H. Gross, J. D. Parsons, B. Hanzelka, U. Mueh, S. Mullin, Y. Liao, A.-L. Grillot, D. Stamos, P. S. Charifson and T. H. Grossman, Antimicrob. Agents Chemother., 2006, 50, 1228. [26] S. P. East, C. B. White, O. Barker, J. Bennett, D. Brown, E. A. Boyd, C. Brennan, C. Chowdhury, I. Collins, E. Convers-Reignier, B. W. Dymock, R. Fletcher, D. J. Haydon, M. Gardiner, S. Hatcher, P. Ingram, P. Lancett, P. Mortenson, K. Papadopoulos, C. Smee, H. B. Thomaides-Brears, H. Tye, J. Workman and L. Czaplewski, Bioorg. Med. Chem. Lett., 2009, 19, 894. [27] D. C. D. Butler, H. Chen, V. R. Hedge, C. Limberakis, R. M. Rasne, R. J. Sciotti and J. T. Starr, WO Patent Application 2006/038116 A2, 2006. [28] R. J. Sciotti, J. T. Starr, C. Richardson, G. W. Rewcastle, B. D. Palmer, H. S. Sutherland, J. A. Spicer and H. Chen, WO Patent Application 2005/089763 A1, 2005. [29] D. R. Haydon, L. G. Czaplewski, N. J. Palmer, D. R. Mitchell, J. F. Atherall, C. R. Steele and T. Ladduwahetty, WO Patent Application 2007/148093 A1, 2007. [30] M. H. Block and W. W. Nichols, Mol. Med. Microbiol., 2002, 1, 609.
Recent Advances in the Inhibition of Bacterial Type II Topoisomerases
395
[31] H. J. Boehm, M. Boehringer, D. Bur, H. Gmuender, W. Huber, W. Klaus, D. Kostrewa, H. Kuehne, T. Luebbers, N. Meunier-Keller and F. Mueller, J. Med. Chem., 2000, 43, 2664. [32] T. Luebbers, P. Angehrn, H. Gmuender, S. Herzig and J. Kulhanek, Bioorg. Med. Chem. Lett., 2000, 10, 821. [33] O. Green, H. Ni, A. Singh, G. Walkup, D. Timms, N. Hales, A. Breeze and A. E. Eakin, F1-2025, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [34] B. Sherer, G. Basarab, K. Hull, S. Hauck, S. Bist and A. Eakin, F1-2026, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [35] K. Hull, O. Green, A. Singh, S. Bist, J. Demeritt, J. Loch, G. Mullen, S. Hauck, B. Sherer, H. Ni and A. Eakin, F1-2027, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [36] R. N. Illingworth, M. Uria-Nickelsen, J. Bryant and A. Eakin, F1-2028, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [37] M. R. Uria-Nickelsen, A. Blodgett and A. Eakin, F1-2029, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [38] M. Oblak, S. Golic Grdadolnik, M. Kotnik, A. Poterszman, R. A. Atkinson, H. Nierengarten, D. Desplancq, D. Moras and T. Solmajer, Biochem. Biophys. Res. Commun., 2006, 349, 1206. [39] A. Plaper, M. Golob, I. Hafner, M. Oblak, T. Solmajer and R. Jerala, Biochem. Biophys. Res. Commun., 2003, 306, 530. [40] H. Gradisar, P. Prisovsek, A. Plaper and R. Jerala, J. Med. Chem., 2007, 50, 264. [41] A. S. Wagman and M. P. Wentland, in Comprehensive Medicinal Chemistry II (eds J. B. Taylor and D. J. Triggle), Elsevier Ltd, Oxford, UK, 2006, p. 567. [42] A. J. Mehlholm and D. A. Brown, Ann. Pharmacother., 2007, 41, 1859. [43] L. S. Almer, J. B. Hoffrage, E. L. Keller, R. K. Flamm and V. D. Shortridge, Antimicrob. Agents Chemother., 2004, 48, 2771. [44] G. A. Pankuch, K. Kosowska-Shick, C. R. King and P. C. Appelbaum, C1-189, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [45] C. R. King, L. Lin and R. Leunk, C1-1971, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [46] H.-J. Yun, Y.-H. Min, Y. W. Jo, M.-J. Shim and E.-C. Choi, Int. J. Antimicrob. Agents, 2005, 25, 334. [47] Y. Asahina, K. Kobayasi, M. Takadoi, A. Jojima, K. Ohata, S. Katayama, T. Komine, A. Nakamura, K. Yokota, O. Nagae, T. Sato, T. Shibue, E. Nagata, H. Takano and Y. Fukuda, F1-2121, 47th Annual ICAAC Meeting, Chicago, IL, September, 2007. [48] W. He, K. Amsler, K. Bush and B. J. Morrow, F1-2035, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [49] B. J. Morrow, B. Foleno, R. Goldschmidt, W. He, K. Amsler, M. Maceielag and K. Bush, F1-2052, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [50] H. W. Boucher, G. H. Talbot, J. S. Bradley, J. E. Edwards, D. Gilbert, L. B. Rice, M. Scheld, B. Spellberg and J. Bartlett, Clin. Infect. Dis., 2009, 48, 1. [51] T. P. Tran, E. L. Ellsworth, M. A. Stier, J. M. Domagala, H. D. Hollis Showalter, S. J. Gracheck, M. A. Shapiro, T. E. Joannides and R Singh, Bioorg. Med. Chem. Lett., 2004, 14, 4405. [52] M. D. Huband, M. A. Cohen, M. Zurack, D. L. Hanna, L. A. Skerlos, M. C. Sulavic, G. W. Gibson, J. W. Gage, E. Ellsworth, M. A. Stier and S. J. Gracheck, Antimicrob. Agents Chemother., 2007, 51, 1191. [53] D. T. W. Chu, P. B. Fernandes, A. K. Claiborne, L. Shen and A. G. Pernet, Drugs Exp. Clin. Res., 1988, 14, 379. [54] W. E. Kohnbrenner, N. Wideburg, D. Weigl, A. Saldivar and D. T. W. Chu, Antimicrob. Agents Chemother., 1992, 36, 81.
396
Gregory S. Bisacchi and Jacques Dumas
[55] Q. Wang, E. Lucien, A. Hashimoto, G. C. G. Pais, D. M. Nelson, Y. Song, J. A. Thanassi, C. W. Marlor, C. L. Thoma, J. Cheng, S. D. Podos, Y. Ou, M. Deshpande, M. J. Pucci, D. D. Buechter, B. J. Bradbury and J. A. Wiles, J. Med. Chem., 2007, 50, 199. [56] M. J. Pucci, C. L. Thoma, S. D. Podos, J. Cheng, J. A. Thanassi, B. J. Bradbury and M. Despande, F1-2021, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [57] J. A. Thanassi, J. Cheng, S. D. Podos, B. J. Bradbury, M. Deshpande and M. J. Pucci, F1-2022, 48th Annual ICAAC Meeting, Washington, DC, October, 2008. [58] L. Pasteur and C. R. Hebd, Seances Acad. Sci., 1853, 37, 162. [59] J. I. Seeman, Angew. Chem. Int. Ed. Engl., 2007, 46, 1378. [60] W. J. Coates, M. N. Gwynn, I. K. Hatton, P. J. Masters, N. D. Pearson, S. S. Rahman, B. Slocombe and J. D. Warrack, WO Patent Application 99/37635 A1, 1999. [61] I. Morrissey, J. Lowther, M. Soltani and J. Northwood, O 457, 17th European Congress of Clinical Microbiology and Infectious Disease, Munich, 2007. [62] M. T. Black, T. Stachyra, D. Platel, A.-M. Girard, M. Claudon, J.-M. Bruneau and C. Miossec, Antimicrob. Agents Chemother., 2008, 52, 3339. [63] P. Levasseur, A. M. Girard and J. Lowther, F1-2125, 47th Annual ICAAC Meeting, Chicago, IL, September, 2007. [64] Press release, June 30, 2008; http://www.novexel.com/ [65] J. J. M. Wiener, L. Gomez, H. Venkatesan, A. Santillan, B. D. Allison, K. L. Schwarz, S. Shinde, L. Tang, M. D. Hack, B. J. Morrow, S. T. Motley, R. M. Goldschmidt, K. J. Shaw, T. K. Jones and C. A Grice, Bioorg. Med. Chem. Lett., 2007, 17, 2718. [66] J. M. Axten, G. Brooks, P. Brown, D. Daview, T. F. Gallager, R. E. Markwell, W. H. Miller, N. D. Pearson and M. Seefeld, WO Patent Application 2004/058144 A1, 2004. [67] T. Kiyoto, J. Ando, T. Tanaka, Y. Tsutsui, M. Yokotani, T. Noguchi, F. Ushiyama, H. Urabe and H. Horikiri, WO Patent Application 2007/138974, 2007. [68] C. Hubschwerlen, G. Rueedi, J.-P. Surivet and C. Zumbrunn Acklin, WO Patent Application 2008/152603, 2008. [69] M. Cronin, B. Geng and F. Reck, WO Patent Application 2009/001126, 2009. [70] J. C. Ruble, A. R. Hurd, T. A. Johnson, D. A. Sherry, M. R. Barbachyn, P. L. Toogood, G. L. Bundy, D. R. Graber and G. M. Kamilar, J. Am. Chem. Soc., 2009, 131, 3991. [71] A. A. Miller, G. L. Bundy, J. E. Mott, J. E. Skepner, T. P. Boyle, A. E. Hromockyj, K. R. Marotti, G. E. Zurenko, J. B. Munzner, M. T. Sweeney, G. F. Bammert, J. C. Hamel, C. W. Ford, W.-Z. Zhong, D. R. Graber, G. E. Martin, F. Han, L. A. Dolak, E. P. Seest, J. C. Ruble, G. M. Kamilar, J. R. Palmer, L. S. Banitt, A. R. Hurd and M. R. Barbachyn, Antimicrob. Agents Chemother., 2008, 52, 2806.
CHAPT ER
20 Progress towards the Discovery and Development of Specifically Targeted Inhibitors of Hepatitis C Virus Nicholas A. Meanwell, John F. Kadow and Paul M. Scola
Contents
1. Introduction 2. Specifically Targeted Inhibitors of HCV 2.1 HCV NS2 protease inhibitors 2.2 HCV NS3/4A protease inhibitors 2.3 HCV NS3 helicase inhibitors 2.4 HCV NS4B replication factor inhibitors 2.5 HCV NS5A replication factor inhibitors 2.6 HCV NS5B polymerase inhibitors 2.7 Emerging mechanisms 3. Conclusions References
397 398 398 400 407 408 409 412 420 427 428
1. INTRODUCTION The incidence of hepatitis C virus (HCV) infection in the United States (U.S.) is estimated to be approaching 5 million, against a backdrop of 150–200 million worldwide, with the majority of individuals unaware of infections that are not sufficiently advanced to cause overt liver disease [1]. Department of Chemistry, Bristol-Myers Squibb Pharmaceutical Research and Development, 5 Research Parkway, Wallingford, CT 06492, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04420-0
r 2009 Elsevier Inc. All rights reserved.
397
398
Nicholas A. Meanwell et al.
The current standard of care (SOC) comprises weekly subcutaneous injections of pegylated interferon-a2a (PEG-IFN-a2a) (Pegasyss) or -a2b (Pegintrons) in combination with twice daily doses of ribavirin, a regimen poorly tolerated due to the association with a significant incidence of side effects. Moreover, this regimen is ineffective in achieving a sustained virologic response (SVR), defined as the absence of viremia 6 months after completion of therapy, in the majority of patients infected with genotype 1 HCV, which is most prevalent in the U.S. [2,3]. Several IFN derivatives designed to improve tolerability are in clinical development [4], including albinterferon-a2b (Albuferons), a protein that fuses IFN-a2b with human albumin to enhance the in vivo half-life and reduce dosing frequency [5], pegylated IFN-l (interleukin 29), which offers increased specificity towards the liver in vivo [6], and Locterons, a sustained release formulation of IFN-a2b [7]. However, the antiviral activity of IFNs is indirect, and ribavirin is a non-specific antiviral agent with inhibitory activity towards some host cell proteins, a circumstance that has contributed to the considerable effort being expended to identify and develop specifically targeted antiviral therapies for HCV (STAT-C) [8–15]. Despite disappointments with the failure of several early candidates in clinical trials [16,17], there is currently a robust and expanding pipeline of specific HCV inhibitors in clinical trials or late-stage preclinical testing, which target a range of viral proteins and RNA, summarized in Table 1 [8–15,18–20]. Of particular concern with STAT-C is the potential for resistance to emerge during or as a consequence of therapy based on the high rate and poor fidelity of replication of HCV [21–27], an anxiety that anticipates combination therapy with two or more antiviral agents [28,29]. However, whether a combination of STAT-C will be able to achieve a SVR in the absence of immune stimulation awaits the results of well-controlled clinical trials. The first clinical assessment of a combination of small molecules was initiated in late 2008 with a 14-day study designed to evaluate the safety and efficacy of an NS3 protease inhibitor and a polymerase inhibitor in treatment-naı¨ve patients infected with genotype 1 virus [30]. In this review, we provide a synopsis of progress made towards identifying and developing inhibitors of HCV during the period January 2007 to April 2009 with the most advanced compounds summarized in Table 1 and discussed in the individual sections.
2. SPECIFICALLY TARGETED INHIBITORS OF HCV 2.1 HCV NS2 protease inhibitors The NS2 protein is essential for the production of infectious virus, and a crystal structure of the protease catalytic domain has revealed it to be a
Table 1
The HCV inhibitor clinical landscape NS3/4A
Phase III Phase II
NS5A
Telaprevir Boceprevir BI-201335
R-7128 — nucleoside analogue active site PF-868554 — thumb domain site 2
ABT-450
BMS-790052
VBY-106
A-831/AZD-2836 A-689/AZD-7295
IDX-136 IDX-316
Albuferon albumin/IFN-a fusion protein Locteron extended release formulation of IFN-a 2b
NIM-811 cyclosporin analogue DEBIO-025 cyclosporin analogue SCY-635 cyclosporin analogue Nitazoxanide host cell target Celgosovir a-glucosidase inhibitor Silibinin VCH-222 — thumb domain site 2 PEG-IFN l (IL-29) liver-targeted IFN derivative MK-3281 — thumb domain site 1 IDX-184 — liver targeting nucleoside prodrug IDX-375 — palm domain site 1 ABT-333 — palm domain site 1 ABT-072 — palm domain site 1 GL-60667 (LDI-133) — thumb domain site 1
399
Preclinical
VCH-759 — thumb domain site 2 VCH-916 — thumb domain site 2 ANA-598 — palm domain site 2 GS-9190 — replication complex inhibitor
Miscellaneous
Discovery and Development of Inhibitors of HCV
ITMN-191/ R-7227 TMC-435350 MK-7009 SCH-900518
Phase I
NS5B
400
Nicholas A. Meanwell et al.
cysteine protease that functions as a dimer [31,32]. The two composite active sites interact cooperatively, with the catalytic histidine and glutamate residues from one domain working in concert with the catalytic cysteine from the second domain. Interestingly, although the protease domain of NS2 is essential for the in vitro production of infectious virus, its enzymatic activity is not required [31,33]. Additional insights into the biological functions of NS2 have been described [34–37], and assays designed to screen for inhibitors of the protease activity have also been disclosed [38,39], but no well-characterized inhibitors of this enzyme have been described to date.
2.2 HCV NS3/4A protease inhibitors Proof-of-concept for NS3 protease inhibitors as antiviral agents with potential for the treatment of HCV was provided by clinical studies conducted with BILN-2061 (1). This compound was administered to patients infected with genotype 1 virus using a dosing regimen of 200 mg BID for 2 consecutive days, which produced a greater than 2–3 log10 reduction in plasma HCV RNA in all patients treated [40]. Since this seminal study, a significant number of protease inhibitors have progressed into clinical trials and results from these investigations continue to shape and refine treatment strategies designed to optimize long-term efficacy (SVR) and minimize the emergence of resistant virus [8,41–45]. Telaprevir (VX-950, 2) is the most advanced protease inhibitor, having recently entered into phase III clinical trials. Telaprevir (2) is a potent and selective, mechanism-based inhibitor of the NS3/4A enzyme that forms a covalent, but reversible complex with the enzyme [46]. Binding studies conducted with 2 and a genotype 1a HCV protease indicate that dissociation of the enzyme–inhibitor complex occurs slowly, with a halflife of approximately 1 h. Telaprevir (2) exhibits low to moderate oral bioavailability in both rat (25%) and dog (41%), with liver exposure in the rat exceeding plasma levels by a factor of 35. Me S MeO
Me
N
N
N
H
H
O H H N
O
N
N O
O OH
N
O N
O
1
H H
O H N H
N
N O Me
O Me Me
2
O
H
N O
N O
Me
Discovery and Development of Inhibitors of HCV
401
In a seminal phase I clinical study, genotype 1 patients receiving a 750-mg TID dose of telaprevir (2) experienced a median 4.4 log10 reduction in HCV RNA after 14 days of treatment, with some patients achieving undetectable levels of HCV RNA [47]. Subsequent clinical studies have confirmed the antiviral activity of telaprevir (2) while also providing evidence for the rapid emergence of resistance during the course of monotherapy [48,49]. In a phase II study, genotype 1 patients were randomly divided into three treatment groups and dosed for 14 days with polyethyleneglycol (PEG)-IFN, telaprevir (750 mg, TID) or telaprevir+PEGIFNa before optionally evolving to a SOC regimen. The median change in HCV RNA observed in each of these dosing groups was 1.09 log10, 3.99 log10 and 5.49 log10, respectively. Importantly, viral breakthrough was observed in four of eight patients dosed with telaprevir (2) alone while, in contrast, all eight patients treated with telaprevir+PEG-IFNa experienced a continuous or sustained drop in viral load. Interestingly, samples of virus from patients who displayed viral rebound were shown to vary in their genetic constitution as a function of time. For example, on day 4, wild-type virus was dominant with the single mutants V36/A/M, R155K/T and A156V/T observed exclusively but at low levels (5–20%). The population of these variants increased in samples isolated on days 8 and 12, but by day 15, these single mutants were replaced with high-level resistant double mutants (V36/R155). A subset of samples from patients receiving both telaprevir (2) and PEG-IFNa showed the presence of a A156T variant that demonstrates high-level resistance in vitro but breakthrough was not observed. In separate studies, the double viral mutant V36/R155 produced on therapy was shown to be slightly more fit than single mutants such as R155K/T [50]. Although these early clinical studies demonstrate the acute antiviral activity of telaprevir (2) and its potential as a therapeutic agent in combination with IFN, they also reveal the limitation of the drug as monotherapy for the treatment of HCV due to the rapid emergence of resistant virus. These general observations have been upheld in subsequent clinical trials with a series of protease inhibitors (vide infra). In a phase II clinical study, the safety and efficacy of telaprevir (2) in combination with PEG-IFNa, with or without ribavirin, in chronic HCV patients was evaluated [51]. The results from this study demonstrate that the addition of telaprevir (2) to PEG-IFNa/ribavirin (RBV) may allow for a significant reduction in overall treatment time from 48 to 24 weeks. Specifically, treatment-naı¨ve genotype 1-infected patients upon administered telaprevir+PEG-IFN/RBV for 12 weeks followed by PEG-IFNa/RBV for an additional 12 weeks (24 weeks total therapy) collectively experienced a SVR rate of 69%, which compared to a 46% SVR in the 48 week SOC arm. In addition, ribavirin was shown to be a critical component of this dosing triad since patients receiving telaprevir+PEGIFNa but not ribavirin experienced significantly greater viral breakthrough
402
Nicholas A. Meanwell et al.
at week 12 when compared to patients receiving telparevir+PEG-IFNa/ RBV. The profile of telaprevir (2) will be confirmed and expanded upon in the ongoing PROVE 3 study [52]. Boceprevir (SCH-503034, 3) has also entered phase III clinical trials. Analogous to telaprevir (2), boceprevir (3) is a mechanism-based inhibitor of the NS3/4A protease that demonstrates a Ki* of 14 nM [53,54]. The selectivity of boceprevir (3) for HCV NS3/4A over human neutrophil elastase is thought to be derived, in part, from the P1 cyclobutylalanine moiety [53,54]. Boceprevir (3) demonstrates an EC90 of 350 nM in a replicon assay and exhibits an oral bioavailability of 34, 26, 30 and 4% in the mouse, rat, dog and monkey, respectively. In rat, liver levels of the compound exceeded that of plasma by a factor of 30. In clinical trials, boceprevir (3) has been studied as add-on therapy to PEG-IFNa/RBV [55,56]. In the SPRINT-1 trial, boceprevir (3) administered to genotype 1 patients at a dose of 800 mg TID in combination with SOC for 28 or 48 weeks produced SVRs of 55 and 66%, respectively, an improvement over the 38% SVR observed in the control group who received PEG-IFNa/RBV for 48 weeks. Two additional cohorts in this study were dosed with a leadin regimen of PEG-IFNa/RBV for 4 weeks, which was followed by treatment with boceprevir (800 mg TID)+PEG-IFNa/RBV for 24 or 44 weeks. The SVRs reported in these studies were 56 and 74%, respectively, which again provided a more favorable outcome than the 38% SVR observed with SOC. The lead-in dosing with SOC appears to reduce the incidence of viral breakthrough in both cohorts. Phase III studies are ongoing to further evaluate treatment options with boceprevir (3). A second-generation protease inhibitor, SCH-900518, that demonstrates improved potency [57] has advanced into clinical trials, and while the structure of this compound has not been released, more general efforts to further optimize this chemotype have been described [58–61]. Me MeO Me
Me Me
Me
N
N O Me
N
Me
O
O
N H
Me
Me H
H
S N
N O O
Me
H
NH2
N
O Me
O N
O
O O S N H
O
Me
3
4
TMC-435350 (4) is a macrocyclic tripeptide derivative that incorporates a functionalized cyclopentyl moiety as a replacement for the P2 proline found in most of the highly potent HCV NS3/4A inhibitors. In addition,
403
Discovery and Development of Inhibitors of HCV
this compound employs an acylsulfonamide as a carboxylic acid isostere that interacts with the catalytic elements and allows the cyclopropyl ring to project into the S1u pocket of the protease [62,63]. In preclinical studies, TMC-435350 (4) demonstrated potent inhibition of a genotype 1b replicon, EC50 ¼ 8 nM, an oral bioavailability of 11–44% and up to 100% in rat and dog, respectively, and a liver to plasma area under the curve (AUC) ratio of greater than 35 in the rat [62,63]. In a phase IIa clinical trial designated OPERA-1, patients infected with genotype 1 virus were dosed daily with either 25 or 75 mg of TMC-435350 (4) for 4 weeks in conjunction with PEG-IFNa/RBV and compared to an arm in which there was a 7-day lead in with TMC-435350 (4) as monotherapy [64]. At day 7, viral loads were reduced by 3.47 and 4.55 log10 in the 25 and 75 mg triple therapy arms, respectively, whereas monotherapy was less efficacious, producing a 2.63 and 3.43 log10 reduction in viremia at the 25 and 75 mg doses, respectively [64]. Further evaluation of TMC-435350 (4) in clinical trials is ongoing. ITMN-191 (R-7227, 5), a tripeptidic acylsulfonamide incorporating the P1 and P3 moieties into a 15-membered macrocyclic ring, is currently in phase II clinical trials. ITMN-191 (5) displays enzyme kinetics that appear to be unique for this structural class, indicative of a slow/tight binding mechanism with slow dissociation, which may suggest a complementary conformational change within the enzyme subsequent to the initial binding event [65,66]. ITMN-191 (5) exhibits excellent potency in a genotype 1b replicon, EC50 ¼ 2 nM, and the concentration of compound in the liver exceeds that of plasma after oral dosing, with an AUCobs ratio of B10 in the rat and B127 in the monkey. In clinical studies, genotype 1 HCV-infected patients were treated with ITMN-191 (5) BID or TID for a period of 14 days, with the most significant antiviral response observed in the 200 mg TID cohort where a 3.9 log10 mean reduction in viral load was observed [67]. Ongoing clinical studies are evaluating this compound as add-on to SOC, whereas a pioneering study initiated in Australia and New Zealand late in 2008 is evaluating the safety and efficacy of ITMN-191 (5) and the polymerase inhibitor R-7128 (vide infra) in treatment-naı¨ve patients infected with genotype 1 virus over 14 days, with promising initial results [30].
N
F
O
N
O
O H
Me Me
H N
O Me
O
N
N O
O
O
O
O O S N H
H Me Me
O
H N O Me
5
N
N O Me Me
6
O
O
O O S
N H
Me
404
Nicholas A. Meanwell et al.
MK-7009 (6), also in phase II clinical trials, incorporates a structurally unique P2-P4 macrocycle designed to improve potency compared to analogous acyclic compounds [68]. MK-7009 (6) is a rapidly reversible enzyme inhibitor that is potently active in a genotype 1b replicon, EC50 ¼ 4 nM. In phase 1 clinical trials conducted in genotype 1 HCVinfected patients, QD and BID dosing regimens of the compound were examined for a duration of 8 days, with the optimal antiviral response observed in the 700 mg BID cohort, which produced a 4.6 log10 mean reduction in HCV RNA [69,70]. Phase II studies are ongoing to determine the efficacy of MK-7009 in combination with SOC. Clinical data have been recently reported for BI-201335 (structure not disclosed) administered QD to genotype-1-infected patients for 14 days at doses of 20, 48, 120 and 240 mg [71]. In each cohort, the maximal viral response was observed between day 2 and day 4, with the greatest effect on mean viral load seen in the 240 mg cohort where HCV RNA declined by 4 log10. However, virologic breakthrough was observed during treatment in the majority of patients at all doses. In a separate study, BI-201335 was administered QD at doses of 48, 120 and 240 mg in combination with PEG-IFNa/RBV for 28 days to treatment-experienced genotype 1 patients, producing median changes in HCV RNA of 5.0 log10, 5.2 log10 and 5.3 log10, respectively [72]. Virologic breakthrough was observed in two of six patients in the 48 mg dose cohort and one of seven in the 120 mg dose group but no breakthroughs were observed in the 240 mg cohort during the 28 days of treatment. At higher doses, an increased incidence of unconjugated hyperbilirubinemia was noted. Several additional HCV NS3 protease inhibitors have entered early clinical development or are in late-stage preclinical studies, including ABT-450, VBY-106, IDX-136 and IDX-316, the structures of which have not been disclosed. HCV NS3 protease inhibition continues to be an area of considerable interest with a number of publications and patent applications describing potent compounds that advance structure–activity relationships. A survey of acylsulfonamides as isosteres of the terminal carboxylic acid moiety revealed that the significantly increased potency in both enzymatic and cell-based assays was dependent on the pKa and implicated a hydrogen bond between one of the sulfone oxygen atoms and the amide NH of Q41 and backbone NH of G137 of the enzyme [73]. Optimal activity was observed with compounds bearing a P1 cyclopropyl amino acid moiety, with a vinyl substituent imparting a significant increase in potency, as demonstrated by the comparison between 7 and 8.
405
Discovery and Development of Inhibitors of HCV MeO
N
Ph Me
Me
O H
O
N H Me
O
N O
N
O
Me Me
O Me
H O O S Ph N
N
H
O
H
N
O
H
H
N
N
R
N
N O
O
O
O
O Me
7, R = H, Ki = 0.055 µM 8, R = CH = CH2, Ki = 0.00076 µM
9
Optimization of ketoamide-based NS3/4A inhibitors has focused on the P4 capping element with the 4,4-dimethyl-substituted glutarimide derivative 9 significantly more potent than a simple tert-butyl urea cap [58,60]. This compound demonstrated an EC90 in the HCV replicon screen of 30 nM, which represents a 12-fold increase in cellular potency compared to boceprevir (3) and appears to be a function of an additional hydrogen bond between one of the imide carbonyl groups and C159 of NS3, observed in the X-ray structure of 9 bound to the enzyme. Further studies in this series have explored P1–P3 macrocycles [58,60]. Tripeptide derivatives that present a boronic acid as the electrophilic trap for the catalytic serine of NS3 have been described, with 10 demonstrating excellent intrinsic inhibitory activity towards HCV NS3, Ki ¼ 200 pM [61]. An X-ray structure of 10 bound to the enzyme revealed a Lewis acid complex engaging the boron atom and the oxygen of S139 of the enzyme. However, the cellular activity of this compound is limited, with an EC90 of W5.0 mM, indicative of poor membrane penetration [61]. Additional patent applications claim acyclic compounds with P2 moieties based on 4-hydroxy proline [74,75]. Me
O O S Me N Me
H
H
N
Me
N
N O
H
OH
N
B
OH
O O
Me Me Me
10
Several novel P2 elements have been elucidated in acyclic tripeptide inhibitors, including the biphenylated proline incorporated in 11, with
406
Nicholas A. Meanwell et al.
optimal compounds claimed to demonstrate enzyme inhibition with EC50 of o50 nM [76,77]. Additional P3 caps have also been described in which a phenyl ring or a heterocycle replaces the more common amide or carbamoyl moiety, with compounds similar to 12 exhibiting NS3 inhibitory activity at concentrations o100 nM [78]. OMe Ph N
Cl OMe H H Me Me
O Me
O Me
N
N
N
O
O
O
O O O S N H
H H MeO
N
N
Me Me
O
Me
O
N
O O S
N
O
H
Me Me
11
12
Efforts to replace the P2 proline group in the tripeptide series have been attempted in the context of the macrocycles 13 and 14, with the former exhibiting a Ki of 76 nM [79,80]. Although these compounds appear to be less potent than the analogous proline derivatives, they may serve as useful starting points for further efforts in this area. MeO
N
Ph
Me
S
Me MeO
N
N
Me
O O H O
Me Me
O
H
N
N N
Me
O
H
13
O
O
O O S N H
O Me N
H N
N H
O
O
O O Me S
N H
14
Interesting macrocyclic inhibitors that explore a tether between the phenyl moiety of an acylsulfonamide P1u and either P1, as exemplified by 15, or P3 have been described although the success of this strategy is not immediately apparent [81,82].
407
Discovery and Development of Inhibitors of HCV
Me
H N S MeO
Me N
N
O H
O
N H Me
N O
N
O O S N H
O O Me
HN
Me 15
ACH-806 (16) has been described as a novel inhibitor of NS3/4A activity with resistance mapping to NS4A [83]. This compound is a potent inhibitor in replicons, fully retaining activity towards replicons resistant to active site inhibitors of NS3. In HCV genotype-1-infected patients, a dose of 300 mg BID of ACH-806 (16) for 5 days produced a 0.9 log10 reduction in HCV RNA, providing clinical proof-of-concept for an NS4A inhibitor [84]. However, development of ACH-806 (16) was terminated due to an elevation of serum creatinine levels, a marker of kidney toxicity [84]. More recently, a second compound, ACH-1095 (undisclosed structure), has entered clinical trials. O O
O
S N H
N H
Me
Ph
O
OH
CF3
N
Me F
N
16
17
2.3 HCV NS3 helicase inhibitors It is well established that the helicase function of NS3 is required for in vivo replication of infectious virus, which is dependent, in part, on the RNA unwinding activity of the enzyme [85]. Although the precise role of this enzyme in the HCV life cycle is not fully understood, the extent of
408
Nicholas A. Meanwell et al.
helicase binding and hydrolysis of adenosine triphosphate (ATP) have been linked to RNA translocation while unwinding studies are suggestive of an inchworm-type mechanism [86]. The NS3 protease and NS3 helicase activities have been shown to be interdependent, with the helicase domain enhancing serine protease activity and the protease domain enhancing helicase activity [87]. Although these findings reflect significant gains in the understanding of NS3 helicase function and mechanism, there are only limited reports of small molecule inhibitors of this essential enzyme despite the availability of an X-ray crystallographic structure [88]. A recent patent application claims a series of indole-based inhibitors, represented by 17, as inhibitors of NS3 helicase function with an IC50 of less than 10 mM [89].
2.4 HCV NS4B replication factor inhibitors HCV NS4B is a protein integral to the endoplasmic reticulum membrane that is predicted to be composed of four transmembrane domains with a topology that projects the N- and C-termini into the cytoplasm of the host cell [18]. Two C-terminal cysteine residues have been shown to be palmitoylated and the protein appears to be capable of forming oligomers. HCV NS4B is critical for viral replication and is postulated to form a web in the membrane that functions as a scaffold for the assembly of the virus replication complex, which includes NS5B and NS3 [90,91]. The HCV NS4B protein has been shown to hydrolyze both ATP and guanosine triphosphate (GTP), with the former the better substrate based on a 25-fold kinetic advantage, and express adenylate kinase activity, properties that map to a Walker nucleotide-binding domain [91]. This domain is required for viral replication and also appears to mediate cellular transformation, providing the basis for a cell-based assay to detect inhibitors of NS4B [92,93]. Small-molecule inhibitors of HCV NS4B are just beginning to emerge. A hypothesis that NS4B binds RNA was substantiated with the development of a binding assay based on microfluidic technology, which revealed a specific and high-affinity interaction with the 3u-terminus of the viral negative RNA strand, Kd ¼ 3.4 nM [94]. The binding sites on the protein mapped to an arginine-rich motif in the last 71 amino acids of the cytoplasmic C terminus. A high-throughput screen identified the histamine H1 antagonist clemizole (18) as a potent inhibitor of NS4BRNA binding, IC50 ¼ 24 nM, that demonstrated activity in a replicon, EC50 ¼ 8 mM. Resistance was mapped to W55R located in the cyctoplasmic amino terminus and R214Q in the C-terminal cytoplasmic domain. Mutant replicons were less fit than wild-type and the individually mutated 4B proteins demonstrated higher affinity for viral RNA, Kd ¼ 0.75 nM for W55R and Kd ¼ 0.6 nM for R214Q. The introduction of methyl substituents at C-5 and C-6 of the benzimidazole leads to increased inhibitory potency in the replicon, EC50 ¼ 1 mM [95,96].
409
Discovery and Development of Inhibitors of HCV N N
H3C N
N N N
Cl CH3
F
N
N N
CH3
N
18
19
O S HN
N CH3
Cl
F
N
O
20
A U.S. patent application discloses a range of structurally diverse compounds claimed to be inhibitors of NS4B, which were discovered by a screening campaign that exploited a binding assay monitoring changes in the intrinsic fluorescence of NS4B [97]. The triazinoindole 19 was active in a 1b replicon assay with an EC50 ¼ 1.13 mM. Resistance mapped to G120V in the second transmembrane domain of NS4B and K52R and A210S located in the N- and C-terminus cytoplasmic domains of the protein. The activity of the pyrazolopyridine anguizole (20), also disclosed in this patent application, has been confirmed [98]. This compound inhibits both 1a and 1b replicons with EC50 of 560 and 310 nM, respectively, with resistance mapped to H94A of NS4B, a residue immediately preceding the first transmembrane domain of the protein.
2.5 HCV NS5A replication factor inhibitors HCV NS5A is a 447 residue phosphoprotein that plays a critically important role in virus replication, assembly and egress, but the function of this protein is extraordinarily complex as a consequence of extensive interactions with both viral and host cell proteins [20,99–101]. HCV NS5A is composed of three domains: domain I, composed of residues 1–213, which incorporates an amphipathic N-terminus helix thought to be involved in membrane association [102], a zinc-binding motif and binding regions for both HCV NS4A and NS5B; domain II that encompasses residues 250–342 and contains the putative IFN sensitivitydetermining region (ISDR) [103] in addition to sites of interaction with HCV NS5B and several host cell factors; domain III, residues 356–447, which has recently been associated with an important role in virus assembly and is regulated by phosphorylation [104–106]. Although HCV NS5A does not express enzymatic function, targeted efforts have focused on disrupting activity associated with the N-terminal helix [102,107] while screening campaigns using HCV replicons have surfaced several classes of inhibitor for which resistance has been mapped to domain I [99]. SWLRDIWDWICEVLSDFK, an 18 mer peptide derived from residues 3–20 of the amphipathic a-helical N-terminus membrane anchor domain of a genotype 1a virus and designated C5A, inhibits replication of JFH1
410
Nicholas A. Meanwell et al.
infectious virus in cell culture with an EC50 of 0.79 mM [107]. The D-amino acid analogue performs similarly but scrambled peptides are inactive. A series of biochemical studies led to the suggestion that these peptides bind to viral membranes and compromise integrity, leading to exposure of the viral genome to exonucleases [107]. Consistent with this hypothesis, C5A also inhibits human immunodeficiency, dengue, respiratory syncytial, West Nile and measles viruses with similar potency [107,108]. Small-molecule inhibitors of HCV NS5A that encompass several structural classes have been disclosed with three compounds, A-831 (AZD-2836), A-689 (AZD-7295) and BMS-790052 (structures not disclosed), advanced into clinical studies [109,110]. However, A-831 was recently withdrawn from clinical evaluation pending reformulation, whereas data on A-689 have not been released. In genotype 1 HCV-infected patients, single 1, 10 and 100 mg doses of BMS-790052 produced impressive 1.8 log10, 3.2 log10 and 3.6 log10 mean reductions in viral load, respectively, with efficacy at the 100 mg dose persisting for 144 h [110]. BMS-790052 is a potent inhibitor of genotype 1a and 1b RNA replication in replicons, EC50 ¼ 50 and 6 pM, respectively, and infectious genotype 2a virus, EC50 ¼ 12 pM [20,110]. Several recent patent applications claim a series of biphenyl-based inhibitors of HCV NS5A, with 21 representative [111–114], whereas an alternative patent estate focuses on two general themes that have been explored broadly, exemplified by the quinazoline 22 and amide 23 [115–120]. NH2 N CH3
N
H3C H3CO2CHN
NH
N
O
O
HN
H N
N
O
CH3 N
N
CH3
21
N F H3C
O
O
N
HN N
H3C
O
N N
N CH3
N
22
CH3
N
N H
NH
O
N
N
O
23
CH3
S O
411
Discovery and Development of Inhibitors of HCV
A series of NS5A inhibitors based on proline has been extensively explored with GL-101267 (24) representative of the basic chemotype, a molecule that demonstrates an EC50 of 80 nM in the replicon assay [99,121–124]. However, the most recent patent application discloses a series of amides, represented by 25, that bear some structural resemblance to 23 [125]. O
N N H
H N
N S
N O
24
O O
Ph
CH3 O
O
S
N
O
N N
N H
N
O S
O
N
NH
H3C S
O 25
A series of HCV NS5A inhibitors based on homoproline and its isosteres have been claimed, with 26 a representative example that demonstrates an EC50 of o1 mM in a genotype 1b HCV replicon assay [126]. However, resistance mapping identified mutations not only in NS5A (Y93H) but also in NS3, NS4A and NS4B.
CH3
H N
HO
O
N O
H N
O
O N
Boc
N
O
HN
OH
Ph
N N
26
27
H3C
Using a replicon assay, a series of piperazine-based inhibitors of replication was identified from which 27 is the most potent, EC50 ¼ 160 nM, with
412
Nicholas A. Meanwell et al.
activity dependent on the presence of the phenol moiety [127]. Resistance was mapped to A92V, Y93H and R157W of NS5A, residues that are located at the dimer interface of domain 1 in the X-ray crystallographic structure, suggesting that these molecules may modulate the dimerization process either by interfering with protein association or exerting a stabilizing effect on the dimer [127,128].
2.6 HCV NS5B polymerase inhibitors HCV NS5B is the virus-encoded RNA-dependent, RNA polymerase (RdRp), possessing enzymatic activity that is responsible for the synthesis of viral RNA, a fundamental and critical step in replication. Inhibitors of NS5B that target the active site (nucleoside derivatives) or one of four distinct allosteric binding sites have been characterized, and representative inhibitors of each site have been clinically validated [8,129–131]. Reflecting the classical nature of this enzyme as an antiviral target and the multiple biochemical opportunities for intervention, studies in the NS5B inhibitor field have been vigorous, spawning a number of clinical candidates [8,16,17]. However, this mechanistic class has been associated with considerable attrition in the early stages of development [8,16,17]. Nevertheless, multiple compounds continue to progress through clinical studies, although the structures of several remain proprietary (Table 1). Panels of sensitive and resistant viruses from multiple HCV genotypes have been used to characterize the binding sites of NS5B inhibitors and understand interactions between the different classes or with other HCV inhibitors in vitro [132,133]. In a comparative study, the intrinsic potency of NS5B inhibitors representing several classes was found to be similar to that of protease inhibitors, IFN or cyclophilinbinding molecules [133]. On the basis of in vitro studies, a failure to detect pre-existing insensitive variants or the S282T mutant that arises in response to several nucleoside inhibitors in clinical studies, resistance to active site NS5B inhibitors may be much less likely to arise from therapy than for allosteric NS5B or NS3 protease inhibitors [27,134]. Moreover, studies in replicons have demonstrated that resistant mutants arising in response to nucleoside inhibitors exhibit reduced fitness compared with replicons resistant to allosteric NS5B inhibitors [27,134]. This raises the specter of clinical resistance with allosteric NS5B inhibitors and the overall consensus of multiple in vitro studies is that, despite excellent potency, the pre-existence or development of relatively fit resistant mutants will dictate that these molecules be used in combination therapy [24]. In vitro studies with combinations of two or three drugs that contain at least one replicase inhibitor have shown synergistic interactions [27,28].
413
Discovery and Development of Inhibitors of HCV
2.6.1 Nucleoside inhibitors The most advanced active site NS5B inhibitor in clinical development is R-1728 (28), the bis-isobutyl ester prodrug of 2u-deoxy-2u-fluoro-2u-Cmethylcytidine (PSI-6130). Interestingly, metabolism of PSI-6130 proceeds through a bifurcated pathway, with HCV inhibitory activity believed to be a function of the triphosphate of both PSI-6130 and the corresponding uridine analogue [135–137]. The latter compound, PSI-7851 (29, precise structure not disclosed), is being developed independently as a phosphoramidate prodrug of the monophosphate [138]. As monotherapy at a dose of 1500 mg BID for 14 days, R-1728 (28) produced a 2.7 log10 reduction in viral RNA, whereas 88% of patients taking 1000 mg BID of the drug in combination with SOC for 28 days followed by SOC for 44 weeks achieved undetectable RNA levels compared to 19% in the placebo group, with viremia reduced by 5.0 log10 at 28 days [139]. NH2 O
O
N
O
N O
R2 O
R3OOC
NH
O N P O H OR1
N
CH3 O O
F 28
O
O CH3 HO
F 29
In preclinical studies, PSI-7851 demonstrated 15–20-fold greater potency than PSI-6130, with an EC50 of 0.31 mM in the replicon compared to 4.8 mM for PSI-6130, and the compound effectively targets the liver in vivo, resulting in high levels of the active triphosphate [138]. Dosing of PSI-7851 (29) in a phase I SAD trial in healthy volunteers was initiated in March 2009. Development of two cytidine-based nucleoside analogues, R-1626 (30) and valopicitabine (31), both being developed as prodrugs, were discontinued due to toxicity [17,140–142]. R-1626 (30) was associated with an unacceptable level of grade 4 neutropenia observed during a 4-week study of the triple combination of the compound (1500 mg BID) in conjunction with SOC while valopicitabine (31) caused gastrointestinal side effects [17]. MK-0608 (32) has also been abandoned for undisclosed reasons despite showing good efficacy in HCV-infected chimpanzees [17,143]. Nevertheless, this area continues to be of interest, and a liver-targeted nucleoside prodrug designated IDX-184 has recently
414
Nicholas A. Meanwell et al.
advanced into phase I clinical studies [144] while R-1728 (28) is being investigated in combination with the NS3 protease inhibitor ITMN-191 (5) [30]. NH3+Cl−
NH2 O
N
N O
N
O
NH2
N
HO
O
N O
HO
N
O
O
O
O N3
O O
O
CH3
CH3
H2N
O
30
N
OH
HO
31
OH
32
2.6.2 Allosteric, non-nucleoside inhibitors 2.6.2.1 Thumb site I inhibitors. Thumb site 1 inhibitors bind to an allosteric pocket normally occupied by the polymerase finger loop, an interaction critical to the initiation of RNA synthesis, and are characterized by resistance arising by mutation at P495, P496 or V499. Clinical proof-ofconcept for inhibitors that target this site was obtained with BILB-1941, a compound of undisclosed structure that was terminated due to gastrointestinal side effects [145]. Inhibitors that target this site have been extensively investigated and are based on chemotypes that incorporate a 6,5 fused heterocyclic core, typically an indole or a benzimidazole, with a cyclohexyl substituent at the one position of the five-membered ring, a polar carboxamide-based substituent at C-6 and an aromatic ring at C-2, as represented by 33 [8,129–131,146,147]. JTK-109 (34) was advanced into clinical studies and is representative of early inhibitors based on this chemotype that have been the focus of further optimization [148]. Cl O
R2 6
Y
R3
X
2
R1
F HO2C
N O N N
33
34
415
Discovery and Development of Inhibitors of HCV
This chemotype has been pursued with some vigor leading to compounds 35 and 36 as representative of a broad patent and literature estate [149–151]. MK-3281 (structure not disclosed) has emerged as a potent inhibitor of a 1b replicon, EC90 in 50%HS ¼ 241 nM, which has been advanced into clinical studies [152]. MK-3281 was generally well tolerated in humans, exhibiting a pharmacokinetic profile predictive of a BID dosing regimen in infected patients, consistent with observations in HCV-infected chimps that experienced a 3.8 log10 (1b) and a mean 1.4 log10 (1a) reduction in viral RNA when dosed at 10 mg/kg BID, with no rapid emergence of resistance observed [153]. NEt2
NMe2 H
O O Me2N
O S
N H
H
HO2C
N
N
N
O
OMe
35
36
GL-60667 (LDI-133, 37), a potent inhibitor of HCV replication in vitro, EC50 ¼ 75 nM (1b) and 130 nM (1a), has been reported to have a preclinical profile suitable for development [154]. A series of patent applications claim polycyclic NS5B inhibitors with structures consistent with a thumb site 1 target, with 38 representative of a cyclopropylamide-based chemotype [155–161]. More recent applications claim compounds in which the amide moiety is replaced by a heterocycle, as exemplified by 39 [162]. O
O O
N O HO2C
O
O H3C
N N
N S
37
H3C
N
S
O N
HN
CH3 CH3
38
N
416
Nicholas A. Meanwell et al.
H3CO N N
H3C O H3C
O
CH3
CH3
CH3 O
S O
N
N
N H
OCH3
39
2.6.2.2 Thumb site II inhibitors. The second thumb site is a hydrophobic pocket at the base of this domain that is characterized by resistant mutations emerging at M423 or L419 in response to exposure of replicons to these inhibitors. The discovery and preclinical profile of filibuvir (40), currently in phase II clinical studies with SOC, has been described in some detail [163,164]. Filibuvir (40) is a potent 1b replicon inhibitor, EC50 ¼ 41 nM, that demonstrates good aqueous solubility, 2.5 mg/mL, and no cytochrome P (CYP) inhibition liabilities [163]. In HCV-infected individuals, filibuvir (40) administered as monotherapy for 8 days, reduced mean plasma viral RNA by 0.97, 1.84, 1.74 and 2.13 log10 at doses of 100, 300, 450 mg BID and 300 mg TID, respectively [165]. A number of subjects experienced a plateau or rebound in HCV RNA after an initial rapid reduction, attributed to mutations arising at M423 [165]. HO OH N Et
O
O
N
CH3
CH3
N
N
N
N
CH3 Et 40
H3C H3C CH 3
S
O COOH 41
The second major class of NS5B inhibitor binding to thumb site II is based on a thiophene carboxylic acid core with 41 selected as representative from a recent patent application [166]. Three compounds
Discovery and Development of Inhibitors of HCV
417
from this series, VCH-916, VCH-759 and VCH-222 (structures not disclosed), have been advanced into clinical trials, with the latter two apparently remaining in active development. Doses of VCH-916 at 200 mg TID for 14 days or 300 or 400 mg BID for 3 days as monotherapy in genotype 1 subjects produced greater than a 1.2 log10 reduction in HCV RNA [167]. Resistant mutants, primarily at L419 or M423, emerged in the 14-day study but not during the 3-day regimen [167]. VCH-222 exhibits improved efficacy with a 750-mg BID dose reducing plasma viral RNA levels by 3.7 log10 at the end of a 3-day study [168]. VCH-222 is a potent inhibitor of genotype 1a and 1b replication in vitro, EC50 ¼ 23 and 12 nM, respectively, and displays good oral bioavailability in rats and dogs, with a liver to plasma ratio of 5 in rats [169,170].
2.6.2.3 Palm site I inhibitors. Palm site I is located at the junction of the thumb and palm domains, proximal to the active site, and inhibitors were originally discovered with a benzothiadiazine chemotype. Resistance to benzothiadiazines mapped to M414T as a signature mutation along with alterations in nearby amino acids, including N411. Several compounds targeting this site have been advanced into clinical studies, including GSK-625433 (42), evolved from a series of pyrrolidine derivatives but discontinued due to hepatotoxicity in preclinical species [171], ANA-598 (structure not disclosed), which has demonstrated efficacy in phase I clinical trials [172], and ABT-072 and ABT-333 (structures not disclosed) [173]. ANA-598 is presumably derived from a series of thiadiazines that have been extensively studied by Anadys with the 5,6-dihydropyridone 43 representing the most recent refinement of the chemotype in the context of optimizing its preclinical profile, particularly oral bioavailability [174]. This compound is a potent inhibitor of HCV polymerase, IC50 o10 nM, that is effective in a replicon, EC50 ¼ 16 nM, with activity dependent on the absolute configuration. The compound exhibits good metabolic stability in monkey liver microsomes and is permeable across a confluent layer of Caco-2 cells, properties that translate into good exposure following oral administration to cynomolgus monkeys [174]. OMe N N HO2C
O
N H
N
OH
N
S
NHSO2Me
OH
N
H
N
O
O H3C
CH3 CH3
H3C CH OMe 3 F
42
43
O S N H
N H
O H3C
O
O S
R
44: R = H 45: R = CH3
NHSO2Me
418
Nicholas A. Meanwell et al.
Another compound from this general chemotype is A-837093 (44), which has been the subject of detailed profiling [175]. The saturated dimethyl butane side chain found in 44 provides improved potency compared to an unsaturated isobutylene moiety, and activity is dependent on chirality since enantiomers are typically 30–40-fold less potent. The introduction of the chiral carbon atom in the ring enhances solubility compared to earlier compounds that incorporate a nitrogen atom at this site and confer a more planar disposition. The tert-butyl homologue 45 shows greater metabolic stability than A-837093 (44), which translates into considerably improved PK characteristics in multiple species following IV dosing, although the advantage was not as apparent after oral dosing [175]. Resistance to A-837093 (44), generated in 1b replicons, mapped to S368A, Y448H, G554D, Y555C and D559G, but these mutants retained sensitivity to a protease inhibitor, a thumb site II non-nucleoside inhibitor and IFN-a [176]. A-837093 (44) produced a maximal viral RNA reduction of 1.4 log10 in a chimpanzee infected with genotype 1a HCV following dosing at 30 mpk BID for 14 days [177]. However, the compound was more efficacious in a 1b-infected animal where a 2.5 log10 reduction in plasma RNA was measured. Viral rebound and partial viral rebound occurred in the 1b- and 1a-infected chimps, respectively, and many of the same resistance mutations observed in in vitro studies were characterized in these animals along with a C316Y mutation in the 1b-infected chimpanzee [177]. Of the two clinical candidates, ABT-333 appears to be the most advanced, but ABT-072 is the more potent [173,178–180]. ABT-333 displays EC50 of 2–7 nM in replicons, with a 10–14-fold shift in the presence of 40% human serum, whereas the corresponding data for ABT072 are EC50 of 0.3–5.3 nM, with an 8–17-fold shift when 40% human serum is added [175,178]. Both compounds elicited resistant mutations at positions C316Y, M414T, Y448H/C or S556G in vitro but most resistant replicons displayed reduced replication capacity. Results from SAD and MAD studies with ABT-333 at doses of 200–400 mg BID to healthy volunteers were encouraging, with exposure of the compound minimally affected by co-dosing with the CYP 3A4 inhibitor ketoconazole [179,180].
2.6.2.4 Palm site II inhibitors. The palm site II pocket partially overlaps with palm site I and is fully formed by a conformational change in the R200 region of the enzyme, which occurs upon binding of inhibitors. Inhibitors that bind to this site select for mutations at C316, I363 or S365. Nesbuvir (HCV 796, 46) is the most prominent palm site II inhibitor that potently blocks HCV genome replication in vitro with EC50 of 5 and 9 nM for 1a and 1b constructs, respectively. In a chimeric mouse model of HCV infection, nesbuvir (46) produced a W2 log10 reduction in viremia, which,
Discovery and Development of Inhibitors of HCV
419
along with a good preclinical PK profile, provided confidence to advance this compound into clinical studies [181]. As monotherapy at a dose of 1000 mg BID, this compound reduced viral RNA by 1.4 log10 at day 4 [43,181]. However, viral rebound occurred during continued treatment, potentially attributed to a C316Y mutant, with the result that the reduction in viral load was only 0.8 log10 at day 14 [43,181]. Co-administration of nesbuvir (46) with PEG-IFNa and ribavirin reduced HCV viral load by up to 3.5 log10 at day 14, an improvement over the 1.7 log10 reduction seen with PEG-IFNa alone [181]. The C316Y mutant has been shown to be 138–166-fold less sensitive to nesbuvir (46) and the molecular mechanisms of resistance and effect on viral fitness in vitro have been thoroughly studied [182,183]. Unfortunately, the discontinuation of nesbuvir (46) was announced in 2007 following the emergence of severe hepatotoxicity in 8% of HCV-infected subjects following 8 weeks of treatment with the compound in combination with SOC [184]. H3C
NH
O
F
HO
O
N O
S O CH3 46
GS-9190 has been advanced into clinical studies and, although the structure has not been disclosed, the compound appears to be derived from a series of imidazo[4,5-c]pyridines of which 47 is prototypical [185]. Compound 47 exhibits an EC50 of 0.010 mM and a CC50 of 108 mM in a replicon system [185]. Compounds from this class of NS5B inhibitor are thought to interact with a fifth allosteric site that is very close to palm site II and which has been referred to as a palm or finger binding site and is characterized by mutations at C445 (F or Y), Y448 and Y452 (H) [128–130]. However, these compounds are somewhat cryptic in their inhibitory activity since they are inactive towards the isolated NS5B enzyme in vitro. GS-9190 displays an EC50 of 0.7 nM towards a genotype 1b replicon and in clinical studies produced peak viral load reductions of 1.4 log10 at 40 mg BID and 1.7 log10 at 120 mg BID following an 8-day treatment regimen [186]. However, cardiac monitoring revealed possible abnormalities leading to a temporary suspension of clinical studies while the QT risk was more fully assessed. GS-9190 is currently in phase IIb trials where it is being evaluated at a dose of 40 mg BID in
420
Nicholas A. Meanwell et al.
conjunction with SOC. Although the structure of GS-9190 has not been revealed, two recent patent applications focus specifically on compound 48 [187,188]. CF3 F F3C
F F
N N
N
47
N
F3C N
N N
N
48
2.7 Emerging mechanisms 2.7.1 HCV entry inhibitors Elucidation of the biochemical steps associated with HCV entry and identification of the cellular proteins involved is a dynamic area of study that has the potential to provide new opportunities for therapeutic intervention [189–193]. An important and apparently final piece in the puzzle of the full repertoire of cellular factors required for viral entry is the identification of the tight junction protein occludin as a co-factor [194,195]. Occludin appears to function in concert with another tight junction protein, claudin 1 [196–200], the tetraspanin CD81 [201,202] and the scavenger receptor (SR) B1 [203–208]. However, although CD81 is a determinant of infectivity both in vitro and in vivo, it does not appear to play a critical role in cell–cell transmission of virus [209,210]. The host cell factor EWI-2wint has been identified as an endogenous partner of CD81 that blocks infection when co-expressed, perhaps an important element in viral tropism since EWI-2wint is not expressed in hepatocytes [211]. Plasma membrane enriched in ceramide leads to CD81 downregulation by ATP-independent internalization [212]. Claudin-1 is internalized by increases in cellular cAMP and protein kinase A activity while increased fatty acid synthase activity reduces claudin-1 expression [213,214]. High avidity antibodies to SR-B1 block infection in the presence of LDL, whereas IFN-a has been shown to downregulate this cellular receptor [215,216]. The heavily glycosylated HCV surface proteins E1 and E2 mediate viral entry, although the precise role of each protein in the carefully choreographed events associated with membrane fusion remains to be determined [217,218]. Nevertheless, the entry process can be studied in the context of pseudoparticles [219] or replicating virus and screens of this type are being used to identify effective inhibitors [220], with the
421
Discovery and Development of Inhibitors of HCV
first drug candidates advancing into clinical studies. Arbidol (49) is a broad-spectrum antiviral agent that has been shown to inhibit pseudoparticle-mediated infection with an IC50 of 11.3 mM by a mechanism that is not specific, since it inhibits HCV replication in a replicon by a mechanism that remains to be elucidated [221]. Benzyl salicylate acid and terfenadine have been characterized as inhibitors of E2-CD81 association, but SAR studies with each molecule failed to significantly improve potency [222,223].
CO2CH3 N
CH3
O
N CH 3
H3C N
N
HO
S Ph N
Br
N
N
Cl
CF3
S
CH3 OCH3
H3CO
49
50
51
PRO-206 (structure not disclosed) is an inhibitor of HCV entry optimized from a lead identified in a library of over 370,000 compounds, with structural refinement affording molecules with favorable antiviral activity and PK properties [224]. PRO-206 inhibits the entry of HCV pseudoparticles expressing a range of viral envelopes with EC50 in the 1–50 nM range for the more sensitive sequences. However, several viral envelopes demonstrated reduced sensitivity with EC50 ranging from 0.186 to W10 mM [224]. Phase 1 clinical trials with the HCV entry inhibitor ITX-5061, a picomolar inhibitor of genotype 1 and 2 HCV entry, have been completed, and the compound is undergoing proof-of-concept trials in HCV-infected patients [225]. Neither the structure nor the mode of action of ITX-5061 has been disclosed but a patent application claims a range of compounds based on variants of a linear tricyclic scaffold that are claimed to block HCV entry, with 50 a representative structure [226]. A patent application claims a broad series of HCV entry inhibitors, with 51 representative, that exhibits an EC50 of o100 nM in an in vitro infection assay in HepG2 cells [227]. JTK-652 (structure not disclosed), an HCV entry inhibitor discovered by screening using a pseudovirus assay, was evaluated in 10 genotype 1 HCV-infected patients at a dose of 100 mg TID as monotherapy for 28 days but failed to exhibit any significant effect on viral load [228].
422
Nicholas A. Meanwell et al.
2.7.2 HCV internal ribosome entry site inhibitors The HCV viral mRNA incorporates an internal ribosome entry site (IRES) that recruits ribosomes and initiation factors directly to the translation start site, avoiding 5u-cap-dependent protein synthesis [229,230]. The functional roles of the domains of the HCV IRES have been mapped, and gross structural information is available from cryo-electron microscopy studies, whereas X-ray crystallographic and NMR structural data have recently been obtained for domain II, the smaller of the two major domains [231–232]. The HCV IRES is viewed as a potentially interesting target for therapeutic intervention, but studies described in detail to date have been largely restricted to RNA-based technologies, including RNA aptamers [233–236], small interfering RNAs (siRNAs) [237–239] and antisense RNA [240,241]. Delivery of an antisense RNA 17 mer to replicon cells was markedly enhanced by conjugation with lipid, octadecanol or cholesterol, to which an alkyne moiety was introduced to allow participation in a copper-catalyzed cycloaddition with an azide at the 5u-position of the oligonucleotide [241]. A bicyclic peptide aptamer with high affinity and good selectivity for the HCV IRES was discovered after multiple rounds of selection from a library [242]. A 27 residue peptide containing three cysteines was exposed to dibromoxylene to afford a bicyclic peptide that bound to the HCV IRES with a Kd of 0.7 nM, close to 10-fold more potent than the acyclic peptide. The IRES recognition element was mapped to the 8N-terminus residues, KCSRGIRC, which demonstrated a Kd of 17.5 nM in the linear form and improved to a Kd of 3.7 nM on cyclization with dibromoxylene [242]. The pokeweed antiviral protein, a 262 residue peptide, and mutated derivatives devoid of cytotoxicity have been claimed to bind to stem loop domains II or IIId of the HCV IRES with single-digit nanomolar affinity, thereby interfering with access by the 40S ribosomal subunit [243]. A series of patent applications claim indole-based inhibitors of the HCV IRES, discovered using proprietary gene-expression modulation by small molecules (GEMS) technology, which inhibit HCV replication in a replicon [244–246]. PS-102123 (52, SCH-1383646) is highlighted as a compound that demonstrates an EC50 of o500 nM in a replicon assay and interacts in a synergistic fashion with the HCV protease inhibitor SCH-446211 (SCH-6) [246,247]. CN
O
N N
CH3
O NH
O
N Cl 52
CH3
423
Discovery and Development of Inhibitors of HCV
2.7.3 HCV assembly and egress inhibitors HCV virion production in human hepatocytes is dependent on the assembly and secretion of triglyceride-rich very low density lipoprotein (VLDL) and vesicles in which HCV replicates are enriched in proteins involved in VLDL assembly, including apolipoprotein B (apoB), apoE, and microsomal triglyceride transfer protein (MTP). MTP is a chaperone protein and BMS-201038 (53) and CP-346086 (54), potent MTP inhibitors, block VLDL assembly and reduce HCV secretion by up to 80% at 100 nM without reducing HCV RNA production, indicating that secretion but not replication depends on VLDL assembly [248,249]. Although dose-dependent effects were seen in both studies, inhibition was incomplete with B20% of HCV still released, possibly by a VLDL-independent pathway [248,249]. siRNA directed towards apoB also blocks HCV secretion [248]. CF3
CF3 O
CF3
O
NH
N
N HN
NH
O
N N
HN
53
54
Narigenin (55), a known inhibitor of VLDL secretion that reduces the expression and activity of MTP and acyl CoA cholesteryl acyl transferase (ACAT), dose dependently reduces HCV virion secretion with an 80% reduction observed at 200 mM [250]. OH HO
O
OH
O 55
O
N
HO
H
CH3
O OH OH 56
2.7.4 a-Glucosidase inhibitors a-Glucosidase inhibitors interfere with virus morphogenesis by reducing protein glycosylation in the ER, leading to the misfolding of viral
424
Nicholas A. Meanwell et al.
proteins that, as a consequence, are targeted for destruction. In addition, protein complex formation is impaired with the result that prebudding complexes of viral proteins are reduced, leading to inhibition of virion production and secretion. As might be anticipated, a-glucosidase inhibitors exhibit broad-spectrum antiviral properties, interfering with the replication of a wide range of viruses including HCV [251,252]. Iminosugars have emerged as the most prominent and effective class of a-glucosidase inhibitors with celgosovir (56), the 6-butanoyl ester prodrug of castanospermine, advanced into clinical trials in HCVinfected patients. In a phase II study, celgosivir, in combination with IFNa-2b and ribavirin, produced a mean HCV viral load reduction of 1.2 log10, which compared with a 0.4 log10 reduction in patients receiving only the IFN-a-2b/ribavirin combination [251].
2.7.5 Inhibitors of host cell targets 2.7.5.1 Cyclosporins. The peptidyl-prolyl cis-trans isomerase cyclophilin B associates with HCV NS5B to stimulate RNA binding, an interaction that is critical for viral replication and represents an interesting host cell target for therapeutic intervention [253–255]. Cyclosporin A (CsA, 57) and non-immunosuppressive cyclosporins inhibit HCV replication, apparently functioning by binding to cyclophilin B and effectively competing with the polymerase. However, cyclophilin A has also been implicated, although with a distinct mechanism [256,257]. The cyclophilin B binding site has been mapped to the C-terminal residues 521–591 of NS5B, with P450 a key amino acid since the P450A mutation eliminates protein–protein association and compromises HCV replication [253–255]. Three non-immunosuppressive cyclosporin derivatives have been examined as inhibitors of HCV – NIM-811 (58) [258,259], DEBIO-025 (59) [260–262] and SCY-635 (60) [263,264]. Resistance to CsA (57) and SCY-635 (60) maps to both HCV NS5B and NS5A, with SCY-635 (60) eliciting T77K and I432V mutations in NS5B and T17A and E295K changes in NS5A [263,265,266]. The I432V change, although located outside of the cyclophilin-binding domain, appears to enhance the NS5Bcycophilin B interaction, and interestingly, this change can rescue the lethal P450A mutation [265]. CsA (57) elicits a different range of mutations, with P538T and S556G in NS5B and several in NS5A [266]. The three non-immunosuppressive cyclosporin analogues are potent inhibitors of HCV genotype 1b replication in vitro, with EC50 less than 1 mM, approximately 10-fold more potent than CsA (57), and interact in an additive to synergistic fashion with IFNa or ribavirin [259,261,263]. In addition, they are not cross-resistant with several STAT-C agents. A genotype 4 replicon is similarly susceptible to NIM-811 (58), but the JFH1 genotype 2a replicon is less sensitive to both NIM-811 and CsA since this replicon does not depend on cyclophilin B for replication [266].
Discovery and Development of Inhibitors of HCV
425
In clinical trials in patients co-infected with HIV, DEBIO-025 (59), at a dose of 1200 mg BID for 15 days, effected a mean 3.63 log10 reduction in viral load in those infected with HCV genotypes 1, 3 and 4 [260,267]. Genotype-3-infected patients were more responsive, achieving a 4.46 log10 reduction in viremia. In a phase IIa dose range–finding study conducted in 90 treatment-naı¨ve patients, DEBIO-025 (59) (1000 mg BID for 7 days as a loading dose followed by 1000 mg QD for 21 days) in conjunction with PEG-IFNa-2a produced a 4.75 log10 reduction in viral load in genotype 1 and 4 patients measured at week 4 [260,268]. This result was superior to either agent alone, with efficacy more pronounced in genotype 2 and 3 patients. SCY-635 (60), which exhibits high affinity for both cyclophilin A and B, IC50 ¼ 7 and 10 nM, respectively, reduced plasma HCV RNA by 2.20 log10 at day 11 and 1.82 log10 at day 15 when administered at a dose of 300 mg TID for 15 days, with lower doses of 100 and 200 mg TID producing inconsistent effects on viremia [269]. CH3 N
N CH3 O HO CH3 O N
N
N
O
O
H N
CH3 N
CH3 O
58
N
CH3 O
O N CH3
O
H N O
57
N H
CH3 O
CH3 N O
N
N
O
N H
O
59 N S
CH3 N
N CH3 O
OH
60
2.7.5.2 Nitazoxanide (Alinias). Nitazoxanide (61) is a nitrothiazole derivative and a prodrug of tizoxanide (62) that was originally developed as a treatment for parasitic infections [270]. However, these compounds have been characterized as inhibitors of both HBV and HCV replication in vitro [271–274] following the observation of clinical efficacy at reducing viral load as monotherapy in HCV genotype-4-infected patients [275]. In HCV replicons, nitazoxanide (61) and tizoxanide (62) inhibit viral RNA replication with EC50 of 210 and 150 nM, respectively, with good therapeutic indexes, CC50 ¼ 38 and 15 mM, respectively [272].
426
Nicholas A. Meanwell et al.
HBV viral DNA production in an in vitro replication assay was inhibited with similar potency, results consistent with the notion that nitazoxanide (61) and tizoxanide (62) modulate a host cell process and supported by resistance studies in HCV replicons [273]. In replicons and Huh-7 cells infected with a genotype 2a virus, nitazoxanide (61) increased the levels of phosphorylated elF2a and its dsRNA-activated protein kinase (PKR), key components of the host antiviral defenses, an effect augmented by IFN [274]. Collectively, these results suggest that nitazoxanide (61) modulates an innate antiviral pathway. In a placebo-controlled clinical trial, 7 of 23 genotype-4-infected patients administered nitazoxanide (61) as monotherapy at a dose of 500 mg BID for 24 weeks had a response at the end of treatment while 4 of 23 patients with a baseline viral load of o400,000 IU/mL had a SVR compared to none in the placebo group [275]. In a follow-up study, also conducted in genotype-4-infected individuals but not blinded, a 12-week lead-in with nitazoxanide (61) (500 mg BID) followed by PEG-IFNa2a/ribavirin or PEG-IFNa2a/ ribavirin/nitazoxanide for 36 weeks was compared with 48 weeks of PEG-IFNa2a/ribavirin [276]. SVR rates at 24 weeks post-treatment were 50% in the control group, 61% in the PEG-IFNa2a/ribavirin group and 79% in those who received triple therapy for 36 weeks [276]. A recent patent application claims analogues of nitazoxanide that broadly explores SAR associated with both the nitro and phenol moieties [277].
OR
O
HO
N N H
NO2 S
H3CO
O
HO 61: R = Ac 62: R = H
OH
O HO 63
O
OH
2.7.5.3 Silymarin and silibinin. A standardized preparation of silymarin, an extract of milk thistle (Silybum marianum) with antioxidant activity that has a history of use as self-medication for liver disease spanning several centuries [278,279], inhibits HCV JFH1 viral infection of Huh 7.5.1 cells in culture in a dose-dependent fashion, with 20 mg/mL demonstrating efficacy comparable to IFN-a [280]. A cocktail therapy comprising silymarin along with seven additional antioxidants demonstrated benefit in HCV-infected individuals who had not responded to IFN therapy, with normalization of liver enzymes in 40–50% of patients but modest
Discovery and Development of Inhibitors of HCV
427
reductions in viral load in only 25% of the cohort [281,282]. Silibinin (63) is one of the six major flavonolignans in silymarin, and this compound showed dose-dependent reductions in viremia when administered intravenously to HCV-infected individuals unresponsive to full doses of PEG-IFNa/ribavirin [283]. A dose of 20 mg/kg infused daily for 7 days was associated with a 3.02 log10 decrease in viral load, with a further reduction to 4.85 log10 below baseline after an additional 7 days in combination with PEG-IFNa and ribavirin [282]. However, oral dosing of silibinin (63) was not effective in maintaining the antiviral effect following the 7 days of infusion [283].
3. CONCLUSIONS The identification and development of specifically targeted inhibitors of HCV has advanced considerably over the past 2 years, and despite setbacks in clinical studies with several early candidates, there is currently a robust pipeline of potential drugs that target a range of mechanisms. The advent of HCV replicons in 1999 was an important catalyst to drug discovery, providing a screening tool to not only confirm the activity of NS3 and NS5B inhibitors but also identify compounds acting at alternative targets. The development of infectious virus offers similar promise for the identification of HCV inhibitors that interfere with virus entry and the process of assembly and egress. Although the majority of compounds in clinical studies are still in the early stages of development, the first steps towards developing a combination of specific antiviral agents to treat HCV unresponsive to SOC, either as addon or replacement therapy, have been taken. With compounds acting at multiple sites in the virus replication cycle in development, there is reason to be optimistic about the potential to identify effective combinations that offer higher rates of therapeutic cure of HCV infection resistant to the current SOC.
NOMENCLATURE AUC PEG RBV ATP GTP CYP VLDL
area under the curve polyethyleneglycol ribavirin adenosine triphosphate guanosine triphosphate cytochrome P (CYP 450 = cytochrome P450) very low density lipoprotein
428
Nicholas A. Meanwell et al.
REFERENCES [1] M. J Alter, World J. Gastroenterol., 2007, 13, 2436. [2] K. Neukam, J. Macı´as, J. A. Mira and J. A. Pineda, Expert. Opin. Pharmacother., 2009, 10, 417. [3] G. Foster and P. Mathurin, Antiviral Ther., 2008, 13(Suppl. 1), 3. [4] S. Chevaliez and J.-M. Pawlotsky, Adv. Drug Deliv. Rev., 2007, 59, 1222. [5] L. A. Sorbera, Drugs Future, 2007, 32, 937. [6] Data available on the Zymogenetics web site, http://www.zymogenetics.com/ products/Interleukin29ProductBackgrounder.htm [7] L. G. J. Leede, J. E. Humphries, A. C. Bechet, E. J. Hoogdalem, R. Verrijk and D. G. Spencer, J. Interferon Cytokine Res., 2008, 28, 113. [8] V. Soriano, M. G. Peters and S. Zeuzem, Clin. Infect. Dis., 2009, 48, 313. [9] A. Thompson, K. Patel, H. Tillman and J. G. McHutchison, J. Hepatol., 2009, 50, 184. [10] R. E. Stauber and H. H. Kessler, Drugs, 2008, 68, 1347. [11] V. Sorianao, A. Madejohn, E. Vispo, P. Labarga, J. Garcia-Samaniego, L. MartinCarbonero, J. Sheldon, M. Bottecchia, P. Tuma and P. Barreiro, Expert. Opin. Emerg. Drugs, 2008, 13, 1. [12] M. P. Manns, G. R. Foster, J. K. Rockstroh, S. Zeuzem, F. Zoulim and M. Houghton, Nat. Rev. Drug Discov., 2007, 6, 991. [13] E. B. Keeffe, Antiviral Ther., 2007, 12, 1015. [14] S. A. Harrison, Am. J. Gastroenterol., 2007, 102, 2332. [15] J.-M. Pawlotsky, S. Chevaliez and J. G. McHutchison, Gastroenterol., 2007, 132, 1979. [16] K. Garber, Nat. Biotechnol., 2007, 25, 1379. [17] N. A. Meanwell and G. Koszalka, Curr. Opin. Investig. Drugs, 2008, 9, 128. [18] D. Moradpour, F. Penin and C. M. Rice, Nat. Rev. Microbiol., 2007, 5, 453. [19] T. L. Tellinghuisen, M. J. Evans, T. Hahn, S. You and C. M. Rice, J. Virol., 2007, 81, 8853. [20] T. P. Holler, T. Parkinson and D. C. Pryde, Expert Opin. Drug Discov., 2009, 4, 293. [21] J. Timm and M. Roggendorf, World J. Gastroenterol., 2007, 13, 4808. [22] G. Koev and W. Kati, Expert. Opin. Investig. Drugs, 2008, 17, 303. [23] B. H. McGovern, B. K. Abu Dayyeh and R. T. Chung, Hepatology, 2008, 48, 1700. [24] T. Kuntzen, J. Timm, A. Berical, N. Lennon, A. M. Berlin, S. K. Young, B. Lee, D. Heckerman, J. Carlson, L. L. Reyor, M. Kleyman, C. M. McMahon, C. Birch, J. Schulze zur Wiesch, T. Ledlie, M. Koehrsen, C. Kodira, A. D. Roberts, G. M. Lauer, H. R. Rosen, F. Bihl, A. Cerny, U. Spengler, Z. Liu, A. Y. Kim, Y. Xing, A. Schneidewind, M. A. Madey, J. F. Fleckenstein, V. M. Park, J. E. Galagan, C. Nusbaum, B. D. Walker, G. V. Lake-Bakaar, E. S. Daar, I. M. Jacobson, E. D. Gomperts, B. R. Edlin, S. M. Donfield, R. T. Chung, A. H. Talal, T. Marion, B. W. Birren, M. R. Henn and T. M. Allen, Hepatology, 2008, 48, 1769. [25] D. J. Bartels, Y. Zhou, E. Z. Zhang, M. Marcial, R. A. Byrn, T. Pfeiffer, A. M. Tigges, B. S. Adiwijaya, C. Lin, A. D. Kwong and T. L. Kieffer, J. Infect. Dis., 2008, 198, 800. [26] M. Cubero, J. I. Esteban, T. Otero, S. Sauleda, M. Bes, R. Esteban, J. Guardia and J. Quer, Virology, 2008, 370, 237. [27] S. Pogam, A. Seshaadri, A. Kosaka, S. Chiu, H. Kang, S. Hu, S. Rajyaguru, J. Symons, N. Cammack and I. Na´jera, J. Antimicrob. Chemother., 2008, 61, 1205. [28] D. L. Wyles, K. A. Kaihara, F. Vaida and R. T. Schooley, J. Virol., 2007, 81, 3005. [29] C. Gru¨nberger, D. L. Wyles, K. A. Kaihara and R. T. Schooley, J. Infect. Dis., 2008, 197, 42. [30] E. J. Gane, S. K. Roberts, C. Stedman, P. W. Angus, B. Ritchie, R. Elston, D. Ipe, L. Baher, P. Morcos, I. Najera, M. Mannino, B. Brennan, M. Berrey, W. Bradford, E. Yetzer, N. Shulman and P. F. Smith, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009,
Discovery and Development of Inhibitors of HCV
[31] [32] [33] [34] [35] [36] [37] [38] [39] [40]
[41] [42] [43] [44] [45] [46]
[47]
[48] [49] [50]
[51]
[52]
[53] [54]
429
Abstract 1046 available at http://www.abstractserver.com/easl2009/planner/sp. php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 2661&EASL2009 ¼ o39t1tihqm 4c4mhf7mp4d7udr1&EASL2009 ¼ o39t1tihqm4c4mhf7mp4d7udr1 C. T. Jones, C. L. Murray, D. K. Eastman, J. Tassello and C. M. Rice, J. Virol., 2007, 81, 8374. I. C. Lorenz, J. Marcotrigiano, T. G. Dentzler and C. M. Rice, Nature, 2006, 442, 831. V. Jirasko, R. Montserret, N. Appel, A. Janvier, L. Eustachi, C. Brohm, E. Steinmann, T. Pietschmann, F. Penin and R. Bartenschlager, J. Biol. Chem., 2008, 283, 28546. C. L. Murray, C. T. Jones, J. Tassello and C. M. Rice, J. Virol., 2007, 81, 10220. V. Schregel, S. Jacobi, F. Penin and N. Tautz, Proc. Natl. Acad. U.S.A., 2009, 106, 5342, early edition, http://www.pnas.org/cgi/reprint/0810950106v1 Y. She, Q. Liao, X. Chen, L. Ye and Z. Wu, Arch. Virol., 2008, 153, 1991. J.-K. Oem, C. Jackel-Cram, Y.-P. Li, H.-N. Kang, Y. Zhou, L. A. Babiuk and Q. Liu, Arch. Virol., 2008, 153, 293. R. Maurice and D. Thibeault, WO Patent Application 2007/048254-A1, 2007. L. Lamarre, L. Lagace´ and D. Thibeault, US Patent Application 2007/0141701-A1, 2007. D. Lamarre, P. C. Andreson, M. Bailey, P. Beaulieu, G. Bolger, P. Bonneau, M. Bo¨s, D. R. Cameron, M. Cartier, M. G. Cordingley, A. M. Faucher, N. Goudreau, S. H. Kawai, G. Kukolj, L. Legace´, S. R. LaPlante, H. Narjes, M. A. Poupart, J. Rancourt, R. E. Sentjens, R. St. George, B. Simoneau, G. Steinmann, D. Thibeault, Y. S. Tsantrizos, S. M. Weldon, C. L. Yong and M. Llina`s-Brunet, Nature, 2003, 426, 186. T. Asselah, Y. Benhamou and P. Marcellin, Liver Int., 2009, 29, 57. Y. Tsantrizos, Acc. Chem. Res., 2008, 41, 1252. A. D. Kwong, L. McNair, I. Jacobson and S. George, Curr. Opin. Pharmacol., 2008, 8, 522. R. Francesco and A. Carfi, Adv. Drug Del. Rev., 2007, 59, 1242. B. Kronenberger and S. Zeuzem, Curr. Gastroenterol. Rep., 2009, 11, 15. R. B. Perni, S. J. Almquist, R. A. Byrn, G. Chandorkar, P. R. Chaturvedi, L. F. Courtney, C. J. Decker, K. Dinehart, C. A. Gates, S. L. Harbeson, A. Heiser, G. Kalkeri, E. Kolaczkowski, K. Lin, Y.-P. Luong, B. G. Rao, W. P. Taylor, J. A. Thomson, R. D. Tung, Y. Wei, A. D. Kwong and C. Lin, Antimicrob. Agents Chemother., 2006, 50, 899. H. W. Reesink, S. Zeuzem, C. J. Weegink, N. Forestier, A. Van Vliet, J. Van de wetering de Rooij, L. Mcnair, S. Purdy, R. Kauffman, J. Alam and P. L. M. Jansen, Gastroenterol., 2006, 131, 997. T. L. Kieffer, C. Sarrazin, J. S. Miller, M. W. Welker, N. Forestier, H. W. Reesink, A. D. Kwong and S. Zeuzem, Hepatology, 2007, 46, 631. N. Forestier, H. W. Reesink, C. J. Weegink, L. McNair, T. L. Kieffer, H.-M. Chu, S. Purdy, P. L. M. Jansen and S. Zeuzem, Hepatology, 2007, 46, 640. C. Sarrazin, T. L. Kieffer, D. Bartels, B. Hanzelka, U. Muh, M. Welker, D. Wincheringer, Y. Zhou, H.-M. Chu, C. Lin, C. Weegink, H. Reesink, S. Zeuzem and A. D. Kwong, Gastroenterol., 2007, 132, 1767. S. Zeuzem, C. Hezode, P. Ferenci, G. M. Dusheiko, K. Alves, L. Bengtsson, S. Gharakhanian, R. Kauffman, J. J. Alam and J.-M. Pawlotsky, Hepatology, 2008, 48(S1), A243. J. G. McHutchison, M. L. Schiffman, N. Terrault, M. P. Manns, A. M. DiBisceglie, I. M. Jacobson, N. H. Afdhal, E. Heathcote, S. Zeuzem, H. W. Reesink, S. George, N. Adda and A. J. Muir, Hepatology, 2008, 48(S1), 431, A269. F. G. Njoroge, K. X. Chen, N.-Y. Shih and J. J. Piwinski, Acc. Chem. Res., 2008, 41, 50. A. J. Prongay, Z. Guo, N. Yao, J. Pichardo, T. Fischmann, C. Strickland, J. Myers, P. C. Weber, B. M. Beyer, R. Ingram, Z. Hong, W. W. Prosise, L. Ramanathan, S. S. Taremi, T. Yarosh-Tomaine, R. Zhang, M. Senior, R.-S. Yang, B. Malcolm, A. Arasappan, F. Bennett, S. L. Bogen, K. Chen, E. Jao, Y.-T. Liu, R. G. Lovey, A. K. Saksena,
430
[55] [56]
[57]
[58]
[59]
[60]
[61] [62]
[63]
[64]
[65] [66]
[67] [68]
[69]
Nicholas A. Meanwell et al.
S. Venkatraman, V. Girijavallabhan, F. G. Njoroge and V. Madison, J. Med. Chem., 2007, 50, 2310. C. Sarrazin, R. Rouzier, F. Wagner, N. Forestier, D. Larrey, S. K. Gupta, M. Hussain, A. Shah, D. Cutler, J. Zhang and S. Zeuzem, Gastroenterol., 2007, 132, 1270. P. Kwo, E. J. Lawitz, J. McCone, E. R. Schiff, J. M. Vierling, D. Pound, M. Davis, J. S. Galati, S. C. Gordon, N. Ravendhran, L. Rossaro, F. H. Anderson, I. M. Jacobson, R. Rubin, K. Koury, E. I. Chaudhri and J. K. Albrecht, Hepatology, 2008, 48(S1), 268A, LB16. X. Tong, A. Arasappan, F. Bennett, R. Chase, B. Feld, Z. Guo, A. Hart, V. Madison, B. Malcolm, J. Pichardo, A. Prongay, R. Ralston, A. Skelton, E. Xia and F. G. Njoroge, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 967 available at http://www.abstractserver.com/easl2009/planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 595&EASL2009 ¼ lb58m4anr8hk3eaqos5519g1d7 &EASL2009 ¼ lb58m4anr8hk3eaqos5519g1d7 S. Venkatraman, F. Velazquez, W. Wu, M. Blackman, K. X. Chen, S. Bogen, L. Nair, X. Tong, R. Chase, A. Hart, S. Agrawal, J. Pichardo, A. Prongay, K.-C. Cheng, V. Girijavallabhan, J. Piwinski, N.-Y. Shih and F. G. Njoroge, J. Med. Chem., 2009, 52, 336. F. Velazquez, S. Venkatraman, M. Blackman, P. Pinto, S. Bogen, M. Sannigrahi, K. Chen, J. Pichardo, A. Hart, X. Tong, V. Girijavallabhan and F. G. Njoroge, J. Med. Chem., 2009, 52, 700. K. X. Chen, L. Nair, B. Vibulbhan, W. Yang, A. Arasappan, S. L. Bogen, S. Venkatraman, F. Bennett, W. Pan, M. L. Blackman, A. I. Padilla, A. Prongay, K.-C. Cheng, X. Tong, N.-Y. Shih and F. G. Njoroge, J. Med. Chem., 2009, 52, 1370. S. Venkatraman, W. Wu, A. Prongay, V. Girijavallabhan and F. G. Njoroge, Bioorg. Med. Chem. Lett., 2009, 19, 180. P. Raboisson, H. Kock, A. Rosenquist, M. Nilsson, L. Salvador-Oden, T.-I. Lin, N. Roue, V. Ivanov, H. Wahling, K. Wickstrom, E. Hamelink, M. Edlund, L. Vrang, S. Vendeville, W. Vreken, D. McGowan, A. Tahri, L. Hu, C. Boutton, O. Lenz, F. Delouvroy, G. Pille, D. Surleraux, P. Wigerinck, B. Samuelsson and K. Simmen, Bioorg. Med. Chem. Lett., 2008, 18, 4853. T.-I. Lin, O. Lenz, G. Fanning, T. Verbinnen, F. Delouvroy, A. Scholliers, K. Vermeiren, A. Rosenquist, M. Edlund, B. Samuelsson, L. Vrang, H. Kock, P. Wigerinck, P. Raboisson and K. Simmen, Antimicrob. Agents Chemother., 2009, 53, 1377. M. P. Manns, H. W. Reesink, C. Moreno, T. Berg, Y. Benhamou, Y. J. Horsmans, G. M. Dusheiko, R. Flisiak, P. Meyvisch, O. Lenz, K. Simmen and R. Verloes, Hepatology, 2008, 48(S1), 2687, LB8. R. Rajagopalan, S. Misialek, S. K. Stevens, D. G. Myszka, B. J. Brandhuber, J. A. Ballard, S. W. Andrews, S. D. Seiwert and K. Kossen, Biochemistry, 2009, 48, 2559. S. D. Seiwert, S. W. Andrews, Y. Jiang, V. Serebryany, H. Tan, K. Kossen, P. T. Ravi Rajagopalan, S. Misialek, S. K. Stevens, A. Stoycheva, J. Hong, S. R. Lim, X. Qin, R. Rieger, K. R. Condroski, H. Zhang, M. Geck Do, C. Lemieux, G. P. Hingorani, D. P. Hartley, J. A. Josey, L. Pan, L. Beigelman and L. M. Blatt, Antimicrob. Agents Chemother., 2008, 52, 4432. N. Forestier, D. G. Larrey, D. Guyader, P. Marcellin, R. Rouzier, A. A. Patat, W. Z. Bradford, S. Porter and S. Zeuzem, Hepatology, 2008, 48(S1), 1132A, A1847. N. J. Liverton, M. K. Holloway, J. A. McCauley, M. T. Rudd, J. W. Butcher, S. S. Carroll, J. DiMuzio, C. Fandozzi, K. F. Gilbert, S.-S. Mao, C. J. McIntyre, K. T. Nguyen, J. J. Romano, M. Stahlhut, B.-L. Wan, D. B. Olsen and J. P. Vacca, J. Am. Chem. Soc., 2008, 130, 4607. D. H. Wright, J. L. Miller, I. Verlinden, C. Cilissen, J. Valentine, P. Sun, M. De Smet, J. de Hoon, M. Depre, L. Cavens, J. Chodakewitz and J. A. Wagner, Hepatology, 2008, 48(S1), 1165A, A1910.
Discovery and Development of Inhibitors of HCV
431
[70] E. J. Lawitz, M. S. Sulkowski, I. M. Jacobson, S. Faruqui, W. K. Kraft, B. Maliakkal, M. Al-Ibrahim, R. H. Ghalib, S. C. Gordon, P. Kwo, J. Rockstroh, M. Miller, P. Hwang, J. Gress and E. Quirk, Hepatology, 2008, 48(S1), 203A, A211. [71] M. P. Manns, M. Bourliere, Y. Benhamou, S. Pol, M. Bonacini, T. Berg, C. Trepo, D. Wright, G. Steinmann, D. B. Huang, J. Mikl, G. Kukolj and J. O. Stern, Hepatology, 2008, 48(S1), 1133A, A1849. [72] M. P. Manns and M. Bourlie`re, Hepatology, 2008, 48(S1), 1151A, A1882. [73] R. Ronn, Y. A. Sabnis, T. Gossas, E. Akerblom, U. H. Danielson, A. Hallberg and A. Johansson, Bioorg. Med. Chem., 2006, 14, 544. [74] D. A. Campbell, D. T. Winn, J. M. Betancort and M. E. Hepperle, WO Patent Application 2007/016476-A2, 2007. [75] D. A. Campbell, M. E. Hepperle, D. T. Winn and J. M. Betancort, WO Patent Application 2007/089618-A2, 2007. [76] A. X. Wang, B. Z. Zheng, S. D’Andrea, Q. Zhao and P. M. Scola, WO Patent Application 2008/064066-A1, 2008. [77] M. D. Bailey, F. Bilodeau, P. Forgione, V. Gorys, M. Llina`s-Brunet, J. Naud, J. O’Meara and M.-A. Poupart, WO Patent Application 2008/098368-A1, 2008. [78] N. Sin, B. L. Venables, L.-Q. Sun, S.-Y. Sit, Y. Chen and P. M. Scola, WO Patent Application 2008/060927-A2, 2008. [79] P. Ortqvist, S. D. Peterson, E. Kerblom, T. Gossas, Y. A. Sabnis, R. Fransson, G. Lindeberg, U. H. Danielson, A. Karlen and A. Sandstrom, Bioorg. Med. Chem., 2007, 15, 1448. [80] C. C. Parsy, F.-R. Alexandre and D. Surleraux, WO Patent Application 2009/014730-A1, 2009. [81] T. Brandl, S. Cottens, C. Ehrhardt, J. Fu, S. Karur, D. T. Parker, M. A. Patane, P. Raman, S. A. Randl, P. Rigollier, M. Seepersaud and O. Simic, WO Patent Application 2008/ 033389-A2, 2009. [82] S. D. Britt, J. Fu, D. T. Parker, M. A. Patane, P. Raman, B. Radetich, M. Seepersaud, A. Yifru, R. Zheng, T. Brandl, S. Cottens, C. Ehrhardt, S. A. Randl, P. Rigollier, N. Schiering and O. Simic, WO Patent Application 2008/101665-A1, 2008. [83] W. Yang, Y. Zhao, J. Fabrycki, X. Hou, X. Nie, A. Sanchez, A. Phadke, M. Deshpande, A. Agarwal and M. Huang, Antimicrob. Agents Chemother., 2008, 52, 2043. [84] M. Huang, Y. Sun, W. Yang, H. Hou, J. Fabrycki, X. Nie, A. Sanchez, Y. Zhao, A. Phadke and M. Deshpande, 42nd Annual Meeting of the European Association for the Study of the Liver, Barcelona, Spain, April 11–15, 2007, http://www.easl.ch/ easl2007/Program/ViewAbstract.asp [85] A. M. I. Lam and D. N. Frick, J. Virol., 2006, 80, 404. [86] S. Dumont, W. Cheng, V. Serebrov, R. K. Beran, I. Tinoco, A. M. Pyle and C. Bustamante, Nature, 2006, 439, 105. [87] R. K. F. Beran and A. M. Pyle, J. Biol. Chem., 2008, 283, 29929. [88] J. L. Kim, K. A. Morgenstern, J. P. Griffith, M. D. Dwyer, J. A. Thomson, M. A. Murcko, C. Lin and P. R. Caron, Structure, 1998, 6, 89. [89] L. Biegelman, B. Buckman, G. Wang, J. Matulic-Adamic, A. Stoycheva, S. W. Andrews, S. M. Misialek, P. T. Ravi Rajagopalan, A. M. Fryer, I. Gunawardana, J. Haas, L. Huang, M. R. Madduru, G. Zhang, K. Kossen and V. Serebryany, WO Patent Application 2008/100867-A2, 2008. [90] A. M. Paredes and K. J. Blight, J. Virol., 2008, 82, 10671. [91] A. A. Thompson, A. Zou, J. Yan, R. Duggal, W. Hao, D. Molina, C. N. Cronin and P. A. Wells, Biochemistry, 2009, 48, 906. [92] S. Einav, E. H. Sklan, H. M. Moon, E. Gehrig, P. Liu, Y. Hao, A. W. Lowe and J. S. Glenn, Hepatology, 2008, 47, 827. [93] J. S. Glenn and S. Einav, WO Patent Application 2009/005615-A1, 2009.
432
Nicholas A. Meanwell et al.
[94] S. Einav, D. Gerber, P. D. Bryson, E. H. Sklan, M. Elazar, S. J. Maerkl, J. S. Glenn and S. R. Quake, Nat. Biotechnol., 2008, 26, 1019. [95] S. R. Quake, S. Einav, J. S. Glenn, R. Mcdowell, W. Yang, D. Gerber and H. DvorySobol, WO Patent Application 2009/039246-A2, 2009. [96] S. Einav, J. S. Glenn, R. Mcdowell and W. Yang, WO Patent Application 2009/039248A2, 2009. [97] S. K. Chunduru, C. A. Benatatos, T. J. Nitz and T. R. Bailey, US Patent Application 2007/ 0269420-A1, 2007. [98] P. D. Bryson, N.-J. Cho, S. Einav, C. Lee, V. Tai, J. Bechtel, M. Sivaraja, C. Roberts, U. Schmitz and J. S. Glenn, 15th International Symposium on Hepatitis C Virus and Related Viruses, San Antonio, TX, USA, October 5–9, 2008, Abstract 213. [99] U. Schmitz and S.-L. Tan, Recent Pat. Antiinfect. Drug Discov., 2008, 3, 77. [100] Y. Huang, K. Staschke, R. Francesco and S.-L. Tan, Virology, 2007, 364, 1. [101] M. Binder, D. Quinkert, O. Bochkarova, R. Klein, N. Kezmic, R. Bartenschlager and V. Lohmann, J. Virol., 2007, 81, 5270. [102] N.-J. Cho, K. H. Cheong, C.-H. Lee, C. W. Frank and J. S. Glenn, J. Virol., 2007, 81, 6682. [103] R. Brillet, F. Penin, C. Hezode, P. Chouteau, D. Dhumeaux and J.-M. Pawlotsky, J. Infect. Dis., 2007, 195, 432. [104] N. Appel, M. Zayas, S. Miller, J. Krijnse-Locker, T. Schaller, P. Friebe, S. Kallis, U. Engel and R. Bartenschlager, PLoS Pathog., 2008, 4, e1000035. [105] T. L. Tellinghuisen, K. L. Foss and J. Treadaway, PLoS Pathog., 2008, 4, e1000032. [106] T. Masaki, R. Suzuki, K. Murakami, H. Aizaki, K. Ishii, A. Murayama, T. Date, Y. Matsuura, T. Miyamura, T. Wakita and T. Suzuki, J. Virol., 2008, 82, 7964. [107] G. Cheng, A. Montero, P. Gastaminza, C. Whitten-Bauer, S. F. Wieland, M. Isogawa, B. Fredericksen, S. Selvarajah, P. A. Gallay, M. R. Ghadiri and F. V. Chisari, Proc. Natl. Acad. Sci. U.S.A, 2008, 105, 3088. [108] M. D. Bobardt, G. Cheng, L. Witte, S. Selvarajah, U. Chatterji, B. E. Sanders-Beer, T. B. H. Geijtenbeek, F. V. Chisari and P. A. Gallay, Proc. Natl. Acad. Sci. U.S.A, 2008, 105, 5525. [109] Arrow Therapeutics’ web site: http://www.arrowt.co.uk/product-hcv.asp [110] R. Nettles, C. Chien, E. Chung, A. Persson, M. Gao, M. Belema, N. A. Meanwell, M. DeMicco, T. C. Marbury, R. Goldwater, P. Northup, J. Coumbis, W. K. Kraft, M. R. Charlton, J. C. Lopez-Talavera and D. M. Grasela, Hepatology, 2008, 48(S1), 267A, LB12. [111] C. Bachand, M. Belema, D. H. Deon, A. C. Good, J. Goodrich, C. A. James, R. Lavoie, O. D. Lopez, A. Martel, N. A. Meanwell, V. N. Nguyen, J. L. Romine, E. H. Ruediger, L. B. Snyder, D. R. St. Laurent, F. Yang, D. R. Langley and L. G. Hamann, WO Patent Application 2008/021928-A2, 2008. [112] C. Bachand, M. Belema, D. H. Deon, A. C. Good, J. Goodrich, C. A. James, R. Lavoie, O. D. Lopez, A. Martel, N. A. Meanwell, V. N. Nguyen, J. L. Romine, E. H. Ruediger, L. B. Snyder, D. R. St. Laurent, F. Yang, D. R. Langley and L. G. Hamann, WO Patent Application 2008/021936-A2, 2008. [113] C. Bachand, M. Belema, D. H. Deon, A. C. Good, J. Goodrich, L. G. Hamann, C. A. James, D. R. Langley, R. Lavoie, O. D. Lopez, A. Martel, N. A. Meanwell, V. N. Nguyen, J. L. Romine, E. H. Ruediger, L. B. Snyder, D. R. St. Laurent, F. Yang and G. Wang, WO Patent Application 2008/144380-A1, 2008. [114] C. Bachand, M. Belema, D. H. Deon, A. C. Good, J. Goodrich, C. A. James, R. Lavoie, O. D. Lopez, A. Martel, N. A. Meanwell, V. N. Nguyen, J. L. Romine, E. H. Ruediger, L. B. Snyder, D. R. St. Laurent, F. Yang, D. R. Langley, G. Wang and L. G. Hamann, WO Patent Application 2008/021927-A2, 2008. [115] C. J. Wheelhouse, A. J. F. Thomas, D. J. Bushnell, J. Lumley, J. I. Salter, M. C. Carter, N. Mathews, C. J. Pilkington and R. M. Angell, WO Patent Application 2007/031791A1, 2007.
Discovery and Development of Inhibitors of HCV
433
[116] N. Mathews, A. J. F. Thomas, K. C. Spencer, H. Dennison, M. C. Barnes, S. S. Chana, L. Jennens and C. J. Pilkington, WO Patent Application 2007/042782-A1, 2007. [117] N. Mathews, A. J. F. Thomas, K. C. Spencer, N. Tiberghien, C. J. Pilkington, L. Jennens, S. Chana and I. J. Fraser, WO Patent Application 2007/080401-A1, 2007. [118] J. Lumley, J. I. Salter, M. C. Carter, N. Mathews, C. J. Pilkington, A. J. F. Thomas and I. Fraser, WO Patent Application 2007/138242-A1, 2007. [119] S. Chana, WO Patent Application 2008/056149-A1, 2008. [120] M. C. Carter, S. Cockerill, S. S. Flack and C. J. Wheelhouse, WO Patent Application 2009/034390-A1, 2009. [121] F. U. Schmitz, C. D. Roberts, A. D. M. Abadi, R. C. Griffith and M. R. Leivers, WO Patent Application 2007/070556-A2, 2007. [122] F. U. Schmitz, C. D. Roberts, A. D. M. Abadi, R. C. Griffith, M. R. Leivers, I. Slobodov and R. Rai, WO Patent Application 2007/070600-A2, 2007. [123] M. R. Leivers, F. U. Schmitz, R. C. Griffith, C. D. Roberts, A. D. M. Abadi, S. A. Chan, R. Rai, I. Slobodov and T. L. Ton, WO Patent Application 2008/064218-A2, 2008. [124] M. R. Leivers, F. U. Schmitz, C. D. Roberts and A. Dehghani Mohammad Abadi, WO Patent Application 2008/070447-A2, 2008. [125] R. Rai, F. U. Schmitz, C. D. Roberts, I. Slobodov and M. R. Leivers, WO Patent Application 2008/154601-A1, 2008. [126] G. Li, R. Fathi, Z. Yang, Y. Liao, Q. Zhu, A. Lam, A. Sandrasagra, K. Nawoschik, H.-J. Cho, J. Cao, W. Ruoqiu and R. C. Wobbe, WO Patent Application 2008/048589-A2, 2008. [127] I. Conte, C. Giuliano, C. Ercolani, F. Narjes, U. Koch, M. Rowley, S. Altamura, R. De Francesco, P. Neddermann, G. Migliaccio and I. Stansfield, Bioorg. Med. Chem. Lett., 2009, 19, 1799. [128] R. A. Love, O. Brodsky, M. J. Hickey, P. A. Wells and C. N. Cronin, J. Virol., 2009, 83, 4395. [129] P. L. Beaulieu, Exp. Opin. Ther. Pat., 2009, 19, 145. [130] E. Tramontano, Mini Rev. Med. Chem., 2008, 8, 1298. [131] U. Koch and F. Narjes, Curr. Topics Med. Chem., 2007, 7, 1302. [132] F. Pauwels, W. Mostmans, L. M. M. Quirynen, L. Helm, C. W. Boutton, A.-S. Rueff, E. Cleiren, P. Raboisson, D. Surleraux, O. Nyanguile and K. A. Simmen, J. Virol., 2007, 81, 6909. [133] J. Paeshuyse, I. Vliegen, L. Coelmont, P. Leyssen, O. Tabarrini, P. Herdewijn, H. Mittendorfer, J. Easmon, V. Cecchetti, R. Bartenschlager, G. Puerstinger and J. Neyts, Antimicrob. Agents Chemother., 2008, 52, 3433. [134] M. F. McCown, S. Rajyaguru, S. Pogam, S. Ali, W.-R. Siang, H. Kang, J. Symons, N. Cammack and I. Najera, Antimicrob. Agents Chemother., 2008, 52, 1604. [135] H. Ma, W.-R. Jiang, N. Robledo, V. Leveque, S. Ali, T. Lara-Jaime, M. Masjedizadeh, D. B. Smith, N. Cammack, K. Klumpp and J. Symons, J. Biol. Chem., 2007, 282, 29812. [136] E. Murakami, H. Bao, M. Ramesh, T. R. McBrayer, T. Whitaker, H. M. Micolochick Steuer, R. F. Schinazi, L. J. Stuyver, A. Obikhod, M. J. Otto and P. A. Furman, Antimicrob. Agents Chemother., 2007, 51, 503. [137] E. Murakami, C. Niu, H. Bao, H. M. Micolochick Steuer, T. Whitaker, T. Nachman, M. J. Sofia, P. Wang, M. J. Otto and P. A. Furman, Antimicrob. Agents Chemother., 2008, 52, 458. [138] P. A. Furman, P. Wang, C. Niu, D. Bao, W. Symonds, D. Nagarathnam, H. M. Steuer, S. Rakakonda, B. S. Ross, M. J. Otto and M. J. Sofia, Hepatology, 2008, 48(S1), 1161A, A1901. [139] M. Rodriguez-Torrez, J. Lazelari, E. J. Gane, E. DeJesus, D. R. Nelson, G. T. Everson, I. M. Jacobson, K. R. Reddy, J. G. McHutchson, A. Beard, S. Walker, W. Symonds and M. M. Berrey, Hepatology, 2008, 48(S1), 1160A, A1899.
434
Nicholas A. Meanwell et al.
[140] D. B. Smith, J. A. Martin, K. Klumpp, S. J. Baker, P. A. Blomgren, R. Devos, C. Granycome, J. Hang, C. J. Hobbs, W.-R. Jiang, C. Laxton, S. Le Pogam, V. Leveque, H. Ma, G. Maile, J. H. Merrett, A. Pichota, K. Sarma, M. Smith, S. Swallow, J. Symons, D. Vesey, I. Najera and N. Cammack, Bioorg. Med. Chem. Lett., 2007, 17, 2570. [141] P. Toniutto, C. Fabris, D. Bitetto, E. Fumolo, E. Fornasiere and M. Pirisi, IDrugs, 2008, 11, 738. [142] P. Toniutto, C. Fabris, D. Bitetto, E. Fornasiere, R. Rapetti and M. Pirisi, Curr. Opin. Investig. Drugs, 2007, 8, 150. [143] S. S. Carroll, S. Ludmerer, L. Handt, K. Koeplinger, N. Rena Zhang, D. Graham, M.-E. Davies, M. MacCoss, D. Hazuda and D. B. Olsen, Antimicrob. Agents Chemother., 2009, 53, 926. [144] Idenix press release, http://www.idenix.com/hepc/drug/ [145] A. Erhardt, K. Deterding, Y. Benhamou, M. Reiser, X. Forns, S. Pol, J. L. Calleja, S. Ross, H. C. Spangenberg, J. Garcia-Samaniego, M. Fuchs, J. Enrı´quez, J. Wiegand, J. Stern, K. Wu, G. Kukolj, M. Marquis, P. Beaulieu, G. Nehmiz and J. Steffgen, Antiviral Ther., 2009, 14, 23. [146] P. D. Patel, M. R. Patel, N. Kaushik-Basu and T. T. Talele, J. Chem. Inf. Model., 2008, 48, 42. [147] H. Cao, R. Cao, H. Zhang, X. Zheng and D. Gao, Curr. Med. Chem., 2008, 15, 1462. [148] S. Hirashima, T. Oka, K. Ikegashira, S. Noji, H. Yamanaka, Y. Hara, H. Goto, R. Mizojiri, Y. Niwa, T. Noguchi, I. Ando, S. Ikeda and H. Hashimoto, Bioorg. Med. Chem. Lett., 2007, 17, 3181. [149] I. Stansfield, M. Pompei, I. Conte, C. Ercolani, G. Migliaccio, M. Jairaj, C. Giuliano, M. Rowley and F. Narjes, Bioorg. Med. Chem. Lett., 2007, 17, 5143. [150] I. Stansfield, C. Ercolani, A. Mackay, I. Conte, M. Pompei, U. Koch, N. Gennari, C. Giuliano, M. Rowley and F. Narjes, Bioorg. Med. Chem. Lett., 2009, 19, 627. [151] J. Habermann, E. Capito, M. D. R. R. Ferreira, U. Koch and F. Narjes, Bioorg. Med. Chem. Lett., 2009, 19, 633. [152] D. Brainard, D. H. Wright, K. Sneddon, C. Cummings, P. Sun, J. Valentine, M. Anderson, S. Warrington, B. Sanderson, J. Chodakewitz and J. Wagner, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 937, http://www.abstractserver.com/easl2009/ planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 464&EASL2009 ¼ grnhh65l25p1sdrskl81trcq53&EASL2009 ¼ grnhh65l25p1sdrskl81trcq53 [153] R. De Francesco, G. Paonessa, D. Olsen, M. Rowley, B. Crescenzi, J. Habermann, F. Narjes and R. Laufer, HepDart, Lahaina, Hawaii, December 9–13, 2007, http:// www.informedhorizons.com/hepdart2007/pdf/hepdart07_presentations/Tues_07_ RDeFrancesco_HepDart_2007export.pdf [154] J. Pouliot, R. N. Pagila, K. Fung, G. Deng, G. Chenail, D. R. Latour, J. Zhang, S. KooMcCoy, M. Sivaraja, J. Bechtel, K. Lin, J. Gralapp, S. G. Alvarez, M. Leivers, U. Schmitz, T. Compton and R. Griffith, Hepatology, 2008, 48(S1), 1165A, A1911. [155] C. P. Bergstrom, J. A. Bender, R. G. Gentles, P. Hewawasam, T. W. Hudyma, J. F. Kadow, S. W. Martin, A. Regueiro-Ren, K.-S. Yeung, Y. Tu, K. A. Grant-Young and X. Zheng, WO Patent Application 2007/033175-A1, 2007. [156] T. W. Hudyma, X. Zheng, F. He, M. Ding, C. P. Bergstrom, P. Hewawasam, S. W. Martin and R. G. Gentles, WO Patent Application 2007/092000-A1, 2007. [157] N. A. Meanwell, R. G. Gentles, M. Ding, J. A. Bender, J. F. Kadow, P. Hewawasam, T. W. Hudyma and X. Zheng, WO Patent Application 2008/111978-A1, 2008. [158] R. G. Gentles, X. Zheng, M. Ding, Y. Tu, Y. Han, P. Hewawasam, J. F. Kadow, J. A. Bender, K.-S. Yeung, K. A. Grant-Young and T. W. Hudyma, WO Patent Application 2008/112473-A1, 2008. [159] J. A. Bender, R. G. Gentles, Y. Han, Y. Tu, Z. Yang, K.-S. Yeung and K. A. Grant-Young, WO Patent Application 2008/112841-A1, 2008.
Discovery and Development of Inhibitors of HCV
435
[160] R. G. Gentles, X. Zheng, M. Ding, Y. Tu, Y. Han, P. Hewawasam, J. F. Kadow, J. A. Bender, K.-S. Yeung, K. A. Grant-Young and T. W. Hudyma, WO Patent Application 2008/112848-A1, 2008. [161] K.-S. Yeung, J. A. Bender, R. G. Gentles, Z. Yang, M. Ding, Y. Tu, P. Hewawasam, Y. Han and J. F. Kadow, WO Patent Application 2008/112851-A1, 2008. [162] S. W. Martin, C. P. Bergstrom, R. G. Gentles and K.-S. Yeung, WO Patent Application 2009/029384-A2, 2009. [163] H. Li, J. Tatlock, A. Linton, J. Gonzalez, T. Jewell, L. Patel, S. Ludlum, M. Drowns, S. V. Rahvendran, H. Skor, R. Hunter, S. T. Shi, K. J. Herlihy, H. Parge, M. Hickey, X. Yu, F. Chau, J. Nonomiya and C. Lewis, J. Med. Chem., 2009, 50, 3969. [164] S. T. Shi, K. J. Herlihy, J. P. Graham, J. Nonomiya, S. V. Rahavendran, H. Skor, R. Irvine, S. Binford, J. Tatlock, H. Li, J. Gonzalez, A. Linton, A. K. Patick and C. Lewis, Antimicrob. Agents Chemother., 2009, 53, 2544. [165] P. Troke, M. Lewis, P. Simpson, E. van der Ryst, J. Hammond, C. Craig, M. Perros and M. Westby, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 968, http:// www.abstractserver.com/easl2009/planner/sp.php?go ¼ abstract&action ¼ abstract_ iplanner&absno ¼ 1795&EASL2009 ¼ n2c2gteknasbmv8149gnu3obj3&EASL2009 ¼ n2c2gteknasbmv8149gnu3obj3 [166] L. Chan Chung Kong, S. Kumar Das, C. G. Yannopoulos, G. Falardeau, L. Vaillancourt and R. Denis, WO Patent Application 2008/058393-A1, 2008. [167] O. Nicolas, I. Boivin, A. Berneche-D’Amours, P. Fex, F. Denis, S. Selliah and J. Bedard, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 961, http://www.abstractserver. com/easl2009/planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 943&EASL2009 ¼ 5qubar83ml3599p7d1vqhmcr21&EASL2009 ¼ 5qubar83ml3599p7 d1vqhmcr21 [168] C. Cooper, R. Larouche, B. Bourgault, N. Chauret and L. Proulx, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 940, http://www.abstractserver.com/easl2009/ planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 1504&EASL2009 ¼ lvfhdsnju362ukbm4c5sv96994&EASL2009 ¼ lvfhdsnju362ukbm4c5sv96994 [169] J. Bedard, O. Nicolas, D. Bilimoria, L. L’Heureux, P. Fex, M. David and L. Chan, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 935, http://www.abstractserver.com/easl2009/ planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 937&EASL2009 ¼ 5qubar83ml3599p7d1vqhmcr21&EASL2009 ¼ 5qubar83ml3599p7d1vqhmcr21 [170] N. Chauret, C. Chagnon-Labelle, M. Diallo, J. Laquerre, J. Laterreur, S. May and L. SteMarie, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 939, http://www.abstractserver. com/easl2009/planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 928&EASL2009 ¼ 2e4ntr9scpfap06bi90l8knv01&EASL2009 ¼ 2e4ntr9scpfap06bi90l8 knv01 [171] M. J. Slater, E. M. Amphlett, D. M. Andrews, G. Bravi, G. Burton, A. G. Cheasty, J. A. Corfield, M. R. Ellis, R. H. Fenwick, S. Fernandes, R. Guidetti, D. Haigh, C. D. Hartley, P. D. Howes, D. L. Jackson, R. L. Jarvest, V. L. H. Lovegrove, K. J. Medhurst, N. R. Parry, H. Price, P. Shah, O. M. P. Singh, R. Stocker, P. Thommes, C. Wilkinson and A. Wonacott, J. Med. Chem., 2007, 50, 897. [172] E. Lawitz, M. Rodriguez-Torres, M. DeMicco, T. Nguyen, E. Godofsky, J. Appleman, M. Rahimy, C. Crowley and J. Freddo, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 1055, http://www.abstractserver.com/easl2009/planner/sp.php?go ¼ abstract
436
[173]
[174]
[175]
[176]
[177]
[178]
[179]
[180]
[181]
[182]
Nicholas A. Meanwell et al.
&action ¼ abstract_iplanner&absno ¼ 2687&EASL2009 ¼ v90uv7ab1nsl3simolr28m6n u2&EASL2009 ¼ v90uv7ab1nsl3simolr28m6nu2 G. Koev, R. Mondal, J. Beyer, T. Reisch, S. Masse, W. Kati, D. Hutchinson, C. Flentge, J. Randolph, P. Donner, A. Krueger, R. Wagner, P. Yan, T. Lin, C. Maring and A. Molla, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 953, http://www.abstractserver. com/easl2009/planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 678&EASL2009 ¼ lam2cg9t20hujsupclpiietmt1&EASL2009 ¼ lam2cg9t20hujsupclpii etmt1 F. Ruebsam, C. V. Tran, L.-S. Li, S. H. Kim, A. X. Xiang, Y. Zhou, J. K. Blazel, Z. Sun, P. S. Dragovich, J. Zhao, H. M. McGuire, D. E. Murphy, M. T. Tran, N. Stankovic, D. A. Ellis, A. Gobbi, R. E. Showalter, S. E. Webber, A. M. Shah, M. Tsan, R. A. Patel, L. A. LeBrun, H. J. Hou, R. Kamran, M. V. Sergeeva, D. M. Bartkowski, T. G. Nolan, D. A. Norris and L. Kirkovsky, Bioorg. Med. Chem. Lett., 2009, 19, 451. R. Wagner, D. P. Larson, D. W. A. Beno, T. D. Bosse, J. F. Darbyshire, Y. Gao, B. D. Gates, W. He, R. F. Henry, L. E. Hernandez, D. K. Hutchinson, W. W. Jiang, W. M. Kati, L. L. Klein, G. Koev, W. A. Kolbrenner, A. C. Krueger, J. Liu, Y. Liu, M. A. Long, C. J. Maring, S. V. Masse, T. Middleton, D. A. Montgomery, J. K. Pratt, P. Stuart, A. Molla and D. J. Kempf, J. Med. Chem., 2009, 52, 1659. L. Lu, T. Dekhtyar, S. Masse, R. Pithawalla, P. Krishnan, W. He, T. Ng, G. Koev, K. Stewart, D. Larson, T. Bosse, R. Wagner, T. Pilot-Matias, H. Mo and A. Molla, Antiviral Res., 2007, 76, 93. C.-H. Chen, Y. He, L. Lu, H. B. Lim, R. L. Tripathi, T. Middleton, L. E. Hernandez, D. W. A. Beno, M. A. Long, W. M. Kati, T. D. Bosse, D. P. Larson, R. Wagner, R. E. Lanford, W. E. Kohlbrenner, D. J. Kempf, T. J. Pilot-Matias and A. Molla, Antimicrob. Agents Chemother., 2007, 51, 4290. C. Maring, R. Wagner, D. Hutchinson, C. Flentge, W. Kati, G. Koev, Y. Liu, D. Beno, J. Shen, Y. Y. Lau, Y. Gao, J. Fischer, S. Vaidyanathan, B. H. Lim, J. Beyer, R. Mondal and A. Molla, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 955, http://www.abstract server.com/easl2009/planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 681&EASL2009 ¼ lam2cg9t20hujsupclpiietmt1&EASL2009 ¼ lam2cg9t20 hujsupclpiietmt1 R. Menon, D. Cohen, A. Nada, E. Olson Dumas, Y.-L. Chiu, T. Podsadecki, W. Awni, B. Bernstein and C. Klein, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 956, http:// www.abstractserver.com/easl2009/planner/sp.php?go ¼ abstract&action ¼ abstract_ iplanner&absno ¼ 2046&EASL2009 ¼ lam2cg9t20hujsupclpiietmt1&EASL2009 ¼ lam 2cg9t20hujsupclpiietmt1 R. Menon, D. Cohen, A. Nada, Y.-L. Chiu, T. Podsadecki, W. Awni, B. Bernstein and C. Klein, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009, Abstract 957, http://www.abstractserver. com/easl2009/planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 2067&EASL2009 ¼ lam2cg9t20hujsupclpiietmt1&EASL2009 ¼ lam2cg9t20hujsupclpi ietmt1 N. M. Kneteman, A. Y. M. Howe, T. Gao, J. Lewis, D. Pevear, G. Lund, D. Douglas, D. F. Mercer, D. L. J. Tyrrell, F. Immermann, I. Chaudhary, J. Speth, S. A. Villano, J. O’Connell and M. Collett, Hepatology, 2009, 49, 745. A. Y. M. Howe, H. Cheng, S. Johann, S. Mullen, S. K. Chunduru, D. C. Young, J. Bard, R. Chopra, G. Krishnamurthy, T. Mansour and J. O’Connell, Antimicrob. Agents Chemother., 2008, 52, 3327.
Discovery and Development of Inhibitors of HCV
437
[183] M. Flint, S Mullen, A. M. Deatly, W. Chen, L. Z. Miller, R. Ralston, C. Broom, E. A. Emini and A. Y. M. Howe, Antimicrob. Agents Chemother., 2009, 53, 401. [184] A. Feldstein, D. Kleiner, D. Kravetz and M. Buck, J. Clin. Gastroenterol., 2009, 43, 374. [185] G. Puerstinger, J. Paeshuyse, S. Heinrich, J. Mohr, N. Schraffl, E. Clercq and J. Neyts, Bioorg. Med. Chem. Lett., 2007, 17, 5111. [186] I. Vliegen, J. Paeshuyse, E. Mabery, B. Peng, I.-H. Shih, L. S. Lehman, H. Dutartre, B. Selisko, B. Canard, S. Bondy, W. Tse, H. Reiser, E. De Clerq, W. A. Lee, G. Pu¨rstinger, W. Zhong and J. Neyts, Hepatology, 2007, 46(S1), 222A, A1385. [187] E. D. Dowdy, K. M. Kent, N. J. Tom and V. Zia, WO Patent Application 2009/009001-A1, 2009. [188] S. S. Bondy, T. C. Dahl, D. A. Oare, R. Oliyai, W. C. Tse and V. Zia, WO Patent Application 2008/005519-A2, 2008. [189] P. Poumbourios and H. E. Drummer, Antiviral Chem. Chemother., 2007, 18, 169. [190] J. M. Timpe and J. A. McKeating, Gut, 2008, 57, 1728. [191] J. Dubuisson, F. Helle and L. Cocquerel, Cell. Microbiol., 2008, 10, 821. [192] F. Helle and J. Dubuisson, Cell. Mol. Life Sci., 2008, 65, 100. [193] T. Hahn and C. M. Rice, J. Biol. Chem., 2008, 283, 3689. [194] S. Liu, W. Yang, L. Shen, J. R. Turner, C. B. Coyne and T. Wang, J. Virol., 2009, 83, 2011. [195] A. Ploss, M. J. Evans, V. A. Gaysinskaya, M. Panis, H. You, Y. P. Jong and C. M. Rice, Nature, 2009, 457, 882. [196] M. J. Evans, T. Hahn, D. M. Tscherne, A. J. Syder, M. Panis, B. Woelk, T. Hatziioannou, J. A. McKeating, P. D. Bieniasz and C. M. Rice, Nature, 2007, 446, 801. [197] M. R. Beard and F. J. Warner, Hepatology, 2007, 46, 277. [198] A. Zheng, F. Yuan, Y. Li, F. Zhu, P. Hou, J. Li, X. Song, M. Ding and H. Deng, J. Virol., 2007, 81, 12465. [199] L. Meertens, C. Bertaux, L. Cukierman, E. Cormier, D. Lavillette, F.-L. Cosset and T. Dragic, J. Virol., 2008, 82, 3555. [200] W. Yang, C. Qiu, N. Biswas, J. Jin, S. C. Watkins, R. C. Montelaro, C. B. Coyne and T. Wang, J. Biol. Chem., 2008, 283, 8643. [201] G. Koutsoudakis, E. Herrmann, S. Kallis, R. Bartenschlager and T. Pietschmann, J. Virol., 2007, 81, 588. [202] M. Brazzoli, A. Bianchi, S. Filippini, A. Weiner, Q. Zhu, M. Pizza and S. Crotta, J. Virol., 2008, 82, 8316. [203] J. Grove, T. Huby, Z. Stamataki, T. Vanwolleghem, P. Meuleman, M. Farquhar, A. Schwarz, M. Moreau, J. S. Owen, G. Leroux-Roels, P. Balfe and J. A. McKeating, J. Virol., 2007, 81, 3162. [204] Z. S. Jia, D. W. Du, Y. F. Lei, X. Wei, W. Yin, L. Ma, J. Q. Lian, P. Z. Wang, D. Li and Y. X. Zhou, J. Int. Med. Res., 2008, 36, 1319. [205] M. Regeard, M. Trotard, C. Lepere, P. Gripon and J. Seyec, J. Viral Hepat., 2008, 15, 865. [206] M. B. Zeisel, G. Koutsoudakis, E. K. Schnober, A. Haberstroh, H. E. Blum, F.-L. Cosset, T. Wakita, D. Jaeck, M. Doffoel, C. Royer, E. Soulier, E. Schvoerer, C. Schuster, F. StollKeller, R. Bartenschlager, T. Pietschmann, H. Barth and T. F. Baumert, Hepatology, 2007, 46, 1722. [207] S. B. Kapadia, H. Barth, T. Baumert, J. A. McKeating and F. V. Chisari, J. Virol., 2007, 81, 374. [208] H. J. Harris, M. J. Farquhar, C. J. Mee, C. Davis, G. M. Reynolds, A. Jennings, K. Hu, F. Yuan, H.-K. Deng, S. G. Hubscher, J. H. Han, P. Balfe and J. A. McKeating, J. Virol., 2008, 82, 5007. [209] P. Meuleman, J. Hesselgesser, M. Paulson, T. Vanwolleghem, I. Desombere, H. Reiser and G. Leroux-Roels, Hepatology, 2008, 48, 1761.
438
Nicholas A. Meanwell et al.
[210] J. Witteveldt, M. J. Evans, J. Bitzegeio, G. Koutsoudakis, A. M. Owsianka, A. G. N. Angus, Z.-Y. Keck, S. K. H. Foung, T. Pietschmann, C. M. Rice and A. H. Patel, J. Gen. Virol., 2009, 90, 48. [211] C. Schuster and T. F. Baumert, J. Hepatol., 2009, 50, 222. [212] C. Voisset, M. Lavie, F. Helle, A. Op De Beeck, A. Bilheu, J. Bertrand-Michel, F. Terce, L. Cocquerel, C. Wychowski, N. Vu-Dac and J. Dubuisson, Cell. Microbiol., 2008, 10, 606. [213] M. J. Farquhar, H. J. Harris, M. Diskar, S. Jones, C. J. Mee, S. U. Nielsen, C. L. Brimacombe, S. Molina, G. L. Toms, P. Maurel, J. Howl, F. W. Herberg, S. C. D. IJzendoorn, P. Balfe and J. A. McKeating, J. Virol., 2008, 82, 8797. [214] W. Yang, B. L. Hood, S. L. Chadwick, S. Liu, S. C. Watkins, G. Luo, T. P. Conrads and T. Wang, Hepatology, 2008, 48, 1396. [215] M. T. Catanese, R. Graziani, T. Hahn, M. Moreau, T. Huby, G. Paonessa, C. Santini, A. Luzzago, C. M. Rice, R. Cortese, A. Vitelli and A. Nicosai, J. Virol., 2007, 81, 8063. [216] K. Murao, H. Imachi, X. Yu, W. M. Cao, T. Nishiuchi, K. Chen, J. Li, R. A. M. Ahmed, N. C. W. Wong and T. Ishida, Gut, 2008, 57, 664. [217] E. Falkowska, F. Kajumo, E. Garcia, J. Reinus and T. Dragic, J. Virol., 2007, 81, 8072. [218] F. Helle, A. Goffard, V. Morel, G. Duverlie, J. McKeating, Z.-Y. Keck, S. Foung, F. Penin, J. Dubuisson and C. Voisset, J. Virol., 2007, 81, 8101. [219] B. Bartosch and F.-L. Cosset, Methods Mol. Biol., 2009, 510, 279. [220] J.-P. Yang, D. Zhou and F. Wong-Staal, Methods Mol. Biol., 2009, 510, 295. [221] E.-I. Pecheur, D. Lavillette, F. Alcaras, J. Molle, Y. S. Boriskin, M. Roberts, F.-L. Cosset and S. J. Polyak, Biochemistry, 2007, 46, 6050. [222] M. Holzer, S. Ziegler, A. Neugebauer, B. Kronenberger, C. D. Klein and R. W. Hartmann, Arch. Pharm. Chem. Life Sci., 2008, 341, 478. [223] M. Holzer, S. Ziegler, B. Albrecht, B. Kronenberger, A. Kaul, R. Bartenschlager, L. Kattner, C. D. Klein and R. W. Hartmann, Molecules, 2008, 13, 1081. [224] G. Coburn, A. Q. Han, J. de Muys, C. Gauss, K. Provoncha, M. Canfield, D. Paul, S. Mohamed, S. Moorji, D. Fisch, J. D. Murga, Y. Rotshteyn, D. Qian, P. J. Maddon and W. C. Olson, Hepatology, 2008, 48(S1), 1162A, A1904. [225] iTherX, press release Febraury 3, 2009, http://www.itherx.com/press.html [226] T. J. Cuthbertson, M. Ibanez, C. A. Rijnbrand, A. J. Jackson, G. K. Mittapalli, F. Zhao, J. E. MacDonald and F. Wong-Staal, WO Patent Application 2008/021745-A2, 2008. [227] H. Ueno, T. Shimada, K. Aoyagi, S. Katoh, H. Shinkai, T. Motomura, Y. Komoda, T. Otsubaki, Y. Soejima and I. Kawahara, WO Patent Application 2007/058392 A1, 2007. [228] J. de Bruijne, J. Bergmann, C. Weegink, K. van Nieuwkerk, R. de Knegt, J. van de Wetering de Rooij, A. van Vliet, R. Molenkamp, J. Schinkel, H. Reesink and H. Janssen, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009. Abstract 822, http://www.kenes.com/ easl2009/posters/Abstract822.htm [229] C. S. Fraser and J. A. Doudna, Nat. Rev. Microbiol., 2007, 5, 29. [230] P. J. Lukavsky, Virus Res., 2009, 139, 166. [231] S. M. Dibrov, H. Johnston-Cox, Y.-H. Weng and T. Hermann, Angew. Chem. Intl. Ed., 2007, 46, 226. [232] Q. Zhao, Q. Han, C. R. Kissinger, T. Hermann and P. A. Thompson, Acta Crystallogr. D Biol. Crystallogr., 2008, D64, 436. [233] C. Alotte, A. Martin, S. A. Caldarelli, A. Di Giorgio, R. Condom, F. Zoulim, D. Durantel and O. Hantz, Antiviral Res., 2008, 80, 280. [234] C. Romero-Lopez, R. Diaz-Gonzalez and A. Berzal-Herranz, Cell. Mol. Life Sci., 2007, 64, 2994. [235] K. Konno, S. Fujita, M. Iizuka, S. Nishikawa, T. Hasegawa and K. Fukuda, Nucleic Acids Symp. Ser., 2008, 52, 493.
Discovery and Development of Inhibitors of HCV
439
[236] K. Konno, S. Nishikawa, T. Hasegawa and K. Fukuda, Nucleic Acids Symp. Ser., 2007, 51, 393. [237] C. Chevalier, A. Saulnier, Y. Benureau, D. Flechet, D. Delgrange, F. Colbere-Garapin, C. Wychowski and A. Martin, Mol. Ther., 2007, 15, 1452. [238] V. Vlassov, B. Korba, K. Farrar, S. Mukerjee, A. A. Seyhan, H. Ilves, R. L. Kaspar, D. Leake, S. A. Kazakov and B. H Johnston, Oligonucleotides, 2007, 17, 223. [239] T. Kanda, R. Steele, R. Ray and R. B. Ray, J. Virol., 2007, 81, 669. [240] V. Guerniou, R. Gillet, F. Berree, B. Carboni and B. Felden, Nucleic Acids Res., 2007, 35, 6778. [241] G. Godeau, C. Staedel and P. Barthelemy, J. Med. Chem., 2008, 51, 4374. [242] A. Litovchick and J. W. Szostak, Proc. Natl. Acad. Sci. U.S.A., 2008, 105, 1529. [243] A. Bhattacharyya, C. R. Trotta and S. W. Peltz, Drug Discov. Today, 2007, 12, 553. [244] R. Di and N. E. Tumer, WO Patent Application 2008/088875-A2, 2008. [245] G. M. Karp, WO Patent Application 2007/084413-A2, 2007. [246] G. M. Karp, P. S. Hwang, J. J. Takasugi, H. Ren, R. G. Wilde, A. A. Turpoff, A. Arefolov, G. Chen and J. A. Campbell, WO Patent Application 2007/084435-A2, 2007. [247] F. C. Lahser and G. M. Karp, WO Patent Application 2007/106317-A2, 2007. [248] H. Huang, F. Sun, D. M. Owen, W. Li, Y. Chen, M. Gale and J. Ye, Proc. Natl. Acad. Sci. U.S.A., 2007, 104, 5848. [249] V. Icard, O. Diaz, C. Scholtes, L. Perrin-Cocon, C Ramie`re, R. Bartenschlager, F. Penin, V. Lotteau and P. Andre´, PLoS ONE, 2009, 4, e4233. [250] Y. Nahmias, J. Goldwasser, M. Casali, D. Poll, T. Wakita, R. T. Chung and M. L. Yarmush, Hepatology, 2008, 47, 1437. [251] D. Durantel, C. Alotte and F. Zoulim, Curr. Opin. Investig. Drugs, 2007, 8, 125. [252] C. Chapel, C. Garcia, B. Bartosch, P. Roingeard, N. Zitzmann, F.-L. Cosset, J. Dubuisson, R. A. Dwek, C. Trepo, F. Zoulim and D. Durantel, J. Gen. Virol., 2007, 88, 1133. [253] R. Flisiak, J.-M. Dumont and R. Crabbe´, Exp. Opin. Investig. Drugs, 2007, 16, 1345. [254] K. Watashi and K. Shimotohno, Rev. Med. Virol., 2007, 17, 245. [255] K. Moriishi and Y. Matsuura, Rev. Med. Virol., 2007, 17, 343. [256] F. Yang, J. M. Robotham, H. B. Nelson, A. Irsigler, R. Kenworthy and H. Tang, J. Virol., 2008, 82, 5269. [257] S. Chinnaswamy, I. Yarbrough, S. Palaninathan, C. T. R. Kumar, V. Vijayaraghavan, B. Demeler, S. M. Lemon, J. C. Sacchettini and C. C. Kao, J. Biol. Chem., 2008, 283, 20535. [258] M. A. El-Farrash, H. H. Aly, K. Watashi, M. Hijikata, H. Egawa and K. Shimotohno, Microbiol. Immunol., 2007, 51, 127. [259] J. E. Mathy, S. Ma, T. Compton and K. Lin, Antimicrob. Agents Chemother., 2008, 52, 3267. [260] R. Crabbe´, G. Vuagniaux, J.-M. Dumont, V. Nicolas-Metral, J. Marfurt and L. Novaroli, Exp. Opin. Investig. Drugs, 2009, 18, 211–220. [261] L. Coelmont, S. Kaptein, J. Paeshuyse, I. Vliegen, J.-M. Dumont, G. Vuagniaux and J. Neyts, Antimicrob. Agents Chemother., 2009, 53, 967. [262] R. G. Ptak, P. A. Gallay, D. Jochmans, A. P. Halestrap, U. T. Ruegg, L. A. Pallansch, M.D. Bobardt, M.-P. Be´thune, J. Neyts, E. Clercq, J.-M. Dumont, P. Scalfaro, K. Besseghir, R. M. Wenger and B. Rosenwirth, Antimicrob. Agents Chemother., 2008, 52, 1302. [263] S. Hopkins, B. Scorneaux, S. M. Mosier, Z. Huang, M. G. Murray and R. R. Harris, Hepatology, 2008, 48(S1), 1117A, A1814. [264] M. Peel, R. Harris, Z. Huang, S. Hopkins, K. Li, M. Peel, T. E. Richardson, B. Scorneaux, A. Scribner and S. Wring, Hepatology, 2008, 48(S1), 1167A, A1915. [265] J. M. Robida, H. B. Nelson, Z. Liu and H. Tang, J. Virol., 2007, 81, 5829.
440
Nicholas A. Meanwell et al.
[266] F. Fernandes, D. S. Poole, S. Hoover, R. Middleton, A.-C. Andrei, J. Gerstner and R. Striker, Hepatology, 2007, 46, 1026. [267] R. Flisiak, A. Horban, P. Gallay, M. Bobardt, S. Selvarajah, A. Wiercinska-Drapalo, E. Siwak, I. Cielniak, J. Higersberger, J. Kierkus, C. Aeschlimann, P. Grosgurin, V. Nicolas-Me´tral, J.-M. Dumont, H. Porchet, R. Crabbe´ and P. Scalfaro, Hepatology, 2008, 47, 817. [268] R. Flisiak, S. V. Feinman, M. Jablowski, A. Horban, W. Kryczka, M. Pawlowska, J. E. Heathcote, G. Mazzella, C. Vandeli, V. Nicolas-Me´tral, P. Grosgurin, J. S. Liz, P. Scalfaro, H. Porchet and R. Crabbe´, Hepatology, 2009, 49, 1460. [269] S. Hopkins, D. Heuman, E. Gavis, J. Lalezari, E. Glutzer, B. DiMassimo, P. Rusnak, S. Wring, C. Smitley and Y. Ribeill, 44th Annual Meeting of the European Association for the Study of the Liver, Copenhagen, Denmark, April 22–26, 2009. Abstract 90, http://www.abstractserver.com/easl2009/planner/sp.php?go ¼ abstract&action ¼ abstract_iplanner&absno ¼ 2400&EASL2009 ¼ plp2pf2nk7dajqfdc38tkjf1l2&EASL 2009 ¼ plp2pf2nk7dajqfdc38tkjf1l2 [270] V. R. Anderson and M. P. Curran, Drugs, 2007, 67, 1947. [271] J. F. Rossignol and E. B. Keeffe, Future Microbiol., 2008, 3, 539. [272] B. E. Korba, A. B. Montero, K. Farrar, K. Gaye, S. Mukerjee, M. S. Ayers and J.-F. Rossignol, Antiviral Res., 2008, 77, 56. [273] B. E. Korba, M. Elazar, P. Lui, J.-F. Rossignol and J. S. Glenn, Antimicrob. Agents Chemother., 2008, 52, 4069. [274] M. Elazar, M. Liu, S. McKenna, P. Liu, E. A. Gehrig, A. Elfert, J. Puglisi, J.-F. Rossignol and J. S. Glenn, Hepatology, 2008, 48(S1), 1151A, A1881. [275] J.-F. Rossignol, S. M. Kabil, Y. El-Gohary, A. Elfert and E. B. Keeffe, Aliment. Pharmacol. Ther., 2008, 28, 574. [276] J.-F. Rossignol, A. Elfert, Y. El-Gohary and E. B. Keeffe, Gastroenterol., 2009, 136, 856. [277] J. E. Semple and J.-F. Rossignol, WO Patent Application 2009/035788-A1, 2009. [278] R. Gazak, D. Walterova and V. Kren, Curr. Med. Chem., 2007, 14, 315. [279] M. Y. Parmar and T. R. Gandhi, Phcog. Rev., 2008, 2, 102. [280] S. J. Polyak, C. Morishima, M. C. Shuhart, C. C. Wang, Y. Liu and D. Y.-W. Lee, Gastroenterol., 2007, 132, 1925. [281] A. Melhem, M. Stern, O. Shibolet, E. Israeli, Z. Ackerman, O. Pappo, N. Hemed, M. Rowe, H. Ohana, G. Zabrecky, R. Cohen and Y. Ilan, J. Clin. Gastroenterol., 2005, 39, 737. [282] E. Gabbay, E. Zigmond, O. Pappo, N. Hemed, M. Rowe, G. Zabrecky, R. Cohen and Y. Ilan, World J. Gastroenterol., 2007, 13, 5317. [283] P. Ferenci, T.-M. Scherzer, H. Kerschner, K. Rutter, S. Beinhardt, H. Hofer, M. Schoeniger-Hekele, H. Holzmann and P. Steindl-Munda, Gastroenterol., 2008, 135, 1561.
CHAPT ER
21 Inhibitors of Nuclear Hormone Receptor/Coactivator Interactions Terry W. Moore and John A. Katzenellenbogen
Contents
1. Introduction 2. Coregulators of Nuclear Hormone Receptors 3. Coactivator Binding Inhibitors 3.1 Peptidic coactivator binding inhibitors 3.2 Small-molecule coactivator binding inhibitors 4. Summary References
443 445 445 445 449 455 455
1. INTRODUCTION Nuclear hormone receptors (NRs) function as transcription factors, and many NRs are activated by endogenous hormones (e.g., estrogens, androgens, and thyroid hormones) as well as by synthetic analogs of these hormones. Upon binding of an agonist to a NR, several structural and functional changes occur; typically, these include receptor dimerization, nuclear translocation, and DNA binding at gene regulatory sites. Another important change is a conformational restructuring, described more fully later, which allows the NR to interact with its associated coregulatory proteins, both coactivators and corepressors. The NR– coregulator complex then modifies chromatin structure and either aids in (coactivates) or blocks (corepresses) the recruitment of the basal transcription machinery for RNA polymerase II (RNA pol II), which Department of Chemistry, University of Illinois at Urbana-Champaign, Roger Adams Laboratory, 600 S. Mathews Ave., Urbana, IL 61801, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04421-2
r 2009 Elsevier Inc. All rights reserved.
443
444
Terry W. Moore and John A. Katzenellenbogen
leads to alterations in the level of gene expression. Although there are 48 members of the human NR superfamily, there are nearly 300 recognized coregulators [1,2]. In addition, coregulatory proteins have various enzymatic activities that can themselves be regulated by post-translational modifications. This complexity, combined with NR dimerization partners and tissue-specific expression, serves to mediate the vast array of physiological functions and tissue specificity of the NRs. Activating or inhibiting the NRs has profound physiological consequences, and, as a result, the NRs have been exploited as therapeutic targets across a number of diseases and physiological states. NR modulators currently used as therapeutic antagonists, such as the breast cancer drug tamoxifen, 1, and the prostate cancer drug flutamide, 2, bind to the ligand-binding pocket of the NR and induce a conformation of the receptor that either disfavors coactivator binding or favors corepressor binding. The inhibitory mechanism of these antagonists is indirect or allosteric, because ligand binding at one site affects protein–protein interaction at a second site. This conventional antagonist therapy is not without drawbacks, however, because many diseases and physiological states affected by an NR often become refractory to the effects of conventional antagonists, examples being the resistance to tamoxifen and flutamide that develops in breast and prostate cancer, respectively. Because it is well-established that blocking the recruitment of coactivators to the NR through this indirect, conventional antagonist mechanism is sufficient to block the transcription of NR-regulated genes, it is logical to assume that directly blocking the NR–coactivator interaction would yield comparable or perhaps superior results and could provide a second line of defense when resistance to conventional antagonist therapies develops. Me N Me O
Me F3C
H N Me O
H
Me 1
O2N 2
Inhibitors of Nuclear Hormone Receptor/Coactivator Interactions
445
Coactivator binding inhibitors (CBIs) are molecules that block the interaction of NRs with coactivators by a direct, competitive mechanism. Such compounds are not without precedent in the literature, and both peptidic and small-molecule CBIs have been reported. From a discovery viewpoint, peptidic CBIs can yield much information about the requirements for disrupting the NR/coactivator interaction, but from a practical standpoint, small-molecule CBIs that would avoid the liabilities of peptides, such as cellular impermeability and metabolic instability, are preferred. Historically, interrupting protein–protein interactions with small molecules has posed significant challenges in drug discovery; however, there have been numerous recent reports of small molecules inhibiting the interaction of two proteins both specifically and potently [3–5]. Herein we outline recent advances in developing NR CBIs.
2. COREGULATORS OF NUCLEAR HORMONE RECEPTORS The NR coregulators constitute a large family of proteins, diverse in size and function, the best known of which are the coactivators of the p160 class (SRC1, 2, and 3) and the corepressors NCoR-1 and SMRT, although there are many others [2]. Additionally, those mentioned earlier are known to interact with other, non-NR transcription factors, a reflection of the complexity of the interactome [6,7]. The coregulator sequence elements, typically LXXLL motifs (NR boxes) in coactivators [8,9] and LXXXIXXXL motifs (corepressor nuclear receptor [CoRNR] boxes) in corepressors [10,11], occur over two and three turns of an a-helix, respectively, and make contacts with the NR by placing the denoted nonpolar residues in a short but deep hydrophobic groove formed by residues from helices 3, 4, 5, 11, and 12 of the NR ligand-binding domain, often termed the activation function 2 (AF2) region. The p160 coactivators, when bound to NR-ligand complexes, function as scaffolds onto which a series of large multi-protein complexes assemble, leading to chromatin modifications, the recruitment of RNA polymerase II, and alteration of gene expression.
3. COACTIVATOR BINDING INHIBITORS 3.1 Peptidic coactivator binding inhibitors Although many studies have focused on disrupting the NR–coactivator interaction with peptides derived from endogenous coactivators, there have been relatively few studies of unnatural or modified peptides that perturb this interaction. Because these latter studies delineate
446
Terry W. Moore and John A. Katzenellenbogen
opportunity space not accessed by the natural coactivators, they can suggest designs for CBIs that might exhibit increased potency and selectivity.
3.1.1 Phage display peptides Various peptides generated by phage-display were shown to bind to estrogen receptor a (ERa) and/or ERb, in the presence of 17b-estradiol (E2), tamoxifen, or in the absence of any ligand [12,13]. In particular, the peptide a/bI (SSNHQSSRLIELLSR) was found to interact with either ERa or ERb in the presence of E2. Interestingly, these probes, obtained from a very large (W109 phage) random library, contain an LXXLL motif, further establishing this as an important consensus sequence for the ER/coactivator interaction. Another peptide, aII (SSLTSRDFGSWYASR), bound to ERa, but not ERb, in the presence of either tamoxifen or E2. Subsequent structural studies showed that this peptide did not bind at the coactivator binding groove, but consistent with the ability of both E2 and tamoxifen to recruit the probe, at a previously unknown binding site on the opposite side of the receptor [14]. A final class of peptide probes, a/bV ([S/M]X[D/E][W/F][W/F]XXXL), was found to bind to ERa or ERb in the presence of tamoxifen, but not E2; however, based on competition experiments, these peptides were believed to bind at the AF2 region, unlike aII. After further rounds of focused phage display, another peptide (BT1) was found to bind to the tamoxifen–ER complex with a higher potency, particularly when helix 12 was removed from ER [15]. The H12 sequence (LYDLLLEML) is a pseudo-CoRNR box motif believed responsible for the poor recruitment of corepressors by antagonist-liganded ER in reconstituted in vitro systems, because it competes with their binding to AF2; deleting this sequence allows these putative corepressor mimics to bind to ER [16]. The sequence of BT1 (ELFDAFQLRQLILRGLQDDIPYH) is also reminiscent of the CoRNR box consensus motif (LXXXIXXXL), so it is believed that a/bV and BT1 are peptidomimetics of corepressors. The respective co-crystal structures of each compound bound to a hydroxytamoxifen–ERa complex, expressed without H12, show that the peptides bind in the coactivator binding area, evidence that corepressors and coactivators bind at overlapping areas of the ER ligand-binding domain [16]. This same approach has also been used to find non-natural peptidic inhibitors of other NRs [17–21].
3.1.2 Synthetic peptides Beyond studies on the random peptides generated by phage display, efforts have been made to rationally design short peptides that might be more potent in inhibiting the NR/SRC interaction. Approaches have involved introduction of constraints on the peptide to enforce the
Inhibitors of Nuclear Hormone Receptor/Coactivator Interactions
447
a-helical structure of the coactivator, as well as incorporation of nonnatural amino acids. A macrolactam constraint between E691 and K695 (Ac-EKHKILcyclo(E691RLLK695)DS-OH) of the coactivator SRC2 (also known as GRIP1) gave a peptide that was 15 times more potent than the unconstrained peptide in disrupting the interaction between thyroid hormone receptor b (TRb) and a peptide derived from NR Box 2 of the coactivator SRC2 [22]. The IC50 value of 0.62 mM for the cyclized peptide was determined by a fluorescence polarization assay. The same constraint (cyclo(E691–K695)), used together with incorporation of non-natural amino acids at the L690, L693, or L694 positions [23,24], gave peptides that were more potent and showed selectivity among three different NRs (ERa, ERb, and TRb). The SRCs themselves bear flanking residues that impart some selectivity, but this study demonstrated that a great degree of selectivity among NRs can be obtained simply by manipulating the side chains of the LXXLL motif itself. These peptides were most often selective for ERa (12 of 37), with one (L690W) showing a W600-fold selectivity for ERa (0.144 mM); one was ERb-selective (0.824 mM; o-Cl-phenylalanine at L693); and one was TRb selective (7.02 mM; p-F-phenylglycine at L690). ERa appeared more capable of binding large and a-branched substituents. Because docking suggested that cooperativity could not be achieved by mutating more than one residue, compounds containing multiple mutations were not examined. This constrained-peptide library was also used to study differential agonist ligand-dependent effects (E2, diethylstilbestrol, and genistein) on peptide recruitment to the AF2 region of ERa and ERb [25]. On the basis of analysis of crystal structures, the authors note differences in the size and shape of the grooves of TRb, ERb, and ERa: a ridge between the leucine-binding subpockets is more substantial in TRb than in ERa, and extra space unique to each NR is also found. For instance, in the ˚ 3 not available in TRb/SRC2-2/triiodothyronine complex, there is 268 A 3 ˚ ERa/SRC2-2/E2, and in ERa there is 357 A of space that is not available in TRb. Lack of available structures made the comparison across other NRs/ligand pairs more difficult. Of 37 inhibitors, 19 were more than 10-fold selective for ERa over ERb when liganded with E2; fifteen and six were ERa-selective when liganded with diethylstilbestrol and genistein, respectively. Only two peptides were selective for ERb, one each for E2 and genistein. Generally, ERa can accommodate replacement of the leucines with aromatic amino acids when an appropriately folded helical scaffold is used. Interestingly, experimental results reveal aspects of binding that the computational studies could not predict; this was attributed to the plasticity of the surface of ERa, which can readily expose new subsites for binding.
448
Terry W. Moore and John A. Katzenellenbogen
In a related report, a panel of 32 unconstrained 20 mers with sequences from NR boxes from various coactivators was assayed against TRb bound with three different ligands. The most potent was the SRC2-2 box sequence bound to a triiodothyronine-liganded TRb (0.7 mM). Most interestingly, ligand structure did influence which coregulator sequences were preferentially recruited, implying that the NR conformation induced by each agonist is sufficiently distinct to target a different coregulator/NR complex. This is likely a molecular determinant of the well-known tissue-selective pharmacology of the NRs. In a related approach, modified peptides were constrained by incorporation of a disulfide bond at various positions. Using a fluorescence-based assay [26], the researchers found that the sequence H-Lys-cyclo(D-Cys-IleLeu-Cys)-Arg-Leu-Leu-Gln–NH2 bound to E2-ERa with an IC50 of 25 nM, but about 15 times more weakly to E2-ERb (390 nM). An X-ray crystal structure of this peptide, referred to as peptidomimetic ER modulator 1 (PERM-1), showed it bound to the AF2 region of ERa [27]. Other disulfideand amide-linked peptides did not bind as well (high nanomolar to micromolar IC50s), but all peptides reported were between 2.4- and 64-fold selective for ERa over ERb. Building on this approach, disulfide-constrained peptides having exquisite affinities for ERa were synthesized, the most potent of which (H-Arg-cyclo(D-Cys-Ile-Leu-Cys)-Arg-Npg-Leu-Gln-NH2, where Npg ¼ neopentylglycine) exhibited a Ki of 0.07 nM for ERa and 1.2 nM for ERb [28]. Almost all peptides published in this report were selective for ERa, but one was fivefold selective for ERb (H-Arg-cyclo(D-Cys-IleLeu-Cys)-Arg-tLeu-Leu-Gln-NH2, where tLeu is t-butylglycine): ERa Ki/ERb Ki ¼ 7/1.2 nM. In a finding consistent with the crystal structure [29], the authors assert that L2 of the LXXLL motif seems less important for binding to the receptor than are L1 and L3, based on the observation that Npg in the L2 position binds much more tightly than Leu (0.07 nM vs. 11 nM) but that Npg would not make any more contacts with the receptor than would Leu (the third methyl points toward the solvent). It is hypothesized that the extra methyl group in Npg stabilizes the conformation through a hydrophobic interaction with Ile (i1). Thus, L2 is more important for stabilizing the helix than for interacting with the receptor. The disulfide linkage in these peptides would likely subvert their use in the reducing environment of the cell. To address this issue, heterodimeric peptides were synthesized that were more likely to be cell permeable by appending, through a disulfide bond, a group that might facilitate uptake (i.e., a decanoyl or hepta-Arg group) [30]. Inside the cell, the disulfide bond would be cleaved, leaving the active monomeric peptide, a compound having a Ki of 60 nM on ERa (1850 nM on ERb) [28]. In a related approach, the stabilizing disulfide of another constrained peptide (H-Arg-cyclo(D-Cys-Ile-Leu-Cys)-Arg-Leu-Leu-Gln–NH2; ERa
Inhibitors of Nuclear Hormone Receptor/Coactivator Interactions
449
Ki/ERb Ki ¼ 11/77 nM) was replaced with a thioether linkage (ERa Ki/ERb Ki ¼ 6.9/64 nM), so that if it were cell-permeable, it would not be reduced by excess glutathione present in the cell [31]. Owing to the most unfortunate, premature death of the principal investigator, Arno Spatola, no further studies have been conducted to date to see whether peptides prepared by either of these approaches have activity in cell-based assays.
3.2 Small-molecule coactivator binding inhibitors Peptidic CBIs are useful as high-affinity probes for investigating NR action, but small molecule CBIs will likely be required for any therapeutic use. There are, however, a number of hurdles that must be overcome in developing high-affinity small-molecule CBIs: Small molecules are typically much smaller than the peptides thus far studied, and their smaller footprint can be an intrinsic disadvantage when inhibiting a high-affinity protein–protein interaction that spans a large surface area. The residues flanking the LXXLL sequence in the coactivator proteins also bolster potency in interaction with NRs; it would be difficult for a small molecule to reach these more distant interaction sites. The LXXLL sequence shared among the three SRCs and other coactivators is used for interaction with many, perhaps all, of the NRs. Therefore, issues regarding the selectivity that a small molecule might have for a particular NR are a concern: Are the slight differences in the coactivator interaction sites of different NRs sufficient so that receptor-selective small-molecule CBIs can be developed? A significant practical challenge in assaying small molecules for NR CBI activity is the need to demonstrate that they have little to no affinity for the ligand-binding pocket of the NR; if they do, they might displace the agonist and block coactivator binding indirectly by functioning as conventional antagonists. Fortunately, with most of the small-molecule CBIs described later, relatively simple control experiments have been used to distinguish between inhibition of coactivator binding by direct and indirect means. Because there are few sites for hydrogen bonding within the coactivator binding region of receptors, small molecules must rely largely on van der Waals and other hydrophobic interactions; they must, however, also maintain sufficient aqueous solubility. Because most high-throughput screening libraries are highly enriched in low molecular weight (i.e., o500 Da) compounds, following the Lipinski specifications for drug-like properties, finding even modest affinity hits for blocking a protein–protein interaction from a highthroughput screen of such libraries — a popular modus operandi for hit identification — can be difficult [32].
450
Terry W. Moore and John A. Katzenellenbogen
Although daunting, these challenges are not impossible to address, and significant work has been done to undertake these and other problems inherent in small molecule CBIs. Published small-molecule CBIs are listed below by NR.
3.2.1 Estrogen receptor The first non-peptidic CBIs for any of the NRs were reported in 2004 [33]. These tri-substituted pyrimidines, although exhibiting only modest affinities (KiB30 mM in a fluorescence polarization assay), provided evidence that small molecules could disrupt the ERa/SRC1 Box 2 interaction without binding in the ligand-binding pocket of the receptor. Each of the iso-propyl-terminated alkyl groups in 3 is thought to mimic one of the three leucines of the LXXLL motif. A library of compounds has been generated around this initial hit, producing compounds that show activity in a time-resolved fluorescence resonance energy transfer (TR-FRET) assay and in an ERa-mediated transcription assay using luciferase as a reporter gene. The structure– activity relationship (SAR) around this hit shows that: (a) substituents at the 2- and 4-positions of the pyrimidine ring need to be linked through an N-atom, and there is a preference for smaller substituents (e.g., isobutyl) at the 2-position; (b) substituents can be linked to the pyrimidine ring at the 6-position through either carbon or nitrogen (but not sulfur or oxygen); (c) the CBI binding pocket in ERa can accommodate no more than two phenyl substituents and no more than one naphthyl substituent on the tri-substituted pyrimidine; and (d) to satisfy what are likely either hydrogen bonding or electrostatic requirements, one of the substituents must be linked through an –NHlinkage (not –NMe-). The best of these compounds have inhibition constants of 2–3 mM [34]. The guanylhydrazone 4, also described in 2004, inhibited the interaction of ERa or ERb (but not progesterone receptor) with SRC1, 2, and 3 with IC50 values B25 mM in an ELISA; moreover, the compound was active in a mammalian two-hybrid assay (Gal4 DNA-binding domain/human ERa ligand-binding domain fusion and SRC1, SRC3, or SRC3/VP16 fusion), exhibiting an IC50 of 5.5 mM [35]. At 10–20 mM, 4 was found to inhibit expression of the ERa-controlled gene pS2 in the ERpositive MCF-7 breast cancer cell line, although cell death occurred at concentrations W20 mM by an unknown, presumably ER-independent mechanism. The best of a small library of guanylhydrazone analogs had IC50 values in reporter gene and mammalian two-hybrid assays of 0.9 and 4.6 mM, respectively [36]. Observations from this study are that: (a) the most potent compounds possessed a chloro substituent at the 1-position; (b) substitution at the phenyl ring had little effect; (c) the second ring was not required (i.e., acyclic compounds showed good
451
Inhibitors of Nuclear Hormone Receptor/Coactivator Interactions
activity), but, if present, it must be saturated; and (d) a ketonic hydrazone, rather than an aldehydic one, was completely inactive.
Me
Me
Me
Me O
HN HN Cl
N Me
N
N H
Me
3
Me
N
NHCOR
Me
H N
NH2
CO2H
Me
NH
Me
Me
4
5
Me
N H2N Me
NH2
Me
N
Me
Me O NH2
OH
6
7
In a TR-FRET assay, a series of bicyclo-[2.2.2]-octanes developed using structure-based design were only partial antagonists of the ERa/SRC3 interaction, with Kis ranging from 7 (i.e., bicyclooctane 5) to 40 M [37]. Given the great structural similarity between the LXXLL motif and the modeled bicyclooctane 5, the weak activity of these molecules illustrates that the precise placement of substituents mimicking the LXXLL leucine residues is certainly not the sole determinant of CBI activity. In a fluorescence polarization assay, the pyridylpyridone 6 was found to inhibit the interaction of ERa/SRC1 NR Box 2 peptide with a Ki of 4.2 mM [38]. A crystal structure of a related analog overlaid with the three substituents of the LXXLL motif shows how 6 is predicted to bind to ER. A series of amphipathic benzenes (e.g., 7) displaying alternating hydrophobic and hydrophilic substituents produces facially amphipathic
452
Terry W. Moore and John A. Katzenellenbogen
molecules that mimic the nonpolar leucine residues of the LXXLL motif on one face of the benzene ring and have polar groups to interact with water on the other [39,40]. Aminoethyl and pentyl groups gave the best combination of polar and hydrophobic substituents, respectively, for inhibiting the ERa/SRC3 protein fragment interaction in a TR-FRET assay (IC50 ¼ 1.7 mM) and the ERa/SRC protein interaction in reporter gene (IC50 ¼ 3.2 mM) and mammalian two-hybrid assays (IC50 ¼ 3.2 mM). In a surprising finding, two molecules of the selective ER modulator hydroxytamoxifen 8 were found in an ERb crystal structure, one in the ligand-binding pocket, but the other bound at the AF2 surface [41]. The unsubstituted phenyl of 8 was buried deep within the coactivatorbinding groove, but no polar interactions to keep the inhibitor in place were evident. Although this crystal structure is consistent with previous biochemical studies suggesting a second binding site for 8 [42], it is not clear whether this second-site binding occurs under physiological conditions.
3.2.2 Thyroid hormone receptor In a high-throughput screen of 138,000 compounds, some Mannich bases (i.e., 9) were found to inhibit the interaction of TRb with a peptide fragment from SRC2 NR Box 2 in a fluorescence polarization assay (IC50 ¼ 2 mM) [43]. The compounds eliminate amines, giving a,bunsaturated ketones that react with cysteine sulfhydryl groups on the TRb surface [44]. A crystal structure showing enone 10 bound noncovalently to the surface of TRb suggests that after elimination of dimethylamine, the compound binds irreversibly with the S-H group ˚ from the electrophilic center. of either C298 or C308, which are B7 A Mutation, fragmentation, and pull-down studies suggest that upon alkylation at C298, both coactivator and corepressor recruitment are irreparably impaired. Through further SAR studies, it was found that p-hexyl or p-heptyl substituents on the core hydrophobic ring were essential for good activity, with highly reactive enones (i.e., lacking large b-substituents) being most potent [45]. Other thiol-reactive electrophilic groups showed similar activity (e.g., a-haloketones, keto-epoxides, b-bromoketones). These changes, however, did not significantly improve compound potency. When assayed against a TR-positive thyroid cancer cell line (ARO) and a TR-negative osteosarcoma epithelial cell line (U2OS), no clear trends in cytotoxicity were seen. Thus, the compounds might also be working at targets other than the TRs. Interestingly, in in vitro assays, all compounds showed up to 18-fold selectivity for TRb over TRa.
Inhibitors of Nuclear Hormone Receptor/Coactivator Interactions
453
Me N Me Me
O R
N
8
Me
Me
O
O
10
9 Me
HO
3.2.3 Androgen receptor After four off-patent drugs were discovered as hits in a highthroughput fluorescence polarization screen for compounds that block androgen receptor (AR)/SRC 2 Box 3 interaction, an X-ray screen of a smaller subset of compounds yielded crystal structures of seven molecules (flufenamic acid, 11; triiodothyronine, T3, 12; Triac, 13; 14; 15; 2-methylindole; and indole-3-carboxylic acid) bound to a previously unrecognized surface of AR [46]. This ‘‘Binding Function 3’’ (BF3) region on AR is near the junction of H1, the H3–H5 loop, and H9 and is adjacent to and nearly as large as AF2. The binding of these molecules to the BF3 region causes a slight restructuring of residues within both BF3 and AF2 regions, and a major repositioning of Arg726 at the AF2 boundary, which induces an AF2 conformation that disfavors SRC binding. Although the in vitro potencies of these molecules are quite modest (i.e., 50 mM), they are active as inhibitors in AR-regulated reporter gene assays at 10–30 mM. Also, although their mode of action is fundamentally different from that of the other CBIs mentioned earlier (i.e., not direct inhibition but allosteric inhibition originating from a site different from the AR ligand-binding pocket), their functional outcomes are similar. Some of these ligands (i.e., T3 and Triac) would cross-react by binding strongly to the ligand-binding pocket of TR; the competing activity of others on their established targets could also be problematic.
HO2C
I H N
CF3 HO2C
11
NH2
I O
I
I
OH
12
HO2C
O
I
I
OH
13
454
Terry W. Moore and John A. Katzenellenbogen
Me NH2
NH2 Me
N
OH
N
N
N
N
N
N
N Me
Me
Me Me
Me
14
Me
15
3.2.4 Pregnane X receptor (or steroid xenobiotic receptor) Xenobiotic activation of the pregnane X receptor (PXR) is known to upregulate genes involved in metabolism, often including metabolism of the inducing xenobiotic itself. Therefore, inhibiting this interaction could enhance the pharmacokinetic profile of a drug that is inducing its own degradation. The antifungal ketoconazole (16) has been reported to inhibit both the PXR/SRC1 and the constitutive androstene receptor (CAR)/SRC1 interactions [47], and it has been reported that the phytoestrogen coumestrol (17) also inhibits the PXR/SRC1 interaction [48]. Ketoconazole exhibited an IC50 of 74 mM in a pulldown assay and similar potency in a mammalian two-hybrid assay. Ketoconazole also delayed the metabolism of the anesthetic tribromoethanol in mice. The Ki of coumestrol in a fluorescence polarization assay (PXR ligand-binding domain/SRC1 peptide) was 1.2–1.3 mM, but this number is believed to be high because of poor compound solubility; lower coumestrol concentrations (i.e., 25 mM) antagonized rifampicin-activated PXR in mammalian two-hybrid assays. Mutational studies suggest that coumestrol is binding to a surface on PXR outside of the ligand-binding pocket. Coumestrol is also an agonist of PXR at 13 mM. O
Cl N
N
O
Cl
O
OH
Me
O
O N N
16
HO O O
17
Inhibitors of Nuclear Hormone Receptor/Coactivator Interactions
455
4. SUMMARY Directly blocking the interaction of NRs and their coactivators could have profound implications in treating a number of different NR-regulated diseases. The work done thus far in developing these CBIs has focused on both peptidic and small-molecule inhibitors. Many peptidic CBIs are potent, and seemingly selective, but, despite work to enhance their cellular uptake and stability, they have not been active in cellular assays. Small-molecule CBIs, thus far, are not as potent as their peptidic counterparts, but many are active in cell-based assays of transcription and gene expression at low micromolar concentrations. Although more work remains to be done, these results provide a promising starting point for developing potent and selective small-molecule CBIs that could provide a valuable and necessary alternative approach to the modulation of NR activity.
REFERENCES ˚ . Gustafsson and V. Laudet, Nat. Rev. Drug Discov., 2004, 3, 950. [1] H. Gronemeyer, J-A [2] D. M. Lonard, R. B. Lanz and B. W. O’Malley, Endocr. Rev., 2007, 28, 575. [3] H. Yin, G. Lee, K. A. Sedey, O. Kutzki, H. S. Park, B. P. Orner, J. T. Ernst, H.-G. Wang, S. M. Sebti and A. D. Hamilton, J. Am. Chem. Soc., 2005, 127, 10191. [4] H. Yin, G. Lee, H. S. Park, G. A. Payne, J. M. Rodriguez, S. M. Sebti and A. D. Hamilton, Angew. Chem. Int. Ed., 2005, 44, 2704. [5] L. T. Vassilev, B. T. Vu, B. Graves, D. Carvajal, F. Podlaski, Z. Filipovic, N. Kong, U. Kammlott, C. Lukacs, C. Klein, N. Fotouhi and E. A. Liu, Science, 2004, 303, 844. [6] S-K. Lee, H-J. Kim, S-Y. Na, T. S. Kim, H-S. Choi, S.-Y. Im and J. W. Lee, J. Biol. Chem., 1998, 273, 16651. [7] S. Werbajh, I. Nojek, R. Lanz and M. A. Costas, FEBS Lett., 2000, 485, 195. [8] D. M. Heery, E. Kalkhoven, S. Hoare and M. G. Parker, Nature, 1997, 387, 733. [9] H. Y. Mak, S. Hoare, P. M. A. Henttu and M. G. Parker, Mol. Cell. Biol., 1999, 19, 3895. [10] J. D. Chen and R. M. Evans, Nature, 1995, 377, 454. [11] A. J. Ho¨rlein, A. M. Na¨a¨r, T. Heinzel, J. Torchia, B. Gloss, R. Kurokawa, A. Ryan, Y. Kamei, M. So¨derstro¨m, C. K. Glass and M. G. Rosenfeld, Nature, 1995, 377, 397. [12] J. D. Norris, L. A. Paige, D. J. Christensen, C.-Y. Chang, M. R. Huacani, D. Fan, P. T. Hamilton, D. M. Fowlkes and D. P. McDonnell, Science, 1999, 285, 744. [13] L. A. Paige, D. J. Christensen, H. Grfn, J. D. Norris, E. B. Gottlin, K. M. Padilla, C.-Y. Chang, L. M. Ballas, P. T. Hamilton, D. P. McDonnell and D. M. Fowlkes, Proc. Natl. Acad. Sci. U.S.A., 1999, 96, 3999. ˚ . Gustafsson, E. Treuter, R. E. Hubbard and A. C. W. Pike, [14] E. H. Kong, N. Heldring, J.-A Proc. Natl. Acad. Sci. U.S.A., 2005, 102, 3593. [15] H.-J. Huang, J. D. Norris and D. P. McDonnell, Mol. Endocrinol., 2002, 16, 1778. ˚ . Gustafsson and A. C. W. Pike, [16] N. Heldring, T. Pawson, D. McDonnell, E. Treuter, J.-A J. Biol. Chem., 2007, 282, 10449. [17] C. Chang, J. Abdo, T. Hartney and D. P. McDonnell, Mol. Endocrinol., 2005, 19, 2478.
456
Terry W. Moore and John A. Katzenellenbogen
[18] C.-Y. Chang, J. D. Norris, H. Grfn, L. A. Paige, P. T. Hamilton, D. J. Kenan, D. Fowlkes and D. P. McDonnell, Mol. Cell. Biol., 1999, 19, 8226. [19] J. M. Hall, C. Chang and D. P. McDonnell, Mol. Endocrinol., 2000, 14, 2010. [20] N. B. Mettu, T. B. Stanley, M. A. Dwyer, M. S. Jansen, J. E. Allen, J. M. Hall and D. P. McDonnell, Mol. Endocrinol., 2007, 21, 2361. [21] J. W. Pike, P. Pathrose, O. Barmina, C.-Y. Chang, D. P. McDonnell, H. Yamamoto and N. K. Shevde, J. Cell. Biochem., 2003, 88, 252. [22] T. R. Geistlinger and R. K. Guy, J. Am. Chem. Soc., 2001, 123, 1525. [23] T. R. Geistlinger and R. K. Guy, J. Am. Chem. Soc., 2003, 125, 6852. [24] T. R. Geistlinger and R. Kiplin Guy, in Methods in Enzymology (eds D. W. Russell and D. J. Mangelsdorf), Vol. 364, Elsevier, San Diego, 2003, p. 223. [25] T. R. Geistlinger, A. C. McReynolds and R. K. Guy, Chem. Biol., 2004, 11, 273. [26] K. S. Bramlett, Y. F. Wu and T. P. Burris, Mol. Endocrinol., 2001, 15, 909. [27] A-M. Leduc, J. O. Trent, J. L. Wittliff, K. S. Bramlett, S. L. Briggs, N. Y. Chirgadze, Y. Wang, T. P. Burris and A. F. Spatola, Proc. Natl. Acad. Sci. U.S.A., 2003, 100, 11273. [28] A. K. Galande, K. S. Bramlett, J. O. Trent, T. P. Burris, J. L. Wittliff and A. F. Spatola, Chembiochem, 2005, 6, 1991. [29] A. C. W. Pike, A. M. Brzozowski and R. E. Hubbard, J. Steroid Biochem. Mol. Biol., 2000, 74, 261. [30] A. K. Galande and A. F. Spatola, Org. Lett., 2003, 5, 3431. [31] A. K. Galande, K. S. Bramlett, T. P. Burris, J. L. Wittliff and A. F. Spatola, J. Pept. Res., 2004, 63, 297. [32] J. A. Wells and C. L. McClendon, Nature, 2007, 450, 1001. [33] A. L. Rodriguez, A. Tamrazi, M. L. Collins and J. A. Katzenellenbogen, J. Med. Chem., 2004, 47, 600. [34] A. A. Parent, J. R. Gunther and J. A. Katzenellenbogen, J. Med. Chem., 2008, 51, 6512. [35] D. Shao, T. J. Berrodin, E. Manas, D. Hauze, R. Powers, A. Bapat, D. Gonder, R. C. Winneker and D. E. Frail, J. Steroid Biochem. Mol. Biol., 2004, 88, 351. [36] A. L. LaFrate, J. R. Gunther, K. E. Carlson and J. A. Katzenellenbogen, Bioorg. Med. Chem., 2008, 16, 10075. [37] H.-B. Zhou, M. L. Collins, J. R. Gunther, J. S. Comninos and J. A. Katzenellenbogen, Bioorg. Med. Chem. Lett., 2007, 17, 4118. [38] J. Becerril and A. D. Hamilton, Angew. Chem. Int. Ed., 2007, 46, 4471. [39] J. R. Gunther, T. W. Moore, M. L. Collins and J. A. Katzenellenbogen, ACS Chem. Biol., 2008, 3, 282. [40] K. V. Kilway and J. S. Siegel, Tetrahedron, 2001, 57, 3615. [41] Y. Wang, N. Y. Chirgadze, S. L. Briggs, S. Khan, E. V. Jensen and T. P. Burris, Proc. Natl. Acad. Sci. U.S.A., 2006, 103, 9908. [42] Y. Berthois, M. Pons, C. Dussert, A. C. de Paulet and P. M. Martin, Mol. Cell. Endocrinol., 1994, 99, 259. [43] L. A. Arnold, E. Este´banez-Perpin˜a´, M. Togashi, N. Jouravel, A. Shelat, A. C. McReynolds, E. Mar, P. Nguyen, J. D. Baxter, R. J. Fletterick, P. Webb and R. K. Guy, J. Biol. Chem., 2005, 280, 43048. [44] E. Este´banez-Perpin˜a´, L. A. Arnold, N. Jouravel, M. Togashi, J. Blethrow, E. Mar, P. Nguyen, K. J. Phillips, J. D. Baxter, P. Webb, R. K. Guy and R. J. Fletterick, Mol. Endocrinol., 2007, 21, 2919. [45] L. A. Arnold, A. Kosinski, E. Este´banez-Perpin˜a´, R. J. Fletterick and R. K. Guy, J. Med. Chem., 2007, 50, 5269.
Inhibitors of Nuclear Hormone Receptor/Coactivator Interactions
457
[46] E. Este´banez-Perpin˜a´, A. A. Arnold, P. Nguyen, E. D. Rodrigues, E. Mar, R. Bateman, P. Pallai, K. M. Shokat, J. D. Baxter, R. K. Guy, P. Webb and R. J. Fletterick, Proc. Natl. Acad. Sci. U.S.A., 2007, 104, 16074. [47] H. Huang, H. Wang, M. Sinz, M. Zoeckler, J. Staudinger, M. R. Redinbo, D. G. Teotico, J. Locker, G. V. Kalpana and S. Mani, Oncogene, 2007, 26, 258. [48] H. Wang, H. Li, L. B. Moore, M. D. L. Johnson, J. M. Maglich, B. Goodwin, O. R. R. Ittoop, B. Wisely, K. Creech, D. J. Parks, J. L. Collins, T. M. Willson, G. V. Kalpana, M. Venkatesh, W. Xie, S. Y. Cho, J. Roboz, M. Redinbo, J. T. Moore and S. Mani, Mol. Endocrinol., 2008, 22, 838.
CHAPT ER
22 Safety Testing of Drug Metabolites Thomas N. Thompson
Contents
1. Introduction 1.1 Importance of metabolites in safety and efficacy 1.2 Brief chronology of events leading up to issuance of guidance 2. Evolution of the MIST Guidance 2.1 The MIST issues are defined and debated 2.2 Key points of the 2005 draft guidance 2.3 Areas of concern after issuance of 2005 draft guidance 2.4 Issuance of the final MIST guidance 3. Potential Issues Related to Implementation of a Sound MIST Strategy 4. Implications of Metabolism in Safety Testing of New Drugs 5. Strategy for Implementation of Best Practices 5.1 What metabolites should be measured? 5.2 What kind of information is needed and when is it needed? 6. Role of the Medicinal Chemist 7. Summary References Appendix: Decision Tree Flow Diagram
459 459 460 461 461 462 463 463 464 466 467 467 468 471 472 472 474
R&D Services Pharma Consulting, 663N. 132nd St. #126, Omaha, NE 68154, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04422-4
r 2009 Elsevier Inc. All rights reserved.
459
460
Thomas N. Thompson
1. INTRODUCTION 1.1 Importance of metabolites in safety and efficacy In the course of modern drug discovery and development, the clear mandate for introducing new drugs is that they be both effective and safe. As such, the safety evaluation program is of paramount importance in the development of any new drug. Indeed, the observance of toxicity in animal tests is a leading cause of drug withdrawal. Safety evaluation of potential new drugs has historically focused on the unchanged form of that drug in various species under varying experimental protocols. Metabolites have not been tested per se because they were assumed to be present during animal testing, and any toxicity attributable to metabolites was accounted for in the overall safety profile of the parent. In addition, the ability to understand the routes and extent of metabolism early enough in development to allow for proper safety evaluation of metabolites has been limited in the past by available analytical methods. With the advent of new and powerful analytical and biochemical techniques, the potential toxicity of metabolites in the overall profile of a drug candidate can now be better considered. In principle, the logic to consider safety testing of metabolites is compelling. After all, metabolites represent new chemicals to which a human is exposed [1]. Administration of a new drug to animals or humans sometimes results in no new metabolites, but more often, there is generation of one to a few to over a dozen metabolites. The concern is that 1) some of these metabolites could have inherent toxicity; 2) they might be formed in humans; and 3) they would not be generated at appreciable levels in animals during toxicity testing. Under a worst case scenario, a metabolite-related toxicity occurs in humans but not in animal species used in non-clinical safety evaluation [2]. Even though the role of metabolites as mediators of toxicity has not always been considered in safety assessment, there have been numerous examples of drugs (e.g., acetaminophen, mephenytoin) that have been known for decades to give rise to toxic metabolites [3]. Indeed, it is often the case that when a drug shows an unexpected off-target adverse effect in clinical trials, there is speculation that the toxicity is attributable to a metabolite. It was against this backdrop that scientists in both the drug industry and the United States Food and Drug Administration (FDA) came to conclude that some form of risk assessment of metabolites is both prudent and necessary. Yet, despite recognizing its importance, significant concerns were raised as to how a policy would be implemented. Some of these concerns were 1) the significant additional cost and development time to implement metabolite studies, and 2) whether such
Safety Testing of Drug Metabolites
461
a program would truly be useful enough in avoiding human toxicity to justify the increased time and cost. On a more practical level, other questions were raised: 1) What degree of metabolism would be the threshold for concern and how would it be measured, for example, by a fraction of circulating drug related material, or by total mass basis? 2) Would these regulations apply to both reactive and stable metabolites, which have different mechanisms of toxicity (e.g., covalent binding in the case of the former versus excessive on or off target pharmacology in the case of the latter)? [2]; 3) What would be the timing of these metabolism studies? and 4) For ‘‘major’’ active metabolites, what toxicity studies would be appropriate and would administration of preformed metabolites really produce meaningful safety data relative to when the metabolites are formed in situ [4]?
1.2 Brief chronology of events leading up to issuance of guidance The initial drug industry position paper on the subject of Metabolites in Safety Testing (MIST) was published by representatives of the Pharmaceutical Research and Manufacturers of America (PhRMA) in 2002 in Toxicology and Applied Pharmacology (TAP) [5]. The FDA response to the PhRMA position paper and the industry group’s response to FDA’s comments were published in 2003 as letters to the editor in TAP [6,7]. In June 2005, the FDA issued their Draft Guidance on Safety Testing of Drug Metabolites issued and requested comments, a compilation of which was completed in August 2005. In December 2005, representatives from PhRMA and FDA attended a joint DruSafe discussion to share feedback on the draft guidance and industry comments. This was followed by several years of ad hoc commentary in the literature by scientists from the industry and FDA [2–4,8–10] and at other meetings such as an open forum sponsored by the Analysis and Pharmaceutical Quality (APQ) section of the American Association of Pharmaceutical Scientists (AAPS) entitled ‘‘MIST and beyond: the role of bioanalysis in the assessment of drug metabolites in safety testing,’’ following the 2006 AAPS annual meeting to further discuss the proposed guidelines. The discussion and commentary culminated in February 2008 when the FDA issued the final version of the ‘‘Guidance for Industry: Safety Testing of Drug Metabolites’’ (herein referred to as the ‘‘MIST Guidance’’) [11]. However, despite the issuance of a final MIST Guidance, the debate and commentary continues [12–14] as the drug industry and the FDA discuss how best to balance the safety concerns potentially posed by metabolites with the enormous strategic and financial costs involved in
462
Thomas N. Thompson
safety testing. In November 2008, in a reprise of the 2006 meeting, the AAPS presented a short course entitled ‘‘Metabolites Data as a Requisite for Regulatory Submissions: Current Practice, Challenges and Case Studies.’’ Finally, as recently as February 2009, Chemical Research in Toxicology devoted an issue largely to commentary, interpretation of the guidance and new methodology on how to best address the crucial issues in the MIST Guidance.
2. EVOLUTION OF THE MIST GUIDANCE 2.1 The MIST issues are defined and debated The initial industry position, as outlined in the 2002 MIST document [5], was intended to be a proactive attempt to provide a framework for when to consider MIST of new molecular entities. Within this framework, a logical sequence of decision points and laboratory studies was outlined that would both address important scientific issues as well as offer a strategy for when and where to direct resources to the problem. One of the main goals was to define a process that was scientifically driven, not merely a rules- or check-box-based list of studies [14]. In essence, the MIST document suggested that if one or more circulating metabolites (i.e., metabolites found in plasma, not excreta) were identified in humans in the definitive radiolabeled Absorption/ Distribution/Metabolism/Excretion (ADME) study, which were major, active, and either unique or disproportionate, then separate toxicology studies that assess the potential adverse effects of such metabolite(s) should be considered. Moreover, such metabolite(s) should be monitored using a validated bioanalytical assay in these special toxicology studies. In this document, ‘‘major’’ human metabolites were defined as those which constituted W25% drug-related material in the human ADME study, whereas unique or disproportionate metabolites are those found to be absent in animals or only present at relatively low concentrations. In addition, it was proposed that the structure of the metabolite should be taken into account in decision making. For example, metabolites that possessed a moiety with a known structural alert should also be considered for evaluation even if they are below the 25% threshold. On the contrary, phase II conjugates are generally not pharmacologically active or chemically reactive, so safety evaluation would not normally be needed (acyl glucuronides being an exception). In addition, the MIST document gave a thorough list of points to consider regarding separate geno-, general, and reproductive toxicity testing of metabolites. Regarding the subject of carcinogenicity testing, the point was made that if carcinogenicity testing is not necessary for
Safety Testing of Drug Metabolites
463
approval of the parent compound, then testing of the metabolite likewise should not be required.
2.2 Key points of the 2005 draft guidance In the initial draft guidance issued in June 2005, FDA was in essential agreement with the industry’s position on many if not most points. However, there were key areas of difference. With regard to the definition of ‘‘major,’’ the draft guidance made the point that 10% of administered dose or systemic exposure, whichever is less, should be the action point rather than the 25% mentioned in the MIST document. The draft guidance advocated the 10% level ostensibly to be consistent with other guidelines and guidances currently used by United States and foreign regulatory agencies. In addition, they cited several cases in which it has been determined that the toxicity of a drug could be attributed to one or more metabolites present at less than 25% of the administered dose, including halothane, felbamate, cyclophosphamide, and acetaminophen. In general, the draft guidance recommended that in vitro-in vivo metabolite correlation studies be conducted as ‘‘early’’ as possible in drug development, both to identify potential species differences and, should qualitative interspecies differences in metabolism be detected, to assist in the selection of the appropriate animal species for toxicological assessments. If the metabolite profile in animals is qualitatively similar and animals form the metabolites in equal or greater amounts than those formed in humans, then the standard toxicology testing program for parent drug is adequate. However, if humans form the metabolite(s) exclusively, or to a greater extent than animals, then other considerations apply. If the unique or disproportionate human metabolite is o10%, then again, the standard toxicology testing program for the parent drug will apply. However, if the human metabolite is W10%, then special toxicity testing is warranted, including a ‘‘bridge’’ general toxicity study, and evaluation of genotoxicity and reprotoxicity. Depending on the genotoxicity results, discrete carcinogenicity testing may be warranted for some metabolites.
2.3 Areas of concern after issuance of 2005 draft guidance After the issuance of the draft guidance and a period for public comments, months of commentary at meetings and in the literature followed. Perhaps the most significant and recurring comment had to do with the definition of ‘‘major’’ and whether it should be based on absolute abundance rather than relative abundance as specified in the draft guidance [8,10]. Another significant point raised was that when a toxicity study on a metabolite is specifically warranted, complications
464
Thomas N. Thompson
may arise due to potential differences in disposition between exogenously and endogenously formed metabolites, which could render the resultant data difficult to interpret [4]. It was also claimed that metabolite levels in excreta should not be relied upon to accurately reflect systemic exposure. Because sensitive analytical methods exist today that allow quantitation in drug-related material in plasma, only circulating metabolites should be of concern and need safety assessment. Other miscellaneous comments included 1) confusion over definition of ‘‘unique’’ and ‘‘major’’; 2) what is meant by saying human ADME studies should be done ‘‘early’’? 3) should the ‘‘free’’ (e.g., unbound to plasma proteins) or total (bound plus unbound to plasma proteins) metabolite concentrations be considered? and 4) should known quantitative structure–activity relationships (QSAR) for reactive metabolites (aka ‘‘structural alerts’’) be considered?
2.4 Issuance of the final MIST guidance In February 2008, nearly 3 years after the draft guidance was issued, the final MIST Guidance was published [11]. The term ‘‘disproportionate’’ drug metabolite was defined in the guidance as a metabolite present only in humans or present at higher plasma concentrations in humans than in the animals used in non-clinical studies. Consistent with the 2005 draft, a ‘‘major’’ metabolite was still defined as one with W10% systemic exposure with the added proviso that exposure should be determined at steady state. Generally speaking, only stable metabolites can be detected in circulation. If reactive metabolites are detected, they may require special consideration, although a strategy for reactive metabolites was not specifically addressed in the MIST Guidance. If at least one of the two species used in toxicology studies of parent drug forms the metabolite at levels that approach the expected human concentrations, then additional toxicity tests are not necessary. However, if exposure to the metabolite in animals is lacking entirely, or it does not form at levels that approach expected human exposure, then additional toxicity tests may be required. This can be accomplished either by administration of parent drug to a species that forms the metabolite or, if this is not possible, by direct administration of the metabolite itself. It should be emphasized that the steps outlined in the MIST Guidance are recommendations, not requirements. The agency encourages active dialogue on a case-by-case basis, especially if the new drug candidate is intended for serious or life-threatening diseases that lack an approved effective therapy. A schema for metabolite testing is shown in the Appendix.
Safety Testing of Drug Metabolites
465
3. POTENTIAL ISSUES RELATED TO IMPLEMENTATION OF A SOUND MIST STRATEGY Regarding the issue of safety testing of drug metabolites, it is the patient’s safety and well-being that are of paramount importance to the drug industry, to the FDA and to all parties involved in drug discovery and development. Accordingly, the MIST Guidance as written charts a conservative action plan and invites pharmaceutical companies into dialogue on a ‘‘case by case’’ basis. Thus, it is very important that the sponsor company provides context regarding the contribution of metabolite(s) in the overall toxicity of their drug candidate. Given the formidable technical capabilities available today, this ownership will ensure an action plan that includes the most appropriate studies, not just those that are technically feasible. It is important to ensure not only the safety of important new medicines but also that the time and resources required to fulfill regulations add value proportionate to their cost [2,9,12]. Taken at face value, the MIST Guidance may appear to call for extensive efforts that may or may not significantly improve human safety of drugs. Obach [14] and others have opined that the regulations unintentionally may have created ‘‘conundrums’’ that both pharmaceutical companies and the FDA must still sort out. These issues will continue to challenge us to better understand the mechanisms of toxicity and to develop better testing procedures and technology to address them. Some of the more challenging issues raised by the MIST Guidelines include the following: 1)
2)
3)
4)
Very potent compounds are a special challenge because they are administered at very low doses. In that case, metabolites that meet the definition of ‘‘major’’ might be present at very low concentrations and not actually cause toxicity owing to their relatively low body burden. Extensively metabolized compounds are also a special challenge. In this case, metabolites in circulation might be ‘‘major’’ relative to parent drug and meet the definition in the MIST Guidance criteria, but in fact represent a negligible percentage of total drug-related material. The requirement to measure metabolites at steady state is technically very challenging as definitive human radiolabeled studies are not normally conducted under steady-state conditions [14]. Reactive metabolites, while mentioned only in passing in the MIST Guidance, can frequently contribute to toxicity [14]. Metabolites in excreta may indicate a body-burden of chemically reactive intermediary metabolites, which can result in toxicity through complex,
466
5)
Thomas N. Thompson
non-specific mechanisms commonly associated with ‘‘covalent binding’’ (e.g., carcinogenicity, immunoallergic response) [10]. In those situations where toxicity studies on specific metabolites are warranted, can we assume that administration of synthetic (e.g., preformed) metabolite to animals will give us meaningful information on the safety of that compound as it is formed in vivo? Several authors have questioned whether such studies are reliable indicators of true toxicity of metabolites [1,4,15–17]. For example, what if new secondary metabolites are formed after direct administration of the metabolite of interest, but these secondary metabolites are not formed in vivo after administration of the parent? If new target organ toxicity ensues from this situation, how do we put that into perspective [1,4]? Pang and others [15–17]have demonstrated that kinetic differences can be observed between metabolites formed in vivo and those that are synthesized and then orally administered.
These are but some of the vexing issues that remain as we attempt to comply with the MIST Guidance. Some insightful new strategies and new technologies have been and are being developed to meet these challenges head on and will be discussed subsequently in this review.
4. IMPLICATIONS OF METABOLISM IN SAFETY TESTING OF NEW DRUGS A circulating metabolite that is observed in human clinical studies but is either absent or only present in animals at very low concentrations can make extrapolation of safety in animals to humans quite complex. Under these circumstances, the 2008 MIST Guidance specifies that either a new species must be identified which forms the metabolite or the metabolite must be administered directly to animals in separate toxicity studies. The MIST Guidance offers the following points to consider when designing such studies [11]. First, the drug metabolite for testing must be not only synthesized but also characterized under good laboratory practice (GLP) conditions as would any new chemical entity destined for GLP toxicology studies. The ensuing toxicology studies must also be conducted under GLP conditions. In planning for these studies, the following factors should also be considered:
Similarity of the metabolite to the parent molecule. Pharmacological or chemical class. Solubility. Stability in stomach pH.
Safety Testing of Drug Metabolites
467
Phase I versus phase II metabolite. Relative amounts detected in humans versus the amounts detected in animals. The following types of studies are all part of a standard battery of toxicology tests. The MIST Guidance has commented on each of these as they may apply to safety testing of metabolites. General toxicity studies. The preferred route of administration would be the same as the parent drug’s intended clinical route of administration. However, if necessary, alternate routes can be used to achieve sufficient exposure to the disproportionate metabolite. The metabolite should be administered at multiples of the human exposure or at least at levels comparable to those measured in humans. The duration of the studies should follow appropriate International Committee on Harmonization (ICH) guidelines. As in all GLP studies, ‘‘it is crucial to gather toxicokinetic data from this type of study using a validated bioanalytical method to ensure adequate exposure.’’ Genotoxicity studies. ‘‘The potential genotoxicity of the drug metabolite should be assessed in an in vitro assay that detects point mutations and in another assay that detects chromosomal aberrations.’’ Embryo-fetal development toxicity studies. ‘‘When a drug is intended for use in a population that includes women of childbearing potential, embryo-fetal development toxicity studies should be performed with the drug metabolite.’’ Carcinogenicity studies. ‘‘Carcinogenicity studies should be conducted on metabolites of drugs that are administered continuously for at least 6 months, or that are used intermittently in the treatment of chronic or recurrent conditions when the carcinogenic potential of the metabolite cannot be adequately evaluated from carcinogenicity studies conducted with the parent drug.’’ It is also important to note that the FDA is willing to negotiate with the sponsor as to the exact plan of studies to expedite drug development for serious or life-threatening diseases other than cancer (e.g., amyotrophic lateral sclerosis (ALS), stroke, human immunodeficiency virus) for those drugs with major beneficial therapeutic advances, and are intended for illnesses that lack an approved effective therapy.
5. STRATEGY FOR IMPLEMENTATION OF BEST PRACTICES 5.1 What metabolites should be measured? Although metabolites that comprise ‘‘10% of systemic exposure at steady state’’ is the action threshold specified in MIST Guidance, it is really up to the sponsor to make a case if this number is not scientifically warranted.
468
Thomas N. Thompson
As was pointed out earlier, there are many instances where ‘‘10%’’ does not make sense based on the particular circumstance, and some of these were discussed in Section 3. FDA has said in both the MIST Guidance and public presentations that it is willing to enter a dialogue with sponsors on a ‘‘case by case’’ basis. It is worth noting here that the MIST Guidance offers no specific recommendations on the subject of reactive metabolites, even though the cases of metabolite-induced toxicity cited by the FDA as justification for the ‘‘10% of systemic exposure’’ action level in their 2005 draft guidance are due to reactive rather than stable metabolites [3,6,14]. Furthermore, in a recent review, the authors considered 14 drugs removed from the market because of human toxicity in which metabolites may have played a role. Of these 14 cases, 11 likely involved reactive metabolites whereas only three involved stable metabolites [10]. Clearly, more work is needed to define an appropriate action plan to evaluate toxicity of potentially reactive metabolites, which may be held to a higher standard of safety than a typical major human metabolite [18]. The issue of reactive metabolites poses a formidable technical challenge. As was pointed out earlier, these metabolites are not going to be detected in plasma as they most likely will not leave the organ in which they were formed, often the liver. In other cases, the presence of a reactive metabolite can be inferred from stable secondary metabolites (e.g., glutathione conjugates) that have been excreted in urine/feces [18]. The following table from a recent review on chemical mechanisms of toxicity lists some of the common structural moieties that are alerts to the possibility of reactive metabolites [19]. Hydrazines and hydrazides Arylacetic or aryl propionic acids Thiophenes, furans, pyrroles Anilines and anilides (Structures that yield) quinones and quinoneimines Medium chain fatty acids Halogenated hydrocarbons and some halogenated aromatics (BrWClWF) Nitroaromatics Moieties that form a,b-unsaturated enol-like structures Thiols, thiono compounds, thiazolidinediones, thioureas Aminothiazoles Although the subject of reactive metabolites will not be treated in detail in this review, the reader can refer to several recent reviews on the role of reactive metabolites in toxicity and methods to detect them [20–27]
Safety Testing of Drug Metabolites
469
5.2 What kind of information is needed and when is it needed? The discovery of a human metabolite not previously seen in animal studies can cause a major disruption in the development program for the parent drug. Once such a metabolite is discovered, not only will safety testing in animals have to be carefully considered, but a validated analytical method may be necessary to monitor the metabolite in subsequent clinical studies. Given these implications, metabolism should be investigated ‘‘as early as possible’’ in drug development, as the MIST Guidance points out. However, ‘‘as early as possible’’ is very subjective and depends on drug class, the potential for filling an unmet need in the market, and the innovator’s risk tolerance. Nevertheless, there are trends regarding the types and timing of studies to generate metabolite information. These are discussed later, as along with some emerging trends and technologies in metabolite profiling. In vitro metabolite profiling. These studies are conducted using unlabeled drug incubated with liver microsomes or hepatocytes from rat, dog, human, and any other appropriate species with unlabeled drug. It is now common to conduct these studies early in the discovery process because they offer a first look at any species differences between animals and human and can help guide selection of the appropriate species for toxicology studies. Modern liquid chromatography-tandem mass spectrometry (LC/MS/MS) methodology has emerged as an indispensable technology for preliminary identification of metabolites before reference standards have been prepared [12]. However, it must be noted that LC/MS/MS methodology alone provides either qualitative or, at best, relatively quantitative information. Certainly, the presence of a unique human metabolite would be noticeable and important information. Moreover, it is possible to infer relative quantitative differences between animals and humans in the formation of a given metabolite. However, it is not possible to quantify absolutely the amount of metabolite formed relative to other metabolites and/or parent drug without synthetic metabolite standards, which are typically not available at this stage. Furthermore, in the absence of corroborating in vivo metabolite profiling data, investigators must always be wary that in vitro studies do not accurately forecast human metabolism [28,29]. The same kinds of in vitro studies may be repeated later once the radiolabeled parent (at a metabolically stable position) is available, usually late in preclinical or early in clinical development. The presence of radiolabel permits absolute quantification, so this not only is another opportunity to verify to what degree animals form the metabolite relative to humans but also provides quantitative data about the abundance of the metabolite relative to the total amount of drug-related material.
470
Thomas N. Thompson
In addition, it is becoming more common to use in vitro methods to produce metabolite ‘‘standards’’ for use in subsequent metabolite profiling. This can be done either by direct isolation of the metabolite or by pooling samples from incubations with labeled drug with samples from studies where unlabeled drug has been administered. In vivo metabolite profiling. These studies are conducted in mice, rats, dogs, or other appropriate species of interest. In principle, samples from any animal pharmacokinetic (PK) or toxicokinetic (TK) study could be analyzed for the presence of a suspected metabolite using LC/MS/MS. Indeed, with the advent of more powerful instrumentation, metabolite profiling of PK and TK studies is becoming more commonplace. However, in practice, it is probably still more common to conduct metabolite profiling as part of the mass balance studies with radiolabeled drug. Timing of these studies varies widely among companies and even programs within companies. Typically, these studies are now conducted in parallel with early clinical development, although in some cases they may be conducted late in preclinical development. In addition to the mass balance studies, it is often common practice to conduct special in vivo studies such as in bile-duct cannulated or even genetically engineered [30] animals to help isolate expected metabolites. It should be noted that the combination of in vitro studies that evaluate animal and human metabolism plus the in vivo metabolite profiles in animals can provide greater clarity in forecasting human metabolites in vivo versus either group of studies alone. Several authors have recently addressed this issue [13,28,29]. Human mass balance study. The human mass balance study with radiolabeled drug has long been considered the ‘‘gold standard’’ when it comes to definitive measurement of human metabolites [5]. The design, strategy, and retrospective analysis of human mass balance studies have been the subject of two recent reviews [31,32]. Typically, this study is conducted in phase 2A since it is highly desirable to have information on human metabolism in place before definitive phase 3 studies are underway. However, it is also common to defer this study to phase 3 to be more certain of the likelihood of success for the program before committing to the expense of the human mass balance study. This approach requires the sponsor to accept the risk that disproportionate human metabolites may be an issue. Although human mass balance studies have typically been conducted using radiolabeled drug at doses up to as much as 100 mCi, with the advent of accelerator mass spectrometry (AMS), a new paradigm is emerging. It is now not only possible but also becoming more common to conduct the human mass balance study with trace levels (E200 nCi or
Safety Testing of Drug Metabolites
471
less) of radiolabeled drug such that the study is not regulated as a radiolabeled study. As such, the time and expense of the preliminary animal dosimetry and the extra clinical costs associated with a radiolabed study can be waived [33–35]. These savings help to offset the added expense of AMS analysis, which is more costly than conventional LC/MS/ MS methods. Early human in vivo metabolite profiling. To avoid late development surprises, it is now more common to conduct metabolite profiling in early clinical studies (e.g., single and multiple ascending dose phase 1 studies), although unlabeled drug was administered. This early profiling can be qualitative, ‘‘semi’’-quantitative, or even quantitative depending on the technique employed. As mentioned earlier, this trend has been facilitated by the introduction of powerful new analytical techniques. For example, recent reviews have described use of a combination of state-of-the-art hardware (e.g., ultra performance liquid chromatography–mass spectrometry analysis, high-resolution time-of-flight mass spectrometer) and software (fractional mass filtering algorithm and computer-assisted structure elucidation software routines) to enable drug metabolites to be identified in plasma samples from a first-in-human study [36,37]. Although this metabolite profiling is still qualitative in nature, it can be made at least ‘‘semiquantitative’’ by use of metabolite standards derived from in vitro or in vivo metabolite profile studies with radiolabeled drugs. The radiolabeled metabolite is added to the ‘‘cold’’ sample and is used to ‘‘calibrate’’ the MS response [36]. Another exciting new area of research that can provide semiquantitative or even quantitative metabolite data is the use of nuclear magnetic resonance (NMR). Improvements in software and hardware have facilitated the use of NMR to determine actual concentrations of isolated metabolites and even measure metabolites from in vivo samples [38,39]. Synthesis of metabolite standards. These exciting bioanalytical techniques notwithstanding, at some point a decision must be made whether synthesis of (a) metabolite standard(s) should be undertaken. The exact timing of this step depends on many factors, including size and risk tolerance of the company, the resources available, and the degree of difficulty of the synthesis. In many cases, metabolites that are relatively easy to synthesize may be prepared early in preclinical development. However, in most cases, the synthesis of metabolite standards are deferred as late as possible, typically until after it has been established unequivocally that the metabolite(s) are deemed ‘‘significant’’ in the context of the MIST Guidance and, therefore, should be monitored in subsequent toxicology and/or clinical studies.
472
Thomas N. Thompson
6. ROLE OF THE MEDICINAL CHEMIST With respect to the MIST Guidance, informed decisions and strategies are best made with full understanding and thoughtful discussion of drug metabolism in the context of the mechanisms of toxicity [40]. Accordingly, the issues presented by the MIST Guidance have now brought drug metabolism and bioanalytical scientists into dialogue with toxicologists. However, it is important to consider that the medicinal chemist can also play a pivotal role in this dialogue at all stages of discovery and development. First, the medicinal chemist’s inherent understanding of the art and science of chemical structure elucidation may be invaluable to help identify metabolite structure. Next, once an understanding of structure is available, the chemist may be in the best position to judge whether the metabolite structure is likely to render it more or less pharmacologically active. Alternately, the chemist’s knowledge of structural alerts will be valuable to help properly predict a metabolite’s risk for toxicity. For example, hidden within the structure of certain metabolites may lie evidence that a reactive intermediate either has been formed or might form by subsequent secondary metabolism. The importance of these structural evaluations cannot be overstated. The MIST Guidance is clear that both abundance and activity (pharmacological or chemical reactivity) of metabolites drive the decision for further safety evaluation. Finally, chemists may be needed either to synthesize a given metabolite(s) for early evaluations or to provide input and/or oversight into batch scale synthesis to support metabolite toxicity testing.
7. SUMMARY That drug metabolites can play an important role in overall drug toxicity is now widely accepted as a significant issue in drug development. Advances in drug metabolism and bioanalytical chemistry have facilitated our ability to detect and quantify metabolites as never before. Given this ability, early identification of human metabolites is a logical approach for a successful drug development program. Although the FDA has recently outlined a generic course of action in their 2008 MIST Guidance, the issues are complex. Accordingly, the agency appears willing to discuss individual cases with sponsors to ensure timely and cost–effective development of important new medicines. However, though dialogue is welcome, the innovator should drive the understanding of metabolism issues and suggest an action plan. New technology gives many options, and sponsors should be ever mindful of
Safety Testing of Drug Metabolites
473
conducting those metabolism studies that should be done, and not necessarily all that can be done. In conclusion, it is clear that gaining a more thorough understanding of how metabolism may affect the toxicity of drugs is not only prudent, but necessary. It is incumbent upon sponsor companies, the FDA, and even academicians to work together using a flexible, science-driven approach to investigate these issues in a timely and cost–effective manner to ensure that the new medicines developed are safe.
REFERENCES [1] J. Polli, Presented in the AAPS Short Course on ‘‘Metabolites Data as a Requisite for Regulatory Submissions: Current Practice, Challenges and Case Studies’’ at the AAPS Annual Meeting, Atlanta GA, November 2008. [2] F. P. Guengerich, Chem. Res. Toxicol., 2006, 19, 1559. [3] K. L. Davis-Bruno and A. Atrakchi, Chem. Res. Toxicol., 2006, 19, 1561. [4] T. Prueksaritanont, J. H. Lin and T. Baillie, Toxicol. Appl. Pharmacol., 2006, 217, 143. [5] T. Baillie, M. N. Cayen, H. Fouda, R. J. Gerson, J. D. Green, S. J. Grossman, L. J. Klunk, B. LeBlanc, D. C. Perkins and L. A. Shipley, Toxicol. Appl. Pharmacol., 2002, 182, 188. [6] K. L. Hastings, J. El-Hage, A. Jacobs, J. Leighton, D. Morse and R. E. Osterberg, Toxicol. Appl. Pharmacol., 2003, 190, 91. [7] T. Baillie, M. N. Cayen, H. Fouda, R. J. Gerson, J. D. Green, S. J. Grossman, L. J. Klunk, B. LeBlanc, D. C. Perkins and L. A. Shipley, Toxicol. Appl. Pharmacol., 2003, 190, 93. [8] D. A. Smith and R. S. Obach, Drug Metab. Dispos., 2005, 33, 1409. [9] W. G. Humphreys and S. E. Unger, Chem. Res. Toxicol., 2006, 19, 1564. [10] D. A. Smith and R. S. Obach, Chem. Res. Toxicol., 2006, 19, 1570. [11] Food and Drug Administration, 2008, Guidance for Industry: Safety Testing of Drug Metabolites, http://www.fda.gov/CDER/GUIDANCE/6897fnl.pdf [12] T. A. Baillie, Chem. Res. Toxicol., 2008, 21, 129. [13] D. Luffer-Atlas, Drug Metab. Rev., 2008, 40, 447. [14] R. S. Obach, Presented in Plenary Session 1 ‘‘Minimizing Drug Toxicities during Drug Discovery and Development, at the 13th North American Regional ISSX Meeting, San Diego, CA, October 2008. [15] K. S. Pang, M. E. Morris and H. Sun, J. Pharm. Pharmacol., 2008, 60, 1247. [16] H. Sun and K. S. Pang, Drug Metab. Dispos., 2009, 37, 187. [17] K. S. Pang, Chem. Biol. Interact., 2009, 179, 45. [18] D. Luffer-Atlas, Presented in the APQ Open Forum ‘‘MIST and Beyond: The Role of Bioanalysis in the Assessment of Drug Metabolites in Safety Testing,’’ at the AAPS Annual Meeting, San Antonio, TX, November 2006. [19] F. P. Guengerich and J. S. MacDonald, Chem. Res. Toxicol., 2007, 20, 344. [20] A. S. Kalgutkar, I. Gardner, R. S. Obach, C. L. Shaffer, E. Callegari, K. R. Henne, A. E. Mutlib, D. K. Dalvie, J. S. Lee, Y. Nakai, J. P. O’Donnell, J. Boer and S. P. Harriman, Curr. Drug Metab., 2005, 6, 161. [21] A. S. Kalgutkar and J. R. Soglia, Expert Opin. Drug Metab. Toxicol., 2005, 1, 91. [22] K. Park, D. P. Williams, D. J. Naisbitt, N. R. Kitteringham and M. Pirmohamed, Toxicol. Appl. Pharmacol., 2005, 207, S425. [23] T. A. Baillie, Chem. Res. Toxicol., 2006, 19, 889. [24] J. C. L. Erve, Expert Opin. Drug Metab. Toxicol., 2007, 2, 923. [25] R. S. Obach, A. S. Kalgutkar, J. R. Soglia and S. X. Zhao, Chem. Res. Toxicol., 2008, 21, 1814.
474
Thomas N. Thompson
[26] H. Takakusa, H. Masumoto, H. Yukinaga, C. Makino, S. Nakayama, O. Okazaki and K. Sudo, Drug Metab. Dispos., 2008, 36, 1770. [27] J. N. Bauman, J. M. Kelly, S. Tripathy, S. X. Zhao, W. W. Lam, A. S. Kalgutkar and R. S. Obach, Chem. Res. Toxicol., 2009, 22, 332. [28] S. Anderson, D. Luffer-Atlas and M. P. Knadler, Chem. Res. Toxicol., 2009, 22, 243. [29] D. Dalvie, R. S. Obach, P. Kang, C. Prakash, C. M. Loi, S. Hurst, A. Nedderman, L. Goulet, E. Smith, H. Z. Bu and D. A. Smith, Chem. Res. Toxicol., 2009, 22, 357. [30] M. W. Powley, C. B. Frederick, F. D. Sistare and J. J. DeGeorge, Chem. Res. Toxicol., 2009, 22, 257. [31] J. H. Beumer, J. H. Beijnen and J. H. M. Schellens, Clin. Pharmacokinet., 2006, 45, 33. [32] S. J. Roffey, R. S. Obach, J. I. Gedge and D. A. Smith, Drug Metab. Rev., 2007, 39, 17. [33] G. Lappin and R. C. Garner, Expert Opin. Drug Metab. Toxicol., 2005, 1, 23. [34] G. Lappin, M. Rowland and R. C. Garner, Expert Opin. Drug Metab. Toxicol., 2006, 2, 419. [35] G. Lappin and L. Stevens, Expert Opin. Drug Metab. Toxicol., 2008, 4, 1021. [36] T. A. Baillie, Chem. Res. Toxicol., 2009, 22, 263. [37] L. Leclercq, F. Cuyckens, G. S. Mannens, R. de Vries, P. Timmerman and D. C. Evans, Chem. Res. Toxicol., 2009, 22, 280. [38] R. Espina, L. Yu, J. Wang, Z. Tong, S. Vashishtha, R. Talaat, J. Scatina and A. Mutlib, Chem. Res. Toxicol., 2009, 22, 299. [39] K. Vishwanathan, K. Babalola, J. Wang, R. Espina, L. Yu, A. Adedoyin, R. Talaat, A. Mutlib and J. Scatina, Chem. Res. Toxicol., 2009, 22, 311. [40] D. A. Smith and R. S. Obach, Chem. Res. Toxicol, 2009, 22, 267.
APPENDIX: DECISION TREE FLOW DIAGRAM Disproportionate Drug Metabolite >10% parent systemic exposure (AUC)
<10% parent systemic exposure (AUC)
Formed in any animal test species? No further testing needed to evaluate metabolite No
Yes How much?
Exposure in animal studies does not approach human exposure Nonclinical testing with the drug metabolite
Exposure in animal studies does approach human exposure
No further testing needed to qualify metabolite
CHAPT ER
23 A Path to Innovation: Gene Knockouts Model New Drug Action Brian P. Zambrowicz and Arthur T. Sands
Contents
1. Introduction 2. Metabolism 3. Gastrointestinal 4. Cardiovascular 5. Immunology/Hematology 6. Rare Genetic Diseases 7. Oncology/Ophthalmology 8. Central Nervous System 9. Can Mouse Knockout Data Guide New Drug Discovery? References
475 477 478 479 481 484 485 487 488 491
1. INTRODUCTION In February of 2001, the initial draft of the human genome was published [1,2]. Perhaps, most surprising at the time was the disclosure that the total number of protein-encoding genes appeared to be approximately 31,000, less than a third of the then prevailing expressed sequence tag (EST)-based estimates of 100,000 or more. After several more rounds through the bioinformatics wash, our favorite set of genes has shrunk even further to a current estimated total of about 20,500 [3], all around a far less imposing genome from the perspective of drug discoverers who Lexicon Pharmaceuticals Incorporated, 8800 Technology Forest Place, The Woodlands, TX 77381, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04423-6
r 2009 Elsevier Inc. All rights reserved.
475
476
Brian P. Zambrowicz and Arthur T. Sands
are exploring this new territory for novel targets with medical applications. A more accurately mapped human genome means fewer mutants to manufacture, proteins to produce, assays to assess, RNAi to synthesize, targets to probe, antibodies to raise, and small molecules to make. Nonetheless, even with the help from Mother Nature on the numbers, the mammalian genome has proven to be rugged terrain for drug discoverers who must forge new pathways, define novel targets, and invent breakthrough drugs aspiring one day to conquer peak sales. It was no coincidence that a little more than 1 year after the first draft of the human genome was completed, the genome of the second mammal was published: the mouse [4]. The mouse genome sequence ‘‘many say holds more promise for our future than even the human genome itself. But why? The laboratory mouse is hailed as holding the experimental key to the human genome. Working on mouse models allows the manipulation of each and every gene to determine its function, and this will give us detailed insights into many aspects of human disease as well as basic human biology’’ [5]. Not surprisingly, the mouse genome was found to contain approximately the same number of protein-encoding genes as humans with a direct orthologous relationship for 99% of human genes. Indeed, the mouse has served as a faithful genetic model for the study of mammalian physiology for many decades. With the historical contribution of chemical mutagenesis, the development of transgenic techniques, the expansion of homologous recombination, and the automation of gene trapping, virtually no mouse gene has been left undisturbed [6]. The mouse has demonstrated it can be relied upon to serve as a ‘‘native guide’’ to the human genome, helping us to experimentally navigate her many mysteries. But what are the implications of using the mouse to translate such knowledge into new targets for drug discovery? In 2003, we first reviewed the topic of the relationship between mouse knockout phenotypes and drug action in humans by analyzing the data for the targets of the 100 best-selling drugs [7]. That retrospective look at the top targets of the pharmaceutical industry indicated there was a strong correlation between knockout mouse phenotypes and drug action in humans. The data suggested that knockout mice create a genetic antagonism of a target that can reliably model chemical or antibodymediated antagonism of the target. Subsequently, we also looked at the drugs from the top 10 pharmaceutical companies with novel mechanisms of action at phase 2 clinical trials or beyond [8]. Again, there was a high degree of correlation between anticipated drug action and knockout mouse phenotypes for the relevant targets. Now 6 years after the initial publication of the mouse genome and many knockouts later, it seems appropriate to take another look at how well the mouse has been serving as an experimental key to the human genome for defining drug targets. The Food and Drug Administration (FDA) has approved drugs
A Path to Innovation: Gene Knockouts Model New Drug Action
477
addressing 30 novel mechanisms from 2002 through 2007. In the present review, we will examine the correlation between the action of these drugs in man and the phenotypes of their corresponding target knockouts in mice. We will also discuss the potential of prospectively using knockouts to guide drug discovery.
2. METABOLISM Type 2 diabetes is a debilitating disease with an alarmingly increased incidence throughout the world. Type 2 diabetes is characterized by hyperglycemia, a defect in insulin secretion and insulin resistance in peripheral tissues such as muscle, liver, and adipose tissue. Januvias (sitagliptin) and Byettas (exenatide) are two new drugs approved to treat diabetes that both act on the incretin signaling pathway. Two incretins, glucagon-like peptide 1 (GLP-1) and glucose-dependant insulinotropic polypeptide (GIP), are released by the gut in response to a meal [9]. By binding to their receptors, GLP-1 receptor and GIP receptor, respectively, on pancreatic b cells, they induce insulin secretion in a glucose-dependent manner. The incretins have a short half-life, being rapidly degraded by dipeptidyl peptidase 4 (DPP4). Byettas is a synthetic form of a peptide found in Gila monster saliva that mimics GLP-1 action by functioning as an agonist at the GLP-1 receptor and stimulating glucose-dependent insulin secretion. Januvias is a small molecule DPP4 inhibitor that prevents the breakdown of both GLP-1 and GIP, thereby extending their action and enhancing glucose-dependent insulin secretion. The mouse knockouts of the GLP-1 receptor [10], GIP receptor [11,12], and DPP4 [13,14] have been highly informative for understanding the function of the incretin signaling pathway and are predictive of drug action in man for both Byettas and Januvias. In response to an oral glucose challenge, both GLP-1 receptor–null and GIP receptor–null animals exhibit elevated blood glucose and lower levels of circulating insulin relative to wild-type controls. While single knockouts of either the GLP-1 receptor or the GIP receptor results in a relatively mild defect in glucose homeostasis, double knockouts of both receptors exhibit a more profound defect in glucose homeostasis in response to an oral glucose challenge due to a distinct b-cell defect in glucose-dependent insulin secretion [15]. In addition, double receptor knockout animals do not respond to either Byettas or DPP4 inhibitors [16]. These data clearly support the rationale for augmenting GLP-1 and GIP activity to enhance glucose-dependent insulin secretion and underpin the observations that GLP-1 or GIP mimetics could have utility for treating type 2 diabetes. DPP4 knockouts exhibit improved glucose homeostasis and reliably model pharmacological antagonism of the DPP4 protein. In response to
478
Brian P. Zambrowicz and Arthur T. Sands
an oral glucose challenge, blood glucose levels do not peak as high and return to baseline more rapidly in DPP4 knockouts relative to wild-type controls. Fifteen minutes after an oral glucose challenge, knockout animals exhibit lower plasma glucose and higher plasma insulin and GLP-1. DPP4 knockouts do not respond to DPP4 inhibitors. Again, these data indicate the importance of DPP4 for inactivation of GLP-1 and suggest that pharmacological inhibition of DPP4 should result in enhanced glucose-dependent insulin secretion and improved glucose homeostasis. Furthermore, the lack of a response to DPP4 inhibitors in DPP4 [13] knockout animals and GLP-1/GIP receptor double knockouts [16] suggests the action of the inhibitors for improving glucosedependent insulin secretion is due to specific antagonism of DPP4 and that all the actions of DPP4 inhibitors on glucose homeostasis are through enhancing GLP-1 and GIP signaling through their receptors. Another newly approved drug for the treatment of type 2 diabetes is Symlins (pramlintide). Symlins is a synthetic version of the peptide amylin containing several amino acid changes. Amylin is a peptide that is co-secreted with insulin by pancreatic b cells. This peptide helps control postprandial blood glucose levels by reducing gastric emptying and decreasing glucagon secretion thereby inhibiting hepatic glucose production. It is also associated with enhanced feelings of satiety and eventually weight loss. In diabetic patients, treatment with Symlins before a meal results in lower postprandial hyperglycemia. Amylin knockout mice express lower levels of insulin in their pancreatic b cells and exhibit faster glucose clearing after an oral glucose challenge [17]. In addition, amylin knockout mice gain weight more rapidly than wild-type controls. Amylin knockout mice are also more sensitive to alloxan, a pancreatic b-cellspecific cytotoxic agent [18]. Upon exposure to this agent, amylin knockouts develop diabetes more rapidly, exhibiting higher basal plasma glucose levels, lower basal insulin levels, and impaired glucose tolerance relative to alloxan-treated wild-type animals. This enhanced diabetic effect in the knockout mice is due to decreased b-cell insulin expression, a greater loss of b-cell containing islets, and a more reduced b-cell mass. By contrast, mice over-expressing human amylin express and store more insulin in their pancreatic b cells. In this case, the knockout suggests that amylin plays a role in postprandial glycemic control, weight gain, and may have a protective effect on pancreatic b cells.
3. GASTROINTESTINAL Two drugs were approved for treatment of gastrointestinal (GI) diseases. Zelnorms (tegaserod) was approved to treat constipation-predominant irritable bowel syndrome. Zelnorms is a 5-hydroxytryptamine 4 (5-HT4)
A Path to Innovation: Gene Knockouts Model New Drug Action
479
receptor agonist that increases GI motility. Not surprisingly, mice with knockout of the 5-HT4 receptor exhibit unusually slow GI motility [19], clearly indicating the important role of this receptor for regulating GI transit. The second GI drug approved is Amatizas (lubiprostone) for chronic idiopathic constipation. Amatizas is an activator of the chloride channel 2 (CLC-2) that is expressed in the GI epithelium. Activation of CLC-2 results in the release of chloride ions into the intestinal lumen. This causes secretion of fluid into the intestine resulting in enhanced GI motility and relief from constipation. The only observable phenotype in CLC-2 knockout mice is degeneration of male germ cells and photoreceptors [20]. It would be interesting to test whether these knockouts might also exhibit any alterations in intestinal secretion or GI motility.
4. CARDIOVASCULAR Vaprisols (conivaptan) has been approved for the treatment of hypervolemic hyponatremia, a condition of electrolyte imbalance most commonly associated with congestive heart failure, cirrhosis, and renal disease. Hypervolemic hyponatremia patients exhibit peripheral edema and have low plasma sodium levels and low total body water. The low total body water causes low plasma volume and triggers antidiuretic hormone (ADH) release. ADH acts on the kidneys to concentrate the urine and retain water, which can further exacerbate the low serum sodium levels. Vaprisol’s s mechanism of action is antagonism of the vasopressin V2 receptor in the renal collecting ducts of the kidney. The result is that the kidneys do not respond to ADH and release more water causing an increase in serum sodium concentration. The mechanism of the vasopressin receptor antagonism is clearly demonstrated by human mutations in this receptor that result in nephrogenic diabetes insipidus [21,22] and targeted mutation of this receptor in mice [23]. Human X-linked diabetes insipidus has been associated with inactivating mutations of the vasopressin V2 receptor gene. Individuals with this condition have frequent urination and excessive thirst and are unresponsive to ADH. Yun et al. produced mice with a targeted nonsense mutation in the vasopressin V2 receptor that is known to cause some cases of human X-linked diabetes insipidus. These mice demonstrated the importance of this receptor for proper salt and water balance. The mutant mice exhibited a decrease in urine osmolality and could not concentrate their urine. They failed to thrive and died within 1 week after birth. In this case, genetic and pharmacological antagonism of the target both result in a lack of response to ADH and an increase in water release by the kidneys.
480
Brian P. Zambrowicz and Arthur T. Sands
The mineralocorticoid receptor is another regulator of salt and water balance. Aldosterone is released by the adrenal glands in response to low blood volume and acts on the kidney to decrease sodium excretion, thereby retaining water, and increase potassium excretion. Again, information from both human and mouse genetics provides important insights into mineralocorticoid receptor function. In humans, mutations in the mineralocorticoid receptor result in autosomal dominant pseuodohypoaldosteronism type I [24]. This disease is characterized by sodium and water wasting by the kidney, hyperkalemia, and hyponatremia. These patients are unresponsive to mineralocorticoids. Likewise in mice, knockout of the mineralocorticoid receptor gene results in the symptoms of pseudohypoaldosteronism including renal sodium and water loss, hyperkalemia, and hyponatremia [25]. The knockout mice die around 10 days after birth due to dehydration. Inspras (eplerenone) binds to the mineralocorticoid receptor and blocks aldosterone binding. Like genetic inhibition of the receptor, the pharmacological inhibition results in an increase loss of sodium and water by the kidney, thereby lowering blood volume and blood pressure. Inspras has been approved for the treatment of hypertension and heart failure post-myocardial infarction. Tekturnas (alisikiren) is the first renin inhibitor approved to treat hypertension. Renin is an aspartyl protease, secreted by the juxtaglomerular cells of the kidney that catalyzes the conversion of angiotensinogen to angiotensin I. Angiotensin I is subsequently converted to angiotensin II by angiotensin converting enzyme (ACE). Angiotensin II through interaction with angiotensin receptors, AT1 and AT2, produces increases in blood pressure by vasoconstriction, aldosterone release, and various other mechanisms. Renin is secreted in response to decreased arterial blood pressure and, as the first and rate-limiting step in the angiotensinogen cascade, represents an exciting mechanism for treating hypertension. The renin/angiotensin system has been a productive pathway for drug development with multiple ACE inhibitors and angiotensin receptor antagonists on the market. Like ACE and the angiotensin receptor AT1, the utility of renin as a target for anti-hypertensive agents is supported by mouse knockout data. While humans have only one renin gene, a duplication event has resulted in two renin genes in some strains of mice, renin-1 and renin-2. Knockout of renin-1 results in a sexually dimorphic decrease in blood pressure [26]. Females deficient in renin-1 exhibit an approximately 13 mm Hg decrease in mean arterial blood pressure, but males exhibit no blood pressure changes. Knockout of renin-2 has no effect on blood pressure [27], presumably due to functional redundancy with renin-1; double knockouts have not yet been reported. Zetias (ezetimibe) is another new drug approved to treat cardiovascular disease. Zetias is used to lower serum cholesterol levels, but unlike the statins that lower serum cholesterol by inhibiting cholesterol
A Path to Innovation: Gene Knockouts Model New Drug Action
481
production, Zetias does it by inhibiting intestinal cholesterol uptake. Zetias inhibits cholesterol absorption by blocking the function of Niemann–Pick C1 Like 1 (NPC1L1) protein. The first indication that Zetias might act through a NPC1L1-dependent pathway came from studies of NPC1L1 knockout mice [28]. Although otherwise normal, these mice exhibited a large reduction in cholesterol absorption and did not respond to treatment with Zetias. NPC1L1 was subsequently demonstrated to be the direct target of Zetias using a binding assay [29]. Zetias was found to bind to enterocyte membranes derived from wild-type, but not NPC1L1 knockout mice. In this case, the knockout not only mimicked the effect of Zetias on cholesterol absorption, it was also critical for identifying the mechanism of action of this drug.
5. IMMUNOLOGY/HEMATOLOGY Five new drugs have been approved in the area of Immunology and Autoimmune Disease. All five are large molecule protein drugs. Cytotoxic T-lymphocyte antigen 4 (CTLA4) is involved in the T-cell co-stimulatory process. CTLA4 and CD28 share the CD80 and CD86 ligands. Binding of these ligands to CD28 results in increased T-cell proliferation and interleukin-2 (IL-2) production. Binding of the ligands to CTLA4 appears to provide an inhibitory signal and/or prevent the stimulatory CD28 signal. Antibodies that cross-link CTLA4 have been demonstrated to decrease the proliferative response and the production of IL-2 by T-cells. Thus, it appears that CD28 and CTLA4 play opposing roles in balancing T-cell response. This important role for CTLA4 in inhibiting T-cell co-stimulation is supported by CTLA4 knockout data [30]. Mice lacking CTLA4 develop a very severe lymphoproliferative disorder resulting in death. CD4+ T-cells are expanded in these mice, and a greater percentage of T-cells are observed in the blast stage. Lymphocytes infiltrate many organs and the animals die at about 2–3 weeks of age. Enhanced binding of CD80 and CD88 to CD28 would be expected to increase T-cell response while decreased binding to CD28 would be expected to have the opposite effect. One way to prevent ligand binding to CD28 is to provide a soluble CTLA4 protein. Orencias (abatacept) is a soluble CTLA4/immunoglobulin Fc fusion protein that can bind CD80 and CD88 and prevent them from interacting with CD28. Orencias has been approved to treat rheumatoid arthritis patients who have not had an adequate response to methotrexate or the anti-tumor necrosis factor a (TNFa) agents. Immunoglobulin (Ig)E plays an important role in allergic response and asthma. Elevated circulating IgE levels were found in asthma patients. IgE binds to high-affinity Fc receptors on mast cells and basophils and once cross-linked with allergen causes these cells to release
482
Brian P. Zambrowicz and Arthur T. Sands
proinflammatory molecules such as histamine, prostaglandins, leukotrienes, and proinflammatory cytokines. The importance of IgE and its binding to the high-affinity FceRI receptor for allergic response has been clearly demonstrated by studying FceRI knockout mice [31]. These knockout mice are resistant to both passive cutaneous anaphylaxis and systemic anaphylaxis. In addition, when subjected to an asthma challenge by immunization with ovalbumin and subsequent challenge with aerosolized ovalbumin, the knockout mice exhibited a decrease in airway hyper-responsiveness [32]. These data suggest that blocking the IgE/FceRI interaction could provide benefit for treating allergic response and asthma. Xolair s (omalizumab) is an antibody to IgE that prevents interaction of IgE with the FceRI receptor. Xolairs has been approved to treat asthma patients who are not responsive to corticosteroid treatment. The other three new drugs in the autoimmune disease area are antibodies that target cell adhesion molecules. Raptivas (efalizumab) is a monoclonal antibody that binds to the CD11a component of lymphocyte function-associated antigen (LFA)-1, a b2-integrin composed of CD11a and CD18 subunits. CD11a is specific to LFA-1 while CD18 is a component of several b2-integrins. LFA-1 on leukocytes interacts with intercellular cell adhesion molecule-1 (ICAM-1) present on endothelial cells and is thought to be involved in attachment and transendothelial migration of leukocytes, a process involved in leukocyte migration to sites of inflammation. Raptivas blocks the interaction of LFA-1 with ICAM-1 and has been approved to treat patients with moderate to severe plaque psoriasis. The importance of the LFA-1/ICAM-1 adhesion for interaction of leukocytes with endothelial cells and in regulating inflammatory response has been clearly illustrated by CD11a, CD18, and ICAM-1 knockout mice. CD11a and CD18 knockout mice exhibit increased leukocyte rolling velocities and reduced leukocyte adhesion efficiency in muscle venules after TNFa challenge [33]. Increased leukocyte rolling velocities are also observed in muscle venules of ICAM-1 knockout mice after inflammation induced by either trauma or TNFa [34]. In addition, both ICAM-1 and LFA-1 knockout mice exhibit alterations in various models of inflammation and autoimmune disease. Knockout of ICAM-1 results in a decreased inflammatory response in both an ovalbumininduced asthma model [35] and a dextran sodium sulfate–induced model of colitis [36]. ICAM-1 knockout mice also exhibit a decreased incidence of collagen-induced arthritis [37]. Knockout of CD11a or CD18 reduces the severity of a mouse model of lupus [38]. CD11a knockout reduces the graft-verses-host reaction toward allogeneic cells [39], and CD18 knockout decreases the inflammatory response in a passive transfer model of arthritis [40]. Importantly, CD18-null mice exhibit reduced response in allergic contact dermatitis and delayed-type hypersensitivity [41]. Extravasation of T-lymphocytes into skin was greatly reduced in both
A Path to Innovation: Gene Knockouts Model New Drug Action
483
allergic contact dermatitis and irritant dermatitis. The number of neutrophils reaching the dermis of the skin in the contact dermatitis challenge in CD18 knockout mice was 98% reduced relative to wild-type animals [42]. Tysabris (natalizumab) is a monoclonal antibody that targets another integrin-based cell adhesion interaction. Tysabris binds the a4 subunit of a4b1 and a4b7 integrin. Tysabris has been approved for the treatment of multiple sclerosis and inflammatory bowel disease (FDA approval, January 2008). a4b1 and a4b7 integrin both bind to vascular cell adhesion molecule-1 (VCAM-1), and a4b7 integrin binds to mucosal addressin cell adhesion molecule-1 (MADCAM-1) as well. The important role for a4b7 integrin and MADCAM-1 for the homing of leukocytes to Peyer’s patches and other gut-associated lymphoid organs has been demonstrated in b7 [43] and NKX2.3 [44] knockout mice. b7 Knockout mice exhibit smaller Peyer’s patches with decreased cellularity and lymphocyte numbers. These knockout animals also had a 10- to 30-fold decrease in IgM- and IgA-positive B cells and CD4-positive T-cells in the lamina propria of the gut. Lymphocytes of b7 knockout mice did not adhere normally to high endothelial venules of Peyer’s patches [45]. The leukocyte rolling velocity was increased in Peyer’s patch venules of b7 knockout mice, and b7 knockout mice had a delayed response in an adoptive transfer model of colitis but not in a genetic model resulting from IL-2 knockout [46]. b7 Knockout mice also exhibit a delayed skin allograft rejection [47]. NKX2.3 knockout mice do not express MADCAM-1 and exhibit Peyer’s patches with reduced size and altered architecture. Knockout of a4 integrin [48], b1 integrin [49,50], and VCAM-1 [51] all result in embryonic lethality precluding the use of these knockout mice for studying leukocyte homing in adult mice. However, when bone marrow cells harboring a conditional knockout of a4 integrin are transplanted to wild-type recipients, they exhibit defective homing to the bone marrow [52]. Amevives (alefacept) is a chimeric fusion protein containing the extracellular portion of LFA-3 fused to the Fc portion of immunoglobulin. It has been approved for the treatment of psoriasis. LFA-3 normally interacts with CD2, and its binding is thought to induce a mitogenic signal through CD2 in lymphocytes and stabilize interactions between T-cells and their targets. Both are important activities for mounting a normal immune response. Amevives binds to CD2 and prevents the normal interaction of LFA-3 with CD2. Knockout of CD2 results in healthy animals with apparently normal populations of T-cells that respond normally to inflammatory challenges [53]. This knockout does not provide supportive data for the utility of Amevive in psoriasis. Perhaps, this is because LFA-3 interacts with other molecules besides CD2; evaluation of an LFA-3 knockout could be used to test such a hypothesis.
484
Brian P. Zambrowicz and Arthur T. Sands
Paroxysmal nocturnal hemaglobinurea (PNH) is a disease characterized by increased susceptibility of red blood cells to complement-mediated hemolysis, thrombosis in large vessels, and deficient hematopoiesis. The disease is due to a genetic deficiency in phosphatidylinositol glycan class A (PIG-A), the enzyme that attaches proteins to cell surfaces through glycosyl-phosphatidylinositol anchors. This deficiency results in decreases or loss of cell-surface proteins including decay accelerating factor, homologous restriction factor, and membrane inhibitor of reactive lysis (MIRL), which play a role in regulating complement-mediated hemolysis. Soliriss (eculizumab) is an antibody to complement protein 5 (C5) that prevents intravascular hemolysis by blocking C5 activation. Mice with knockouts of MIRL [54] and complement components 3 (C3) [55] have been produced, and a naturally occurring mutation in complement component C5 has been identified [56]. Mice with knockout of PIG-A in some hematopoietic cells [57,58] and mice with a null mutation in MIRL [54] both exhibit spontaneous intravascular hemolysis that is less severe than that observed in human PNH. The hemolysis in MIRL knockout mice can be rescued by a knockout of C3 [59]. C5 knockout mice have not been studied for effects on complement-mediated hemolysis, but it would be interesting to cross the MIRL knockout or the PIG-A mice with C5 knockouts to determine whether hemolysis can be rescued.
6. RARE GENETIC DISEASES We have excluded enzyme replacement therapies from this discussion. Tyrosinemia Type I is caused by mutations in the fumarylacetoacetate hydrolase (FAH) gene. FAH catalyzes the final step in the tyrosine catabolic pathway. Individuals with Tyrosinemia Type I die of liver cirrhosis and kidney disease. Knockout of FAH results in neonatal lethality due to hypoglycemia and liver dysfunction [60]. Plasma alanine aminotransferase (ALT) levels correlate with liver damage and FAH knockout mice exhibit elevated plasma ALT levels. FAH knockout mice can be genetically rescued by a knockout of 4-hydroxyphenylpyruvate dioxygenase (HPD), another tyrosine catabolic enzyme upstream of FAH [61]. HPD knockout prevents the production of maleylacetoacetate and fumarylacetocetate, which, after conversion to succinylacetone and succinylacetoacetate, cause the liver and kidney toxicities observed in Tyrosinemia Type I. The double knockout mice did not exhibit any liver or kidney problems. These data suggest that inhibition of HPD could provide therapeutic utility for the treatment of Tyrosinemia Type I. Orfadins (nitisinon) is a small molecule inhibitor of HPD approved to treat patients with Tyrosinemia Type I. Increlexs (mecasermin) is a recombinant human insulin-like growth factor-1 (IGF-1) approved to treat growth retardation due to primary IGF-1 deficiency. The role of IGF-1 for normal growth is clear from
A Path to Innovation: Gene Knockouts Model New Drug Action
485
studies of IGF-1 knockout mice as well as knockout of the IGF-1 receptor [62,63]. IGF-1 knockout mice are born at about 60% of normal size while heterozygotes are born at about 80% of normal size. IGF-1 receptor knockout mice show a more severe phenotype [62,64]. They are born at about 45% of normal size and die at birth.
7. ONCOLOGY/OPHTHALMOLOGY A large number of drugs with a seemingly disparate array of new targets have recently been approved to treat cancer. Analysis of the knockouts of these targets, however, reveals commonalities of function at the level of cell and developmental biology: phenotypes tend to include potent effects on cell viability and embryonic lethality. Two new small molecule drugs approved to treat cancer are multi-kinase inhibitors. Nexavars (sorafenib) inhibits vascular endothelial growth factor receptor (VEGFR) 2 and 3, platelet-derived growth factor receptor beta (PDGFR-b), KIT (CD117, a cytokine receptor), Fms-related tyrosine kinase 3 (FLT-3), C receptorassociated factor (CRAF), B receptor-associated factor (BRAF). Of these seven targets, knockout of all but FLT-3 results in embryonic lethality [65–73]. Likewise, Sutents (sunitinib) is an inhibitor of multiple kinases including PDGFR-a and PDGFR-b, VEGFR1, 2, and 3, KIT, FLT-3, colonystimulating factor (CSF)-1R and rearranged during transfection protooncogene (RET). Of these nine targets, knockouts of all but CSF-1R and FLT-3 result in embryonic lethality [65–71,74–77]. For perspective on the potency of these targets, VEGFR1, 2, and 3 knockouts all result in embryonic lethality with defects in angiogenesis, and knockout of their ligand VEGF results in the rare phenotype of heterozygous lethality [78,79]. This means that having only one of two Vegf gene copies is incompatible with life. While embryonic stem (ES) cells normally form tumors in nude mice, VEGF knockout ES cells exhibit a dramatic reduction in their ability to form these tumors [79]. Likewise, BRAF knockout results in embryonic lethality due to vascular defects that are likely the result of apoptosis of vascular endothelial cells [73], and CRAF knockouts are embryonic lethal and show dramatically increased liver apoptosis indicating the normal role of CRAF in inhibition of apoptosis [72]. Another inhibitor of multiple kinases is Tykerbs (lapatinib), a dual human epidermal growth factor receptor 2 (HER2) (ERBB2) and epidermal growth factor receptor (EGFR) inhibitor. Mutations in both of these kinases are associated with human tumors with elevated expression or mutation of HER2 being associated with breast cancer. Tykerbs has been approved to treat breast cancer with elevated HER2 expression. Knockout of either of these targets results in an embryonic or neonatal lethal phenotype [80–82]. Two drugs represent the first compounds to target epigenetic mechanisms. These compounds target the DNA modifications of
486
Brian P. Zambrowicz and Arthur T. Sands
acetylation and methylation. Zolinzas (vorinostat) is an inhibitor of histone deacetylaces 1, 2, 3, and 6 (HDAC1–3,6). HDAC1 knockout results in embryonic lethality and HDAC1 appears to be required for cell proliferation through repression of cell-cycle inhibitors [83]. Vidaza (azacytidine) is the first DNA methyltranferase inhibitor approved to treat certain types of myelodysplastic syndrome. Although there are many DNA methyltranferases, the cytotoxic effect of azacytidine appears to be primarily mediated by DNMT3a and DNMT3b. This was demonstrated with the use of ES cells with mutations in DNMT genes [84]. While ES cells null for both DNMT3a and DNMT3b are highly resistant to the cytotoxic effects of azacytidine, ES cells with single mutations in DNMT1, DNMT3a, or DNMT3b exhibit normal sensitivity to azacytidine. Knockout of either DNMT3a or DNMT3b results in an embryonic lethal phenotype [85]. For both of these epigenetic mechanisms, rapidly dividing cells such as cancer cells appear to be more sensitive to inhibition than normal cells. A series of rather high-profile new drugs approved for cancer also inhibit targets that, when knocked out, cause embryonic lethality. These include four drugs targeting the EGFR. Iressas (gefitinib) and Tarcevas (erlotinib) are small molecule inhibitors of the EGFR kinase while Erbituxs (cetuximab) and Vectibixs (panitumumab) are antibodies that bind to the receptor. Avastins (bevacizumab) and Lucentiss (ranibizumab) are an antibody and a Fab fragment, respectively, that bind VEGF and are approved for treating cancer and age-related macular degeneration (AMD) through inhibition of angiogenesis. Macugens (pegaptanib) is an aptamer that also acts as a VEGF antagonist and is approved for the treatment of AMD. The roles of VEGF and its receptors in angiogenesis were described above. Finally, Velcades (bortezomib) is a small molecule proteosome inhibitor approved to treat cancer. In the latter case, while there have not been publications of knockouts of proteosome subunits in mice, mutations of any one of multiple proteosome subunits in other genetic model organisms such as fly, worm, and yeast [86–92] result in severe phenotypes that are incompatible with life. Finally, in the area of cancer supportive care, palifermin was approved to treat the mucositis associated with chemotherapy. Kepivances is recombinant human keratinocyte growth factor 1 that enhances the proliferation, differentiation, and migration of epithelial cells in the skin. Knockouts of keratinocyte growth factor 2, the most closely related family member, exhibit a decreased proliferation and aberrant terminal differentiation of the epidermis, confirming the importance of keratinocyte growth factors for promoting cell growth and division in skin and supporting the basis for protein therapy as a growth promoting agent for epithelial cells [93]. In spite of the desire to find increasingly targeted therapies such as Gleevecs (imatinib), Herceptins (trastuzumab), and Iressas, which hit
A Path to Innovation: Gene Knockouts Model New Drug Action
487
drivers of tumor formation or progression, the majority of current oncology drugs continue to target potent mechanisms that rapidly dividing cells are slightly more dependent on than normal cells. Although new tumor drivers will continue to be identified, as with the EML4-ALK fusion that appears to be a driver of cancer in a percentage of non-small cell lung cancer patients [94], the majority of cancer targets will likely continue to be mechanisms critical for the rapid cell division in cancer. Mouse knockout data suggest that this process may be best modeled by the rapid cell division observed during early embryonic development. In reviewing these data, one must ask whether embryonic lethality should be a prerequisite for any new oncology target. In other words, in the absence of targeting a clear tumor driver, would targeting a mechanism that does not cause embryonic lethality provide a potent enough approach for treating cancer?
8. CENTRAL NERVOUS SYSTEM Four new drugs with novel mechanisms of action have been approved that act on the central nervous system (CNS). Prialts (ziconatide) is a synthetic version of a marine cone snail peptide toxin that inhibits N-type calcium channels. It has been approved for treating severe chronic pain through intrathecal delivery. Knockout of the N-type calcium channel [95–97] did not appear to have a major effect on acute pain as knockout mice exhibited a normal response in mechanical pain induced by von Frey hairs and thermal pain in the paw flick and hot plate tests with only a slight increase in latency to respond in the tail flick test. However, the knockout mice exhibited a decreased response in models of visceral pain and neuropathic pain. The knockouts had a decreased writhing response in an acetic acid–induced model of visceral pain and a decreased response in the early half of phase 2 of the formalin paw test, a measure of neuropathic pain. Another model of neuropathic pain is the hyperalgesia resulting from spinal nerve ligation. The threshold for mechanical and thermal stimuli was markedly increased in the knockout animals following nerve ligation. These data clearly indicate the importance of N-type calcium channels for neuropathic pain. Namendas (memantine) is an N-methyl-D-aspartic acid (NMDA) receptor antagonist approved to treat moderate to severe Alzheimer’s type dementia. Namendas acts as a cognitive enhancer. The role of the NMDA receptor in learning and memory has been demonstrated through the study of mice with a tissue-specific knockout of the NMDA receptor in the CA1 region of the hippocampus [98–102]. These studies demonstrate the importance of the NMDA receptor in the hippocampus for spatial and nonspatial memory as well as for memory consolidation.
488
Brian P. Zambrowicz and Arthur T. Sands
Rozerems (ramelteon) is a melatonin 1 and 2 receptor dual agonist approved to treat insomnia with difficulty falling asleep. Both receptors have been knocked out. While knockouts of the melatonin 1 receptor exhibit decreased expression levels of several circadian rhythm genes including Per1, Cry1, Clock, and Bmal1 in pars tuberalis [103], neither receptor knockout results in an altered circadian rhythm [103,104]. Melatonin 1 receptor knockout mice also exhibited enhanced anxiety-like behavior in the open field by spending less time in center, enhanced depressive-like behavior by increased immobility in the forced swim test, and impaired sensory motor gating as indicated by a decrease in prepulse inhibition of the acoustic startle [105]. It would be interesting to study melatonin 1 and 2 receptor double knockout mice for any effects on circadian rhythm or sleep. Emends (aprepitant) is a neurokinin 1 receptor antagonist approved to treat chemotherapy-induced nausea and vomiting. Both the NK1 receptor and its ligand encoded by the TAC1 gene have been knocked out [106,107]. The NK1 receptor knockout mice exhibited increased time spent in the open arms of an elevated plus maze and decreased latency to feed in novelty-suppressed feeding [106], both are anxiolytic-like behaviors. These receptor knockout mice also exhibited decreased visceral pain and hyperalgesia [108]. TAC1 knockouts also exhibited a decrease in anxiolytic- and depressive-like behavior [107]. Knockouts had increased activity in the open arms of an elevated plus maze and a decreased immobility time in the forced swim and tail suspension tests. TAC1 knockouts also exhibited decreased latency to feed in a noveltysuppressed feeding test and increased social activity in males when exposed to a novel male, two additional anxiolytic-like behavioral effects. Although NK1 receptor agonists initially showed promise in treating depression, additional studies failed to confirm this efficacy. Meanwhile, a number of companies are continuing to study their utility in this area. Unfortunately, mice and rats do not have an emetic response, making it difficult to directly study the effects of these knockouts on nausea and vomiting. However, chemotherapeutic agents have been found to induce a type of behavior called pica, which is characterized by the increased consumption of kaolin, a type of clay, in mouse and rats [109,110]. Antiemetic agents reduce this behavior. It would be of great interest to study pica in NK1 receptor and TAC1 knockout mice.
9. CAN MOUSE KNOCKOUT DATA GUIDE NEW DRUG DISCOVERY? Overall, there is a high degree of correlation between the action of recently approved drugs in humans and the phenotypes for the
A Path to Innovation: Gene Knockouts Model New Drug Action
489
knockouts of their corresponding targets in mice. Twenty-five of these thirty new mechanisms are supported by the knockout data if one accepts that embryonic lethality is an important phenotype for directing oncology drug development. Only one knockout discussed in this review required a conditional mutation to understand the relevant target biology, the hippocampus-specific NMDA receptor knockout. The high degree of correlation between knockout phenotypes and drug action suggests that gene function for the majority of viable drug targets cannot be replaced through compensatory mechanisms in knockout mice. This makes sense since such mechanisms might be expected to compensate for both the genetic and the pharmacological inhibition of the target. In fact, lack of compensatory mechanisms may be an important if not essential quality for a good drug target. One area where it can be challenging to model current drug action using knockout mice is in neuroscience. There may be multiple reasons for this, but among the most likely are (1) many drugs that act on the CNS modulate more than one target; (2) difficulty correlating mouse behavior with therapeutic utility in humans; and (3) a related issue, the lack of reliable biomarkers by which to guide preclinical and clinical research. The atypical antipsychotics, for example, acting through multiple dopamine, serotonin, and histamine receptors, antidepressants such as bupropion acting through multiple monoamine transporters, and ramelteon mentioned above all exert their effects through modulation of multiple targets making their overall effect difficult to model in a single gene knockout mouse. Additionally, it is challenging to measure parameters such as anxiety, depression, and psychosis-related behavior in knockout mice. Targets outside of the CNS tend to have more discreet mechanisms of action and tend to regulate more basic, quantitative, and clinically relevant parameters of physiology such as blood glucose, body composition, bone mineral density, and blood pressure. Cognizant of the challenges, knockouts probably still offer one of the best opportunities to identify those rare targets that may provide a highly specific control point for particular neurobehavioral functions. This potential value of knockouts is further emphasized by the difficulty in identifying human disease– associated genes for CNS illnesses such as autism and schizophrenia. Over the past 6 years, drugs addressing 30 novel mechanisms of action have been commercialized. While the number of new molecular entities (NMEs) approved each year by the FDA is often used as a sign of industry productivity, the number of drugs addressing novel mechanisms may be a better measure of both productivity and innovation. Using this measure, the increase in yearly FDA approvals of drugs targeting novel mechanisms, from about two to three per year over the previous decade [36] to five per year based on the past six years, represents a dramatic increase in innovation. It is also worthwhile to consider that 12
490
Brian P. Zambrowicz and Arthur T. Sands
of these novel mechanisms are being addressed with peptide, protein, or antibody drugs, so called biologics. Additionally, four of these drugs (excluding enzyme replacement therapy) are for rare indications and may represent the forerunners of more personalized medicine. Together, this information suggests an increase in innovation in the pharmaceutical and biotechnology industry that is being enhanced by the increased development of biologics and drugs for orphan indications. The knockouts of the targets associated with recently approved drugs were most likely created concomitantly and independently of the drug development process. With the proliferation of mouse knockout technology and the now general recognition of its impact on the understanding of human biology (underscored by the 2007 Nobel Prize in Medicine), it is reasonable to assume that, in the future, knockouts of targets will precede the development of therapeutic agents addressing those targets. The saturation of the mammalian genome with knockouts is eminently achievable given that the total number of protein-encoding genes is now recognized to number only slightly more than 20,000. Indeed, the number of genes encoding proteins of unknown function that are amenable to small molecule or antibody intervention (the so-called druggable genome) is even less, probably around 5,000 in total. Given that we, as a single entity, have already completed the knockout and phenotypic analysis of 4,650 genes in this category, tackling the whole genome is clearly within the capabilities of a global knockout effort [111]. With access to a knockout for every gene, drug discovery researchers should look forward to a new era in which it will be possible to select targets based on predefined in vivo biologic function. This presupposes, however, not only the creation of large numbers of mouse knockouts, but more importantly, the in-depth phenotypic analysis, effective distribution, and widespread dissemination of the scientific discoveries made using knockout mice. While knockouts have been increasingly accepted as a method to study gene function, evidence supporting the concept of using them to guide drug discovery is as yet less well established. This is most likely due to the relatively recent advancement of knockout technology: the very first mouse knockout created through homologous recombination was published in 1990 [112]. Due to the additional time required for the technology to become widespread and the extended timeframes for drug development, there has not been sufficient time to bring novel drugs initiated based on knockout validation to market. However, one class of drug candidates showing positive phase 2 clinical data is the Orexin receptor antagonists for treating insomnia. The functions of Orexin [113] and its receptor, Orexin receptor 2 [114], were published simultaneously in 1999. The knockout of Orexin was found to result in a phenotype remarkably similar to canine narcolepsy. Narcolepsy is characterized by
A Path to Innovation: Gene Knockouts Model New Drug Action
491
daytime sleepiness, cataplexy, and sudden transitions to rapid eye movement (REM) sleep. Positional cloning was used to identify Orexin receptor 2 as the gene mutated in canine narcolepsy. These data caused the authors to suggest that Orexin and its receptors play an important role in the regulation of sleep and wakefulness [113]. Rare cases of human narcolepsy were subsequently found to be due to mutations in Orexin [115]. A logical hypothesis is that pharmacological antagonists of the Orexin receptors could have utility in promoting sleep. Actelion reported the development of a specific Orexin receptor 1 and 2 antagonist, ACT-078573, that promoted sleep when administered to rats, dogs, and humans [116] and is currently in phase 3 clinical studies. The list of drug discovery programs initiated due to knockout validation of a target will certainly grow over the coming years. Multiple other examples exist where the knockout phenotype of a target supported the further drug discovery efforts for a program. These include Cathepsin K inhibitors for osteoporosis [117] and DPP4 inhibitors for diabetes [13]. At the very least, specific knockout models can allow one to develop useful hypotheses regarding mechanism of action that can be tested, once small molecule modulators or antibodies have been obtained, in animal pharmacology experiments and ultimately in human clinical trials. Applied on a genome scale, knockouts hold the promise of providing a reliable path to innovation for the biopharmaceutical industry.
REFERENCES [1] E. S. Lander, L. M. Linton, B. Birren, C. Nusbaum, M. C. Zody, J. Baldwin, K. Devon, K. Dewar, M. Doyle, W. Fitzhugh, R. Funke, D. Gage, K. Harris, A. Heaford, J. Howland, L. Kann, J. Lehoczky, R. Levine, P. Mcewan, K. Mckernan, J. Meldrim, J. P. Mesirov, C. Miranda, W. Morris, J. Naylor, C. Raymond, M. Rosetti, R. Santos, A. Sheridan, C. Sougnez, N. Stange-Thomann, N. Stojanovic, A. Subramanian, D. Wyman, J. Rogers, J. Sulston, R. Ainscough, S. Beck, D. Bentley, J. Burton, C. Clee, N. Carter, A. Coulson, R. Deadman, P. Deloukas, A. Dunham, I. Dunham, R. Durbin, L. French, D. Grafham, S. Gregory, T. Hubbard, S. Humphray, A. Hunt, M. Jones, C. Lloyd, A. Mcmurray, L. Matthews, S. Mercer, S. Milne, J. C. Mullikin, A. Mungall, R. Plumb, M. Ross, R. Shownkeen, S. Sims, R. H. Waterston, R. K. Wilson, L. W. Hillier, J. D. Mcpherson, M. A. Marra, E. R. Mardis, L. A. Fulton, A. T. Chinwalla, K. H. Pepin, W. R. Gish, S. L. Chissoe, M. C. Wendl, K. D. Delehaunty, T. L. Miner, A. Delehaunty, J. B. Kramer, L. L. Cook, R. S. Fulton, D. L. Johnson, P. J. Minx, S. W. Clifton, T. Hawkins, E. Branscomb, P. Predki, P. Richardson, S. Wenning, T. Slezak, N. Doggett, J. F. Cheng, A. Olsen, S. Lucas, C. Elkin, E. Uberbacher, M. Frazier, R. A. Gibbs, D. M. Muzny, S. E. Scherer, J. B. Bouck, E. J. Sodergren, K. C. Worley, C. M. Rives, J. H. Gorrell, M. L. Metzker, S. L. Naylor, R. S. Kucherlapati, D. L. Nelson, G. M. Weinstock, Y. Sakaki, A. Fujiyama, M. Hattori, T. Yada, A. Toyoda, T. Itoh, C. Kawagoe, H. Watanabe, Y. Totoki, T. Taylor, J. Weissenbach, R. Heilig, W. Saurin, F. Artiguenave, P. Brottier, T. Bruls, E. Pelletier, C. Robert, P. Wincker, D. R. Smith, L. Doucette-Stamm, M. Rubenfield, K. Weinstock, H. M. Lee, J. Dubois, A. Rosenthal, M. Platzer, G. Nyakatura, S. Taudien,
492
Brian P. Zambrowicz and Arthur T. Sands
A. Davis, H. Yang, J. Yu, J. Wang, G. Huang, J. Gu, L. Hood, L. Rowen, A. Madan, S. Qin, R. W. Davis, N. A. Federspiel, A. P. Abola, M. J. Proctor, R. M. Myers, J. Schmutz, M. Dickson, J. Grimwood, D. R. Cox, M. V. Olson, R. Kaul, C. Raymond, N. Shimizu, K. Kawasaki, S. Minoshima, G. A. Evans, M. Athanasiou, R. Schultz, B. A. Roe, F. Chen, H. Pan, J. Ramser, H. Lehrach, R. Reinhardt, W. R. Mccombie, M. De La Bastide, N. Dedhia, H. Blocker, K. Hornischer, G. Nordsiek, R. Agarwala, L. Aravind, J. A. Bailey, A. Bateman, S. Batzoglou, E. Birney, P. Bork, D. G. Brown, C. B. Burge, L. Cerutti, H. C. Chen, D. Church, M. Clamp, R. R. Copley, T. Doerks, S. R. Eddy, E. E. Eichler, T. S. Furey, J. Galagan, J. G. Gilbert, C. Harmon, Y. Hayashizaki, D. Haussler, H. Hermjakob, K. Hokamp, W. Jang, L. S. Johnson, T. A. Jones, S. Kasif, A. Kaspryzk, S. Kennedy, W. J. Kent, P. Kitts, E. V. Koonin, I. Korf, D. Kulp, D. Lancet, T. M. Lowe, A. Mclysaght, T. Mikkelsen, J. V. Moran, N. Mulder, V. J. Pollara, C. P. Ponting, G. Schuler, J. Schultz, G. Slater, A. F. Smit, E. Stupka, J. Szustakowski, D. Thierry-Mieg, J. Thierry-Mieg, L. Wagner, J. Wallis, R. Wheeler, A. Williams, Y. I. Wolf, K. H. Wolfe, S. P. Yang, R. F. Yeh, F. Collins, M. S. Guyer, J. Peterson, A. Felsenfeld, K. A. Wetterstrand, A. Patrinos, M. J. Morgan, P. Jong, J. J. Catanese, K. Osoegawa, H. Shizuya, S. Choi and Y. J. Chen, Nature, 2001, 409(6822), 860. [2] J. C. Venter, M. D. Adams, E. W. Myers, P. W. Li, R. J. Mural, G. G. Sutton, H. O. Smith, M. Yandell, C. A. Evans, R. A. Holt, J. D. Gocayne, P. Amanatides, R. M. Ballew, D. H. Huson, J. R. Wortman, Q. Zhang, C. D. Kodira, X. H. Zheng, L. Chen, M. Skupski, G. Subramanian, P. D. Thomas, J. Zhang, G. L. Gabor Miklos, C. Nelson, S. Broder, A. G. Clark, J. Nadeau, V. A. Mckusick, N. Zinder, A. J. Levine, R. J. Roberts, M. Simon, C. Slayman, M. Hunkapiller, R. Bolanos, A. Delcher, I. Dew, D. Fasulo, M. Flanigan, L. Florea, A. Halpern, S. Hannenhalli, S. Kravitz, S. Levy, C. Mobarry, K. Reinert, K. Remington, J. Abu-Threideh, E. Beasley, K. Biddick, V. Bonazzi, R. Brandon, M. Cargill, I. Chandramouliswaran, R. Charlab, K. Chaturvedi, Z. Deng, V. Di Francesco, P. Dunn, K. Eilbeck, C. Evangelista, A. E. Gabrielian, W. Gan, W. Ge, F. Gong, Z. Gu, P. Guan, T. J. Heiman, M. E. Higgins, R. R. Ji, Z. Ke, K. A. Ketchum, Z. Lai, Y. Lei, Z. Li, J. Li, Y. Liang, X. Lin, F. Lu, G. V. Merkulov, N. Milshina, H. M. Moore, A. K. Naik, V. A. Narayan, B. Neelam, D. Nusskern, D. B. Rusch, S. Salzberg, W. Shao, B. Shue, J. Sun, Z. Wang, A. Wang, X. Wang, J. Wang, M. Wei, R. Wides, C. Xiao, C. Yan, A. Yao, J. Ye, M. Zhan, W. Zhang, H. Zhang, Q. Zhao, L. Zheng, F. Zhong, W. Zhong, S. Zhu, S. Zhao, D. Gilbert, S. Baumhueter, G. Spier, C. Carter, A. Cravchik, T. Woodage, F. Ali, H. An, A. Awe, D. Baldwin, H. Baden, M. Barnstead, I. Barrow, K. Beeson, D. Busam, A. Carver, A. Center, M. L. Cheng, L. Curry, S. Danaher, L. Davenport, R. Desilets, S. Dietz, K. Dodson, L. Doup, S. Ferriera, N. Garg, A. Gluecksmann, B. Hart, J. Haynes, C. Haynes, C. Heiner, S. Hladun, D. Hostin, J. Houck, T. Howland, C. Ibegwam, J. Johnson, F. Kalush, L. Kline, S. Koduru, A. Love, F. Mann, D. May, S. Mccawley, T. Mcintosh, I. Mcmullen, M. Moy, L. Moy, B. Murphy, K. Nelson, C. Pfannkoch, E. Pratts, V. Puri, H. Qureshi, M. Reardon, R. Rodriguez, Y. H. Rogers, D. Romblad, B. Ruhfel, R. Scott, C. Sitter, M. Smallwood, E. Stewart, R. Strong, E. Suh, R. Thomas, N. N. Tint, S. Tse, C. Vech, G. Wang, J. Wetter, S. Williams, M. Williams, S. Windsor, E. Winn-Deen, K. Wolfe, J. Zaveri, K. Zaveri, J. F. Abril, R. Guigo, M. J. Campbell, K. V. Sjolander, B. Karlak, A. Kejariwal, H. Mi, B. Lazareva, T. Hatton, A. Narechania, K. Diemer, A. Muruganujan, N. Guo, S. Sato, V. Bafna, S. Istrail, R. Lippert, R. Schwartz, B. Walenz, S. Yooseph, D. Allen, A. Basu, J. Baxendale, L. Blick, M. Caminha, J. Carnes-Stine, P. Caulk, Y. H. Chiang, M. Coyne, C. Dahlke, A. Mays, M. Dombroski, M. Donnelly, D. Ely, S. Esparham, C. Fosler, H. Gire, S. Glanowski, K. Glasser, A. Glodek, M. Gorokhov, K. Graham, B. Gropman, M. Harris, J. Heil, S. Henderson, J. Hoover, D. Jennings, C. Jordan, J. Jordan, J. Kasha, L. Kagan, C. Kraft, A. Levitsky, M. Lewis, X. Liu, J. Lopez, D. Ma, W. Majoros, J. Mcdaniel, S. Murphy, M. Newman, T. Nguyen, N. Nguyen, M. Nodell, S. Pan, J. Peck, M. Peterson, W. Rowe, R. Sanders, J.
A Path to Innovation: Gene Knockouts Model New Drug Action
[3] [4]
[5] [6]
[7] [8] [9] [10] [11]
[12]
493
Scott, M. Simpson, T. Smith, A. Sprague, T. Stockwell, R. Turner, E. Venter, M. Wang, M. Wen, D. Wu, M. Wu, A. Xia, A. Zandieh and X. Zhu, Science, 2001, 291(5507), 1304. M. Clamp, B. Fry, M. Kamal, X. Xie, J. Cuff, M. F. Lin, M. Kellis, K. Lindblad-Toh and E. S. Lander, Proc. Natl. Acad. Sci. USA, 2007, 104(49), 19428. R. H. Waterston, K. Lindblad-Toh, E. Birney, J. Rogers, J. F. Abril, P. Agarwal, R. Agarwala, R. Ainscough, M. Alexandersson, P. An, S. E. Antonarakis, J. Attwood, R. Baertsch, J. Bailey, K. Barlow, S. Beck, E. Berry, B. Birren, T. Bloom, P. Bork, M. Botcherby, N. Bray, M. R. Brent, D. G. Brown, S. D. Brown, C. Bult, J. Burton, J. Butler, R. D. Campbell, P. Carninci, S. Cawley, F. Chiaromonte, A. T. Chinwalla, D. M. Church, M. Clamp, C. Clee, F. S. Collins, L. L. Cook, R. R. Copley, A. Coulson, O. Couronne, J. Cuff, V. Curwen, T. Cutts, M. Daly, R. David, J. Davies, K. D. Delehaunty, J. Deri, E. T. Dermitzakis, C. Dewey, N. J. Dickens, M. Diekhans, S. Dodge, I. Dubchak, D. M. Dunn, S. R. Eddy, L. Elnitski, R. D. Emes, P. Eswara, E. Eyras, A. Felsenfeld, G. A. Fewell, P. Flicek, K. Foley, W. N. Frankel, L. A. Fulton, R. S. Fulton, T. S. Furey, D. Gage, R. A. Gibbs, G. Glusman, S. Gnerre, N. Goldman, L. Goodstadt, D. Grafham, T. A. Graves, E. D. Green, S. Gregory, R. Guigo, M. Guyer, R. C. Hardison, D. Haussler, Y. Hayashizaki, L. W. Hillier, A. Hinrichs, W. Hlavina, T. Holzer, F. Hsu, A. Hua, T. Hubbard, A. Hunt, I. Jackson, D. B. Jaffe, L. S. Johnson, M. Jones, T. A. Jones, A. Joy, M. Kamal, E. K. Karlsson, D. Karolchik, A. Kasprzyk, J. Kawai, E. Keibler, C. Kells, W. J. Kent, A. Kirby, D. L. Kolbe, I. Korf, R. S. Kucherlapati, E. J. Kulbokas, D. Kulp, T. Landers, J. P. Leger, S. Leonard, I. Letunic, R. Levine, J. Li, M. Li, C. Lloyd, S. Lucas, B. Ma, D. R. Maglott, E. R. Mardis, L. Matthews, E. Mauceli, J. H. Mayer, M. Mccarthy, W. R. Mccombie, S. Mclaren, K. Mclay, J. D. Mcpherson, J. Meldrim, B. Meredith, J. P. Mesirov, W. Miller, T. L. Miner, E. Mongin, K. T. Montgomery, M. Morgan, R. Mott, J. C. Mullikin, D. M. Muzny, W. E. Nash, J. O. Nelson, M. N. Nhan, R. Nicol, Z. Ning, C. Nusbaum, M. J. O’connor, Y. Okazaki, K. Oliver, E. Overton-Larty, L. Pachter, G. Parra, K. H. Pepin, J. Peterson, P. Pevzner, R. Plumb, C. S. Pohl, A. Poliakov, T. C. Ponce, C. P. Ponting, S. Potter, M. Quail, A. Reymond, B. A. Roe, K. M. Roskin, E. M. Rubin, A. G. Rust, R. Santos, V. Sapojnikov, B. Schultz, J. Schultz, M. S. Schwartz, S. Schwartz, C. Scott, S. Seaman, S. Searle, T. Sharpe, A. Sheridan, R. Shownkeen, S. Sims, J. B. Singer, G. Slater, A. Smit, D. R. Smith, B. Spencer, A. Stabenau, N. Stange-Thomann, C. Sugnet, M. Suyama, G. Tesler, J. Thompson, D. Torrents, E. Trevaskis, J. Tromp, C. Ucla, A. Ureta-Vidal, J. P. Vinson, A. C. Von Niederhausern, C. M. Wade, M. Wall, R. J. Weber, R. B. Weiss, M. C. Wendl, A. P. West, K. Wetterstrand, R. Wheeler, S. Whelan, J. Wierzbowski, D. Willey, S. Williams, R. K. Wilson, E. Winter, K. C. Worley, D. Wyman, S. Yang, S. P. Yang, E. M. Zdobnov, M. C. Zody and E. S. Lander, Nature, 2002, 420(6915), 520. C. Gunter and R. Dhand, Nature, 2002, 420(6915), 509. B. P. Zambrowicz, K. H. Holt, W. D. Walke, L. L. Kirkpatrick and D. E. Eberhart, in Target Validation in Drug Discovery (eds B. W. Metcalf and S. Dillon), Academic Press, Burlington, MA, 2007, p. 3. B. P. Zambrowicz and A. T. Sands, Nat. Rev. Drug. Discov., 2003, 2(1), 38. B. P. Zambrowicz, C. A. Turner and A. T. Sands, Curr. Opin. Pharmacol., 2003, 3(5), 563. M. C. Riddle and D. J. Drucker, Diabetes Care, 2006, 29(2), 435. L. A. Scrocchi, T. J. Brown, N. Maclusky, P. L. Brubaker, A. B. Auerbach, A. L. Joyner and D. J. Drucker, Nat. Med., 1996, 2(11), 1254. K. Miyawaki, Y. Yamada, H. Yano, H. Niwa, N. Ban, Y. Ihara, A. Kubota, S. Fujimoto, M. Kajikawa, A. Kuroe, K. Tsuda, H. Hashimoto, T. Yamashita, T. Jomori, F. Tashiro, J. Miyazaki and Y. Seino, Proc. Natl. Acad. Sci. USA, 1999, 96(26), 14843. N. Pamir, F. C. Lynn, A. M. Buchan, J. Ehses, S. A. Hinke, J. A. Pospisilik, K. Miyawaki, Y. Yamada, Y. Seino, C. H. Mcintosh and R. A. Pederson, Am. J. Physiol. Endocrinol. Metab., 2003, 284(5), E931.
494
Brian P. Zambrowicz and Arthur T. Sands
[13] D. Marguet, L. Baggio, T. Kobayashi, A. M. Bernard, M. Pierres, P. F. Nielsen, U. Ribel, T. Watanabe, D. J. Drucker and N. Wagtmann, Proc. Natl. Acad. Sci. USA, 2000, 97(12), 6874. [14] S. L. Conarello, Z. Li, J. Ronan, R. S. Roy, L. Zhu, G. Jiang, F. Liu, J. Woods, E. Zycband, D. E. Moller, N. A. Thornberry and B. B. Zhang, Proc. Natl. Acad. Sci. USA, 2003, 100(11), 6825. [15] F. Preitner, M. Ibberson, I. Franklin, C. Binnert, M. Pende, A. Gjinovci, T. Hansotia, D. J. Drucker, C. Wollheim, R. Burcelin and B. Thorens, J. Clin. Invest., 2004, 113(4), 635. [16] T. Hansotia, L. L. Baggio, D. Delmeire, S. A. Hinke, Y. Yamada, K. Tsukiyama, Y. Seino, J. J. Holst, F. Schuit and D. J. Drucker, Diabetes, 2004, 53(5), 1326. [17] S. Gebre-Medhin, H. Mulder, M. Pekny, G. Westermark, J. Tornell, P. Westermark, F. Sundler, B. Ahren and C. Betsholtz, Biochem. Biophys. Res. Commun., 1998, 250(2), 271. [18] H. Mulder, S. Gebre-Medhin, C. Betsholtz, F. Sundler and B. Ahren, Am. J. Physiol. Endocrinol. Metab., 2000, 278(4), E684. [19] M. D. Gershon, Aliment. Pharmacol. Ther., 2004, 20(Suppl. 7), 3. [20] M. R. Bosl, V. Stein, C. Hubner, A. A. Zdebik, S. E. Jordt, A. K. Mukhopadhyay, M. S. Davidoff, A. F. Holstein and T. J. Jentsch, EMBO J., 2001, 20(6), 1289. [21] W. Rosenthal, A. Antaramian, S. Gilbert and M. Birnbaumer, J. Biol. Chem., 1993, 268(18), 13030. [22] E. J. Holtzman and D. A. Ausiello, Hosp. Pract. (Off. Ed), 1994, 29(3), 89. [23] J. Yun, T. Schoneberg, J. Liu, A. Schulz, C. A. Ecelbarger, D. Promeneur, S. Nielsen, H. Sheng, A. Grinberg, C. Deng and J. Wess, J. Clin. Invest., 2000, 106(11), 1361. [24] D. S. Geller, J. Rodriguez-Soriano, A. Vallo Boado, S. Schifter, M. Bayer, S. S. Chang and R. P. Lifton, Nat. Genet., 1998, 19(3), 279. [25] S. Berger, M. Bleich, W. Schmid, R. Greger and G. Schutz, Kidney Int., 2000, 57(4), 1295. [26] A. F. Clark, M. G. Sharp, S. D. Morley, S. Fleming, J. Peters and J. J. Mullins, J. Biol. Chem., 1997, 272(29), 18185. [27] M. G. Sharp, D. Fettes, G. Brooker, A. F. Clark, J. Peters, S. Fleming and J. J. Mullins, Hypertension, 1996, 28(6), 1126. [28] H. R. Davis, Jr., L. J. Zhu, L. M. Hoos, G. Tetzloff, M. Maguire, J. Liu, X. Yao, S. P. Iyer, M. H. Lam, E. G. Lund, P. A. Detmers, M. P. Graziano and S. W. Altmann, J. Biol. Chem., 2004, 279(32), 33586. [29] M. Garcia-Calvo, J. Lisnock, H. G. Bull, B. E. Hawes, D. A. Burnett, M. P. Braun, J. H. Crona, H. R. Davis, Jr., D. C. Dean, P. A. Detmers, M. P. Graziano, M. Hughes, D. E. Macintyre, A. Ogawa, K. A. O’neill, S. P. Iyer, D. E. Shevell, M. M. Smith, Y. S. Tang, A. M. Makarewicz, F. Ujjainwalla, S. W. Altmann, K. T. Chapman and N. A. Thornberry, Proc. Natl. Acad. Sci. USA, 2005, 102(23), 8132. [30] C. A. Chambers, T. J. Sullivan and J. P. Allison, Immunity, 1997, 7(6), 885. [31] D. Dombrowicz, V. Flamand, K. K. Brigman, B. H. Koller and J. P. Kinet, Cell, 1993, 75(5), 969. [32] S. I. Mayr, R. I. Zuberi, M. Zhang, J. Sousa-Hitzler, K. Ngo, Y. Kuwabara, L. Yu, W. P. Fung-Leung and F. T. Liu, J. Immunol., 2002, 169(4), 2061. [33] J. L. Dunne, C. M. Ballantyne, A. L. Beaudet and K. Ley, Blood, 2002, 99(1), 336. [34] D. A. Steeber, M. A. Campbell, A. Basit, K. Ley and T. F. Tedder, Proc. Natl. Acad. Sci. USA, 1998, 95(13), 7562. [35] C. A. Hatfield, J. R. Brashler, G. E. Winterrowd, F. P. Bell, R. L. Griffin, S. F. Fidler, K. P. Kolbasa, J. L. Mobley, K. L. Shull, I. M. Richards and J. E. Chin, Am. J. Physiol., 1997, 273(3 Pt 1), L513. [36] F. Bendjelloul, P. Maly, V. Mandys, M. Jirkovska, L. Prokesova, L. Tuckova and H. Tlaskalova-Hogenova, Clin. Exp. Immunol., 2000, 119(1), 57. [37] D. C. Bullard, L. A. Hurley, I. Lorenzo, L. M. Sly, A. L. Beaudet and N. D. Staite, J. Immunol., 1996, 157(7), 3153.
A Path to Innovation: Gene Knockouts Model New Drug Action
495
[38] C. G. Kevil, M. J. Hicks, X. He, J. Zhang, C. M. Ballantyne, C. Raman, T. R. Schoeb and D. C. Bullard, Am. J. Pathol., 2004, 165(2), 609. [39] P. Shier, G. Otulakowski, K. Ngo, J. Panakos, E. Chourmouzis, L. Christjansen, C. Y. Lau and W. P. Fung-Leung, J. Immunol., 1996, 157(12), 5375. [40] G. M. Watts, F. J. Beurskens, I. Martin-Padura, C. M. Ballantyne, L. B. Klickstein, M. B. Brenner and D. M. Lee, J. Immunol., 2005, 174(6), 3668. [41] S. Grabbe, G. Varga, S. Beissert, M. Steinert, G. Pendl, S. Seeliger, W. Bloch, T. Peters, T. Schwarz, C. Sunderkotter and K. Scharffetter-Kochanek, J. Clin. Invest., 2002, 109(2), 183. [42] J. P. Mizgerd, H. Kubo, G. J. Kutkoski, S. D. Bhagwan, K. Scharffetter-Kochanek, A. L. Beaudet and C. M. Doerschuk, J. Exp. Med., 1997, 186(8), 1357. [43] N. Wagner, J. Lohler, E. J. Kunkel, K. Ley, E. Leung, G. Krissansen, K. Rajewsky and W. Muller, Nature, 1996, 382(6589), 366. [44] O. Pabst, R. Forster, M. Lipp, H. Engel and H. H. Arnold, EMBO J., 2000, 19(9), 2015. [45] E. J. Kunkel, C. L. Ramos, D. A. Steeber, W. Muller, N. Wagner, T. F. Tedder and K. Ley, J. Immunol., 1998, 161(5), 2449. [46] B. C. Sydora, N. Wagner, J. Lohler, G. Yakoub, M. Kronenberg, W. Muller and R. Aranda, Clin. Exp. Immunol., 2002, 129(1), 35. [47] X. Sun, H. Qiao, J. Shi, J. R. Kanwar, W. Mueller, N. Wagner and G. W. Krissansen, Transplantation, 2002, 74(8), 1202. [48] J. T. Yang, H. Rayburn and R. O. Hynes, Development, 1995, 121(2), 549. [49] L. E. Stephens, A. E. Sutherland, I. V. Klimanskaya, A. Andrieux, J. Meneses, R. A. Pedersen and C. H. Damsky, Genes Dev., 1995, 9(15), 1883. [50] R. Fassler and M. Meyer, Genes Dev., 1995, 9(15), 1896. [51] G. C. Gurtner, V. Davis, H. Li, M. J. Mccoy, A. Sharpe and M. I. Cybulsky, Genes Dev., 1995, 9(1), 1. [52] L. M. Scott, G. V. Priestley and T. Papayannopoulou, Mol. Cell. Biol., 2003, 23(24), 9349. [53] N. Killeen, S. G. Stuart and D. R. Littman, EMBO J., 1992, 11(12), 4329. [54] D. S. Holt, M. Botto, A. E. Bygrave, S. M. Hanna, M. J. Walport and B. P. Morgan, Blood, 2001, 98(2), 442. [55] M. B. Fischer, M. Ma, S. Goerg, X. Zhou, J. Xia, O. Finco, S. Han, G. Kelsoe, R. G. Howard, T. L. Rothstein, E. Kremmer, F. S. Rosen and M. C. Carroll, J. Immunol., 1996, 157(2), 549. [56] R. A. Wetsel, D. T. Fleischer and D. L. Haviland, J. Biol. Chem., 1990, 265(5), 2435. [57] V. Rosti, G. Tremml, V. Soares, P. P. Pandolfi, L. Luzzatto and M. Bessler, J. Clin. Invest., 1997, 100(5), 1028. [58] G. Tremml, C. Dominguez, V. Rosti, Z. Zhang, P. P. Pandolfi, P. Keller and M. Bessler, Blood, 1999, 94(9), 2945. [59] X. Qin, M. Dobarro, S. J. Bedford, S. Ferris, P. V. Miranda, W. Song, R. T. Bronson, P. E. Visconti and J. A. Halperin, J. Immunol., 2005, 175(10), 6294. [60] M. Grompe, M. Al-Dhalimy, M. Finegold, C. N. Ou, T. Burlingame, N. G. Kennaway and P. Soriano, Genes Dev., 1993, 7(12A), 2298. [61] F. Endo and M. S. Sun, J. Inherit. Metab. Dis., 2002, 25(3), 227. [62] J. P. Liu, J. Baker, A. S. Perkins, E. J. Robertson and A. Efstratiadis, Cell, 1993, 75(1), 59. [63] L. Powell-Braxton, P. Hollingshead, C. Warburton, M. Dowd, S. Pitts-Meek, D. Dalton, N. Gillett and T. A. Stewart, Genes Dev., 1993, 7(12B), 2609. [64] K. T. Coschigano, A. N. Holland, M. E. Riders, E. O. List, A. Flyvbjerg and J. J. Kopchick, Endocrinology, 2003, 144(9), 3799. [65] F. Shalaby, J. Rossant, T. P. Yamaguchi, M. Gertsenstein, X. F. Wu, M. L. Breitman and A. C. Schuh, Nature, 1995, 376(6535), 62. [66] D. J. Dumont, L. Jussila, J. Taipale, A. Lymboussaki, T. Mustonen, K. Pajusola, M. Breitman and K. Alitalo, Science, 1998, 282(5390), 946.
496
Brian P. Zambrowicz and Arthur T. Sands
[67] P. Soriano, Genes Dev., 1994, 8(16), 1888. [68] S. Rajaraman, L. K. Wood, D. K. Willhite, L. B. Russell and M. A. Bedell, Mamm. Genome, 2003, 14(3), 168. [69] V. Munugalavadla and R. Kapur, Crit. Rev. Oncol. Hematol., 2005, 54(1), 63. [70] K. Mackarehtschian, J. D. Hardin, K. A. Moore, S. Boast, S. P. Goff and I. R. Lemischka, Immunity, 1995, 3(1), 147. [71] H. J. Mckenna, K. L. Stocking, R. E. Miller, K. Brasel, T. De Smedt, E. Maraskovsky, C. R. Maliszewski, D. H. Lynch, J. Smith, B. Pulendran, E. R. Roux, M. Teepe, S. D. Lyman and J. J. Peschon, Blood, 2000, 95(11), 3489. [72] M. Mikula, M. Schreiber, Z. Husak, L. Kucerova, J. Ruth, R. Wieser, K. Zatloukal, H. Beug, E. F. Wagner and M. Baccarini, EMBO J., 2001, 20(8), 1952. [73] L. Wojnowski, A. M. Zimmer, T. W. Beck, H. Hahn, R. Bernal, U. R. Rapp and A. Zimmer, Nat. Genet., 1997, 16(3), 293. [74] X. M. Dai, G. R. Ryan, A. J. Hapel, M. G. Dominguez, R. G. Russell, S. Kapp, V. Sylvestre and E. R. Stanley, Blood, 2002, 99(1), 111. [75] G. H. Fong, J. Rossant, M. Gertsenstein and M. L. Breitman, Nature, 1995, 376 (6535), 66. [76] P. Soriano, Development, 1997, 124(14), 2691. [77] A. Schuchardt, V. D’agati, L. Larsson-Blomberg, F. Costantini and V. Pachnis, Nature, 1994, 367(6461), 380. [78] P. Carmeliet, V. Ferreira, G. Breier, S. Pollefeyt, L. Kieckens, M. Gertsenstein, M. Fahrig, A. Vandenhoeck, K. Harpal, C. Eberhardt, C. Declercq, J. Pawling, L. Moons, D. Collen, W. Risau and A. Nagy, Nature, 1996, 380(6573), 435. [79] N. Ferrara, K. Carver-Moore, H. Chen, M. Dowd, L. Lu, K. S. O’shea, L. PowellBraxton, K. J. Hillan and M. W. Moore, Nature, 1996, 380(6573), 439. [80] W. Lin, H. B. Sanchez, T. Deerinck, J. K. Morris, M. Ellisman and K. F. Lee, Proc. Natl. Acad. Sci. USA, 2000, 97(3), 1299. [81] D. W. Threadgill, A. A. Dlugosz, L. A. Hansen, T. Tennenbaum, U. Lichti, D. Yee, C. Lamantia, T. Mourton, K. Herrup and R. C. Harris, Science, 1995, 269(5221), 230. [82] P. J. Miettinen, J. R. Chin, L. Shum, H. C. Slavkin, C. F. Shuler, R. Derynck and Z. Werb, Nat. Genet., 1999, 22(1), 69. ´ carroll, M. Rembold, H. Khier, J. Tischler, G. Weitzer, B. [83] G. Lagger, D. O Schucttengruber, C. Hauser, R. Brunmeir, T. Jenuwein and C. Seiser, EMBO J., 2002, 21(11), 2672. [84] M. Oka, A. M. Meacham, T. Hamazaki, N. Rodic, L. J. Chang and N. Terada, Oncogene, 2005, 24(19), 3091. [85] M. Okano, D. W. Bell, D. A. Haber and E. Li, Cell, 1999, 99(3), 247. [86] E. Georgatsou, T. Georgakopoulos and G. Thireos, FEBS Lett., 1992, 299(1), 39. [87] D. H. Lee, K. Tanaka, T. Tamura, C. H. Chung and A. Ichihara, Biochem. Biophys. Res. Commun., 1992, 182(2), 452. [88] T. Fujiwara, K. Tanaka, E. Orino, T. Yoshimura, A. Kumatori, T. Tamura, C. H. Chung, T. Nakai, K. Yamaguchi, S. Shin, et al., J. Biol. Chem., 1990, 265(27), 16604. [89] T. Szlanka, L. Haracska, I. Kiss, P. Deak, E. Kurucz, I. Ando, E. Viragh and A. Udvardy, J. Cell Sci., 2003, 116(Pt 6), 1023. [90] K. A. Smyth and J. M. Belote, Genetics, 1999, 151(1), 211. [91] K. J. Saville and J. M. Belote, Proc. Natl. Acad. Sci. USA, 1993, 90(19), 8842. [92] M. Takahashi, H. Iwasaki, H. Inoue and K. Takahashi, Biol. Chem., 2002, 383(7–8), 1263. [93] K. Suzuki, K. Yamanishi, O. Mori, M. Kamikawa, B. Andersen, S. Kato, T. Toyoda and G. Yamada, FEBS Lett., 2000, 481(1), 53. [94] M. Soda, Y. L. Choi, M. Enomoto, S. Takada, Y. Yamashita, S. Ishikawa, S. Fujiwara, H. Watanabe, K. Kurashina, H. Hatanaka, M. Bando, S. Ohno, Y. Ishikawa, H. Aburatani, T. Niki, Y. Sohara, Y. Sugiyama and H. Mano, Nature, 2007, 448(7153), 561.
A Path to Innovation: Gene Knockouts Model New Drug Action
497
[95] H. Saegusa, Y. Matsuda and T. Tanabe, Neurosci. Res., 2002, 43(1), 1. [96] H. Saegusa, T. Kurihara, S. Zong, A. Kazuno, Y. Matsuda, T. Nonaka, W. Han, H. Toriyama and T. Tanabe, EMBO J., 2001, 20(10), 2349. [97] C. Kim, K. Jun, T. Lee, S. S. Kim, M. W. Mcenery, H. Chin, H. L. Kim, J. M. Park, D. K. Kim, S. J. Jung, J. Kim and H. S. Shin, Mol. Cell. Neurosci., 2001, 18(2), 235. [98] J. Z. Tsien, P. T. Huerta and S. Tonegawa, Cell, 1996, 87(7), 1327. [99] E. Shimizu, Y. P. Tang, C. Rampon and J. Z. Tsien, Science, 2000, 290(5494), 1170. [100] L. Rondi-Reig, M. Libbey, H. Eichenbaum and S. Tonegawa, Proc. Natl. Acad. Sci. USA, 2001, 98(6), 3543. [101] K. Nakazawa, L. D. Sun, M. C. Quirk, L. Rondi-Reig, M. A. Wilson and S. Tonegawa, Neuron, 2003, 38(2), 305. [102] L. Rondi-Reig, G. H. Petit, C. Tobin, S. Tonegawa, J. Mariani and A. Berthoz, J. Neurosci., 2006, 26(15), 4071. [103] A. Jilg, J. Moek, D. R. Weaver, H. W. Korf, J. H. Stehle and C. Von Gall, Eur. J. Neurosci., 2005, 22(11), 2845. [104] X. Jin, C. Gall, R. L. Pieschl, V. K. Gribkoff, J. H. Stehle, S. M. Reppert and D. R. Weaver, Mol. Cell. Biol., 2003, 23(3), 1054. [105] Z. M. Weil, A. K. Hotchkiss, M. L. Gatien, S. Pieke-Dahl and R. J. Nelson, Brain Res. Bull., 2006, 68(6), 425. [106] L. Santarelli, G. Gobbi, P. C. Debs, E. T. Sibille, P. Blier, R. Hen and M. J. Heath, Proc. Natl. Acad. Sci. USA, 2001, 98(4), 1912. [107] A. Bilkei-Gorzo, I. Racz, K. Michel and A. Zimmer, J. Neurosci., 2002, 22(22), 10046. [108] J. M. Laird, T. Olivar, C. Roza, C. De Felipe, S. P. Hunt and F. Cervero, Neuroscience, 2000, 98(2), 345. [109] K. Yamamoto, S. Matsunaga, M. Matsui, N. Takeda and A. Yamatodani, Methods Find. Exp. Clin. Pharmacol., 2002, 24(3), 135. [110] K. Yamamoto, M. Nakai, K. Nohara and A. Yamatodani, Eur. J. Pharmacol., 2007, 554(1), 34. [111] C. P. Austin, J. F. Battey, A. Bradley, M. Bucan, M. Capecchi, F. S. Collins, W. F. Dove, G. Duyk, S. Dymecki, J. T. Eppig, F. B. Grieder, N. Heintz, G. Hicks, T. R. Insel, A. Joyner, B. H. Koller, K. C. Lloyd, T. Magnuson, M. W. Moore, A. Nagy, J. D. Pollock, A. D. Roses, A. T. Sands, B. Seed, W. C. Skarnes, J. Snoddy, P. Soriano, D. J. Stewart, F. Stewart, B. Stillman, H. Varmus, L. Varticovski, I. M. Verma, T. F. Vogt, H. Melchner, J. Witkowski, R. P. Woychik, W. Wurst, G. D. Yancopoulos, S. G. Young and B. Zambrowicz, Nat. Genet., 2004, 36(9), 921. [112] K. R. Thomas and M. R. Capecchi, Nature, 1990, 346(6287), 847. [113] R. M. Chemelli, J. T. Willie, C. M. Sinton, J. K. Elmquist, T. Scammell, C. Lee, J. A. Richardson, S. C. Williams, Y. Xiong, Y. Kisanuki, T. E. Fitch, M. Nakazato, R. E. Hammer, C. B. Saper and M. Yanagisawa, Cell, 1999, 98(4), 437. [114] L. Lin, J. Faraco, R. Li, H. Kadotani, W. Rogers, X. Lin, X. Qiu, P. J. Jong, S. Nishino and E. Mignot, Cell, 1999, 98(3), 365. [115] C. Peyron, J. Faraco, W. Rogers, B. Ripley, S. Overeem, Y. Charnay, S. Nevsimalova, M. Aldrich, D. Reynolds, R. Albin, R. Li, M. Hungs, M. Pedrazzoli, M. Padigaru, M. Kucherlapati, J. Fan, R. Maki, G. J. Lammers, C. Bouras, R. Kucherlapati, S. Nishino and E. Mignot, Nat. Med., 2000, 6(9), 991. [116] C. Brisbare-Roch, J. Dingemanse, R. Koberstein, P. Hoever, H. Aissaoui, S. Flores, C. Mueller, O. Nayler, J. Van Gerven, S. L. De Haas, P. Hess, C. Qiu, S. Buchmann, M. Scherz, T. Weller, W. Fischli, M. Clozel and F. Jenck, Nat. Med., 2007, 13(2), 150. [117] P. Saftig, E. Hunziker, O. Wehmeyer, S. Jones, A. Boyde, W. Rommerskirch, J. D. Moritz, P. Schu and K. Von Figura, Proc. Natl. Acad. Sci. USA, 1998, 95(23), 13453.
CHAPT ER
24 Discovery of Novel Positron Emission Tomography Tracers Dennis J. McCarthy*, Christer Halldin**, Jan D. Andersson** and M. Edward Pierson*
Contents
1. 2. 3. 4.
Introduction PET Imaging Discovery of a New PET Tracer for Serotonin 5-HT1B Steps Needed to Develop PET Tracers 4.1 Physical properties 4.2 Tracer validation 4.3 Regulatory considerations 5. PET Tracers as Translational Tools 5.1 Receptor occupancy (getting the dose correct) 5.2 PET microdosing 5.3 Pharmacodynamic measurements 5.4 Studying disease pathophysiology 6. New PET Tracers 7. Conclusions References
501 502 502 506 506 506 507 508 508 508 509 509 509 511 512
1. INTRODUCTION Molecular imaging is a non-invasive technique that allows for the imaging of particular targets or pathways in a living organism. This technique has application in our understanding of biological processes in * AstraZeneca Pharmaceuticals, 1800 Concord Pike, Wilmington, DE 19803, USA ** Karolinska Institutet, SE-171 77 Stockholm, Sweden Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04424-8
r 2009 Elsevier Inc. All rights reserved.
501
502
Dennis J. McCarthy et al.
a living system, with particular application in understanding disease etiology, progression, and diagnosis. Molecular imaging methods cover a wide range of techniques, including positron emission tomography (PET), magnetic resonance imaging (MRI), single photon emission computed tomography (SPECT), ultrasound, and optical imaging. In this review, we will focus on PET imaging, the discovery of novel PET tracers, and their use in speeding drug development.
2. PET IMAGING In PET, a molecule is tagged with a short-lived positron-emitting isotope. During the radioactive decay, a proton in the nucleus transforms into a neutron and a highly energetic positron that is ejected from the nucleus and the positron collides with a nearby electron resulting in the emission of two 511-keV annihilation photons emitted 180 degrees from each other. A scanner then estimates the density of positron annihilations in a specific area of the body. Computer algorithms are used to visualize the density of the original molecule in the areas of interest. Typical radioisotopes used for clinical investigations include 11C, 13N, 15O, and 18 F. Using appropriate radiotracers, PET can be used to study interactions with specific drug targets, assess pharmacodynamic (PD) response, visualize disease pathology, and monitor the effect of treatment. The radiotracer is administered at such a low dose that it generates no pharmacological effects. A disadvantage of PET is that the radiotracers must be produced with a cyclotron. Since the radioisotopes have halflives measured in minutes or hours the cyclotron, in most cases, needs to be located at the site of the study. An onsite cyclotron is necessary for the shorter-lived isotopes such as 11C (t1/2 B 20 min), 13N (t1/2 B 10 min), and 15O (t1/2 B 2 min). Tracers labeled with the longer lived 18 F (t1/2 B 110 min) can be synthesized at one facility and then transported to a remote imaging facility, thus enabling greater clinical access to 18F PET tracers.
3. DISCOVERY OF A NEW PET TRACER FOR SEROTONIN 5-HT1B The development of new PET tracers requires a multidisciplinary approach combining the expertise of many fields, encompassing radiochemistry, medicinal chemistry, biology, pharmacology, drug metabolism, mathematics, and regulatory affairs. In addition, new tracer development requires a dedicated technical staff with access to a cyclotron and PET camera. By way of illustration of these many aspects,
503
Discovery of Novel Positron Emission Tomography Tracers
CH3 O
CH3 O
O N N H
O
H N
[11C]MeOTf
H N
O N N O
N
O
N O
11CH 3
Figure 1 Synthesis of [11C]AZ10419369.
we describe the process that led to the discovery of a novel PET tracer for the serotonin 5-HT1B receptor. AZ10419369 (5-methyl-8-(4-methyl-piperazin-1-yl)-4-oxo-4H-chromene2-carboxylic acid (4-morpholin-4-yl-phenyl)-amide) is a selective receptor antagonist, which has been labeled with carbon-11 and characterized as a PET radiotracer for visualization of 5-HT1B receptors (Figure 1). [11C]AZ10419369 has been administered to both monkeys and humans [1]. The discovery of [11C]AZ10419369 began with a brainstorming session that included the serotonin 5-HT1B project team, radiochemists, medicinal chemists, and PET experts. Structure-activity relationships were analyzed among a large set of analogs within the project database. This led to the selection of a group of three chemical structures as potential PET tracers, based on their physical properties, binding affinity, selectivity, metabolism, and ease of synthesis (Table 1). The team then considered various options for differentiation of the three, including the conventional option of stepwise synthesis of [3H]-labeled versions of each compound followed by administration to rodents and assessment of tissue binding. This approach would have added a year to the development of a PET tracer. The team decided on an alternative innovative approach involving [11C]-labeling of all three compounds in parallel, injecting into monkey, and performing PET scans. The necessary in vivo experiments were completed in less than 1 week and enabled the rapid discovery of a novel PET tracer. The top three candidate compounds were labeled using a single [11C] labeling step using [11C]methyltrifalate [2]. The N-H piperazine precursors were prepared using t-Boc protected piperazine to ensure no non-radiolabeled N-methyl piperazine contaminated the precursor. Despite their similar chemical structures and physical properties, each compound displayed different distribution properties when injected into monkeys (Figure 2). [11C]AZ11562215 showed the lowest brain uptake among the three compounds tested (Figure 3). [11C]AZ11562215 was quickly washed from
504
Table 1
Chemical structures of potential 5-HT1B PET tracers
PET criteria O
O
CH3
N
O O
N
N N
H N
H N
H N
O N
CH3
N
O N
O
O
N O
N
N
CH3
CH3
Chemical name
AZ10419369
AZ11562215
AZ12305176
MW Log P LogD7.4 Solubility (mM) Binding affinity (nM) Permeability Efflux ratio (BA/AB) Micro. Clint (mL/min/mg) Protein binding human (%) Rat iv Cl (mL/min/kg) Rat iv VDss (L/kg) Rat iv PK t1/2 (h)
463 2.1 2.5 160 0.57 20 1.2
449 1.6 1.8 9.5 0.2 35 1.1 12 49 ND ND ND
446 3.7 3.4 8.9 0.15 5.4 0.3 59 97.4 8 2.6 ND
90.2 7.9 0.9 1.1
O
Note: ND, Not determined; Micro. Clint, In vitro metabolism (clearance) by liver microsomes; VDss, In vivo tissue volume of distribution; PK, pharmacokinetics.
Dennis J. McCarthy et al.
CH3
Discovery of Novel Positron Emission Tomography Tracers
505
Figure 2 Color-coded PET images showing distribution of radioactivity in the monkey brain after injection of 53, 53, and 51 MBq, respectively, of (A) [11C]AZ10419369, (B) [11C]AZ12305176, and (C) [11C]AZ11562215. Summation images are shown from the transaxial plane near the occipital cortex from 9- to 93-min post-injection. (See Color Plate 24.2 in Color Plate Section.) 5 AZ10419369 AZ12305176
Whole Brain Uptake (% ID)
4
AZ11562215 3
2
1
0 0
20
40
60
80
100
Time (min)
Figure 3 Whole brain uptake of [11C]AZ11562215, [11C]AZ12305176, and [11C]AZ10419369 into monkey brain. (See Color Plate 24.3 in Color Plate Section.)
the brain and displayed high uptake in the pituitary gland, suggesting non-specific peripheral binding. [11C]AZ12305176 displayed slow washout (Figure 3) and low specific binding with measurable levels in the reference tissue (cerebellum). High non-specific binding results from a molecule interaction with the
506
Dennis J. McCarthy et al.
cell’s lipid bilayer as reflected in AZ12305176 higher volume of distribution. The higher log P for AZ12305176 might also explain its higher level of non-specific binding in tissues, a property not ideal for good PET tracers (Table 1). Among the three compounds tested, [11C]AZ10419369 displayed the best properties for a PET tracer, including high uptake in the regions known to possess serotonin 5-HT1B receptors, low binding in reference tissues known to be devoid of receptors (Figure 2), and optimal time for specific binding peak equilibrium (Figure 3).
4. STEPS NEEDED TO DEVELOP PET TRACERS 4.1 Physical properties The minimum requirements for a good PET tracer include high affinity (low nanomolar or picomolar) at the target receptor, selectivity versus other receptors (Bmax/Kd of at least 10-fold), good permeability, and low background binding. (Bmax is the maximal binding that is approached asymptotically as radiotracer concentration is increased. Bmax is the density of the receptor in the tissue being studied. Kd is the concentration of radiotracer required to occupy 50% of the binding sites.) For tracers being developed for targets in the brain, the tracer needs to be a poor substrate for p-glycoprotein (Pgp). P-gp is a transport protein expressed at the capillary endothelial cells that make up the blood-brain barrier (BBB) [3]. It can transport a large variety of drugs out of the brain, contributing to what misleadingly seems to be poor penetration of the BBB. This results in a reduced percentage of the injected dose entering the brain. Rapid metabolism of the PET tracer may be an advantage, if the radioactive metabolites are not capable of penetrating the BBB. Log P was thought to be an important chemical property for good PET tracers. This observation is inferred from the early structure activity work of Hansch et al. [4]. We have found, however, that low hydrogen bonding plays a much more important role in predicting good PET tracers (Ulf Norinder, AstraZeneca, unpublished results). Finally, there needs to be feasible labeling chemistry preferably with the introduction of the label late in the synthesis.
4.2 Tracer validation Validation studies are required before use of a new tracer. AZ10419369 was successfully radiolabeled by N-methylation using carbon-11 methyl triflate. The radiochemical purity was W99% and specific radioactivity was W6,000 Ci/mmol. [11C]AZ10419369 demonstrated good brain
Discovery of Novel Positron Emission Tomography Tracers
507
uptake in the primate. The highest uptake of radioactivity in brain tissue was observed in the occipital cortex, thalamus, caudate, and putamen. Uptake was moderate in the temporal and frontal cortex regions and lowest in the cerebellum. Specific binding of [11C]AZ10419369 to 5-HT1B was demonstrated using PET with ratios of 4 in the occipital cortex to cerebellum and 3 in the putamen and caudate nucleus to cerebellum. Pre-treatment with a selective reference 5-HT1B antagonist, AR-A000002, ((R)-N-[5-methyl-8-(4-methylpipperazin-1-yl)-1,2,3,4-tetrahydro-2-naphthyl]4-morpholinobenzamide) decreased the specific binding in a dosedependent manner [1]. For high-quality PET tracers, the time to binding equilibrium should be long relative to washout of non-specifically bound tracer, but short relative to isotope decay. There should also be an adequate window of observation time, during which the receptor signal-to-background ratio should be at least 4. If the labeling method is one routinely used at the PET center (as was described in our example), little or no prior chemistry development is required. The overall radiochemical yields can be as low as a few percent and synthesis times can be as high as three half-lives. The tracer must provide adequate signal-to-background ratios in regions of interest and must be blocked by pharmacological doses of a known selective compound.
4.3 Regulatory considerations Both the Food and Drug Administration (FDA) and the European Agency for the Evaluation of Medicinal Products (EMEA) have introduced initiatives to speed the drug R&D process. Such initiatives are the FDA’s Critical Path (2004) and the EMEA’s medicines legislation and the Road Map to 2010. The main focus of these initiatives is to support early clinical evaluation of drug candidates in a safe and efficacious manner including the use of imaging tools [5]. In the European Union (EU), clinical studies using PET imaging require a Clinical Trial Application (CTA) and are covered by a Position Paper on microdosing studies [6]. In the United States, PET studies are covered by the Investigational New Drug Application (IND) regulations (21 CFR 312), the exploratory IND (eIND) guidance (Guidance for Industry, Investigators, and Reviewers-Exploratory IND Studies, January 2006), or by the Radioactive Drug Research Committee (RDRC) regulations (21 CFR 361.1). No specific guidelines for microdosing studies have been issued in Japan. For human PET tracers, a limited safety package is required usually conducted with single intravenous dose acute toxicology in two species. In addition, safety pharmacology core battery, local irritation, and blood compatibility studies should be performed. A more extensive overview of the both the European and the US regulatory requirements for PET tracers can be
508
Dennis J. McCarthy et al.
found in a publication by Farde et al. [7]. Once regulatory criteria are established, data can be collected in non-human primate for dosimetry calculations for human studies. The amount of radioactive material administered to human subjects should be the smallest radiation dose with which it is practical to perform the study without jeopardizing the benefits to be obtained from the study.
5. PET TRACERS AS TRANSLATIONAL TOOLS Drug development is a time-consuming and costly endeavor. Overall, the discovery and development of a new medicine takes about 12–15 years. Estimates of the cost of vary widely, from $800 million to nearly $2 billion per drug [8]. We now benefit from the discovery of new PET tracers that can be used to streamline this process. In addition, recent advancements in scanner technology and labeling chemistry have advanced our use of PET as a translational tool [9].
5.1 Receptor occupancy (getting the dose correct) Dose prediction in humans is particularly difficult for central nervous system (CNS) drugs. Many clinical trials have failed due to doses that are either too low (thus ineffective) or too high (thus producing unwanted side effects). In addition, there are examples of an inverted U-shape dose response where the efficacy of a drug can be lost if the dose is too high. A powerful approach to assist dose selection for CNS drugs is to use PET radiotracers suitable for quantitative study of drug binding to target receptors in the brain. Here, the drug itself is not radiolabeled. Instead, the degree at which a candidate drug inhibits radiotracer binding is determined at pharmacological doses. The degree of inhibition is commonly referred to as receptor occupancy. Correlations among plasma exposure, receptor occupancy, and functional changes in PD animal models can then be used to estimate the likely pharmacological dose in humans for a candidate drug. In addition, data from such studies can be used to optimize formulations and determine dosing regimens [10,11]. An example is the PET imaging of dopamine D2 receptor occupancy that has been successfully used to facilitate the selection of doses of antipsychotics that are effective in treating psychosis without inducing extra-pyramidal side effects. The resulting defined dose range can then be used in a translational fashion to measure PD changes (e.g., transmitter release) in animal and human in vivo [11,12].
Discovery of Novel Positron Emission Tomography Tracers
509
5.2 PET microdosing PET microdosing refers to the administration of a drug candidate that is labeled with a radiotracer. The labeled drug is given to human subjects at subpharmacological doses to obtain in vivo distribution [7] providing a sensitive method for studying the metabolism and distribution of a potential medicine. The use of tracer doses has the added advantage of requiring less extensive safety data for human dosing [9].
5.3 Pharmacodynamic measurements PET microdosing can also be used to measure a drug’s effect on important PD responses in the brain. For example, the binding of [11C]MNPA or [11C]raclopride to the dopamine D2 receptor has been shown to be sensitive to the release of endogenous dopamine in the brains of monkeys or humans after dosing with amphetamine or nicotine [13,14]. These PET imaging tools can thus provide proof of mechanism by the assessment of a PD response.
5.4 Studying disease pathophysiology In recent years, there has been increasing interest in the development of PET radiotracers as biomarkers for disease pathology in various neurological diseases. Particular progress has been made in the development of PET tracers that bind to amyloid plaque for the study of Alzheimer’s disease [15]. Such an approach offers another method by which to further understanding of the pathophysiology and early diagnosis of this debilitating disease, as well as stratification of patients [16]. Importantly, the technology is non-invasive and allows for longitudinal studies to follow disease progression and monitor treatment responses in patients.
6. NEW PET TRACERS The number of novel PET tracers has increased in parallel with advances in synthetic radiochemistry and our understanding of the properties necessary for good PET tracers. Here, we present a small list of recently reported novel PET tracers and their targets. The novel PET tracer [11C]SB207145, selective for the serotonin 5-HT4 receptor, was recently reported [17,18] (Figure 4). This receptor has been detected in the brains of rats, mice, pigs, monkeys, and human. The highest 5-HT4 receptor densities are in the hippocampus and striatum. The receptor is thought to play a role in cognition and may also be involved in the etiology of major depressive disorder [19]. 5-HT4 agonists
510
Dennis J. McCarthy et al.
NH2 Cl
O O O
O N11 CH3
Figure 4
[11C]SB207145: A novel PET tracer for the serotonin 5-HT4 receptor.
11
11
CH3 O
N N
O
N H11 C 3
N
CH3
N
O O N
N
O O
O N
[11C]DPA-713
Figure 5
N
N
[11C]AC-5216
[11C]PBR28
Novel peripheral benzodiazepine receptor PET tracers.
have also been shown to reverse memory deficits induced by the muscarinic antagonists and may have utility in the treatment of psychiatric diseases. There are a number of reports of novel PET tracers for the peripheral benzodiazepine receptor including [11C]-DPA-713, [11C]-AC-5216, and [11C]-PBR28 [20,21,22] (Figure 5). These tracers have been developed for non-invasive studies of inflammation, including microglia and macrophage activation in the brain, lung, and heart. All three of these tracers are reported to have superior properties compared to an older tracer [11C]PK11195. Glycine transporter 1 (GlyT-1) inhibitors are being developed to treat schizophrenia. A novel PET tracer, [11C]GSK931145, has recently been reported that is selective for this transporter [23] (Figure 6). Investigators reported heterogeneous distribution of [11C]GSK931145 in the brains of non-human primates and pigs. The distribution of the tracer was consistent with the known levels of GlyT-1 in the brain. This tracer will support the clinical development of GlyT-1 inhibitors and be used to investigate the role of GlyT-1 in the pathogenesis of schizophrenia. AstraZeneca and the Karolinska Institutet have jointly published data on a potentially more sensitive radiotracer for the PET visualization of b-amyloid deposits in living human brain [24]. Such tracers would have use in the early diagnosis of Alzheimer’s disease. AZD2184 (2-[6-(methylamino)pyridin-3-yl]-1,3-benzothiazol-6-ol) has a chemical
511
Discovery of Novel Positron Emission Tomography Tracers
11CH 3
H3C N
O
CH3
N H H3C [11C]GSK931145
Figure 6 Novel GlyT1 transporter PET tracer [11C]GSK931145, N-[1(S)-[1-([11C]dimethylamino)cyclopentyl]-1-phenylmethyl]-2,6-dimethylbenzamide.
HO
S N AZD2184
H N N
HO
S
H N 11C
N [11C]PIB
Figure 7 Chemical structure of AZD2184 a potential sensitive tracer for amyloid plaque and [11C]PIB.
structure that is similar to Pittsburg compound B (PIB) (2-[4-(methyl11 C-amino)phenyl]-6-hydroxybenzothiazole), an extensively studied tracer (Figure 7). Patients with Alzheimer’s disease have a twofold to threefold retention of [11C]PIB compared to age-matched normal healthy volunteers [25–27]. Compared to PIB, AZD2184 displays a higher signalto-background ratio due to its low levels of non-specific binding. More sensitive PET tracers are needed for the early detection of b-amyloid deposits and such biomarkers would allow for early diagnosis and treatment of this devastating disease.
7. CONCLUSIONS Innovation results when groups with differing perspectives collaborate to address unresolved issues. For example, radiochemists are able to view a chemical database from a different perspective than a medicinal chemist. The radiochemist can identify chemical structures that have properties amenable to being a good radiotracer. They also consider novel synthetic strategies that allow for the incorporation of the radiolabel in the fewest synthetic steps. The medicinal chemist is able to assess the physical properties of chemical series to select compounds with optimal chemical space to maximize drug-like properties. The biologist ensures that the compounds bind to their targets with the
512
Dennis J. McCarthy et al.
appropriate affinity and selectivity. With all of our knowledge, we still cannot predict with 100% certainty whether a particular compound will have the predicted properties when observed in complex biological systems. This was demonstrated by our example of the 5-HT1B PET tracer. Three compounds from the same chemical series had significantly different properties when compared side by side in a living animal. In the end, it takes experimentation for new breakthroughs to occur. Novel discoveries occur by bringing together different disciplines with different perspectives on the same problem. This type of collaborative interaction can result in the discovery of new PET tracers that can be used to advance our understanding of receptor biology.
REFERENCES [1] E. M. Pierson, J. Andersson, S. Nyberg, D. J. McCarthy, S. J. Finnema, K. Varna¨s, A. Takano, P. Karlsson, B. Gulya´s, A. Medd, C.-M. Lee, M. E. Powell, R. Heys, W. Potts, N. Seneca, L. Mrzljak, L. Farde and C. Halldin, Neuroimage, 2008, 41, 1075. [2] J. Sandell, O. Langer, P. Larsen, F. Dolle, F. Vaufrey, S. Demphel, C. Crouzel and C. Halldin, J. Label. Compd. Radiopharm., 2000, 43, 331. [3] T. Terasaki and K. Hosoya, Adv. Drug Deliv. Rev., 1999, 36, 195. [4] C. Hansch and J. M. Clayton, J. Pharm. Sci., 1973, 62, 1. [5] C.-P. Milne, Expert Opin. Drug Discov., 2006, 1, 11. [6] A. Verbruggen, H. H. Coenen, J-R. Deverre, D. Guilloteau, B. Langstrom, P. A. Salvadori and C. Halldin, Eur. J. Nucl. Med. Mol. Imaging, 2008, 35, 2144. [7] L. Farde, S. Heminway, C.-M. Lee, D. J. McCarthy, I. Nordgren and S. Nyberg, Regul. Rapporteur, 2007, 4, 2. [8] N. Masia, Focus On: Intellectual Property Rights, U. S. Department of State, 2006, p. 86. [9] C. Halldin, B. Gulya´s and L. Farde, Curr. Pharm. Des., 2001, 7, 1907. [10] D. J. Brooks, NeuroRx., 2005, 2, 226. [11] C.-M. Lee and L. Farde, Trends Pharmacol. Sci., 2006, 27, 310. [12] L. Farde, Trends Neurosci., 1996, 19, 211. [13] S. Marenco, R. E. Carson, K. F. Berman, P. Herscovitch and D. L. Weinberger, Neuropsychopharmcology, 2004, 29, 259. [14] N. Seneca, S. J. Finnema, L. Farde, B. Gulyas, H. V. Wikstrom, C. Halldin and R. B. Innis, Synapse, 2006, 59, 260. [15] A. Lockhart, Drug Discov. Today, 2006, 11, 1093. [16] G. W. Small, V. Kepe and J. R. Barrio, Curr. Opin. Psychiatry, 2006, 19, 564. [17] A. D. Gee, L. Martarello, J. Passchier, M. Wishart, C. Parker, J. Matthews, R. Comley, R. Hopper and R. Gunn, Curr. Radiopharm., 2008, 1, 110. [18] B. R. Kornum, N. M. Lind, N. Gillings, L. Marner, F. Andersen and G. M. Knudsen, J. Cereb. Blood Flow Metab., 2008, 29, 186. [19] J. L. Warner-Schmidt, M. Flajolet, A. Maller, E. Y. Chen, H. Qi, P. Svenningsson and P. Greengard, J. Neurosci., 2009, 29, 1937. [20] H. Boutin, F. Chauveau, C. Thominiaux, M. C. Gre´goire, M. L. James, R. Trebossen, P. Hantraye, F. Dolle´, B. Tavitian and M. Kassiou, J. Nucl. Med., 2007, 48, 573. [21] M. R. Zhang, K. Kumata, J. Maeda, K. Yanamoto, A. Hatori, M. Okada, M. Higuchi, S. Obayashi, T. Suhara and K. Suzuki, J. Nucl. Med., 2007, 48, 1853. [22] M. Fujita, M. Imaizumi, S. S. Zoghbi, Y. Fujimura, A. G. Farris, T. Suhara, J. Hong, V. W. Pike and R. B. Innis, Neuroimage, 2008, 40, 43.
Discovery of Novel Positron Emission Tomography Tracers
513
[23] N. V. Murthy, J. Passchier, R. N. Gunn, G. E. Searle, S. Bullich, M. Suarez, R. Herance, M. Farre, H. R. Herdon, R. Porter, S. Sutherland, R. Fagg, M. Neve, M. Slifstein, M. Laruelle and A. M. Catafau, Neuroimage, 2008, 41(Suppl. 2), T21. [24] A. E. Johnson, F. Jeppsson, J. Sandell, D. Wensbo, J. A. M. Neelissen, A. Jure´us, P. Stro¨m, H. Norman, L. Farde and S. P. S. Svensson, J. Neurochem., 2009, 108, 1177. [25] W. E. Klunk, H. Engler, A. Nordberg, Y. Wang, G. Blomquist, D. Holt, M. Bergstrom, I. Savitchera, G.-F. Huang, S. Estrada, M. L. Debnah, J. Barletta, J. C. Price, J. Sandell, B. J. Lopresti, A. Wall, P. Koivisto, G. Antoni, B. Ausen, C. A. Mathis and B. Langstrom, Ann. Neurol., 2004, 55, 306. [26] K. E. Pike, S. Savage, V. L. Villemange, S. Ng, S. A. Moss, P. Maruff, C. A. Mathis, W. E. Klunk, C. L. Masters and C. C. Rowe, Brain, 2007, 130, 2837. [27] A. Forsberg, H. Engler, O. Almkvist, G. Blomquist, G. Hagman, A. Wall, A. Ringheim, B. La¨ngstro¨m and A. Nordberg, Neurobiol. Aging, 2007, 29, 1456.
CHAPT ER
25 The Use of Isotopically Labeled Compounds in Drug Discovery Charles S. Elmore
Contents
1. Introduction 1.1 Radioactive waste 1.2 Storage of radiolabeled compounds 2. Synthesis of Radiolabeled Compounds 2.1 Tritium-labeled compounds 2.2 Carbon-14-labeled compounds 2.3 Sulfur-35 2.4 Iodine-125 2.5 Stable isotope–labeled compounds 3. Human ADME Studies 4. Conclusions References
515 517 517 518 518 523 526 527 528 530 532 532
1. INTRODUCTION Isotopically labeled compounds are critical to the drug development process for use as radioligands in lead discovery [1], as metabolic tracers in development [2], and in phase IV clinical studies [3]. Large pharmaceutical companies typically have a radiochemistry group to synthesize isotopically labeled compounds and coordinate the use of contract laboratories for custom synthesis when required. In contrast, smaller companies generally outsource this work to contract laboratories. This review familiarizes the reader with the uses of long-lived Isotope Chemistry, CNS Chemistry, AstraZeneca Pharmaceuticals LP, Wilmington, DE 19350, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04425-X
r 2009 Elsevier Inc. All rights reserved.
515
516
Charles S. Elmore
radiolabeled compounds (those compounds with half-lives greater than 1 month) and stable isotope–labeled (SIL) compounds and enables them to effectively plan radiochemical syntheses. The synthesis and use of short-lived positron-emitting radiolabeled compounds (those labeled with 11C, 13N, 15O, 18F) used in nuclear medicine imaging studies is discussed in a separate review in this journal. Radioisotopes are used to improve the sensitivity and ease of detection of drug molecules in both in vitro and in vivo studies. The low levels of radioactivity in the environment provide a very low background against which the radiolabeled drug can be detected. Additionally, since the amount of radiation released by the radioactive drug molecule is independent of structure, accurate quantitation of the parent drug molecule and its metabolites can be done without the use of reference standards, provided the radioactive atom is retained in the metabolites. Liquid scintillation counting (LSC) is the main method by which longlived isotopes are quantitatively detected. LSC uses a photomultiplier tube to detect light emissions of a fluor; the fluor is excited by the absorption of radiation, and the light is released when the excited fluor relaxes to the ground state. The amount of light emitted by a specified amount of radioactive material can be directly correlated to the amount of radioactivity present. Accelerator mass spectrometry has also been used to detect low levels of radioactivity in biological samples but is infrequently used due to the expense of the equipment and the difficulty in sample preparation [4,5]. Radioactive compounds are generally quantified in terms of radioactivity rather than, or in addition to, mass. The SI unit of measure is the Becquerel (Bq), which is 1 disintegration/s. whereas the traditional unit of measure is the Curie (Ci), which corresponds to the amount of radioactivity produced by 1 g of Ra-222 (2.022 1012 disintegration/min or dpm). One milliCurie (1 mCi) is equal to 37 megaBecquerel (MBq). The correlation between mass and radioactivity depends on the radioisotope and the amount of radioactive nuclei present in the molecule. A compound that is labeled with approximately 100% of a radioisotope is referred to as carrier free; in practice, this refers to a labeling of W80% for C-14 and W50% for tritium. Some commonly used radioisotopes and their properties are given in Table 1. The more rapid the radiochemical decay (the shorter the half-life), the higher the amount of radioactivity/atom. The amount of radioactivity/ molecule is termed specific activity (SA) and is expressed in mCi/mmol, MBq/mg, or similar units. As can be seen in Equation 1, SA varies inversely with the half-life of the atom. The SA of a molecule governs its use. High SA radioisotopes (e.g., S-35, I-125, H-3) are used in applications that require low mass loadings such as radioligand binding assays, autoradiography studies, plasma protein binding studies, and
The Use of Isotopically Labeled Compounds in Drug Discovery
Table 1
517
Commonly used radioisotopes and their properties
Radionuclide
Emission type
Half-life
Maximum specific activity
Carbon-14 Tritium Sulfur-35 Iodine-125
Beta Beta Beta Gamma
5730 years 12.43 years 87.4 days 60 days
62.4 mCi/mmol 28.8 Ci/mmol 1498 Ci/mmol 2175 Ci/mmol
permeability assays. Low SA radioisotopes (predominantly C-14) are used for studies in which therapeutic doses of the drug are given such as absorption, distribution, metabolism, and excretion (ADME) studies and covalent binding screens. Maximum specific activity ðfor one atomÞ 4:17 10e23 disintegrations=mmol ¼ t1=2
ð1Þ
Compounds labeled with stable isotopes (non-radioactive isotopes) have many applications within the pharmaceutical industry. By far, the most common use is as internal standards for mass spectroscopy studies of biological fluids where the isotopically labeled compound allows a more accurate determination of the concentration of the drug. The stable isotope labels used in these studies are typically C-13, N-15, and H-2. The oxygen isotope, O-18, is seldom used due to the propensity of oxygen atoms to exchange with water; however, a few elegant studies have utilized O-18-labeled compounds for mechanistic elucidation [6].
1.1 Radioactive waste Although waste disposal for even non-radioactive chemical processes is an issue, disposal of radioactive waste can pose an even greater challenge. The disposal of mixed waste such as flammable or highly toxic radioactive waste (e.g., mercury or lead waste) can be extremely expensive, and in some cases, it is not even possible to find a vendor willing to handle the disposal. Therefore, the radioactive waste to be produced by a reaction and the availability of proper radioactive waste disposal options must be assessed before undertaking a synthetic sequence.
1.2 Storage of radiolabeled compounds By definition, radiochemical compounds are in a constant state of decomposition. When they decompose, they emit a high-energy particle (a gamma ray or beta particle), which can interact with other nearby drug
518
Charles S. Elmore
molecules leading to further decomposition. Compounds that are good radical scavengers (such as indoles) or prone to oxidation (such as thioethers) tend to decompose at accelerated rates when compared to compounds without those functionalities. This complicates the requirement for radiolabeled compounds to be of high purity and emphasizes the importance of using the radiotracers soon after their preparation. Furthermore, the purity of the tracer should be monitored if a significant amount of time has elapsed between the synthesis and use [7]. A few principles that have been developed as a guide to aid in maximizing the stability of radiochemicals during storage are summarized here [8]: Protect the chemical from light and store under N2 or Ar. Store at low temperatures (o80 1C). Store as a solution (alcohols, especially ethanol, and acetonitrile are good solvents, water should be avoided). 4. If storing as a solid, store as a crystal and not in amorphous form. These principles are merely a guideline; every compound has a different stability profile and may benefit from alternative storage conditions. 1. 2. 3.
2. SYNTHESIS OF RADIOLABELED COMPOUNDS The synthesis of radiolabeled compounds is very similar to nonradiolabeled syntheses with a few notable exceptions (especially for tritium). (1) The synthesis often needs to be altered from the previously developed routes to allow for late-stage incorporation of radioactivity into the target molecule. This allows for a higher radiochemical yield due to fewer reaction steps and generates less radioactive waste. The stoichiometry of the reactions is often reversed as the most precious reagent in the radiochemical reaction is often the radioactive reagent rather than the custom made intermediate. (2) If the tracer is to be used in vivo as an ADME tracer, a metabolically stable location in the molecule must be identified for the radiolabel. This often requires the label to be incorporated in the central core of the molecule, thus posing a synthetic challenge. (3) The purity of the final molecule must be higher than those found for typical medicinal chemistry molecules due to the low detection limits of radioactivity. This requirement is generally W98%, and for many tracers, no single impurity can be W0.4%.
2.1 Tritium-labeled compounds The ubiquitous occurrence of hydrogen in drug molecules and the ease with which a carbon–hydrogen bond can be formed make the use of
The Use of Isotopically Labeled Compounds in Drug Discovery
519
radioactive hydrogen isotopes attractive. However, a necessary corollary to the ease of forming the carbon–hydrogen bond is the ease with which that bond can also be cleaved. If the molecule is labeled with tritium (the radioisotope of hydrogen) at the cleavage site, the label will be lost. This loss can occur either through a chemical route (e.g., placement of the label alpha to a ketone) or through a biochemical means (e.g., metabolism). Many metabolic and chemical losses of the tritium label can be predicted by analyzing the structure of a drug, but other metabolic fates are often elucidated only by appropriate labeling of the drug. Tritium can also lead to metabolic shifting due to the primary isotope effect associated with the breaking of a tritium–carbon bond. This can artificially decrease the amount of metabolites formed by the cleavage of the carbon–tritium bond while increasing other metabolites not requiring this bond cleavage [9]. Despite these drawbacks, tritium is used extensively in support of drug discovery efforts. Owing to the volume of requests and the limited budgets of many discovery organizations, the syntheses of tritium-labeled compounds should be rapid and simple. Ideally, the synthesis should be completed in one or two steps with tritium label incorporation being the final step. As a result, many methods have been developed, which allow for direct and indirect conversion of the unlabeled target compound to the tritiated species. Tritium can be incorporated into molecules using a reduction with tritium gas, reduction with a tritide source, transition metal mediated exchange, or by the incorporation of commercially available reagents such as tritiated methyl iodide (CT3I). Tritiated products are primarily analyzed by HPLC with radioactive detection and LC/MS; however, tritium is a spin active nucleus, which makes 3H NMR analysis also feasible. When the site of labeling of the tritium label is critical, 3H NMR is an invaluable tool in confirming that the tritium atom is in the location anticipated. Tritium gas is a dangerous and difficult substance to manipulate, and relatively small volumes of tritium gas contain many Curies of radioactivity. A Toepler pump — an apparatus that used mercury displacement to transfer known amounts of tritium gas — was originally used to deliver tritium gas into reaction vessels. This method was cumbersome, messy, and dangerous and resulted in large amounts of mercury waste contaminated with tritium. Additionally, the excess tritium from the reaction mixtures was difficult to recover after use. Thus, a critical advancement allowing routine, high SA tritiations to be safely performed in the laboratory was the development of tritium manifold technology [10,11]. These manifolds are now commercially available and allow for known amounts of tritium gas to be delivered to reaction mixtures and the residual tritium gas to be recovered from the reaction upon completion [12,13]. These manifolds rely upon the reversible formation
520
Charles S. Elmore
of uranium (III) tritide. The application of heat to 238UT3 releases tritium gas at a rate that can be accurately controlled by varying the temperature. The formation of UT3 is reversible so that once cooled, the 3H2 gas remaining after the reaction can be condensed upon the uranium bed once again. The volume of the manifold is accurately determined so that by measuring the pressure of the tritium gas, the number of moles of tritium gas present can be determined using the ideal gas law. The gas in the manifold is then exposed to the liquid nitrogen cooled reaction mixture by opening and then rapidly closing the manifold. The drop in pressure in the manifold is used to determine the amount of tritium transferred. The tritium gas remaining in the manifold can then be recovered onto the U-bed.
2.1.1 Tritiodehalogenation One of the most versatile and common routes to tritium labeling is the use of aryl halides. Aryl iodides, bromides, and chlorides can be readily reduced to the corresponding tritiated arenes with heterogeneous (generally Pd/C) catalysts. This route is particularly attractive as it is tolerant of nearly all functional groups that cannot be readily reduced (notable exceptions are ArNO2 and unsaturated bonds), and selectivity can be achieved for reducing an iodide in the presence of other halides or a bromide in the presence of a chloride by controlling the reaction conditions. The most powerful aspect of this synthetic route is that there are many ways to convert an unlabeled drug target directly into an aryl halide using one-step halogenations including N-iodosuccinimide (NIS) in triflic or trifluoroacetic acid, ICl, and Br2 in CHCl3, among others. When this is feasible, a specific precursor need not be prepared solely for radiolabeling. Additionally, these halogenations require only small amounts of material and are rapidly conducted so that they are well suited for discovery efforts [14] (Scheme 1).
2.1.2 Organoiridium-catalyzed ortho-hydrogen isotope exchange Certain iridium-containing catalysts have been shown to insert into the C-H bond of arenes with a predictable regioselectivity; these catalysts are O
O
N O N H
O
N O
NIS CF3SO3H
N H
Pd/CaCO3, DMF I
T I
Scheme 1
O
T2, Pd/C,
N H
N
I
Synthesis of [3H3](S)-mephenytoin [14].
T
T
The Use of Isotopically Labeled Compounds in Drug Discovery
521
directed to specific sites in the molecule by directing groups such as ketones, esters, amides, and the like [15]. The basic mechanism is that the iridium catalyst first complexes to the carbonyl oxygen (or pyridine nitrogen) and then oxidatively insets into the ortho aromatic C-H bond to give a 5- or 6-member heterocycle. The Ir-H bond rapidly scrambles with the Ir-3H2 and reductive elimination gives a tritium-labeled arene with the tritium ortho to the directing group. Some of the more commonly used catalysts include [(COD)Ir(pyr)PCy3]BF4 [16] and [(COD)Ir(PPh3)2]BF4 [17]. Recently, several N-heterocyclic-containing catalysts have been reported to be very effective at performing exchange reactions [18,19]. [(Z5-Cp)(PMe3)Ir(Me)(CH2Cl2)][B(3,5-C6H3(CF3)2]4 has been reported to obviate the need for directing groups in the labeling of aromatic rings, which makes this an extremely powerful tool for labeling [20,21]. The precise level of tritium incorporation varies with the substrate so for optimal results a range of catalysts need to be investigated. Often multiple sites are labeled with hydrogen isotope exchange reactions including aliphatic protons (3H NMR can identify the sites of labeling and the relative amounts of tritium in each site) [22] (Scheme 2). This can limit the utility of this reaction as does the requirement for the reactions to be performed in CH2Cl2.
2.1.3 Methylation The reaction of tritiated methyl iodide, CT3I or tritiated methyl triflate, CT3OTf (or other tritiated sulfonates) with amines, alcohols, or acids is a well-established method for the labeling of drug molecules [23]. The reactions are fast, easy to run, and frequently high yielding, thus producing less waste. However, the products of these reactions (methyl ethers, amines, or esters) are often rapidly metabolized to the desmethyl adduct, making the radiotracer unsuitable for many applications. Despite not being metabolically stable, these compounds work well in in vitro binding studies and in vivo receptor occupancy studies [24].
N HN
22% T
CF3
N HN
O N O N
O T
42%
T2, CH2Cl2,
T [(COD)Ir(pyr)PCy3]BF4
N O
18%
18%
N T
Scheme 2 Ir-catalyzed hydrogen–tritium exchange of a complex molecule and identification of the sites of tritium incorporation by 3H NMR [22].
CF3
522
Charles S. Elmore
2.1.4 Reduction of unsaturated bonds Reduction of unsaturated bonds provides access to tritium-labeled compounds, but is infrequently used due to the need to prepare the substrate. This method generally gives good chemical yields; however, if there are methylene carbons adjacent to the unsaturated bond, the radioactive label can be spread over more than the two Pi-bonded carbons. This propensity of palladium to ‘‘walk’’ down the carbon chain can result in multiple labeled sites and therefore render the tracer unsuitable for use [25,26] (Scheme 3). The use of Ir-based catalysts to effect the reduction can often limit or eliminate the additional labeling of the carbon chain.
2.1.5 Reduction with tritium hydrides Tritide sources such as LiBT4, NaBT4, and LiAlT4 are very useful and seemingly underutilized in the pharmaceutical industry [27]. Borotritide reagents are relatively stable and readily available from commercial sources. Generally, this methodology is applied when tritiation with 3H2 gas is impractical such as in the synthesis of bilirubin [28] (Scheme 4).
CO2H
CO2H Pd/C, D2 THF SO2NH2
SO2NH2
Dn
OPh
OPh D0 = 18.5% D1 = 25.2% D2 = 24.3%
D3 = 16.9% D4 = 9.7% D5 = 5.5%
Scheme 3 Reduction of a carbon–carbon double bond showing propensity for incorporation of more than two deuterium atoms/molecule [26].
CH2CH2CO2H
HO2CH2CH2C NH
T
HO2CH2CH2C
N
NH
CH2CH2CO2H N
NaBT4, EtOH NH
HN
NH
HN
O
O
O
O
[3H]Bilirubin
Scheme 4
3
Synthesis [ H]bilirubin by NaBT4 reduction [28].
The Use of Isotopically Labeled Compounds in Drug Discovery
523
Several recent reports highlight the utility of using sodium borotritide as a 3H2 source in tritiations with Pd/C [29,30]. This methodology generates much less waste than tritiations using 3H2 gas and can give excellent chemoselectivity, which is difficult to achieve using other methods.
2.2 Carbon-14-labeled compounds The synthesis of C-14-labeled compounds is generally much more difficult and time consuming than the synthesis of tritium-labeled compounds. The most common uses of C-14-labeled compounds are for ADME studies and reactive metabolite screens, which require that the C-14 label be incorporated in locations that will not be metabolically cleaved [31]. Incorporation of the C-14 label into these sites generally requires longer and more difficult syntheses. There are many C-14labeled starting materials commercially available, but the price of the reagents increases dramatically with the complexity of the material. A significant challenge faced by radiochemists is how to incorporate the typical one-carbon C-14-labeled building blocks into an existing synthetic route that has been developed for other purposes. Sometimes this requires the development of a novel synthesis, but often the methodology developed by the medicinal chemist can be utilized by the radiochemist. However, to improve the radiochemical yield, an adjustment to the stoichiometry of a reaction is often made to reflect the higher value of the radiochemical.
2.2.1 [14C]Barium carbonate and [14C]carbon dioxide Ba14CO3 is the universal precursor for all C-14-labeled compounds. It is converted into 14CO2 by treatment with acid (generally H2SO4) or by heating with PbCl2, although the later leads to radioactive waste mixed with Pb. The handling of 14CO2 is easily conducted using a gas manifold as carbon dioxide condenses at liquid nitrogen temperatures and is thus easily manipulated under high vacuum. Recently, technology similar to the tritium manifold has been adapted for use with 14CO2. 14CO2 is reversibly bound to molecular sieves and released with heating or trapped with cooling [32]. 14 CO2 can be reacted with organolithium reagents, Grignard reagents, and other organometallics to give carboxylation products. On larger scales (W1 mmol), 14CO2 is frequently used rather than BaCO3 as it is more convenient and the price is comparable. Acetic acid and bromoacetic acid are extremely useful intermediates that can be produced from 14CO2 and are frequently used as starting materials for the synthesis of C-14-labeled compounds.
524
Charles S. Elmore
2.2.2 [14C]Methyliodide 14
CH3I is a versatile one-carbon synthon that has seen considerable use in radiosyntheses. 14CH3I has been used directly in alkylations [33] and has been converted to methylating reagents of many forms. Reaction of C-14 methyl iodide with n-BuLi gives 14CH3Li, which can be further elaborated into more complex reagents [34] or used directly as a methylating agent [35]. 14CH3Li has been used to prepare C-14 methylated boronate esters that readily react with aryl iodides to give C-14-labeled toluenes; however, the synthesis of the precursor to the boronate requires a long sequence, thus limiting its utility [36] (Scheme 5). 14CH3I has also been converted to [14C]methylsulfones by reaction with a thiol to give the corresponding thiomethylether, which was subsequently oxidized to the sulfone [37] (Scheme 5).
2.2.3 [14C]Cyanides Carbon-14 cyanide is used in many forms including K14CN, Na14CN, Zn(14CN)2, and Cu14CN. K14CN and Na14CN are utilized primarily in nucleophilic reactions. Cu14CN can be used directly [38] or can be formed in situ from K14CN [39] to generate aryl nitriles from activated arenes through an SN2Ar reaction. These reactions are rather intolerant of functional groups though due to the high temperature required to affect S14CH3
RSCH2O2CCF3 NaOH, MeOH, THF, CH2Cl2
SO214CH3
Cl
Cl N
N N
N HN
14
[ C]etoricoxib
14
CH3I
O N N H314C 14
CH3Li
O
O N HN
OMe
B
CO2H
H NH SO2
[14C]L-738167 14
14
14
Scheme 5 Syntheses of [ C]L-738167[36] and [ C]etoricoxib[37] from [14C]iodomethane.
CH3
525
The Use of Isotopically Labeled Compounds in Drug Discovery
the transformation. A milder method to generate aryl [14C]nitriles has been developed using Pd-catalysis and Zn(14CN)2 [40] (Scheme 6). Use of wet dimethylformamide (DMF) ensures that both cyanide molecules of the Zn(14CN)2 are transferred in the reaction so high radiochemical yields can be achieved under relatively mild conditions that tolerate a wide range of functionality. A procedure to convert K14CN to methyl[14C]isothiocyanate that was then converted to several C-14labeled triazoles in high yield has recently been published [41].
2.2.4 [14C]Carbon monoxide The use of 14CO has been reported several times this past decade in C-14 carbonylations [42]. 14CO is expensive and unstable but a method has been reported that allows for its generation from lithium formate, which upon immediate use affords a range of carbonylation products including amides, acids, esters, and ketones. The methodology is extremely tolerant of functional groups and thus can be used late in syntheses. This methodology has been adapted to the synthesis of a C-14 GNRH antagonist [43] (Scheme 7).
2.2.5 [14C]Aromatics Substituted aromatics incorporated with C-14 are also available commercially and have been used in many radiosyntheses. There are many potential routes to these radiochemical starting materials, but the majority start from calcium [14C2]carbide. This results in multiple C-14 atoms in the aromatic ring and thus multiple C-14 labels in the target molecule. This is not generally a problem, but can complicate LC/MS analyses. Typically, industrial groups purchase substituted C-14-labeled
Zn(14CN)2
RI, Pd(PPh3)4
N
N 14
N
DMSO Cl
CN
O N
Scheme 6
Synthesis of [14C]L-778123 using Zn(14CN)2 [40]. Cl
14CO
2H
NBoc 14CO
LiBEt3H 2
H14CO
2Li
H2SO4
14CO
RI, Pd(Dppf)Cl2 KOAc, DMSO
HN
O
O
Scheme 7
Toward a [14C]GNRH antagonist through C-14 carbonylation [43].
526
Charles S. Elmore
aromatics due to the inefficiency of performing these reactions on a small scale [44].
2.2.6 [14C]Dimethylformamide [14C]DMF has been utilized in many syntheses [45], but is often avoided due to its volatility and water solubility. [14C]Dibutylformamide [46,47] and [14C]N-formylpiperidine [48], which have similar reactivity to [14C]DMF but much less volatility and water solubility, have been reported as substitutes for DMF for Vilsmeier and anion chemistry. These compounds can be prepared from H14CO2H, which is itself prepared from 14CO2.
2.2.7 Other common [14C]synthons Nitromethane [49], thiocyanate [50], [2-14C] or [1,3-14C]malonic acid [51], paraformaldehyde [52], and phosgene [53] are also readily available in C-14-labeled form and have been used to good effect. 14CS2 has been used to convert a phenol into a C-14-labeled trifluoromethoxy group through a xanthate ester [54].
2.3 Sulfur-35 The use of S-35-labeled compounds is limited due to the limited availability of S-35 precursors and infrequent occurrence of sulfur in drug molecules; however, due to its high SA, S-35 makes a very attractive target if synthetically feasible [55]. The relatively short half-life of S-35 requires the radioligand to be used rapidly to avoid contamination with radiochemical impurities. Two reagents similar to the Bolton–Hunter reagent (Section 2.4.2) have been developed for the labeling of lysine residues in proteins, N-succinimidyl-4-(methanesulfonylaminomethyl)benzoate (SMSB) and 4-(Methanesulfonylaminomethyl)-phenylpropylaldehyde (MSAPPA), and provide nice alternatives to the use of I-125-labeled proteins [56]. OH
MeO235SHN
MeO235SHN
125I
O
O
O
N
O
O
O
N
O
O
H
O
Bolton-Hunter Reagent
SMSB
MSAPPA
527
The Use of Isotopically Labeled Compounds in Drug Discovery
2.3.1 [35S]Sulfonic acids S-35 methane and benzene sulfonic acid are readily available at high SA. The acid can be converted to the corresponding S-35 sulfonyl chloride and then coupled with an appropriate nucleophile. This nucleophile is almost always an amine or aniline thus yielding a S-35 labeled sulfonamide [44]. The Friedel–Crafts reaction of S-35-labeled methanesulfonyl chloride with anisole to afford aryl sulfones has been reported [57] (Scheme 8). Conditions favoring formation of the para isomer (BF3) were reported while alternate conditions (Bi(OTf)3) gave close to a 1:1 mixture of ortho and para isomers.
2.3.2 [35S]Elemental sulfur [35S8]Elemental sulfur has been reported to react with Grignard reagents or lithiates providing access to a much more diverse pool of S-35-labeled compounds [58] (Scheme 9). These reactions give rather complex reaction products due to dimerization of the S-35-labeled thiols; however, a onepot oxidation–chlorination procedure in which the sulfonyl chloride is generated directly and then used to make complex radioligands has been reported [58].
2.4 Iodine-125 The use of I-125 is generally restricted to the labeling of peptides (and photoaffinity labels) because few drug molecules contain iodine, and while iodine is seldom found in proteins, addition of an iodine atom to 35SO
35SO
2Me
Me35SO2Cl BBr3
2Me
Me35SO2Cl
+
Bi(OTf)3
OMe
OH
35SO
OMe
23
OMe
:
2Me
17
35
Scheme 8 Lewis acid catalyzed reaction of anisole with [ S]methanesulfonic acid [57].
X
35SH 35S 8
R Scheme 9
35SO
2Cl
SO2Cl2, KNO3 R
CH2Cl2
Reaction of [35S8]sulfur with a lithiate or Grignard reagent [58].
R
528
Charles S. Elmore
the peptide often leaves the binding of a peptide to the target receptor unperturbed.
2.4.1 Oxidative methods Nearly all iodine labeling methodology is oxidative. In these methods, an electrophilic iodinated species is generated which in turn reacts with nucleophilic amino acids such as tyrosine. Typical protocols use Na125I and chloramine T [59], lactoperoxidase [60], or iodogen [61] to effect the iodination.
2.4.2 Conjugation to nucleophilic peptidic residues The other major route to I-125-labeled compounds is the conjugation of an I-125-labeled molecule to a nucleophilic portion of the peptide such as the nitrogen of a lysine or the N-terminus of the peptide. The main reagent used for this purpose is the Bolton–Hunter reagent [62,63]. A major drawback of this methodology is that the Bolton–Hunter adduct must be located in a site in the peptide which can tolerate bulk.
2.5 Stable isotope–labeled compounds The principles involved with synthesizing a SIL compound are very similar to those of a C-14-labeled compound. The scale is generally a bit larger, but the expense of the starting materials is slightly lower. One major difference between SIL and C-14 syntheses is that typically the isotopic label need not be located in a metabolically stable location since the SIL compound is generally used only to quantitate the parent compound in biological matrixes. A known amount of the SIL drug is added to the biological sample, and after processing to prepare the sample for LC/MS, the ratio of SIL to unlabeled drug is determined. This ratio allows one to determine the absolute amount of unlabeled drug present in the biological sample, and by adding the SIL compound before processing, the loss of unlabeled compound due to these manipulations is accounted for. Since the molecule will be used in LC/MS/MS studies, it is preferable to have the isotopic label in the daughter fragment to improve the signal to noise ratio in the MS, but it is not critical as the same signal to noise ratio can be obtained by the addition of more SIL compound to the biological sample. Unlike compounds labeled with C-14, SIL compounds invariably require more than a single isotopic label to give sufficient resolution between the mass spectrum of the parent drug and the SIL drug. The isotopes incorporated may be of a single element or a mixture of two or more elements. The minimum mass increase required to achieve appropriate resolution is three, while four is preferred. If the molecule contains a Br, Cl, or S, a higher mass increase is warranted to give proper
The Use of Isotopically Labeled Compounds in Drug Discovery
529
resolution from the parent molecule. A key synthetic consideration is that the isotopically labeled material must contain no (or very low levels of) unlabeled material or the bioanalysis becomes complex. Generally N-15, C-13, and H-2 are used to prepare SIL compounds (with C-13 and N-15 being preferred by bioanalysts). This preference arises from the fact that C-13 and N-15 seldom, if ever, fractionate from their unlabeled isotopomers under typical chromatographic conditions. However, H-2-labeled and -unlabeled compounds can have different retention times on LC if the label is not properly located [64]. This effect can be minimized by avoiding placing the deuterium label within two bonds of nitrogen. The H-2 label must also be placed in a location that is non-exchangeable (e.g., not alpha to a carbonyl). Although the starting materials for preparing SIL compounds are expensive, a wider range is readily available than for C-14 syntheses. Frequently, the SIL synthesis is used to model the C-14 synthesis. In many companies, the responsibility for preparing the SIL material resides with medicinal chemistry rather than a group specializing in isotopic synthesis since no radioactivity is involved.
2.5.1 Deuterium-labeled compounds Deuterium is often used in internal standards, but it suffers from the chromatographic problems described earlier. Despite these problems, it is often used owing to the ease of incorporation of the label into molecules, the plethora of deuterium-labeled reagents commercially available, and the relative low cost of those reagents when compared to those labeled with C-13 or N-15.
2.5.2 Carbon-13- and nitrogen-15-labeled compounds The reagents used to prepare compounds labeled with C-13 and/or N-15 are similar to those available for C-14 labeling, and a similar cost structure exists, with the less functionalized reagents being the least expensive. Typical reagents include K13C15N [65], [13C]paraformaldehyde [66], [13C6]aromatics [44], glycine and other amino acids, and Ba13CO3.
2.5.3 Oxygen-18-labeled compounds Oxygen-18 is seldom used in the pharmaceutical industry as carbonyl oxygens in ketones, amides, and esters are known to exchange with water, which makes these locations unsuitable for labeling as internal standards. Although the oxygen atoms in ethers and phenols are not subject to the same limitations, the supply of O-18-labeled starting materials available for purchase is much more limited. Oxygen-18 has been utilized in mechanistic investigations such as the elegant work performed on SIL rofecoxib, which was used in the elucidation of the mechanism of the enterohepatic recycling of the drug. In vivo, rofecoxib is
530
Charles S. Elmore
O O
O
O p-450
liver
O
gut
OH
MeO2S
O OH
O Gut OH
OH MeO2S
Scheme 10
O-Glu MeO2S
MeO2S
Rofecoxib
O
O MeO2S
Metabolism and reuptake of rofecoxib [6].
metabolized to the 5-hydroxyrofecoxib and subsequently converted to the glucuronide which is secreted in the bile. In the gut, the glucuronide was cleaved and the ring-opened aldehyde form of 5-hydroxyrofecoxib was reduced to the alcohol. The alcohol then spontaneously cyclized to give the parent drug, which could be re-absorbed into the blood stream [6] (Scheme 10).
3. HUMAN ADME STUDIES Before submission of a New Drug Application (NDA), a human ADME study must be run; this is generally performed in late phase I or in phase II clinical trials. The objectives of these studies are to determine the recovery of the administered dose (mass balance), the route of elimination, and the metabolic profile of the drug in plasma. Although earlier in vitro and in vivo studies provide metabolism data, the radiolabeled ADME study is the first to assess these data in vivo in humans. Unanticipated metabolic data in the form of a new metabolite or unexpected abundance of an existing one can signal the need for additional safety studies before continuing with clinical development [67]. Human ADME studies generally use 4–10 healthy subjects or patients (for oncology products) who are given a single oral or parenteral dose (generally 50–200 mCi). Plasma, urine, and feces are then collected over a timeframe that is expected to allow for the majority of the radiolabel to be excreted. These studies require a large amount of planning
The Use of Isotopically Labeled Compounds in Drug Discovery
531
and coordination between the radiochemistry group and the safety group, so at least 6 months should be allotted for the synthesis. Generally, carbon14 is used as the radiolabel choice because it often avoids the potential loss of the label or metabolic isotope fractionation of the metabolites. Tritium has been used for many studies when C-14 labeling is impractical or otherwise does not suit the needs of the study [68]. The synthesis of radiolabeled material for human administration poses challenges not encountered in the generation of non-radioactive active pharmaceutical ingredient (API). Since radioactive compounds are unstable, the radiochemical stability of the labeled API must be assessed by conducting a stability study; this entails storing a batch of radiolabeled tracer for a period of time and monitoring its impurity profile. If the decomposition is too rapid, alternative storage conditions, formulations, or SA may be needed to slow the rate of decomposition. If significant radioactive impurities develop in the sample (W0.2%), they may need to be identified. This can be a significant challenge if the impurities are different from those found in the parent drug [69]. Formulation of the API can also present a challenge as radiolabeled compounds are typically dosed as solids or liquids in capsules for human ADME studies and not as tablets. This is done to avoid radioactive contamination of the tablet press. Owing to the scale difference between the non-radioactive and radioactive human preparations, complete replication of the process research synthetic route is impossible. This can result in either a difference in particle size and/or alternate crystal forms. Any type of particle size reduction technique used in the manufacturing of the non-radioactive API can be extremely difficult to replicate, and differences in particle size can lead to differing exposures of the tracer than expected. Sonication has been demonstrated to be an alternative to milling to reduce particle size in some cases [70]. Oral solutions can be used if the compound has an appropriate solubility in the dosing solution, but this is suboptimal as the dosing solution may alter the metabolism of the tracer. In general, the method of dosing the radiolabeled API should be as close as possible to the unlabeled material. The actual synthesis also presents challenges. Since the radiolabeled API is required only for one study, only one batch will be produced, typically in laboratories that also produce non-GMP materials. This precludes application of GMP requirements; however, GMP-like conditions have been accepted by the regulatory agencies so long as the subject is appropriately protected. Another difficulty in preparing a compound for a human ADME study is the analytical testing that must be performed on the labeled compound. The analytical data typically acquired on a radiolabeled substrate must be gathered (HPLC, LC/MS, NMR), but other data are also required to ensure the quality of the material for human administration
532
Charles S. Elmore
including weight percent, residual solvent, particle size, residual metal, IR, and similar assays based on the release criteria for the non-radioactive API. These tests can consume a significant amount of the radiolabeled material especially if the material is to be dosed at high SA.
4. CONCLUSIONS The synthesis of radiolabeled compounds can be an integral component of any drug discovery effort from the earliest high-throughput screens to human ADME to post-marketing studies. A basic understanding of labeling methods and the uses to which the radiotracers are put will better equip the medicinal chemist when interacting with radiochemists either internal to the company or within a contract laboratory.
REFERENCES [1] [2] [3] [4] [5]
[6]
[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]
A. P. Davenport and F. D. Russell, Dev. Nucl. Med., 1996, 30, 169. P. H. Marathe, W. C. Shyu and W. G. Humphreys, Curr. Pharma. Design, 2004, 10, 2991. L. Huang, Y. Wang and S. Grimm, Drug Metab. Dispos., 2006, 34, 738. M. Palmblad, B. A. Buchholz, D. J. Hillegonds and J. S. Vogel, J. Mass Spectrom., 2005, 40, 154. T. J. Ognibene and J. S. Vogel, in Synthesis and Applications of Isotopically Labelled Compounds, Proceedings of the International Symposium Volume 8 (eds D. C. Dean, C. N. Filer, and K. E. McCarthy), Wiley, Chichester, UK, 2004, p. 293. T. A. Baillie, R. A. Halpin, B. K. Matuszewski, L. A. Geer, C. M. Chavez-Eng, D. Dean, M. Braun, G. Doss, A. Jones, T. Marks, D. Melillo and K. P. Vyas, Drug Metab. Dispos., 2001, 29, 1614. A. N. Jones, D. C. Dean, H. J. Jenkins, D. G. Melillo, R. P. Nargund and M. A. Wallace, J. Labelled Compd. Radiopharm., 1996, 38, 561. In Guide to the Self-Decomposition of Radiochemicals, Amersham International Plc, Buckinghamshire, 1992. G. T. Miwa and A. Y. H. Lu, BioEssays, 1987, 7, 215. H. Morimoto and P. G. Williams, Fusion Tech., 1992, 21, 256. W. J. S. Lockley, J. Labelled Compd. Radiopharm., 1987, 24, 1509. E. Rapkin, G. Steele and R. Schavey, Am. Lab., 1995, 27, 31. K. Pfanner and A. Zeller, J. Labelled Compd. Radiopharm., 1998, 41, 1033. A. Di Marco, A. Cellucci, A. Chaudhary, M. Fonsi and R. Laufer, Drug Metab. Dispos., 2007, 35, 1737. J. R. Heys, J. Labelled Compd. Radiopharm., 2007, 50, 770. D. Hesk, P. R. Das and B. Evans, J. Labelled Compd. Radiopharm., 1995, 36, 497. W. Chen, K. T. Garnes, S. H. Levinson, D. Saunders, S. G. Senderoff, A. Y. L. Shu, A. J. Villani and J. R. Heys, J. Labelled Compd. Radiopharm., 1997, 39, 291. M. E. Powell, C. S. Elmore, P. N. Dorff and J. R. Heys, J. Labelled Compd. Radiopharm., 2007, 50, 523. J. A. Brown, S. Irvine, A. R. Kennedy, W. J. Kerr, S. Andersson and G. N. Nilsson, Chem. Comm., 2008, 1115. M. B. Skaddan, C. M. Yung and R. G. Bergman, Org. Lett., 2004, 6, 11.
The Use of Isotopically Labeled Compounds in Drug Discovery
533
[21] C. M. Yung, M. B. Skaddan and R. G. Bergman, J. Amer. Chem. Soc., 2004, 126, 13033. [22] A. Y. L. Shu, D. Saunders, S. H. Levinson, S. W. Landvatter, A. Mahoney, S. G. Senderoff, J. F. Mack and J. R. Heys, J. Labelled Compd. Radiopharm., 1999, 42, 797. [23] W. J. Wheeler and D. K. Clodfelter, J. Labelled Compd. Radiopharm., 2008, 51, 175. [24] S. Hintermann, I. Vranesic, H. Allgeier, A. Bru¨lisauer, D. Hoyer, M. Lemaire, T. Moenius, S. Urwyler, S. Whitebread, F. Gasparini and Y. P. Auberson, Bioorg. Med. Chem., 2007, 15, 903. [25] H. J. Dutton, C. R. Scholfield, E. Selke and W. K. Rohwedder, J. Catalysts, 1968, 10, 316. [26] S. D. Wyrick, S. Morris-Natschke and P. K. Lauf, J. Labelled Compd. Radiopharm., 1984, 21, 173. [27] M. Saljoughian and P. G. Williams, Curr. Pharma. Design, 2000, 6, 1029. [28] B. Latli, M. Hrapchak, D. Krishnamurthy and C. Senanayake, J. Labelled Compd. Radiopharm., 2008, 51, 106. [29] A. Chaudhary, Y. Tang, D. Dean, W. Ashton and D. Melillo, in Synthesis and Applications of Isotopically Labelled Compounds, Proceedings of the International Symposium, Volume 8 (eds D. C. Dean, C. N. Filer, and K. E. McCarthy), Wiley, Chichester, UK, 2004, p. 425. [30] Y. S. Tang, W. Liu, A. Chaudhary, D. G. Melillo and D. C. Dean, in Synthesis and Applications of Isotopically Labelled Compounds, Proceedings of the International Symposium Volume 8 (eds D. C. Dean, C. N. Filer, and K. E. McCarthy), Wiley, Chichester, UK, 2004, p. 71. [31] D. C. Evans, A. P. Watts, D. A. Nicoll-Griffith and T. A. Baillie, Chem. Res. Toxicol., 2004, 17, 3. [32] E. Bannwart, A. Zeller, P. Stro¨m and M. Skrinjar, in Synthesis and Applications of Isotopically Labelled Compounds, Volume 7 (eds U. Pleiss and R. Voges), Wiley, Chichester, UK, 2001, p. 664. [33] S. R. Prakash and R. L. Ellsworth, J. Labelled Compd. Radiopharm., 1988, 25, 815. [34] K. G. Cuevas-Licea, N. X. Yu, S. J. Staskiewicz and C. E. Raab, J. Labelled Compd. Radiopharm., 2007, 50, 513. [35] M. A. Tetrick, T. D. Crenshaw and N. J. Benevenga, Anal. Biochem., 1997, 248, 1. [36] M. P. Braun, D. C. Dean and D. G. Melillo, J. Labelled Compd. Radiopharm., 1999, 42, 469. [37] C. S. Elmore, D. C. Dean, T. M. Marks, M. P. Braun, M. A. Egan and D. G. Melillo, in Synthesis and Applications of Isotopically Labelled Compounds, Volume 8 (eds D. C. Dean, C. N. Filer, and K. E. McCarthy), Wiley, Chichester, UK, 2004, p. 15. [38] T. Werner, S. Berg and R. Johansson, J. Labelled Compd. Radiopharm., 2004, 47, 175. [39] Y. Gong, D. C. Hoerr, L. E. Weaner and R. Lin, J. Labelled Compd. Radiopharm., 2008, 51, 268. [40] J. Z. Ho, C. S. Elmore and M. P. Braun, J. Labelled Compd. Radiopharm., 2008, 51, 399. [41] P. Stro¨m and J. Malmquist, J. Labelled Compd. Radiopharm., 2008, 51, 419. [42] C. S. Elmore, D. C. Dean and D. G. Melillo, J. Labelled Compd. Radiopharm., 2000, 43, 1135. [43] C. S. Elmore, D. C. Dean, R. J. Devita and D. G. Melillo, J. Labelled Compd. Radiopharm., 2003, 46, 993. [44] C. F. Lavey, D. Hesk, S. Hendershot, D. Koharski, S. Saluja and P. Mc Namara, J. Labelled Compd. Radiopharm., 2007, 50, 264. [45] J. T. Kendall, J. Labelled Compd. Radiopharm., 1999, 42, 477. [46] C. S. Elmore, D. C. Dean, K. Liu, A. B. Jones and D. G. Melillo, in Synthesis and Applications of Isotopically Labelled Compounds, Volume 7 (eds U. Pleiss and R. Voges), Wiley, Chichester, UK, 2001, p. 224. [47] J. Z. Ho, C. S. Elmore, M. A. Wallace, D. Yao, M. P. Braun, D. C. Dean, D. G. Melillo and C.-Y. Chen, Helv. Chim. Acta, 2005, 88, 1040. [48] J. T. Kendall and M. K. May, in Synthesis and Applications of Isotopically Labelled Compounds, Volume 7 (eds U. Pleiss and R. Voges), Wiley, Chichester, UK, 2001, p. 221.
534
Charles S. Elmore
[49] D. Hendry, N. S. Nixon, B. S. Roughley, P. Skagestad and P. M. Winton, in Synthesis and Applications of Isotopically Labelled Compounds, Volume 8 (eds D. C. Dean, C. N. Filer, and K. E. McCarthy), Wiley, Chichester, UK, 2004, p. 25. [50] B. Latli, M. Hrapchak, V. Gorys, C. A. Busacca and C. Senanayake, J. Labelled Compd. Radiopharm., 2005, 48, 447. [51] S. K. Johansen, J. Labelled Compd. Radiopharm., 2005, 48, 1025. [52] Y. Li and S. R. Prakash, J. Labelled Compd. Radiopharm., 2005, 48, 323. [53] J. F. Eggler, C. A. Gabel, K. Zandi, J. D. Weaver, H. McKechney, M. A. Dombroski, H. Peurano and N. Hawryluk, J. Labelled Compd. Radiopharm., 2002, 45, 785. [54] C. E. Raab, D. C. Dean and D. G. Melillo, J. Labelled Compd. Radiopharm., 2001, 44, 815. [55] D. C. Dean, R. P. Nargund, S.-S. Pong, L.-Y. P. Chaung, P. Griffin, D. G. Melillo, R. L. Ellsworth, L. H. T. Van Der Ploeg, A. A. Patchett and R. G. Smith, J. Med. Chem., 1996, 39, 1767. [56] S. Ren, P. McNamara, D. Koharski, D. Hesk and S. Borges, J. Labelled Compd. Radiopharm., 2007, 50, 395. [57] M. A. Wallace, C. Raab, D. Dean and D. Melillo, J. Labelled Compd. Radiopharm., 2007, 50, 347. [58] M. A. Wallace, C. E. Raab, D. C. Dean and D. G. Melillo, J. Labelled Compd. Radiopharm., 2005, 48, 275. [59] M. Murai, K. Sekiguchi, T. Nishioka and H. Miyoshi, Biochemistry, 2009, 48, 688. [60] S. V. Dhuria, L. R. Hanson and W. H. Frey, J. Pharmacol. Exp. Ther., 2009, 328, 312. [61] A. T. Yordanov, M. Hens, C. Pegram, D. D. Bigner and M. R. Zalutsky, Nucl. Med. Biol., 2007, 34, 173. [62] S. H. Song, K.-H. Jung, J.-Y. Paik, B.-H. Koh, J.-S. Bae, Y. S. Choe, K.-H. Lee and B.-T. Kim, Nucl. Med. Biol., 2005, 32, 845. [63] A. E. Bolton and W. M. Hunter, Biochem. J., 1973, 133, 529. [64] A. Valleix, S. Carrat, C. Caussignac, E. Le´once and A. Tchapla, J. Chromatogr. A, 2006, 1116, 109. [65] C. S. Elmore, D. C. Dean, Y. Zhang and D. G. Melillo, J. Labelled Compd. Radiopharm., 2004, 47, 837. [66] C. C. Vu and L. A. Peterson, J. Labelled Compd. Radiopharm., 2005, 48, 117. [67] D. A. Smith and R. S. Obach, Chem. Res. Toxicol., 2009, 22, 267. [68] Y. Hong, S. J. Bonacorsi, Jr., Y. Tian, S. Gong, D. Zhang, W. G. Humphreys, B. Balasubramanian, E. H. Cheesman, Z. Zhang, J. F. Castner and P. D. Crane, J. Labelled Compd. Radiopharm., 2008, 51, 113. [69] A. N. Jones, M. Braun, D. Dean, C. Elmore, Y. Jakubowski, H. Jenkins, D. Melillo, R. Miller, S. Staskiewicz and M. Wallace, in Synthesis and Applications of Isotopically Labelled Compounds, Volume 8 (eds D. C. Dean, C. N. Filer, and K. E. McCarthy), Wiley, Chichester, UK, 2004, p. 289. [70] D. J. Schenk, A. N. Jones, M. P. Braun, D. Yao, M. A. Wallace, R. Marques, H. J. Jenkins, A. Chang, X. Jia, L. S. Crocker, D. C. Dean and D. G. Melillo, J. Labelled Compd. Radiopharm., 2004, 47, 399.
CHAPT ER
26 Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450 Bernard P. Murray
Contents
1. Introduction 2. Reversible, Irreversible, and Quasi-Irreversible Mechanism-Based Inhibition 3. Practical Determination 4. Inhibition of Cytochromes P450 4.1 Functional groups and target enzymes 4.2 Sequelae of mechanism-based inhibition 5. Clinical Relevance 5.1 Magnitude of effect in vivo 5.2 Practical use of mechanism-based inhibitors 6. Conclusions Acknowledgments References
535 536 539 543 543 547 549 549 549 551 551 551
1. INTRODUCTION Drugs that have the potential to cause drug–drug interactions through enzyme inhibition are generally at a competitive disadvantage compared to those with cleaner profiles, so one of the main tasks of research scientists during compound optimization is to minimize this liability. Drug metabolizing cytochromes P450, such as CYP3A4, are a particularly important class of target enzymes in this regard. For compounds that Drug Metabolism Department, Gilead Sciences, Foster City, CA, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04426-1
r 2009 Elsevier Inc. All rights reserved.
535
536
Bernard P. Murray
interact reversibly with the affected enzyme, quantitative prediction of the extent and duration of effect are relatively straightforward, as these can be calculated from the pharmacokinetics of the ‘‘perpetrator’’ drug and the Ki for inhibition of the enzyme. The Ki values can be determined in relatively simple in vitro studies and there are well-characterized assays available for this purpose [1]. The situation is more complex for agents that inactivate enzymes, as the time course for inhibition and recovery is decoupled from the pharmacokinetic profile of the inhibitor and is dependent on the rates of degradation and resynthesis of the affected protein. Assaying the kinetic constants (kinact and KI) of a mechanism-based inhibitor (MBI) is also more challenging and time-consuming than measuring the IC50 or Ki values for a reversible inhibitor, but the extra effort can usually be justified. For example, in a recent case study [2], a kinetic and mechanistic understanding of mechanism-based inhibition caused by 1 led to rational modification to 2, eliminating the inhibitory liability. Cl
H N X O N
O
NH2
O
NH
1X=H 2 X = CH3
2. REVERSIBLE, IRREVERSIBLE, AND QUASI-IRREVERSIBLE MECHANISM-BASED INHIBITION Compounds that inactivate cytochromes P450 in a selective manner (as opposed to agents that chemically modify the enzyme directly) require metabolism by the enzyme and are most commonly known as mechanism-based inhibitors/inactivators, but they may also be termed
Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450
537
suicide inhibitors or (preincubation) time-dependent inhibitors (TDIs). Metabolism of a substrate can lead to inhibition of an enzyme in various ways. While it is theoretically possible for a cytochrome P450 substrate to be metabolized to a form that interacts more potently, but still freely reversibly, with the enzyme, there are no clear examples of such an agent. This is likely because the products of metabolism by a cytochrome P450 are generally less hydrophobic than the substrate and therefore will bind less strongly to the enzyme. For example, CYP3A4 can convert itraconazole (itself a potent CYP3A4 inhibitor) to several metabolites that inhibit the enzyme and have been shown to contribute to clinical drug interactions [3], but the affinities of these metabolites for CYP3A4 are lower than that of the parent and it is their concentrations, rather than their relative potencies, that determines their contributions to the drug interaction. According to a recent literature survey [4], there is some evidence that the involvement of metabolites in inhibitory clinical drug interactions may be more prevalent than is currently appreciated, but this hypothesis remains to be tested. For true mechanism-based inhibition, the off-rate for dissociation of the metabolite from the enzyme should be much slower than the rate of elimination of the parent drug and should instead be of the same order as the rate of turnover of the affected protein. Even highly potent reversible ligands have off-rates of the order of W103 s1. This gives a dissociation half-life of o12 min, not slow enough to achieve as prolonged an effect as a MBI. For example, surface plasmon resonance analysis of the interaction of the low-spin (‘‘type II’’) inhibitor ketoconazole with CYP3A4 found off-rates for the two enantiomers of 6.1 103 and 7.1 103 s1 [5] (yielding dissociation half-lives of 114 and 98 s, respectively), confirming that although the dissociation rate was classified as ‘‘slow,’’ it is still much faster than the rate of elimination of the compound in vivo. Although a long duration of effect (hours or days) is needed, covalent attachment to P450 apoprotein or heme is not a prerequisite for mechanism-based inhibition as metabolites of so-called quasi-irreversible inhibitors can also form tight complexes with the enzyme (vide infra). This distinction between covalent and non-covalent modification is important, as degradation of proteins modified by covalent binding inhibitors can result in the generation of potentially immunogenic peptide adducts that could elicit hypersensitivity reactions to the inhibitor, while quasi-irreversible inhibitors leave the protein untouched. Two classes of quasi-irreversible MBIs discussed below are methylenedioxyphenyl compounds, such as isosafrole (3), and alkylamines, such as erythromycin (4) and troleandomycin. After treatment of rats with these compounds in vivo, the enzyme-inhibitory metabolite complexes are sufficiently stable that they remain intact during preparation of the microsomal fractions and purification of the enzymes (CYP1A2 and
538
Bernard P. Murray
CYP3A, respectively). However, active enzyme can be regenerated using relatively mild conditions such as treatment with potassium ferricyanide or displacement of bound metabolite with a high-affinity ligand such as ethoxycoumarin [6,7]. OH O O N
OH O O O
O O OH
HO
O
OH
O
O
3
4
Other compounds cause mechanism-based inhibition through irreversible changes to the protein. These changes include covalent binding to the apoprotein, binding to (or loss of ) the heme, or destruction of the heme leading to modification of the protein by heme fragments. In some cases, there are multiple mechanisms for inactivation by a single compound, as has been demonstrated [8,9] with secobarbital (5) and dihydropyridine derivatives such as DDEP (6). In almost all cases, the enzyme cannot be repaired, and therefore, the rate of recovery of enzyme activity is dependent on the rate of synthesis of new protein. The exception to this is when the heme is destroyed without modification of the apoprotein, as this may allow reconstitution of holoenzyme by introduction of fresh heme. This process is only possible with some inactivated enzymes, and a role for the heat shock protein, GRP94, in the reconstitution of CYP2B1 inactivated by allylisopropyl acetamide (7) has been shown [10].
O
O O
NH O 5
N H
O O N H
O 6
H2N
O 7
Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450
539
The aim of this chapter is to summarize recent advances in mechanismbased inhibition of cytochromes P450, with CYP3A4, the major catalyst of human drug metabolism, as the primary illustration but with comparisons with other enzymes as required. General aspects of mechanism-based inhibition, pharmacokinetic enhancement, and a catalog of various drugs associated with the phenomenon have been the subject of recent reviews [e.g., 11–15]. After a brief overview, the focus of this chapter is on recent advances in the quantitative prediction of clinical liability, studies giving insight into active site modifications by inhibitors, and studies on postinactivation events.
3. PRACTICAL DETERMINATION Silverman [16] proposed a widely used list of criteria for assignment of an agent as a MBI including requirements such as determination of the stoichiometry for inactivator/enzyme binding, but for most purposes for cytochromes P450, the focus is on demonstration of time and cofactor (NADPH) dependence of the inactivation. Since metabolism of the inhibitor is a prerequisite for mechanism-based inhibition, the kinetics of inactivation are described in a similar manner to those for normal metabolism of substrates. The Henri–Michaelis–Menten equation for metabolism of a substrate, S, by enzyme, E, with velocity, v, can be written as follows: v¼
kcat ½E½S Km þ ½S
(1)
The catalytic constant (or turnover number) is kcat and the Michaelis affinity constant is Km. It is common to combine kcat and the enzyme concentration as the Vmax: Vmax ¼ ½Ekcat
(2)
Enzyme inactivation kinetics are described in a similar manner, except that the enzyme is consumed in the reaction as well as the substrate, and the velocity of the reaction refers primarily to the rate of enzyme loss. Formal determination of the kinetics for MBIs commonly involves a two-step procedure: (1) a ‘‘preincubation’’ step intended to allow optimal conditions for inactivation of the enzyme in the absence of a competing substrate and (2) an enzyme assay step, wherein the remaining activity of the enzyme is determined. Since MBIs are substrates for the target enzyme, and can thus act as competitive inhibitors, dilution of the reaction mixture is required between the two steps to reduce the concentration of the residual inhibitor in the enzyme assay reaction (e.g., 10–20 ). Varying the inhibitor concentration, [I], and preincubation time, t, during the first
540
Bernard P. Murray
step should result in a family of curves showing a preincubation timedependent log-linear loss of enzyme activity [E]t from the initial value [E]0 described by the following equation: ½Et ¼ ½E0 ekobs t ,
(3)
where kobs is the inactivation rate at a given inhibitor concentration: kobs ¼
kinact ½I KI þ ½I
(4)
The enzyme inactivation parameters, kinact and KI (analogous to the Vmax and Km for the reaction), can be illustrated with a Kitz–Wilson plot [17] and the values determined by non-linear curve fitting. Example data from such an analysis are shown in Figure 1. Since the compound can usually be converted to non-inhibitory metabolites as well as causing mechanism-based inhibition, the relative rates for the two processes can be compared using the partition ratio, r, which can vary from r ¼ 0 (kcat ¼ 0, so every catalytic cycle leads to inactivation) to r ¼ N (kinact ¼ 0, so no detectable inactivation): r¼
kcat kinact
(5)
For quasi-irreversible inhibitors, another potential way to determine inactivation kinetics is by measurement of spectral changes in the enzyme [18]. Binding of the metabolite results in the formation of a metabolic intermediate complex (MIC) that perturbs the heme resulting in a change in the visible spectrum for the enzyme (a Soret peak with increased absorption at B455 nm). Computational models to predict MIC formation between CYP3A4 and ligands have been developed [19], and a pharmacophore containing at least four hydrophobic interactions and a hydrogen bond acceptor has been suggested. The enzyme inhibition assay procedure works well when the inhibitor is relatively water soluble, as this permits the use of high inhibitor concentrations to saturate the enzyme and thus allows accurate determination of kinact values. It is also easier when the reversible inhibition by the compound is weak without preincubation, as this reduces interference by residual inhibitor in the second step. High metabolic stability and low microsomal binding also make determining kinetics easier. In this regard, erythromycin is relatively well-behaved, but in contrast, the HIV protease inhibitor, ritonavir (8), is relatively insoluble, is a potent inhibitor of CYP3A even in the absence of preincubation, and is metabolically unstable at low concentrations. Accordingly, there is some variability in the kinetics for inactivation by ritonavir that have been determined in different laboratories [20–23]. An alternative approach to overcome some of these issues is to use progress curve analysis, as has been demonstrated for
Enzyme activity remaining (%)
Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450
541
125 100
0 x KI 0.1 x KI
75
0.3 x KI
50
1.0 x KI
25
10 x KI
3.0 x KI
0 0
4
12
8
16
Preincubation time
(a) 0.20
kobs
0.15 0.10 0.05 0.00 0
20
40
60 [I]
(b)
80
100
120
60 50
1/kobs
40 30 20 10
-0.2 (c)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1/[I]
Figure 1 Simulated data from an enzyme inactivation experiment. Panel A shows a plot of enzyme activity remaining with a range of preincubation times and with various concentrations of inhibitor (multiples of the KI for inactivation). Panel B shows the resulting inactivation rate (kobs) values plotted against inhibitor concentration. The data are fitted to a rectangular hyperbola. Panel C shows Kitz–Wilson transformation of the data from Panel B.
542
Bernard P. Murray
CYP1A2 [24], in which direct inhibition and metabolic instability are modeled as part of the inactivation process.
O
O H N
N
S N
N H
N H O
OH
O
S N
8
A common adaptation to avoid inhibition by residual inhibitor is to use a relatively high substrate concentration in the second step (e.g., 5–10 Km). When the enzyme to be assayed is CYP3A, this approach should be used with caution as CYP3A enzymes can display atypical kinetics with some substrates [25]. Similarly, increasing the duration of the preincubation leads to confounding effects as CYP3A4 is particularly susceptible to autocatalytic inactivation due to futile cycling [26]. The formal determination described above is time-consuming, even when fast mass spectrometry-based assays are used to quantify the analytes. Some degree of automation can be employed [27,28], but accuracy and precision usually suffer somewhat when potent inhibitors are used. A common general strategy for dealing with MBIs in a drug discovery setting is to employ faster, more minimal, assays to steer projects away from this liability and then to determine inactivation kinetics formally to check representative analogs and the final candidates. A common approach for a faster assay is the ‘‘IC50 shift,’’ which is an adaptation of the procedure for determining a simple IC50 value, wherein a step with preincubation of enzyme cofactor and inhibitor is included before the addition of substrate to the same reaction [29]. Another method is to use a ‘‘2 2’’ format (7 a single concentration of inhibitor and 7 a single preincubation time) [23]. In some formats, the assay is made faster by omission of the dilution step. Because of their compatibility with highthroughput screening techniques, protocols using substrates yielding fluorescent or luminescent metabolites can be useful [30]. There are few such substrates that are sufficiently enzyme selective, and therefore, individual recombinant enzymes have to be used as the catalysts. Caution in interpretation is also required due to the potential for direct quenching of the fluorescent or luminescent signal by the inhibitor.
Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450
543
4. INHIBITION OF CYTOCHROMES P450 4.1 Functional groups and target enzymes Excellent reviews that discuss the functional groups that give rise to mechanism-based inhibition of cytochromes P450 are available [12,31,32]. As discussed above, such lists overlap to some extent with those of functional groups that are potential precursors of reactive metabolites [13,33,34], as mechanism-based inhibitors that covalently modify P450 heme or apoprotein are almost always reactive metabolites. As mentioned above, exceptions are the moieties that result in quasi-irreversible inhibition rather than covalent modification. The main classes of functional groups conferring quasi-irreversible inhibition are methylenedioxyphenyl derivatives [35,36], such as 3, piperonyl butoxide (9), methylenedioxymethamphetamine (MDMA, 10), and paroxetine (11), which are likely metabolized to heme-binding carbene derivatives, and amine derivatives, such as 4, SKF-525A (12), diltiazem (13), and verapamil (14), which are likely metabolized to nitroso derivatives. Although for these classes the focus is on non-covalent interactions, methylenedioxyphenyl compounds may not be completely innocuous, as demethylenation to a catechol derivative is a common alternative pathway and there can then be further conversion to reactive metabolites [37]. On the contrary, cyclopropylamines such as 15 were originally thought to be metabolized through single-electron transfer to cyclopropylidene forms that bind covalently to the enzyme, but recent data demonstrated that for cytochromes P450, the mechanism-based inhibition results in MIC formation and inhibition that could be reversed by ferricyanide treatment, strongly suggesting that it is quasi-irreversible [38]. O O
O
O
O
O
9
HN
O
10
H N O
O O
F 11
544
Bernard P. Murray
S O
N
N
O N
O
O
O O 12
13
O O
N
N
O
N H
O 14
15
For quasi-irreversible inhibitors, the complexes between the inhibitory metabolite and the cytochrome P450 heme will be broken when the holoenzyme is degraded. In contrast, examples of moieties leading to covalent modification during inhibition include the following: – Furan derivatives, such as the grapefruit juice constituents 6u,7u-dihydroxybergamottin (16), paradisin A (17), and paradisin B, all of which can inactivate CYP3A4, likely through the formation of a furanoepoxide intermediate. O OH HO
O
O
O 16
545
Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450 O
OH O
O
O
O
O
O
OH O O
17
– Thiophene-containing compounds, such as tienilic acid (18), ticlopidine (19), and suprofen (20), which inhibit enzymes of the CYP2C subfamily, likely through S-oxidation or formation of a thiophenoexpoxide. A recent study showed covalent binding of another thiophene derivative, OSI-930 (21), to CYP3A4, but the possibility of mechanismbased inhibition was not examined [39].
OH OH
S O
O
N
Cl
S
Cl
O
Cl
18
O
S
O
19
20
O
CF3
O N H S
NH
N 21
– Alkenes as well as alkynes, such as 5, 17a-ethynyl estradiol (22) and mifepristone (23), which can be activated to reactive epoxides that can bind to the P450 apoprotein and/or heme.
546
Bernard P. Murray
N HO
HO
HO
O 22
23
– Other miscellaneous functional groups including sulfur-containing compounds, such as the thiol metabolite of spironolactone (24), which can undergo thiol oxidation, and cyclic tertiary amines such as phencyclidine (PCP, 25), which can be oxidized to an iminium species. A similar mechanism likely plays a role with the piperidine derivative, SCH 66712, which has been shown to inactivate CYP2D6 [40].
O O N
O
SH 24
25
Simple inspection of the structure may not always be sufficient to identify the moiety responsible for inhibition. When furafylline (26) was found to be a MBI of human CYP1A2, it was commonly thought that epoxidation of the furan ring was responsible, an assumption that has persisted [e.g., 32], despite later studies showing that oxidation at the 8-methyl position is involved [41]. Some molecules contain more than one group that can be associated with mechanism-based inhibition. For example, inhibition of CYP2B1 by clorgyline (27) can be reversed by ferricyanide treatment, suggesting that quasi-irreversible inhibition occurs after metabolism at the amine moiety, while inhibition of CYP1A2 is irreversible and can be reduced by including glutathione in the reaction, suggesting that the terminal alkyne is the site of metabolism. PH-302 (28) contains methylenedioxyphenyl and alkylamine groups, as
547
Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450
well as an imidazole that could inhibit P450 directly through type II complex formation, and therefore, it is not surprising that interactions with CYP3A4 are complex [42]. The pathways for enzyme inactivation by some well-known MBIs still remain to be fully elucidated. For example, carbamazepine (29) may require generation of the 3-hydroxy metabolite to inactivate CYP3A4 [43].
O O
N
N
N
Cl N
N H
O Cl
O 26
H2N
N
N
O
27
O O
N
N N
N
N H2N
28
O
29
For CYP3A4, the nature of the drug-protein adducts resulting from inactivation by covalent binding inhibitors is the subject of great interest. Most recently, several lines of evidence point to Cys239 as being the target nucleophile for activated raloxifene and carbamazepine [44,45] with the absence of this residue in the related enzyme, CYP3A5, being the likely explanation as to why it is resistant to inactivation. Another recent publication also described covalent binding by MBIs that inactivate CYP3A4 but not CYP3A5 [46], so this may be an emerging theme.
4.2 Sequelae of mechanism-based inhibition Drug metabolizing cytochromes P450, like all endoplasmic reticulum proteins, are synthesized and degraded continuously, even in relatively quiescent tissues such as the liver, and mechanism-based inhibition of
548
Bernard P. Murray
cytochromes P450 has the potential to change both the rate and the route of turnover of the affected enzymes. While bulk endoplasmic reticulum is turned over every few days [47], cytochromes P450 are turned over asynchronously with rates varying from enzyme to enzyme [48,49]. The two main mechanisms for degradation are through the macroautophagic-lysosomal and ubiquitin-proteasomal pathways [50], and under normal circumstances, the choice of pathway may be intrinsic to each protein, for example, due to the presence of degrons within the primary sequence or tertiary structure. Inactivation of a cytochrome P450 by a MBI can have a variety of effects on the turnover of the protein: 1.
2.
3.
Accelerated degradation: For example, as is seen with intestinal CYP3A and grapefruit juice [51] and rat hepatic CYP3A and DDEP [52]. Stabilization: For example, treatment of rats with 3 or 4 results in accumulation of complexes with CYP1A2 or CYP3A enzymes, respectively (vide supra). No apparent effect: For example, treatment with the quasiirreversible inhibitor diltiazem inactivates intestinal CYP3A without loss of immunodetectable protein [53].
Inactivation can also change the route of degradation. For example, native CYP2B1 appears to be turned over primarily by the lysosomal route, but inactivation can result in a shift to the ubiquitin-proteasomal pathway [54]. The determinants for these responses to inactivation are still the subject of study. When the rate of degradation is accelerated, it is likely that there has been a structural change that is recognized by the endoplasmic reticulum quality control process, and thus, the inactivated protein is targeted for endoplasmic reticulum-associated degradation (ERAD) [55–57]. This requires ubiquitination and extraction from the endoplasmic reticulum membrane, but the details of the process are not yet fully known. For CYP3A enzymes, recent results suggest that phosphorylation by cytosolic kinases, ubiquitination by gp78 and CHIP E3 ligases, extraction involving the AAA-ATPase, p97, and then degradation by the 26S proteasome [50,58–60] are involved. Since native CYP3A4 also appears to be turned over by the ubiquitin proteasomal pathway [61], inactivation of CYP3A enzymes likely changes their rates of degradation but not the route. Since adducts with the cytochrome P450 protein will survive proteolytic degradation, inhibitors that inactivate enzymes through covalent modification can potentially lead to neoantigen production and hypersensitivity [33,62]. Examples include tienilic acid, which inactivates CYP2C9, and carbamazepine, which inactivates CYP3A4 and result in antibodies against the adducted proteins.
Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450
549
5. CLINICAL RELEVANCE 5.1 Magnitude of effect in vivo Calculation of the likely magnitude of the clinical effect of an inhibitor can be performed relatively simply for a reversible inhibitor. For example, an estimation of the change in clearance of a ‘‘victim’’ drug from the uninhibited state (CL0) to the inhibited state (CLi) can be made as follows: CLi 1 ¼ CL0 1 þ ð½I=Ki Þ
(6)
where [I] is a representative plasma concentration (Cmax, Caverage, etc.) and Ki is the inhibition constant measured in vitro. The equation can be adapted further for situations where clearance of the victim drug by the inhibited enzyme is not the only route of elimination. Other more sophisticated pharmacokinetic-pharmacodynamic models can replace the use of a single concentration of inhibitor with values that change over a dosing interval. For a MBI, the prediction of the clinical effect is more complex. A commonly used approach [18] calculates the change in clearance of a victim drug in a manner analogous to that used in Equation 6 for a reversible inhibitor: kdeg CLMBI ¼ CL0 kdeg þ kobs
(7)
where kobs is the inactivation rate from Equation 4 using a representative plasma concentration and kdeg is the rate constant for degradation of the target enzyme. The equation can also be adapted further as described for Equation 6. After the plasma concentration of the inhibitor declines, the enzyme activity will return to baseline at a rate determined by the rate of synthesis of the enzyme, ksynth. Problems arise in the prediction if the inhibitor is also an inducer (e.g., 5, 7, 29) as chronic treatment results in an increase in ksynth. It is clear that the in vivo kinetics of mechanism-based inhibition and recovery (both the magnitude of the effect and the time course) are very dependent on the rate of turnover of the target enzyme. Somewhat surprisingly, these rates are still the subject of much debate [49].
5.2 Practical use of mechanism-based inhibitors For most drug candidates, mechanism-based inhibition is a property to be avoided, and steps are taken to eliminate compounds with this liability. However, in some cases, inhibitors can be used to enhance the pharmacokinetics of co-administered agents. In this regard, MBIs have
550
Bernard P. Murray
an advantage as their inhibitory effects will persist after plasma concentrations (and the risk of off-target effects) have declined, and therefore, a long pharmacokinetic half-life is not required. With half-lives for cytochrome P450 turnover in the 12- to 48-h range, once-daily dosing is feasible for all but the most metabolically labile co-dosed agents unless there is significant induction. Arguably the first use of MBIs of cytochrome P450 as pharmacokinetic enhancers was over 60 years ago when they were employed as insecticide synergists [63]. Agents such as 9 inhibit insect P450 enzymes (likely those of the CYP6A subfamily) and increase the concentration and thus effectiveness of agents such as pyrethroids. A more clinically relevant historical example of mechanism-based inhibition exploited the CYP3A inhibitory properties of 13 or grapefruit juice in ‘‘cyclosporine sparing’’ regimens in the late 1980s and early 1990s when supplies of this important immunosuppressant were limited [64]. More recently, the HIV protease inhibitor ritonavir has seen regular use in the therapy of HIV as a pharmacokinetic enhancer, acting through mechanism-based inhibition of CYP3A, and is co-dosed with HIV protease inhibitors [65], the integrase inhibitor elvitegravir [66] and the CCR5 inhibitor aplaviroc [67]. Although it has been assumed that mechanism-based inhibition is due to reactive metabolite generation [15], there is currently no evidence for this. Planned use of an inhibitor-containing regimen during drug discovery may allow exploration of chemical space that would otherwise be inaccessible as there would no longer be a requirement for high metabolic stability of partner compounds, as long as they were largely eliminated by the inhibited enzyme. For example, the potent HIV protease inhibitor lopinavir is metabolically extremely unstable and it thus can only be co-dosed with ritonavir [68]. Regimens featuring inhibition should also have the advantage that interindividual variability in the pharmacokinetics of the enhanced agent will be greatly reduced and toxicity due to metabolites should also be decreased. Initial concern that chronic inhibition of CYP3A enzymes would result in adverse effects due to impairment of metabolism of endogenous substrates appears unfounded, as pharmacokinetic enhancement has now been used for over a decade. It thus seems that CYP3A4 activity is not inherently essential, and, to add further evidence, knockout of all murine CYP3A genes results in animals that are viable and physiologically unimpaired [69]. Another use of MBIs is in understanding the relative roles of hepatic and intestinal enzymes in the bioavailability of drugs. Grapefruit juice contains compounds such as 16 and 17 that inactivate intestinal CYP3A enzymes but do not affect hepatic activity. When the results of treatment with grapefruit juice are compared to those seen with ritonavir or clarithromycin (drugs that inactivate both hepatic and intestinal CYP3A), the contributions of the two tissues can be determined [70].
Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450
551
For clinical mechanism-based inhibition, a quasi-irreversible inhibitor should be preferred as the lack of covalent attachment to the protein means that there is less chance of eliciting a hypersensitivity reaction when the enzyme is degraded. Quasi-irreversible inhibition also has the advantage that it will be effective against all substrates, as it targets the catalytic mechanism directly by blocking the heme. As demonstrated [45] with 29, MBIs that covalently modify the protein may only affect a subset of substrates as they may only partially occlude the large active site. Also, as discussed above, covalent binding inhibitors of CYP3A4 may be overly specific and only inactivate that enzyme and leave partner drugs susceptible to metabolism by CYP3A5.
6. CONCLUSIONS In conclusion, recent advances in the understanding of MBIs of cytochromes P450 have provided a framework for the prediction of in vivo effects from in vitro data and have given us insight into the biology of the turnover of the proteins. The potent, long-lasting effects of MBIs of enzymes such as CYP3A4 will continue to be a liability for some drugs and a positive advantage for others, and therefore, a thorough understanding of the mechanism and kinetics of inhibition will always remain important.
ACKNOWLEDGMENTS I take the opportunity to show my appreciation for my mentors and colleagues, past and present, for many enjoyable discussions over the years. In particular, I thank my postdoctoral mentor, Prof. M. Almira Correia, as well as Prof. Alan Boobis, Dr. Liping Pan, and Dr. Mingxiang Liao.
REFERENCES [1] T. D. Bjornsson, J. T. Callaghan, H. J. Einolf, V. Fischer, L. Gan, S. Grimm, J. Kao, S. P. King, G. Miwa, L. Ni, G. Kumar, J. McLeod, R. S. Obach, S. Roberts, A. Roe, A. Shah, F. Snikeris, J. T. Sullivan, D. Tweedie, J. M. Vega, J. Walsh and S. A. Wrighton, Pharmaceutical Research and Manufacturers of America (PhRMA) Drug Metabolism/ Clinical Pharmacology Technical Working Group; FDA Center for Drug Evaluation and Research (CDER), Drug Metab. Dispos., 2003, 31, 815. [2] W. Tang, R. A. Stearns, R. W. Wang, R. R. Miller, Q. Chen, J. Ngui, R. K. Bakshi, R. P. Nargund, D. C. Dean and T. A. Baillie, Xenobiotica, 2008, 38, 1437. [3] I. E. Templeton, K. E. Thummel, E. D. Kharasch, K. L. Kunze, C. Hoffer, W. L. Nelson and N. Isoherranen, Clin. Pharmacol. Ther., 2008, 83, 77. [4] N. Isoherranen, H. Hachad, C. K. Yeung and R. H. Levy RH, Chem. Res. Toxicol., 2009, 22, 294. [5] J. T. Pearson, J. J. Hill, J. Swank, N. Isoherranen, K. L. Kunze and W. M. Atkins, Biochemistry, 2006, 45, 6341.
552 [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]
[22] [23] [24] [25] [26] [27] [28] [29]
[30] [31]
[32] [33]
[34] [35] [36] [37] [38] [39] [40]
Bernard P. Murray
D. E. Ryan, P. E. Thomas and W. Levin, J. Biol. Chem., 1980, 255, 7941. G. Danan, V. Descatoire and D. Pessayre, J. Pharmacol. Exp. Ther., 1981, 218, 509. K. He, A. M. Falick, B. Chen, F. Nilsson and M. A. Correia, Chem. Res. Toxicol., 1996, 9, 614. K. Sugiyama, K. Yao, A. E. Rettie and M. A. Correia, Chem. Res. Toxicol., 1989, 2, 400. V. G. Zgoda, B. Arison, S. Mkrtchian, M. Ingelman-Sundberg and M. A. Correia, Arch. Biochem. Biophys., 2002, 408, 58. S. F. Zhou, C. C. Xue, X. Q. Yu, C. Li and G. Wang, Ther. Drug Monit., 2007, 29, 687. P. F. Hollenberg, U. M. Kent and N. N. Bumpus, Chem. Res. Toxicol., 2008, 21, 189. A. S. Kalgutkar, R. S. Obach and T. S. Maurer, Curr. Drug Metab., 2007, 8, 407. K. H. Grime, J. Bird, D. Ferguson and R. J. Riley, Eur. J. Pharm. Sci., 2009, 36, 175. L. Xu and M. C. Desai, Curr. Opin. Invest. Drugs, 2009, 10, 775. R. B. Silverman, in Mechanism-Based Enzyme Inactivation: Chemistry and Enzymology 1, CRC Press, Boca Raton, FL, 1988, p. 3. R. Kitz and I. B. Wilson, J. Biol. Chem., 1962, 237, 3245. B. S. Mayhew, D. R. Jones and S. D. Hall, Drug Metab. Dispos., 2000, 28, 1031. D. R. Jones, S. Ekins, L. Li and S. D. Hall, Drug Metab. Dispos., 2007, 35, 1466. T. Koudriakova, E. Iatsimirskaia, I. Utkin, E. Gangl, P. Vouros, E. Storozhuk, D. Orza, J. Marinina and N. Gerber, Drug Metab. Dispos., 1998, 26, 552. G. Luo, J. Lin, W. D. Fiske, R. Dai, T. J. Yang, S. Kim, M. Sinz, E. LeCluyse, E. Solon, J. M. Brennan, I. H. Benedek, S. Jolley, D. Gilbert, L. Wang, F. W. Lee and L. S. Gan, Drug Metab. Dispos., 2003, 31, 1170. C. S. Ernest, S. D. Hall and D. R. Jones, J. Pharmacol. Exp. Ther., 2005, 312, 583. R. S. Obach, R. L. Walsky and K. Venkatakrishnan, Drug Metab. Dispos., 2007, 35, 246. D. A. Fairman, C. Collins and S. Chapple, Drug Metab. Dispos., 2007, 35, 2159. K. R. Korzekwa, N. Krishnamachary, M. Shou, A. Ogai, R. A. Parise, A. E. Rettie, F. J. Gonzalez and T. S. Tracy, Biochemistry, 1998, 37, 4137. S. Puntarulo and A. I. Cederbaum, Free Radic. Biol. Med., 1998, 24, 1324. H. K. Lim, N. Duczak, Jr., L. Brougham, M. Elliot, K. Patel and K. Chan, Drug Metab. Dispos., 2005, 33, 1211. A. Watanabe, K. Nakamura, N. Okudaira, O. Okazaki and K. Sudo, Drug Metab. Dispos., 2007, 35, 1232. E. S. Perloff, A. K. Mason, S. S. Dehal, A. P. Blanchard, L. Morgan, T. Ho, A. Dandeneau, R. M. Crocker, C. M. Chandler, N. Boily, C. L. Crespi and D. M. Stresser, Xenobiotica, 2009, 39, 99. S. Fowler and H. Zhang, AAPS J., 2008, 10, 410. M. A. Correia and P. R. Oritiz de Montellano, in Cytochrome P450: Structure, Mechanism, and Biochemistry (ed. P. R. Ortiz de Montellano), 3rd edition, Academic/Plenum, Kluwer New York, Boston, Dordrecht, London, Moscow, 2004, p. 247. E. Fontana, P. M. Dansette and S. M. Poli, Curr. Drug Metab., 2005, 6, 413. A. S. Kalgutkar, I. Gardner, R. S. Obach, C. L. Shaffer, E. Callegari, K. R. Henne, A. E. Mutlib, D. K. Dalvie, J. S. Lee, Y. Nakai, J. P. O’Donnell, J. Boer and S. P. Harriman, Curr. Drug Metab., 2005, 6, 161. G. A. Doss and T. A. Baillie, Drug Metab. Rev., 2006, 38, 641. M. Nakajima, M. Suzuki, R. Yamaji, H. Takashina, N. Shimada, H. Yamazaki and T. Yokoi, Xenobiotica, 1999, 29, 1191. M. Murray, Curr. Drug Metab., 2000, 1, 67. J. M Hutzler, R. J. Melton, J. M. Rumsey, D. C. Thompson, D. A. Rock and L. C. Wienkers, Chem. Res. Toxicol., 2008, 21, 1125. M. A. Cerny and R. P. Hanzlik, J. Am. Chem. Soc., 2006, 128, 3346. C. Medower, L. Wen and W. W. Johnson, Chem. Res. Toxicol., 2008, 21, 1570. J. R. Palamanda, C. N. Casciano, L. A. Norton, R. P. Clement, L. V. Favreau, C. Lin and A. A. Nomeir, Drug Metab. Dispos., 2001, 29, 863.
Mechanism-Based Inhibition of CYP3A4 and Other Cytochromes P450
553
[41] J. K. Racha, A. E. Rettie and K. L. Kunze, Biochemistry, 1998, 37, 7407. [42] J. M. Hutzler, R. J. Melton, J. M. Rumsey, M. E. Schnute, C. W. Locuson and L. C. Wienkers, Chem. Res. Toxicol., 2006, 19, 1650. [43] R. E. Pearce, W. Lu, Y. Wang, J. P. Uetrecht, M. A. Correia and J. S. Leeder, Drug Metab. Dispos., 2008, 36, 1637. [44] J. T. Pearson, J. L. Wahlstrom, L. J. Dickmann, S. Kumar, J. R. Halpert, L. C. Wienkers, R. S. Foti and D. A. Rock, Chem. Res. Toxicol., 2007, 20, 1778. [45] P. Kang, M. Liao, M. R. Wester, J. S. Leeder, R. E. Pearce and M. A. Correia, Drug Metab. Dispos., 2008, 36, 490. [46] C. Sridar, U. M. Kent, K. Noon, A. McCall, B. Alworth, M. Foroozesh and P. F. Hollenberg, Drug Metab. Dispos., 2008, 36, 2234. [47] T. Omura, P. Siekevitz and G. E. Palade, J. Biol. Chem., 1967, 242, 2389. [48] M. A. Correia, Drug Metab. Rev., 2003, 35, 107. [49] J. Yang, M. Liao, M. Shou, M. Jamei, K. R. Yeo, G. T. Tucker and A. Rostami-Hodjegan, Curr. Drug Metab., 2008, 9, 384. [50] M. A. Correia and M. Liao, Expert Opin. Drug Metab. Toxicol., 2007, 3, 33. [51] P. Schmiedlin-Ren, D. J. Edwards, M. E. Fitzsimmons, K. He, K. S. Lown, P. M. Woster, A. Rahman, K. E. Thummel, J. M. Fisher, P. F. Hollenberg and P. B. Watkins, Drug Metab. Dispos., 1997, 25, 1228. [52] H. F. Wang, M. E. Figueiredo Pereira and M. A. Correia, Arch. Biochem. Biophys., 1999, 365, 45. [53] A. G. Pinto, J. Horlander, N. Chalasani, M. Hamman, A. Asghar, D. Kolwankar and S. D. Hall, Br. J. Clin. Pharmacol., 2005, 59, 440. [54] K. K. Korsmeyer, S. Davoll, M. E. Figueiredo-Pereira and M. A. Correia, Arch. Biochem. Biophys., 1999, 365, 31. [55] S. Raasi and D. H. Wolf, Semin. Cell. Dev. Biol., 2007, 18, 780. [56] K. Nakatsukasa and J. L. Brodsky, Traffic, 2008, 9, 861. [57] R. Y. Hampton and R. M. Garza, Chem. Rev., 2009, 109, 1561. [58] S. Faouzi, K. F. Medzihradszky, C. Hefner, J. J. Maher and M. A. Correia, Biochemistry, 2007, 46, 7793. [59] Y. Wang, M. Liao, N. Hoe, P. Acharya, C. Deng, A. N. Krutchinsky and M. A. Correia, J. Biol. Chem., 2009, 284, 5671. [60] M. K. Pabarcus, N. Hoe, S. Sadeghi, C. Patterson, E. Wiertz and M. A. Correia, Arch. Biochem. Biophys., 2009, 483, 66. [61] B. P. Murray and M. A. Correia, Arch. Biochem. Biophys., 2001, 393, 106. [62] T. Mizutani, M. Shinoda, Y. Tanaka, T. Kuno, A. Hattori, T. Usui, N. Kuno and T. Osaka, Drug Metab. Rev., 2005, 37, 235. [63] A. Tozzi, in Piperonyl Butoxide: The Insecticide Synergist (ed. D. G. Jones), Academic Press, San Diego, CA, 1998, p. 1. [64] T. E. Jones, Clin. Pharmacokinet., 1997, 32, 357–367. [65] M. J. Youle, J. Antimicrob. Chemother., 2007, 60, 1195. [66] A. A. Mathias, S. West, J. Hui and B. P. Kearney, Clin. Pharmacol. Ther., 2009, 85, 64. [67] K. K. Adkison, A. Shachoy-Clark, L. Fang, Y. Lou, V. R. Otto, M. M. Berrey and S. C. Piscitelli, Br. J. Clin. Pharmacol., 2006, 62, 336. [68] G. N. Kumar, V. K. Jayanti, M. K. Johnson, J. Uchic, S. Thomas, R. D. Lee, B. A. Grabowski, H. L. Sham, D. J. Kempf, J. F. Denissen, K. C. Marsh, E. Sun and S. A. Roberts, Pharm. Res., 2004, 21, 1622. [69] A. E. van Herwaarden, E. Wagenaar, C. M. van der Kruijssen, R. A. van Waterschoot, J. W. Smit, J. Y. Song, M. A. van der Valk, O. van Tellingen, J. W. van der Hoorn, H. Rosing, J. H. Beijnen and A. H. Schinkel, J. Clin. Invest., 2007, 117, 3583. [70] M. Gertz, J. D. Davis, A. Harrison, J. B. Houston and A. Galetin, Curr. Drug Metab., 2008, 9, 785.
CHAPT ER
27 Nonclinical Toxicogenomics in the Pharmaceutical Environment William R. Foster, Stefan U. Ruepp and Bruce D. Car
Contents
1. 2. 3. 4.
Introduction Evaluation of Liver Toxicogenomics Transcriptional Markers of Pharmacology Deriving Diagnostic Transcriptional Signatures as Biomarkers 5. Percentage Global Transcriptional Change with Drug Treatment 6. Correlating Transcriptomics with Histopathology 7. Transcriptional Changes Associated with General Toxicity 8. Ruling in and out Mechanisms of Toxicity with Toxicogenomics 9. Assigning Cause or Effect to Transcriptional Alterations 10. Critical Assessment of In Vitro Toxicogenomics Applications 11. Transcriptional Responses to Diet 11.1 Transcriptional response to fasting 11.2 Transcriptional effects of feeding in multiple-dose studies 11.3 Transcriptional effects of overfeeding and obesity 12. Toxicogenomics, Genotoxicity, and Carcinogenicity Prediction 13. Providing Feedback to Discovery Working Groups 14. Conclusion References
555 556 557 559 559 559 561 563 565 566 566 567 567 568 569 569 571 572
Discovery Toxicology, Pharmaceutical Candidate Optimization, Bristol-Myers Squibb Co., Princeton, NJ 08543-4000, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04427-3
r 2009 Elsevier Inc. All rights reserved.
555
556
William R. Foster et al.
1. INTRODUCTION Traditional assessments of toxicology have included clinical signs, body weight, electrocardiograms, ophthalmoscopy, clinical chemistry, hematology, urinalysis, organ weights, and histopathology in toxicology studies from a single day to 2 years duration. These endpoints are generally well understood and their interpretation has changed little in years. Assessments employed in investigative toxicology however have better matched technologic advancements in the broader field of biology. These newer assessments, including toxicogenomics, are now also applied in a nonroutine manner to toxicology studies, enhancing the interpretations provided by traditional endpoints and occasionally presenting important new findings. Medicinal chemists are assailed with new technologies, assays, and approaches to triaging their compounds. The incremental addition of such assays to a compound selection algorithm is both expensive and timeconsuming. The expense and cycle-time issues must be balanced appropriately with the positive impact on compound quality. These approaches fall across a spectrum of validity from little to rigorous and may be viewed with some skepticism by a chemist being told that his compound scored positively in a new, unfamiliar assay, as for a transcriptomic signature for an obscure liability, and should be eliminated from advancement. When toxicogenomics was a nascent technology, the predictive power of in vitro assessments was much touted. Experts publishing in this field and at focused meetings now generally acknowledge the many shortcomings of global in vitro transcriptional profiling with the possible exception of applications to drug metabolizing genes and have generally become more cautious relative to the potential for toxicogenomics to replace traditional approaches [1–5]. Given the complexity of toxicogenomics as applied to traditional toxicology assessments, the presentation of these evaluations may include unusual statistical plots, Venn diagrams, or other novel visualization tools, potentially representing a blackbox for the uninitiated. The interpretation of toxicogenomics data should include strengths and weaknesses; this transparency is particularly important for a new technology. The purpose of this chapter is to provide an experimentally rather than theoretically grounded projection of the applications and potential impact of this new technology.
2. EVALUATION OF LIVER TOXICOGENOMICS Rodent liver from nonclinical toxicity studies has been the most extensively transcriptionally analyzed tissue in systematic surveys [6–9]. The compelling advantage of evaluation of the liver is the technical
Nonclinical Toxicogenomics in the Pharmaceutical Environment
557
simplicity of obtaining hepatic mRNA because the organ size provides for analysis by many techniques concurrently, the liver’s homogeneity enables direct comparisons of findings across assays, and the high yields of mRNA per milligram of tissue provide sufficient material from even small samples. Hepatotoxicity is also a common and important nonclinical and clinical finding [10]. The liver as viewed through the lens of whole genome transcription could hypothetically be a reporter tissue for the general health of the animal [11]. In addition, in the most favorable circumstances, liver might act as a surrogate tissue or biosensor capable of detecting patterns of change predictive of a wide range of toxicities, even toxicities peculiar to other tissues, species, higher exposures, or longer durations of dosing than that of the study analyzed. This original vision for liver toxicogenomics by technology enthusiasts, (e.g., omic early adopters, bioinformaticians) has begun to be replaced with traditional methodologies with a more well-defined level of impact and higher level of expert-driven interpretive input for hepatic transcription [4]. Table 1 contrasts the original approach to interpreting hepatic transcriptomics versus the new emerging standard. A narrow interpretive focus based on a finite set of processes with established connections to hepatic transcription, as with inflammation and drug metabolism, can be rapidly recognized, developed, and scientifically defended by toxicologists/pathologists to drug discovery team members in terms of direct references to individual genes, established science, and by correlation with other study findings. The interpretation of omics data from nonclinical studies by toxicologists and pathologists provides a broadening of the assay interpretation into basic physiology. For example, pathologists can readily recognize the tissue they are viewing and judge by light microscopy if there are lesions present whose maturity suggests they should be interpreted as pre-existing or caused by dosing with a xenobiotic. A pathologist or toxicologist can also recognize with relatively little training the tissue type hybridized to each DNA microarray based on abundance of transcripts. The experienced pathologist/toxicologist can also identify, by cross-correlating omics with the study records, a large constellation of changes associated with the study conduct and study pathology including individual animal outliers by sex, normal versus dysregulated diurnal variation, drug exposure versus response, or the fasted state of an animal prior to necropsy.
3. TRANSCRIPTIONAL MARKERS OF PHARMACOLOGY Whenever there are known transcriptional markers of pharmacology, it is straightforward to evaluate them in a transcriptomic study as any DNA microarray–based approach typically includes measurements across
558 Table 1
William R. Foster et al.
Evolution of the interpretation of hepatic transcriptional changes
Bioinformatics-based transcriptional interpretation
Knowledge-based transcriptional interpretation
Use of historical transcriptomic data to create predictivity algorithms for a large range of adverse events
Hepatic transcription as a sensor of a subset of events with accepted relationships to transcription (e.g., drug metabolism or transport, genotoxicity, pharmacology, inflammation, cholesterol biosynthesis) or for detection of broad black/white events (e.g., marked transcriptional changes absent/ present prior to or coincident with pathology) Interpretation based on manual expert– driven integration of traditional and transcriptomic findings, the literature, established pharm/tox transcriptional markers, and historical collections of transcriptional data
Interpretation based on automated computer scoring of multidimensional classifiers of transcriptional data regardless of traditional endpoint findings Example finding: Compound is a hepatotoxicant. Interpretation: the automated classifier has a false-positive/negative rate of X/Y%. Examination of individual gene changes for mechanism is not appropriate (gene contributions depend on the entire historical database, and gene changes relative to control may be statistically insignificant) Example finding: Compound is predicted to cause hepatomegaly
Example finding: Compound causes statistically significant and coordinate hepatic inflammatory transcriptional changes relative to control across a broad set of established markers of inflammation Interpretation: These findings are considered drug-related and secondary to a minimal myopathy. The combination of transcriptomic and light microscope findings adds to the weight of evidence for a positive finding of drug-related myopathy at the lowest dose Example finding: Drug metabolism and transport transcripts are up-regulated and correlate to a decrease in drug exposure and increase in liver weight Interpretation: These findings are drugrelated, consistent with observed in vitro biotransformation pathways
Nonclinical Toxicogenomics in the Pharmaceutical Environment
559
transcripts of most known genes. One caveat is that the omic assay (e.g., DNA microarray) is usually less sensitive than RT-PCR, potentially underestimating the fold-change and significance level of a finding. The presence or absence of markers of pharmacology in the drug safety study may be useful for structuring assessments of therapeutic index, particularly when biomarkers of efficacy are not available for the nonclinical species. The detection of hepatic pharmacology markers is most straightforward for anti-inflammatory treatments due to the significant ‘‘basal inflammatory tone’’ in normal rat liver due to intestinal bacteria [12]. A significantly more difficult area for evaluation of pharmacology in rat liver is metabolic diseases targets due to the multiplicity of convergent inputs other than drug pharmacology that can modulate transcription of the same kinds of genes, for instance, by food intake, diurnal rhythm, inflammation, drug metabolism, and toxicity [13–16].
4. DERIVING DIAGNOSTIC TRANSCRIPTIONAL SIGNATURES AS BIOMARKERS Gene signatures are designed to robustly capture, as an algebraic algorithm, an association between drug-related changes in genomic markers and a diagnosis or prediction of pathology or pharmacology. The most complex, ambitious signature is one where a multi-dimensional automated computer algorithm is applied solely to genomic data (ignoring all other study findings or human input), and a prediction of future pathology is made as a numerical probability. A less ambitious but potentially more practically applied signature is where single markers are used to diagnose pharmacology or toxicity, and the algorithm used to drive interpretation is the experience and judgment of the toxicologist/ pathologist. Table 2 contrasts features of the two extremes of gene signature types.
5. PERCENTAGE GLOBAL TRANSCRIPTIONAL CHANGE WITH DRUG TREATMENT A low specificity general marker of potentially high utility is the percentage of all transcripts assayed that change in a drug treated versus control group at a given level of significance (typically a t-test of po0.01). Such a global evaluation provides a facile view of off-target activities and a reasonable surrogate for the presence of pathology. In random normally distributed data, 1% of transcripts are expected to change by chance.
560 Table 2
William R. Foster et al.
Features of transcriptional profiling signatures for predictivity
Feature
Single maker/expertdriven, integrating traditional study data
Multi-dimensional, automated, predictive marker
Grounded in biological, toxicological literature? Formally derived from a standardized set of historical data? Is marker accessible by average drug discovery scientist? Is marker accessible by toxicologist/ pathologist? Accessible hypothesis between marker change and pharm/tox diagnosis/prediction? Quantitative relationship between marker foldchange, p-value and toxicity/pharmacology call? Subject to systematic biases of one/few historical data collections? Subject to computational overfitting?a Cost of construction?
Yes
No
No
Yes
Somewhat
No
Yes
No
Yes
No
No
Yes
No
Yes
No
Yes
Experience/ literature review Yes
Years/millions of dollars No
Difficult
Simple
Subject to investigator bias Ease of application to a data set a
Likely to yield a positive but scientifically ungrounded, misleading result due to exclusion of critical data.
Figure 1 summarizes expectations for an analysis with this transcriptomic study summary statistic [4]. An additional expectation of the percent po0.01 statistic is that if significant change (W1% at po0.01) is present in the study, there should be more change with higher drug exposure.
Nonclinical Toxicogenomics in the Pharmaceutical Environment
marked
Do not analyze: high %p<0.01 is uninterpetable multiple changes secondary to toxicity/pathology
Not observed*. (unifeasible?) severity of pathology, toxicity
minimal
Do not analyze: high variablity in controls or no meaningful changes.
Promising TGX global signal present, tractable analysis, less mature pathology.
1.0
561
3.0
Not observed*. (unfeasible?)
8.0
% transcripts p<0.01 *Not observed in properly controlled studies, e.g. same vehicle, feeding, age, dissection procedure, time of day in treated and control groups.
Figure 1 The empirical relationship of pathology to global transcriptional changes in multiple dose studies. Histopathologic alterations may be in the tissue assayed or may be elsewhere in the animal. Marked pathology without global transcriptional changes or broad transcriptional changes in the absence of pathology are generally not observed.
6. CORRELATING TRANSCRIPTOMICS WITH HISTOPATHOLOGY In drug safety studies, transcriptomics adds to the weight of evidence that pathology exists or is absent by presenting an additional correlation or lack of correlation across multiple endpoints. These comparative assessments can be helpful when the team is simply seeking additional evidence beyond the interpretation of the light microscopic view and particularly when the drug-relatedness of a minimal finding is in question for therapeutic targets with low risk tolerance. Table 3 provides examples from nonclinical studies of such findings of pharmacologic or toxicity interest, a study correlate, and the resulting combined interpretation. The examples of Table 3 include cases where either the traditional endpoints are the findings of interest and the mRNA is a correlate or the mRNA is the finding of interest and traditional endpoints are the correlate. The resulting interpretation is contingent on the analyst’s knowledge of the study findings, traditional pathology, and the scientific literature.
562 Table 3
William R. Foster et al.
Non-clinical study findings, study correlate, and interpretation
Minimal/uncertain finding
Study correlate
Interpretation
Myopathy in skeletal but not cardiac muscle
mRNA markers of degeneration/ regeneration in skeletal but not heart samples Marked global and dose-related brain transcriptional changes within hours following a single dose Alterations in a broad set of transcripts known as responsive to inflammation, generally increased in drug-treated animals Marked increase in splenic mRNA for reticulocyte transcripts correlates animal by animal with the hematology findings Efficacious exposure estimate (crossspecies in vitro potency and protein binding adjusted)
Unacceptable minimal cardiac myopathy is absent with drug treatment
Time of necropsy animal by animal correlates with mRNA changes and literature descriptions of normal diurnal variation of hepatic glucokinase mRNA
Transcriptional marker is not indicative of pharmacology but of time of necropsy
Decreased activity in rodent after a single dose, seizure after multiple doses
Minimal, highly variable increase in hepatic inflammatory foci with drug treatment
Minimal decrease in RBC counts and increase in reticulocytes in some drug-treated animals Dose-related repression of inflammationrelated mRNAs for an antiinflammatory drug Alteration in hepatic glucokinase mRNA post-administration of a drug for metabolic diseases
Behavioral changes correlate to marked brain molecular changes that may potentiate for seizure Minimal, drug-related hepatic inflammation is correctly diagnosed relative to inflammation in control animals Diagnosis of minimal hematology finding is confirmed by mRNA
Pharmacologic response in the drug safety species is in the expected drug concentration range
Nonclinical Toxicogenomics in the Pharmaceutical Environment
563
7. TRANSCRIPTIONAL CHANGES ASSOCIATED WITH GENERAL TOXICITY General toxicity as defined by decreased activity, decreased weight gain and food consumption, and mild effects on clinical pathology parameters can present with a large number of transcriptional changes (top right quadrant of Figure 1). The investigation of such transcriptomic profiles is likely to be time-consuming and is generally unlikely to reveal useful insights. The difficulty in interpretation is due to the sheer number of findings and the multiplicity of likely secondary sources of transcriptional change other than drug action or pathogenesis (decreased food intake, dehydration, consequences of dysregulated blood pH, poor tissue perfusion, etc.). An additional complication of interpretation is that observed changes may simply represent a repression by toxicity or pathology of the normal physiological response to decreased food intake, pharmacology, or diurnal rhythm as illustrated for two anti-proliferative therapies in Table 4. Integration of simple summary data like that of Table 4 helps develop perspectives about the relationships among dose, pharmacology, toxicity, and maintenance of normal physiology in the nonclinical species.
8. RULING IN AND OUT MECHANISMS OF TOXICITY WITH TOXICOGENOMICS Before embarking on extensive toxicogenomics studies, careful consideration should be given to the appropriateness of toxicogenomics to the evaluation of the findings of interest. Factors that complicate mechanistic investigations include the presence of general toxicity as manifested, for example, in overt clinical signs, substantial reduction in food intake, and body weight loss. The presence of profound pathological changes or excessive amounts of transcriptional changes will also confound the ability to meaningfully interpret transcriptomic data sets (Figure 1). Study duration can have a substantial effect on the number, magnitude, and type of gene changes observed [17]. Mechanistic investigations are more likely to succeed in situations where transcription is assessed at a time when pathology is in the process of emerging rather than fully established or even worse, when tissue remodeling and regeneration processes are dominating the transcriptome. Given aggressive timelines for typical drug discovery and development programs, the use of toxicogenomics in mechanistic studies is of particular value for rapid ruling out of potential mechanistic causes [4].
564
Comparison of toxicity markers
Treatment
30 mg/kg cmpd 1 100 mg/kg cmpd 1 5 mg/kg cmpd 2 30 mg/kgb cmpd 2 a
Hepatic inflammation/ necrosis?
Neutrophil counts fold over control
mRNA % po0.01
Inflammatory marker (Lbp mRNA) fold over control
Proliferative marker (RRM2 mRNA) fold over control
No/No
1
2
1
4
7
Minimal/No
5
8
4
14
7
No/No
3
10
6
9
11
Mild/Mild
3
20
10
5
1
Low- and high-dose groups were necropsied concurrently but later in the day than controls. Dose group had 13% weight loss and fourfold elevation of ALT, other groups showed no weight or ALT changes relative to controls.
b
Diurnal Rhythm Marker (Dbp) fold over controla
William R. Foster et al.
Table 4
Nonclinical Toxicogenomics in the Pharmaceutical Environment
565
A combination of utilizing a gene expression compendium, predictive algorithms, and confirmatory in vitro experiments was useful in investigating a series of compounds with antidiabetic properties administered up to 5 days to rats. Utilizing a software algorithm, three separate compounds (Rx08, Rx09, Rx10) were found to have substantial transcriptional similarities with a structurally diverse set of compounds known to produce microvesicular steatosis within a similar timeframe. Histopathology indicated the presence of microvesicular steatosis for Rx08 and Rx09, whereas Rx10 did not cause histological changes in the liver. Indeed, a 14-day study in dogs with Rx10 caused microvesicular steatosis in the dog liver, confirming the toxicogenomic and biochemical assessments derived from rat [18]. While long-term animal studies may also be used to unravel such liabilities, a combination of toxicogenomics and confirmatory assays as discussed above will require less compound and has the potential to be faster and cheaper. Another example utilizing a gene expression compendium to suggest a potential mechanism of toxicity was recently published [19]. Inhibitors of acetyl CoA carboxylase (ACC) 2 intended to treat type 2 diabetes and obesity were administered for 14 days to ob/ob mice. The compound, A-908292 (S), markedly reduced de novo lipid synthesis and increased lipid oxidation triglyceride levels in serum as well as serum glucose. A-875400 (R), the near inactive (50-fold lower ACC2 activity) enantiomer of A-908292 (S), unexpectedly caused a very similar time-dependent decrease in glucose and triglycerides, suggesting that these effects were not mediated by the pharmacological target, ACC2. In a 3-day rat study, a dose-related increase in cholesterol was observed with A-908292 (S), but triglycerides or glucose were not affected by treatment with either compound after this relatively short dosing period. Similar transcriptional expression changes were seen with both compounds and correlated with several known peroxisome proliferator-activated receptor alpha (PPAR-a) agonists in the reference database DrugMatrixs [20]. Subsequently, rats were dosed with the PPAR-a agonist, benzafibrate. Western blot and immunohistochemistry showed an increase in PMP70, a marker for peroxisome proliferation for A-908292 (S), A-875400 (R), and benzafibrate, and subsequent electron microscopy confirmed peroxisome proliferation.
9. ASSIGNING CAUSE OR EFFECT TO TRANSCRIPTIONAL ALTERATIONS Carefully designed time-course studies using toxicogenomics in combination with phenotypic anchoring are often particularly helpful in mechanistic studies.
566
William R. Foster et al.
In a mouse study investigating the effects of acetaminophen overdosing, livers were collected from 15 min to 4 h after dosing, and molecular effects on the transcriptome and proteome were evaluated. Transcriptional and morphological effects manifesting as a swelling of the mitochondria were observed as early as 15 min after dosing, suggesting that observed effects were caused by direct action of the reactive acetaminophen metabolite, N-acetyl-p-benzo-quinoneimine (NAPQI). Therefore, transcriptional effects were likely to be secondary to the chemical damage and thus mechanistically less informative [21].
10. CRITICAL ASSESSMENT OF IN VITRO TOXICOGENOMICS APPLICATIONS In vitro systems are a useful tool to investigate mechanisms of toxicity first observed in in vivo studies, but the utility of the approach is limited to pathways and processes conserved in the selected in vitro system. For whole transcriptome analysis, it has been established that the baseline in gene expression is markedly different in vivo and in vitro and that there are considerable differences between various in vitro systems in different culture conditions [5,22,23]. Thus, the usefulness and mechanistic relevance of in vitro toxicogenomics studies has to be carefully evaluated on a case-by-case basis. Since in vitro models are designed to capture a key, but limited aspect of biology, the application of a global technology like whole genome transcriptional profiling as an assay endpoint to in vitro evaluations is an odd juxtaposition. A cell line’s primary global phenotype is growth/ inhibition of apoptosis with transcription likely to detect a vast network of highly redundant changes that respond sensitively to cell culture conditions of confluency, temperature, media changes, and so on, as well as to drug treatment. When global transcriptional profiles of in vitro samples detect only growth or viability-related transcriptional changes, they provide no advantage over simpler, more direct assays [3]. Cell lines and primary cells exhibit basal patterns of gene expression as distinct from their source tissue as one tissue’s basal pattern differs from another [22,24]. In the case of immortalized cells, chromosomal duplications and deletions are present, while for primary cells, the isolation procedures create profound wounding and inflammatory responses. For both cell types, the growth conditions and resulting homeostasis are distinct from the in vivo situation. For well-characterized pathways that are preserved in the selected in vitro system, transcriptional analysis can be valuable to rapidly characterize and differentiate drug candidates. A low-density array to investigate compound-mediated effects on selected genes in rat and in
Nonclinical Toxicogenomics in the Pharmaceutical Environment
567
human hepatocytes can be used as a rapid assessment for potential effects on a variety of nuclear receptors, transporters, and cytochromes P450s, which allow prioritizing specified cytochromes P450 for a more labor-intensive enzyme activity determination [25].
11. TRANSCRIPTIONAL RESPONSES TO DIET Fasting by itself can have profound effects on various physiological parameters. It has been known for decades that prolonged moderate calorie restriction in rodents has a positive impact on lifespan and various diseases including metabolic parameters, cancer, and the immune system [26–28]; however, severe calorie restriction leads to organ pathologies and has to be considered adverse [29]. Food effects in toxicology studies, as in short-term fasting or over-night fasting before necropsy and prolonged calorie restriction, typically associated with pharmacological target modulation or secondary to toxicological effects manifesting in decreased appetite, are quite common.
11.1 Transcriptional response to fasting The overnight fast before necropsy or blood collection is a routine procedure in many toxicology studies and can affect the toxicity of compounds studied. A well-known example is the increased hepatotoxicity of acetaminophen in mice due to a decrease in glutathione content in the liver as a consequence of fasting. As whole transcriptome analysis provides a very high level of resolution for altered cellular processes, differences in feeding status may have a profound effect in itself. To characterize and understand consequences of this routine, over-night fasting procedure is necessary to accurately interpret toxicogenomics data. Fasting male and female Sprague–Dawley rats for 16 h had substantial effects on several parameters of toxicological interest. Fasted rats lost 7–9% bodyweight in females and males, respectively, while their ad libitum fed counterparts gained 3–4% [30]. Approximately 7% of transcripts in males and 10% in females changed at po0.01 in fasted rats compared to controls. These effects exceeded effects typically seen in preclinical toxicology studies, even compared to compounds associated with hepatotoxicity or other severe pathologies [4]. Major transcriptional changes included genes with roles in fatty acid, carbohydrate, cholesterol, and bile acid metabolism indicating decreased activity in glycolytic pathways and a shift toward increased utilization of fatty acids. Typically, multiple genes within these metabolic pathways, including key rate limiting genes, changed simultaneously and those changes were correlative to changes in clinical pathology parameters.
568
William R. Foster et al.
Repressed key mRNAs included 3-hydroxy-3-methylglutaryl-coenzyme A reductase and mevalonate kinase. Reduction in cholesterol 7-ahydroxylase (CYP7A1), which catalyzes the first reaction and ratelimiting step in bile acid synthesis and is known to be negatively regulated at the transcriptional level by bile acids, was consistent with observed increases in serum bile acids [31].
11.2 Transcriptional effects of feeding in multiple-dose studies In many toxicology studies, changes in food uptake and in body weight are observed. A decrease in food consumption may be dose-related, complicating the task of determining whether effects are related to test compound or are a consequence of inanition. As the level of decreased food uptake is not easy to predict, and as pair fed controls are seldom used, the toxicogenomics analyst may be confronted with data sets derived from control and treated animals with a very different dietary status. Thus, the characterization of different levels of food restriction is expected to greatly assist the identification of effects that are likely to be associated with caloric deficits and may not be associated with the pharmacologic properties of the compound itself. Several investigators used slightly different experimental protocols to study transcriptional effects in rodents as a consequence of reduced calorie intake. As expected, transcriptional changes were influenced by the level and duration of calorie restriction. Overall increases in lipolysis were most comparable between different study designs. Changes in glycolysis/gluconeogenesis, cholesterol biosynthesis, and effects on cell cycle and apoptosis were also prominently represented, and also, some effects on transporters and cytochrome P450s were noted; however, study duration and severity of caloric restriction have a substantial effect on phases and timing of metabolic adaptation [16,32]. Substantial transcriptional effects in response to different fasting regimes are not limited to the liver but have also been described for other organs [33,34]. Calorie restriction leads also to a clear reduction in inflammatory genes in white adipose tissue and in liver [32,35].
11.3 Transcriptional effects of overfeeding and obesity Rodents are overfed in diet-induced obesity (DIO) models to generate obese or diabetic animals that serve as efficacy models for the corresponding diseases. Given the cost associated with these models, it is tempting to collect as much information as possible including not only efficacy but also toxicity and pathology endpoints. As for other disease models, it is important to understand differences between these animals and ‘‘normal’’ animals.
Nonclinical Toxicogenomics in the Pharmaceutical Environment
569
To characterize a commonly used mouse DIO model, mice were fed a high-fat diet for half a year leading to an almost 50% increase in body weight relative to chow-fed controls. The altered physiological status was associated with elevated serum enzymes (alanine aminotransferase, ALT, approximately threefold and aspartate aminotransferase, AST, approximately twofold) correlating with histopathology findings in the liver characterized as widespread intracellular accumulation of variably sized lipid droplets, consistent with micro- and macro-vesicular steatosis. Just over 12% of genes on the microarray used were changed as defined by a p-value in a t-test of o0.01. Most remarkable were changes in the cholesterol and sterol biosynthesis pathways where the majority of the pathway was down-regulated. Livers in these animals showed a proinflammatory state indicated by upregulation of many immune and inflammation-related genes including serum amyloid A (SAA1, SAA2, SAA3), lipocalin, and chemokine (C-C motif) receptor 2 (CCR2, CD192). Genes affected by diet, potentially impacting exposure of drugs included the major P450s Cyp 1A, Cyp 2B, Cyp3A, Cyp 4A, and the transporters Abcb4 (MDR/TAP), Abcd2, Abcg1, and OAT (Figure 2). Lean chow-fed rats and DIO rats received the same doses (in mg/kg) of a candidate drug for the treatment of obesity. Plasma drug concentrations as area under the curve (AUC), and Cmax were two- to threefold higher in obese versus lean rats. This observation was in agreement with in vitro metabolism data for the compound indicating approximately a twofold increase in drug half-life in liver microsomes from DIO versus chow-fed rats. Effects on body weight, liver enzyme elevations, and liver histopathology were milder than in the mouse study (Table 5) correlating with a reduced effect on the liver transcriptome. Effects on cholesterol biosynthesis and fatty acid metabolism were observed in both mice and rats, but effects on fatty acid metabolism were more pronounced in the rat than the mouse. Overall, the host of transcriptional changes indicates that feeding status alone may confound the ability to detect other transcriptomic alterations of interest.
12. TOXICOGENOMICS, GENOTOXICITY, AND CARCINOGENICITY PREDICTION Work led by Ellinger-Ziegelbauer and since confirmed by a variety of groups has provided evidence for the predictivity of transcriptomic profiles of liver for both genotoxic and non-genotoxicant-induced changes that ultimately result in carcinogenicity [36,37]. A Critical Path initiative is attempting to collate the work of these groups into RT-PCR
William R. Foster et al.
Rat Mouse
570
The color scale indicates the level of expression in DIO relative to chow fed animals. Upregulation is in red, downregulation in green and white indicates that p-value criteria were not met (p<0.05).
Figure 2 Selected genes of interest affected by DIO in mouse and rat. (See Color Plate 27.2 in Color Plate Section.)
signatures that may be used to inform the potential for carcinogenicity in late discovery or early development compounds [37].
13. PROVIDING FEEDBACK TO DISCOVERY WORKING GROUPS The forum in which most medicinal chemists in pharmaceutical companies discuss new information is the Discovery Working Group or equivalent. Counterparts to this group also exist in the Development environment.
Nonclinical Toxicogenomics in the Pharmaceutical Environment
Table 5
571
Similarities between mouse and rat DIO studies
Time on diet Diet Strain, sex Liver % po0.01 Body weight versus control Clin. Path. Histopathology Transcriptional phenotype
DIO mouse
DIO rat
28 weeks Lard-based diet versus normal chow C57Bl/6J, male 12 +47%
14 weeks Plant-based diet/choice versus normal chow Sprague–Dawley male 5 +13%
ALT (m3 , AST(m2 ) Marked/severe steatosis Cholesterol metabolismm FA metabolismm Drug metabolizing enzymes and transporters (mk) Inflammationm Tissue remodelingm
ALT (m1.5 ) Minimal findings Cholesterol metabolismm FA metabolismm Drug metabolizing enzymes & transporters (mk) Inflammationk
When toxicogenomic or simple transcriptomic endpoints are included in studies of pharmacology models, profiling tissue samples by highdensity array analysis is typically a process much slower than several rounds of a chemical synthesis. The intent to conduct such analyses needs to be reported well ahead of time, and a reasonable estimate of time till presentation of data should be provided. Typically, toxicogenomics is not a facile enough technology upon which to base rapid cycles of structure-activity relationship (SAR). Perhaps, more importantly, relative to typically low nanomolar potency at the desired pharmacologic target, multiple toxicophores may exist in micromolar or higher ranges. This leads to transcriptional changes relating to toxicity intermingled with those of pharmacology and environment (fasting, feeding, diurnal rhythm, etc., as discussed above). Rendering this data cleanly into a structure-toxicity relationship can be extremely difficult, if not impossible.
14. CONCLUSION Toxicogenomics is a maturing research tool in its application to toxicology. While prudent use of toxicogenomics is able to focus mechanistic investigations and indirectly provide potential counterscreens for toxicity and secondary pharmacology, the direct predictive
572
William R. Foster et al.
application of toxicogenomics must be carefully validated for specific toxicities. The sensitivity of this tool to environment and other influences in its application to animal models as well as toxicology studies must always be considered in the interpretation of data. A large number of the transcriptional changes observed in tissues from animals given xenobiotics are not readily interpretable given the current state of knowledge of the discipline of toxicology. Fostering this discipline will unquestionably improve the quality and breadth of toxicologic interpretations applied to studies in nonclinical species.
REFERENCES [1] [2] [3] [4] [5]
[6] [7]
[8] [9] [10] [11]
[12] [13] [14] [15] [16] [17] [18] [19]
A. Lu¨he, L. Suter, S. Ruepp, T. Singer, T. Weiser and S. Albertini, Mutat. Res., 2005, 575, 102. E. A. Blomme, Y. Yang and J. F. Waring, Toxicol. Lett., 2009, 186, 22. T. P. Ryan, J. L. Stevens and C. E. Thomas, Curr. Opin. Pharmacol., 2008, 8, 654. W. R. Foster, S. J. Chen, A. He, A. Truong, V. Bhaskaran, D. M. Nelson, D. M. Dambach, L. D. Lehman-McKeeman and B. D. Car, Toxicol. Pathol., 2007, 35, 621. M. Schug, T. Heise, A. Bauer, D. Storm, M. Blaszkewicz, E. Bedawy, M. Brulport, B. Geppert, M. Hermes, W. Follmann, K. Rapp, L. Maccoux, W. Schormann, K. E. Appel, A. Oberemm, U. Gundert-Remy and J. G. Hengstler, Arch. Toxicol., 2008, 82, 923. G. Natsoulis, C. I. Pearson, J. Gollub, B. P. Eynon, J. Ferng, R. Nair, R. Idury, M. D. Lee, M. R. Fielden, R. J. Brennan, A. H. Roter and K. Jarnagin, Mol. Syst. Biol., 2008, 4, 175. K. R. Hayes, A. L. Vollrath, G. M. Zastrow, B. J. McMillan, M. Craven, S. Jovanovich, D. R. Rank, S. Penn, J. A. Walisser, J. K. Reddy, R. S. Thomas and C. A. Bradfield, Mol. Pharmacol., 2005, 67, 1360. G. Steiner, L. Suter, F. Boess, R. Gasser, M. C. de Vera, S. Albertini and S. Ruepp, Environ. Health Perspect., 2004, 112, 1236. N. Kiyosawa, K. Ito, K. Watanabe, M. Kanbori, N. Niino, S. Manabe and T. Yamoto, Toxicol. Lett., 2006, 163, 161. B. D. Car, Am. Drug Discov., 2006, 1, 53. K. T. Morgan, Z. Jayyosi, M. A. Hower, M. V. Pino, T. M. Connolly, K. Kotlenga, J. Lin, M. Wang, H. L. Schmidts, M. S. Bonnefoi, T. C. Elston and G. A. Boorman, Toxicol. Pathol., 2005, 33, 136. T. R. Billiar, M. A. Maddaus, M. A. West, D. L. Dunn and R. L. Simmons, J. Surg. Res., 1988, 44, 397. F. L. Lim, R. A. Currie, G. Orphanides and J. G. Moggs, Xenobiotica, 2006, 36, 1140. G. A. Boorman, P. E. Blackshear, J. S. Parker, E. K. Lobenhofer, D. E. Malarkey, M. K. Vallant, D. K. Gerken and R. D. Irwin, Toxicol. Sci., 2005, 86, 185. T. W. Gant, P. R. Baus, B. Clothier, J. Riley, R. Davies, D. J. Judah, R. E. Edwards, E. George, P. Greaves and A. G. Smith, EHP Toxicogenomics, 2003, 111, 37. R. Pohjanvirta, P. C. Boutros, I. D. Moffat, J. Linden, D. Wendelin and A. B. Okey, Toxicol. Appl. Pharmacol., 2008, 230, 41. M. E. Andersen, H. J. Clewell, 3rd, E. Bermudez, G. A. Willson and R. S. Thomas, Toxicol. Sci., 2008, 105, 368. S. Ruepp, F. Boess, L. Suter, M. C. de Vera, G. Steiner, T. Steele, T. Weiser and S. Albertini, Toxicol. Appl. Pharmacol., 2005, 207, 161. J. F. Waring, Y. Yang, C. H. Healan-Greenberg, A. L. Adler, R. Dickinson, T. McNally, X. Wang, M. Weitzberg, X. Xu, A. Lisowski, S. E. Warder, Y. G. Gu, B. A. Zinker, E. A. Blomme and H. S. Camp, J. Pharmacol. Exp. Ther., 2008, 324, 507.
Nonclinical Toxicogenomics in the Pharmaceutical Environment
573
[20] B. Ganter, S. Tugendreich, C. I. Pearson, E. Ayanoglu, S. Baumhueter, K. A. Bostian, L. Brady, L. J. Browne, J. T. Calvin, G. J. Day, N. Breckenridge, S. Dunlea, B. P. Eynon, L. M. Furness, J. Ferng, M. R. Fielden, S. Y. Fujimoto, L. Gong, C. Hu, R. Idury, M. S. Judo, K. L. Kolaja, M. D. Lee, C. McSorley, J. M. Minor, R. V. Nair, G. Natsoulis, P. Nguyen, S. M. Nicholson, H. Pham, A. H. Roter, D. Sun, S. Tan, S. Thode, A. M. Tolley, A. Vladimirova, J. Yang, Z. Zhou and K. Jarnagin, J. Biotechnol., 2005, 119, 219. [21] S. U. Ruepp, R. P. Tonge, J. Shaw, N. Wallis and F. Pognan, Toxicol. Sci., 2002, 65, 135. [22] F. Boess, M. Kamber, S. Romer, R. Gasser, D. Muller, S. Albertini and L. Suter, Toxicol. Sci., 2003, 73, 386. [23] L. Hultin-Rosenberg, S. Jagannathan, K. C. Nilsson, S. A. Matis, N. Sjogren, R. D. Huby, A. H. Salter and J. D. Tugwood, Xenobiotica, 2006, 36, 1122. [24] R. Shyamsundar, Y. H. Kim, J. P. Higgins, K. Montgomery, M. Jorden, A. Sethuraman, M. Rijn, D. Botstein, P. O. Brown and J. R. Pollack, Genome Biol., 2005, 6, R22. [25] L. Richert, G. Tuschl, C. Abadie, N. Blanchard, D. Pekthong, G. Mantion, J. C. Weber and S. O. Mueller, Toxicol. Appl. Pharmacol., 2009, 235, 86. [26] C. M. McCay, M. F. Crowell and L. A. Maynard, J. Nutr., 1935, 10, 63. [27] D. Kritchevsky, Toxicol. Sci., 1999, 52, 13. [28] R. Weindruch, Toxicol. Pathol., 1996, 24, 742. [29] S. Levin, D. Semler and Z. Ruben, Toxicol. Pathol., 1993, 21, 1. [30] S. Ruepp, W. Foster, S. Stryker and D. Robertson. Changes in Rat Liver Transcriptome by Over Night Fasting. American College of Toxicology, Annual Meeting, Tucson, AZ, November 9–12, 2008. [31] M. I. Ramirez, D. Karaoglu, D. Haro, C. Barillas, R. Bashirzadeh and G. Gil, Mol. Cell. Biol., 1994, 14, 2809. [32] M. Bauer, A. C. Hamm, M. Bonaus, A. Jacob, J. Jaekel, H. Schorle, M. J. Pankratz and J. D. Katzenberger, Physiol. Genomics, 2004, 17, 230. [33] C. Selman, N. D. Kerrison, A. Cooray, M. D. Piper, S. J. Lingard, R. H. Barton, E. F. Schuster, E. Blanc, D. Gems, J. K. Nicholson, J. M. Thornton, L. Partridge and D. J. Withers, Physiol. Genomics, 2006, 27, 187. [34] M. Sokolovic, D. Wehkamp, A. Sokolovic, J. Vermeulen, L. A. Gilhuijs-Pederson, R. I. van Haaften, Y. Nikolsky, C. T. Evelo, A. H. van Kampen, T. B. Hakvoort and W. H. Lamers, BMC Genomics, 2007, 8, 361. [35] Y. Higami, J. L. Barger, G. P. Page, D. B. Allison, S. R. Smith, T. A. Prolla and R. Weindruch, J. Nutr., 2006, 136, 343. [36] H. Ellinger-Ziegelbauer, J. Aubrecht, J. C. Kleinjans and H. J. Ahr, Toxicol. Lett., 2009, 186, 36. [37] M. R. Fielden, A. Nie, M. McMillian, C. S. Elangbam, B. A. Trela, Y. Yang, R. T. Dunn, 2nd, Y. Dragan, R. Fransson-Stehen, M. Bogdanffy, S. P. Adams, W. R. Foster, S. J. Chen, P. Rossi, P. Kasper, D. Jacobson-Kram, K. S. Tatsuoka, P. J. Wier, J. Gollub, D. N. Halbert, A. Roter, J. K. Young, J. F. Sina, J. Marlowe, H. J. Martus, J. Aubrecht, A. J. Olaharski, N. Roome, P. Nioi, I. Pardo, R. Snyder, R. Perry, P. Lord, W. Mattes and B. D. Car, Toxicol. Sci., 2008, 103, 28.
CHAPT ER
28 To Market, To Market — 2008 Shridhar Hegde and Michelle Schmidt
Contents
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
Introduction Alvimopan (Postoperative Ileus) [4–8] Biolimus Drug-Eluting Stent (Anti-Restenotic) [9–13] Blonanserin (Antipsychotic) [14–17] Ceftobiprole Medocaril (Antibiotic) [18–21] Certolizumab Pegol (Crohn’s Disease) [22–25] Choline Fenofibrate (Dyslipidemia) [26–29] Clevidipine (Antihypertensive) [30–34] Dabigatran Etexilate (Anti-Coagulant) [35–38] Desvenlafaxine (Antidepressant) [39–43] Etravirine (Antiviral) [44–48] Fesoterodine (Overactive Bladder) [49–54] Fosaprepitant Dimeglumine (Antiemetic) [55–57] Icatibant (Hereditary Angiodema) [58–60] Lacosamide (Anticonvulsant) [61–63] Methylnaltrexone Bromide (Opioid-Induced Constipation) [64–66] 17. Pirfenidone (Idiopathic Pulmonary Fibrosis) [67–70] 18. Rilonacept (Genetic Autoinflammatory Syndromes) [71–73] 19. Rivaroxaban (Anticoagulant, Venous Thromboembolism) [74–75] 20. Romiplostim (Antithrombocytopenic) [76–78] 21. Sitafloxacin Hydrate (Antibacterial) [79–82] 22. Sugammadex (Reversal of Neuromuscular Blockade) [83–87] 23. Tafluprost (Antiglaucoma) [88–90] 24. Thrombin Alfa (Hemostat) [91–92] 25. Thrombomodulin (Recombinant) (Anticoagulant) [93–95] References
577 583 585 587 588 591 593 595 598 600 602 604 605 607 609 612 614 615 617 619 621 623 625 627 628 630
Pfizer Global Research & Development, St. Louis, MO 63017, USA Annual Reports in Medicinal Chemistry, Volume 44 ISSN: 0065-7743, DOI 10.1016/S0065-7743(09)04428-5
r 2009 Elsevier Inc. All rights reserved.
577
578
Shridhar Hegde and Michelle Schmidt
1. INTRODUCTION The pharmaceutical market in 2008 saw the launch of 24 new molecular entities (NMEs) for therapeutic use. Although the NMEs of the previous few years were dominated by anticancer and anti-infective drugs, there was a significant shift in this trend this past year. The field of cardiovascular and hematological agents was by far the most prolific area of new product introductions with the launch of seven NMEs, including the only first-in-class therapy of the year [1–3]. The United States and the European Union were the most active markets with nine and eight new product launches, respectively. Of the remaining NMEs, four products reached their first markets in Japan, and three were introduced in Canada. New biological entities (NBEs) accounted for approximately 20% of the NMEs with five product launches. In addition to these NBEs, the year also saw the market entry of four new vaccines, most notably an immunotherapy to prevent recurrence in stage II colon cancer patients following surgical tumor resection. As in previous years, the focus on combinations of existing drugs to enhance patient benefit continued this past year with the launch of several innovative products in the fields of antihypertensives, cholesterol-lowering agents, and antidiabetics. Keeping with the growing trend of recent years, the number of new combinations, new formulations, and new indications of existing drugs continued to grow rapidly. Although these line extensions as well as the new vaccines of the year are not elaborated in this review of NMEs, they comprised a significant portion of the new products in 2008. Among the cardiovascular and hematological NMEs launched past year, Amgen’s recombinant fusion protein Nplatet (romiplostim) attracted significant attention due to its therapeutic novelty. Romiplostim, a thrombopoietin receptor (TpoR) agonist, represented the only first-in-class therapy of the year. It is indicated for the treatment of thrombocytopenia in patients with idiopathic thrombocytopenic purpura (ITP), an autoimmune blood disorder characterized by autoantibodymediated platelet destruction. Promactas (eltrombopag), a small molecule TpoR agonist by GlaxoSmithKline (GSK), also received approval this past year for the treatment of ITP; however, it had not yet been launched by year-end. Other significant additions to the hematological field included two new oral anticoagulants, Pradaxas (dabigatran etexilate) and Xareltos (rivaroxaban), for the prevention of venous thromboembolic (VTE) events in patients who have undergone elective total hip or total knee replacement surgery. Dabigatran etexilate is a pro-drug of dabigatran, which is a direct inhibitor of thrombin, whereas rivaroxaban is a factor Xa inhibitor. A significant unmet need in the thrombosis market is for a new oral anticoagulant to replace the vitamin K antagonists (e.g., warfarin), a class that is efficacious, but suffers from narrow therapeutic
To Market, To Market — 2008
579
index and requires frequent monitoring. With the brief exception of ximelagatran (Exantas), which was launched in 2004 but withdrawn two years later owing to severe elevations in liver enzymes, warfarin has remained unchallenged for almost 50 years. Although dabigatran etexilate and rivaroxaban have entered a relatively narrow section of the thrombosis market, they represent a substantial advancement toward the ultimate goal of safer and broadly applicable oral anticoagulants. The year’s new hematological agents also included two additional biologics: Recomodulint (recombinant thrombomodulin a) as an intravenous anticoagulant for the treatment of disseminated intravascular coagulation (DIC) syndrome and Recothromt (recombinant thrombin) as a locally applied hemostat in surgical procedures. In addition, an albumin- and antibody-free version of the previously marketed, recombinant antihemophilic factor ReFacto was launched past year under the trade name Xynthat. The cardiovascular NMEs of last year included a new intravenous antihypertensive agent Cleviprexs (clevidipine), an ultrashort-acting calcium channel blocker (CCB) for the acute management of hypertension when oral therapy is not feasible or not desirable. Clevidipine has a rapid onset and offset of action and a very short plasma half-life, which allows for titration of the drug to achieve precise control of blood pressure (BP) in critical care setting. The field of drug-eluting coronary stents saw an important advancement with the launch of BioMatrixs, a novel stent system combining a biodegradable polylactic acid polymer (PLA) coating and a new immunosuppressant drug biolimus. The bioresorbable drug coating used in the stent is expected to result in improved safety and vessel-healing characteristics when compared to previous designs that employ durable polymer coatings to affix the immunosuppressant drug on the stent surface. In addition to the NMEs, the cardiovascular field had two new combination products enter the market: Tekturna HCTs (aliskiren fumarate/hydrochlorothiazide), a single-tablet combination of two antihypertensive drugs, and Atocor-R (atorvastatin/ramipril), a fixed-dose combination of a statin and an ACE inhibitor for the treatment of patients with comorbid hypercholesterolemia and essential hypertension. The anti-infective NMEs of 2008 included a new cephalosporin antibiotic Zefterat (ceftobiprole medocaril), a new fluoroquinolone antibacterial Gracevits (sitafloxacin), and a new anti-HIV drug Intelencet (etravirine). Ceftobiprole medocaril is an extended-spectrum cephalosporin with activity against a wide range of difficult-to-treat Gram-positive and Gram-negative infections including methicillinresistant Staphylococcus aureus (MRSA). It was launched past year in Canada for the treatment of complicated skin and skin structure infections (cSSSI) and diabetic foot infection. Sitafloxacin, a fourthgeneration fluoroquinolone for treating a wide range of respiratory and
580
Shridhar Hegde and Michelle Schmidt
urinary tract infections, has the advantage of being effective against many Gram-positive, Gram-negative, and anaerobic clinical isolates, including strains resistant to other fluoroquinolones. Etravirine, a secondgeneration non-nucleoside reverse transcriptase inhibitor (NNRTI), differentiates from its predecessors due to a significantly higher genetic barrier to resistance development. Whereas single mutations of the reverse transcriptase enzyme are capable of conferring marked reductions in susceptibility to first-generation NNRTIs, the development of resistance to etravirine has been shown to require multiple mutations. Another significant addition to the anti-infective domain past year, besides these three NMEs, was the approval and launch of the previously marketed HIV drug Vireads (tenofovir disoproxil) for treating chronic hepatitis B as a new indication. The central nervous system (CNS) area was represented by the entry of four new drugs: Vimpats (lacosamide) for the treatment of partialonset seizures in epilepsy, Pristiqt (desvenlafaxine) for the treatment of major depressive disorder (MDD), Lonasens (blonanserin) for the treatment of psychosis and schizophrenia, and fosaprepitant (IVEmends) for the treatment of chemotherapy-induced nausea and vomiting (CINV). Lacosamide is thought to exert its anticonvulsant effect by selectively enhancing slow inactivation of voltage-gated sodium channels, which results in stabilization of hyperexcitable neuronal membranes. Additionally, lacosamide binds to collapsin response mediator protein 2 (CRMP2), which is involved in neuronal differentiation, control of axonal outgrowth, and possibly epileptogenesis. As compared to traditional anti-epileptic agents, the advantages of lacosamide include a low potential for drug–drug interactions and the availability of an intravenous formulation that may be used for replacement therapy in patients temporarily unable to take oral medication. Desvenlafaxine, a serotoninnorepinephrine reuptake inhibitor (SNRI), is an active metabolite of the previously marketed antidepressant venlafaxine (Effexor XRs). Desvenlafaxine differentiates from its predecessor by the fact that it does not require CYP2D6 metabolism to generate the active drug. This could be beneficial in minimizing interactions with other commonly prescribed medications that are prone to CYP2D6 metabolism. Blonanserin is the newest addition to the growing class of atypical antipsychotics that act by dual antagonism of dopamine D2 and serotonin 5-HT2 receptors. The dual antagonism is a key factor in ensuring effectiveness against both the positive and negative symptoms of schizophrenia while minimizing extrapyramidal symptoms (EPS). Blonanserin is expected to have minimal sedative and hypotensive side effects since its adrenaline-a1 receptor-blocking function is weak. Fosaprepitant, an injectable version of Merck’s NK-1 antagonist aprepitant (Emends), is the phosphate
To Market, To Market — 2008
581
pro-drug of aprepitant, and it is indicated for the prevention of acute and delayed nausea and vomiting associated with highly and moderately emetogenic cancer chemotherapy including high-dose cisplatin. In the area of metabolic diseases, two NMEs were introduced past year: Firazyrs (icatibant) as a novel treatment for acute attacks of hereditary angioedema (HAE), and TriLipixt (choline fenofibrate) as a new fibrate derivative for the treatment of dyslipidemia. HAE is a debilitating and potentially life-threatening genetic disease characterized by unpredictable recurrent swelling attacks in the hands, feet, face, larynx, and abdomen. The culprit in the flare-up of the disease is shown to be increased levels of bradykinin, an endogenous peptide hormone that transmits both pathophysiological and beneficial effects through the G-protein-coupled B2 receptor. Icatibant blocks the action of bradykinin through potent and selective antagonism of the B2 receptor. Both FDA and EMEA have granted orphan drug status to icatibant for the treatment of angioedema. Choline fenofibrate, a salt formulation of fenofibric acid, is indicated both as monotherapy and in combination with a statin to lower triglycerides (TG) and low-density lipoprotein (LDL) cholesterol and to raise high-density lipoprotein (HDL) cholesterol in patients with dyslipidemia. Fenofibric acid is also the active metabolite of Tricors (fenofibrate), which has been previously marketed for this indication. The primary mode of action of fenofibric acid is through the activation of the nuclear transcription factor peroxisome proliferatorsactivated receptor a (PPARa). The gastrointestinal (GI) field saw the entry of two new small molecule drugs, Enteregs (alvimopan) and Relistort (methylnaltrexone bromide) and a new biological agent Cimzias (certolizumab pegol). Alvimopan, a polar molecule with a zwitterionic structure and methylnaltrexone bromide, a quaternary ammonium salt, are both m-opioid receptor antagonists with a high level of peripheral restriction. Consequently, when used during opioid therapy, they are able to negate the peripheral side effects without compromising the centrally mediated analgesia of opioids. Some of the most commonly encountered peripheral side effects of traditional opioids include decreased GI motility and the ensuing complications. Alvimopan is indicated for accelerating GI recovery following bowel resection surgery, and methylnaltrexone bromide is indicated for treating opioid-induced constipation. Certolizumab pegol, a pegylated anti-tumor necrosis factor a (TNF-a) antibody, was launched for treating Crohn’s disease. It is the third anti-TNF-a biologic to be marketed for this indication behind Remicades (infliximab) and Humiras (adalimumab). An anti-a4-integrin biologic, Tysabris (natalizumab), was also approved past year for the treatment of Crohn’s disease as a new indication. Natalizumab was previously marketed for the treatment of multiple sclerosis.
582
Shridhar Hegde and Michelle Schmidt
One new urologic drug entered its first market past year with the launch of Toviazs (fesoterodine) for treating symptoms associated with overactive bladder (OAB). Fesoterodine is an orally active ester prodrug that is hydrolyzed in vivo to the potent muscarinic antagonist 5-hydroxymethyltolterodine (5-HMT). 5-HMT is also an active metabolite of Detrols (tolterodine), which has been marketed for the same indication since 1998. However, unlike tolterodine, the metabolism of fesoterodine to 5-HMT is not mediated by CYP2D6; hence, it affords significantly lower inter-patient variability in average plasma concentrations of 5-HMT with different CYP2D6 phenotypes. The area of chronic inflammatory diseases had one new biologic drug reach the market in 2008. Arcalystt (rilonacept), a dimeric fusion protein, was launched by Regeneron for the long-term treatment of two cryopyrin-associated periodic syndrome (CAPS) disorders: familial cold autoinflammatory syndrome (FCAS) and Muckle–Wells syndrome (MWS). Symptoms of both of these rare genetic disorders include joint pain, rash or skin lesions, fever and chills, eye redness or eye pain, and fatigue. MWS is furthermore associated with more severe inflammation and may include hearing loss or deafness as well as amyloidosis. The target of rilonacept is interleukin-1b (IL-1b), which has been shown to play a key role in the inflammation seen in CAPS patients. Rilonacept blocks IL-1b signaling by acting as a soluble decoy receptor that binds IL-1b and prevents interaction with cell surface receptors. Rilonacept has been granted orphan drug status in the United States. A second biologic Actemras (tocilizumab) was also launched past year in Japan for the new indications of rheumatoid arthritis, juvenile idiopathic arthritis, and systemic-onset juvenile idiopathic arthritis. Tocilizumab is a humanized anti-IL-6 monoclonal antibody, and it has been marketed in Japan since 2005 for the indication of Castleman’s disease. The 2008 launch of a new respiratory drug Pirespas (pirfenidone) in Japan represents a highly significant advancement in the treatment of idiopathic pulmonary fibrosis (IPF), a disease characterized by progressive fibrosis in the interstitial space of the lungs and eventual loss of pulmonary function. Until recently, IPF therapy consisted of a combination of corticosteroids and immunosuppressants to target the inflammation that was believed to be the pathogenic stimulus. Since IPF is now considered to be predominantly a disorder of fibroproliferation, agents that intervene in fibrogenesis have moved to the forefront of treatment options. Pirfenidone is an orally-active small molecule that inhibits collagen synthesis, downregulates the production of multiple cytokines, and blocks fibroblast proliferation and stimulation in response to cytokines.
To Market, To Market — 2008
583
The field of anesthesia was highlighted by the launch of Bridions (sugammadex sodium), a g-cyclodextrin derivative designed specifically to reverse the effects of amino-steroid class of neuromuscular blocking (NMB) drugs such as rocuronium and vecuronium. NMB drugs are used to facilitate endotracheal intubation for mechanical ventilation and to provide muscle relaxation in the operating room and intensive care unit. To decrease the risk for postoperative residual paralysis, NMB is reversed at the end of surgery. Sugammadex acts by forming a very tight host–guest complex with the NMB drug and rendering it inactive. Unlike the currently available agents, sugammadex can reverse within minutes both moderate and deep muscle relaxation induced by rocuronium or vecuronium. In the area of ophthalmic drugs, a new preservative-free prostaglandin (PG), Taflotant (tafluprost), was introduced last year for the reduction of elevated intraocular pressure (IOP) in open-angle glaucoma and ocular hypertension. This product is expected to provide benefits to patients who suffer side effects when using eye drops with preservatives, especially patients with dry or sensitive eyes. The most notable addition to the arsenal of vaccines in 2008 was the anticancer product OncoVAXs. It was launched in Switzerland for the treatment of colorectal cancer. OncoVAXs immunotherapy is prepared from a patient’s own tumor cells, which are dissociated, irradiated to make them non-tumorigenic, and administered to the patient by three weekly injections, beginning 4 weeks after surgery. A booster injection is administered 6 months later. Other additions to the field included three new influenza vaccines: Optaflus, a novel cell-culture vaccine, Prepandrixt, a prepandemic H5N1 influenza vaccine, and Panvaxs, a pandemic H5N1 influenza vaccine. Finally, a new contrast agent also entered the market past year. Lexiscant (regadenoson), an adenosine A2A receptor agonist, was launched as an injectable pharmacologic stress agent in radionuclide myocardial perfusion imaging (MPI) in patients unable to undergo adequate exercise stress. MPI tests, commonly called cardiac stress tests, identify areas of poor blood flow in the heart and facilitate detection and characterization of coronary artery disease. Many patients exercise on a treadmill to generate the increase in coronary blood flow necessary to perform an MPI study. However, in patients who are unable to exercise adequately because of medical conditions, a pharmacologic stress agent that temporarily increases blood flow through the coronary arteries may be used as an alternative. Regadenoson is the first adenosine A2A receptor agonist approved for use as a pharmacologic stress agent.
584
Shridhar Hegde and Michelle Schmidt
2. ALVIMOPAN (POSTOPERATIVE ILEUS) [4–8] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
United States Lilly United States
O H3C
GSK/Adolor Entereg 156053-89-3 424.53
N
N H
COOH
HO CH3
Alvimopan, a peripherally restricted antagonist of the m-opioid receptor, is marketed for the oral treatment of postoperative ileus (POI) following bowel resection surgery. Ileus is defined as a transient decrease in the motility of the GI tract. It is a complication that affects almost all patients that undergo major bowel surgery. The signs and symptoms of POI include abdominal discomfort, inability to tolerate solid food, nausea, and vomiting. POI is a major reason for prolonged hospitalization. Alvimopan is specifically indicated for the acceleration of time to GI recovery following large or small bowel resection surgery with primary anastomosis, and it is the first pharmacotherapy to be approved for this indication. The pathophysiology of POI is complex. However, it is thought that an important contributory factor is the activation of m-opioid receptors in the GI tract by endogenous opioids that are released in response to the stress caused by surgery, as well as by the use of opioid analgesics for the treatment of pain. Activation of these peripheral mopioid receptors leads to an increase in colonic muscle tone and a decrease in propulsive activity in the GI tract. Hence, administration of opioids for postoperative pain relief can prolong POI. The adverse effect of opioids on the GI tract can be reversed with previously marketed m-opioid receptor antagonists such as naloxone and nalmefene. However, these drugs are also active in the CNS, which results in the inhibition of analgesic effects of the opioids as well. By contrast, the zwitterionic nature and the high polarity of alvimopan restrict its passage through the blood–brain barrier, thereby limiting its activity to the periphery. Alvimopan has a higher binding affinity for m-opioid than d- and k-opioid receptors (Ki ¼ 0.4 versus 4.4 and 40 nM, respectively). Following oral administration of alvimopan, the amide hydrolysis compound is present in the systemic circulation, which is also a m-opioid receptor antagonist with a Ki of 0.8 nM. The active metabolite is thought to be produced exclusively by intestinal flora metabolism; its contribution to the clinical efficacy of alvimopan has not been determined. The
To Market, To Market — 2008
585
peripheral restriction of alvimopan was ascertained by studying its effect on opioid-induced analgesia and inhibition of GI motility in healthy volunteers. Thus, oral alvimopan (3 mg, t.i.d.) completely reversed the GI transit effects of oral morphine (30 mg, b.i.d.) without affecting analgesia or pupil constriction. The pharmacokinetics of alvimopan in healthy volunteers is dose proportional at 6–18 mg oral doses, with peak plasma concentrations achieved in about 2 h. The oral bioavailability is low (approximately 6%). After administration of multiple oral doses, the terminal half-life of alvimopan is 10–17 h and of the metabolite is 10–18 h. Mean plasma protein binding of alvimopan and the metabolite are 80% and 94%, respectively. The volume of distribution of alvimopan at steady state is about 30 L, and the average plasma clearance is 402 mL/min. Alvimopan is primarily eliminated through biliary secretion, with the unabsorbed and unchanged drug being hydrolyzed to the metabolite. The metabolite is eliminated in the feces and urine as the unchanged metabolite, the glucuronide conjugate of the metabolite and other minor metabolites. The clinical efficacy of alvimopan in POI was assessed in five randomized, double-blind, placebo-controlled studies. Patients (aged X18 years) undergoing partial large or small bowel resection surgery with primary anastomosis were randomized to receive either alvimopan 12 mg or placebo as a single oral preoperative dose, followed by twice-daily administration for up to 7 days postoperatively. All patients were scheduled to receive either intravenous patient-controlled opioid analgesia or bolus parenteral administration of opioids (intravenous or intramuscular). The primary end point was time to resolution of POI, measured by time to toleration of solid food and time to first bowel movement. In all the studies, alvimopan accelerated the time to recovery of GI function compared with placebo, with the mean time to GI recovery ranging from 11 to 26 h sooner with alvimopan than with placebo. Across all five studies, alvimopan did not reverse opioid analgesia as measured by visual analog scale pain intensity scores and/ or amount of postoperative opioids administered. The most common adverse events associated with alvimopan treatment were constipation, hypokalemia, flatulence, dyspepsia, anemia, urinary retention, and back pain. The recommended dosage for alvimopan is one 12-mg capsule given just before surgery, followed by 12 mg twice daily for up to 7 days. Patients should not receive more than 15 doses of alvimopan. It is only available for short-term use in hospitalized patients, and the hospitals are mandated not to dispense alvimopan to patients after discharge. Alvimopan is contraindicated in patients who have received therapeutic doses of opioids for more than 7 days immediately before the surgery. The use of alvimopan is not recommended in patients with severe hepatic impairment or with end-stage renal disease.
586
Shridhar Hegde and Michelle Schmidt
3. BIOLIMUS DRUG-ELUTING STENT (ANTI-RESTENOTIC) [9–13] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
United States Biosensors European Union Biosensors Biomatrix 851536-75-9 986.28
O OCH2CH3
H H
OCH3
H O
N
CH3
O
O
O
O H3C
OH O
H
H3C OH H3C
H3CO
CH3
O H3CO CH3 CH3
Coronary stents have dramatically improved the success rate of interventional cardiology in recent years, and stent implantation has become the standard of care in percutaneous coronary interventions. However, the long-term success of coronary stenting is hampered by a high rate of restenosis (i.e., recurrence of stenosis, or reblocking), which is caused by proliferation and migration of smooth muscle cells and production of extracellular matrix. Drug-eluting stents (DESs) are the latest advancements in the field to minimize in-stent restenosis. They contain a coating of an antiproliferative drug on a polymer surface and provide controlled release of the drug for extended period of time. Currently marketed DESs contain a coating of either paclitaxel (taxol) or sirolimus (rapamycin) on a durable polymer. These first-generation DESs have significantly reduced the rate of restenosis compared with baremetal stents. However, restenosis still occurs, and there are emerging concerns about late-stent thrombosis caused by delayed neointimal growth, enhanced platelet aggregation, and local hypersensitivity reaction against its polymeric coatings. As a result, improvements in the safety and efficacy of DES continue to be sought, with emphasis on more tissue-specific drug substances and more biocompatible polymer surfaces. The Biomatrixs DES is a novel stent system combining a biodegradable PLA and the new anti-restenoic drug biolimus. Biolimus is a semi-synthetic analog of sirolimus wherein the hydroxyl moiety at position 42 is modified to an ethoxyethyl ether group. As with rapamycin, the mechanism of action of biolimus consists of forming
587
To Market, To Market — 2008
a complex with intracellular 12-kDa FK506-binding protein (FKBP-12), which binds to the mammalian target of rapamycin (mTOR) and reversibly inhibits cell-cycle transition of proliferating smooth muscle cells. The antiproliferative potency of biolimus is similar to that of sirolimus; however, it is approximately 10-fold more lipophilic than sirolimus, which results in rapid absorption of the drug into fatty tissues and reduced systemic exposure. The Biomatrixs DES is produced by the absorption of a 1:1 combination of biolimus and PLA on a flexible stainless steel stent. The precision automated coating method used in the production of the stent ensures the PLA–biolimus combination is applied solely to the abluminal surface of the stent. PLA is co-released with biolimus over 6–9 months, and biodegraded initially to lactic acid, and eventually to carbon dioxide and water. On the basis of in vivo studies, PLA is fully converted to lactic acid at 6 months; however, the effects of PLA release in the surrounding tissues can be detected by a sustained rise in lactic acid concentrations for up to 9 months. A comparative clinical study has demonstrated the efficacy and safety of biolimus DES relative to uncoated stent. A total of 120 patients with 122 de novo coronary lesions were randomized to receive the biolimus stent (n ¼ 80, 82 lesions) or the control uncoated stent (n ¼ 40). At 6 months, a decrease in late lumen loss was observed with the biolimus stent both in the stent (0.26 versus 0.74 mm) and in the segment (0.14 versus 0.40 mm) compared with the control. In-stent restenosis was 3.9% in the biolimus stent group and 7.7% in the control group. Safety and efficacy demonstrated at 6 months was sustained at 1, 2, and 3 years, and there was no incidence of sub-acute or late-stent thrombosis. Biolimus is chemically derived from rapamycin by alkylation with 2-ethoxyethyl triflate in the presence of N,N-diisopropylethylamine in dichloromethane.
4. BLONANSERIN (ANTIPSYCHOTIC) [14–17] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
Japan Sumitomo Japan Dainippon Lonasen 132810-10-7 367.50
N N
F
N
CH3
588
Shridhar Hegde and Michelle Schmidt
Blonanserin, a dual antagonist of dopamine D2 and serotonin 5-HT2 receptors, was launched past year in Japan for the oral treatment of psychosis and schizophrenia. It is the latest entry in the class of atypical antipsychotic agents to reach the market. Schizophrenia is diagnosed by the presence of positive psychotic symptoms such as delusional thoughts, auditory hallucinations, and irrational fears. In addition to these symptoms, many patients also experience the negative symptoms of social alienation, difficulty articulating, apathy, and lack of energy. The first-generation drugs, known as typical antipsychotics, antagonize only the dopamine D2 receptors. Chlorpromazine and haloperidol fall into this category, and although they demonstrate efficacy against psychosis, they are ineffective against the negative symptoms, and dose-limiting extrapyramidal side effects (EPS) are a concern. The second-generation antipsychotics, also known as atypical antipsychotics, antagonize both the mesolimbic pathway dopamine D2 receptors and the serotonin 5-HT2A receptors in the prefrontal cortex. This dual antagonism retains the antipsychotic activity with an improvement in the negative symptoms and a reduction in EPS. Previously marketed atypical antipsychotics include olanzapine (Zyprexas), risperidone (Risperdals), quetiapine (Seroquels), paliperidone (Invegas), and ziprasidone (Geodons). Blonanserin exhibits high affinity for D2 and 5-HT2 receptors (IC50 ¼ 23.6 and 9.85 nM, respectively). Its affinity for D2 receptors is very close to that of haloperidol and risperidone, whereas the affinity for 5-HT2 receptors is about 7 times higher than that of haloperidol and 10 times lower than that of risperidone. In murine models of schizophrenia, both blonanserin and haloperidol (0.1 mg/kg, p.o.) significantly block sensitization to phencyclidine (PCP)-induced hyperlocomotion, a positive symptom of schizophrenia. However, only blonanserin (1 mg/kg, p.o.) attenuates PCP-induced enhancement of immobility in the forced swimming test, a negative symptom of schizophrenia. The recommended initial dose of blonanserin is 4 mg twice daily for adults, and the dose may be increased gradually up to 8 mg twice daily. In pharmacokinetic studies with healthy volunteers, single (0.25–4 mg) and multiple (1 mg/day) oral doses of bloanserin resulted in tmax of 1.4–1.7 h and t1/2 of 4.1–4.8 h, and the increase in AUC was dose proportional. The clinical efficacy and safety of blonanserin was compared with haloperidol in an 8-week, double-blind study that included 263 patients with schizophrenia. Treatment with blonanserin at doses of 8–24 mg/day was at least as effective as haloperidol (4–12 mg/ day) in improving the Final Global Improvement Rating (FGIR) score (61.2% versus 51.3% of patients having at least moderate improvement) and was associated with a lower incidence of extrapyramidal adverse reactions (52.7% versus 75.4%). In addition, an 8-week, randomized, double-blind, phase III trial compared the efficacy of blonanserin with
589
To Market, To Market — 2008
that of risperidone for the treatment of cognitive impairment in 26 schizophrenic patients who received either blonanserin (8–24 mg/day) or risperidone (2–6 mg/day). Both drugs significantly improved Wechsler Memory Scale-Revised (WMS-R) logical memory I/II and Positive and Negative Syndrome Scale (PANSS) test scores at endpoint compared with baseline. In addition, blonanserin significantly improved the Wechsler Adult Intelligence Scale-Revised (WAIS-R) digit symbol test score. No significant differences between the two groups for all measures were observed. Blonanserin is chemically derived in three steps starting with a polyphosphoric acid–mediated condensation reaction of cyclooctanone with 4-fluorobenzoylacetonitrile. The resultant 4-fluorophenylcycloocta[b]pyridin-2-one intermediate is converted to the corresponding 2-chloro derivative by treatment with phosphoryl chloride and subsequently condensed with 1-ethylpiperazine to afford blonanserin.
5. CEFTOBIPROLE MEDOCARIL (ANTIBIOTIC) [18–21] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
Switzerland Roche Canada Basilea/ J&J Zeftera 376653-43-9 690.66
OH N
H N
N
H S
O
CH3
S N H2N
O
N
N
O COOH
O
N
O
O O O
Ceftobiprole medocaril is a new injectable cephalosporin antibiotic with broad-spectrum activity against a wide range of difficult-to-treat Gram-positive and Gram-negative hospital- and community-acquired infections including MRSA. It was launched past year in Canada for the treatment of cSSSI, including diabetic foot infection. Ceftobiprole medocaril is a water-soluble pro-drug of ceftobiprole, which is a pyrrolidinone-3-ylidenemethyl cephem with a 3-pyrrolidinyl side chain. The pro-drug is derived by the attachment of a carbamoyl ester group on the pyrrolidine nitrogen of ceftobiprole. Like all beta-lactam antibiotics, ceftobiprole produces its bactericidal effects by inhibiting cell wall
590
Shridhar Hegde and Michelle Schmidt
synthesis through the prevention of the cross-linking of peptides on the mucosaccharide chains that make up the cell wall. This is accomplished by binding to and inhibiting the penicillin-binding proteins (PBPs). As a result, bacterial cell walls are weakened and susceptible to osmotic pressure and cell lysis. The broad-spectrum activity of ceftobiprole is due to the inhibition of PBPs in both Gram-positive and Gram-negative bacteria. MRSA contains a supplemental peptidoglycan transpeptidase, PBP2a, which is different from normal PBPs and is responsible for its methicillin resistance. Although most available beta-lactams do not efficiently inhibit PBP2a, ceftobiprole binds tightly to its active site and forms a stable acyl–enzyme complex that is resistant to hydrolysis. In addition, ceftobiprole inhibits PBP2x in penicillin-resistant Streptococcus pneumoniae as well as PBP3 and other PBPs in certain Gram-negative bacteria such as Escherichia coli and Pseudomonas aeruginosa. The minimum inhibitory concentration (MIC) of ceftobiprole for methicillin-susceptible S. aureus (MSSA) ranges from r0.25 to 0.5 mg/mL, while those for MRSA range from 0.25 to 4 mg/mL, with a 90% MIC (MIC90) of 2 mg/mL. Vancomycin-resistant strains of S. aureus are also susceptible in vitro. Among Gram-positive pathogens, it has excellent activity against coagulase-negative staphylococci (both methicillin-susceptible and methicillin-resistant strains), S. pneumoniae (including penicillinresistant strains) and other streptococcal species, and Enterococcus faecalis (MIC90 ¼ 4 mg/mL). Ceftobiprole is also active in vitro against a broad range of aerobic Gram-negative bacilli, including P. aeruginosa, against which its activity is similar to that of cefepime. Like other cephalosporins, ceftobiprole demonstrates only modest activity against Acinetobacter spp. and little activity against extended spectrum beta-lactamase (ESBL) or carbapenemase-producing Enterobacteriaceae and non-fermentative bacilli. Like cefepime, but in contrast to other advanced generation cephalosporins, ceftobiprole is a poor substrate for AmpC cephalosporinases and consequently demonstrates only modest activity against Enterobacter spp. With the exception of Fusobacterium nucleatum, the in vitro activity of ceftobiprole against anaerobic Gram-negative bacilli, particularly those that produce b-lactamase, is poor. Ceftobiprole is, however, active against many anaerobic Gram-positive organisms, including Clostridium perfringens. Ceftobiprole medocaril is rapidly and almost completely converted to ceftobiprole by plasma esterases. Diacetyl and carbon dioxide are also formed as byproducts of bioconversion. Peak levels of the active drug in plasma are observed at the end of a 30-min infusion. Afterwards, concentrations in plasma decline in a biphasic manner consistent with a rapid distribution of ceftobiprole from the systemic circulation into other body compartments. Steadystate concentrations are attained on the first day of dosing, with no appreciable accumulation when administered three times daily or twice
To Market, To Market — 2008
591
daily in subjects with normal renal function. The pharmacokinetics of ceftobiprole are linear following single and multiple infusions of 125–1000 mg. It has low protein binding (16%), a volume of distribution at steady state equivalent to the extracellular water compartment (18–20 L), and an elimination half-life in healthy volunteers of approximately 3 h. Clearance is mostly by renal excretion and corresponds to the glomerular filtration rate (4.0–4.5 L/h). Ceftobiprole is not hepatically metabolized. It does not significantly induce or inhibit cytochrome P 450 enzymes and is neither a substrate nor an inhibitor of P-glycoprotein. On the basis of ceftobiprole’s renal clearance, patients with impaired renal function will likely require dose adjustment. The efficacy of ceftobiprole medocaril in comparison with either vancomycin or a combination of vancomycin and ceftazidime in treating patients with cSSSIs has been studied in two non-inferiority clinical trials. Results from the first comparative study demonstrated that ceftobiprole medocaril was not inferior to vancomycin in 784 patients with skin and soft tissue infections caused by Gram-positive infections, including MRSA. In this randomized, double-blind trial, patients were treated with either ceftobiprole medocaril 500 mg every 12 h (n ¼ 397) or vancomycin 1 g every 12 h (n ¼ 387), twice daily for 7–14 days. Overall cure rates in clinically evaluable population were 93.3% and 93.5% for ceftobiprole and vancomycin, respectively. Cure rates in clinically evaluable population with MRSA infections were 91.8% and 90% for ceftobiprole and vancomycin, respectively. The second comparative trial showed that ceftobiprole medocaril was not inferior to a standard combination therapy regimen comprising vancomycin and ceftazidime in patients with cSSSIs due to both Gram-positive and Gram-negative pathogens, including diabetic foot infections. This trial enrolled a total of 828 such patients who were randomized in a 2:1 ratio to receive ceftobiprole medocaril, or vancomycin and ceftazidime. Almost one-third of the patients had diabetic foot infections, of which 75% were moderate to severe cases. Of the clinically evaluable population, 91% and 90% of patients were cured with ceftobiprole medocaril and combination therapy, respectively. The corresponding clinical response rates in patients with diabetic foot infections were 86% with ceftobiprole medocaril and 82% with combination therapy. More than 20% of the microbiologically evaluable population had confirmed MRSA infection. The clinical cure rates in patients with MRSA were 91% with ceftobiprole medocaril and 86% with combination therapy. Among one-third of the total population who had Gram-negative infection, microbiological eradication rates were 84% in both treatment groups. The chemical synthesis of ceftobiprole medocaril starts with the acylation of 7-(R)-amino-3-(hydroxymethyl)-3-cephem-4carboxylic acid with (Z)-5-(amino-[1,2,4]-thiadiazol-3-yl)-trityloxyiminothioacetic acid S-benzothiazol-2-yl ester. The carboxylic acid group on the
592
Shridhar Hegde and Michelle Schmidt
resultant cephem intermediate is then protected as the diphenylmethyl ester by reaction with diphenyldiazomethane, and the hydroxymethyl group is oxidized using either sodium hypochlorite in the presence of 2,2,6,6-tetramethyl-1-piperidinyloxy radical (TEMPO) or manganese dioxide to yield the corresponding aldehyde. Subsequently, the aldehyde is condensed with a phosphonium ylide to give an N-boc-protected pyrrolidinone-3-ylidenemethyl cephem. All three protecting groups (boc, trityl, and diphenylmethyl) are then cleaved by treatment with triethylsilane to produce ceftobiprole. Finally, ceftobiprole medocaril is derived by the reaction of ceftobiprole with 5-methyl-2-oxo-[1,3]dioxolol4-ylmethyl ester 4-nitro phenyl ester.
6. CERTOLIZUMAB PEGOL (CROHN’S DISEASE) [22–25] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no:
United Kingdom UCB Celltech Switzerland UCB Cimzia 428863-50-7
Class:
Humanized pegylated Fab’ Type: anti-TNFa Molecular weight: 91 kDa Expression system: E. Coli Manufacturer: UCB
Certolizumab pegol, a tumor necrosis factor a (TNF-a) blocker, was launched past year for the treatment of moderate-to-severe Crohn’s disease. It is specifically indicated for reducing signs and symptoms and maintaining clinical response in adult patients who have had an inadequate response to conventional therapy. Certolizumab pegol is the third anti-TNF-a biologic to be marketed for this indication behind infliximab (Remicades) and adalimumab (Humiras). Certolizumab pegol is a recombinant, humanized antibody Fab’ fragment linked to approximately 40 kDa polyethylene glycol (PEG). The Fab’ fragment is composed of a light chain with 214 amino acids and a heavy chain with 229 amino acids. It is manufactured in E. coli and is subsequently subjected to purification and conjugation to PEG, to generate certolizumab pegol. The addition of PEG moiety significantly enhances the plasma half-life of the antibody, allowing for less frequent dosing. Certolizumab pegol neutralizes both soluble and membrane-associated forms of human TNF-a. It binds to human TNF-a with slightly higher affinity (KD ¼ 0.09 nM) than infliximab (KD ¼ 0.23 nM) and adalimumab (KD ¼ 0.16 nM). In addition, in cellular assays, certolizumab pegol is more potent in preventing TNF-a-induced killing of L929 fibroblasts than infliximab and adalimumab (IC50 ¼ 0.35 ng/mL versus 5 and 6 ng/mL,
To Market, To Market — 2008
593
respectively). Unlike the full-length antibodies infliximab and adalimumab, certolizumab pegol does not contain a fragment crystallizable (Fc) region and, therefore, does not induce complement activation, antibodydependent cellular cytotoxicity, or apoptosis in vitro. Certolizumab pegol is administered through subcutaneous (s.c.) injection. Peak plasma concentrations are attained 54–171 h after injection, and steady-state concentrations range from 0.5 to 90 mg/mL. The steady-state volume of distribution is about 6.4 L. The bioavailability of certolizumab pegol is approximately 80% following s.c. administration, and the elimination half-life is about 14 days. The recommended dose of certolizumab pegol is 400 mg (given as two s.c. injections of 200 mg) at weeks 0, 2, and 4 for the induction phase, followed by a maintenance regimen of 400 mg every 4 weeks. The safety and efficacy of certolizumab pegol have been assessed in two placebocontrolled clinical studies in adult patients with moderate-to-severe Crohn’s disease, as defined by a Crohn’s Disease Activity Index (CDAI) of 220–450 points. In the first study (n ¼ 662), patients received certolizumab or placebo at weeks 0, 2, and 4, and then every 4 weeks up to week 24. Clinical response was defined as at least a 100-point reduction in the CDAI score compared with baseline, and clinical remission was defined as an absolute CDAI score of 150 points or lower. Clinical response rates at weeks 6 and 26 were significantly higher for certolizumab pegol (35% and 37%, respectively) compared to placebo (27% at both time points). Rate of clinical remission was also higher at week 26 in the certolizumab group (29%) compared to placebo (18%). The second study was a randomized treatment-withdrawal trial in clinical responders to induction therapy with 400-mg certolizumab pegol at weeks 0, 2, and 4. At week 6, non-responders were withdrawn from the study, and responders (n ¼ 428) were randomized to receive either certolizumab pegol (400 mg) or placebo, every 4 weeks starting at week 8, as maintenance therapy through to week 24. The evaluation of CDAI scores at week 26 demonstrated a higher response rate and remission rate for the certolizumab group (63% and 48%, respectively) compared to placebo (36% and 29%, respectively). Baseline use of immunosuppressants or corticosteroids had no impact on the clinical response to certolizumab pegol. The most common adverse reactions with certolizumab were upper respiratory infections (20%), urinary tract infections (7%), and anthralgias (6%). Direct comparisons of efficacy and safety of certolizumab with other TNF-a blockers are not available. TNF-a blockers as a class have been linked to increased risk of opportunistic infections and malignancy. Tuberculosis, invasive fungal infections, and other serious infections have occurred in patients receiving certolizumab pegol, and the drug label contains a black-box warning of the risk.
594
Shridhar Hegde and Michelle Schmidt
7. CHOLINE FENOFIBRATE (DYSLIPIDEMIA) [26–29] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
United States Abbott United States Abbott/Solvay TriLipix 856676-23-8 421.91
O Cl
O O(-) H3C
CH3 (+)
O
N
HO
Choline fenofibrate, a salt formulation of fenofibric acid, is a lipid regulating agent available as delayed release capsules for oral administration. It is indicated in combination with a statin to reduce TG and increase high-density lipoprotein cholesterol (HDL-C) in patients with mixed dyslipidemia and coronary heart disease (CHD) or a CHD-risk equivalent who are optimal for statin therapy to achieve their lowdensity lipoprotein cholesterol (LDL-C) goal. In addition, choline fenofibrate is indicated as monotherapy in patients with severe hypertriglyceridemia to reduce TG, and in patients with primary hyperlipidemia or mixed dyslipidemia to reduce elevated LDL-C, total cholesterol, TG, apolipoprotein B, and to increase HDL-C. Fenofibric acid is also the active metabolite of fenofibrate (Tricors), which has been previously marketed for the treatment of hypercholesterolemia and hypertriglyceridemia. The primary mode of action of fenofibric acid is through the activation of the nuclear transcription factor PPARa, predominantly expressed in tissues that metabolize fatty acids, such as the liver, kidney, heart, and muscle. On activation by binding of the fibrate, PPARa binds as heterodimers with retinoid X receptor (RXR), which subsequently recognizes and binds to specific PPARa-response elements leading to modulation of expression of the target genes. In particular, the activity of lipoprotein lipase is increased and synthesis of apoprotein C-III is decreased, which together enhance the clearance of circulating TG-rich lipoproteins. The resulting fall in TG produces an alteration in the size and composition of LDL from small dense particles to large buoyant particles. These larger particles have a greater affinity for cholesterol receptors and are catabolized rapidly. Following oral administration of the delayed-release capsules, peak fenofibric acid concentrations are reached within 4–7 h, with an absolute bioavailability of approximately 81%. Steady-state concentrations are reached within 8 days. Plasma half-life of fenofibric acid is about 20 h. Protein binding of fenofibric acid is high (approximately 99%). It does not undergo cytochrome P450–mediated metabolism. Fenofibric acid is primarily
To Market, To Market — 2008
595
conjugated with glucuronic acid and then excreted in urine. A small amount of fenofibric acid is reduced at the carbonyl moiety to a benzhydrol metabolite, which is also excreted in urine as the glucuronide. Dose reductions are recommended in patients with mild to moderate renal impairment. Choline fenofibrate may be taken without regard to food. The recommended dosage in patients with mixed dyslipidemia is 135 mg once daily. In patients with hypertriglyceridemia, a dosage of 45–135 mg once daily is recommended. In patients with mild to moderate renal impairment, the recommended dosage is 45 mg once daily; the dosage should only be increased after an evaluation of the effects on renal function and lipid levels at this dosage. Choline fenofibriate acid is contraindicated in patients with severe renal impairment. The clinical efficacy of choline fenofibrate in combination with low-dose and moderate-dose statins has been demonstrated in three 12-week phase III studies, and one 52-week, long-term, open-label extension study. A total of 2,698 subjects with mixed dyslipidemia were enrolled in these studies. In the three 12-week studies, subjects received choline fenofibrate co-administered with 10-mg or 20-mg rosuvastatin, or 20-mg or 40-mg simvastatin, or 20-mg or 40-mg atorvastatin. The primary efficacy endpoints for all three studies were mean percent changes from baseline to final value in HDL-C, TG, and LDL-C. For each statin dose co-administered with choline fenofibrate, there were three primary comparisons. For HDL-C and TG, the combination therapy was compared with statin monotherapy at the corresponding dose, and for LDL-C, it was compared with choline fenofibrate monotherapy. In the pooled analysis, choline fenofibrate co-administered with both lowdose statins and moderate-dose statins resulted in mean percent HDL-C increases of 18.1% and 17.5%, respectively, and mean percent TG decreases of 43.9% and 42.0%, respectively. These changes were significantly greater than the corresponding dose of statin monotherapy (7.4% and 8.7% for HDL-C; 16.8% and 23.7% for TG). In addition, both doses of combination therapy resulted in mean percent LDL-C decreases of 33.1% and 34.6%, respectively, which were significantly greater than choline fenofibrate monotherapy (5.1%). The subjects who completed the 12-week treatment period of any of the studies were eligible to participate in the 52-week long-term extension study. In this study, the subjects received choline fenofibrate co-administered with the moderate dose of the statin that had been used in the 12-week study in which they were enrolled. Mean 52-week values and mean percent change from baseline were 91.7 mg/dL (38.2%) for LDL-C, 47.3 mg/dL (+24.0%) for HDL-C, and 135.5 mg/dL (47.6%) for TG. The most common adverse reactions observed with the use of fenofibric acid alone or in combination with a statin are headache, back pain, nasopharyngitis, nausea, myalgia, diarrhea, and upper respiratory tract infection. Choline
596
Shridhar Hegde and Michelle Schmidt
fenofibrate is chemically derived from p-anisole in four steps through Friedel–Crafts acylation with p-chlorobenzoyl chloride to the corresponding benzophenone derivative, followed by methyl ether cleavage with hydrogen bromide, alkylation of the phenolic hydroxyl group with 2-bromo-2-methylpropanoic acid using sodium hydroxide, and finally salt formation with choline.
8. CLEVIDIPINE (ANTIHYPERTENSIVE) [30–34] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
United Kingdom AstraZeneca United States The Medicines Co. Cleviprex 167221-71-8 456.32
Cl
Cl O
O H3CO H3C
O O
N H
O
CH3
CH3
Clevidipine is an ultra-short-acting vasodilator of the dihydropyridine (DHP) class of CCB. It is indicated as an intravenous treatment for the acute management of hypertension when oral therapy is not feasible or not desirable. Clevidipine is the second intravenous CCB to be marketed behind nicardipine for this indication. Both agents are primarily used for urgent treatment of hypertension in cardiac surgical setting, intensive care units, and emergency departments. Other intravenous agents currently on the market for this indication include sodium nitroprusside, nitroglycerin, fenoldopam, hydralazine, esmolol, and labetalol. Clevidipine is a potent, arterial-specific, vasodilator with very little or no effect on the myocardial contractility and venous capacitance. It has a rapid onset and offset of action and a very short plasma half-life, which allows for titration of the drug to achieve precise BP control. Sodium nitroprusside and nitroglycerin are also short-acting, but lack arterial selectivity. Hydralazine and labetalol both exhibit significantly longer plasma half-lives than clevidipine. In addition, unlike several of its counterparts, clevidipine is exclusively metabolized in the blood and the tissues. It does not undergo any renal or hepatic metabolism and does not accumulate in the body. Clevidipine is a blocker of L-type calcium channels, which mediate the influx of calcium during depolarization in arterial smooth muscle. Experiments in anesthetized rats and dogs show
To Market, To Market — 2008
597
that clevidipine reduces mean arterial BP by decreasing systemic vascular resistance. It does not reduce cardiac filling pressure (pre-load), confirming lack of effects on the venous capacitance vessels. Clevidipine is administered as a continuous intravenous infusion through a central or peripheral line. The dose is titrated to achieve the desired BP reduction and individualized depending on the BP response and target BP. Clevidipine infusion is initiated at 1–2 mg/h, with the dose doubled at 90-s intervals initially. As BP approaches the goal, the increase in doses should be less than doubling, and time between dose adjustments should be increased to every 5–10 min. An approximately 1–2 mg/h increase generally produces an additional 2–4 mm Hg decrease in systolic pressure. The desired therapeutic response for most patients occurs at doses of 4–6 mg/h. Patients with severe hypertension may require doses up to 16–32 mg/h; however, experience with 32 mg/h dose is very limited. Clevidipine exhibits linear pharmacokinetics in healthy volunteers over a wide range of doses from 0.12 to 48 nmol/min/kg. The pharmacokinetic parameters of R and S enantiomers are essentially similar to each other and to those of racemic clevidipine. Clevidipine is rapidly distributed and metabolized resulting in a very short half-life. The initial phase half-life is approximately 1 min and accounts for 85–90% of clevidipine elimination. The terminal half-life is approximately 15 min. Clevidipine has high protein binding (W99.5%), high clearance (69 mL/min/kg), and low volume of distribution (0.19 L/kg). It is rapidly metabolized by hydrolysis of the butanoic ester linkage, primarily by esterases in the blood and extravascular tissues. The primary metabolites are the DHP-carboxylic acid and formaldehyde. The DHP-carboxylic acid metabolite is inactive as an antihypertensive, and it is further metabolized by glucuronidation or oxidation to the corresponding pyridine derivative. In a clinical study with radiolabeled clevidipine, 83% of the drug was excreted as metabolites in urine and feces. More than 90% of the recovered radioactivity is excreted within the first 72 h of collection. The clinical efficacy of clevidipine in controlling perioperative hypertension has been evaluated in two placebo-controlled, double-blind, randomized trials. In the first study, 152 patients with current or recent hypertension undergoing cardiac surgery were randomized to clevidipine or placebo preoperatively. Among the patients with systolic blood pressure (SBP) X160 mm Hg, 92.5% of the clevidinetreated group achieved a SBP reduction goal of X15% from baseline, compared with 17.3% of patients on placebo. The target decrease in BP with clevidipine occurred in a median of 6 min. The second study in 110 patients with SBP X140 mm Hg showed that clevidipine administered postoperatively following cardiac surgery, achieved a 91.8% success rate in attaining SBP reduction goal of X15%, compared with 20.4% of patients on placebo. The response was achieved in a median of 5 min.
598
Shridhar Hegde and Michelle Schmidt
In yet another prospective, open label study in 126 patients presenting in an emergency department or intensive care unit with severe hypertension (SBP W180 mm Hg or diastolic blood pressure W115 mm Hg), clevidipine lowered BP to the target range within 30 min in 89% of the patients. The most commonly reported adverse events associated with clevidipine include headache, nausea, and vomiting. Hypotension and reflex tachycardia are potential consequences of a rapid upward titration of clevidipine. If either occurs, the dose must be reduced. Beta-blockers are not recommended for the treatment of clevidipine-induced tachycardia. Clevidipine is contraindicated in patients with severe aortic stenosis, defective lipid metabolism, and allergies to eggs, egg products, soybeans, or soy products. The chemical synthesis of clevidipine entails the esterification of 4-(2,3-dichlorophenyl)-2,6-dimethyl-1,4-dihydropyridine-3,5-dicarboxylic acid monomethyl ester with chloromethyl butyrate by the action of potassium bicarbonate in refluxing acetonitrile. The DHP-monoester starting material is obtained in two steps through condensation of methyl 2,3-dichlorobenzylidine acetoacetate and 2-cyanoethyl ester of 2-amino-2-propenoic acid to the DHP diester, followed by base catalyzed cleavage of the cyanoethyl group. Clevidipine is practically insoluble in water and is formulated in an oil-inwater emulsion. In addition to the active ingredient, the injectable formulation also contains soybean oil, glycerin, and purified egg yolk phospholipids.
9. DABIGATRAN ETEXILATE (ANTI-COAGULANT) [35–38] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
Germany Boehringer Ingelheim United Kingdom Boehringer Ingelheim Pradaxa 211915-06-9 627.73 O
O
H3C N H2N O N
N H
N
O
N
CH3
N
H3C
O
Dabigatran etexilate is an orally administered pro-drug of dabigatran, which is a direct inhibitor of thrombin and a potent anticoagulant. The
To Market, To Market — 2008
599
serine protease thrombin is the final mediator in the coagulation cascade that leads to the production of fibrin, the main protein component of blood clots. Thrombin is also a potent activator of platelets. Consequently, thrombin inhibitors have found utility as anticoagulants in treating arterial and venous thrombosis. Historically, antithrombotic therapy has been mainly based on vitamin K antagonists (e.g., warfarin), or indirect thrombin inhibitors such as low-molecular-weight heparins (e.g., enoxaparin). Despite their high clinical effectiveness at reducing deep-vein thrombosis (DVT) and pulmonary embolism (PE), each of these therapies has limitations. Warfarin, although orally available, has a narrow therapeutic index and requires frequent monitoring and dosage adjustment. Low-molecular-weight heparins provide better safety profile and less inter-patient variability; however, these agents are only available for parenteral administration. Direct thrombin inhibitors (DTIs) are a new class of anticoagulants that act by directly binding to thrombin at its catalytic or fibrinogen-binding sites, or both. Unlike the heparins, DTIs do not require the activation of secondary factors such as antithrombin to derive their activity, which makes their action more predictable. In addition, their ability to inhibit both free and clot-bound thrombin predisposes them for enhanced anticoagulation effect. The first orally available DTI, ximelagatran (Exantas), was approved in Europe in 2004, but was subsequently withdrawn from the market owing to issues with liver toxicity. Dabigatran etexilate is the next oral DTI to reach the market. It is specifically indicated for use in the prevention of VTE events in adult patients who have undergone elective total hip or total knee replacement surgery. In contrast to warfarin, dabigatran etexilate can be administered as a fixed dose with a rapid onset of action. It provides a predictable and consistent anticoagulation effect without the need for monitoring. Dabigatran is a potent inhibitor of thrombin (Ki ¼ 4.5 nM). Oral dabigatran etexilate is rapidly and almost completely hydrolyzed to dabigatran by plasma esterases. The bioconversion consists of hydrolyzing both the ethyl and the hexyl ester groups of the pro-drug. Peak plasma concentrations of dabigatran occur at 0.5–2 h after dosing in healthy volunteers and 7–9 h in postsurgical patients on the operative day, falling to 2 h on subsequent days. The half-life of dabigatran is 11 h in young healthy subjects and 14–17 h in the elderly. Dabigatran has low protein binding (35%) and a very high volume of distribution (60–70 L). Dabigatran is excreted mostly unchanged in the urine (85%). The efficacy of dabigatran etexilate was evaluated in two randomized, double-blind trials; one involving patients undergoing knee replacement surgery (n ¼ 2,101), and another with patients undergoing hip replacement surgery (n ¼ 3,493). Patients were randomized to either dabigatran etexilate (75 or 110 mg per day orally) or low-molecular-weight heparin enoxaparin (40 mg/day as a subcutaneous injection). In the trial for
600
Shridhar Hegde and Michelle Schmidt
patients undergoing knee replacement, treatment was for 6–10 days, and in the trial for patients undergoing hip replacement, treatment was for 28–35 days. The primary efficacy end point was a composite of incidences of total VTE events (including DVT and PE) and all-cause mortality during the treatment period. The results of both studies showed that 220-mg and 150-mg doses of dabigatran etexilate and 40-mg dose of enoxaparin achieve similar primary efficacy endpoints (incidences of 6.0%, 8.6%, and 6.7%, respectively, in patients undergoing hip replacement, and 36.4%, 40.5%, and 37.7%, respectively, in patients undergoing knee replacement). There were no significant differences in major bleeding rates with either dose of dabigatran etexilate compared with enoxaparin in either study. The recommended oral dose of dabigatran etexilate is 220 mg once daily taken as two capsules of 110 mg. Treatment is initiated within 1–4 h of completed surgery with a single capsule and subsequently continued with 220 mg/d dosages for a total of 10 days for knee replacement surgery and 28–35 days for hip replacement surgery. Dabigatran etexilate exerts a predictable anticoagulant effect, and at recommended dosages, routine coagulation monitoring is not required. Patients with moderate renal impairment (glomerular filtration rate [GFR] between 30 and 50 mL/min) have higher plasma levels of dabigatran and are at increased risk of bleeding and therefore require lower dosages. Dabigatran is contraindicated in patients with severe renal impairment (GFRo30 mL/min). Additionally, it is not recommended in patients with liver enzymes W2-fold higher than the upper limit of normal (ULN). The chemical synthesis of dabigatran etexilate starts with the acylation of ethyl 3-(2-pyridylamino)propionate with 4-(methylamino)-3-nitrobenzoyl chloride to produce the corresponding amide. Subsequent reduction of the nitro group by catalytic hydrogenation, and cyclization of the resultant phenylenediamine with N-(4-cyanophenyl)glycine leads to a benzimidazole intermediate. The cyano group is then transformed into an amidine by employing the Pinner reaction. Finally, acylation of the amidino group with hexyl chloroformate gives rise to dabigatran etexilate.
10. DESVENLAFAXINE (ANTIDEPRESSANT) [39–43] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
CH3
United States Wyeth United States Wyeth Pristiq 93413-62-8 263.38
N H3C OH
HO
To Market, To Market — 2008
601
Desvenlafaxine is a new member of the SNRI class of antidepressants. It is marketed as an extended release tablet for once-daily oral treatment of MDD in adult patients. Desvenlafaxine (O-desmethylvenlafaxine) is the major active metabolite of the previously marketed antidepressant venlafaxine (Effexors). Similar to venlafaxine, desvenlafaxine is a more potent inhibitor of serotonin and norepinephrine reuptake (IC50 ¼ 47 and 531 nM, respectively) than that of dopamine (62% inhibition at 100 mM) in in vitro studies. It does not have significant in vitro affinity for other neurotransmitter reporters such as a1-adrenergic, H1-histaminergic, or muscarine-cholinergic receptors. Desvenlafaxine is marketed as a racemic mixture. Specific pharmacological properties of the enantiomers of desvenlafaxine have not been reported. The recommended dose for desvenlafaxine is 50 mg once daily, with or without food. In clinical studies, doses of 50–400 mg/day were shown to be effective, although no additional benefit was demonstrated at doses greater than 50 mg/day. In addition, adverse events and discontinuations were more frequent at higher doses. After oral administration, the absolute bioavailability of desvenlafaxine is about 80%. Mean time to peak plasma concentrations (Tmax) is about 7.5 h, and the mean terminal half-life is approximately 11 h. With once-daily dosing, steady-state plasma concentrations are achieved within approximately 4–5 days. The plasma protein binding of desvenlafaxine is low (30%) and is independent of drug concentration. The volume of distribution at steady-state following intravenous administration is 3.4 L/kg, indicating distribution into non-vascular compartments. Desvenlafaxine is primarily metabolized through conjugation mediated by uridine diphosphate-glucuronosyltransferase (UGT) and, to a lesser extent, through oxidative N-demethylation mediated by CYP3A4. Approximately 45% of desvenlafaxine is excreted unchanged in urine at 72 h after oral administration. Approximately 19% of the administered dose is excreted as the glucuronide metabolite and o5% as the oxidative metabolite in urine. Desvenlafaxine has been evaluated in multiple clinical trials; however, efficacy results using the approved dosage of 50 mg/day are published for only two studies. These trials compared the efficacy of 50- and 100-mg doses of desvenlafaxine to placebo over 8-week treatment period. The enrollment consisted of MDD patients (n ¼ 959) who had a total score of X20 on the 17-item Hamilton Rating Scale for Depression (HAM-D17). The first of these studies found that the mean decrease in HAM-D17 from baseline was 11.5 and 11.0 with 50- and 100-mg doses of desvenlafaxine, respectively, as compared to 9.5 with placebo. In the second study, the decrease from baseline was 10.7 with placebo, and 13.2 and 13.7 with 50- and 100-mg doses of desvenlafaxine, respectively. Comparative studies to evaluate the safety and efficacy of desvenlafaxine relative to venlafaxine have not been reported. The most common adverse effects of desvenlafaxine include nausea, insomnia, dry mouth, dizziness, sweating, nervousness, anorexia, constipation, and asthenia. As with other antidepressants from the
602
Shridhar Hegde and Michelle Schmidt
SNRI class, desvenlafaxine carries a black box warning showing an increased risk of suicidal ideation and behavior compared to placebo for patients under the age of 24. Desvenlafaxine is chemically derived from 4-benzyloxyphenylacetic acid, by first converting to the acid chloride with oxalyl chloride and then condensing with dimethylamine to produce the corresponding N,N-dimethyl amide. Deprotonation of the amide with butyllithium, followed by condensation with cyclohexanone gives a b-hydroxyamide intermediate, which is reduced with alane, and debenzylated by using palladium on carbon and 1,4-cyclohexadiene to produce desvenlafaxine.
11. ETRAVIRINE (ANTIVIRAL) [44–48] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
United States Janssen United States Tibotec Intelence 269055-15-4 435.28
CH3
Br O
NC
NH2
N CH3
N HN
CN
Etravirine is a second-generation NNRTI. It is indicated for use in combination with other antiretroviral agents for treating HIV-1 infection in treatment-experienced adult patients who have evidence of viral replication and HIV-1 strains resistant to the currently available NNRTIs and other antiretroviral agents. The NNRTIs, along with nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs/NtRTIs), are important components of the combination regimens currently used to treat HIV-1 infection. The NRTIs/NtRTIs act by competing with the natural nucleotide substrates of reverse transcriptase for incorporation into viral DNA and subsequent chain termination. By contrast, the NNRTIs bind to an allosteric site of the enzyme and disrupt the DNA polymerase function by inducing conformational changes to the catalytic site. The allosteric binding nature of NNRTIs generally results in improved safety profile since there is no known human homolog for the drug-binding site of the enzyme. The first-generation NNRTIs currently on the market include efavirenz (Sustivas), nevirapine (Viramunes), and delavirdine (Rescriptors). Although these NNRTIs have been successfully used for almost a decade, their major drawback is the low genetic barrier to
To Market, To Market — 2008
603
resistance development, with single reverse transcriptase mutations capable of conferring marked reductions in drug susceptibility and cross-resistance to other agents in the class. Mutations at residues surrounding the NNRTI binding site, in particular K103N, are shown to be the most prevalent causes of resistance development. The secondgeneration NNRTI, etravirine, appears to be less sensitive to these mutations. It has potent in vitro activity against wild-type HIV (EC50 ¼ 1.4 nM; EC90 ¼ 2.9 nM), as well as viruses with the common NNRTI mutations (EC50 ¼ 1–10 nM). For example, EC50 values against HIV strains carrying the K103N, Y188L, and L100I mutations are 1.3, 4, and 3 nM, respectively. Against the same strains, the values for delavirdine are 1697, 187, and 144 nM, respectively, and for efavirenz are 38, 39, and 144 nM, respectively. The inherent conformational flexibility of etravirine structure is believed to allow multiple binding configurations and hence enable the agent to overcome the effects of some resistance mutations to a greater extent than other NNRTIs with more rigid structure. Development of resistance to etravirine has been shown to require multiple mutations of reverse transcriptase enzyme. Following oral administration, etravirine is absorbed with a Tmax of about 2.5–4 h. The Cmax and AUC are decreased by about 50% when administered under fasting conditions, as compared to when administered following a meal. Etravirine is highly protein bound (99.9%), primarily to plasma albumin and alpha-1-acid glycoprotein. Distribution of etravirine into compartments other than plasma has not been evaluated. The mean terminal plasma elimination half-life is 30–40 h. Etravirine steady-state concentration is reached after 4–5 days. Etravirine is excreted mainly unchanged in the feces. Less than 20% of the drug is metabolized, with CYP-mediated oxidation as the primary route of metabolism. The major metabolites, formed by hydroxylation of methyl groups in the dimethylbenzonitrile moiety, are at least 90% less active than etravirine against wild-type HIV in cellular assays. The recommended oral dose of etravirine is 200 mg taken twice daily following a meal. The safety and efficacy of etravirine were evaluated in two similar randomized, double-blind, placebo-controlled phase III trials (n ¼ 1,203). Eligible subjects were treatment-experienced patients with plasma HIV-1 RNA W5,000 copies/mL while on a stable antiretroviral regimen for at least 8 weeks. In addition, the subjects had at least one documented NNRTI mutation and three or more primary PI mutations. These studies compared 200 mg twice-daily etravirine to placebo in patients who were on a background regimen of darunavir/ritonavir and at least two other antiretroviral agents. Response to therapy at 24 weeks was defined as o50 HIV-1 RNA copies/mL. At week 24, pooled data from both trials showed that 59.8% of the patients in the etravirine group responded to therapy as compared to 40.2% in the placebo group. The most common adverse reactions of etravirine treatment included rash, nausea, peripheral
604
Shridhar Hegde and Michelle Schmidt
neuropathy, and hypertension. As with other NNRTIs, etravirine has many drug–drug interactions. It is a substrate of CYP3A4, CYP2C9, and CYP2C19, an inducer of 3A4, and an inhibitor of 2C9 and 2C19. Caution should be used with co-administration of inducers, inhibitors, or substrates of these enzymes. Etravirine can be synthesized starting from 5-bromo-2,4,6-trichloropyrimidine through three successive nucleophilic substitution reactions. Initial displacement with 4-aminobenzonitrile using Hu¨nig’s base, followed by reaction with sodium salt of 4-hydroxy-3,5dimethylbenzonitrile, and subsequent ammonolysis reaction with ammonia in dioxane under pressure affords etravirine.
12. FESOTERODINE (OVERACTIVE BLADDER) [49–54] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
Germany Schwarz Pharma United Kingdom Pfizer Toviaz 286930-02-7 411.58
O H3C
O
CH3
CH3
N H3C
CH3 CH3
OH
Fesoterodine, launched last year for the treatment of OAB, is an orally active pro-drug that is converted in vivo to its active metabolite 5-HMT through hydrolysis by non-specific esterases. 5-HMT is also an active metabolite of tolterodine (Detrol s), which has been marketed for the treatment of OAB since 1998. 5-HMT is a potent muscarinic antagonist, with essentially equivalent affinity for M1, M2, M3, M4, and M5 receptors (Ki ¼ 0.32, 0.63, 1.26, 2, and 0.63 nM, respectively). The binding of 5-HMT is stereoselective; the corresponding S-enantiomer has at least 100 times lower binding affinity for all five receptors. Fesoterodine is supplied as its fumarate salt in an extended release tablet form. The recommended starting dose is 4 mg once daily. On the basis of individual response and tolerability, the dose may be increased to 8 mg once daily. Following oral administration, fesoterodine is not detected in the peripheral blood, thereby indicating a rapid and complete bioconversion to 5-HMT. Bioavailability of 5-HMT is about 52%. After single- or multiple-dose oral administration of fesoterodine in doses from 4 to 28 mg, plasma concentrations of 5-HMT are proportional to the dose. Maximum plasma levels are reached after approximately 5 h. No accumulation occurs after multiple-dose administration. 5-HMT is
To Market, To Market — 2008
605
further metabolized in the liver through oxidation of the hydroxymethyl group and oxidative cleavage of N-alkyl groups mediated by CYP2D6 and CYP3A4. However, none of these secondary metabolites contribute significantly to the antimuscarinic activity of fesoterodine. 5-HMT and its metabolites are primarily eliminated through renal excretion. The terminal half-life of 5-HMT is approximately 4 h following an intravenous administration. The apparent terminal half-life following oral administration is approximately 7 h. Although 5-HMT is also a metabolite of tolterodine, there is an important distinction between these two drugs. Unlike fesoterodine, tolterodine metabolism to 5-HMT is mediated by CYP2D6; hence the Cmax and AUC of 5-HMT are subject to larger inter-patient variability depending on CYP2D6 phenotype of patients. In contrast, average plasma concentrations of 5-HMT following oral administration of fesoterodine differ by less than a factor of 2 in subjects with different CYP2D6 phenotypes. The efficacy of fesoterodine in OAB was assessed in two phase III, randomized, double-blind, placebo-controlled, 12-week trials. Patients were randomized to treatment with fesoterodine 4 or 8 mg/day or placebo. Between the two studies, 554 patients were treated with fesoterodine 4 mg/day, 566 were treated with fesoterodine 8 mg/day, and 554 were treated with placebo. The primary end points were mean change in the number of urge urinary incontinence episodes per 24 h and mean change in the number of micturitions per 24 h. In the first study, the number of urge incontinence episodes per 24 h showed a change from baseline of 1.20, 2.06, and 2.27 for placebo and fesoterodine 4-mg and 8-mg arms, respectively (p ¼ 0.001 versus placebo). The number of micturitions per 24 h showed a change from baseline of 1.02, 1.74, and 1.94 for placebo, fesoterodine 4-mg, and fesoterodine 8-mg arms, respectively (po0.001 versus placebo). In the second study, the number of urge incontinence episodes per 24 h showed a change from baseline of 1.00, 1.77, and 2.42 for placebo and fesoterodine 4-mg (po0.003 versus placebo) and 8-mg (po0.001 versus placebo) arms, respectively. The number of micturitions per 24 h showed a change from baseline of 1.02, 1.86, and 1.94 for placebo and fesoterodine 4-mg (p ¼ 0.032) and 8-mg (po0.001) arms, respectively. In both studies, the number of urge urinary incontinence episodes per 24 h was reduced as early as 2 weeks after initiation of fesoterodine therapy. The most common adverse events associated with fesoterodine include dry mouth, constipation, and dyspepsia. Fesoterodine is contraindicated in patients with urinary retention, gastric retention, or uncontrolled narrow-angle glaucoma. The chemical synthesis of fesoterodine starts with the cyclization of 4-(hydroxymethyl)phenol with cinnamaldehyde by means of piperazine in refluxing toluene to produce 2-hydroxy-6-(hydroxymethyl)-4-phenyl-3,4-dihydro-2H-1benzopyran, which is subjected to reductive amination with
606
Shridhar Hegde and Michelle Schmidt
diisopropylamine by means of hydrogen over palladium hydroxide catalyst. The resultant racemic amine intermediate is optically resolved with (R)-(2)-acetoxy-2-phenylacetic acid to yield the corresponding (R)enantiomer, which is finally acylated with isobutyryl chloride to afford fesoterodine.
13. FOSAPREPITANT DIMEGLUMINE (ANTIEMETIC) [55–57] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
United States Merck United Kingdom Merck IvEmend 265121-04-8 1004.83
O O HO
OH
CH3
O
NH P
N N
N
CF3
OH
O
OH
H N OH
Me CF3
OH
OH
2
F
One of the most debilitating side effects of cancer treatment is CINV. Although concomitant anti-emetic agents have been prevalent for the past two decades, 50–60% of patients are still plagued by CINV. The large refractory population has led the National Comprehensive Cancer Network (NCCN) and the Multinational Association Supportive Care in Cancer (MASCC) to revise the anti-emetic guidelines. The current recommendation is a three-drug regimen including a traditional 5-HT3 antagonist, dexamethasone, and previously the only neurokinin-1 (NK-1) antagonist aprepitant. Since many of these patients cannot tolerate orallyadministered medication, the option of intravenous anti-emetics is essential. Although intravenous 5HT3 and dexamethasone have been available, aprepitant only existed as an oral agent. To meet this need, fosaprepitant dimeglumine has been launched as a water-soluble, phosphoryl pro-drug of aprepitant for intravenous applications. This pro-drug is rapidly converted to aprepitant in vivo, which has a high affinity for the NK-1 receptor (IC50 ¼ 90750 pM for the displacement of [125I]-substance P from hNK-1 expressed in CHO cells); antagonism
To Market, To Market — 2008
607
of NK-1 prevents binding of the endogenous substance P that is known to induce vomiting. Fosaprepitant is prepared by the regioselective phosphorylation of aprepitant with dibenzylphosphoryl chloride followed by catalytic hydrogenolysis in the presence of N-methyl-Dglucamine. It is approved for the prevention of acute and delayed nausea and vomiting associated with initial and repeat courses of highly and moderately emetogenic cancer chemotherapy. Given 30 min before chemotherapy on day 1, an infusion of 115 mg of fosaprepitant over 15 min is a comparable substitute for 125 mg of oral aprepitant. Following a single IV injection, the pharmacokinetic parameters reflect the parent aprepitant due to its efficient emergence. The mean AUC is 31.7 mg h/mL, and the Cmax is 3.27 mg/mL. Aprepitant is greater than 95% plasma protein bound, and the mean volume of distribution at steady state is approximately 70 L. The terminal half-life ranges from 9 to 13 h with elimination resulting in nearly equal recovery in the urine and feces (57% and 45%, respectively). Metabolism is predominantly by CYP3A4 (oxidation of the morpholine ring and side chains) and to a lesser extent CYP1A2 and CYP2C19. Not only is aprepitant a substrate for CYP3A4, it is also a moderate inhibitor and inducer of CYP3A4, as well as an inducer of CYP2C9. When fosaprepitant is coadministered with CYP3A4 substrates, plasma concentrations of these agents may be elevated. In particular, the dexamethasone dose should be reduced by 50% while cisapride and pimozide are contraindicated with concomitant aprepitant. Conversely, aprepitant plasma concentrations may increase when coadministered with strong CYP3A4 inhibitors, such as ketoconazole, itraconazole, clarithromycin, nelfinavir, and ritonavir. As an inducer of CYP2C9, aprepitant may also decrease plasma levels of the CYP2C9 substrate warfarin creating the need for prothrombin time monitoring for 2 weeks following fosaprepitant treatment. The efficacy of oral contraceptives may also be compromised during fosaprepitant administration; therefore, alternative contraceptive methods are recommended during and for the month following the last dose of fosaprepitant. In a phase II study enrolling 53 cisplatin-naı¨ve patients, IV fosaprepitant was compared to IV ondansetron in a randomized, double-blind manner. Although there was no statistical difference in the prevention of emesis in the first 24 h after chemotherapy (37% versus 52%), fosaprepitant was better at controlling delayed emesis (72% versus 30% for ondansetron). A second phase II study involved 177 cisplatin-naı¨ve patients that were randomized to three groups: group 1 received fosaprepitant (100 mg) and dexamethasone (20 mg) on day 1, followed by 300 mg aprepitant on days 2–5: in group 2, placebo was given instead of aprepitant on days 2–5; and group 3 received ondansetron (20 mg) and dexamethasone (20 mg) on day 1, followed by 4 days of placebo. Although it should be noted that this study was accomplished before the effects of aprepitant
608
Shridhar Hegde and Michelle Schmidt
on dexamethasone dose were elucidated and before an optimal fosaprepitant dose was established, the results indicated that the fosaprepitant/aprepitant-dexamethasone therapy was better at preventing delayed emesis versus acute emesis; the ondansetron–dexamethasone combination proved superior in managing acute emesis (84% versus 47–49%). The major adverse events, occurring in levels of at least 10%, included asthenia/fatigue, headache, alopecia, nausea, constipation, anorexia, diarrhea, and hiccups.
14. ICATIBANT (HEREDITARY ANGIODEMA) [58–60] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight: H2N
Germany Jerini United Kingdom, Germany Jerini Firazyr 138614-30-9 1304.52
NH NH
H2N
NH
OH
O H2N
O
H N
N
N
N H
O
NH
O
S H N
O N H
O
O N H
O OH
. CH CO H 3 2
O N
N H
OH O
H
NH H2N
NH
Although it is a rare disease, HAE is an autosomal dominant disorder that can be potentially life-threatening when swelling manifests itself within the upper airways. Other target areas such as the face, hands, feet, and abdomen can lead to temporary disfiguration. Attacks may occur without a distinguishable trigger although stress has often been implicated. If left untreated, symptoms usually abate within 2–5 days. The culprit in the flare-up of the disease is increased levels of bradykinin, an endogenous kinin peptide that transmits both pathophysiological and beneficial effects through the G-protein-coupled B2 receptor. A defect or absence of C1 esterase inhibitor is the hallmark of HAE that perpetuates elevated bradykinin levels. Stimulation of the B2 receptor by bradykinin mediates increased vascular permeability, vasodilation, hypotension, and inflammatory pain. Antagonism of the constitutively expressed B2 receptor, therefore, would not only be beneficial in the treatment of HAE
To Market, To Market — 2008
609
but also find application in inflammatory, allergic, and hyperalgesic disorders. Designed to possess enhanced metabolic stability with the inclusion of five non-proteinogenic amino acids, icatibant is a competitive and selective B2 receptor antagonist (devoid of activity against the inducible B1 receptor) with a Ki of 2.270.7 nM, and it is the first downstream targeting agent to receive regulatory approval. It joins purified C1 esterase inhibitor, danazol (an attenuated androgen), and tranexamic acid (an antifibrinolytic agent) as acute and prophylactic treatments for HAE. The peptidomimetic is prepared by standard solidphase technology on Wang resin employing Fmoc chemistry. Following trifluoroacetic acid cleavage with appropriate scavengers and HPLC purification, icatibant is ultimately isolated as its acetate salt. It is formulated as a 30-mg, clear, colorless solution (3-mL volume) in prefilled syringes for subcutaneous injection, preferably in the abdomen, by a medical professional. The recommended dose is one 30-mg injection following an acute HAE attack. In cases of insufficient relief or recurrence of symptoms, a second dose may be administered after 6 h with no more than three injections to be given in a 24-h period. In humans, subcutaneous injection of icatibant results in excellent bioavailability (approximately 90%) with Cmax achieved within 30 min. Regardless of the route of administration (IV or SC), the elimination half-life was 1.2–1.5 h, and although the mechanism of elimination is not known, it is presumed to be metabolized similarly to kinin-related peptides. The efficacy and safety of icatibant were evaluated in two randomized, double-blind phase III studies; one utilized tranexamic acid as the comparator while the other study was placebo-controlled. Of the 130 patients enrolled, 63 received a 30-mg subcutaneous injection of icatibant, 38 patients were given the comparator tranexamic acid, and 29 patients were randomized to placebo. If patients experienced subsequent episodes of HAE, they were treated with icatibant in an open-label extension study along with patients presenting with laryngeal angiodema. With median time to onset of symptom relief as the primary efficacy end point in both studies, icatibant was more efficient than tranexamic acid (2 h compared to 12 h) and placebo (2.5 h compared to 4.6 h). The laryngeal cases in the open-label study experienced similar results with median time to onset of symptom relief ranging from 0.6 to 1.0 h. Almost all patients treated with subcutaneous icatibant experienced some injection site irritation including erythema, swelling, burning, itching, warm sensation, and/or cutaneous pain, but these usually subsided within 1 h. Since metabolism does not involve the CYP450 isozymes, drug–drug interactions are not anticipated. Icatibant is contraindicated in patients with a hypersensitivity to the active substance or any of the excipients. Caution should be observed in patients with ischemic heart disease since antagonism of the B2 receptor
610
Shridhar Hegde and Michelle Schmidt
could result in deterioration of cardiac function and a decrease in coronary blood flow. Also, icatibant may attenuate the neuroprotective effects of bradykinin in the weeks following a stroke, so careful consideration should be given to administration of icatibant to stroke patients.
15. LACOSAMIDE (ANTICONVULSANT) [61–63] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
United States Harris FRC Germany Schwarz Pharma Vimpat 175481-36-4 250.30
H N
H3C O
O N H O CH3
Although epilepsy is a neurological disorder with varying etiology and severity, the common feature is unprovoked, recurring seizures. Whether classified as generalized, involving both cerebral hemispheres, or partial with only localized portions of brain participation at onset, effective treatment relies on accurate assessment of syndrome type to optimally decrease the frequency, duration, and severity of seizures. Although traditional anti-epileptic drugs (AEDs) such as phenobarbital, phenytoin, carbamazepine, and valproate control seizure activity, their effectiveness is hampered by considerable side effects and the tendency for drug–drug interactions. In addition, about a third of patients are refractory to existing therapeutics necessitating elucidation of novel approaches. The latest weapon against partial onset epilepsy is lacosamide, formerly known as harkoseride and erlosamide. Although the precise mechanism of action is not clear, lacosamide stabilizes hyperexcitable neuronal membranes in vitro by the selective enhancement of slow inactivation of voltage-gated sodium channels. The data also indicate that lacosamide binds to collapsing response mediator protein 2 (CRMP2); CRMP2 is involved in neuronal differentiation, control of axonal outgrowth, and possibly epileptogenesis. Furthermore, lacosamide is heralded as having a dual mode of action as it has also displayed efficacy against diabetic neuropathy, possibly as a result of stabilization of neuronal hyperexcitability. An earlier paper in 2004 suggested that lacosamide was a glycinesite NMDA antagonist; however, Errington et al. later demonstrated no significant displacement of radioligands by lacosamide at various binding sites of the NMDA receptor, including the glycine site. Currently, lacosamide is approved as adjunctive treatment of partial onset seizures
To Market, To Market — 2008
611
in patients 17 years or older and is in development as a monotherapy for epilepsy and for neuropathic pain. Its preclinical efficacy as an anticonvulsant was determined using the maximal electroshock (MES) seizure test in mice and rats. Compared to an ED50 for seizure control of 6.5 mg/kg for phenytoin, an ED50 of 4.5 mg/kg was obtained for lacosamide following i.p. dosing. Similarly, an ED50 of 3.9 mg/kg was observed following oral dosing in rats. Lacosamide may be prepared by several routes, but the simplest involves the initial acetylation of D-serine followed by isobutyl chloroformate-mediated condensation with benzylamine. The final product is obtained by a Williamson-type ether synthesis (methylation with methyl iodide and silver oxide). The drug is formulated in tablets of 50, 100, 150, or 200 mg for convenient dose alteration. Lacosamide is typically administered orally at 50 mg twice daily, but the dose may be adjusted at weekly intervals by 100 mg/day up to a daily dose of 200–400 mg/day, based on clinical response and tolerability. When oral administration is temporarily not feasible, an injectable formulation of lacosamide is available for intravenous treatment. The pharmacokinetics, safety, and tolerability of lacosamide were evaluated in healthy patients ranging in age from 18 to 87. Following oral administration, the bioavailability was approximately 100%; food did not affect the rate and extent of absorption. Furthermore, the pharmacokinetics were dose proportional (100–800 mg) with low inter- and intra-patient variability. For doses of 400 and 600 mg of lacosamide, the peak plasma concentrations were 8.7 and 14.3 mg/mL, respectively with Tmax ranging from 1 to 4 h. The corresponding AUC values were 143 and 231 mg h/mL, respectively. The volume of distribution was close to the volume of total body water at 0.6 L/kg, and lacosamide was less than 15% bound to plasma protein. The terminal half-life was 13 h, and lacosamide was primarily eliminated from systemic circulation by renal clearance and biotransformation. After administration of [14C]-lacosamide, approximately 95% of the dose was recovered in the urine with the parent constituting 40%, its O-desmethyl metabolite accounting for 30%, and the remaining percentage composed of a structurally unknown polar component. The major metabolite, O-desmethyl-lacosamide, was also present in the plasma, but it is not pharmacologically active. Although lacosamide is a substrate for CYP2C19, the contribution of other CYP isoforms or non-CYP enzymes in the metabolism of the drug is not clear. Overall, lacosamide appeared safe and was well tolerated at doses of 400 mg or less; however, dose reductions were necessary at higher doses due to adverse CNS-related events. The efficacy of lacosamide was evaluated in three randomized, placebo-controlled clinical trials. All three trials involved the enrollment of patients with a mean duration of partial-onset seizures of 24 years that were not adequately controlled with one or more AEDs. Each study was
612
Shridhar Hegde and Michelle Schmidt
designed with an 8-week baseline phase, and patients were required to have Z4 partial-onset seizures during this period with no seizure-free period exceeding 21 days for inclusion. Following the baseline phase, the first study randomized 418 patients taking one or two concomitant AEDs to placebo or titrated lacosamide (200, 400, or 600 mg). For titration, treatment was initiated at 50 mg twice daily with weekly adjustments of 100 mg/day until the desired dose was reached. This titration phase was followed by a 12-week maintenance phase. The responder rate was defined as a 50% or greater reduction in seizure frequency from baseline to maintenance. Compared to placebo (22%), a significant improvement in responder rate was achieved with lacosamide 400 mg (41%) and lacosamide 600 mg (38%). A second trial evaluated the doses of 200 or 400 mg in 485 patients taking up to three concomitant AEDs. Again, the 400-mg lacosamide dose had a responder rate of 41% compared to placebo at 26%. The third trial enrolled 405 patients on up to three concomitant AEDs with similar responder rates for the two doses tested (38% for the 400-mg dose and 41% for the 600-mg dose). These results were significantly better than placebo (18% responder rate). The most common adverse events were diplopia, headache, dizziness, and nausea. As typical with AEDs, lacosamide may increase the risk of suicidal thoughts or behavior. Patients should, therefore, be monitored for the emergence or worsening of depression. Caution should also be exercised in patients with known conduction problems or severe cardiac disease (myocardial ischemia or heart failure) since dose-dependent prolongations in PR interval have been observed in clinical studies. In this patient population, an ECG before and after completion of drug titration is recommended. Regarding drug–drug interactions, lacosamide did not affect the pharmacokinetics of any of the concomitant AEDs evaluated in the clinical studies and vice versa. Although lacosamide is a CYP2C19 substrate, it had no influence on the pharmacokinetics of omeprazole, a CYP2C19 substrate and inhibitor. Although concomitant omeprazole likewise did not require dose adjustment of lacosamide, it did reduce the formation of the O-desmethyl metabolite by about 60%.
16. METHYLNALTREXONE BROMIDE (OPIOID-INDUCED CONSTIPATION) [64–66] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
United States University of Chicago Canada Wyeth/Progenics Relistor 073232-52-7 436.34
Br-
HO
O
N+ CH 3 OH
O
To Market, To Market — 2008
613
The widespread efficacy of opioids in treating patients with moderate to severe acute and chronic pain is often accompanied by untoward side effects. In particular, opioid-induced bowel dysfunction is one of the more common and debilitating consequences afflicting up to 50% of patients. To counteract the peripherally-mediated adverse effects, opioid antagonists such as naloxone, naltrexone, and nalmephene are sometimes prescribed. Unfortunately, since these agents also penetrate the CNS, they have the potential to reverse analgesia and/or to precipitate opioid withdrawal symptoms. The latest market entry exploits a strategic modification of naltrexone to lower its lipid solubility and increase its polarity: quaternization of the amine of naltrexone by methylation (methyl bromide) prevents crossing of the blood–brain barrier thereby creating an effective peripheral antagonist. Despite a loss of potency upon methylation, methylnaltrexone antagonizes opioid binding at m-opioid receptors with an IC50 of 70 nM. Its affinity for k-opioid receptors is approximately eightfold less (IC50 ¼ 575 nM) with no significant binding to d-opioid, orphanin, or non-opioid receptors. Methylnaltrexone bromide has been approved for the treatment of opioid-induced constipation in patients with advanced illness receiving palliative care. It is administered subcutaneously every other day as needed with no more than a single dose given in a 24-h period. Following subcutaneous injection, methylnaltrexone bromide is rapidly absorbed with Cmax achieved within a half hour. The pharmacokinetics increase dose proportionally with a Cmax of 117 ng/mL at a dose of 0.15 mg/kg, 239 ng/mL at a dose of 0.30 mg/kg, and 392 ng/mL at a dose of 0.50 mg/kg. The corresponding AUC values are 175, 362, and 582 ng h/mL, respectively. At steady state, the volume of distribution is approximately 1.1 L/kg, and the plasma–protein binding ranges from 11 to 15%. Regarding metabolism, methylnaltrexone bromide is eliminated primarily as intact drug (85% based on administered radioactivity) by slightly more renal than hepatic clearance. Although N-demethylation is an issue in rats, it is not a significant pathway of metabolism in humans. Of the five distinct metabolites, none exceeded 6% of administered radioactivity with conversion to methyl 6-naltrexol isomers (5% of total) and methylnaltrexone sulfate (1.3% of total) as most notable. The terminal half-life is around 8 h. The efficacy, tolerability, and safety of methylnaltrexone bromide were evaluated in a phase III study of patients with advanced medical illness. Of the 154 patients enrolled, the majority were cancer patients (80%), but the distribution also included diagnoses of emphysema, end-stage AIDS, and heart failure; all received palliative opioid therapy and were experiencing opioid-induced constipation (Z 48 h with no laxation). The study was designed with a single, double-blind, subcutaneous dose (0.15 or 0.30 mg/kg of methylnaltrexone bromide or placebo) followed by a 4-week, open-label dosing period, and the primary endpoint was the percentage of patients with rescue-free laxation within 4 h of the dose. Compared to placebo (14%), the methylnaltrexone bromide-treated patients had significantly higher percentages of
614
Shridhar Hegde and Michelle Schmidt
laxation. With both treatment groups registering about 60% response, the higher 0.30 mg/kg dose provided no significant advantage. Furthermore, there did not appear to be an effect on opioid pain management, and there were no reports of opioid withdrawal. The most common adverse events were abdominal pain and flatulence followed by nausea, dizziness, and diarrhea. Although methylnaltrexone bromide is a weak inhibitor of CYP2D6, it did not significantly affect the metabolism of the CYP2D6 substrate dextromethorphan. Finally, methylnaltrexone bromide is contraindicated in patients with known or suspected mechanical GI obstruction.
17. PIRFENIDONE (IDIOPATHIC PULMONARY FIBROSIS) [67–70] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
United States Marnac Japan Shinogi Pirespa 53179-13-8 185.22
N
O
H 3C
As intangible as the underlying cause of IPF, an effective treatment has been equally elusive. IPF is characterized initially by dyspnea followed by anomalies in lung function tests, confirmation of distortion of lung architecture by high-resolution computer tomography (CT), and the histological presence of usual interstitial pneumonia (UIP). Morbidity is high with a median survival of only 3 years, barring a lung transplant. Until recently, IPF therapy consisted of a combination of corticosteroids and immunosuppressive agents (azathioprine and cyclophosphamide) to target the inflammation that was believed to be the pathogenic stimulus. Since IPF is now considered to be predominantly a disorder of fibroproliferation, agents that intervene in fibrogenesis have moved to the forefront of treatment options. With demonstrated efficacy in a bleomycin-induced lung fibrosis animal model, pirfenidone has been developed and launched as an approved therapy for IPF. Its antifibrotic activity is derived from the inhibition of p38 MAP kinase that is upstream of transforming growth factor-b (TGF-b), a cytokine implicated in the stimulation of collagen synthesis and the inhibition of its degradation. Pirfenidone also inhibits the expression of TNF-a, IL-1, and ICAM-1, so it possesses the dual benefit of an anti-inflammatory and antifibrotic agent. This relatively simplistic drug is constructed by the copper-catalyzed reaction of 5-methyl-2(1H)-pyridinone with bromo- or chlorobenzene. The pharmacokinetics of pirfenidone were evaluated in healthy volunteers following single (200, 400, or 600 mg) and multiple
To Market, To Market — 2008
615
doses (400 mg tid). Absorption was rapid with a Tmax 0.33–1 h and an elimination half-life of 2–2.5 h. Food prolonged the rate of absorption (Tmax ¼ 1.570.4 h) while peak plasma concentrations were decreased [Cmax ¼ 9.272.9 mg/L (fed) versus 13.071.8 mg/L (fasted)]. The AUC was also affected by food decreasing from 46.6716.8 mg h/L under fasted conditions to 37.4715.4 mg h/L with food. Under fasted conditions, both Cmax and AUC values increased linearly with dose. For patients receiving the 400-mg dose three times daily, the pharmacokinetic parameters were similar to the single dose under fed conditions, indicating no significant accumulation of pirfenidone with multiple dosing. At the higher 400- and 600-mg single doses in the fasted state, the most common adverse effects were GI (nausea and vomiting). Interestingly, no adverse events were reported in the group receiving multiple doses with food over 5 consecutive days, suggesting that concomitant food intake could ameliorate side effects. Although metabolism data are not available in humans, the two major metabolites in animals (mice, horse, and sheep) are derived from oxidation of the methyl moiety to generate hydroxypirfenidone and carboxypirfenidone. Glucuronidation and acetylation of the hydroxyl provides secondary metabolites. The route of metabolism has not been disclosed, so the potential for drug–drug interactions in humans is not clear. With pirfenidone receiving fast-track evaluation by the FDA, it is anticipated that this is not a major concern. In a single center, open-label, phase II trial, pirfenidone (40 mg/kg/day up to 3,600 mg/day) was administered to patients (n ¼ 54) with advanced IPF to evaluate its efficacy and safety. Following a 2-week titration period, the maximum dose of 3,600 mg/day was maintained for 2 years in most of the patients. At 6 months, 29 patients possessed either improved or stabilized lung function; survival was 78% at 1 year and 63% at 2 years. In this study, the common adverse events were nausea, fatigue, and photosensitive skin rash. The latter, experienced by 24% of patients, caused four to discontinue treatment.
18. RILONACEPT (GENETIC AUTOINFLAMMATORY SYNDROMES) [71–73] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no:
United States Regeneron United States Regeneron Arcalyst 501081-76-1
Class:
Recombinant fusion Protein Type: IL-1 trap Molecular weight: 252 kDa Expression system: CHO cells Manufacturer: Regeneron
616
Shridhar Hegde and Michelle Schmidt
The symptoms of inherited autoinflammatory syndromes, which include recurrent fever, joint pain, and rash, are a result of the aberrant signaling of the proinflammatory cytokine IL-1b. Patients are clinically diagnosed by an observed mutation in the NLRP3 gene (nucleotide-binding domain, leucine-rich family, pyrin domain containing 3; also known as NALP3 and CIASI) that leads to the overproduction of the protein cryopyrin. Diseases of this type such as FCAS and MWS are, therefore, also characterized as CAPS. As the central component of the cryopyrin inflammasome, excess cryopyrin leads to stimulation of the inflammasome resulting in the activation of the other major component of the inflammasone, caspase 1, which is responsible for the processing of proIL-1b to its functional form. Rilonacept’s mechanism of intervention in the pathogenesis of CAPS is the blockade of the excessive IL-1b signaling. As an engineered fusion protein, rilonacept combines the extracellular binding domains of the human IL-1 receptor component (IL-1R1) and IL-1 receptor accessory protein (IL-1RAcP) in a single chain, with two of these chains joined to the Fc portion of human immunoglobulin G (IgG) creating a dimeric molecule. Rilonacept serves as an effective soluble IL-1b sink or trap since the coupled receptor components bind IL-1b with higher affinity than either individual receptor. The equilibrium dissociation constants for IL-1b and IL-1a are 0.5 and 1.4 pM, respectively. Compared to other IL-1 antagonists, rilonacept may provide improved efficacy with its offering of dual receptor occupancy; agents containing a single soluble receptor have been hypothesized to potentially induce immune responses since IL-1 exerts its signaling through a multi-component receptor system. Likewise, drugs that target only one of the components, such as the IL-1R1 antagonist anakinra, have been less effective in treating CAPS, possibly due to receptor occupancy issues. With its weekly subcutaneous dosing regimen, rilonacept may have better patient compliance over anakinra that is administered daily. The pharmacokinetics of rilonacept were evaluated in a single-dose phase I study in patients with rheumatoid arthritis (n ¼ 20). The Cmax values increased dose proportionally ranging from 218 ng/mL at the 50 mg/kg SC dose to 1896 ng/mL at the 400 mg/kg SC dose as did terminal half-lives (128 and 182 h, respectively). The safety and efficacy of rilonacept were examined in FCAS and MWS patients in a two-part, sequential, randomized, double-blind, placebocontrolled trial. Part A lasted for 6 weeks, and patients were randomized to receive 160-mg rilonacept subcutaneously, after an initial 320-mg loading dose, or placebo. Part B, which immediately ensued, consisted of a 9-week period where all patients blindly received rilonacept (160 mg weekly) followed by a 9-week, double-blind withdrawal interval in which patients were randomized to continue rilonacept treatment or to receive placebo. At the end of the study, patients were given the option
617
To Market, To Market — 2008
of enrolling in a 24-week rilonacept treatment extension phase. Using a daily questionnaire, patients scored five symptoms of CAPS on a scale of 0 (no severity) to 10 (very severe). The symptoms for rating included joint pain, eye redness/pain, rash, sensation of fever/chills, and fatigue. Using the change from baseline to the end of treatment, a mean symptom score was the primary readout. Compared to placebo, patients receiving rilonacept in part A experienced a greater reduction in the mean symptom score (2.4 for rilonacept compared to 0.5 for placebo). Correspondingly, mean symptom scores increased more in patients randomized to placebo in part B compared to patients remaining on rilonacept (0.9 versus 0.1, respectively). In addition, both serum amyloid A (SAA) and C-reactive protein (CRP) levels, typically elevated in patients with CAPS, decreased versus baseline while there were no changes for patients on placebo. The most common adverse reactions included injection site irritation followed by upper respiratory infection. Although specific drug interaction studies have not been conducted with rilonacept, concomitant administration of other IL-1 blockers and antiTNF-a agents has been associated with an increased risk of serious infection and neutropenia. Also, it is not recommended to co-administer rilonacept with other IL-1 antagonists. Since IL-1 blockade may interfere with the immune response to infections, patients should not initiate rilonacept treatment with an existing infection. As increases in total cholesterol and TGs have been observed in CAPS patients receiving rilonacept, lipid levels should be monitored during the course of treatment, and lipid-lowering adjustments should be made as warranted. Patients with chronic inflammation also tend to have suppressed formation of CYP450 enzymes. Treatment with rilonacept is expected to normalize CYP450 distribution, so plasma levels of co-administered drugs that are CYP450 substrates with narrow therapeutic indexes should be carefully monitored for potential dose modification.
19. RIVAROXABAN (ANTICOAGULANT, VENOUS THROMBOEMBOLISM) [74–75] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
Germany Bayer Canada Bayer/Ortho-McNeil Xarelto 366789-02-8 435.88
O O N
O
H N
S O
N O
Cl
618
Shridhar Hegde and Michelle Schmidt
Venous thromboembolism continues to be a major health concern despite conventional anticoagulation therapies. Of the 900,000 VTE events recorded in 2005, nearly a third of these were fatal, the majority being the result of PE. Intervention by traditional anticoagulants, such as oral antivitamin K, unfractionated heparin, and low-molecular-weight heparins, is by indirect thrombin inhibition and is not without inconvenience (regular monitoring) and side effects. As a convergence point of both the intrinsic and the extrinsic coagulation pathways, factor Xa (FXa) is an attractive target as it catalyzes the conversion of prothrombin to thrombin, blocking the burst of thrombin-mediated activation of coagulation. Rivaroxaban is a recent market introduction that directly inhibits FXa with high potency (Ki ¼ 0.4 nM; IC50 ¼ 0.7 nM) and selectivity, W10,000-fold over other related serine proteases (thrombin, trypsin, plasmin, FVIIa, FIXa, FXIa, urokinase, and activated protein C). From the X-ray crystal structure, the central oxazolidinone moiety anchors the drug through two hydrogen bonds to Gly219 and directs the morpholinone group into the S4 pocket and the chlorothiophene portion into the S1 pocket. These key components may be coupled together synthetically by a couple of routes. Condensation of 3-morpholinone with 4-fluoronitrobenzene followed by catalytic hydrogenation provides N-(p-aminophenyl)morpholinone for subsequent reaction with (S)-2-(phthalimidomethyl)oxirane. With establishment of the aminoalcohol adduct, cyclization with 1,1’-carbonyldiimidazole generates the central oxazolidinone. Deprotection and acylation with 5-chlorothiophene-2-carbonyl chloride affords rivaroxaban. The drug is formulated in 10-mg tablets for once-daily oral administration to patients undergoing elective hip or knee replacement surgery to prevent VTE. Regarding the pharmacokinetic properties, the 10-mg dose has high bioavailability (80%–100%) and reaches Cmax (141 mg/L) within 2–4 h. The AUC is 1020 mg.h/L, and food intake does not affect either AUC or Cmax. At higher doses, dissolution-limited absorption results in decreased bioavailability and decreased absorption rate. Although the steady-state volume of distribution is moderate at approximately 50 L, the plasma protein binding is high (92%–95%). The mean terminal halflife is around 9 h. About two-thirds of the administered dose is metabolized through CYP3A4-, CYP2J2-, and CYP-independent mechanisms, and the metabolites, predominantly derived from morpholino moiety oxidation and amide cleavage, are excreted both renally and hepatically, whereas unchanged parent is eliminated exclusively by renal excretion. The prophylactic efficacy and safety of rivaroxaban were evaluated in patients undergoing orthopedic surgery. One study focused on the prevention of thrombosis in 621 patients undergoing elective knee replacement. Initiated 6–8 h postsurgery, patients were randomized to oral rivaroxaban (2.5, 5, 10, 20, and 30 mg b.i.d.) or subcutaneous enoxaparin (30 mg b.i.d.) at 12–24 h postsurgery. Treatment was
To Market, To Market — 2008
619
continued for 5–9 days after surgery until bilateral venography was performed. At all doses of rivaroxaban, the rate of symptomatic VTE (symptomatic DVT, non-fatal PE, and VTE-related death) was lower than in patients treated with enoxaparin. Regarding safety, there was no statistical difference in the incidence of major postoperative bleeding between any of the rivaroxaban dose groups and enoxaparin although there did appear to be a dose dependency in the rivaroxaban set. In addition to bleeding and subsequent posthemorrhagic anemia, presenting as weakness, paleness, asthenia, dizziness, headache, or unexplained swelling, other common adverse events included nausea, increased GGT, and an increase in transglutaminase. Owing to its mechanism of action, there is a bleeding risk, so the drug is contraindicated in patients with clinically active bleeding. Rivaroxaban is also contraindicated in pregnant and breast-feeding women and in patients with hepatic disease associated with coagulopathy and clinically relevant bleeding risk. As a CYP3A4 substrate, plasma levels may be reduced with concomitant use of strong CYP3A4 inducers (rifampicin, phenytoin, carbamazepine, phenobarbital, or St. John’s Wort). Conversely, concomitant administration of strong CYP3A4 and P-glycoprotein inhibitors result in increases in AUC and Cmax that could increase the bleeding risk. The use of ketoconazole, other azole-antimycotics, and HIV protease inhibitors is, therefore, not recommended with rivaroxaban treatment and vice-versa. Owing to increased bleeding risk, caution is also advised in the coadministration of other anticoagulants, NSAIDS, or platelet aggregation inhibitors.
20. ROMIPLOSTIM (ANTITHROMBOCYTOPENIC) [76–78] Country of origin: United States Originator: Amgen First introduction: United States Introduced by: Trade name: CAS registry no:
Amgen Nplate 267639-76-9
Class:
Recombinant fusion protein Type: TPO receptor agonist Molecular weight: 60 kDa Expression system: E. coli Manufacturer: Amgen
Romiplostim has been developed and launched for the treatment of thrombocytopenia in patients with ITP, an autoimmune blood disorder in which there is autoantibody-mediated platelet destruction. Coating of the platelets by autoantibodies results in accelerated ingestion by macrophages. In addition to the destruction component, ITP is also characterized by impaired platelet production. The primary physiological regulator of platelet production is thrombopoietin (TPO), a growth factor that orchestrates its effects through the TPO (Mpl) receptor.
620
Shridhar Hegde and Michelle Schmidt
Agonists of this receptor could, thereby, remediate the platelet production contribution of ITP. By binding to and activating the TPO receptor, romiplostim serves as a TPO receptor agonist. As a recombinant fusion protein, in this case coined as a peptibody, it was engineered to contain two identical single-chain subunits, each consisting of a TPO-binding domain linked to the C-terminus of a human IgG1 Fc domain designed to increase the half-life of the protein. Although it binds to the TPO receptor, romiplostim has no sequence homology to endogenous TPO, which should mitigate the risks encountered with the first generation predecessor, recombinant megakaryocyte growth and development factor (MGDF); this truncated, non-glycosylated form of TPO conjugated to PEG was halted during clinical studies because of the immunogenicity of PEG-MGDF resulting in cross-reacting (neutralizing) antibodies against endogenous TPO and subsequent severe, unrelenting thrombocytopenia. Romiplostim also avoids the side effects associated with the generalized immunosuppressive agents previously employed for the treatment of ITP and may prevent splenectomy, another common treatment option. In clinical studies, dose-dependent increases in platelet count were achieved with romiplostim in the range of 1–10 mg/kg delivered subcutaneously. Over a 2- to 3-week period, peak platelet count was 1.3–14.9 times greater than baseline. In a dose range of 0.3–10 mg/kg administered intravenously, the pharmacokinetics were non-linear. The plasma half-life was 1.50, 2.41, and 13.8 h for doses of 0.3, 1.0, and 10 mg/kg, respectively. The AUC increased from 0.964 to 1530 mg h/mL, whereas the maximum serum concentrations increased from 2,810 to 210,000 pg/mL. The Tmax ranged from 13 to 15 days. Following a subcutaneous dose, absorption appeared to be slow with a peak serum concentration of 2.0 mg/kg occurring between 24 and 36 h. The efficacy and safety of romiplostim were evaluated in two double-blind, placebocontrolled studies encompassing 125 patients with chronic ITP. For enrollment, patients were required to complete at least one prior treatment (corticosteroids, Igs, or rituximab) and have had a platelet count of r30 109 per liter. Patients were randomized to receive either romiplostim (1 mg/kg subcutaneously) once a week or placebo for a duration of 24 weeks. Individual dose adjustments were permitted to maintain platelet counts between 50 109 and 200 109 per liter. Patients were further segregated into groups based on whether they had undergone a previous splenectomy. In the non-splenectomized group, 61% (25 of 41) of patients receiving romiplostim achieved the primary efficacy endpoint of durable platelet response of a weekly platelet count Z50 109 per liter for any 6 of the last 8 weeks of the 24-week treatment period. Only one of the 21 patients in the placebo group reached the primary efficacy endpoint. In the splenectomized group, 38% (16 of 42) of patients receiving romiplostim achieved a durable platelet response compared to none of the 21 patients in the placebo group. The most
621
To Market, To Market — 2008
common adverse events (W5%) reported in these two clinical studies were headache, arthralgia, dizziness, insomnia, myalgia, pain in extremity, abdominal pain, shoulder pain, dyspepsia, and paresthesia. Of the patients with positive antibodies to romiplostim or to TPO, only one patient had neutralizing activity to romiplostim, and none had neutralizing activity to TPO. When initiating romiplostim treatment, a 1-mg/kg dose is recommended once weekly subcutaneously. The dose may be adjusted weekly in increments of 1 mg/kg to achieve and maintain a platelet count Z50 109 per liter, but the maximum weekly dose should not exceed 10 mg/kg. Romiplostim increases the risk for the development of reticulin fiber deposition within the bone marrow, so patients should be monitored for this condition and discontinue use if detected. Romiplostim should only be used in pregnant and breastfeeding women when the potential benefit justifies the potential risk. Formal drug interaction studies have not been conducted.
21. SITAFLOXACIN HYDRATE (ANTIBACTERIAL) [79–82] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
Japan Daiichi Sankyo Japan Daiichi Sankyo H 2N Gracevit 127254-12-0 409.81 (+ H2O ¼ 427.83)
O
O
F
OH
N
.H2O
N Cl F
Patients in Japan have a new treatment option to combat respiratory and urinary tract infections. Sitafloxacin hydrate is the newest member (fourth generation) of the fluoroquinolone family of antibiotics that exhibits broad spectrum activity against many Gram-positive, Gramnegative, and anaerobic clinical isolates, including strains resistant to other fluoroquinolones. Since the launch of the first fluoroquinolone norfloxacin (patented in 1978), 20 other versions have made it to market with ciprofloxacin and levofloxacin experiencing the most prevalent usage. The mechanism of action involves inhibition of bacterial type II topoisomerases, both DNA gyrase and topoisomerase IV. By inhibiting these enzymes and preventing DNA supercoiling, cell division is disrupted leading to cell death. In Gram-negative bacteria, the primary target appears to be gyrase, whereas topoisomerase IV is involved in Gram-positive bacteria. Dual inhibition is attractive for widespread activity and avoidance of resistance as both encoding genes would have to acquire mutations. The IC50 (mg/L) against DNA gyrase from E. coli is 0.13 and from S. pneumoniae is 1.16. Against topoisomerase IV, the IC50
622
Shridhar Hegde and Michelle Schmidt
(mg/L) is 0.39 from S. aureus and 1.88 from S. pneumoniae. Compared to the antimicrobial activities of ciprofloxacin, sparfloxacin, levofloxacin, cefotaxime, imipenem, and vancomycin against nine penicillinsusceptible, -intermediate, and -resistant pneumococci, sitafloxacin had the lowest MIC value (0.064 mg/mL) and the most rapid killing among all the compounds tested. The most challenging aspect of the synthesis of sitafloxacin was the construction of the cis-oriented (1R,2S)-2-fluorocyclopropylamine moiety. One of the first explored routes exploited the diastereoface selective cyclopropanation of (4R,5S)-4,5-diphenyl-3vinyl-2-oxazolidinone with zinc-monofluorocarbenoid followed by hydrogenolytic deprotection. After a thorough investigation of five synthetic routes, however, the preferred method involved diazonium coupling, separation of volatile stereoisomers by distillation, and eventually chiral HPLC. Although the 8-chloro substituent of the quinolone core appears to be essential for the enhanced activity, it may also contribute to the phototoxicity potential of sitafloxacin. Following oral administration, the pharmacokinetic parameters were evaluated in healthy Japanese men (n ¼ 36). Both Cmax and AUC were approximately proportional to dose. The Cmax increased from 0.29 mg/L at the 25-mg dose to 0.51 mg/L at 50 mg, 1.00 mg/L at 100 mg, and 1.86 mg/L at 200 mg. The corresponding AUC values ranged from 1.52 mg.h/L at the 25-mg dose to 2.62 mg. h/L at 50 mg, 5.55 at 100 mg, and 12.03 at 200 mg. Regardless of dose, Tmax was about 1.2 h, and the terminal half-life was 4.5–5 h. The volume of distribution ranged from 1.46–1.88 L/kg, suggesting reasonable tissue penetration, and sitafloxacin was approximately 50% protein-bound. No drug accumulation was observed upon multiple dosing. The oral bioavailability of sitafloxacin was 89%, and food did not significantly affect rate and extent of absorption. Since the primary route of excretion was renal, with 99% of the drug being eliminated intact, dose adjustment may be necessary in patients with renal deficiency. Glucuronidation of sitafloxacin by UGT1A1 was observed in human liver microsomes, but no clinical data is available. Sitafloxacin does not appear to interact with the cytochrome P450 system, so the potential of interactions with other drugs metabolized by this system is minimal. In a Phase II open-label study in patients with severe infections caused by MRSA or vancomycin-resistant Enterococcus faecium (VRE), all but one of the patients had failed previous antibiotic therapy. Sitafloxacin was effective in 4 of 11 of MRSA patients, and 5 of 9 patients with VRE were deemed clinically cured. In another phase II randomized, openlabel, multi-center study, sitafloxacin (400 mg once daily) was compared to imipenem/cilastatin (500 mg three times daily) for the treatment of pneumonia in South Africa (69 hospitalized patients). Both treatment groups had a W90% cure rate with 90–95% of sitafloxacin patients having a satisfactory bacteriologic response at the follow-up assessment versus 100% for the comparator. As a class, the major adverse events for fluoroquinolones are cardiac arrhythmia (due to QT interval
To Market, To Market — 2008
623
prolongation), major phototoxicity, CNS disturbances (seizures, dizziness, and headaches), and tendonitis. The GI side effects, common to most antimicrobials, include Clostridium difficile-associated diarrhea (CDAD) and alterations in glucose homeostasis. From the clinical safety profile of sitafloxacin (1,059 patients receiving either 50 mg b.i.d. or 100 mg b.i.d.), about a third of patients experienced an adverse event with the most common being diarrhea, liver enzyme elevations, and headaches; however, the risk of QT prolongation, hypoglycemia, and hepatotoxicity were all considered to be low. Phototoxicity appears to be the limiting toxicity, particularly in non-Asian patients. Although doses of 50 or 200 mg b.i.d. did not cause clinically significant phototoxicity compared to placebo in Asian patients, sitafloxacin administered at a dose of 100 mg b.i.d. in Caucasian volunteers showed mild ultraviolet 1-dependent phototoxicity that was maximal at 24 h with normalization by 24 h after drug cessation. Although currently only approved for use in Japan, future use in Caucasians may be relegated to hospital settings where the risk for phototoxicity is diminished.
22. SUGAMMADEX (REVERSAL OF NEUROMUSCULAR BLOCKADE) [83–87] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: CAS registry no: Molecular weight:
Netherlands Organon Sweden Schering-Plough Bridion 343306-71-8 (free acid) 343306-79-6 (sodium salt) 2178.01 (sodium salt)
+Na-O
S
O
O O
O
O O
+Na-O
OH HO OH OH
S O O
O O OH HO
OH
O
O OH
S O
OH OH O
+Na-O
S O
O
HO
HO OH HO
O
O
O
O S
O-Na+
S
HO
O
O-Na+
S
HO
O
+Na-O
O-Na+
S
O O O-Na+
624
Shridhar Hegde and Michelle Schmidt
To facilitate tracheal intubation, mechanical ventilation, and surgical access, NMB agents are frequently used as non-anesthetic adjuncts in surgical procedures. Classified as either depolarizing or non-depolarizing, non-depolarizing blockers prevent cell membrane depolarization by serving as an antagonist of the nicotinic acetylcholine receptor at the neuromuscular junction, and clinically, non-depolarizing agents are preferable due to fewer side effects. Since residual blockade at the end of surgery is typical, it is essential to reverse the effect to ensure full neuromuscular recovery. Acetylcholinesterase inhibitors have traditionally been used as antagonists of non-depolarizing neuromuscular blockade, but their side effects and limitations, not effective against the depolarizing class, validate the search for improved reversal agents. As a g-cyclodextrin, sugammadex provides a novel mechanism of action employing encapsulation or chelation of the residual blocking drug. With the common non-depolarizing blocker rocuronium, the pair forms an extremely tight 1:1 host–guest complex with an association constant of approximately 107 M1. Sugammadex is able to function as a pharmacologic sink of rocuronium and vecuronium, another non-depolarizing neuromuscular blocker, without the cardiovascular adverse effects experienced with reversal agents that directly interact with the cholinergic system. The g-cyclodextrin has been designed to enhance binding of the guest by incorporating acidic side chains to promote an electrostatic interaction with the positive nitrogen of the blocker. Starting with g-cyclodextrin, these side chains are readily installed by first halogenating with iodine or bromine to provide a handle for nucleophilic displacement with either 3-mercaptopropionic acid in the presence of sodium hydride or with 3-mercaptopropionic acid methyl ester and cesium carbonate. The latter requires hydrolysis with sodium hydroxide to generate sugammadex sodium. The drug is formulated as a 100 mg/mL solution for intravenous injection with a pH between 7 and 8 and osmolality between 300 and 500 mOsm/kg. In volunteers who had not received a NMB agent, 0.1–8.0 mg/kg doses of sugammadex resulted in a clearance rate of 88 mL/min, an elimination half-life of 1.8 h, and a volume of distribution of 11–14 L. About 75% of the dose was excreted intact in the urine; metabolism was virtually non-existent. The complex with rocuronium also undergoes rapid urinary excretion with 70% excreted in 6 h and 90% within 24 h. After the administration of 4–8 mg/kg, the renal elimination of rocuronium was increased by more than 100% over 24 h. The safety and dose–response relationship of sugammadex were evaluated in 30 patients who had at least a 2.5-h anesthesia duration. Following a randomization of sugammadex doses of 0.5–6.0 mg/kg to reverse neuromuscular blockade, there was a dose dependency to recovery time with the fastest recovery reached in 1.6 min. In another study, sugammadex reversal of rocuronium-induced
625
To Market, To Market — 2008
neuromuscular blockade was compared to the reversal of cis-atracuriuminduced blockade by neostigmine. A 2 mg/kg dose of sugammadex provided faster reversal of neuromuscular blockade than the 50 mg/kg dose of neostigmine (1.9 min compared to 7.2 min, respectively). In both studies, sugammadex was well tolerated. The most common adverse effects included coughing, movement, vomiting, dry mouth, dysgeusia, hypotension, parosmia, and a sensation of changed temperature. Reversal of neuromuscular blockade while anesthesia is continued may obviate the need to provide additional doses of anesthesia to maintain this state. Although formal drug interaction studies have not been conducted, modeling suggests that certain concomitant drugs could cause displacement of rocuronium or vecuronium from sugammadex, such as toremifene, fusidic acid, and high-dose flucloxacillin, resulting in significantly longer recovery times. In contrast, sugammadex can affect the efficacy of hormonal contraceptives; administration of a bolus dose of sugammadex is considered to be equivalent to missing one daily dose of the oral contraceptive. In this situation, the patient should follow the advice given in the package insert of the contraceptive regarding a missed dose. Sugammadex is recommended for the reversal of neuromuscular blockade induced by rocuronium or vecuronium; it should not be used with other steroidal blocking agents since safety and efficacy data are not available for these situations. It should also not be partnered with non-steroidal neuromuscular blockers such as succinylcholine or benzylisoquinolinium-based drugs. Since sugammadex is predominantly cleared by the kidneys, care should be taken when administering this reversal agent in patients with renal impairment.
23. TAFLUPROST (ANTIGLAUCOMA) [88–90] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no: Molecular weight:
Japan HO Santen/Asahi Glass Denmark Santen/Asahi Glass Taflotan HO 209860-87-7 452.53
O O O
CH3 CH3
F F
Glaucoma is second only to cataracts as a causative factor of blindness. By 2010, it is estimated that approximately 60 million people worldwide will be afflicted by glaucoma, so effective treatments should garner a large market. Although no cure for glaucoma exists, blindness may be prevented with early detection and treatment. Diagnosis is typically
626
Shridhar Hegde and Michelle Schmidt
made by the measurement of IOP, and as such, treatment usually involves reduction in IOP to prevent further damage and to maintain remaining vision. PG analogs have been widely used for lowering IOP by increasing uveoscleral outflow through agonism of the prostanoid FP receptor, and currently marketed versions include latanoprost, unoprostone isopropyl ester, bimatoprost, and travoprost. As a first-line therapy, latanoprost is not devoid of side effects (conjunctival hyperemia, irritation, and headache), so next generation agents should differentiate on minimizing these events. Not only is the recently launched tafluprost a stronger IOP-lowering agent than latanoprost, it also boasts fewer and milder local side effects. Compared to the carboxylic acid of latanaprost (Ki ¼ 4.7 nM), the carboxylic acid of tafluprost displayed a 10-fold greater affinity for the prostanoid FP receptor (Ki ¼ 0.4 nM). The synthesis of tafluprost begins with a Wittig condensation of the protected bicyclic lactone carbaldehyde with a dimethyl phosphonate ketone derivative. The bottom appendage is then completed by the fluorination of the ketone with morpholino-sulfur trifluoride. Hydrolysis of the benzoate ester protecting group liberates the hydroxy group, and reduction of the lactone is accomplished with aluminum hydride to generate the lactol. Condensation of this intermediate with the phosphonium salt of the acid side chain generates the free acid, or active ingredient, which is subsequently esterified with 2-iodopropane in the presence of DBU. The isopropyl ester pro-drug is formulated as a 15 mg/mL, preservative-free solution, a first for an ocular PG drug, to be administered as a single drop (0.0015%) once daily in the affected eye(s). Previously, there was a perception that a preservative, such as benzalkonium chloride, was necessary for ocular penetration and respectable drug activity. A randomized, single-center, cross-over phase I study proved that for tafluprost there was no statistical significance in pharmacokinetic parameters between preserved and preservative-free formulations. Both formulations were administered once daily for a duration of 8 days in 16 healthy volunteers with a washout of at least 4 weeks. Evaluating the free acid active metabolite, AUC values were 581.17529.9 (preservative) versus 431.97457.8 (preservative-free) pg/min/mL, and Cmax were 31.4719.5 (preservative) and 26.6718.0 (preservative-free) pg/mL. The Tmax was 10 min for both formulations, and plasma levels were low at all time points. Systemic bioavailability was comparable for both, and both formulations were cleared rapidly from the circulation, presumably through typical PG mechanisms. Although the incidence of ocular hyperemia was similar for both formulations, the severity was deemed moderate for preserved and mild for preservative-free tafluprost. The efficacy and safety of tafluprost were evaluated in several clinical studies. In a 6-month study compared to timolol 0.5% twice daily, tafluprost 0.0015% once daily was at least as effective with IOP reductions of
To Market, To Market — 2008
627
5–7 mmHg for tafluprost and 4–6 mmHg for timolol. Another 6-month study compared tafluprost 0.0015% once daily with latanoprost 0.005% once daily. IOP reductions were similar (6–8 mmHg for tafluprost versus 7–9 mmHg for latanoprost). As seen in the phase 1 study, hyperemia was the most common adverse event. Tafluprost is recommended as monotherapy in patients intolerant to first-line treatment or as adjunct to beta-blockers. It is anticipated to have a clinical benefit for patients who might experience preservative-related side effects, particularly those patients with dry or sensitive eyes. The drug should not be used in women of child-bearing age unless adequate contraceptive measures are taken. Regarding drug interactions, formal studies have not been conducted.
24. THROMBIN ALFA (HEMOSTAT) [91–92] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no:
United States ZymoGenetics United States ZymoGenetics Recothrom 869858-13-9
Class: Type: Molecular weight: Expression system: Manufacturer:
Recombinant human protein Factor IIa agonist 33.8 kDa CHO cells ZymoGenetics
Thrombin alfa is a recombinant human thrombin, identical in amino acid sequence and structure to the endogenous protein, which has been launched for use as a locally applied hemostat in surgical procedures. When applied topically to a site of bleeding, the highly specific serine protease promotes blood clot formation by activating platelets and catalyzing the conversion of fibrinogen to fibrin. It is indicated as an aid to hemostasis when standard techniques are ineffective or impractical, and it may be used in conjunction with an absorbable gelatin sponge. The coagulation protein is supplied as a lyophilized powder that is reconstituted with sterile 0.9% sodium chloride to yield a solution containing 1,000 IU/mL before use. The efficacy, safety, and immunogenicity of thrombin alfa were evaluated in a multiple-site, randomized, double-blind phase III study involving 411 patients undergoing one of the four following surgeries — spinal, hepatic resection, peripheral arterial bypass, and arteriovenous graft formation for hemodialysis access. Patients were randomized to either thrombin alfa or bovine thrombin, both in combination with an absorbable gelatin sponge, and efficacy was assessed by the incidence of hemostasis within 10 min. Thrombin alfa displayed comparable efficacy to bovine thrombin with 95% of patients in each group achieving the primary efficacy endpoint.
628
Shridhar Hegde and Michelle Schmidt
Adverse events were similar in both groups with incision site complication being the most common complaint followed by procedural pain and nausea. Regarding immunogenicity, thrombin alfa had a statistically lower incidence of product-specific antibody production; three of 198 patients in the thrombin alfa group developed specific antithrombin product antibodies compared to 43 of 200 patients in the bovine thrombin set that developed antibodies against bovine thrombin. Although none of the antibodies in the thrombin alfa group neutralized native human thrombin, the antibodies against bovine thrombin were not tested for neutralization. No adverse events such as excessive bleeding were directly associated with the development of antibodies in either group. Thrombin alfa is contraindicated for the treatment of massive or brisk arterial bleeding. It should not be directly injected into the circulatory system as this would create a potential risk of thrombosis if absorbed systemically. Thrombin alfa should not be administered to patients with known hypersensitivity to this agent, any of its excipients, or hamster proteins. There may also be a potential for allergic reaction in patients with known hypersensitivity to snake proteins. Since reproduction studies have not been conducted with thrombin alfa, it should only be administered to pregnant women if clinically required. Finally, drug interaction studies have not been performed with thrombin alfa.
25. THROMBOMODULIN (RECOMBINANT) (ANTICOAGULANT) [93–95] Country of origin: Originator: First introduction: Introduced by: Trade name: CAS registry no:
Japan Asahi Kasei Pharma Japan Asahi Kasei Pharma Recomodulin 112414-64-9
Class:
Recombinant human glycoprotein Type: Antithrombotic Molecular weight: 62 kDa Expression system: CHO Cells Manufacturer: Asahi Kasei Pharma
Recombinant thrombomodulin, an anticoagulant, was approved and launched in Japan last year for the intravenous treatment of DIC. DIC is a blood clotting disorder in which the blood coagulation system becomes overactive, leading to blockage of capillaries, ischemia and other damage to vascular endothelial cells, and organ damage. Concurrently, the use and subsequent depletion of platelets and coagulation proteins resulting from the ongoing coagulation may induce severe bleeding. DIC frequently occurs as a complication of malignant tumors and infections. Bleeding is typically the presenting symptom in patients with DIC,
To Market, To Market — 2008
629
a factor that can complicate decisions about treatment. Thrombomodulin is an endothelial cell membrane glycoprotein that neutralizes thrombin pro-coagulant activity and accelerates the thrombin-catalyzed activation of anticoagulant protein C, thus allowing potential use in the treatment of thrombotic disorders such as DIC. Thrombomodulin contains 557 amino acids composed of an N-terminal lectin-like domain, six epidermal growth factor–like domains, an O-glycosylation site domain, a transmembrane domain, and a cytoplasmic domain. The last three epidermal growth factor–like domains of thrombomodulin are essential for thrombin binding and protein C activation. Recombinant thrombomodulin is a negatively charged soluble form of human thrombomodulin comprising all its extracellular domains, including the three epidermal growth factor–like structure domains that are essential for thrombin binding. It consists of 498 amino acids and sugar chains and has a molecular weight of about 62,000. It is produced by CHO cells containing cDNA encoding human soluble thrombomodulin inserted in the expression vector. Recombinant thrombomodulin binds to thrombin at its fibrinogen recognition site. The resultant thrombin/thrombomodulin complex is unable to generate fibrin or activate platelets, but instead becomes a potent activator of protein C. The activated form of protein C is an anticoagulant protease that selectively inactivates coagulation factors Va and VIIIa, providing an essential feedback mechanism to prevent excessive coagulation. Thrombomodulin is a reversible and direct inhibitor of thrombin, with an ability to inhibit both free and clotbound thrombin. Other anticoagulants such as heparin derivatives inhibit thrombin activity indirectly through the anti-thrombin III pathway, which is less effective on clot-bound thrombin than on free thrombin. In in vitro assays, recombinant thrombomodulin inhibits fibrin formation with an IC50 of 1.6 mg/mL and Ki of 22 nM. The apparent Kd for thrombin is about 4 nM. In human pharmacokinetic studies in healthy volunteers, single-dose administration of intravenous recombinant thrombomodulin at doses of 0.03, 0.1, and 0.3 mg resulted in Cmax range of 10–122 ng/mL, with Tmax of about 2 h. The Cmax and the AUC showed a linear and dose-dependent pattern. The plasma half-lives were 2.92–3.97 h (t1/2a) and 17.75–20.48 h (t1/2b). The elimination pattern had good linearity, and more than 50% of the dose was excreted in urine within 48 h. In a multiple-dose study, recombinant thrombomodulin (0.2 mg intravenous) was infused over 2 h once daily for 3 days. The plasma t1/2 and total clearance after each dose did not significantly differ from those assessed after a single dose, indicating that there is no accumulation of recombinant thrombomodulin in plasma. A phase III, randomized, parallel-group trial compared the efficacy and safety of recombinant thrombomodulin to those of low-dose heparin for the treatment of DIC associated with hematological malignancy or
630
Shridhar Hegde and Michelle Schmidt
infection. DIC patients (n ¼ 234) were assigned to receive recombinant thrombomodulin (0.06 mg/kg for 30 min, once daily) or heparin sodium (8 U/kg/h for 24 h) for 6 days. The primary efficacy endpoint was DIC resolution rate. The secondary endpoints included clinical course of bleeding symptoms and mortality rate at 28 days. DIC was resolved in 66.1% of the recombinant thrombomodulin group, as compared with 49.9% of the heparin group. Recombinant thrombomodulin was also superior to heparin in terms of the clinical course of bleeding symptoms and the disappearance rate of bleeding symptoms (35.2% versus 20.9%). Bleeding-related adverse events occurred in 43.1% and 56.5% of patients given recombinant thrombomodulin and heparin, respectively, up to 7 days after treatment initiation. The difference in the reduction of mortality at 28 days between the two groups was statistically not significant.
REFERENCES [1] The collection of new therapeutic entities first launched in 2008 originated from the following sources: (a) Integritys, Prous Science database. (b) Iddb, Current Drugs database. (c) IMS R&D Focus. (d) Adis Business Intelligence R&D Insight. (e) Pharmaprojects. (f) Package Inserts of individual NMEs. [2] A. I. Graul, L. Revel, M. Barrionuevo, E. Cruces, E. Rosa, C. Verge´s, B. Lupone, N. Diaz and R. Castanˇer, Drug News Perspect., 2009, 22, 7. [3] B. Hughes, Nat. Rev. Drug Discov., 2009, 8, 93. [4] M. P. Curran, G. W. Robins, L. J. Scott and C. M. Perry, Drugs, 2008, 68, 2011. [5] C. P. Delaney, U. Yasothan and P. Kirkpatrick, Nat. Rev. Drug Discov., 2008, 7, 727. [6] P. Neary and C. P. Delaney, Expert Opin. Invest. Drugs, 2005, 14, 479. [7] J. B. Leslie, Drugs Today, 2007, 43, 611. [8] J. A. Werner, L. R. Cerbone, S. A. Frank, J. A. Ward, P. Labib, R. W. Tharp-Taylor and C. W. Ryan, J. Org. Chem., 1996, 61, 587. [9] J. Steffel, F. R. Eberli, T. F. Lu¨scher and F. C. Tanner, Ann. Med., 2008, 40, 242. [10] S. Windecker, P. W. Serruys, S. Wandel, P. Buszman, S. Trznadel, A. Linke, K. Lenk, T. Ischinger, V. Klauss, F. Eberli, R. Corti, W. Wijns, M. Morice, C. di Mario, S. Davies, R. van Geuns, P. Eerdmans, G. van Es, B. Meier and P. Ju¨ni, Lancet, 2008, 372, 1163. [11] E. Grube and L. Buellesfeld, Exp. Rev. Med. Devices, 2006, 3, 731. [12] R. E. Betts, D. R. Savage and J. E. Shulze, US Patent 0131008, 2005. [13] M. Isozaki and T. Kawanishi, US Patent 7193078, 2007. [14] M. Oka and K. Hino, Drugs Future, 1992, 17, 9. [15] M. Oka, Y. Noda, Y. Ochi, K. Furukawa, T. Une, S. Kurumiya, K. Hino and T. Karasawa, J. Pharmacol. Exp. Ther., 1993, 264, 158. [16] Y. Noda, S. Kurumiya, Y. Miura and M. Oka, J. Pharmacol. Exp. Ther., 1993, 265, 745. [17] C. E. Heading, Curr. Opin. CPNS Investig. Drugs, 2000, 2, 79. [18] L. A. Sorbera, J. Castan˜er and R. M. Castan˜er, Drugs Future, 2005, 30, 11. [19] M. G. P. Page, Expert Opin. Drug Discov., 2007, 2, 115. [20] S. D. Anderson and J. G. Gums, Ann. Pharmacother., 2008, 42, 806. [21] J. L. Del Pozo and R. Patel, Drugs Today, 2008, 44, 801. [22] V. V. Kaushik and R. J. Moots, Expert Opin. Biol. Ther., 2005, 5, 601. [23] L. A. Sorbera and P. A. Leeson, Drugs Future, 2005, 30, 1087.
To Market, To Market — 2008
631
[24] G. Y. Melmed, S. R. Targan, U. Yasothan, D. Hanicq and P. Kirkpatrick, Nat. Rev. Drug Discov., 2008, 7, 641. [25] W. J. Sandborn, B. C. Feagan, S. Stoinov, P. J. Honiball, P. Rutgeerts, D. Mason, R. Bloomfield and S. Schreiber, N. Engl. J. Med., 2007, 357, 228. [26] T. Filippatos and H. J. Milionis, Expert Opin. Investig. Drugs, 2008, 17, 1599. [27] A. C. Goldberg, H. E. Bays, C. M. Ballantyne, M. T. Kelly, S. M. Buttler, C. M. Setze, D. J. Sleep and J. C. Stolzenbach, Am. J. Cardiol., 2009, 103, 515. [28] S. M. Mohiuddin, C. J. Pepine, M. T. Kelly, S. M. Buttler, C. M. Setze, D. J. Sleep and J. C. Stolzenbach, Am. Heart J., 2009, 157, 195. [29] R. D. Davis, R. N. Fitzgerald and J. Guo, Synthesis, 2004, 1959. [30] I. Noviawaty, G. Uzun and A. I. Qureshi, Expert Opin. Pharmacother., 2008, 9, 2519. [31] L. A. Sorbera and J. Castan˜er, Drugs Future, 2004, 29, 105. [32] J. Varon and J. E. Angeles, Therapy, 2008, 5, 751. [33] J. Varon, Drugs, 2008, 68, 283. [34] A. Mattson, C. Svensson, K. Thornblom and C. Odman, WO Patent 00031035, 2000. [35] L. A. Sorbera, J. Bozzo and J. Castan˜er, Drugs Future, 2005, 30, 877. [36] B. I. Eriksson, H. Smith, U. Yasothan and P. Kirkpatrick, Nat. Rev. Drug Discov., 2008, 7, 557. [37] M. Sanford and G. L. Plosker, Drugs, 2008, 68, 1699. [38] N. H. Hauel, H Nar, H. Priepke, U. Ries, J.-M. Stassen and W. Wienen, J. Med. Chem., 2002, 45, 1757. [39] L. A. Sorbera, J. Bolos and N. Serradell, Drugs Future, 2006, 31, 304. [40] B. A. Sproule, M. Hazra and B. G. Pollock, Drugs Today, 2008, 44, 475. [41] F. W. Lohoff and K. Rickels, Expert Opin. Pharmacother., 2008, 9, 2129. [42] M. A. Spopko, Jr., M. J. Ehret and M. Grgas, Ann. Pharmacother., 2008, 42, 1439. [43] L. P. H. Yang and G. L. Plosker, CNS Drugs, 2008, 22, 1061. [44] R. Haubrich, S. Gubernick, U. Yasothan and P. Kirkpatrick, Nat. Rev. Drug Discov., 2008, 7, 287. [45] M. C. Michel, Expert Opin. Pharmacother., 2008, 9, 3083. [46] E. D. Deeks and G. M. Keating, Drugs, 2008, 68, 2357. [47] S. L. Davies, J. Castan˜er, J. S. Silvestre and M. Baye´s, Drugs Future, 2005, 30, 462. [48] B. M. Knoll, S. Vento and Z. Temesgen, Drugs Today, 2008, 44, 23. [49] M. C. Michel, Expert Opin. Pharmacother., 2008, 9, 1787. [50] L. A. Sorbera, J. Castan˜er and P. A. Leeson, Drugs Future, 2003, 28, 647. [51] R. Browne and M. Kilkelly, WO Patent 07137799, 2007. [52] V. W. Nitti, R. Dmochowski, K. P. Sand, H.-T. Forst, C. Haag-Molkenteller, U. Massow, J. Wang, M. Brodsky and T. Bavendam, J. Urol., 2007, 178, 2488. [53] P. Ney, R. K. Pandita, D. T. Newgreen, A. Breidenbach, T. Sto¨hr and K.-E. Andersson, BJU Int., 2008, 101, 1036. [54] C. R. Chapple, P. E. van Kerrebroeck, K.-P. Junemann, J. T. Wang and M. Brodsky, BJU Int., 2008, 102, 1128. [55] S. J. P. van Belle and V. Cocquyt, Expert Opin. Pharmacother., 2008, 9, 3261. [56] R. M. Navari, Expert Rev. Anticancer Ther., 2008, 8, 1733. [57] J. J. Hale, S. G. Mills, M. MacCoss, C. P. Dorn, P. E. Finke, R. J. Budhu, R. A. Reamer, S.-E. W. Huskey, D. Luffer-Atlas, B. J. Dean, E. M. McGowan, W. P. Feeney, S.-H. Lee Chiu, M. A. Cascieri, G. G. Chicchi, M. M. Kurtz, S. Sadowski, E. Ber, F. D. Tattersall, N. M. J. Rupniak, A. R. Williams, W. Rycroft, R. Hargreaves, J. M. Metzger and D. E. MacIntyre, J. Med. Chem., 2000, 43, 1234. [58] L. A. Sorbera, D. Ferna´ndez-Forner and M. Baye´s, Drugs Future, 2008, 31, 101. [59] N. L. M. Cruden and D. E. Newby, Expert Opin. Pharmacother., 2008, 9, 2383. [60] K. Bork, U. Yasothan and P. Kirkpatrick, Nat. Rev. Drug Discov., 2008, 7, 801. [61] E. Ben-Menachem, Drugs Today, 2008, 44, 35.
632
Shridhar Hegde and Michelle Schmidt
[62] J. A. McIntyre, J. Castan˜er and L. Martin, Drugs Future, 2004, 29, 992. [63] E. Perucca, U. Yasothan, G. Clincke and P. Kirkpatrick, Nat. Rev. Drug Discov., 2008, 7, 973. [64] C.-S. Yuan and R. J. Israel, Expert Opin. Investig. Drugs, 2006, 15, 541. [65] T.-Y. Chen, C. E. Rosow, N. Serradell and J. Bolo´s, Drugs Future, 2007, 32, 771. [66] J. Thomas, S. Karver, G. A. Cooney, B. H. Chamberlain, C. K. Watt, N. E. Slatkin, N. Stambler, A. B. Kremer and R. J. Israel, N. Engl. J. Med., 2008, 358, 2332. [67] V. J. Thannickal, K. R. Flaherty, F. J. Martinez and J. P. Lynch, Expert Opin. Pharmacother., 2004, 5, 1671. [68] S. A. Antonia, Expert Opin. Investig. Drugs, 2006, 15, 823. [69] V. J. Thannickal, K. R. Flaherty, R. C. Hyzy and J. P. Lynch, 3rd, Expert Opin. Emerg. Drugs, 2005, 10, 707. [70] S. Shi, J. Wu, H. Chen, H. Chen, J. Wu and F. Zeng, J. Clin. Pharmacol., 2007, 47, 1268. [71] R. Goldbach-Mansky, S. D. Shroff, M. Wilson, C. Snyder, S. Plehn, B. Barham, T.-H. Pham, F. Pucino, R. A. Wesley, J. H. Papadopoulos, S. P. Weinsein, S. J. Mellis and D. L. Kastner, Arthritis Rheum., 2008, 58, 2432. [72] L. A. Sorbera, J. Bozzo and M. Baye´s, Drugs Future, 2007, 32, 411. [73] H. M. Hoffman, U. Yasothan and P. Kirkpatrick, Nat. Rev. Drug Discov., 2008, 7, 385. [74] G. Escolar, J. Villalta, F. Casals, J. Bozzo, N. Serradell and J. Bolo´s, Drugs Future, 2006, 31, 484. [75] P. Kakar, T. Watson and G. Y. H. Lip, Drugs Today, 2007, 43, 129. [76] D. B. Cines, U. Yasothan and P. Kirkpatrick, Nat. Rev. Drug Discov., 2008, 7, 887. [77] R. Stasi, M. L Evangelista and S. Amadori, Drugs, 2008, 68, 901. [78] D. J. Kuter, Annu. Rev. Med., 2009, 60, 193. [79] M. Biava, G. C. Porretta, D. Deidda and R. Pompei, Infect. Disorders, 2006, 6, 159. [80] D. L. Anderson, Drugs Today, 2008, 44, 489. [81] R. C. Owens, Jr. and P. G. Ambrose, Clin. Infect. Dis., 2005, 41, S144. [82] O. Tamura, M. Hashimoto, Y. Kobayashi, T. Katoh, K. Nakatani, M. Kamada, I. Hayakawa, T. Akiba and S. Terashima, Tet. Lett., 1992, 24, 3487. [83] J. A. McIntyre and J. Castan˜er, Drugs Future, 2005, 30, 780. [84] F. Donati, Expert Opin. Pharmacother., 2008, 9, 1375. [85] A. L. Kovac, Formulary, 2009, 44, 13. [86] M. Naguib and S. J. Brull, Expert Rev. Clin. Pharmacol., 2009, 2, 37. [87] G. Tayal, S. Kundra and A. Grewal, J. Anaesth. Clin. Pharmacol., 2008, 24, 211. [88] Y. Wang, J. Bolo´s and N. Serradell, Drugs Future, 2006, 31, 788. [89] Y. Takagi, T. Nakajima, A. Shimazaki, M. Kageyama, T. Matsugi, Y. Matsumura, B. T. Gabelt, P. L. Kaufman and H. Hara, Exp. Eye Res., 2004, 78, 767. [90] A. Sutton, P. Gouws and A. Ropo, Int. J. Clin. Pharmacol. Ther., 2008, 46, 400. [91] S. Jiang, R. I. Senderoff and J. D. Meyer, WO Patent 06009989, 2006. [92] P. D. Bishop, K. B. Lewis, J. Schultz and K. M. Walker, Semin. Thromb. Haemost., 2006, 32(Suppl. 1), 86. [93] M. Mohri, Cardiovasc. Drug Rev., 2000, 18, 312. [94] H. Saito, I. Maruyama, S. Shimazaki, Y. Yamamoto, N. Aikawa, R. Ohno, A. Hirayama, T. Matsuda, H. Asakura, M. Nakashima and N. Aoki, J. Thromb. Haemost, 2007, 5, 31. [95] S. Moll, C. Lindley, S. Pescatore, D. Morrison, K. Tsuruta, M. Mohri, M. Serada, M. Sata, H. Shimizu, K. Yamada and G. C. White, II, J. Thromb. Haemost, 2004, 2, 1745.
COMPOUND NAME, CODE NUMBER AND SUBJECT INDEX, VOLUME 44
A1 AR antagonists, 266, 268 A2A AR agonists, 268 A2B AR antagonists, 270–272 A3 AR antagonists, 273 A-796260, 229–230 A-836339, 229–230 A-928605, 285 Ab, 51–54, 56–58, 65 Absorption/Distribution/Metabolism/ Excretion (ADME), 462, 464 ABT-072, 399, 417–418 ABT-333, 399, 417–418 acetylcholine, 71, 73–75, 79, 88–89 ACH-806, 407 activation function 2, 445 active, 461–462, 464, 472 acute coronary syndrome (ACS), 190–191, 196 adenosine receptors, 266 ADME, 515, 517–518, 523, 530–532 AEZS-126, 347 airway remodeling, 272 aliskiren (Tekturnas, Rasilezs), 105–106, 108, 119, 121–124 Allergy, 267–269, 271–272, 274 Allosteric HCV NS5B inhibitor, 412 Alzheimer’s disease, 4, 51–53, 58 p-allylpalladium, 309–310 Amatiza, 479 Amevive, 483 amlodipine, 124 amyloid, 4–5, 19 amyotrophic lateral sclerosis, 8 androgen receptor, 453 angiotensin AT1 receptor blockers (ARBs), 106–107 angiotensin converting enzyme inhibitors (ACEi), 106–107, 119, 123 Anguizole, 409 Anthraquinones, 61 Anti-allodynia, 234, 239 antibacterial, 379–381, 385, 387, 390–391 antibiotic, 379–381, 392–393 anti-cancer, 324 anticoagulant, 190, 195, 198–201 antifungal, 359, 370–372 anti-hyperalgesia, 230
Antimitotic, 310 antipsychotic, 20 antithrombotic, 191–192, 199, 201, 203 antiviral, 359, 370–371 Anxiety, 129–130, 132, 134, 143 APD-597,166 APD-668, 166 apixaban (BMS-562247), 191, 193 AR231453, 152–153 Artemisinin, 359–363, 365–368, 370–375 AS-252424, 350 aspartyl protease, 111–112, 114, 120 aspirin, 191 assisted reproductive therapy, 172 asthma, 266–268, 270–272, 274–275 atherosclerosis, 123 atrial fibrilation (AF), 190–192, 195–196 Avastin, 486 AVP, 130–131, 139–140, 143–144 axin, 6–7 AXL-1717, 295 [11C]AZ10419369, 503, 505 AZD0837, 195 AZD6482, 349 a4b2 nAChR pharmacophore, 91 a-glucosidase inhibitor, 423–424 a-helix, 445 a-synuclein, 51, 52, 57, 63–65 Baicalein, 63–64 basal cell carcinoma (BCC), 324, 326, 332–333 benazepril, 123 beta-catenin, 6, 7, 10, 12, 16 betrixaban (PRT-54021), 192–193 BG-9928, 265–268 BGT226, 345 BI-201335, 399, 404 bicyclic benzazepine, 71, 87–88, 92, 94 bilirubin, 522 BILN-2061, 400 binding function 3, 453 bioinformatics, 558 biomarker, 559
635
636
Compound Name, Code Number and Subject Index, Volume 44
blood pressure, 106 BMS-201038, 423 BMS-247550, 310 BMS-536924, 290 BMS-695735, 290 BMS-754807, 288, 295 BMS-790052, 399, 410 BMS-833923 (XL-139), 333 Boceprevir, 399, 402, 405 Bolton-Hunter reagent, 526 bone-marrow derived macrophage (BMDM), 215–217, 219, 221 breast cancer, 301–302, 304, 315–316, 318–319, 444, 450 buprenorphine, 73 bupropion, 73, 98 bupropion SR, 98 Byetta, 477 CAL-101, 347–348 Cancer, 339–341, 343, 345–348, 350, 359–360, 365, 370, 372–373 Carbamazepine, 547–548 carcinogenicity, 462–463, 466–467 cardio-renoprotection, 123 Catecholamines, 64 CB2 agonist, 227–229, 231–233, 235–238, 240–242 CB2 receptor, 227–233, 236–237, 239–241 Celgosovir, 399, 424 central nervous system (CNS), 46–47 CGP29287, 108–109 CGP38560, 109–110, 121 CGS-21680, 265, 268–269 champix, 71 Chantix, 71 chloroquine, 364–365 choriogonadotropin (CG), 171–173 Chung model, 37 ciprofloxacin, 388–390 circulating, 461–462, 464, 466 cKit, 215–218, 220–221 clavatadine A, 203 Clemizole, 408 clinical trials, 41, 43, 47 clopidogrel, 191 clorobiocin, 379, 381–382, 386 coactivator, 443–447, 449, 451–453 coactivator binding inhibitor, 443, 445, 449 collagen-induced arthritis, 214–215 Colony Stimulating Factor-1, 211–221
Colony Stimulating Factor-1 Receptor, 211 congestive heart failure, 130, 142 Congo red, 57, 63, 65 conivaptan, 141 contraception, 171, 177–178, 181, 183–184 COPD, 266, 268–269, 272, 274–275 corepressor, 444–446, 452 coumestrol, 454 Covalent binding, 537–538, 545, 547, 551 CP-346086, 423 CP-690,550, 250–253, 255–256, 259, 261 Cur61414, 326–327, 333 CVT-6883, 265, 270–271 cyclopamine, 325–326, 332–333 Cyclosporin derivative, 424 cyclothialidine, 379–381, 383 CYP3A4, 535, 537, 539–545, 547–551 cytisine, 71, 74–75, 77–89, 91–95, 99 Cytochrome P450, 537, 544, 548, 550 cytotoxic, 304, 310–311, 318 dabigatran, 196, 203 dabigatran etexilate, 196 dalteparin, 191 db/db mouse, 165–166 DEBIO-025, 399, 424–425 deep vein thrombosis (DVT), 191, 195 delafloxacin, 389 Depression, 129–131, 134, 143 Designed Multiple Ligand (DML), 47 diet, 555, 567–569, 571 Dihydroartemisinin (DHA), 362, 373 Diltiazem, 543, 548 dimer, 29 dimerization inhibitor, 41 disproportionate, 462–464, 467, 470 DNA gyrase, 380–381, 383, 386–387, 392–393 DPP-IV inhibitor, 150, 154, 156, 161, 163, 165 Drug adduct, 547 drug discrimination, 76–78, 81–82 Drug interactions, 535, 537 Dual Action, 27, 43–45, 47 dysmenorrhea, 130–131, 133, 143 edoxaban (Du-176b), 191, 195 Emend, 488 endocannabinoids, 227
Compound Name, Code Number and Subject Index, Volume 44
endoethelial Nitric Oxide Synthase (eNOS), 28–32, 34–41, 45 enoxaparin, 192 Enzyme inactivation, 539–541, 547 Enzyme inhibition, 535, 540 enzyme, 250–251, 257, 259 epigallocathechin gallate, 388 epothilone, 301–311, 317–319 ERAD, 548 Erbitux, 486 eribaxaban (PD-348292), 192 Erythromycin, 537, 540 estrogen receptor, 446, 450 17b-estradiol, 446 etoricoxib, 524 euglycemic clamp, 163 experimental autoimmune encephalomyelitis, 227, 231 factor VIIa/tissue factor inhibitors, 199–200 factor IXa inhibitors, 201–203 factor Xa inhibitors, 189–190, 198 factor XIa inhibitors, 201–202 fasting, 555, 567–568, 571 fermentation, 301–303, 305, 308 Fibrils, 56, 58, 63–65 FK-838, 265–268 Flavonoids, 61, 64 flufenamic acid, 453 fluorescence polarization, 447, 450–454 fluoroquinolone, 380, 389–392 flutamide, 444 FMN, 28 FMS, 211–221 folate-conjugate, 318 follicle stimulating hormone (FSH), 171–173, 175–176, 179, 184 follicle stimulating hormone receptor (FSHR), 171–173, 175–181, 183–185 fragment-based approaches, 379, 385 FSHR agonist, 171, 175–177, 179–180, 183, 185 FSHR antagonist, 171, 173, 177, 179–181, 184 GANT58, 331–332 GANT61, 331–332 gatifloxacin, 388 GDC-0449, 328, 333 GDC-0941, 345 general toxicity, 463, 467
637
genotoxicity, 463, 467, 555, 558, 569 GIP, 149–150, 152, 154, 156, 162 Gli, 324–329, 331–332 Gli-Luciferase reporter gene assay, 331 Gli-mediated transcription, 331 GLP-1, 150, 153–154, 156 glucose-dependent insulin release, 151–152, 156 glycogen synthase kinase-3, 5 GPR119, 149–167 granulosa cells, 172, 179 Grapefruit juice, 544, 548, 550 GRC10693, 242 grepafloxacin, 388 GS-9190, 399, 419–420 GSK1904529A, 288 GSK252A, 161–162 GSK-3 crystal structure, 9 GSK-3 inhibitor, 6–11, 13–14, 16–20, 22 GSK-3 isoforms, 9 GSK-615, 346–347 GW274150, 39, 42–43 GW328276, 265,268–270 GW405833, 228 GW833972A, 232 GW842166, 242 GyrA, 380–381, 389, 391, 393 GyrB, 380–381, 383–385, 387–388, 390, 393 HCV assembly and egress inhibitor, 423 HCV entry inhibitor, 420–421 HCV IRES inhibitor, 422 HCV NS2 inhibitors, 397–399 HCV NS3 helicase inhibitor, 397, 407 HCV NS3/4A inhibitors, 402 HCV NS4B replication factor inhibitor, 397, 408 HCV NS5A replication factor inhibitor, 397, 409 HCV NS5B polymerase inhibitor, 397, 412 HCV-796 (nesbuvir), 418–419 headache, 28, 33, 39, 41–42 heart failure, 106, 124 Hedgehog, 323–325, 331–332 Hedgehog pathway inhibitors, 323–324, 326 helix-groove interaction, 447 Heme, 28–29, 33, 41 heparin, 201–202 hepatotoxicity, 557, 567 Hh signaling pathway, 324–325, 334
638
Compound Name, Code Number and Subject Index, Volume 44
Hh-Antag691, 327–328 HIT-T15, 153, 155, 165 HU-308, 231–232 hydrochlorothiazide, 124 hyperalgesia, 37, 44, 47 hyperphosphorylation, 4–6, 8 hypersensitivity, 537, 548, 551 hypertension, 105–107, 123–124 hyponatremia, 140–141 hypothermia, 238 IC-87114, 347–348 IGF-1R, 281–295 IL6, 216, 220 imatinib, 220 immunosuppression, 261 in vitro metabolite profiling, 469 in vivo metabolite profiling, 469–471 in vivo microdialysis, 95 INCB18424, 251–253, 255 Increlex, 484 Indoles, 56 inducible Nitric Oxide Synthase (iNOS), 28–29, 31–43, 45–47 infertility, 171–173, 176–178, 182–185 Inflammation, 248, 252, 261, 339, 341, 351, 557–559, 562, 564, 569 inflammatory pain, 229–230, 233, 236, 238–240, 242 INSM-18, 294–295 Inspra, 480 Insulin, 281–283, 287, 288, 290, 292, 294, 296 Insulin Like-Growth Factor, 281–295 intra-renal RAAS, 122 IPI-926, 326, 335 IR, 282–284, 286–288, 290, 292–294 Iressa, 486 Isozyme selectivity, 349–350 ITMN-191 (R-7227), 399, 403, 416 ITX-5061, 421 ixabepilone, 301–303, 310, 314–318 IXEMPRA, 301–319 JAK3, 247–261 Janus Kinase 3, 247 Januvia, 477 JK184, 332 JNJ-17308616, 132 ketoconazole, 454 KF-26777, 265, 272–274
kinase, 5–7, 9–11, 13–22, 247–249, 253–256, 261, 281–286, 290–294, 340–343, 345 Kinase inhibitors, 342–343 knockout, 475–491 L-738167, 524 L-778123, 525 L-97-1, 265–266 LAF237, 163 L-Arginine, 28-30, 33–34, 38–40 LAS38096, 265, 270–272 LDE225, 333 left ventricular hypertrophy, 123 leishmaniasis, 359–361, 367–368 levofloxacin, 388–391 Leydig cells, 172 LHR agonist, 171, 174–176, 183, 185 LHR antagonist, 171, 175 Ligand-bound structures, 343 Liquid chromatography/tandem mass spectrometry (LC/MS/MS), 469–470 lithium, 8, 10 liver, 555–559, 565, 567–569, 571 lixivaptan, 140 L-NMMA, 33, 41–43 LPS, 216, 220 Lucentis, 486 lung injury, 271 luteinizing hormone (LH), 171–173 luteinizing hormone receptor (LHR), 171–176, 183, 185 LY294002, 342 LY517717, 192 macrolactam, 307, 310, 319 macrolide, 308, 319 Macrophage, 211–216, 221 Macugen, 486 major, 461–465, 467–469 malaria, 360–363, 365, 367, 370–371, 374 mass balance studies, 470 mast cells, 268, 272, 274 MBX-2982, 166–167 MBX-3152, 161 MDA-7, 231 mecamylamine, 73–74, 76–77 Mechanism-based inhibition, 535–540, 543, 546–547, 549–551 mechanisms, 555, 563, 566
Compound Name, Code Number and Subject Index, Volume 44
medulloblastoma, 324, 326–327, 330, 333–334 melagatran, 195 Memoquin, 57 Metabolic intermediate complex, 540 Metabolism studies, 461, 473 Metabolites, 461–474 Metabolites in Safety Testing (MIST), 461–469, 471–472 (5-methyl-8-(4-methyl-piperazin-1-yl)-4oxo-4H-chromene-2-carboxylic acid (4-morpholin-4-yl-phenyl)-amide), 503 Methylene blue (methythioioninium chloride), 62 (S)-Mephenytoin, 520 Microdosing, 501, 507, 509 microtubule, 4–6, 302, 317, 319 migraine, 28, 33, 39, 41–43 MIST guidance, 459, 461–462, 464–469, 471–472 MK-7009, 399, 404 morphine, 87–88 mouse medulloblastoma allograft model, 330 moxifloxacin, 388–391 mozavaptan, 141 MRE-2029F20, 265, 272–274 MRE-3008-F20, 265, 272–274 MRS-1523, 265, 272–273 MRSA, 389–391, 393 MSAPPA, 526 mTOR, 340–341, 343, 345–347, 350–351 MTP inhibitor, 423 MTR-106, 43 mucus production, 266, 272–274 multidrug resistance, 319 multiple sclerosis, 215, 222, 231 myxobacteria, 302 NADPH, 28 nalidixic acid, 388 Namenda, 487 natural product, 301–304, 306, 309, 317 NDGA, 294 NECA, 268–271 needle screening, 387 nemonoxacin, 389 nephropathy, 123–124 neurodegeneration, 4, 52, 59 neurofibrillary tangles, 4
639
neuronal Nitric Oxide Synthase (nNOS), 28–41, 43–45, 47 neuropathic pain, 37, 46, 229–230, 234, 239–240, 242 neuropathy, 315–316, 318 neuroprotection, 19 new mechanism of action, 489 Nexavar, 485 nicotine addiction, 72, 98 nicotine dependence, 73, 98 nicotine replacement therapy, 72 nicotinic acetylcholine receptor, 71, 73 nicotinic partial agonist, 80, 84 NIM-811, 399, 424 Nitazoxanide, 399, 425–426 Nitric Oxide (NO), 27–28 N-Oleoyl dopamine, 151 noncompetitve inhibitor, 18 non-pyridone cytisine, 85–90, 94 nordihydroguaiaretic acid, 294 novobiocin, 379–381, 383–384, 386, 393 N-Phenylamines, 61 nuclear hormone receptor, 443–445 Nuclear magnetic resonance (NMR), 471 Nucleoside HCV NS5B inhibitor, 412, 424 NVP-ADW742, 284 NVP-AEW-541, 284 NVP-BEZ235, 345 NXN-188, 43 obesity, 555, 565, 568–569 Oleoyl ethanolamide, 151 Oligomers, 51–54, 59, 63–65 OPC-21268, 131, 132 OPC-51803, 139 Orencia, 481 Orfadin, 484 Org 41841, 174–175 Org 43553, 174–175, 183, 185 organ transplantation, 247–248, 252 OSI-906, 285, 295 OSIP339391, 265, 270–272 osteoarthritis, 229, 238, 242 Osteoclast, 212–213, 217 osteoporosis, 232 oxgenase domain, 28, 29, 38 p110a, 341, 345–346 p110b, 341 p110d, 341, 348 p110g, 341, 345
640
Compound Name, Code Number and Subject Index, Volume 44
paclitaxel, 302, 304, 310–314, 319 paired helical filament, 5 Palifermin, 486 pannus, 213–215 ParC, 380–381, 389, 391 ParE, 380, 385, 390 Parkinson’s disease, 51–52, 63 Paroxetine, 543 Patched, 323–324 pathology, 557–559, 561, 563, 567–568 patupilone, 317 PCI-27483, 199 peptidergic GPCR, 129, 131 PET tracer, 501–504, 506–512 PF-03550096, 231 phage display, 446 Pharmaceutical Research and Manufacturers of America (PhRMA), 461 Pharmacokinetic enhancement, 539, 550 Pharmacokinetic modeling, 549 Pharmacokinetics, 470 pharmacology, 555, 557–560, 561–563, 571 Phase 1 studies, 471 Phase 3 studies, 470, Phenothiazines, 61–62 Phenylthiazolyl-hydrazides, 59, 61 phosphorylation, 5–10, 18–19 phototoxicity, 389 physalins, 331 Physical Properties, 501, 503, 506, 511 PI3K, 339–351 PI3Ka, 341, 350 PI3Kb, 341 PI3Kd, 347–348 PI3Kg, 341–343, 348, 350 picropodophyllotoxin, 295 PIK-75, 350 Piperonyl butoxide, 543 plasma renin activity (PRA), 107, 109, 115–116, 118, 120 PNU-286607, 392 polyketide, 303, 319 Polyphenols, 61 polyuria, 139 Porphyrins, 61–62 Post traumatic stress disorder (PTSD), 130–131 post-operative pain, 234 PPP, 295 PQ401, 291
PQIP, 285 pregnane X receptor, 454 Prialt, 487 prostate cancer, 315, 319, 444 Proteasome, 548 Protein aggregation, 52, 65 Protein turnover, 548, 551 Protein-protein interactions, 51–52, 62, 66, 445 PRS-211375, 242 PSI-7851, 413 PSN119-1, 164 PSN119-1M, 164 PSN375963, 153 PSN632408, 153 PSN-821, 167 psoriasis, 247, 251, 253 pulmonary inflammation, 271 purines, 247, 258 purinones, 247, 258 PX-866, 345 pyridines, 247, 259 pyrimidines, 247, 259 pyrrolopyridines, 247, 258 pyrrolopyrimidines, 247, 255, 257 quinolone-resistance determining region (QRDR), 389–392 Quasi-irreversible, 535–537, 540, 543–544, 546, 551 quercetin, 388 quinine, 391 quinolone, 379–381, 388–391, 393 R-1626, 413 R-1728, 413–414 R-348, 251–252, 260 radiochemistry, 502, 509 Radioisotopes, 516–517 radiolabeled, 462, 465, 469–471 Radiolabeled compounds, 515–518, 531–532 Raptiva, 482 Rasilezs (aliskiren), 105–106, 124 rat, 559, 565–566, 569–571 Raynaud’s syndrome, 130–131 razaxaban, 191, 193–194 reactive, 461–462, 464–465, 468, 472 Receptor Occupancy, 501, 508 reductase domain, 28–29 relcovaptan (SR49059), 131–132
Compound Name, Code Number and Subject Index, Volume 44
remikiren (RO-42-5892), 109–110, 119 renin, 105–112, 114–120, 122–124 renin inhibitor, 105–112, 114–116, 119, 123–124 renin-angiotensin double transgenic rats (dTGR), 108, 122–123 renin-angiotensin-aldosterone system (RAAS), 106–108, 122–123 reproductive tract disorders, 180 reprotoxicity, 463 rheumatoid arthritis, 212, 247, 251–253, 261 Rhodanines, 60 Ritonavir, 540, 550 rivaroxaban (BAY-59-7939), 190, 194, 203 Robotnikin, 331–332 rofecoxib, 529–530 Rozerem, 488 RWJ-339489, 142 RWJ-676070, 142 Safety evaluation, 460, 462, 472 sagopilone, 317 satavaptan, 140–141 SC-56525, 110 SCY-635, 399, 424–425 Scyllo-inositol, 53–54 Selective NOS Inhibitor, 27, 33, 41–43 semisynthesis, 301, 305, 307–310 serine proteases, 190–192, 201, 203 Serotonin 5-HT1B, 501–503, 506 Sertoli cells, 172 severe combined immunodeficiency, 248 SF1126, 343 Shh (Sonic hedgehog) inhibitors, 324 shistosomiasis, 359–361, 368 signaling pathway, 7 Silibinin, 399, 426–427 Silymarin, 426–427 smoking cessation therapy, 73, 77–78, 97 Smoothened, 323–325, 327, 329–331, 333 Smoothened inhibitors, 323, 325, 327, 331, 333 SMSB, 526 Sociality, 129–131, 144 Soliris, 484 Sorangium cellulosum, 302 Specific activity, 516–517 Spinal Nerve Ligation (SNL), 37 SPP100 (aliskiren), 119 SRX246, 133 SRX251, 133
641
SSR149415, 130, 134–135, 138 Stable isotopes, 517 stable, 461, 464, 468–469 steroid receptor coactivator, 454 streptococcal cell wall-induced arthritis, 219 Structure Based Design, 38, 40, 47 sunitinib, 220 suramine, 179 Sutent, 485 Symlin, 478 Tabex, 79 TAK-442, 192 tamoxifen, 444, 446 tanogitran (BIBT 986), 198 Tarceva, 486 target validation, 491 Tau, 51–52, 58–62 Tau protein, 8–9 Taxol, 304 TEA 226, 287 Tekturna, 480 Tekturnas (aliskiren), 105–106, 124 Telaprevir, 399–402 temafloxacin, 388 tetrahydrobiopterin (BH4) site, 28–29, 33–34, 38 TG(mRen-2)27 rats, 122–123 TG100-115, 348 TGX-221, 349 thapsigargin, 362 Theca cells, 172 Thiacarbocyanine dye, 60 thioglycollate-elicited peritoneal macrophages (TEPM), 216 thrombin inhibitors, 189, 195–197, 199 Thrombosis, 341, 350–351 thyroid hormone receptor, 447, 452 Tienilic acid, 545, 548 time-resolved fluorescence resonance energy transfer, 450 TMC-435350, 399, 402–403 TNF alpha, 214–216, 220 tolvaptan, 140–141 topoisomerase, 382–381, 391–392 Toxicity testing, 460, 462–463, 472 toxicogenomics, 555–557, 563, 565–569, 571–572 Toxicokinetics, 470 toxicology, 555–556, 567–568, 571–572
642
Compound Name, Code Number and Subject Index, Volume 44
toxoplasma, 366–367 Tramiprosate (3-amino-1propanesulfonic acid; APS), 57 transcription inhibitors, 324 transcriptomics, 555, 557, 561 Translational Science, 508 TRIAC, 453 Tricyclic pyrones and pyridinones, 54 triiodothyronine, 447–448, 453 Troleandomycin, 537 trovafloxacin, 388 Trypanosoma, 369–370 TTP889, 201 tubulin, 304, 306–311, 313, 317, 319 tumor regression, 333 Tykerb, 485 Type 2 diabetes mellitus, 149 tyrosine kinase, 211, 215 tyrphostins, 293 Tysabri, 483 Ubiquitination, 548 UK-371104, 265, 268 unique, 462–464, 469 United States Food and Drug Administration (FDA), 460–461, 463, 465, 467–468, 472–473 V1a, 129–133, 135–136, 138–146 V1b, 129–132, 134–141, 143–144 V2, 129–130, 132, 137, 139–143 Valopicitabine, 413 valsartan, 122–124
Vaprisol, 479 varenicline, 71, 73, 92, 94–99 Vasopressin, 129–130, 143 Velcade, 486 venous thromboembolism (VTE), 190–192, 196, 201 Verapamil, 543 Vidaza, 486 VRT-752586, 384–385 VX-509, 251 warfarin, 190, 192, 195 WAY-151932 (VNA-932), 139 wnt, 5–9 Wortmannin, 342, 345 WRC-0571, 265–268 ximelagatran, 195 XL147, 346 XL-228, 286, 295 XL765, 346 Xolair, 482 YM150, 192 zankiren (A-72517), 110, 119 ZDF rat, 164 Zelnorm, 478 Zetia, 480–481 Zolinza, 486 ZSTK474, 347 zyban, 98
CUMULATIVE CHAPTER TITLES KEYWORD INDEX, VOLUME 1–44
acetylcholine receptors, 30, 41; 40, 3 acetylcholine transporter, 28, 247 acyl sulfonamide anti-proliferatives, 41, 251 adenylate cyclase, 6, 227, 233; 12, 172; 19, 293; 29, 287 adenosine, 33, 111 adenosine, neuromodulator, 18, 1; 23, 39 adenosine receptor ligands, 44, 265 A3 adenosine receptors, 38, 121 adjuvants, 9, 244 ADME by computer, 36, 257 ADME, computational models, 42, 449 ADME properties, 34, 307 adrenal steroidogenesis, 2, 263 adrenergic receptor antagonists, 35, 221 b-adrenergic blockers, 10, 51; 14, 81 b-adrenergic receptor agonists, 33, 193 b2-adrenoceptor agonists, long acting, 41, 237 aerosol delivery, 37, 149 affinity labeling, 9, 222 b3-agonists, 30, 189 AIDS, 23, 161, 253; 25, 149 AKT kinase inhibitors, 40, 263 alcohol consumption, drugs and deterrence, 4, 246 aldose reductase, 19, 169 alkaloids, 1, 311; 3, 358; 4, 322; 5, 323; 6, 274 allergic eosinophilia, 34, 61 allergy, 29, 73 alopecia, 24, 187 Alzheimer’s Disease, 26, 229; 28, 49, 197, 247; 32, 11; 34, 21; 35, 31; 40, 35 Alzheimer’s Disease Research, 37, 31 Alzheimer’s Disease Therapies, 37, 197; 40, 35 aminocyclitol antibiotics, 12, 110 b-amyloid, 34, 21 amyloid, 28, 49; 32, 11 amyloidogenesis, 26, 229 analgesics (analgetic), 1, 40; 2, 33; 3, 36; 4, 37; 5, 31; 6, 34; 7, 31; 8, 20; 9, 11; 10, 12; 11, 23; 12, 20; 13, 41; 14, 31; 15, 32; 16, 41; 17, 21; 18, 51; 19, 1; 20, 21; 21, 21; 23, 11; 25, 11; 30, 11; 33, 11 androgen action, 21, 179; 29, 225 androgen receptor modulators, 36, 169 anesthetics, 1, 30; 2, 24; 3, 28; 4, 28; 7, 39; 8, 29; 10, 30, 31, 41 angiogenesis inhibitors, 27, 139; 32, 161 angiotensin/renin modulators, 26, 63; 27, 59 animal engineering, 29, 33 animal healthcare, 36, 319 animal models, anxiety, 15, 51
645
646
Cumulative Chapter Titles Keyword Index, Volume 1–44
animal models, memory and learning, 12, 30 Annual Reports in Medicinal Chemistry, 25, 333 anorexigenic agents, 1, 51; 2, 44; 3, 47; 5, 40; 8, 42; 11, 200; 15, 172 antagonists, calcium, 16, 257; 17, 71; 18, 79 antagonists, GABA, 13, 31; 15, 41; 39, 11 antagonists, narcotic, 7, 31; 8, 20; 9, 11; 10, 12; 11, 23 antagonists, non-steroidal, 1, 213; 2, 208; 3, 207; 4, 199 Antagonists, PGD2, 41, 221 antagonists, steroidal, 1, 213; 2, 208; 3, 207; 4, 199 antagonists of VLA-4, 37, 65 anthracycline antibiotics, 14, 288 antiaging drugs, 9, 214 antiallergy agents, 1, 92; 2, 83; 3, 84; 7, 89; 9, 85; 10, 80; 11, 51; 12, 70; 13, 51; 14, 51; 15, 59; 17, 51; 18, 61; 19, 93; 20, 71; 21, 73; 22, 73; 23, 69; 24, 61; 25, 61; 26, 113; 27, 109 antianginals, 1, 78; 2, 69; 3, 71; 5, 63; 7, 69; 8, 63; 9, 67; 12, 39; 17, 71 anti-angiogenesis, 35, 123 antianxiety agents, 1, 1; 2, 1; 3, 1; 4, 1; 5, 1; 6, 1; 7, 6; 8, 1; 9, 1; 10, 2; 11, 13; 12, 10; 13, 21; 14, 22; 15, 22; 16, 31; 17, 11; 18, 11; 19, 11; 20, 1; 21, 11; 22, 11; 23, 19; 24, 11 antiapoptotic proteins, 40, 245 antiarrhythmic agents, 41, 169 antiarrhythmics, 1, 85; 6, 80; 8, 63; 9, 67; 12, 39; 18, 99, 21, 95; 25, 79; 27, 89 antibacterial resistance mechanisms, 28, 141 antibacterials, 1, 118; 2, 112; 3, 105; 4, 108; 5, 87; 6, 108; 17, 107; 18, .29, 113; 23, 141; 30, 101; 31, 121; 33, 141; 34, 169; 34, 227; 36, 89; 40, 301 antibacterial targets, 37, 95 antibiotic transport, 24, 139 antibiotics, 1, 109; 2, 102; 3, 93; 4, 88; 5, 75, 156; 6, 99; 7, 99, 217; 8, 104; 9, 95; 10, 109, 246; 11, 89; 11, 271; 12, 101, 110; 13, 103, 149; 14, 103; 15, 106; 17, 107; 18, 109; 21, 131; 23, 121; 24, 101; 25, 119; 37, 149; 42, 349 antibiotic producing organisms, 27, 129 antibodies, cancer therapy, 23, 151 antibodies, drug carriers and toxicity reversal, 15, 233 antibodies, monoclonal, 16, 243 antibody drug conjugates, 38, 229 anticancer agents, mechanical-based, 25, 129 anticancer drug resistance, 23, 265 anticoagulants, 34, 81; 36, 79; 37, 85 anticoagulant agents, 35, 83 anticoagulant/antithrombotic agents, 40, 85 anticonvulsants, 1, 30; 2, 24; 3, 28; 4, 28; 7, 39, 8, 29; 10, 30; 11, 13; 12, 10; 13, 21; 14, 22; 15, 22; 16, 31; 17, 11; 18, 11; 19, 11; 20, 11; 21, 11; 23, 19; 24, 11 antidepressants, 1, 12; 2, 11; 3, 14; 4, 13; 5, 13; 6, 15; 7, 18; 8, 11; 11, 3; 12, 1; 13, 1; 14, 1; 15, 1; 16, 1; 17, 41; 18, 41; 20, 31; 22, 21; 24, 21; 26, 23; 29, 1; 34, 1 antidepressant drugs, new, 41, 23 antidiabetics, 1, 164; 2, 176; 3, 156; 4, 164; 6, 192; 27, 219 antiepileptics, 33, 61 antifungal agents, 32, 151; 33, 173, 35, 157 antifungal drug discovery, 38, 163; 41, 299 antifungals, 2, 157; 3, 145; 4, 138; 5, 129; 6, 129; 7, 109; 8, 116; 9, 107; 10, 120; 11, 101; 13, 113; 15, 139; 17, 139; 19, 127; 22, 159; 24, 111; 25, 141; 27, 149
Cumulative Chapter Titles Keyword Index, Volume 1–44
647
antiglaucoma agents, 20, 83 anti-HCV therapeutics, 34, 129; 39, 175 antihyperlipidemics, 15, 162; 18, 161; 24, 147 antihypertensives, 1, 59; 2, 48; 3, 53; 4, 47; 5, 49; 6, 52; 7, 59; 8, 52; 9, 57; 11, 61; 12, 60; 13, 71; 14, 61; 15, 79; 16, 73; 17, 61; 18, 69; 19, 61; 21, 63; 22, 63; 23, 59; 24, 51; 25, 51 antiinfective agents, 28, 119 antiinflammatory agents, 28, 109; 29, 103 anti-inflammatories, 37, 217 anti-inflammatories, non-steroidal, 1, 224; 2, 217; 3, 215; 4, 207; 5, 225; 6, 182; 7, 208; 8, 214; 9, 193; 10, 172; 13, 167; 16, 189; 23, 181 anti-ischemic agents, 17, 71 antimalarial inhibitors, 34, 159 antimetabolite cancer chemotherapies, 39, 125 antimetabolite concept, drug design, 11, 223 antimicrobial drugs—clinical problems and opportunities, 21, 119 antimicrobial potentiation, 33, 121 antimicrobial peptides, 27, 159 antimitotic agents, 34, 139 antimycobacterial agents, 31, 161 antineoplastics, 2, 166; 3, 150; 4, 154; 5, 144; 7, 129; 8, 128; 9, 139; 10, 131; 11, 110; 12, 120; 13, 120; 14, 132; 15, 130; 16, 137; 17, 163; 18, 129; 19, 137; 20, 163; 22, 137; 24, 121; 28, 167 anti-obesity agents, centrally acting, 41, 77 antiparasitics, 1, 136, 150; 2, 131, 147; 3, 126, 140; 4, 126; 5, 116; 7, 145; 8, 141; 9, 115; 10, 154; 11, 121; 12, 140; 13, 130; 14, 122; 15, 120; 16, 125; 17, 129; 19, 147; 26, 161 antiparkinsonism drugs, 6, 42; 9, 19 antiplatelet therapies, 35, 103 antipsychotics, 1, 1; 2, 1; 3, 1; 4, 1; 5, 1; 6, 1; 7, 6; 8, 1; 9, 1; 10, 2; 11, 3; 12, 1; 13, 11; 14, 12; 15, 12; 16, 11; 18, 21; 19, 21; 21, 1; 22, 1; 23, 1; 24, 1; 25, 1; 26, 53; 27, 49; 28, 39; 33, 1 antiradiation agents, 1, 324; 2, 330; 3, 327; 5, 346 anti-resorptive and anabolic bone agents, 39, 53 anti-retroviral chemotherapy, 25, 149 antiretroviral drug therapy, 32, 131 antiretroviral therapies, 35, 177; 36, 129 antirheumatic drugs, 18, 171 anti-SARS coronavirus chemistry, 41, 183 antisense oligonucleotides, 23, 295; 33, 313 antisense technology, 29, 297 antithrombotics, 7, 78; 8, 73; 9, 75; 10, 99; 12, 80; 14, 71; 17, 79; 27, 99; 32, 71 antithrombotic agents, 29, 103 antitumor agents, 24, 121 antitussive therapy, 36, 31 antiviral agents, 1, 129; 2, 122; 3, 116; 4, 117; 5, 101; 6, 118; 7, 119; 8, 150; 9, 128; 10, 161; 11, 128; 13, 139; 15, 149; 16, 149; 18, 139; 19, 117; 22, 147; 23, 161; 24, 129; 26, 133; 28, 131; 29, 145; 30, 139; 32, 141; 33, 163; 37, 133; 39, 241 antitussive therapy, 35, 53 anxiolytics, 26, 1 apoptosis, 31, 249 aporphine chemistry, 4, 331 arachidonate lipoxygenase, 16, 213 arachidonic acid cascade, 12, 182; 14, 178
648
Cumulative Chapter Titles Keyword Index, Volume 1–44
arachidonic acid metabolites, 17, 203; 23, 181; 24, 71 artemisinin derivatives, 44, 359 arthritis, 13, 167; 16, 189; 17, 175; 18, 171; 21, 201; 23, 171, 181; 33, 203 arthritis, immunotherapy, 23, 171 aspartyl proteases, 36, 247 asthma, 29, 73; 32, 91 asymmetric synthesis, 13, 282 atherosclerosis, 1, 178; 2, 187; 3, 172; 4, 178; 5, 180; 6, 150; 7, 169; 8, 183; 15, 162; 18, 161; 21, 189; 24, 147; 25, 169; 28, 217; 32, 101; 34, 101; 36, 57; 40, 71 atherosclerosis HDL raising therapies, 40, 71 atherothrombogenesis, 31, 101 atrial natriuretic factor, 21, 273; 23, 101 attention deficit hyperactivity disorder, 37, 11; 39, 1 autoimmune diseases, 34, 257; 37, 217 autoreceptors, 19, 51 BACE inhibitors, 40, 35 bacterial adhesins, 26, 239 bacterial genomics, 32, 121 bacterial resistance, 13, 239; 17, 119; 32, 111 bacterial toxins, 12, 211 bacterial virulence, 30, 111 basophil degranulation, biochemistry, 18, 247 Bcl2 family, 31, 249; 33, 253 behavior, serotonin, 7, 47 benzodiazepine receptors, 16, 21 biofilm-associated infections, 39, 155 bioinformatics, 36, 201 bioisosteric groups, 38, 333 bioisosterism, 21, 283 biological factors, 10, 39; 11, 42 biological membranes, 11, 222 biological systems, 37, 279 biopharmaceutics, 1, 331; 2, 340; 3, 337; 4, 302; 5, 313; 6, 264; 7, 259; 8, 332 biosensor, 30, 275 biosimulation, 37, 279 biosynthesis, antibotics, 12, 130 biotechnology, drug discovery, 25, 289 biowarfare pathegens, 39, 165 blood-brain barrier, 20, 305; 40, 403 blood enzymes, 1, 233 bone, metabolic disease, 12, 223; 15, 228; 17, 261; 22, 169 bone metabolism, 26, 201 bradykinin-1 receptor antagonists, 38, 111 bradykinin B2 antagonists, 39, 89 brain, decade of, 27, 1 C5a antagonists, 39, 109 calcium antagonists/modulators, 16, 257; 17, 71; 18, 79; 21, 85 calcium channels, 30, 51 calmodulin antagonists, SAR, 18, 203 cancer, 27, 169; 31, 241; 34, 121; 35, 123; 35, 167
Cumulative Chapter Titles Keyword Index, Volume 1–44
649
cancer chemosensitization, 37, 115 cancer chemotherapy, 29, 165; 37, 125 cancer cytotoxics, 33, 151 cancer, drug resistance, 23, 265 cancer therapy, 2, 166; 3, 150; 4, 154; 5, 144; 7, 129; 8, 128; 9, 139, 151; 10, 131; 11, 110; 12, 120; 13, 120; 14, 132; 15, 130; 16, 137; 17, 163; 18, 129; 21, 257; 23, 151; 37, 225; 39, 125 cannabinoid receptors, 9, 253; 34, 199 cannabinoid, receptors, CB1, 40, 103 cannabinoid (CB2) selective agonists, 44, 227 carbohydrates, 27, 301 carboxylic acid, metalated, 12, 278 carcinogenicity, chemicals, 12, 234 cardiotonic agents, 13, 92; 16, 93; 19, 71 cardiovascular, 10, 61 case history: Chantix (varenicline tartrate), 44, 71 case history: Ixabepilone (ixempras), 44, 301 case history - JANUVIAs, 42, 95 case history - Tegaserod, 42, 195 case history: Tekturnas/rasilezs (aliskiren), 44, 105 caspases, 33, 273 catalysis, intramolecular, 7, 279 catalytic antibodies, 25, 299; 30, 255 Cathepsin K, 39, 63 CCR1 antagonists, 39, 117 CCR2 antagonists, 42, 211 CCR3 antagonists, 38, 131 cell adhesion, 29, 215 cell adhesion molecules, 25, 235 cell based mechanism screens, 28, 161 cell cycle, 31, 241; 34, 247 cell cycle kinases, 36, 139 cell invasion, 14, 229 cell metabolism, 1, 267 cell metabolism, cyclic AMP, 2, 286 cellular pathways, 37, 187 cellular responses, inflammatory, 12, 152 chemical tools, 40, 339 cheminformatics, 38, 285 chemogenomics, 38, 285 chemoinformatics, 33, 375 chemokines, 30, 209; 35, 191; 39, 117 chemotaxis, 15, 224; 17, 139, 253; 24, 233 chemotherapy of HIV, 38, 173 cholecystokinin, 18, 31 cholecystokinin agonists, 26, 191 cholecystokinin antagonists, 26, 191 cholesteryl ester transfer protein, 35, 251 chronic obstructive pulmonary disease, 37, 209 chronopharmacology, 11, 251 circadian processes, 27, 11
650
Cumulative Chapter Titles Keyword Index, Volume 1–44
Clostridium difficile treatments, 43, 269 CNS medicines, 37, 21 CNS PET imaging agents, 40, 49 coagulation, 26, 93; 33, 81 co-crystals in drug discovery, 43, 373 cognition enhancers, 25, 21 cognitive disorders, 19, 31; 21, 31; 23, 29; 31, 11 collagenase, biochemistry, 25, 177 collagenases, 19, 231 colony stimulating factor, 21, 263 combinatorial chemistry, 34, 267; 34, 287 combinatorial libraries, 31, 309; 31, 319 combinatorial mixtures, 32, 261 complement cascade, 27, 199; 39, 109 complement inhibitors, 15, 193 complement system, 7, 228 conformation, nucleoside, biological activity, 5, 272 conformation, peptide, biological activity, 13, 227 conformational analysis, peptides, 23, 285 congestive heart failure, 22, 85; 35, 63 contrast media, NMR imaging, 24, 265 corticotropin-releasing factor, 25, 217; 30, 21; 34, 11; 43, 3 corticotropin-releasing hormone, 32, 41 cotransmitters, 20, 51 CXCR3 antagonists, 40, 215 cyclic AMP, 2, 286; 6, 215; 8, 224; 11, 291 cyclic GMP, 11, 291 cyclic nucleotides, 9, 203; 10, 192; 15, 182 cyclin-dependent kinases, 32, 171 cyclooxygenase, 30, 179 cyclooxygenase-2 inhibitors, 32, 211; 39, 99 cysteine proteases, 35, 309; 39, 63 cystic fibrosis, 27, 235; 36, 67 cytochrome P-450, 9, 290; 19, 201; 32, 295 cytochrome P-450 inhibition, 44, 535 cytokines, 27, 209; 31, 269; 34, 219 cytokine receptors, 26, 221 database searching, 3D, 28, 275 DDT-type insecticides, 9, 300 dermal wound healing, 24, 223 dermatology and dermatological agents, 12, 162; 18, 181; 22, 201; 24, 177 designer enzymes, 25, 299 diabetes, 9, 182; 11, 170; 13, 159; 19, 169; 22, 213; 25, 205; 30, 159; 33, 213; 39, 31; 40, 167 diabetes targets, G-Protein coupled receptors, 42, 129 Diels-Alder reaction, intramolecular, 9, 270 dipeptidyl, peptidase 4, inhibitors, 40, 149 discovery indications, 40, 339 distance geometry, 26, 281 diuretic, 1, 67; 2, 59; 3, 62; 6, 88; 8, 83; 10, 71; 11, 71; 13, 61; 15, 100 DNA binding, sequence-specific, 27, 311; 22, 259
Cumulative Chapter Titles Keyword Index, Volume 1–44
651
DNA vaccines, 34, 149 docking strategies, 28, 275 dopamine, 13, 11; 14, 12; 15, 12; 16, 11, 103; 18, 21; 20, 41; 22, 107 dopamine D3, 29, 43 dopamine D4, 29, 43 DPP-IV Inhibition, 36, 191 drug abuse, 43, 61 drug abuse, CNS agents, 9, 38 drug allergy, 3, 240 drug carriers, antibodies, 15, 233 drug carriers, liposomes, 14, 250 drug delivery systems, 15, 302; 18, 275; 20, 305 drug design, 34, 339 drug design, computational, 33, 397 drug design, knowledge and intelligence in, 41, 425 drug design, metabolic aspects, 23, 315 drug discovery, 17, 301; 34, ; 34, 307 drug discovery, bioactivation in, 41, 369 drug disposition, 15, 277 drug metabolism, 3, 227; 4, 259; 5, 246; 6, 205; 8, 234; 9, 290; 11, 190; 12, 201; 13, 196, 304; 14, 188; 16, 319; 17, 333; 23, 265, 315; 29, 307 drug receptors, 25, 281 drug resistance, 23, 265 drug safety, 40, 387 dynamic modeling, 37, 279 dyslipidemia and insulin resistance enzyme targets, 42, 161 EDRF, 27, 69 elderly, drug action, 20, 295 electrospray mass spectrometry, 32, 269 electrosynthesis, 12, 309 enantioselectivity, drug metabolism, 13, 304 endorphins, 13, 41; 14, 31; 15, 32; 16, 41; 17, 21; 18, 51 endothelin, 31, 81; 32, 61 endothelin antagonism, 35, 73 endothelin antagonists, 29, 65, 30, 91 enzymatic monooxygenation reactions, 15, 207 enzyme induction, 38, 315 enzyme inhibitors, 7, 249; 9, 234; 13, 249 enzyme immunoassay, 18, 285 enzymes, anticancer drug resistance, 23, 265 enzymes, blood, 1, 233 enzymes, proteolytic inhibition, 13, 261 enzyme structure-function, 22, 293 enzymic synthesis, 19, 263; 23, 305 epitopes for antibodies, 27, 189 erectile dysfunction, 34, 71 estrogen receptor, 31, 181 estrogen receptor modulators, SERMS, 42, 147 ethnobotany, 29, 325 excitatory amino acids, 22, 31; 24, 41; 26, 11; 29, 53
652
Cumulative Chapter Titles Keyword Index, Volume 1–44
ex-vivo approaches, 35, 299 factor VIIa, 37, 85 factor Xa, 31, 51; 34, 81 factor Xa inhibitors, 35, 83 Fc receptor structure, 37, 217 fertility control, 10, 240; 14, 168; 21, 169 filiarial nematodes, 35, 281 FMS kinase inhibitors, 44, 211 formulation in drug discovery, 43, 419 forskolin, 19, 293 fragment-based lead discovery, 42, 431 free radical pathology, 10, 257; 22, 253 fungal nail infections, 40, 323 fungal resistance, 35, 157 G-proteins, 23, 235 G-proteins coupled receptor modulators, 37, 1 GABA, antagonists, 13, 31; 15, 41 galanin receptors, 33, 41 gamete biology, fertility control, 10, 240 gastrointestinal agents, 1, 99; 2, 91; 4, 56; 6, 68; 8, 93; 10, 90; 12, 91; 16, 83; 17, 89; 18, 89; 20, 117; 23, 201, 38, 89 gastrointestinal prokinetic agents, 41, 211 gastrointestinal tracts of mammals, 43, 353 gender based medicine, 33, 355 gene expression, 32, 231 gene expression, inhibitors, 23, 295 gene knockouts in mice as source of new targets, 44, 475 gene targeting technology, 29, 265 gene therapy, 8, 245; 30, 219 genetically modified crops, 35, 357 gene transcription, regulation of, 27, 311 genomic data mining, 41, 319 genomics, 34, 227; 40, 349 ghrelin receptor modulators, 38, 81 glucagon, 34, 189 glucagon, mechanism, 18, 193 glucagon receptor antagonists, 43, 119 b-D-glucans, 30, 129 glucocorticoid receptor modulators, 37, 167 glucocorticosteroids, 13, 179 Glucokinase Activators, 41, 141 glutamate, 31, 31 glycoconjugate vaccines, 28, 257 glycogen synthase kinase-3 (GSK-3), 40, 135; 44, 3 glycolysis networks model, 43, 329 glycopeptide antibiotics, 31, 131 glycoprotein IIb/IIIa antagonists, 28, 79 glycosylation, non-enzymatic, 14, 261 gonadal steroid receptors, 31, 11 gonadotropin receptor ligands, 44, 171
Cumulative Chapter Titles Keyword Index, Volume 1–44
gonadotropin releasing hormone, 30, 169; 39, 79 GPIIb/IIIa, 31, 91 gpr119 agonists, 44, 149 GPR40 (FFAR1) modulators, 43, 75 G-Protein coupled receptor inverse agonists, 40, 373 G protein-coupled receptors, 35, 271 growth factor receptor kinases, 36, 109 growth factors, 21, 159; 24, 223; 28, 89 growth hormone, 20, 185 growth hormone secretagogues, 28, 177; 32, 221 guanylyl cyclase, 27, 245 hallucinogens, 1, 12; 2, 11; 3, 14; 4, 13; 5, 23; 6, 24 HDL cholesterol, 35, 251 HDL modulating therapies, 42, 177 health and climate change, 38, 375 heart disease, ischemic, 15, 89; 17, 71 heart failure, 13, 92; 16, 93; 22, 85 hedgehog pathway inhibitors, 44, 323 HCV antiviral agents, 39, 175 helicobacter pylori, 30, 151 hemoglobinases, 34, 159 hemorheologic agents, 17, 99 hepatitis C viral inhibitors, 44, 397 herbicides, 17, 311 heterocyclic chemistry, 14, 278 high throughput screening, 33, 293 histamine H3 receptor agents, 33, 31; 39, 45 histamine H3 receptor antagonists, 42, 49 histone deacetylase inhibitors, 39, 145 hit-to-lead process, 39, 231 HIV co-receptors, 33, 263 HIV prevention strategies, 40, 277 HIV protease inhibitors, 26, 141; 29, 123 HIV reverse transcriptase inhibitors, 29, 123 HIV therapeutics, 40, 291 HIV vaccine, 27, 255 HIV viral entry inhibitors, CCR5 and CXCR4, 42, 301 homeobox genes, 27, 227 hormones, glycoprotein, 12, 211 hormones, non-steroidal, 1, 191; 3, 184 hormones, peptide, 5, 210; 7, 194; 8, 204; 10, 202; 11, 158; 16, 199 hormones, steroid, 1, 213; 2, 208; 3, 207; 4, 199 host modulation, infection, 8, 160; 14, 146; 18, 149 Hsp90 inhibitors, 40, 263 5-HT2C receptor modulator, 37, 21 human dose projections, 43, 311 human gene therapy, 26, 315; 28, 267 human retrovirus regulatory proteins, 26, 171 hybrid antibacterial agents, 43, 281 11 b-hydroxysteroid dehydrogenase type 1 inhibitors, 41, 127
653
654
Cumulative Chapter Titles Keyword Index, Volume 1–44
5-hydroxytryptamine, 2, 273; 7, 47; 21, 41 5-hydroxytryptamine -5-HT5A, 5-HT6, and 5-HT7, 43, 25 hypercholesterolemia, 24, 147 hypersensitivity, delayed, 8, 284 hypersensitivity, immediate, 7, 238; 8, 273 hypertension, 28, 69 hypertension, etiology, 9, 50 hypnotics, 1, 30; 2, 24; 3, 28; 4, 28; 7, 39; 8, 29; 10, 30; 11, 13; 12, 10; 13, 21; 14, 22; 15, 22, 16; 31; 17, 11; 18, 11; 19, 11; 22, 11 ICE gene family, 31, 249 IgE, 18, 247 IkB kinase inhibitors, 43, 155 Immune cell signaling, 38, 275 immune mediated idiosyncratic drug hypersensitivity, 26, 181 immune system, 35, 281 immunity, cellular mediated, 17, 191; 18, 265 immunoassay, enzyme, 18, 285 immunomodulatory proteins, 35, 281 immunophilins, 28, 207 immunostimulants, arthritis, 11, 138; 14, 146 immunosuppressants, 26, 211; 29, 175 immunosuppressive drug action, 28, 207 immunosuppressives, arthritis, 11, 138 immunotherapy, cancer, 9, 151; 23, 151 immunotherapy, infectious diseases, 18, 149; 22, 127 immunotherapy, inflammation, 23, 171 infections, sexually transmitted, 14, 114 infectious disease strategies, 41, 279 inflammation, 22, 245; 31, 279 inflammation, immunomodulatory approaches, 23, 171 inflammation, proteinases in, 28, 187 inflammatory bowel disease, 24, 167, 38, 141 inhibitors, AKT/PKB kinase, 42, 365 inhibitors and modulators, amyloid secretase, 42, 27 inhibitors, anti-apoptotic proteins, 40, 245 inhibitors, cathepsin K, 42, 111 inhibitors, complement, 15, 193 inhibitors, connective tissue, 17, 175 inhibitors, dipeptidyl peptidase 4, 40, 149 inhibitors, enzyme, 13, 249 inhibitors, gluthathione S-transferase, 42, 321 inhibitors, HCV, 42, 281 inhibitors, histone deacetylase, 42, 337 inhibitors, influenza neuraminidase, 41, 287 inhibitors, irreversible, 9, 234; 16, 289 inhibitors. MAP kinases, 42, 265 inhibitors, mitotic kinesin, 41, 263 inhibitors, monoamine reuptake, 42, 13 inhibitors, PDEs, 42, 3 inhibitors, platelet aggregation, 6, 60
Cumulative Chapter Titles Keyword Index, Volume 1–44
inhibitors, proteolytic enzyme, 13, 261 inhibitors, renin, 41, 155 inhibitors, renin-angiotensin, 13, 82 inhibitors, reverse transcription, 8, 251 inhibitors, spleen tyrosine kinase (Syk), 42, 379 inhibitors, transition state analogs, 7, 249 inorganic chemistry, medicinal, 8, 294 inosine monophosphate dehydrogenase, 35, 201 inositol triphosphate receptors, 27, 261 insecticides, 9, 300; 17, 311 insomnia treatments, 42, 63 in silico approaches, prediction of human volume of distribution, 42, 469 insulin, mechanism, 18, 193 insulin-like growth factor receptor (IGF-1R) inhibitors, 44, 281 integrins, 31, 191 b2 –integrin Antagonist, 36, 181 integrin alpha 4 beta 1 (VLA-4), 34, 179 intellectual property, 36, 331 interferon, 8, 150; 12, 211; 16, 229; 17, 151 interleukin-1, 20, 172; 22, 235; 25, 185; 29, 205, 33, 183 interleukin-2, 19, 191 interoceptive discriminative stimuli, animal model of anxiety, 15, 51 intracellular signaling targets, 37, 115 intramolecular catalysis, 7, 279 ion channel modulators, 37, 237 ion channels, ligand gated, 25, 225 ion channels, voltage-gated, 25, 225 ionophores, monocarboxylic acid, 10, 246 ionotropic GABA receptors, 39, 11 iron chelation therapy, 13, 219 irreversible ligands, 25, 271 ischemia/reperfusion, CNS, 27, 31 ischemic injury, CNS, 25, 31 isotopes, stable, 12, 319; 19, 173 isotopically labeled compounds in drug discovery, 44, 515 JAK3 Inhibitors, 44, 247 JAKs, 31, 269 ketolide antibacterials, 35, 145 b-lactam antibiotics, 11, 271; 12, 101; 13, 149; 20, 127, 137; 23, 121; 24, 101 b-lactamases, 13, 239; 17, 119; 43, 247 LDL cholesterol, 35, 251 learning, 3, 279; 16, 51 leptin, 32, 21 leukocyte elastase inhibitors, 29, 195 leukocyte motility, 17, 181 leukotriene biosynthesis inhibitors, 40, 199 leukotriene modulators, 32, 91 leukotrienes, 17, 291; 19, 241; 24, 71 LHRH, 20, 203; 23, 211 lipid metabolism, 9, 172; 10, 182; 11, 180; 12, 191; 13, 184; 14, 198; 15, 162
655
656
Cumulative Chapter Titles Keyword Index, Volume 1–44
lipoproteins, 25, 169 liposomes, 14, 250 lipoxygenase, 16, 213; 17, 203 LXR agonists, 43, 103 lymphocytes, delayed hypersensitivity, 8, 284 macrocyclic immunomodulators, 25, 195 macrolide antibacterials, 35, 145 macrolide antibiotics, 25, 119 macrophage migration inhibitor factor, 33, 243 magnetic resonance, drug binding, 11, 311 malaria, 31, 141; 34, 349, 38, 203 male contraception, 32, 191 managed care, 30, 339 MAP kinase, 31, 289 market introductions, 19, 313; 20, 315; 21, 323; 22, 315; 23, 325; 24, 295; 25, 309; 26, 297; 27, 321; 28, 325; 29, 331; 30, 295; 31, 337; 32, 305; 33, 327 mass spectrometry, 31, 319; 34, 307 mass spectrometry, of peptides, 24, 253 mass spectrometry, tandem, 21, 213; 21, 313 mast cell degranulation, biochemistry, 18, 247 matrix metalloproteinase, 37, 209 matrix metalloproteinase inhibitors, 35, 167 mechanism based, anticancer agents, 25, 129 mechanism, drug allergy, 3, 240 mechanisms of antibiotic resistance, 7, 217; 13, 239; 17, 119 medicinal chemistry, 28, 343; 30, 329; 33, 385; 34, 267 melanin-concentrating hormone, 40, 119 melanocortin-4 receptor, 38, 31 melatonin, 32, 31 melatonin agonists, 39, 21 membrane function, 10, 317 membrane regulators, 11, 210 membranes, active transport, 11, 222 memory, 3, 279; 12, 30; 16, 51 metabolism, cell, 1, 267; 2, 286 metabolism, drug, 3, 227; 4, 259; 5, 246; 6, 205; 8, 234; 9, 290; 11, 190; 12, 201; 13, 196, 304; 14, 188; 23, 265, 315 metabolism, lipid, 9, 172; 10, 182; 11, 180; 12, 191; 14, 198 metabolism, mineral, 12, 223 metabonomics, 40, 387 metabotropic glutamate receptor, 35, 1, 38, 21 metal carbonyls, 8, 322 metalloproteinases, 31, 231; 33, 131 metals, disease, 14, 321 metastasis, 28, 151 microbial genomics, 37, 95 microbial products screening, 21, 149 microtubule stabilizing agents, 37, 125 microwave-assisted chemistry, 37, 247 migraine, 22, 41; 32, 1
Cumulative Chapter Titles Keyword Index, Volume 1–44
mitogenic factors, 21, 237 mitotic kinesin inhibitors, 39, 135 modified serum lipoproteins, 25, 169 molecular diversity, 26, 259, 271; 28, 315; 34, 287 molecular libraries screening center network, 42, 401 molecular modeling, 22, 269; 23, 285 monoclonal antibodies, 16, 243; 27, 179; 29, 317 monoclonal antibody cancer therapies, 28, 237 monoxygenases, cytochrome P-450, 9, 290 mTOR inhibitors, 43, 189 multi-factorial diseases, basis of, 41, 337 multivalent ligand design, 35, 321 muscarinic agonists/antagonists, 23, 81; 24, 31; 29, 23 muscle relaxants, 1, 30; 2, 24; 3, 28; 4, 28; 8, 37 muscular disorders, 12, 260 mutagenicity, mutagens, 12, 234 mutagenesis, SAR of proteins, 18, 237 myocardial ischemia, acute, 25, 71 narcotic antagonists, 7, 31; 8, 20; 9, 11; 10, 12; 11, 23; 13, 41 natriuretic agents, 19, 253 natural products, 6, 274; 15, 255; 17, 301; 26, 259; 32, 285 natural killer cells, 18, 265 neoplasia, 8, 160; 10, 142 neurodegeneration, 30, 31 neurodegenerative disease, 28, 11 neurokinin antagonists, 26, 43; 31, 111; 32, 51; 33, 71; 34, 51 neurological disorders, 31, 11 neuronal calcium channels, 26, 33 neuronal cell death, 29, 13 neuropathic pain, 38, 1 neuropeptides, 21, 51; 22, 51 neuropeptide Y, 31, 1; 32, 21; 34, 31 neuropeptide Y receptor modulators, 38, 61 neuropeptide receptor antagonists, 38, 11 neuroprotection, 29, 13 neuroprotective agents, 41, 39 neurotensin, 17, 31 neurotransmitters, 3, 264; 4, 270; 12, 249; 14, 42; 19, 303 neutrophic factors, 25, 245; 28, 11 neutrophil chemotaxis, 24, 233 nicotinic acetylcholine receptor, 22, 281; 35, 41 nicotinic acetylcholine receptor modulators, 40, 3 nitric oxide synthase, 29, 83; 31, 221; 44, 27 NMR, 27, 271 NMR in biological systems, 20, 267 NMR imaging, 20, 277; 24, 265 NMR methods, 31, 299 NMR, protein structure determination, 23, 275 non-enzymatic glycosylation, 14, 261 non-HIV antiviral agents, 36, 119, 38, 213
657
658
Cumulative Chapter Titles Keyword Index, Volume 1–44
non-nutritive, sweeteners, 17, 323 non-peptide agonists, 32, 277 non-peptidic d-opinoid agonists, 37, 159 non-steroidal antiinflammatories, 1, 224; 2, 217; 3, 215; 4, 207; 5, 225; 6, 182; 7, 208; 8, 214; 9, 193; 10, 172; 13, 167; 16, 189 non-steroidal glucocorticoid receptor agonists, 43, 141 novel analgesics, 35, 21 NSAIDs, 37, 197 nuclear hormone receptor/steroid receptor coactivator inhibitors, 44, 443 nuclear orphan receptors, 32, 251 nucleic acid-drug interactions, 13, 316 nucleic acid, sequencing, 16, 299 nucleic acid, synthesis, 16, 299 nucleoside conformation, 5, 272 nucleosides, 1, 299; 2, 304; 3, 297; 5, 333; 39, 241 nucleotide metabolism, 21, 247 nucleotides, 1, 299; 2, 304; 3, 297; 5, 333; 39, 241 nucleotides, cyclic, 9, 203; 10, 192; 15, 182 obesity, 1, 51; 2, 44; 3, 47; 5, 40; 8, 42; 11, 200; 15, 172; 19, 157; 23, 191; 31, 201; 32, 21 obesity therapeutics, 38, 239 obesity treatment, 37, 1 oligomerisation, 35, 271 oligonucleotides, inhibitors, 23, 295 oncogenes, 18, 225; 21, 159, 237 opioid receptor, 11, 33; 12, 20; 13, 41; 14, 31; 15, 32; 16, 41; 17, 21; 18, 51; 20, 21; 21, 21 opioids, 12, 20; 16, 41; 17, 21; 18, 51; 20, 21; 21, 21 opportunistic infections, 29, 155 oral pharmacokinetics, 35, 299 organocopper reagents, 10, 327 osteoarthritis, 22, 179 osteoporosis, 22, 169; 26, 201; 29, 275; 31, 211 oxazolidinone antibacterials, 35, 135 oxytocin antagonists and agonists, 41, 409 P38a MAP kinase, 37, 177 P-glycoprotein, multidrug transporter, 25, 253 parallel synthesis, 34, 267 parasite biochemistry, 16, 269 parasitic infection, 36, 99 patents in medicinal chemistry, 22, 331 pathophysiology, plasma membrane, 10, 213 PDE IV inhibitors, 31, 71 PDE7 inhibitors, 40, 227 penicillin binding proteins, 18, 119 peptic ulcer, 1, 99; 2, 91; 4, 56; 6, 68; 8, 93; 10, 90; 12, 91; 16, 83; 17, 89; 18, 89; 19, 81; 20, 93; 22, 191; 25, 159 peptide-1, 34, 189 peptide conformation, 13, 227; 23, 285 peptide hormones, 5, 210; 7, 194; 8, 204; 10, 202; 11, 158, 19, 303 peptide hypothalamus, 7, 194; 8, 204; 10, 202; 16, 199 peptide libraries, 26, 271
Cumulative Chapter Titles Keyword Index, Volume 1–44
659
peptide receptors, 25, 281; 32, 277 peptide, SAR, 5, 266 peptide stability, 28, 285 peptide synthesis, 5, 307; 7, 289; 16, 309 peptide synthetic, 1, 289; 2, 296 peptide thyrotropin, 17, 31 peptidomimetics, 24, 243 periodontal disease, 10, 228 peroxisome proliferator – activated receptors, 38, 71 PET, 24, 277 PET imaging agents, 40, 49 PET ligands, 36, 267 pharmaceutics, 1, 331; 2, 340; 3, 337; 4, 302; 5, 313; 6, 254, 264; 7, 259; 8, 332 pharmaceutical innovation, 40, 431 pharmaceutical productivity, 38, 383 pharmaceutical proteins, 34, 237 pharmacogenetics, 35, 261; 40, 417 pharmacogenomics, 34, 339 pharmacokinetics, 3, 227, 337; 4, 259, 302; 5, 246, 313; 6, 205; 8, 234; 9, 290; 11, 190; 12, 201; 13, 196, 304; 14, 188, 309; 16, 319; 17, 333 pharmacophore identification, 15, 267 pharmacophoric pattern searching, 14, 299 phosphatidyl-inositol-3-kinases (PI3Ks) inhibitors, 44, 339 phosphodiesterase, 31, 61 phosphodiesterase 4 inhibitors, 29, 185; 33, 91; 36, 41 phosphodiesterase 5 inhibitors, 37, 53 phospholipases, 19, 213; 22, 223; 24, 157 physicochemical parameters, drug design, 3, 348; 4, 314; 5, 285 pituitary hormones, 7, 194; 8, 204; 10, 202 plants, 34, 237 plasma membrane pathophysiology, 10, 213 plasma protein binding, 31, 327 plasma protein binding, free drug principle, 42, 489 plasminogen activator, 18, 257; 20, 107; 23, 111; 34, 121 plasmon resonance, 33, 301 platelet activating factor (PAF), 17, 243; 20, 193; 24, 81 platelet aggregation, 6, 60 polyether antibiotics, 10, 246 polyamine metabolism, 17, 253 polyamine spider toxins, 24, 287 polymeric reagents, 11, 281 positron emission tomography, 24, 277; 25, 261; 44, 501 potassium channel activators, 26, 73 potassium channel antagonists, 27, 89 potassium channel blockers, 32, 181 potassium channel openers, 24, 91, 30, 81 potassium channel modulators, 36, 11 potassium channels, 37, 237 pregnane X receptor and CYP3A4 enzyme, 43, 405 privileged structures, 35, 289
660
Cumulative Chapter Titles Keyword Index, Volume 1–44
prodrugs, 10, 306; 22, 303 prodrug discovery, oral, 41, 395 profiling of compound libraries, 36, 277 programmed cell death, 30, 239 prolactin secretion, 15, 202 prostacyclin, 14, 178 prostaglandins, 3, 290; 5, 170; 6, 137; 7, 157; 8, 172; 9, 162; 11, 80; 43, 293 prostanoid receptors, 33, 223 prostatic disease, 24, 197 protease inhibitors for COPD, 43, 171 proteases, 28, 151 proteasome, 31, 279 protein C, 29, 103 protein growth factors, 17, 219 proteinases, arthritis, 14, 219 protein kinases, 18, 213; 29, 255 protein kinase C, 20, 227; 23, 243 protein phosphatases, 29, 255 protein-protein interactions, 38, 295; 44, 51 protein structure determination, NMR, 23, 275 protein structure modeling, 39, 203 protein structure prediction, 36, 211 protein structure project, 31, 357 protein tyrosine kinases, 27, 169 protein tyrosine phosphatase, 35, 231 proteomics, 36, 227 psoriasis, 12, 162; 32, 201 psychiatric disorders, 11, 42 psychoses, biological factors, 10, 39 psychotomimetic agents, 9, 27 pulmonary agents, 1, 92; 2, 83; 3, 84; 4, 67; 5, 55; 7, 89; 9, 85; 10, 80; 11, 51; 12, 70; 13, 51; 14, 51; 15, 59; 17, 51; 18, 61; 20, 71; 21, 73; 22, 73; 23, 69; 24, 61; 25, 61; 26, 113; 27, 109 pulmonary disease, 34, 111 pulmonary hypertension, 37, 41 pulmonary inflammation, 31, 71 pulmonary inhalation technology, 41, 383 purine and pyrimide nucleotide (P2) receptors, 37, 75 purine-binding enzymes, 38, 193 purinoceptors, 31, 21 QT interval prolongation, 39, 255 quantitative SAR, 6, 245; 8, 313; 11, 301; 13, 292; 17, 281 quinolone antibacterials, 21, 139; 22, 117; 23, 133 radioimmunoassays, 10, 284 radioisotope labeled drugs, 7, 296 radioimaging agents, 18, 293 radioligand binding, 19, 283 radiosensitizers, 26, 151 ras farnesyltransferase, 31, 171 ras GTPase, 26, 249 ras oncogene, 29, 165
Cumulative Chapter Titles Keyword Index, Volume 1–44
661
receptor binding, 12, 249 receptor mapping, 14, 299; 15, 267; 23, 285 receptor modeling, 26, 281 receptor modulators, nuclear hormone, 41, 99 receptor, concept and function, 21, 211 receptors, acetylcholine, 30, 41 receptors, adaptive changes, 19, 241 receptors, adenosine, 28, 295; 33, 111 receptors, adrenergic, 15, 217 receptors, b-adrenergic blockers, 14, 81 receptors, benzodiazepine, 16, 21 receptors, cell surface, 12, 211 receptors, drug, 1, 236; 2, 227; 8, 262 receptors, G-protein coupled, 23, 221, 27, 291 receptors, G-protein coupled CNS, 28, 29 receptors, histamine, 14, 91 receptors, muscarinic, 24, 31 receptors, neuropeptide, 28, 59 receptors, neuronal BZD, 28, 19 receptors, neurotransmitters, 3, 264; 12, 249 receptors, neuroleptic, 12, 249 receptors, opioid, 11, 33; 12, 20; 13, 41; 14, 31; 15, 32; 16, 41; 17, 21 receptors, peptide, 25, 281 receptors, serotonin, 23, 49 receptors, sigma, 28, 1 recombinant DNA, 17, 229; 18, 307; 19, 223 recombinant therapeutic proteins, 24, 213 renal blood flow, 16, 103 renin, 13, 82; 20, 257 reperfusion injury, 22, 253 reproduction, 1, 205; 2, 199; 3, 200; 4, 189 resistant organisms, 34, 169 respiratory syncytial virus, 43, 229 respiratory tract infections, 38, 183 retinoids, 30, 119 reverse transcription, 8, 251 RGD-containing proteins, 28, 227 rheumatoid arthritis, 11, 138; 14, 219; 18, 171; 21, 201; 23, 171, 181 rho-kinase inhibitors, 43, 87 ribozymes, 30, 285 RNAi, 38, 261 safety testing of drug metabolites, 44, 459 SAR, quantitative, 6, 245; 8, 313; 11, 301; 13, 292; 17, 291 same brain, new decade, 36, 1 schizophrenia, treatment of, 41, 3 secretase inhibitors, 35, 31; 38, 41 sedative-hypnotics, 7, 39; 8, 29; 11, 13; 12, 10; 13, 21; 14, 22; 15, 22; 16, 31; 17, 11; 18, 11; 19, 11; 22, 11 sedatives, 1, 30; 2, 24; 3, 28; 4, 28; 7, 39; 8, 29; 10, 30; 11, 13; 12, 10; 13, 21; 14, 22; 15; 22; 16, 31; 17, 11; 18, 11; 20, 1; 21, 11
662
Cumulative Chapter Titles Keyword Index, Volume 1–44
semicarbazide sensitive amine oxidase and VAP-1, 42, 229 sequence-defined oligonucleotides, 26, 287 serine protease inhibitors in coagulation, 44, 189 serine proteases, 32, 71 SERMs, 36, 149 serotonergics, central, 25, 41; 27, 21 serotonergics, selective, 40, 17 serotonin, 2, 273; 7, 47; 26, 103; 30, 1; 33, 21 serotonin receptor, 35, 11 serum lipoproteins, regulation, 13, 184 sexually-transmitted infections, 14, 114 SH2 domains, 30, 227 SH3 domains, 30, 227 silicon, in biology and medicine, 10, 265 sickle cell anemia, 20, 247 signal transduction pathways, 33, 233 skeletal muscle relaxants, 8, 37 sleep, 27, 11; 34, 41 slow-reacting substances, 15, 69; 16, 213; 17, 203, 291 SNPs, 38, 249 sodium/calcium exchange, 20, 215 sodium channel blockers, 41, 59; 43, 43 sodium channels, 33, 51 solid-phase synthesis, 31, 309 solid state organic chemistry, 20, 287 solute active transport, 11, 222 somatostatin, 14, 209; 18, 199; 34, 209 sphingomyelin signaling path, 43, 203 sphingosine 1 receptor modulators, 42, 245 spider toxins, 24, 287 SRS, 15, 69; 16, 213; 17, 203, 291 Statins, 37, 197; 39, 187 Statins, pleiotropic effects of, 39, 187 STATs, 31, 269 stereochemistry, 25, 323 steroid hormones, 1, 213; 2, 208; 3, 207; 4, 199 stroidogenesis, adrenal, 2, 263 steroids, 2, 312; 3, 307; 4, 281; 5, 192, 296; 6, 162; 7, 182; 8, 194; 11, 192 stimulants, 1, 12; 2, 11; 3, 14; 4, 13; 5, 13; 6, 15; 7, 18; 8, 11 stroke, pharmacological approaches, 21, 108 stromelysin, biochemistry, 25, 177 structural genomics, 40, 349 structure-based drug design, 27, 271; 30, 265; 34, 297 substance P, 17, 271; 18, 31 substituent constants, 2, 347 suicide enzyme inhibitors, 16, 289 superoxide dismutases, 10, 257 superoxide radical, 10, 257 sweeteners, non-nutritive, 17, 323 synthesis, asymmetric, 13, 282
Cumulative Chapter Titles Keyword Index, Volume 1–44
synthesis, computer-assisted, 12, 288; 16, 281; 21, 203 synthesis, enzymic, 23, 305 systems biology and kinase signaling, 42, 393 T-cells, 27, 189; 30, 199; 34, 219 tachykinins, 28, 99 target identification, 41, 331 taxol, 28, 305 technology, providers and integrators, 33, 365 tetracyclines, 37, 105 thalidomide, 30, 319 therapeutic antibodies, 36, 237 thrombin, 30, 71, 31, 51; 34, 81 thrombolytic agents, 29, 93 thrombosis, 5, 237; 26, 93; 33, 81 thromboxane receptor antagonists, 25, 99 thromboxane synthase inhibitors, 25, 99 thromboxane synthetase, 22, 95 thromboxanes, 14, 178 thyrotropin releasing hormone, 17, 31 tissue factor pathway, 37, 85 TNF-a, 32, 241 TNF-a converting enzyme, 38, 153 topical microbicides, 40, 277 topoisomerase, 21, 247; 44, 379 toxicity, mathematical models, 18, 303 toxicity reversal, 15, 233 toxicity, structure activity relationships for, 41, 353 toxicogenomics, 44, 555 toxicology, comparative, 11, 242; 33, 283 toxins, bacterial, 12, 211 transcription factor NF-kB, 29, 235 transcription, reverse, 8, 251 transcriptional profiling, 42, 417 transgenic animals, 24, 207 transgenic technology, 29, 265 transient receptor potential modulators, 42, 81 translational control, 29, 245 transporters, drug, 39, 219 traumatic injury, CNS, 25, 31 trophic factors, CNS, 27, 41 TRPV1 vanilloid receptor, 40, 185 tumor classification, 37, 225 tumor necrosis factor, 22, 235 type 2 diabetes, 35, 211; 40, 167 tyrosine kinase, 30, 247; 31, 151 urinary incontinence, 38, 51 urokinase-type plasminogen activator, 34, 121 urotensin-II receptor modulators, 38, 99 vanilloid receptor, 40, 185 vascular cell adhesion molecule-1, 41, 197
663
664
Cumulative Chapter Titles Keyword Index, Volume 1–44
vascular proliferative diseases, 30, 61 vasoactive peptides, 25, 89; 26, 83; 27, 79 vasoconstrictors, 4, 77 vasodilators, 4, 77; 12, 49 vasopressin antagonists, 23, 91 vasopressin receptor ligands, 44, 129 vasopressin receptor modulators, 36, 159 veterinary drugs, 16, 161 viruses, 14, 238 vitamin D, 10, 295; 15, 288; 17, 261; 19, 179 waking functions, 10, 21 water, structures, 5, 256 wound healing, 24, 223 xenobiotics, cyclic nucleotide metabolism, 15, 182 xenobiotic metabolism, 23, 315 X-ray crystallography, 21, 293; 27, 271
CUMULATIVE NCE INTRODUCTION INDEX, 1983–2008 GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
abacavir sulfate abarelix abatacept acarbose aceclofenac acemannan acetohydroxamic acid acetorphan acipimox acitretin acrivastine actarit adalimumab adamantanium bromide adefovir dipivoxil adrafinil AF-2259 afloqualone agalsidase alfa alacepril alclometasone dipropionate alefacept alemtuzumab alendronate sodium alfentanil HCl alfuzosin HCl alglucerase alglucosidase alfa aliskiren alitretinoin alminoprofen almotriptan anakinra anidulafungin alosetron hydrochloride alpha-1 antitrypsin alpidem alpiropride alteplase
antiviral anticancer rheumatoid arthritis antidiabetic antiinflammatory wound healing agent hypoammonuric antidiarrheal hypolipidemic antipsoriatic antihistamine antirheumatic rheumatoid arthritis antiseptic antiviral psychostimulant antiinflammatory muscle relaxant fabry’s disease antihypertensive topical antiinflammatory plaque psoriasis anticancer osteoporosis analgesic antihypertensive enzyme Pompe disease antihypertensive anticancer analgesic antimigraine antiarthritic antifungal irritable bowel syndrome protease inhibitor anxiolytic antimigraine thrombolytic
1999 2004 2006 1990 1992 2001 1983 1993 1985 1989 1988 1994 2003 1984 2002 1986 1987 1983 2001 1988 1985 2003 2001 1993 1983 1988 1991 2006 2007 1999 1983 2000 2001 2006 2000 1988 1991 1988 1987
35, 40, 42, 26, 28, 37, 19, 29, 21, 25, 24, 30, 39, 20, 38, 22, 23, 19, 37, 24, 21, 39, 37, 29, 19, 24, 27, 42, 43, 35, 19, 36, 37, 42, 36, 24, 27, 24, 23,
333 446 509 297 325 259 313 332 323 309 295 296 267 315 348 315 325 313 259 296 323 267 260 332 314 296 321 511 461 333 314 295 261 512 295 297 322 296 326
667
668
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
alvimopan ambrisentan
post-operative ileus pulmonary arterial hypertension antiinflammatory cytoprotective topical antiinflammatory antipsychotic antiasthmatic antihypertensive topical antifungal antihypertensive antiinflammatory antiviral cardiotonic antineoplastic antineoplastic antiinflammatory hematological antineoplastic anticancer adjuvant cognition enhancer antidote
2008 2007
44, 584 43, 463
1986 1995 1990 1986 1987 1990 1991 1988 1994 1999 1983 2002 1987 1993 1997 1995 1994 1993 2002
22, 31, 26, 22, 23, 26, 27, 24, 30, 35, 19, 38, 23, 29, 33, 31, 30, 29, 38,
315 338 298 316 327 298 322 297 296 334 314 349 327 332 328 338 296 333 350
1987 1988 2003 1987 1996 1990 2007
23, 24, 39, 23, 32, 26, 43,
326 297 268 326 306 298 465
1990 1999 2002 1986 2000 1987 1987 1983 1985 2003 2003
26, 35, 38, 22, 36, 23, 23, 19, 21, 39, 39,
299 335 350 316 296 327 328 314 324 269 270
1997 2000
33, 328 36, 297
amfenac sodium amifostine aminoprofen amisulpride amlexanox amlodipine besylate amorolfine HCl amosulalol ampiroxicam amprenavir amrinone amrubicin HCl amsacrine amtolmetin guacil anagrelide HCl anastrozole angiotensin II aniracetam anti-digoxin polyclonal antibody APD apraclonidine HCl aprepitant APSAC aranidipine arbekacin arformoterol argatroban arglabin aripiprazole arotinolol HCl arteether artemisinin aspoxicillin astemizole astromycin sulfate atazanavir atomoxetine atorvastatin calcium atosiban
calcium regulator antiglaucoma antiemetic thrombolytic antihypertensive antibiotic chronic obstructive pulmonary disease antithromobotic anticancer neuroleptic antihypertensive antimalarial antimalarial antibiotic antihistamine antibiotic antiviral attention deficit hyperactivity disorder dyslipidemia preterm labor
669
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
atovaquone auranofin azacitidine azelnidipine azelaic acid azelastine HCl azithromycin azosemide aztreonam balofloxacin balsalazide disodium bambuterol barnidipine HCl beclobrate befunolol HCl belotecan benazepril HCl benexate HCl benidipine HCl beraprost sodium betamethasone butyrate prospinate betaxolol HCl betotastine besilate bevacizumab bevantolol HCl bexarotene biapenem bicalutamide bifemelane HCl bimatoprost binfonazole binifibrate biolimus drug-eluting stent bisantrene HCl bisoprolol fumarate bivalirudin blonanserin bopindolol bortezomib bosentan brimonidine brinzolamide brodimoprin bromfenac sodium
antiparasitic chrysotherapeutic anticancer antihypertensive antiacne antihistamine antibiotic diuretic antibiotic antibacterial ulcerative colitis bronchodilator antihypertensive hypolipidemic antiglaucoma anticancer antihypertensive antiulcer antihypertensive platelet aggreg. inhibitor topical antiinflammatory
1992 1983 2004 2003 1989 1986 1988 1986 1984 2002 1997 1990 1992 1986 1983 2004 1990 1987 1991 1992 1994
28, 19, 40, 39, 25, 22, 24, 22, 20, 38, 33, 26, 28, 22, 19, 40, 26, 23, 27, 28, 30,
326 143 447 270 310 316 298 316 315 351 329 299 326 317 315 449 299 328 322 326 297
antihypertensive antiallergic anticancer antihypertensive anticancer antibacterial antineoplastic nootropic antiglaucoma hypnotic hypolipidemic anti-restenotic antineoplastic antihypertensive antithrombotic antipsychotic antihypertensive anticancer antihypertensive antiglaucoma antiglaucoma antibiotic NSAID
1983 2000 2004 1987 2000 2002 1995 1987 2001 1983 1986 2008 1990 1986 2000 2008 1985 2003 2001 1996 1998 1993 1997
19, 36, 40, 23, 36, 38, 31, 23, 37, 19, 22, 44, 26, 22, 36, 44, 21, 39, 37, 32, 34, 29, 33,
315 297 450 328 298 351 338 329 261 315 317 586 300 317 298 587 324 271 262 306 318 333 329
670
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
brotizolam brovincamine fumarate bucillamine bucladesine sodium budipine budralazine bulaquine bunazosin HCl bupropion HCl buserelin acetate buspirone HCl butenafine HCl butibufen butoconazole butoctamide butyl flufenamate cabergoline cadexomer iodine cadralazine calcipotriol camostat mesylate candesartan cilexetil capecitabine captopril carboplatin carperitide carumonam carvedilol caspofungin acetate cefbuperazone sodium cefcapene pivoxil cefdinir cefditoren pivoxil cefepime cefetamet pivoxil HCl cefixime cefmenoxime HCl cefminox sodium cefodizime sodium cefonicid sodium ceforanide cefoselis cefotetan disodium cefotiam hexetil HCl cefozopran HCl
hypnotic cerebral vasodilator immunomodulator cardiostimulant antiParkinsonian antihypertensive antimalarial antihypertensive antidepressant hormone anxiolytic topical antifungal antiinflammatory topical antifungal hypnotic topical antiinflammatory antiprolactin wound healing agent hypertensive antipsoriatic antineoplastic antihypertension antineoplastic antihypertensive agent antibiotic congestive heart failure antibiotic antihypertensive antifungal antibiotic antibiotic antibiotic oral cephalosporin antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic injectable cephalosporin
1983 1986 1987 1984 1997 1983 2000 1985 1989 1984 1985 1992 1992 1986 1984 1983 1993 1983 1988 1991 1985 1997 1998 1982 1986 1995 1988 1991 2001 1985 1997 1991 1994 1993 1992 1987 1983 1987 1990 1984 1984 1998 1984 1991 1995
19, 22, 23, 20, 33, 19, 36, 21, 25, 20, 21, 28, 28, 22, 20, 19, 29, 19, 24, 27, 21, 33, 34, 13, 22, 31, 24, 27, 37, 21, 33, 27, 30, 29, 28, 23, 19, 23, 26, 20, 20, 34, 20, 27, 31,
315 317 329 316 330 315 299 324 310 316 324 327 327 318 316 316 334 316 298 323 325 330 319 086 318 339 298 323 263 325 330 323 297 334 327 329 316 330 300 316 317 319 317 324 339
671
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
cefpimizole cefpiramide sodium cefpirome sulfate cefpodoxime proxetil cefprozil ceftazidime cefteram pivoxil ceftibuten ceftobiprole medocaril cefuroxime axetil cefuzonam sodium celecoxib celiprolol HCl centchroman centoxin cerivastatin certolizumab pegol cetirizine HCl cetrorelix cetuximab cevimeline hydrochloride chenodiol CHF-1301 choline alfoscerate choline fenofibrate cibenzoline ciclesonide cicletanine cidofovir cilazapril cilostazol cimetropium bromide cinacalcet cinildipine cinitapride cinolazepam ciprofibrate ciprofloxacin cisapride cisatracurium besilate citalopram cladribine clarithromycin clevidipine clevudine
antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antibiotic antiarthritic antihypertensive antiestrogen immunomodulator dyslipidemia crohn’s disease antihistamine female infertility anticancer anti-xerostomia anticholelithogenic antiparkinsonian nootropic dyslipidemia antiarrhythmic asthma, COPD antihypertensive antiviral antihypertensive antithrombotic antispasmodic hyperparathyroidism antihypertensive gastroprokinetic hypnotic hypolipidemic antibacterial gastroprokinetic muscle relaxant antidepressant antineoplastic antibiotic antihypertensive hepatitis B
1987 1985 1992 1989 1992 1983 1987 1992 2008 1987 1987 1999 1983 1991 1991 1997 2008 1987 1999 2003 2000 1983 1999 1990 2008 1985 2005 1988 1996 1990 1988 1985 2004 1995 1990 1993 1985 1986 1988 1995 1989 1993 1990 2008 2007
23, 21, 28, 25, 28, 19, 23, 28, 44, 23, 23, 35, 19, 27, 27, 33, 44, 23, 35, 39, 36, 19, 35, 26, 44, 21, 41, 24, 32, 26, 24, 21, 40, 31, 26, 29, 21, 22, 24, 31, 25, 29, 26, 44, 43,
330 325 328 310 328 316 330 329 589 331 331 335 317 324 325 331 592 331 336 272 299 317 336 300 594 325 443 299 306 301 299 326 451 339 301 334 326 318 299 340 311 335 302 596 466
672
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
clobenoside cloconazole HCl clodronate disodium clofarabine clopidogrel hydrogensulfate cloricromen clospipramine HCl colesevelam hydrochloride colestimide colforsin daropate HCl conivaptan crotelidae polyvalent immune fab cyclosporine cytarabine ocfosfate dabigatran etexilate dalfopristin dapiprazole HCl daptomycin darifenacin darunavir dasatinib decitabine defeiprone deferasirox defibrotide deflazacort delapril delavirdine mesylate denileukin diftitox denopamine deprodone propionate desflurane desloratadine desvenlafaxine dexfenfluramine dexibuprofen dexmedetomidine hydrochloride dexmethylphenidate HCl dexrazoxane dezocine diacerein didanosine dilevalol
vasoprotective topical antifungal calcium regulator anticancer antithrombotic antithrombotic neuroleptic hypolipidemic hypolipidaemic cardiotonic hyponatremia antidote
1988 1986 1986 2005 1998 1991 1991 2000 1999 1999 2006 2001
24, 22, 22, 41, 34, 27, 27, 36, 35, 35, 42, 37,
300 318 319 444 320 325 325 300 337 337 514 263
immunosuppressant antineoplastic anti-coagulant antibiotic antiglaucoma antibiotic urinary incontinence HIV anticancer myelodysplastic syndromes iron chelator chronic iron overload antithrombotic antiinflammatory antihypertensive antiviral anticancer cardiostimulant topical antiinflammatory anesthetic antihistamine antidepressant antiobesity antiinflammatory sedative
1983 1993 2008 1999 1987 2003 2005 2006 2006 2006 1995 2005 1986 1986 1989 1997 1999 1988 1992 1992 2001 2008 1997 1994 2000
19, 29, 44, 35, 23, 39, 41, 42, 42, 42, 31, 41, 22, 22, 25, 33, 35, 24, 28, 28, 37, 44, 33, 30, 36,
317 335 598 338 332 272 445 515 517 519 340 446 319 319 311 331 338 300 329 329 264 600 332 298 301
psychostimulant cardioprotective analgesic antirheumatic antiviral antihypertensive
2002 1992 1991 1985 1991 1989
38, 28, 27, 21, 27, 25,
352 330 326 326 326 311
Cumulative NCE Introduction Index, 1983–2008
673
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
dirithromycin disodium pamidronate divistyramine docarpamine docetaxel dofetilide dolasetron mesylate donepezil HCl dopexamine doripenem dornase alfa dorzolamide HCL dosmalfate doxacurium chloride doxazosin mesylate doxefazepam doxercalciferol doxifluridine doxofylline dronabinol drospirenone drotrecogin alfa droxicam droxidopa duloxetine dutasteride duteplase eberconazole ebastine ebrotidine ecabet sodium eculizumab edaravone efalizumab efavirenz efonidipine egualen sodium eletriptan emedastine difumarate emorfazone emtricitabine enalapril maleate enalaprilat encainide HCl enfuvirtide
antibiotic calcium regulator hypocholesterolemic cardiostimulant antineoplastic antiarrhythmic antiemetic anti-Alzheimer cardiostimulant antibiotic cystic fibrosis antiglaucoma antiulcer muscle relaxant antihypertensive hypnotic vitamin D prohormone antineoplastic bronchodilator antinauseant contraceptive antisepsis antiinflammatory antiparkinsonian antidepressant 5a reductase inhibitor anticougulant antifungal antihistamine antiulcer antiulcerative hemoglobinuria neuroprotective psoriasis antiviral antihypertensive antiulcer antimigraine antiallergic/antiasthmatic analgesic antiviral antihypertensive antihypertensive antiarrhythmic antiviral
1993 1989 1984 1994 1995 2000 1998 1997 1989 2005 1994 1995 2000 1991 1988 1985 1999 1987 1985 1986 2000 2001 1990 1989 2004 2002 1995 2005 1990 1997 1993 2007 2001 2003 1998 1994 2000 2001 1993 1984 2003 1984 1987 1987 2003
29, 336 25, 312 20, 317 30, 298 31, 341 36, 301 34, 321 33, 332 25, 312 41, 448 30, 298 31, 341 36, 302 27, 326 24, 300 21, 326 35, 339 23, 332 21, 327 22, 319 36, 302 37, 265 26, 302 25, 312 40, 452 38, 353 31, 342 41, 449 26 302 33, 333 29, 336 43, 468 37, 265 39, 274 34, 321 30, 299 36, 303 37, 266 29, 336 20, 317 39, 274 20, 317 23, 332 23, 333 39, 275
674
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
enocitabine enoxacin enoxaparin enoximone enprostil entacapone entecavir epalrestat eperisone HCl epidermal growth factor epinastine epirubicin HCl eplerenone epoprostenol sodium eprosartan eptazocine HBr eptilfibatide erdosteine erlotinib ertapenem sodium erythromycin acistrate erythropoietin escitalopram oxolate esmolol HCl esomeprazole magnesium eszopiclone ethyl icosapentate etizolam etodolac etoricoxibe etravirine everolimus exemestane exenatide exifone ezetimibe factor VIIa factor VIII fadrozole HCl falecalcitriol famciclovir famotidine fasudil HCl felbamate felbinac
antineoplastic antibacterial antithrombotic cardiostimulant antiulcer antiparkinsonian antiviral antidiabetic muscle relaxant wound healing agent antiallergic antineoplastic antihypertensive platelet aggreg. inhib. antihypertensive analgesic antithrombotic expectorant anticancer antibacterial antibiotic hematopoetic antidepressant antiarrhythmic gastric antisecretory hypnotic antithrombotic anxiolytic antiinflammatory antiarthritic/analgesic antiviral immunosuppressant anticancer anti-diabetic nootropic hypolipidemic haemophilia hemostatic antineoplastic vitamin D antiviral antiulcer neuroprotective antiepileptic topical antiinflammatory
1983 1986 1987 1988 1985 1998 2005 1992 1983 1987 1994 1984 2003 1983 1997 1987 1999 1995 2004 2002 1988 1988 2002 1987 2000 2005 1990 1984 1985 2002 2008 2004 2000 2005 1988 2002 1996 1992 1995 2001 1994 1985 1995 1993 1986
19, 22, 23, 24, 21, 34, 41, 28, 19, 23, 30, 20, 39, 19, 33, 23, 35, 31, 40, 38, 24, 24, 38, 23, 36, 41, 26, 20, 21, 38, 44, 40, 36, 41, 24, 38, 32, 28, 31, 37, 30, 21, 31, 29, 22,
318 320 333 301 327 322 450 330 318 333 299 318 276 318 333 334 340 342 454 353 301 301 354 334 303 451 303 318 327 355 602 455 304 452 302 355 307 330 342 266 300 327 343 337 320
675
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
felodipine fenbuprol fenoldopam mesylate fenticonazole nitrate fesoterodine fexofenadine filgrastim finasteride fisalamine fleroxacin flomoxef sodium flosequinan fluconazole fludarabine phosphate flumazenil flunoxaprofen fluoxetine HCl flupirtine maleate flurithromycin ethylsuccinate flutamide flutazolam fluticasone furoate fluticasone propionate flutoprazepam flutrimazole flutropium bromide fluvastatin fluvoxamine maleate follitropin alfa follitropin beta fomepizole fomivirsen sodium fondaparinux sodium formestane formoterol fumarate fosamprenavir fosaprepitant dimeglumine foscarnet sodium fosfosal fosfluconazole fosinopril sodium fosphenytoin sodium fotemustine fropenam
antihypertensive choleretic antihypertensive antifungal overactive bladder antiallergic immunostimulant 5a-reductase inhibitor intestinal antiinflammatory antibacterial antibiotic cardiostimulant antifungal antineoplastic benzodiazepine antag. antiinflammatory antidepressant analgesic antibiotic
1988 1983 1998 1987 2008 1996 1991 1992 1984 1992 1988 1992 1988 1991 1987 1987 1986 1985 1997
24, 19, 34, 23, 44, 32, 27, 28, 20, 28, 24, 28, 24, 27, 23, 23, 22, 21, 33,
302 318 322 334 604 307 327 331 318 331 302 331 303 327 335 335 320 328 333
antineoplastic anxiolytic anti-allergy antiinflammatory anxiolytic topical antifungal antitussive hypolipaemic antidepressant fertility enhancer fertility enhancer antidote antiviral antithrombotic antineoplastic bronchodilator antiviral anti-emetic antiviral analgesic antifungal antihypertensive antiepileptic antineoplastic antibiotic
1983 1984 2007 1990 1986 1995 1988 1994 1983 1996 1996 1998 1998 2002 1993 1986 2003 2008 1989 1984 2004 1991 1996 1989 1997
19, 20, 43, 26, 22, 31, 24, 30, 19, 32, 32, 34, 34, 38, 29, 22, 39, 44, 25, 20, 40, 27, 32, 25, 33,
318 318 469 303 320 343 303 300 319 307 308 323 323 356 337 321 277 606 313 319 457 328 308 313 334
676
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
frovatriptan fudosteine fulveristrant gabapentin gadoversetamide gallium nitrate gallopamil HCl galsulfase ganciclovir ganirelix acetate garenoxacin gatilfloxacin gefitinib gemcitabine HCl gemeprost gemifloxacin gemtuzumab ozogamicin gestodene gestrinone glatiramer acetate glimepiride glucagon, rDNA GMDP goserelin granisetron HCl guanadrel sulfate gusperimus halobetasol propionate halofantrine halometasone histrelin hydrocortisone aceponate hydrocortisone butyrate ibandronic acid ibopamine HCl ibudilast ibutilide fumarate ibritunomab tiuxetan icatibant idarubicin HCl idebenone idursulfase
antimigraine expectorant anticancer antiepileptic MRI contrast agent calcium regulator antianginal mucopolysaccharidosis VI antiviral female infertility anti-infective antibiotic antineoplastic antineoplastic abortifacient antibacterial anticancer
2002 2001 2002 1993 2000 1991 1983 2005 1988 2000 2007 1999 2002 1995 1983 2004 2000
38, 37, 38, 29, 36, 27, 19, 41, 24, 36, 43, 35, 38, 31, 19, 40, 36,
357 267 357 338 304 328 319 453 303 305 471 340 358 344 319 458 306
progestogen antiprogestogen Multiple Sclerosis antidiabetic hypoglycemia immunostimulant hormone antiemetic antihypertensive immunosuppressant topical antiinflammatory antimalarial topical antiinflammatory precocious puberty topical antiinflammatory topical antiinflammatory osteoporosis cardiostimulant antiasthmatic antiarrhythmic anticancer hereditary angiodema antineoplastic nootropic mucopolysaccharidosis II (Hunter syndrome) platelet aggreg. inhibitor
1987 1986 1997 1995 1993 1996 1987 1991 1983 1994 1991 1988 1983 1993 1988 1983 1996 1984 1989 1996 2002 2008 1990 1986 2006
23, 22, 33, 31, 29, 32, 23, 27, 19, 30, 27, 24, 19, 29, 24, 19, 32, 20, 25, 32, 38, 44, 26, 22, 42,
335 321 334 344 338 308 336 329 319 300 329 304 320 338 304 320 309 319 313 309 359 608 303 321 520
1992
28, 332
iloprost
677
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
imatinib mesylate imidafenacin imidapril HCl imiglucerase imipenem/cilastatin imiquimod incadronic acid indalpine indeloxazine HCl indinavir sulfate indisetron indobufen influenza virus (live) insulin lispro interferon alfacon-1 interferon gamma-1b interferon, gamma interferon, gamma-1a interferon, b-1a interferon, b-1b interleukin-2 ioflupane ipriflavone irbesartan irinotecan irsogladine isepamicin isofezolac isoxicam isradipine itopride HCl itraconazole ivabradine ivermectin ixabepilone ketanserin ketorolac tromethamine kinetin
antineoplastic overactive bladder antihypertensive Gaucher’s disease antibiotic antiviral osteoporosis antidepressant nootropic antiviral antiemetic antithrombotic antiviral vaccine antidiabetic antiviral immunostimulant antiinflammatory antineoplastic multiple sclerosis multiple sclerosis antineoplastic diagnosis CNS calcium regulator antihypertensive antineoplastic antiulcer antibiotic antiinflammatory antiinflammatory antihypertensive gastroprokinetic antifungal angina antiparasitic anticancer antihypertensive analgesic skin photodamage/ dermatologic antihypertensive anticonvulsant gastric antisecretory antiviral anticonvulsant antiarrhythmic
2001 2007 1993 1994 1985 1997 1997 1983 1988 1996 2004 1984 2003 1996 1997 1991 1989 1992 1996 1993 1989 2000 1989 1997 1994 1989 1988 1984 1983 1989 1995 1988 2006 1987 2007 1985 1990 1999
37, 43, 29, 30, 21, 33, 33, 19, 24, 32, 40, 20, 39, 32, 33, 27, 25, 28, 32, 29, 25, 36, 25, 33, 30, 25, 24, 20, 19, 25, 31, 24, 42, 23, 43, 21, 26, 35,
267 472 339 301 328 335 335 320 304 310 459 319 277 310 336 329 314 332 311 339 314 306 314 336 301 315 305 319 320 315 344 305 522 336 473 328 304 341
1991 2008 2000 1995 1990 2002
27, 44, 36, 31, 26, 38,
330 610 307 345 304 360
lacidipine lacosamide lafutidine lamivudine lamotrigine landiolol
678
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
lanoconazole lanreotide acetate lansoprazole lapatinib laronidase latanoprost lefunomide lenalidomide
antifungal acromegaly antiulcer anticancer mucopolysaccaridosis I antiglaucoma antiarthritic myelodysplastic syndromes, multiple myeloma antibiotic immunostimulant anticoagulant antihyperintensive anticancer hormone nootropic antiasthmatic antiepileptic antiglaucoma local anesthetic
1994 1995 1992 2007 2003 1996 1998 2006
30, 31, 28, 43, 39, 32, 34, 42,
302 345 332 475 278 311 324 523
1987 1986 1997 1997 1996 1984 1986 1999 2000 1985 2000
23, 22, 33, 33, 32, 20, 22, 35, 36, 21, 36,
336 322 336 337 311 319 322 341 307 328 308
antihistamine antihistamine antitussive antibiotic heart failure antiperistaltic antithrombotic antibiotic topical antifungal ADHD antihypertensive antiinflammatory antiallergic ophthalmic antibiotic antimigraine antineoplastic antiviral hypnotic cardiostimulant antibiotic antihistamine NSAID antihypertensive antiallergic ophthalmic
1991 2001 1988 1993 2000 1984 1988 2000 2000 2007 1987 1986 1992 1989 1999 1987 2000 1983 1996 1992 1988 1997 1994 1998
27, 37, 24, 29, 36, 20, 24, 36, 36, 43, 23, 22, 28, 25, 35, 23, 36, 19, 32, 28, 24, 33, 30, 34,
330 268 305 340 308 320 306 309 309 477 337 322 333 315 342 337 310 321 312 333 306 337 302 324
lenampicillin HCl lentinan lepirudin lercanidipine letrazole leuprolide acetate levacecarnine HCl levalbuterol HCl levetiracetam levobunolol HCl levobupivacaine hydrochloride levocabastine HCl levocetirizine levodropropizine levofloxacin levosimendan lidamidine HCl limaprost linezolid liranaftate lisdexamfetamine lisinopril lobenzarit sodium lodoxamide tromethamine lomefloxacin lomerizine HCl lonidamine lopinavir loprazolam mesylate loprinone HCl loracarbef loratadine lornoxicam losartan loteprednol etabonate
679
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
lovastatin loxoprofen sodium lulbiprostone
hypocholesterolemic antiinflammatory chronic idiopathic constipation antifungal anti-inflammatory vaccine bronchodilator hepatoprotective antihypertensive anti-infective – HIV topical antineoplastic vitamin D antihypertensive antidepressant antimalarial hypolipidemic hypocholesterolemic antiarthritic analeptic analgesic carbapenem antibiotic anxiolytic antidepressant opioid-induced constipation anxiolytic antifungal abortifacient antidiabetic gaucher’s disease antidepressant cardiostimulant topical antineoplastic antibiotic antidepressant antiulcer antidiabetic antineoplastic muscle relaxant hepatoprotectant antihistamine immunosuppressant antidepressant idiopathic hypersomnia antihypertensive
1987 1986 2006
23, 337 22, 322 42, 525
2005 2005 1999 1986 1985 1990 2007 1992 2000 1997 1986 1985 1983 1984 1996 1984 1983 1994 1987 1984 2008 1984 2002 1988 1998 2003 1997 1989 1993 1985 1994 1985 2004 1984 1992 1999 1998 1984 1990 1994 1995
41, 41, 35, 22, 21, 26, 43, 28, 36, 33, 22, 21, 19, 20, 32, 20, 19, 30, 23, 20, 44, 20, 38, 24, 34, 39, 33, 25, 29, 21, 30, 21, 40, 20, 28, 35, 34, 20, 26, 30, 31,
luliconazole lumiracoxib Lyme disease mabuterol HCl malotilate manidipine HCl maraviroc masoprocol maxacalcitol mebefradil HCl medifoxamine fumarate mefloquine HCl meglutol melinamide meloxicam mepixanox meptazinol HCl meropenem metaclazepam metapramine methylnaltrexone bromide mexazolam micafungin mifepristone miglitol miglustat milnacipran milrinone miltefosine miokamycin mirtazapine misoprostol mitiglinide mitoxantrone HCl mivacurium chloride mivotilate mizolastine mizoribine moclobemide modafinil moexipril HCl
454 455 342 323 329 304 478 333 310 338 323 329 321 320 312 320 321 303 338 320 612 321 360 306 325 279 338 316 340 329 303 329 460 321 334 343 325 321 305 303 346
680
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
mofezolac mometasone furoate montelukast sodium moricizine HCl mosapride citrate moxifloxacin HCL moxonidine mozavaptan mupirocin muromonab-CD3 muzolimine mycophenolate mofetil mycophenolate sodium nabumetone nadifloxacin nafamostat mesylate nafarelin acetate naftifine HCl naftopidil nalmefene HCl naltrexone HCl naratriptan HCl nartograstim natalizumab nateglinide nazasetron nebivolol nedaplatin nedocromil sodium nefazodone nelarabine nelfinavir mesylate neltenexine nemonapride nepafenac neridronic acide nesiritide neticonazole HCl nevirapine nicorandil nifekalant HCl nilotinib nilutamide nilvadipine nimesulide
analgesic topical antiinflammatory antiasthma antiarrhythmic gastroprokinetic antibiotic antihypertensive hyponatremia topical antibiotic immunosuppressant diuretic immunosuppressant immunosuppressant antiinflammatory topical antibiotic protease inhibitor hormone antifungal dysuria dependence treatment narcotic antagonist antimigraine leukopenia multiple sclerosis antidiabetic antiemetic antihypertensive antineoplastic antiallergic antidepressant anticancer antiviral cystic fibrosis neuroleptic anti-inflammatory calcium regulator congestive heart failure topical antifungal antiviral coronary vasodilator antiarrythmic anticancer – CML antineoplastic antihypertensive antiinflammatory
1994 1987 1998 1990 1998 1999 1991 2006 1985 1986 1983 1995 2003 1985 1993 1986 1990 1984 1999 1995 1984 1997 1994 2004 1999 1994 1997 1995 1986 1994 2006 1997 1993 1991 2005 2002 2001 1993 1996 1984 1999 2007 1987 1989 1985
30, 23, 34, 26, 34, 35, 27, 42, 21, 22, 19, 31, 39, 21, 29, 22, 26, 20, 35, 31, 20, 33, 30, 40, 35, 30, 33, 31, 22, 30, 42, 33, 29, 27, 41, 38, 37, 29, 32, 20, 35, 43, 23, 25, 21,
304 338 326 305 326 343 330 527 330 323 321 346 279 330 340 323 306 321 344 347 322 339 304 462 344 305 339 347 324 305 528 340 341 331 456 361 269 341 313 322 344 480 338 316 330
681
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
nimodipine nimotuzumab nipradilol nisoldipine nitisinone nitrefazole nitrendipine nizatidine nizofenzone fumarate nomegestrol acetate norelgestromin norfloxacin norgestimate OCT-43 octreotide ofloxacin olanzapine olimesartan Medoxomil olopatadine HCl omalizumab omeprazole ondansetron HCl OP-1 orlistat ornoprostil osalazine sodium oseltamivir phosphate oxaliplatin oxaprozin oxcarbazepine oxiconazole nitrate oxiracetam oxitropium bromide ozagrel sodium paclitaxal palifermin paliperidone palonosetron panipenem/betamipron panitumumab pantoprazole sodium parecoxib sodium paricalcitol parnaparin sodium paroxetine
cerebral vasodilator anticancer antihypertensive antihypertensive antityrosinaemia alcohol deterrent hypertensive antiulcer nootropic progestogen contraceptive antibacterial progestogen anticancer antisecretory antibacterial neuroleptic antihypertensive antiallergic allergic asthma antiulcer antiemetic osteoinductor antiobesity antiulcer intestinal antinflamm. antiviral anticancer antiinflammatory anticonvulsant antifungal nootropic bronchodilator antithrombotic antineoplastic mucositis antipsychotic antiemetic carbapenem antibiotic anticancer antiulcer analgesic vitamin D anticoagulant antidepressant
1985 2006 1988 1990 2002 1983 1985 1987 1988 1986 2002 1983 1986 1999 1988 1985 1996 2002 1997 2003 1988 1990 2001 1998 1987 1986 1999 1996 1983 1990 1983 1987 1983 1988 1993 2005 2007 2003 1994 2006 1995 2002 1998 1993 1991
21, 42, 24, 26, 38, 19, 21, 23, 24, 22, 38, 19, 22, 35, 24, 21, 32, 38, 33, 39, 24, 26, 37, 34, 23, 22, 35, 32, 19, 26, 19, 23, 19, 24, 29, 41, 43, 39, 30, 42, 30, 38, 34, 29, 27,
330 529 307 306 361 322 331 339 307 324 362 322 324 345 307 331 313 363 340 280 308 306 269 327 339 324 346 313 322 307 322 339 323 308 342 461 482 281 305 531 306 364 327 342 331
682
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
pazufloxacin pefloxacin mesylate pegademase bovine pegaptanib
antibacterial antibacterial immunostimulant age-related macular degeneration antineoplastic acromegaly anticancer antiasthmatic antiviral antineoplastic antiparkinsonian antihypertensive neuroleptic antithrombotic immunostimulant topical antiinflammatory antiarrhythmic topical antiinflammatory immunosuppressant heart failure antihypertensive antidiabetic antineoplastic idiopathic pulmonary fibrosis antiarrhythmic antiinflammatory hypocholesterolemic antidepressant antiulcer antiulcer antineoplastic adjuvant antifungal antiParkinsonian cognition enhancer anti-diabetic antiasthmatic antilipidemic topical antiinflammatory antiepileptic vulnery anticonvulsant progestogen analgesic antiviral
2002 1985 1990 2005
38, 21, 26, 41,
364 331 307 458
1994 2003 2004 1991 1996 1992 1988 1988 2001 1987 1993 1984 1991 1984 2002 1994 1987 1999 1988 2008 1994 1988 2003 1997 1987 1994 1993 2006 1997 1993 2005 1995 1989 1986 2004 1996 1985 1983 1986 1994
30, 39, 40, 27, 32, 28, 24, 24, 37, 23, 29, 20, 27, 20, 38, 30, 23, 35, 24, 44, 30, 24, 39, 33, 23, 30, 29, 42, 33, 29, 41, 31, 25, 22, 40, 32, 21, 19, 22, 30,
306 281 463 331 314 334 308 309 270 340 343 322 332 322 365 307 340 346 309 614 307 309 282 341 340 307 343 532 341 343 460 347 316 325 464 314 331 323 325 308
pegaspargase pegvisomant pemetrexed pemirolast potassium penciclovir pentostatin pergolide mesylate perindopril perospirone HCL picotamide pidotimod piketoprofen pilsicainide HCl pimaprofen pimecrolimus pimobendan pinacidil pioglitazone HCL pirarubicin pirfenidone pirmenol piroxicam cinnamate pitavastatin pivagabine plaunotol polaprezinc porfimer sodium posaconazole pramipexole HCl pramiracetam H2SO4 pramlintide pranlukast pravastatin prednicarbate pregabalin prezatide copper acetate progabide promegestrone propacetamol HCl propagermanium
683
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
propentofylline propionate propiverine HCl propofol prulifloxacin pumactant quazepam quetiapine fumarate quinagolide quinapril quinfamide quinupristin rabeprazole sodium raloxifene HCl raltegravir raltitrexed ramatroban ramelteon ramipril ramosetron ranibizumab
cerebral vasodilator urologic anesthetic antibacterial lung surfactant hypnotic neuroleptic hyperprolactinemia antihypertensive amebicide antibiotic gastric antisecretory osteoporosis anti-infective – HIV anticancer antiallergic insomnia antihypertensive antiemetic age-related macular degeneration antineoplastic antiulcer angina muscle relaxant parkinson’s disease antiulcer antidepressant analgesic antipsychotic antidiabetic antiallergic anti-infective fibrinolytic anticoagulant antibacterial antibacterial antibiotic antibiotic hypnotic antihypertensive genetic autoinflammatory syndromes neuroprotective antiviral
1988 1992 1986 2002 1994 1985 1997 1994 1989 1984 1999 1998 1998 2007 1996 2000 2005 1989 1996 2006
24, 28, 22, 38, 30, 21, 33, 30, 25, 20, 35, 34, 34, 43, 32, 36, 41, 25, 32, 42,
310 335 325 366 308 332 341 309 317 322 338 328 328 484 315 311 462 317 315 534
1987 1995 2006 1999 2005 1990 1997 1996 1990 1998 1987 2007 1996 1993 1992 1988 1985 1987 1989 1988 2008
23, 31, 42, 35, 41, 26, 33, 32, 26, 34, 23, 43, 32, 29, 28, 24, 21, 23, 25, 24, 44,
341 348 535 347 464 308 342 316 308 329 341 486 316 344 335 310 332 341 317 310 615
1996 1987
32, 316 23, 342
ranimustine ranitidine bismuth citrate ranolazine rapacuronium bromide rasagiline rebamipide reboxetine remifentanil HCl remoxipride HCl repaglinide repirinast retapamulin reteplase reviparin sodium rifabutin rifapentine rifaximin rifaximin rilmazafone rilmenidine rilonacept riluzole rimantadine HCl
684
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
rimexolone rimonabant risedronate sodium risperidone ritonavir rivaroxaban
antiinflammatory anti-obesity osteoporosis neuroleptic antiviral anticoagulant, venous thromboembolism anti-Alzheimer antimigraine neuromuscular blocker antiarthritic antibiotic antithrombocytopenic immunostimulant hypolipidemic antiParkinsonian anesthetic antiulcer antidiabetic hypocholesterolemic parkinson’s disease antiulcer antiulcer anticonvulsant antibacterial antiallergic antibiotic bronchodilator
1995 2006 1998 1993 1996 2008
31, 42, 34, 29, 32, 44,
348 537 330 344 317 617
1997 1998 1994 1999 1986 2008 1991 1986 1996 1996 1985 1999 2003 2006 1986 1987 2007 1992 2003 1989 1990
33, 34, 30, 35, 22, 44, 27, 22, 32, 32, 21, 35, 39, 42, 22, 23, 43, 28, 39, 25, 26,
342 330 309 347 325 619 332 326 317 318 332 348 283 538 326 342 488 335 284 318 308
hyperphenylalaninemia antiviral immunostimulant platelet antiaggregant immunostimulant antiasthmatic topical antifungal neuroleptic antihistamine antidepressant antidepressant anesthetic antiobesity male sexual dysfunction dysuria hypocholesterolemic
1992 1995 1991 1993 1985 1995 1992 1996 1987 1989 1990 1990 1998 1998 2006 1988
8, 336 31 349 27, 332 29, 344 22, 326 31, 349 28, 336 32, 318 23, 342 25, 318 26, 309 26, 309 34, 331 34, 331 42, 540 24, 311
rivastigmin rizatriptan benzoate rocuronium bromide rofecoxib rokitamycin romiplostim romurtide ronafibrate ropinirole HCl ropivacaine rosaprostol rosiglitazone maleate rosuvastatin rotigotine roxatidine acetate HCl roxithromycin rufinamide rufloxacin HCl rupatadine fumarate RV-11 salmeterol hydroxynaphthoate sapropterin HCl saquinavir mesvlate sargramostim sarpogrelate HCl schizophyllan seratrodast sertaconazole nitrate sertindole setastine HCl setiptiline setraline HCl sevoflurane sibutramine sildenafil citrate silodosin simvastatin
685
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
sitafloxacin hydrate sitagliptin sitaxsentan sivelestat SKI-2053R sobuzoxane sodium cellulose PO4 sofalcone solifenacin somatomedin-1
antibacterial antidiabetic pulmonary hypertension anti-inflammatory anticancer antineoplastic hypocalciuric antiulcer pollakiuria growth hormone insensitivity growth hormone hormone anticancer antiviral antibiotic antihypertensive antiulcer antiviral osteoporosis chelator analgesic reversal of neuromuscular blockade b-lactamase inhibitor topical antifungal antibiotic antimigraine anticancer antiallergic analgesic respiratory surfactant topical antipsoriatic Alzheimer’s disease immunosuppressant male sexual dysfunction antiglaucoma anticancer antiParkinsonian CNS stimulant anticancer antiprostatic hypertrophy anxiolytic anticancer antiallergic
2008 2006 2006 2002 1999 1994 1983 1984 2004 1994
44, 42, 42, 38, 35, 30, 19, 20, 40, 30,
621 541 543 366 348 310 323 323 466 310
1994 1987 2005 1993 1993 1995 1987 1994 2004 1991 1983 2008
30, 23, 41, 29, 29, 31, 23, 30, 40, 27, 19, 44,
310 343 466 345 345 349 343 311 466 333 323 623
1986 1985 1987 1991 2006 1995 1983 1987 1993 1993 1993 2003 2008 2004 1996 2000 2005 1993 1996 1999 1990
22, 21, 23, 27, 42, 31, 19, 23, 29, 29, 29, 39, 44, 40, 32, 36, 41, 29, 32, 35, 26,
326 332 343 333 544 350 324 344 346 346 347 284 625 469 318 311 467 347 319 349 309
somatotropin somatropin sorafenib sorivudine sparfloxacin spirapril HCl spizofurone stavudine strontium ranelate succimer sufentanil sugammadex sulbactam sodium sulconizole nitrate sultamycillin tosylate sumatriptan succinate sunitinib suplatast tosilate suprofen surfactant TA tacalcitol tacrine HCl tacrolimus tadalafil tafluprost talaporfin sodium talipexole taltirelin tamibarotene tamsulosin HCl tandospirone tasonermin tazanolast
686
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
tazarotene tazobactam sodium tegaserod maleate teicoplanin telbivudine telithromycin telmesteine telmisartan temafloxacin HCl temocapril temocillin disodium temoporphin
antipsoriasis b-lactamase inhibitor irritable bowel syndrome antibacterial hepatitis B antibiotic mucolytic antihypertensive antibacterial antihypertensive antibiotic antineoplastic/ photosensitizer anticancer anticancer antiviral
1997 1992 2001 1988 2006 2001 1992 1999 1991 1994 1984 2002
33, 28, 37, 24, 42, 37, 28, 35, 27, 30, 20, 38,
1999 2007 2001
35, 349 43, 490 37, 271
antiinflammatory antiulcer antihypertensive antifungal antifungal antihypertensive hemostat anticoagulant
1987 1984 1984 1991 1983 1987 2008 2008
23, 20, 20, 27, 19, 23, 44, 44,
344 323 323 334 324 344 627 628
immunomodulator antiepileptic antihypertensive antidepressant anabolic antibiotic antihypertensive Paget’s disease neuroleptic nasal decongestant antifungal urolithiasis bronchodilator HIV antispasmodic antiarrhythmic subarachnoid hemorrhage antithrombotic
1985 1996 1988 1983 1988 2005 1992 1995 1984 1988 1983 1989 2002 2005 1984 1990 1995
21, 32, 24, 19, 24, 41, 28, 31, 20, 24, 19, 25, 38, 41, 20, 26, 31,
333 319 311 324 312 468 337 350 323 312 324 318 368 470 324 310 351
1998
34, 332
temozolomide temsirolimus tenofovir disoproxil fumarate tenoxicam teprenone terazosin HCl terbinafine HCl terconazole tertatolol HCl thrombin alfa thrombomodulin (recombinant) thymopentin tiagabine tiamenidine HCl tianeptine sodium tibolone tigecycline tilisolol HCl tiludronate disodium timiperone tinazoline tioconazole tiopronin tiotropium bromide tipranavir tiquizium bromide tiracizine HCl tirilazad mesylate tirofiban HCl
343 336 270 311 546 271 337 349 334 311 323 367
687
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
tiropramide HCl tizanidine tolcapone toloxatone tolrestat topiramate topotecan HCl torasemide toremifene tositumomab tosufloxacin tosylate trabectedin trandolapril travoprost treprostinil sodium tretinoin tocoferil trientine HCl trimazosin HCl trimegestone trimetrexate glucuronate
antispasmodic muscle relaxant antiParkinsonian antidepressant antidiabetic antiepileptic anticancer diuretic antineoplastic anticancer antibacterial anticancer antihypertensive antiglaucoma antihypertensive antiulcer chelator antihypertensive progestogen Pneumocystis carinii pneumonia antidiabetic antiemetic antibiotic
1983 1984 1997 1984 1989 1995 1996 1993 1989 2003 1990 2007 1993 2001 2002 1993 1986 1985 2001 1994
19, 20, 33, 20, 25, 31, 32, 29, 25, 39, 26, 43, 29, 37, 38, 29, 22, 21, 37, 30,
1997 1992 1998
33, 344 28, 337 34, 332
antiulcer immunostimulant erectile dysfunction antiglaucoma
1986 1987 2005 1994
22, 23, 41, 30,
327 345 472 312
antiviral antiarthritic antiviral anticancer antihypertensive male sexual dysfunction nicotine-dependence antidepressant photosensitizer cardiostimulant anticonvulsant antidiabetic antineoplastic antidiabetic antifungal
1995 2002 2001 1999 1996 2003 2006 1994 2000 1990 1989 2007 1989 1994 2002
31, 38, 37, 35, 32, 39, 42, 30, 36, 26, 25, 43, 25, 30, 38,
352 369 273 350 320 286 547 312 312 310 319 494 320 313 370
troglitazone tropisetron trovafloxacin mesylate troxipide ubenimex udenafil unoprostone isopropyl ester valaciclovir HCl vadecoxib vaglancirclovir HCL valrubicin valsartan vardenafil varenicline venlafaxine verteporfin vesnarinone vigabatrin vildagliptin vinorelbine voglibose voriconazole
324 324 343 324 319 351 320 348 319 285 310 492 348 272 368 348 327 333 273 312
688
Cumulative NCE Introduction Index, 1983–2008
GENERAL NAME
INDICATION
YEAR INTRO.
ARMC VOL., PAGE
vorinostat xamoterol fumarate ximelagatran zafirlukast zalcitabine zaleplon zaltoprofen zanamivir ziconotide zidovudine zileuton zinostatin stimalamer ziprasidone hydrochloride zofenopril calcium zoledronate disodium zolpidem hemitartrate zomitriptan zonisamide zopiclone zuclopenthixol acetate
anticancer cardiotonic anticoagulant antiasthma antiviral hypnotic antiinflammatory antiviral severe chronic pain antiviral antiasthma antineoplastic neuroleptic antihypertensive hypercalcemia hypnotic antimigraine anticonvulsant hypnotic antipsychotic
2006 1988 2004 1996 1992 1999 1993 1999 2005 1987 1997 1994 2000 2000 2000 1988 1997 1989 1986 1987
42, 24, 40, 32, 28, 35, 29, 35, 41, 23, 33, 30, 36, 36, 36, 24, 33, 25, 22, 23,
549 312 470 321 338 351 349 352 473 345 344 313 312 313 314 313 345 320 327 345
CUMULATIVE NCE INTRODUCTION INDEX, 1983–2008 (BY INDICATION)
GENERIC NAME
INDICATION
YEAR INTRO.
ARMC VOL., (PAGE)
gemeprost mifepristone lanreotide acetate pegvisomant lisdexamfetamine pegaptanib
ABORTIFACIENT
1983 1988 1995 2003 2007 2005
19 24 31 39 43 41
(319) (306) (345) (281) (477) (458)
2006 1983 2003 1993 1984 1988 1984 1983 1983 1991 1984 1987 2002 1985 1984 1990 1983 1994 2002 1986 1996 1983 1983 1992 1986 1996 1990 2000
42 19 39 29 20 24 20 19 19 27 20 23 38 21 20 26 19 30 38 22 32 19 19 28 22 32 26 36
(534) (322) (280) (346) (322) (312) (320) (314) (314) (326) (317) (334) (355) (328) (319) (304) (321) (304) (364) (325) (316) (323) (324) (329) (325) (318) (309) (308)
2006 2006 1989 2000 1993
42 42 25 36 29
(522) (535) (310) (297) (336)
ranibizumab nitrefazole omalizumab tacrine HCl quinfamide tibolone mepixanox alfentanil HCl alminoprofen dezocine emorfazone eptazocine HBr etoricoxib flupirtine maleate fosfosal ketorolac tromethamine meptazinol HCl mofezolac parecoxib sodium propacetamol HCl remifentanil HCl sufentanil suprofen desflurane propofol ropivacaine sevoflurane levobupivacaine hydrochloride ivabradine ranolazine azelaic acid betotastine besilate emedastine difumarate
ACROMEGALY ADHD AGE-RELATED MACULAR DEGENERATION ALCOHOL DETERRENT ALLERGIC ASTHMA ALZHEIMER’S DISEASE AMEBICIDE ANABOLIC ANALEPTIC ANALGESIC
ANESTHETIC
ANESTHETIC, LOCAL ANGINA ANTIACNE ANTIALLERGIC
691
692
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
epinastine fexofenadine nedocromil sodium olopatadine hydrochloride ramatroban repirinast suplatast tosilate tazanolast lodoxamide tromethamine rupatadine fumarate fluticasone furoate loteprednol etabonate donepezil hydrochloride rivastigmin gallopamil HCl cibenzoline dofetilide encainide HCl esmolol HCl ibutilide fumarate landiolol moricizine hydrochloride nifekalant HCl pilsicainide hydrochloride pirmenol tiracizine hydrochloride anakinra celecoxib etoricoxib meloxicam leflunomide rofecoxib valdecoxib amlexanox emedastine difumarate ibudilast levalbuterol HCl montelukast sodium pemirolast potassium seratrodast
INDICATION
YEAR INTRO.
ARMC VOL., (PAGE)
1994 1996 1986 1997
30 32 22 33
(299) (307) (324) (340)
ANTIALLERGIC
2000 1987 1995 1990 1992
36 23 31 26 28
(311) (341) (350) (309) (333)
ANTI-ALLERGY OPHTHALMIC
2003 2007 1998
39 (284) 43 (469) 34 (324)
ANTI-ALZHEIMERS
1997
33 (332)
1997 1983 1985 2000 1987 1987 1996 2002 1990
33 19 21 36 23 23 32 38 26
1999 1991
35 (344) 27 (332)
1994 1990
30 (307) 26 (310)
2001 1999 2002 1996 1998 1999 2002 1987 1993
37 35 38 32 34 35 38 23 29
(261) (335) (355) (312) (324) (347) (369) (327) (336)
1989 1999 1998 1991
25 35 34 27
(313) (341) (326) (331)
1995
31 (349)
ANTIANGINAL ANTIARRHYTHMIC
ANTIARTHRITIC
ANTIASTHMATIC
(342) (319) (325) (301) (333) (334) (309) (360) (305)
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
zafirlukast zileuton balofloxacin biapenem ciprofloxacin enoxacin ertapenem sodium fleroxacin gemifloxacin norfloxacin ofloxacin pazufloxacin pefloxacin mesylate pranlukast prulifloxacin rifabutin rifapentine rufloxacin hydrochloride sitafloxacin hydrate teicoplanin temafloxacin hydrochloride tosufloxacin tosylate arbekacin aspoxicillin astromycin sulfate azithromycin aztreonam brodimoprin carboplatin carumonam cefbuperazone sodium cefcapene pivoxil cefdinir cefepime cefetamet pivoxil hydrochloride cefixime cefmenoxime HCl cefminox sodium cefodizime sodium cefonicid sodium ceforanide cefoselis cefotetan disodium cefotiam hexetil hydrochloride
INDICATION
ANTIBACTERIAL
ANTIBIOTIC
693
YEAR INTRO.
ARMC VOL., (PAGE)
1996 1997 2002 2002 1986 1986 2002 1992 2004 1983 1985 2002 1985 1995 2002 1992 1988 1992
32 33 38 38 22 22 38 28 40 19 21 38 21 31 38 28 24 28
2008 1988 1991
44 (621) 24 (311) 27 (334)
1990 1990 1987 1985 1988 1984 1993 1986 1988 1985
26 26 23 21 24 20 29 22 24 21
(310) (298) (328) (324) (298) (315) (333) (318) (298) (325)
1997 1991 1993 1992
33 27 29 28
(330) (323) (334) (327)
1987 1983 1987 1990 1984 1984 1998 1984 1991
23 19 23 26 20 20 34 20 27
(329) (316) (330) (300) (316) (317) (319) (317) (324)
(321) (344) (351) (351) (318) (320) (353) (331) (458) (322) (331) (364) (331) (347) (366) (335) (310) (335)
694
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
cefpimizole cefpiramide sodium cefpirome sulfate cefpodoxime proxetil cefprozil ceftazidime cefteram pivoxil ceftibuten ceftobiprole medocaril cefuroxime axetil cefuzonam sodium clarithromycin dalfopristin dirithromycin doripenem erythromycin acistrate flomoxef sodium flurithromycin ethylsuccinate fropenam gatifloxacin imipenem/cilastatin isepamicin lenampicillin HCl levofloxacin linezolid lomefloxacin loracarbef miokamycin moxifloxacin HCl quinupristin rifaximin rifaximin rokitamycin RV-11 sparfloxacin sultamycillin tosylate telithromycin temocillin disodium tigecycline trovafloxacin mesylate meropenem panipenem/betamipron mupirocin nadifloxacin abarelix
INDICATION
ANTIBIOTIC, CARBAPENEM ANTIBIOTIC, TOPICAL ANTICANCER
YEAR INTRO.
ARMC VOL., (PAGE)
1987 1985 1992 1989
23 21 28 25
(330) (325) (328) (310)
1992 1983 1987 1992 2008 1987 1987 1990 1999 1993 2005 1988 1988 1997
28 19 23 28 44 23 23 26 35 29 41 24 24 33
(328) (316) (330) (329) (589) (331) (331) (302) (338) (336) (448) (301) (302) (333)
1997 1999 1985 1988 1987 1993 2000 1989 1992 1985 1999 1999 1985 1987 1986 1989 1993 1987
33 35 21 24 23 29 36 25 28 21 35 35 21 23 22 25 29 23
(334) (340) (328) (305) (336) (340) (309) (315) (333) (329) (343) (338) (332) (341) (325) (318) (345) (343)
2001 1984 2005 1998
37 20 41 34
(271) (323) (468) (332)
1994 1994 1985 1993 2004
30 30 21 29 40
(303) (305) (330) (340) (446)
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
alemtuzumab alitretinoin arglabin azacitidine belotecan bevacizumab bexarotene bortezomib cetuximab clofarabine dasatinib denileukin diftitox erlotinib exemestane fulvestrant gemtuzumab ozogamicin ibritumomab tiuxetan ixabepilone lapatinib letrazole nelarabine nimotuzumab OCT-43 oxaliplatin panitumumab pemetrexed raltitrexed SKI-2053R sorafenib sunitinib talaporfin sodium tamibarotene tasonermin temozolomide temsirolimus topotecan HCl tositumomab trabectedin valrubicin vorinostat angiotensin II nilotinib chenodiol dabigatran etexilate duteplase lepirudin parnaparin sodium
INDICATION
ANTICANCER ADJUVANT ANTICANCER – CML ANTICHOLELITHOGENIC ANTICOAGULANT
695
YEAR INTRO.
ARMC VOL., (PAGE)
2001 1999 1999 2004 2004 2004 2000 2003 2003 2005 2006 1999 2004 2000 2002 2000
37 35 35 40 40 40 36 39 39 41 42 35 40 36 38 36
2002
38 (359)
2007 2007 1996 2006 2006 1999 1996 2006 2004 1996 1999 2005 2006 2004 2005 1999 1999 2007 1996 2003 2007 1999 2006 1994 2007 1983 2008 1995 1997 1993
43 43 32 42 42 35 32 42 40 32 35 41 42 40 41 35 35 43 32 39 43 35 42 30 43 19 44 31 33 29
(260) (333) (335) (447) (449) (450) (298) (271) (272) (444) (517) (338) (454) (304) (357) (306)
(473) (475) (311) (528) (529) (345) (313) (531) (463) (315) (348) (466) (544) (469) (467) (349) (350) (490) (320) (285) (492) (350) (549) (296) (480) (317) (598) (342) (336) (342)
696
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
reviparin sodium rivaroxaban thrombomodulin (recombinant) ximelagatran lacosamide lamotrigine oxcarbazepine progabide rufinamide vigabatrin zonisamide bupropion HCl citalopram desvenlafaxine duloxetine escitalopram oxalate fluoxetine HCl fluvoxamine maleate indalpine medifoxamine fumarate metapramine milnacipran mirtazapine moclobemide nefazodone paroxetine pivagabine reboxetine setiptiline sertraline hydrochloride tianeptine sodium toloxatone venlafaxine acarbose epalrestat exenatide glimepiride insulin lispro miglitol mitiglinide nateglinide pioglitazone HCl pramlintide repaglinide
INDICATION
ANTICOAGULANT
ANTICONVULSANT
ANTIDEPRESSANT
ANTIDIABETIC
YEAR INTRO.
ARMC VOL., (PAGE)
1993 2008 2008
29 (344) 44 (617) 44 (628)
2004 2008 1990 1990 1985 2007 1989 1989 1989 1989 2008 2004 2002
40 44 26 26 21 43 25 25 25 25 44 40 38
1986 1983
22 (320) 19 (319)
1983 1986
19 (320) 22 (323)
1984 1997 1994 1990 1994 1991 1997 1997 1989 1990
20 33 30 26 30 27 33 33 25 26
(320) (338) (303) (305) (305) (331) (341) (342) (318) (309)
1983 1984 1994 1990 1992 2005 1995 1996 1998 2004 1999 1999 2005 1998
19 20 30 26 28 41 31 32 34 40 35 35 41 34
(324) (324) (312) (297) (330) (452) (344) (310) (325) (460) (344) (346) (460) (329)
(470) (610) (304) (307) (331) (488) (319) (320) (310) (311) (600) (452) (354)
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
rosiglitazone maleate sitagliptin tolrestat troglitazone vildagliptin voglibose acetorphan anti-digoxin polyclonal antibody crotelidae polyvalent immune fab fomepizole aprepitant dolasetron mesylate fosaprepitant dimeglumine granisetron hydrochloride indisetron ondansetron hydrochloride nazasetron palonosetron ramosetron tropisetron felbamate fosphenytoin sodium gabapentin levetiracetam pregabalin tiagabine topiramate centchroman anidulafungin caspofungin acetate eberconazole fenticonazole nitrate fluconazole fosfluconazole itraconazole lanoconazole luliconazole micafungin naftifine HCl oxiconazole nitrate
INDICATION
ANTIDIARRHEAL ANTIDOTE
ANTIEMETIC
ANTIEPILEPTIC
ANTIESTROGEN ANTIFUNGAL
697
YEAR INTRO.
ARMC VOL., (PAGE)
1999
35 (347)
2006 1989 1997 2007 1994 1993 2002
42 25 33 43 30 29 38
2001
37 (263)
1998 2003 1998 2008
34 39 34 44
1991
27 (329)
2004 1990
40 (459) 26 (306)
1994 2003 1996 1992 1993 1996
30 39 32 28 29 32
(305) (281) (315) (337) (337) (308)
1993 2000 2004 1996 1995 1991 2006 2001 2005 1987 1988 2004 1988 1994 2005 2002 1984 1983
29 36 40 32 31 27 42 37 41 23 24 40 24 30 41 38 20 19
(338) (307) (464) (320) (351) (324) (512) (263) (449) (334) (303) (457) (305) (302) (454) (360) (321) (322)
(541) (319) (344) (494) (313) (332) (350)
(323) (268) (321) (606)
698
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
posaconazole terbinafine hydrochloride terconazole tioconazole voriconazole amorolfine hydrochloride butenafine hydrochloride butoconazole cloconazole HCl liranaftate flutrimazole neticonazole HCl sertaconazole nitrate sulconizole nitrate apraclonidine HCl befunolol HCl bimatroprost brimonidine brinzolamide dapiprazole HCl dorzolamide HCl latanoprost levobunolol HCl tafluprost travoprost unoprostone isopropyl ester acrivastine astemizole azelastine HCl cetirizine HCl desloratadine ebastine levocabastine hydrochloride levocetirizine loratadine mizolastine setastine HCl alacepril alfuzosin HCl aliskiren amlodipine besylate amosulalol aranidipine arotinolol HCl
INDICATION
ANTIFUNGAL, TOPICAL
ANTIGLAUCOMA
ANTIHISTAMINE
ANTIHYPERTENSIVE
YEAR INTRO.
ARMC VOL., (PAGE)
2006 1991
42 (532) 27 (334)
1983 1983 2002 1991
19 19 38 27
1992
28 (327)
1986 1986 2000 1995 1993 1992 1985 1988 1983 2001 1996 1998 1987 1995 1996 1985 2008 2001 1994
22 22 36 31 29 28 21 24 19 37 32 34 23 31 32 21 44 37 30
(318) (318) (309) (343) (341) (336) (332) (297) (315) (261) (306) (318) (332) (341) (311) (328) (625) (272) (312)
1988 1983 1986 1987 2001 1990 1991
24 19 22 23 37 26 27
(295) (314) (316) (331) (264) (302) (330)
2001 1988 1998 1987 1988 1988 2007 1990 1988 1996 1986
37 24 34 23 24 24 43 26 24 32 22
(268) (306) (325) (342) (296) (296) (461) (298) (297) (306) (316)
(324) (324) (370) (322)
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
azelnidipine barnidipine hydrochloride benazepril hydrochloride benidipine hydrochloride betaxolol HCl bevantolol HCl bisoprolol fumarate bopindolol bosentan budralazine bunazosin HCl candesartan cilexetil carvedilol celiprolol HCl cicletanine cilazapril cinildipine clevidipine delapril dilevalol doxazosin mesylate efonidipine enalapril maleate enalaprilat eplerenone eprosartan felodipine fenoldopam mesylate fosinopril sodium guanadrel sulfate imidapril HCl irbesartan isradipine ketanserin lacidipine lercanidipine lisinopril losartan manidipine hydrochloride mebefradil hydrochloride moexipril HCl moxonidine nebivolol
INDICATION
699
YEAR INTRO.
ARMC VOL., (PAGE)
2003 1992
39 (270) 28 (326)
1990
26 (299)
1991
27 (322)
1983 1987 1986 1985 2001 1983 1985 1997 1991 1983 1988 1990 1995 2008 1989 1989 1988 1994 1984 1987 2003 1997 1988 1998
19 23 22 21 37 19 21 33 27 19 24 26 31 44 25 25 24 30 20 23 39 33 24 34
(315) (328) (317) (324) (262) (315) (324) (330) (323) (317) (299) (301) (339) (596) (311) (311) (300) (299) (317) (332) (276) (333) (302) (322)
1991 1983 1993 1997 1989 1985 1991 1997 1987 1994 1990
27 19 29 33 25 21 27 33 23 30 26
(328) (319) (339) (336) (315) (328) (330) (337) (337) (302) (304)
1997
33 (338)
1995 1991 1997
31 (346) 27 (330) 33 (339)
700
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
nilvadipine nipradilol nisoldipine olmesartan medoxomil perindopril pinacidil quinapril ramipril rilmenidine spirapril HCl telmisartan temocapril terazosin HCl tertatolol HCl tiamenidine HCl tilisolol hydrochloride trandolapril treprostinil sodium trimazosin HCl valsartan zofenopril calcium captopril daptomycin garenoxacin retapamulin maraviroc raltegravir aceclofenac AF-2259 amfenac sodium ampiroxicam amtolmetin guacil butibufen deflazacort dexibuprofen droxicam etodolac flunoxaprofen fluticasone propionate interferon, gamma isofezolac isoxicam lobenzarit sodium loxoprofen sodium lumiracoxib nabumetone
INDICATION
ANTIHYPERTENSIVE AGENT ANTI INFECTIVE
ANTI-INFECTIVE – HIV ANTIINFLAMMATORY
YEAR INTRO.
ARMC VOL., (PAGE)
1989 1988 1990 2002
25 24 26 38
(316) (307) (306) (363)
1988 1987 1989 1989 1988 1995 1999 1994 1984 1987 1988 1992
24 23 25 25 24 31 35 30 20 23 24 28
(309) (340) (317) (317) (310) (349) (349) (311) (323) (344) (311) (337)
1993 2002 1985 1996 2000 1982 2003 2007 2007 2007 2007 1992 1987 1986 1994 1993 1992 1986 1994 1990 1985 1987 1990
29 38 21 32 36 13 39 43 43 43 43 28 23 22 30 29 28 22 30 26 21 23 26
(348) (368) (333) (320) (313) (086) (272) (471) (486) (478) (484) (325) (325) (315) (296) (332) (327) (319) (298) (302) (327) (335) (303)
1989 1984 1983 1986 1986 2005 1985
25 20 19 22 22 41 21
(314) (319) (320) (322) (322) (455) (330)
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
nepafenac nimesulide oxaprozin piroxicam cinnamate rimexolone sivelestat tenoxicam zaltoprofen fisalamine osalazine sodium alclometasone dipropionate aminoprofen betamethasone butyrate propionate butyl flufenamate deprodone propionate felbinac halobetasol propionate halometasone hydrocortisone aceponate hydrocortisone butyrate propionate mometasone furoate piketoprofen pimaprofen prednicarbate pravastatin arteether artemisinin bulaquine halofantrine mefloquine HCl almotriptan alpiropride eletriptan frovatriptan lomerizine HCl naratriptan hydrochloride rizatriptan benzoate sumatriptan succinate
INDICATION
ANTIINFLAMMATORY, INTESTINAL ANTIINFLAMMATORY, TOPICAL
ANTILIPIDEMIC ANTIMALARIAL
ANTIMIGRAINE
701
YEAR INTRO.
ARMC VOL., (PAGE)
2005 1985 1983 1988
41 21 19 24
(456) (330) (322) (309)
1995 2002 1987 1993 1984 1986 1985
31 38 23 29 20 22 21
(348) (366) (344) (349) (318) (324) (323)
1990 1994
26 (298) 30 (297)
1983 1992
19 (316) 28 (329)
1986 1991
22 (320) 27 (329)
1983 1988
19 (320) 24 (304)
1983
19 (320)
1987 1984 1984 1986 1989 2000 1987 2000 1988 1985 2000 1988 2001 2002 1999 1997
23 20 20 22 25 36 23 36 24 21 36 24 37 38 35 33
1998 1991
34 (330) 27 (333)
(338) (322) (322) (325) (316) (296) (327) (299) (304) (329) (295) (296) (266) (357) (342) (339)
702
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
zolmitriptan dronabinol amrubicin HCl amsacrine anastrozole bicalutamide bisantrene hydrochloride camostat mesylate capecitabine cladribine cytarabine ocfosfate docetaxel doxifluridine enocitabine epirubicin HCl fadrozole HCl fludarabine phosphate flutamide formestane fotemustine geftimib gemcitabine HCl idarubicin hydrochloride imatinib mesylate interferon gamma-1a interleukin-2 irinotecan lonidamine mitoxantrone HCl nedaplatin nilutamide paclitaxal pegaspargase pentostatin pirarubicin ranimustine sobuzoxane temoporphin toremifene vinorelbine zinostatin stimalamer porfimer sodium masoprocol
INDICATION
ANTINAUSEANT ANTINEOPLASTIC
ANTINEOPLASTIC ADJUVANT ANTINEOPLASTIC, TOPICAL
YEAR INTRO.
ARMC VOL., (PAGE)
1997 1986 2002 1987 1995 1995 1990
33 22 38 23 31 31 26
(345) (319) (349) (327) (338) (338) (300)
1985 1998 1993 1993 1995 1987 1983 1984 1995 1991
21 34 29 29 31 23 19 20 31 27
(325) (319) (335) (335) (341) (332) (318) (318) (342) (327)
1983 1993 1989 2002 1995 1990
19 29 25 38 31 26
(318) (337) (313) (358) (344) (303)
2001 1992
37 (267) 28 (332)
1989 1994 1987 1984 1995 1987 1993 1994 1992 1988 1987 1994 2002 1989 1989 1994
25 30 23 20 31 23 29 30 28 24 23 30 38 25 25 30
1993
29 (343)
1992
28 (333)
(314) (301) (337) (321) (347) (338) (342) (306) (334) (309) (341) (310) (367) (319) (320) (313)
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
miltefosine dexfenfluramine rimonabant orlistat sibutramine atovaquone ivermectin budipine CHF-1301 droxidopa entacapone pergolide mesylate pramipexole hydrochloride ropinirole HCl talipexole tolcapone lidamidine HCl gestrinone cabergoline tamsulosin HCl acitretin calcipotriol tazarotene tacalcitol amisulpride blonanserin paliperidone remoxipride hydrochloride zuclopenthixol acetate biolimus drug-eluting stent actarit diacerein octreotide adamantanium bromide drotecogin alfa cimetropium bromide tiquizium bromide tiropramide HCl romiplostim argatroban bivalirudin defibrotide cilostazol
INDICATION
703
YEAR INTRO.
ARMC VOL., (PAGE)
1993 1997 2006 1998 1998 1992 1987 1997 1999 1989 1998 1988 1997
29 33 42 34 34 28 23 33 35 25 34 24 33
(340) (332) (537) (327) (331) (326) (336) (330) (336) (312) (322) (308) (341)
1996 1996 1997 1984 1986 1993 1993
32 32 33 20 22 29 29
(317) (318) (343) (320) (321) (334) (347)
1989 1991 1997 1993 1986 2008 2007 1990
25 27 33 29 22 44 43 26
(309) (323) (343) (346) (316) (587) (482) (308)
1987
23 (345)
ANTI-RESTENOTIC
2008
44 (586)
ANTIRHEUMATIC
1994 1985 1988 1984 2001 1985
30 21 24 20 37 21
(296) (326) (307) (315) (265) (326)
1984 1983 2008 1990 2000 1986 1988
20 19 44 26 36 22 24
(324) (324) (619) (299) (298) (319) (299)
ANTIOBESITY
ANTIPARASITIC ANTIPARKINSONIAN
ANTIPERISTALTIC ANTIPROGESTOGEN ANTIPROLACTIN ANTIPROSTATIC HYPERTROPHY ANTIPSORIATIC
ANTIPSORIATIC, TOPICAL ANTIPSYCHOTIC
ANTISECRETORY ANTISEPTIC ANTISEPSIS ANTISPASMODIC
ANTITHROMBOCYTOPENIC ANTITHROMBOTIC
704
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
clopidogrel hydrogensulfate cloricromen enoxaparin eptifibatide ethyl icosapentate fondaparinux sodium indobufen limaprost ozagrel sodium picotamide tirofiban hydrochloride flutropium bromide levodropropizine nitisinone benexate HCl dosmalfate ebrotidine ecabet sodium egualen sodium enprostil famotidine irsogladine lansoprazole misoprostol nizatidine omeprazole ornoprostil pantoprazole sodium plaunotol polaprezinc ranitidine bismuth citrate rebamipide rosaprostol roxatidine acetate HCl roxithromycin sofalcone spizofurone teprenone tretinoin tocoferil troxipide abacavir sulfate adefovir dipivoxil amprenavir atazanavir
INDICATION
ANTITUSSIVE ANTITYROSINAEMIA ANTIULCER
ANTIVIRAL
YEAR INTRO.
ARMC VOL., (PAGE)
1998
34 (320)
1991 1987 1999 1990 2002
27 23 35 26 38
(325) (333) (340) (303) (356)
1984 1988 1988 1987 1998 1988 1988 2002 1987 2000 1997 1993 2000 1985 1985 1989 1992 1985 1987 1988 1987 1994
20 24 24 23 34 24 24 38 23 36 33 29 36 21 21 25 28 21 23 24 23 30
(319) (306) (308) (340) (332) (303) (305) (361) (328) (302) (333) (336) (303) (327) (327) (315) (332) (329) (339) (308) (339) (306)
1987 1994 1995
23 (340) 30 (307) 31 (348)
1990 1985 1986
26 (308) 21 (332) 22 (326)
1987 1984 1987 1984 1993 1986 1999 2002 1999 2003
23 20 23 20 29 22 35 38 35 39
(342) (323) (343) (323) (348) (327) (333) (348) (334) (269)
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
cidofovir delavirdine mesylate didanosine efavirenz emtricitabine enfuvirtide entecavir etravirine famciclovir fomivirsen sodium fosamprenavir foscarnet sodium ganciclovir imiquimod indinavir sulfate interferon alfacon-1 lamivudine lopinavir nelfinavir mesylate nevirapine oseltamivir phosphate penciclovir propagermanium rimantadine HCl ritonavir saquinavir mesylate sorivudine stavudine tenofovir disoproxil fumarate valaciclovir HCl zalcitabine zanamivir zidovudine influenza virus live cevimeline hydrochloride alpidem buspirone HCl etizolam flutazolam flutoprazepam metaclazepam mexazolam tandospirone ciclesonide atomoxetine
INDICATION
ANTIVIRAL VACCINE ANTI-XEROSTOMIA ANXIOLYTIC
ASTHMA, COPD ATTENTION DEFICIT HYPERACTIVITY DISORDER
705
YEAR INTRO.
ARMC VOL., (PAGE)
1996 1997
32 (306) 33 (331)
1991 1998 2003 2003 2005 2008 1994 1998 2003 1989 1988 1997 1996 1997 1995 2000 1997 1996 1999 1996 1994 1987 1996 1995 1993 1994 2001
27 34 39 39 41 44 30 34 39 25 24 33 32 33 31 36 33 32 35 32 30 23 32 31 29 30 37
(326) (321) (274) (275) (450) (602) (300) (323) (277) (313) (303) (335) (310) (336) (345) (310) (340) (313) (346) (314) (308) (342) (317) (349) (345) (311) (271)
1995 1992 1999 1987 2003 2000
31 28 35 23 39 36
(352) (338) (352) (345) (277) (299)
1991 1985 1984 1984 1986 1987 1984 1996 2005 2003
27 21 20 20 22 23 20 32 41 39
(322) (324) (318) (318) (320) (338) (321) (319) (443) (270)
706
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
INDICATION
YEAR INTRO.
ARMC VOL., (PAGE)
flumazenil bambuterol doxofylline formoterol fumarate mabuterol HCl oxitropium bromide salmeterol hydroxynaphthoate tiotropium bromide APD clodronate disodium disodium pamidronate gallium nitrate ipriflavone neridronic acid dexrazoxane bucladesine sodium denopamine docarpamine dopexamine enoximone flosequinan ibopamine HCl loprinone hydrochloride milrinone vesnarinone amrinone colforsin daropate HCL xamoterol fumarate cefozopran HCL
BENZODIAZEPINE ANTAG. BRONCHODILATOR
1987 1990 1985 1986 1986 1983 1990
23 26 21 22 22 19 26
2002 1987 1986
38 (368) 23 (326) 22 (319)
1989
25 (312)
1991 1989 2002 1992 1984 1988 1994 1989 1988 1992 1984 1996
27 25 38 28 20 24 30 25 24 28 20 32
(328) (314) (361) (330) (316) (300) (298) (312) (301) (331) (319) (312)
1989 1990 1983 1999
25 26 19 35
(316) (310) (314) (337)
1988 1995
24 (312) 31 (339)
1994 1986 1985 1988 1991 1986 1983 2006
30 22 21 24 27 22 19 42
2005 2007
41 (446) 43 (465)
1983 2000
19 (314) 36 (311)
CALCIUM REGULATOR
CARDIOPROTECTIVE CARDIOSTIMULANT
CARDIOTONIC
CEPHALOSPORIN, INJECTABLE cefditoren pivoxil CEPHALOSPORIN, ORAL brovincamine fumarate CEREBRAL VASODILATOR nimodipine propentofylline succimer CHELATOR trientine HCl fenbuprol CHOLERETIC lulbiprostone CHRONIC IDIOPATHIC CONSTIPATION deferasirox CHRONIC IRON OVERLOAD arformoterol CHRONIC OBSTRUCTIVE PULMONARY DISEASE auranofin CHRYSOTHERAPEUTIC taltirelin CNS STIMULANT
(335) (299) (327) (321) (323) (323) (308)
(297) (317) (330) (310) (333) (327) (318) (525)
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
707
GENERIC NAME
INDICATION
YEAR INTRO.
ARMC VOL., (PAGE)
aniracetam pramiracetam H2SO4 carperitide nesiritide drospirenone norelgestromin nicorandil certolizumab pegol dornase alfa neltenexine amifostine nalmefene HCL
COGNITION ENHANCER
1993 1993
29 (333) 29 (343)
CONGESTIVE HEART FAILURE CONTRACEPTIVE
1995 2001 2000 2002 1984 2008 1994 1993 1995 1995
31 37 36 38 20 44 30 29 31 31
(339) (269) (302) (362) (322) (592) (298) (341) (338) (347)
2000 1986 1983 1993 1997 1997 2008 1999 2006 1991 2005 1995 2001 2001 1999 2000 1996 1996 1996 2000
36 22 19 29 33 33 44 35 42 27 41 31 37 37 35 36 32 32 32 36
(306) (316) (321) (348) (328) (331) (594) (343) (540) (321) (472) (342) (267) (259) (336) (305) (307) (308) (316) (303)
2000 1998 1990 1988 1995 1998 1994 2003 2008
36 34 26 24 31 34 30 39 44
(307) (328) (301) (299) (344) (326) (301) (279) (615)
1994 1994
30 (310) 30 (310)
ioflupane azosemide muzolimine torasemide atorvastatin calcium cerivastatin choline fenofibrate naftopidil silodosin alglucerase udenafil erdosteine fudosteine agalsidase alfa cetrorelix ganirelix acetate follitropin alfa follitropin beta reteplase esomeprazole magnesium lafutidine rabeprazole sodium cinitapride cisapride itopride HCL mosapride citrate imiglucerase miglustat rilonacept
somatotropin somatomedin-1
CORONARY VASODILATOR CROHN’S DISEASE CYSTIC FIBROSIS CYTOPROTECTIVE DEPENDENCE TREATMENT DIAGNOSIS CNS DIURETIC
DYSLIPIDEMIA
DYSURIA ENZYME ERECTILE DYSFUNCTION EXPECTORANT FABRY’S DISEASE FEMALE INFERTILITY FERTILITY ENHANCER FIBRINOLYTIC GASTRIC ANTISECRETORY
GASTROPROKINETIC
GAUCHER’S DISEASE GENETIC AUTOINFLAMMATORY SYNDROMES GROWTH HORMONE GROWTH HORMONE INSENSITIVITY
708
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
INDICATION
YEAR INTRO.
ARMC VOL., (PAGE)
factor VIIa levosimendan pimobendan anagrelide hydrochloride erythropoietin eculizumab thrombin alfa factor VIII telbivudine clevudine malotilate mivotilate Icatibant darunavir tipranavir buserelin acetate goserelin leuprolide acetate nafarelin acetate somatropin zoledronate disodium cinacalcet sapropterin hydrochloride quinagolide cadralazine nitrendipine binfonazole brotizolam butoctamide cinolazepam doxefazepam eszopiclone loprazolam mesylate quazepam rilmazafone zaleplon zolpidem hemitartrate zopiclone acetohydroxamic acid sodium cellulose PO4 divistyramine
HAEMOPHILIA HEART FAILURE
1996 2000 1994 1997
32 36 30 33
(307) (308) (307) (328)
1988 2007 2008 1992 2006 2007 1985 1999 2008 2006 2005 1984 1987 1984
24 43 44 28 42 43 21 35 44 42 41 20 23 20
(301) (468) (627) (330) (546) (466) (329) (343) (608) (515) (470) (316) (336) (319)
HYPERCALCEMIA
1990 1987 2000
26 (306) 23 (343) 36 (314)
HYPERPARATHYROIDISM HYPERPHENYL-ALANINEMIA
2004 1992
40 (451) 28 (336)
HYPERPROLACTINEMIA HYPERTENSIVE
1994 1988 1985 1983 1983 1984 1993 1985 2005 1983
30 24 21 19 19 20 29 21 41 19
(309) (298) (331) (315) (315) (316) (334) (326) (451) (321)
1985 1989 1999 1988
21 25 35 24
(332) (317) (351) (313)
HYPOAMMONURIC
1986 1983
22 (327) 19 (313)
HYPOCALCIURIC
1983
19 (323)
HYPOCHOLESTEROLEMIC
1984
20 (317)
HEMATOLOGIC HEMATOPOETIC HEMOGLOBINURIA HEMOSTAT HEMOSTATIC HEPATITIS B HEPATOPROTECTIVE HEREDITARY ANGIODEMA HIV HORMONE
HYPNOTIC
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
lovastatin melinamide pitavastatin rosuvastatin simvastatin glucagon, rDNA acipimox beclobrate binifibrate ciprofibrate colesevelam hydrochloride colestimide ezetimibe fluvastatin meglutol ronafibrate conivaptan mozavaptan modafinil pirfenidone bucillamine centoxin thymopentin filgrastim GMDP interferon gamma-1b lentinan pegademase bovine pidotimod romurtide sargramostim schizophyllan ubenimex cyclosporine everolimus gusperimus mizoribine muromonab-CD3 mycophenolate sodium mycophenolate mofetil pimecrolimus tacrolimus ramelteon
INDICATION
HYPOGLYCEMIA HYPOLIPIDEMIC
HYPONATREMIA IDIOPATHIC HYPERSOMNIA IDIOPATHIC PULMONARY FIBROSIS IMMUNOMODULATOR
IMMUNOSTIMULANT
IMMUNOSUPPRESSANT
INSOMNIA
709
YEAR INTRO.
ARMC VOL., (PAGE)
1987 1984 2003 2003 1988 1993 1985 1986 1986 1985 2000
23 20 39 39 24 29 21 22 22 21 36
(337) (320) (282) (283) (311) (338) (323) (317) (317) (326) (300)
1999 2002 1994 1983 1986 2006 2006 1994
35 38 30 19 22 42 42 30
(337) (355) (300) (321) (326) (514) (527) (303)
2008
44 (614)
1987 1991 1985 1991 1996 1991
23 27 21 27 32 27
(329) (325) (333) (327) (308) (329)
1986 1990 1993 1991 1991 1985 1987 1983 2004 1994 1984 1986 2003
22 26 29 27 27 22 23 19 40 30 20 22 39
(322) (307) (343) (332) (332) (326) (345) (317) (455) (300) (321) (323) (279)
1995
31 (346)
2002 1993 2005
38 (365) 29 (347) 41 (462)
710
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
INDICATION
YEAR INTRO.
ARMC VOL., (PAGE)
defeiprone alosetron hydrochloride tegasedor maleate sulbactam sodium tazobactam sodium nartograstim pumactant sildenafil citrate
IRON CHELATOR IRRITABLE BOWEL
1995 2000
31 (340) 36 (295)
SYNDROME b-LACTAMASE INHIBITOR
2001 1986 1992 1994 1994 1998
37 22 28 30 30 34
(270) (326) (336) (304) (308) (331)
2000 1992 2003 2005 2006
36 28 39 41 42
(304) (337) (278) (453) (520)
2005 1996 1993 1997 2004 1983 1995
41 32 29 33 40 19 31
(461) (311) (339) (334) (462) (313) (340)
1991
27 (326)
1983 1992
19 (318) 28 (334)
1999 1984 2006
35 (347) 20 (324) 42 (519)
2006
42 (523)
1984 1988 2002 1991
20 24 38 27
1991 1996 2001
27 (331) 32 (313) 37 (270)
1997 1993 1996 1984 2000
33 29 32 20 36
gadoversetamide telmesteine laronidase galsulfase idursulfase
LEUKOPENIA LUNG SURFACTANT MALE SEXUAL DYSFUNCTION MRI CONTRAST AGENT MUCOLYTIC MUCOPOLYSACCARIDOSIS MUCOPOLYSACCHARIDOSIS VI MUCOPOLYSACCHARIDOSIS II (HUNTER SYNDROME) MUCOSITIS MULTIPLE SCLEROSIS
palifermin interferon X-1a interferon X-1b glatiramer acetate natalizumab afloqualone MUSCLE RELAXANT cisatracurium besilate doxacurium chloride eperisone HCl mivacurium chloride rapacuronium bromide tizanidine decitabine MYELODYSPLASTIC SYNDROMES lenalidomide MYELODYSPLASTIC SYNDROMES, MULTIPLE MYELOMA naltrexone HCl NARCOTIC ANTAGONIST tinazoline NASAL DECONGESTANT aripiprazole NEUROLEPTIC clospipramine hydrochloride nemonapride olanzapine perospirone hydrochloride quetiapine fumarate risperidone sertindole timiperone ziprasidone hydrochloride
(322) (312) (350) (325)
(341) (344) (318) (323) (312)
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
INDICATION
YEAR INTRO.
ARMC VOL., (PAGE)
rocuronium bromide
NEUROMUSCULAR
1994
30 (309)
1995 1995 1996 2006 1987 1990 1988 1986 1988 1986 1988
37 31 32 42 23 26 24 22 24 22 24
1987 1997 1997
23 (339) 33 (329) 33 (337)
2008
44 (612)
2001 1993
37 (269) 29 (332)
1996 1997 1998
32 (309) 33 (335) 34 (328)
1998 2004 2008 2007 1995
34 40 44 43 31
(330) (467) (604) (472) (350)
2005 2006 2003 2003 2002 2000 2003 1992 1983
41 42 39 39 38 36 39 28 19
(464) (538) (284) (286) (367) (312) (267) (326) (318)
1992 1993 1994
28 (332) 29 (344) 30 (312)
2004 2006 2008
40 (466) 42 (511) 44 (584)
edaravone fasudil HCL riluzole varenicline bifemelane HCl choline alfoscerate exifone idebenone indeloxazine HCl levacecarnine HCl nizofenzone fumarate oxiracetam bromfenac sodium lornoxicam methylnaltrexone bromide OP-1 alendronate sodium ibandronic acid incadronic acid raloxifene hydrochloride risedronate sodium strontium ranelate fesoterodine imidafenacin tiludronate disodium rasagiline rotigotine tadalafil vardenafil temoporphin verteporfin alefacept beraprost sodium epoprostenol sodium iloprost sarpogrelate HCl trimetrexate glucuronate solifenacin alglucosidase alfa alvimopan
BLOCKER NEUROPROTECTIVE
NICOTINE-DEPENDENCE NOOTROPIC
NSAID
OPIOID-INDUCED CONSTIPATION OSTEOINDUCTOR OSTEOPOROSIS
OVERACTIVE BLADDER PAGET’S DISEASE PARKINSON’S DISEASE PDE5 INHIBITOR PHOTOSENSITIZER PLAQUE PSORIASIS PLATELET AGGREG. INHIBITOR
PLATELET ANTIAGGREGANT PNEUMOCYSTIS CARINII PNEUMONIA POLLAKIURIA POMPE DISEASE POST-OPERATIVE ILEUS
(265) (343) (317) (547) (329) (300) (302) (321) (304) (322) (307)
711
712
Cumulative NCE Introduction Index, 1983–2008 (by Indication)
GENERIC NAME
INDICATION
YEAR INTRO.
ARMC VOL., (PAGE)
histrelin atosiban gestodene nomegestrol acetate norgestimate promegestrone trimegestone alpha-1 antitrypsin nafamostat mesylate adrafinil dexmethylphenidate HCl dutasteride efalizumab ambrisentan
PRECOCIOUS PUBERTY PRETERM LABOR PROGESTOGEN
1993 2000 1987 1986 1986 1983 2001 1988 1986 1986 2002
29 36 23 22 22 19 37 24 22 22 38
2002 2003 2007
38 (353) 39 (274) 43 (463)
2006 1992 1987 2008
42 28 23 44
2006 2003 2000
42 (509) 39 (267) 36 (301)
2005 1999
41 (473) 35 (341)
1995
31 (351)
ULCERATIVE COLITIS
1987 1987 1997
23 (326) 23 (326) 33 (329)
URINARY INCONTINENCE UROLITHIASIS UROLOGIC
2005 1989 1992
41 (445) 25 (318) 28 (335)
VACCINE VASOPROTECTIVE VENOUS THROMBOEMBOLISM VITAMIN D
1999 1988 2008
35 (342) 24 (300) 44 (617)
2001 2000 1998 1999 1996
37 36 34 35 32
2001 1983 1987
37 (257) 19 (316) 23 (333)
sitaxsentan finasteride surfactant TA sugammadex
abatacept Adalimumab dexmedetomidine hydrochloride ziconotide kinetin tirilazad mesylate APSAC alteplase balsalazide disodium darifenacin tiopronin propiverine hydrochloride Lyme disease clobenoside rivaroxaban
PROTEASE INHIBITOR PSYCHOSTIMULANT
PSORIASIS PULMONARY ARTERIAL HYPERTENSION PULMONARY HYPERTENSION 5a-REDUCTASE INHIBITOR RESPIRATORY SURFACTANT REVERSAL OF NEUROMUSCULAR BLOCKADE RHEUMATOID ARTHRITIS SEDATIVE SEVERE CHRONIC PAIN SKIN PHOTODAMAGE/ DERMATOLOGIC SUBARACHNOID HEMORRHAGE THROMBOLYTIC
falecalcitriol maxacalcitol paricalcitol doxercalciferol VITAMIN D PROHORMONE prezatide copper VULNERARY acetate acemannan WOUND HEALING AGENT cadexomer iodine epidermal growth factor
(338) (297) (335) (324) (324) (323) (273) (297) (323) (315) (352)
(543) (331) (344) (623)
(266) (310) (327) (339) (314)
