suffering certain types of cancer, spontaneous activation of the immune system can lead to complete recovery. The concept of immunotherapy is based on that notion. The intention is to stimulate T cells directed against the cancer, leading to the eradication of cancer cells. However, at the moment the methods to stimulate T cell immunity are functional in small laboratory animals only. This thesis introduces a novel T cell vaccination method that uses a tattoo machine to inject DNA in the skin of the vaccinee. In comparison to other experimental vaccination methods DNA tattooing is very strong: besides small laboratory animals also large animals mount strong T cell responses upon tattoo DNA vaccination. Future tests in melanoma patients will point out whether DNA tattoo vaccination is equally efficient in humans and whether it may have a therapeutic effect in cancer patients.
Adriaan D. Bins (Nijmegen, 1973) studied Medicine and Biology at Leiden University Medical Center. In 2001 he started the research leading to this thesis at the Netherlands Cancer Institute / Antoni van Leeuwenhoek Hospital in Amsterdam. The work done in this period was rewarded in 2006 with the NKI/Avl Award. In 2006 Leiden University has initiated a series Leiden Dissertations at Leiden University Press. This series affords an opportunity to those who have recently obtained their doctorate to publish the results of their doctoral research so as to ensure a wide distribution among colleagues and the interested public. The dissertations will become available both in printed and in digital versions. Books from this LUP series can be ordered through www.lup.nl. The large majority of Leiden dissertations from 2005 onwards is available digitally on www.dissertation.leidenuniv.nl.
l u p d i s s e rtat i o n s
9 789087 280116
Bins_DEF.indd 1
Adriaan D. Bins · Induction and analysis of antigen-specific T cell responses in melanoma patients and animal models
In the last century, it became clear that in patients
LUP
Adriaan D. Bins
Induction and analysis of antigen-specific T cell responses in melanoma patients and animal models
leiden universit y press
29-1-2007 20:37:53
Induction and analysis of antigen-specific T cell responses in melanoma patients and animal models
1
The publication of this book is made possible by grants from the Netherlands Cancer Institute / Antoni van Leeuwenhoek Hospital and the Dutch Cancer Foundation Cover illustration: Elaine Bell, Nature Reviews Immunology. 20 July 2005 Cover design: Randy Lemaire, Utrecht isbn 978 90 8728 011 4 nur 870 © Leiden University Press, 2007 2nd edition All rights reserved. Without limiting the rights under copyright reserved above, no part of this book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the written permission of both the copyright owner and the author of the book. 2
Induction and analysis of antigen-specific T cell responses in melanoma patients and animal models
PROEFSCHRIFT
ter verkrijging van de graad van Doctor aan de Universiteit Leiden, op gezag van de Rector Magnificus prof.mr. P.F. van der Heijden, volgens besluit van het College voor Promoties te verdedigen op donderdag15 maart 2007 klokke 16:15 uur door
Adriaan Dirk Bins geboren te Nijmegen in 1973 3
Promotiecommissie: Promotor: Prof. Dr. T.N.M. Schumacher
Co-Promotor: Dr. J.B.A.G. Haanen
the Netherlands Cancer Institute / Antoni van Leeuwenhoek Hospital, Amsterdam
Referent: Prof. Dr. T.H.M. Ottenhoff
Overige Leden: Prof. mr. P.F. van der Heijden Prof. Dr. J.J. Neefjes Dr. M.H.M. Heemskerk Prof. Dr. G.J. Adema
4
Radboud University Nijmegen, Nijmegen
To my laboratory animals.......
5
6
Contents Introduction
Chapter 1
An introduction to the immune system and to the principles of tumor vaccination. An introduction to the work described in this thesis.
9
On the Role of Melanoma-Specific CD8+ T-Cell Immunity in Disease Progression of Advanced-Stage Melanoma Patients.
24
17
Clin Cancer Res 2004 July 15;10(14):4754-60
Chapter 2
Phase I clinical study with multiple peptide vaccines in combination with tetanus toxoid and GM-CSF in advanced-stage HLA-A*0201-positive melanoma patients.
43
J.Immunotherapy, in print
Chapter 3
A rapid and potent DNA vaccination strategy defined by in vivo monitoring of antigen expression
57
Nat Med 2005 August;11(8):899-904.
Chapter 4
Intradermal DNA tattooing primes robust T cell responses in maccaca mulatta.
73
Manuscript in preparation
Chapter 5
In vivo antigen stability affects vaccine efficacy
82
Submitted
Chapter 6
High-throughput intravital imaging of fluorescent markers and FRET probes by DNA tattooing
104
BMC Biotechnol 2007 January 3;7(1):2.
Chapter 7
Design and use of conditional MHC class I ligands
118
Nat Med 2006 February;12(2):246-51.
Discussion
138
Summary in Dutch
147
Curriculum Vitae & Acknowledgements
148 7
8
i
I N T R O D U C T I O N
An Introduction to the immune system and to the principles of cancer vaccination. The immune system T Cells and B cells The system that we refer to as “the immune system” is in fact a collection of many smaller systems, each with specialized features that serve to protect us against different types of virusses and germs (pathogens). The cells that orchestrate the production of antibodies (B cells) are the bestknown “sub-system”. Antibodies are proteins that stick to a pathogen, thereby neutralizing it. Each B cell produces a type of antibody that has unique sticking properties, making it stick to specific pathogens. In order to do so, the gene in each B cell that encodes the antibody, is randomly compiled from a set of genomic building blocks at the birth of the B cell in the bone marrow. This compilation process is called genetic recombination. Consequently, each of the millions of B cells that the bone marrow generates through life, produces antibodies with a random and unique target specificity. Time will tell whether they can stick to an invading pathogen, or whether they are useless. The only thing that the bone marrow makes sure about their specificity, is that they do not stick to “self” proteins (proteins that belong to oneself). B cells that do so are killed before they are released in the system (so-called “negative B cell selection”). What is left recognizes “non-self” proteins only. Such “foreign” non-self proteins are called “antigens”. By the production of antibodies, B cells can kill from a long distance. When released, antibodies easily disperse in all tissues, sticking in large amounts to the surface of the pathogen that matches their specificity. Once bound by antibodies, other “sub-systems” will kill the pathogen. However, antibodies can only stick to the outside of invaders and infected cells, since they cannot pass through cellular walls (“cell membranes”). Therefore, pathogens that disguise their surface molecules to look like our own or pathogens that hide inside cells of the patient cannot be fought with antibodies. For this purpose, we have T cells. These cells are part of a sub-system that can recognize non-self proteins expressed inside our own cells. The work described in this thesis exclusively concerns T cells. T cells have a T cell receptor (TCR) that is recombined just like antibodies and T cells are subsequently negatively selected just like B cells are. Contrary to B cells, the TCR of a T cell is not secreted in the circulation but remains attached to the T cell. Furthermore, it is not designed to stick to the enemy but to so-called MHC-molecules: molecules that any cell in 9
AN INTRODUCTION
TO THE IMMUNE
SYSTEM AND TO THE PRINCIPLES OF TUMOR VACCINATION
our body constantly replaces on its surface, with always a fresh part of the cells entrails right on top of it (a fragment of a protein: a peptide). Upon binding of the TCR to the MHC molecule, this peptide is sandwiched in between them. Therefore it determines the strength of the TCR - MHC interaction. As mentioned previously, T cells with a TCR that can bind to a MHC molecule with a self-peptide on top of it have been parsed by negative-T-cell selection during their development in the thymus. Hence, any cell presenting an MHC/peptide complex that can be bound by a TCR is non-self by exclusion; therefore it is killed instantly by the T cell carrying this TCR. Due to the negative selection processes, neither the T nor the B cell system will react to self-proteins. This non-responsiveness originating from negative T and B cell selection is referred to as “central self-tolerance”. There are other forms of immunological tolerance that do not come about by negative selection in the thymus (T cells) or the bone marrow (B cells). These are referred to as “peripheral tolerance”. In order for inexperienced B and T cells (freshly produced, “naive” T or B cells) to start killing cells that express (produce) a non self (“foreign”) protein, the mere presence of this protein will not suffice. In short two things are needed to activate them: 1) the foreign protein, and 2) alarm signals (“danger signals”). Danger signals are provided by another very important subsystem: the Antigen presenting cells (APCs). These cells are capable of sensing the presence of invading pathogens and ‘signal’ their presence to the rest of the immune system. If both the foreign protein and danger signals are present, T and B- cells will start to divide and to prepare for killing (T cells) or antibody production (B cells). The “activated” T cells can be identified by molecules that either appear (e.g. CD45RO) or disappear (e.g. CD27) from their cell membrane and by the secretion of substances (e.g. IFN-γ, IL-2 and IL-4) that support the fulfillment of their killer (“cytotoxic”) duties. Importantly, the presence of a foreign protein in the absence of danger signals usually leads to tolerization of naïve T cells specific for this protein instead of activation. This is a very important peripheral tolerance mechanism. The B cell en the T-cell subsystems are sometimes referred to as the humoral and cellular immune systems, respectively. They are the most complex and evolutionary advanced sub-systems of our immune system. Together they form the “adaptive” part of our immune system, referring to their “memory”-function: once a foreign protein has been encountered, the B and T cell systems adapt to become hyper vigilant for pathogens expressing this protein. This is partly due to a small portion of the B and T cells that persists once the pathogen has been cleared after having participated in the immune attack. Some B cells continue to produce antibodies and some T cells continue to patrol the body. Importantly, contrary to inexperienced T or B cells, the re-appearance of the pathogen in the absence of danger signals can reactivate these “experienced” T and B cells. 10
i Furthermore, a distinct sub-type of T cells ‘helps’ in the maintenance of both cellular an humoral memory. These cells (the T helper- or Th cells) are characterized by the presence of a ‘CD4’ molecule on their membrane. Altogether, the adaptations leading to immunological memory are responsible for us getting our childhood diseases only once. None of the other sub-systems of our immune-system has this capacity. The other systems always remain as they were at birth, therefore they are collectively named “innate-immunity”. APCs APCs are part of our innate immunity and reside in all parts of our body, especially those parts that border the exterior world. These borders are lined with so-called epithelia: special cells that form a barrier. Roughly, there are 4 types: lining the gut, the airways, the urogenital system or the body surface (the skin or dermis). As any infectious disease starts with a bug crossing one of these borders, APCs are constantly on guard here for intruders. The upper layer of the skin (epidermis) is equipped with a dense network of specialized APCs, called Langerhans cells (LC). In deeper layers of the skin dermal dendritic cells (DCs) are on guard. Both these cells have a spider-like appearance projecting long protrusions of their body (dendrites) deep in the surrounding tissue. Using these dendrites, they sample all molecules present in the tissue. Furthermore, APCs that do not have dendrites, called macrophages, patrol the skin as well. These cells are able to ingest larger particles, including whole cells. APCs are pivotal to T and B cell activation. If the pathogen recognition receptors of an APCs sense danger (i.e. the presence of an intruder) the APC will change its “phenotype” (properties) and start to make its way to the lymph node that drains the infected area (the draining lymph node, DLN). This is named the “maturation” of an APC. In the DLN, the antigens that were taken up in the tissues are presented by the APC to naive T and B cells, leading to their activation. For presentation to T cells the antigens are hashed, stuck on top of the MHC molecules and brought to the membrane of the APC. For B cells the APC just glues some of the antigen to its surface, for B cells to snoop it of. Importantly, the presentation by APCs of antigens produced in other cells is called “cross presentation”1. Interestingly, from all the antigens present in the infected tissue, only some will be efficiently cross presented by the APC. Activating the immune system to fight against a tumor Tumor antigens Contrary to self proteins, antigens can be recognized by antibodies and T cells. However, tumors are exclusively build from self proteins, since all cancers originate from one of our own cells that has gone through 11
AN INTRODUCTION
TO THE IMMUNE
SYSTEM AND TO THE PRINCIPLES OF TUMOR VACCINATION
many uncontrolled divisions, creating a lump of meat consisting of its offspring. If the T and B cell systems are to fight a cancer, what non self proteins should they recognize then? Tumor derived proteins that T or B cells can recognize are named “tumor antigens”. Different classes of tumor antigens can be identified2: 1) Some cancers are caused by viruses. Especially the human papilloma virus (HPV) is renowned for its potential to cause cervical carcinoma in sexually active women. The HPV derived proteins driving the uncontrolled cell division in cervical carcinomas are non-self and can be recognized by T and B cells. These antigens are referred to as “virally derived tumor antigens” 2) Usually, the cell that started the cancer did so because a chemical agent or a radiation beam struck one of its genes, mutating it in a gene driving uncontrolled cell-division. In some cases, the DNA mutation results in the production of a mutated protein that (if it differs enough from the normal version) can be recognized by T and B cells as a “tumor specific antigen”. In many cases however, the mutation occurs in DNA that does not encode a protein, but solely has a regulatory function in the cell division process. 3) During their uncontrolled divisions, tumor cells make many mistakes in the copy-process of their genes, leading to more mutations and further deregulation of their behavior. This may cause tumor cells to start producing proteins that were produced only during embryonic development of the patient and were never produced since. As the B and T cell systems develop for a large part after birth, T- and B cells are not negatively selected against these proteins. Therefore, they may recognize these so-called “differentiation-antigens”. 4) Sometimes a TCR or an antibody can recognize a self-protein that the tumor produces in abnormal and unphysiological amounts (so-called “overexpressed antigens”). This happens just because there is suddenly so much of it. The efficacy of the central tolerization process varies per self-protein: for some self-proteins the central tolerance is absolute while for other self-proteins overexpression may lead to the activation of T and B cells. 5) Some tissues, the testis in particular, produce many proteins that are not protected by central tolerance but by local (peripheral) tolerance mechanisms. In the testis this is because during spermatogenesis the genes of our parents are recombined to new variants, leading to the expression of foreign proteins. When testis antigens are accidentally expressed by tumors without peripheral tolerance mechanisms, T and B cells may recognize them. Such antigens are called “cancer/testis antigens”.
12
i 6) As mentioned previously, central tolerance can be incomplete for self proteins in tissues outside the testis as well. In these cases this is compensated for by peripheral tolerance mechanisms. Potentially, when such tissues give rise to a tumor that breaches peripheral tolerance (by eliciting strong danger signals) suddenly these proteins can activate T and B cells. The Mart-1 and Gp100 tumor antigens in pigment cells of the skin (melanocytes) belong to this group. It may seem that T and B cells of cancer patients have a high chance to recognize fragments of tumor antigens (peptides) presented to them on top of MHC molecules. Unfortunately this is not the case, as most tumors do not express any tumor antigen. Moreover, regarding T cells, only peptides that can stick on top of a MHC molecule can be presented. Such peptides are called “T cell epitopes”. Since MHC molecules vary from person to person, each individual will recognize other T cell epitopes in a tumor antigen. Hence, the T cells of a patient with an ‘unlucky’ set of MHC molecules may in theory fail to recognize a good tumor antigen. T cells and cancer Occasionally, tumors regress spontaneously without any form of therapy. Many such spontaneous regressions are thought to be immune mediated3. Since these regressions are mainly restricted to certain cancer types, e.g. renal cell carcinoma, neuroblastoma, lymphomas and melanomas (cancer from melanocytes)4;5, from early days the focus of tumor immunology has been on these cancers. Experiments in the eighties, in which immune cells were isolated from cancer patients and reinjected after ex vivo (in a culture dish) stimulation, suggested that T cells are a key player in the immune mediated anti tumor effect6. Extensive analyses of antitumor immune responses in patients with spontaneous tumor regressions have reinforced this notion since7;8. Cancer immunotherapy aims at the eradication of a tumor by activation of tumor specific T cells (so-called “immunization”). To this purpose various immunization strategies are employed. Here I’ll categorize these T cell inducing immunization methods against cancer. Protective or therapeutic immunization The majority of experimental cancer immunization strategies consists of “therapeutic immunization”. This means that once the patient has the disease, tumor vaccines are employed to mount a sufficiently early, sizable and fully differentiated T cell response. Conversely, “protective immunization” strives to generate tumor specific T cell immunity before onset of the disease. Theoretically, as tumor antigens – except virally derived antigens – are self proteins, there is a risk for autoimmunity (an immune attack against friendly self tissue) upon immunization against non-virally derived tumor antigens. In case of protective vaccination this risk is unjustifiably big compared to the small chance to develop the 13
AN INTRODUCTION
TO THE IMMUNE
SYSTEM AND TO THE PRINCIPLES OF TUMOR VACCINATION
tumor. For this and other reasons this strategy is currently pursued for virally induced tumors only, e.g. for hepatocellular carcinoma (caused by the hepatitis B virus) and for cervical carcinoma (caused by HPV). aspecific or specific immunization Most tumors are notoriously unalarming to APCs, leading to peripheral tolerization of tumor specific T cells (if there are any). To breach such tolerization, ‘ancient’ cancer vaccines tried to jam the alarm bells. These vaccines are referred to as “aspecific cancer vaccines”, since they do not provide an antigen. More recent cancer vaccines provide both antigens and danger signals. These vaccines are referred to as “specific cancer vaccines”. In specific vaccines, the components meant to provide the danger signal, are referred to as ‘adjuvants’. In order to mislead APCs to make alarm, some adjuvants consist of molecules derived form notorious pathogens (so called PAMPs, pathogen associated molecular patterns). Specialized trouble sensors of APCs (Toll-like receptors, TLRs) can detect PAMPs, resulting in danger signals. Examples of PAMPs are bacterial cell wall components (LPS, sticks to TLR4) and unique bacterial DNA sequences (CpGs, stick to TLR9). Other adjuvants mimic signal molecules that have a function in the alarm-cascade. Some adjuvants consist of these molecules themselves (e.g. IL-2 and GM-CSF) and other adjuvants manipulate their function by sticking to them (e.g. CD40 agonistic (stimulating) antibodies, CTLA-4 antagonistic (blocking) antibodies and CD28 agonistic antibodies). passive or active immunization Roughly there are two methods that lead to immunity against a (tumor) antigen: activating the patients own naive T and B cells or injecting antigen experienced T cells or antibodies straight away. The first is called active immunization and the second passive immunization. “Vaccination” is a synonym for active immunization. Most passive immunization methods do not result in immune memory. However, passive immunization works very fast and leads to strong, albeit short, immunity. More importantly, passive immunization circumvents both central and peripheral tolerization of T and B cells. This makes it very suitable for therapeutic cancer vaccination. For the induction of protective long-term immune memory, active immunization methods are the strategy of choice. Nevertheless, many active cancer immunization methods are being tested in cancer patients all over the world. This may be due to the relatively low costs of active cancer vaccines. Up to this moment however, the therapeutic effect of these vaccines has been disappointingly low. On the other hand, the immunological knowledge that has resulted from the 14
i pursuit of active cancer vaccines has been useful for the development of other vaccines as well. Currently, passive cancer immunization is an accepted therapeutic strategy in for certain types of cancer. Examples are the antibodies directed against breast cancer antigens (herceptin) and certain lymphoma antigens (rituximab) that are used in the treatment of patients suffering from these cancers. Although the infusion of such antibodies never leads to the complete eradication of the tumor, it can considerably delay disease progression. Regarding T cells, the experimental infusion of tumor specific T cells has lead to impressive results in end stage melanoma patients8. Furthermore, the implantation of a gene that encodes a tumor specific TCR in T cells of a cancer patient (“TCR gene transfer”), is a novel form of passive immunization9. The first clinical tests with this type of gene-therapy are currently ongoing. Various means to include antigens and adjuvants in T cell vaccines Cancer specific T cell vaccines combine an adjuvant to provide danger signals with an antigen to direct the response towards the tumor. The first successful T cell vaccine ever, protected against smallpox7. The recipient was inoculated with viable cow pox virus isolated from diseased cows, resulting in a non-lethal infectious disease similar to smallpox, but milder. While the viral activity provided danger signals, the antigens of the cow-pox virus (sufficiently similar to those of the human smallpox virus) induced a T cell response that would protect the child against future smallpox infection. Since cow pox infection can be fatal in immunocompromized patients (patients with a weakened immune system), the method became obsolete. The smallpox virus has eventually been exterminated worldwide by a very similar, but safer, vaccination strategy10. Currently, some cancer patients are experimentally vaccinated analogous to this approach. Patients are injected with an viable (but weakened “attenuated”) pathogen (the “vector”) to provide danger signals, while the expression of tumor antigens is accomplished by genetic modification of this vector. In many such vaccines, an attenuated pox-virus (MVA, Modified Vaccinia Ankara) is employed for this purpose11. Inherently, these vaccines contain many non tumor specific vector-derived antigens. This disadvantage is absent in “autologous tumor vaccination”. In this procedure, the patients own tumor material, containing all relevant tumor antigens is excised, homogenized and irradiated to preclude further cell divisions. Before reinjection of this material, adjuvant is added (e.g. BCG, an inactivated form of Mycobacterium Bovis, a pathogen very similar to the one that causes tuberculosis) in order to provide danger signals12. 15
AN INTRODUCTION
TO THE IMMUNE
SYSTEM AND TO THE PRINCIPLES OF TUMOR VACCINATION
The results of tests with MVA based and autologous cancer vaccines have been disappointing13. In order to increase vaccine strength, many experimental tumor vaccines try to focus the T cell response to a limited and defined number of tumor antigens. Such tumor vaccines incorporate a single or a small set of antigens, or even a single antigenic peptide. Peptide vaccines are inherently MHC-restricted, since the peptides that they contain can only be presented to T cells if the MHC of the vaccinee is appropriate. Moreover, designing such vaccines requires knowledge about the antigenic epitopes that are present in the tumor. The antigen of an active specific vaccine can also be administered in the form of DNA. DNA holds (“encodes”) the blueprints for all our proteins and is contained in the nucleus (the central compartment) of each cell. Apart from one extra step, the mechanism of “DNA vaccination” is the same as that of conventional protein and peptide vaccines. This step implies so-called “transfection”: upon injection of the DNA, the DNA molecules enter cells of the patient surrounding the injection site. Subsequently, once the DNA enters the nucleus, the transfected cell will mistake the plasmid DNA for its own and will start to produce the antigen that the DNA vaccine encodes. For optimal transfection in living tissues (“in vivo”) the DNA string should have no loose ends (it should be a circle, a ‘plasmid’) and it should be a lot of DNA. After cells have been transfected with the DNA vaccine and are producing the antigen that it encodes, the immune system will respond to this antigen just as it responds to the antigens in a protein vaccine. Yet another immunization method that is currently pursued, is the injection of autologous APCs that have been “loaded” in a culture dish (“in vitro”) with antigens or peptides. Such APCs can also be transfected in vitro with DNA encoding these peptides. Although the method is complicated an laborious, it has proven to be both a potent immunization method and a powerful tool for research on the mechanisms involved in T cell priming. Developing and Testing vaccines The immune system is highly complex. Although most sub-systems may have been identified and dissected, our understanding of their cellular interplay is incomplete. Moreover, our insight in the internal wiring of immune cells is just evolving. Therefore, we cannot simulate the immune system in a test tube and use this to test experimental vaccines; either we experiment on Homo Sapiens or we use a comparably complex substitute organism to develop new vaccines.
16
i The most convenient substitute for Homo Sapiens is the Mus Musculus; the mouse. It breeds fast, doesn’t need much space and has a comparably complex immune system. However, there are considerable differences. As far as anatomy is concerned there are obvious differences in skin anatomy that are particularly relevant for the development of dermal DNA vaccines: mice have a thin skin with extensive hair growth whereas human skin is thicker and virtually hairless. Furthermore, the weight of an adult mouse is approximately 30g, which is 2500 times less than that of a human adult. Calculating the dose of a vaccine to be used in humans by multiplying the dose in mice with this factor, results in unacceptably large doses for humans (or unworkably small doses for mice). Biochemically, there are many differences as well. For example the lack of TLR9 expression in human skin and the high resistance of mice to LPS, a toxic contamination in DNA preparations. Collectively these issues make the “translation” of a mouse vaccine to a human vaccine a difficult task. Altogether, mice are suitable for uncovering fundamental immunological mechanisms and for the first steps in the development of experimental vaccines. For a trustworthy assessment of the applicability of an experimental vaccine in humans, it is wise to switch to an animal model that better resembles humans, for example monkeys such as the rhesus macaque (Macaca Mulatta).
17
INTRODUCTION TO THE WORK DESCRIBED IN THIS THESIS
Introduction to the work described in this thesis Five reasons why T cells fail to reject tumors spontaneously Unfortunately, in the vast majority of patients suffering from a cancer T cells will not spontaneously eradicate the tumor. Although it is tempting to speculate that many tumors are eradicated by our cellular immune system before they are diagnosed, this is not the case. Based on the comparison of cancer incidence in healthy and immunocompromized individuals, the role of the immune system in such “tumor-immunosurveillance” is modest at most14 and mainly confined to virally induced tumors. Here I will give a brief overview of the hurdles that stand between T cells and their eradication of a tumor. 1a) Tumors may not express tumor antigens. To put it differently: the T cell repertoire of the patient may be centrally tolerized towards all proteins expressed by the tumor to an extent that hampers their activation. If this is the case, no T cell response will ever occur during the disease. 1b) Neither is a spontaneous T cell response likely to occur if the T cell repertoire is peripherally tolerized towards the tumor antigens. Although in some instances danger signals provided by the tumor may breach this tolerance. 2) Since all tumor antigens – except virally derived antigens – are inherently self-proteins, the tumor specific response usually consists of T cells with suboptimal, low-affinity TCRs. Hence, these T cells may get insufficiently stimulated to fully differentiate to effector cells and exert their cytotoxic effects15;16. 3) If fully differentiated tumor specific T cells do occur in a cancer patient, they can be anergized or even killed by interactions with the tumor cells or the tumor stroma17;18. The expression of FasL by tumor cells or the secretion of TGF-ß by the tumor stroma are notorious examples. 4) If such fully differentiated T cells are not hindered by the tumor or the tumor stroma and can fulfill their cytotoxic duties, then the size and expansion rate of the tumor mass may preclude T cell mediated tumor eradication. To put it differently, the cells may arrive too late or the number of T cells may not be sufficient 19, this thesis. 5) Even if fully differentiated T cells, unhindered by interactions with the tumor or the tumor stroma arrive sufficiently early, they may encounter a tumor that is refractory to T cell killing due to loss of MHC
18
i expression. This renders the tumor specific T cell response useless at once20. Only if all these hurdles have been passed, T cells may spontaneously eradicate the tumor. But even than the patient may not be cured. The uncontrolled division in tumor cells will have spawned progeny with varied phenotypes; when any of these cells possesses a phenotypic trait that hampers T cell killing, this cell will survive the T cell attack and give rise to a new tumor that will be resistant to T cell killing. This phenomenon has been termed “cancer immuno editing”21. To investigate the quantity and the quality of spontaneously occurring melanoma specific T cell responses in light of these issues, we studied melanoma specific T cell responses in 62 end stage melanoma patients. The results are described in the first chapter of this thesis. Challenges for cancer vaccines Cancer vaccines have to face all the challenges mentioned above. Concerning the first hurdle, active immunization will clearly have no benefit for tumors that do not express tumor antigens (hurdle 1a). In this case passive immunization is the immunotherapy of choice; by transfer of tumor specific T cells that either may or may not have been transduced with TCR encoding genes. Theoretically, a sufficiently potent active vaccination strategy might be able to induce tumor specific T cells, if they are present but reside in a peripherally tolerized state (hurdle 1b). Experimental cancer vaccines cannot yet breach peripheral tolerance towards tumor antigens, without breaching that to self proteins as well. But one day the growing insight in peripheral tolerance mechanisms, combined with stronger vaccination methods, may make this possible. Chances for successful tumor specific vaccination are higher if the cancer patient possesses a non-tolerized T cell repertoire that is non functional because it consists of T cells with a low-affinity TCR (hurdle 2). In this case, the extra stimulation provided by the vaccine may help these T cells to differentiate to full-blown effector cells. In cases where such cells are hindered by tumor-mediated immunomodulation (hurdle 3) or faced with an overwhelming tumor load (hurdle 4) active immunization may be beneficial as well, simply by activating T cells at an earlier stage of the disease. However, in most patients a combination of low TCR affinity, high tumor load and tumor immuno-modulation will make successful active immunization a very difficult feat22;23.
19
INTRODUCTION TO THE WORK DESCRIBED IN THIS THESIS
A more realistic target to pursue with the active immunization methods that so far have been developed, is to design vaccines against virally derived tumors expressing highly immunogenic tumor antigens (e.g. cervical carcinoma). With the current state of the art, for most other cancers passive immunization methods are more likely to be successful8. However, it should be kept in mind that when the tumor cells do not express MHC (hurdle 5), all T cell mediated therapies will fail to impact on the tumor. Still, NK cell based immuno therapies may have potential in this situation. However, NK cell mediated therapies fall outside the scope of this thesis. Although up to this day active therapeutic immunization in patients suffering from non-virally induced tumors has never generated satisfactory results, we tested the effect of a peptide vaccine adjuvanted with GM-CSF and tetanus toxoid in end stage melanoma patients in the hope that this combination might sort some effect. In the second chapter the results of this study are presented and discussed. The development and evaluation of a novel active immunization method Active specific immunization using DNA offers several advantages over peptide vaccines: low production costs, flexibility in the use of antigen (resulting from the relief of constraints posed by large scale protein or peptide production), and remarkable keeping qualities. However, a major disadvantage of DNA vaccines is their modest potency and the length of the canonical DNA vaccination regimen24. In addition, there have been doubts about its safety, as theoretically the vaccine could integrate in the genome of host cells. These concerns however were soothed in recent trials. Since the first report of successful DNA vaccination in the early nineties25;26, most DNA vaccines have been administered either intramuscularly by injection or intradermally using a gene gun27. Originally, the gene-gun principle was developed to shoot DNA through the tough cellulose wall of plant cells, using DNA coated to gold or tungsten bullets28. Although animal cells do not have this cellulose shell, the bullet principle became a standard for human tissue transfection as well. Through the years however, many other means of dermal DNA injection have been developed. Examples are the Med-E-Jet29, the Pigjet30 the Biojector31 and the Medijector32. Most of these tools pneumatically inject an aerosolized DNA solution in the skin. Yet other methods include topical skin creams33 and even the use of polynucleotide coated suture34. Furthermore, oral and intranasal administration of DNA vaccines are currently explored as
20
i well35. All these dermal administration methods are high-tech, requiring dedicated instruments and batches of specially prepared DNA. In the work described in chapter 3 we set out to improve the main weaknesses of DNA vaccination: its time consuming regimen and its feeble potency. In order to achieve this we explored antigen production and presentation after intramuscular and dermal DNA vaccination. To deliver the DNA intradermally we employed a very low-tech administration method: a tattoo device. Since most DNA immunization methods that were proven to be effective in mice failed in larger species (i.e. in apes, monkeys and humans36) we tested the DNA tattoo method introduced in chapter 3 in a small Macaca Mulatta immunization study. The results of this study are described in chapter 4. Requirements for antigens encoded in DNA vaccines An increasing amount of evidence in mouse models points out that dermal DNA vaccines depend on cross presentation37. Additionally, cross presentation has been identified as an important step in the induction of T cell responses against tumors38. What determines the efficacy of the cross-presentation of antigens? In other words, what protein characteristics are required for optimal transfer from the antigen producing cells (the “antigen donor cells”) to the antigen presenting cells, and why? Understanding the selectivity in cross-presentation could reveal some tricks that pathogens or tumors may play to avoid presentation. Moreover, it would be very useful for the optimization of dermal DNA vaccines as well. In chapter 5 the results of a study on the impact of the antigen stability on the efficiency of its cross presentation are presented. These data shed light on the in vivo mechanism of cross presentation. Furthermore, they are highly relevant for the design of DNA vaccines. The development of tools for vaccine research The efforts to improve intradermal DNA vaccination are hindered by a lack of knowledge about its mechanism. To clarify the mechanisms involved in dermal DNA vaccination we developed a tool that enables us to follow the cellular events in the dermis after DNA tattooing. The tool comprises a intravital microscopy method and is presented in chapter 6. Since the efficacy of any T cell inducing vaccine is predominantly defined by the number and the cytotoxicity of the T cells that it induces, sensitive and specific T cell detection tools are pivotal for the development of vaccines. Historically, T cells directed towards a certain antigen 21
could only be discriminated from the rest of the T cells in ‘functional’ assays: T cells produce cytokines upon incubation with their cognate antigens, discerning them from naive or non-specific T cells. With the advent of fluorochrome labeled MHC/peptide tetramers T cells could be discriminated from each other directly, by the specificity of their TCR, irrespective of their functionality. Moreover, the ease of tetramer stainings facilitated the analysis of the kinetics of T cell responses39. However, tetramer analysis in laboratory animals and patients requires knowledge of the MHC haplotype of the individual, since tetramers are inherently MHC restricted. Combined with the complexity of their production, this MHC restriction has discouraged the use of tetramers in outbred (non MHC identical) primates and humans. Furthermore, it has hindered the development of high throughput methods for T cell analysis, for example by tetramer arrays40. A significant improvement in the production process of tetramers is described in the seventh and last chapter of this thesis. Besides facilitating the use of tetramer analysis in outbred populations, it will facilitate novel means of high-throughput T cell analysis. Overview of the work described in this thesis The first chapter, entitled “On the role of melanoma-specific CD8+ T cell immunity in disease progression of advanced-stage melanoma patients”, describes the meager impact of T cells on disease progression in malignant melanoma patients. Subsequently, the second chapter “Phase I clinical study with multiple peptide vaccines in combination with tetanus toxoid and GM-CSF in advanced-stage HLA-A*0201-positive melanoma patients”, reports the results of a peptide vaccination study in such patients. After reading through the disappointing results of this trial, the spirit may be brightened by the prospect of a new experimental immunization method. The article reporting on the efficacy of this “ DNA tattoo” method in mice is entitled “A rapid and potent DNA vaccination strategy defined by in vivo monitoring of antigen expression”. Chapter 4 “Intradermal DNA tattooing primes robust T cell responses in Macaca Mulatta” reports on the results of a small rhesus macaque study, intended to evaluate this immunization method in larger animals. Chapter 5 addresses a fundamental immunological question: what properties should a vaccine encoded protein that is produced in the upper layer of the skin upon DNA tattooing have for efficient shipping to the DLN and presentation to T cells that reside there? The paper entitled “In vivo antigen stability affects vaccine efficacy” at the end of this chapter identifies one important property: it should be a stable protein.
22
i The sixth chapter, “High-throughput intravital imaging of fluorescent markers and FRET probes by DNA tattooing”, describes a method to microscopically study the cellular events in the skin of living mice following DNA tattooing. The insights gained from these studies may be useful for a better insight in immunological processes in the skin and especially for the improvement of dermal DNA immunization methods. The final chapter, entitled “Design and use of conditional MHC class I ligands”, describes a novel method to produce MHC-peptide tetramers. The technology facilitates new methods for high-throughput enumeration of T cells with multiple specificities.
23
C H A P T E R
1
On the Role of Melanoma-Specific CD8+ T-Cell Immunity in Disease Progression of AdvancedStage Melanoma Patients. Adriaan Bins1, Monique van Oijen1, Sjoerd Elias2, Johan Sein2, Pauline Weder1, Gijsbert de Gast2, Henk Mallo1, Maarten Gallee3, Harm van Tinteren4, Ton Schumacher1 and John Haanen1,2 Divisions of 1 Immunology, 2 Medical Oncology, 3 Oncologic Diagnostics, and Statistics, The Netherlands Cancer Institute, Amsterdam, the Netherlands
4
Abstract Cytotoxic T-cell immunity directed against melanosomal differentiation antigens is arguably the best-studied and most prevalent form of tumor-specific T-cell immunity in humans. Despite this, the role of T-cell responses directed against melanosomal antigens in disease progression has not been elucidated. To address this issue, we have related the presence of circulating melanoma-specific T cells with disease progression and survival in a large cohort of patients with advanced-stage melanoma who had not received prior treatment. In 42 (68%) of 62 patients, melanoma-specific T cells were detected, sometimes in surprisingly large numbers. Disease progression during treatment was more frequent in patients with circulating melanoma-specific T cells, and mean survival of patients with circulating melanoma-specific T cells was equal to the survival of patients without melanoma-specific T cells. These data suggest that the induction of melanosomal differentiation antigen-specific T-cell reactivity in advanced stage melanoma is a late event most likely due to antigen load and spreading and is not accompanied by a clinically significant antitumor effect. These melanoma-specific T cells may be functionally distinct from T cells raised during spontaneous regression or upon vaccination. 24
Introduction Melanoma is considered one of the most immunogenic tumors. Spontaneous remissions occasionally occur in melanoma patients albeit infrequently and immunotherapy-induced remissions correlate with autoimmune skin depigmentation (1, 2, 3, 4). These phenomena have, at least in part, been attributed to a cellular immune response, because the presence of tumor-infiltrating T lymphocytes in primary melanoma and in melanoma lymph node metastases are independent positive prognostic factors (5 , 6). In past years, extensive efforts to identify target molecules for melanoma-reactive T cells have resulted in the identification of a large set of melanoma-associated antigens (7). These antigens can be classified as melanocyte lineage-specific antigens (MART-1/Melan-A, tyrosinase, gp100) and antigens derived from genes expressed in testis and a variety of cancers (including MAGE-family, NY-ESO-1, PRAME). Melanocyte lineage antigens are expressed in a large fraction of melanomas, and a substantial number of epitopes from these antigens that are recognized by cytotoxic T cells have been mapped (8). With the aid of soluble tetramerized MHC complexes containing these epitopes, melanosomal antigen-specific CD8+ T cells have been detected in peripheral blood from melanoma patients (9). Large expansions of these T cells are primarily observed in the peripheral blood or tumor-infiltrated lymph nodes from stage III and particularly stage IV melanoma patients. In patients with tumors that are confined to the primary site, such as in stage I and II melanoma, MHC tetramer-positive T cells have been found, but at significantly lower frequencies (10) . Despite the presence of naturally occurring melanoma-specific T-cell immunity in a large proportion of melanoma patients, spontaneous remissions in this group of patients are extremely rare and it is currently unclear what the significance is of spontaneous melanoma-specific T-cell immunity for these patients. Patients & Methods Patients. Peripheral blood samples were taken from 62 HLA-A*0201 advanced-stage melanoma patients (stage IV) at the time that they had been diagnosed with metastatic melanoma. None of the patients had undergone prior treatment for metastatic melanoma. Patients were treated with standard chemotherapy with or without cytokine therapy mostly as part of a clinical trial (11), as indicated in Table 1 MHC Tetramers. MHC class I tetrameric complexes were prepared as previously described with minor modifications (12). HLA-A2.1-peptide complexes were generated with the following five antigenic peptides: (a) influenza-A matrix58–66; (b) GILGFVFTL, MART126–35; (c) ELAGIGILTV (A27L variant 13); (d) tyrosinase368–376, YMDGTMSQV (N370D variant 14); and (e) gp100280–288, YLEPGPVT (A288V variant 15). MHC class I-peptide complexes were subsequently purified, biotinylated by biotin
25
1
ON THE ROLE OF MELANOMA-SPECIFIC CD8+ T-CELL IMMUNITY IN PROGRESSION OF ADVANCED-STAGE MELANOMA
ligase, purified and converted to tetramers by the addition of phycoerythrin-labeled streptavidin (Molecular Probes), and stored at –20°C in 16% glycerol/0.5% BSA. MHC-Tetramer Staining. It has previously been shown that CD8 can play a critical role in the binding of MHC tetramers to antigen-specific T cells (16, 17, 18) . In particular, in case of lowaffinity T-cell receptors, the simultaneous binding of CD8 to the MHC α3 domain is necessary to obtain a sufficiently stable complex between MHC tetramers and antigen-specific T cells, and in such cases, costaining with CD8 can block MHC tetramer binding (16, 17, 18) . In line with this, we observed diminished MHC tetramer staining of a tyrosinase-specific T-cell clone in the presence of CD8 monoclonal antibody. Likewise, a large fraction of the tyrosinase-specific CD8+ T cells in the peripheral blood of advanced-stage melanoma patients became undetectable by MHC tetramer staining in the presence of CD8 monoclonal antibody (data not shown). To overcome this issue, peripheral blood mononuclear cells (PBMCs) were stained with a set of lineage (lin) marker antibodies specific for B-cells (CD19), natural killer cells (CD16), monocytes (CD14) granulocytes (CD13), and CD4+ T cells (CD4), and the linnegative lymphocyte subset was used for analysis. Linnegative lymphocytes are >95% CD8-positive and HLA-A2.1/tyrosinase tetramerpositive cells within the linnegative population are likewise >95% CD8-positive, thereby validating this strategy (Fig. 1) Thawed PBMC samples were incubated overnight at 37°C in Iscove’ s medium with 10% FCS to recover PBMCs and eliminate apoptotic cells (9) . After washing, the cells were incubated for 5 min in cold PBS with 0.5% BSA and 1% normal mouse serum to block Fc-receptors. Two million cells per sample were incubated for 10 min with 2 µg/ml phycoerythrin-labeled MHC-tetramer at 37°C (19). CD8+ T cells were negatively selected by staining with a large set of FITC-labeled lineage marker antibodies (CD4, CD13, CD14, CD16, CD19; Becton-Dickinson). Cells were stained with propidium iodide to be able to gate out dead cells. Samples were analyzed by flow cytometry using a FACScalibur and Cell Quest software (Becton Dickinson). Forward and side scatter parameters were used to define lymphocyte populations. Background and detection limits were separately established for each tetramer by staining of PBMCs from four HLA-A2-negative individuals with phycoerythrinlabeled MHC-tetramers and the panel of FITC-labeled antibodies. If more than 50,000 CD3+/CD8+ cells could be analyzed by flow cytometry in the HLA-A2positive samples, the mean percentage of MHC-tetramer positive cells +3x SD of four HLA-A2-negative PBMC samples was used as the detection limit for positive MHC-tetramer staining. If less than 50,000 CD3+/CD8+ cells could be analyzed, the mean percentage +6x SD was used as the detection limit. Immunohistochemistry.
26
Formalin-fixed paraffin-embedded material of melanoma metastases was available from 17 of 62 patients. Sections (4-µm) were stained for MART-1 (Ab-3; Neomarkers; 1:250), tyrosinase (clone T311; Neomarkers; 1:100), gp100 (HMB-45; DAKO; 1:200) or HLA-A (HC-A2; generously donated by J. Neefjes, Netherlands Cancer Institute, Amsterdam; 1:400). For all stainings except for gpl00, microwave antigen retrieval in citrate buffer was performed. Stainings were performed with standard alkaline-phosphatase three-step immunohistochemistry. Statistics. End point of the study was overall survival, measured from start of treatment until death or last follow-up. Patients alive at the time of analyses were censored. Survival estimates and curves were made with the Kaplan-Meier technique, and differences between groups were tested by a log-rank test. Cox proportional hazard analysis was used to estimate the size of the effect (hazard ratio) with 95% confidence intervals. The relation of MART-1 with response (in three categories) was tested by a Cochrane-Armitage trend test. Wilcoxon test was used to perform a statistical analysis on the phenotypical characteristics that distinguish naïve and effector/memory MART-1-specific T cells. Results Patients. Between 1998 and 2002, 62 HLA-A*0201-positive patients were enrolled (Table 1). All patients had advanced-stage melanoma (stage IV) with a median organ involvement of 2 (range 1–5). The majority (42 of 62) of the patients had visceral metastases, mostly in lung and/or liver. Nonvisceral metastases (20 of 62) were most often located in lymph nodes and subcutis. Five patients were treated with standard dacarbazine (800 mg/m2, every 3 weeks), 10 patients were treated with temozolomide (200 mg/m2 for 5 days, every 4 weeks) and 47 patients were treated with combined chemobiotherapy (temozolomide for 5 days, triple cytokine interleukin 2, interferon-α, granulocyte macrophage colony-stimulating factor for 12 days, every 4 weeks). The latter two treatments were given as part of Phase II and III clinical trials. Analysis of Patient-Derived Peripheral Blood Samples. Peripheral blood samples were obtained before initiation of therapy and were analyzed for the presence of melanoma-specific CD8+ T cells (Fig. 2) . In 33 patients (53%) MART-1-specific CD8+ T cells were detectable, and 17 patients (27%) displayed expansions of tyrosinase-specific CD8+ T cells. Eight of these latter patients also had MART-1-specific T cells. Only 1 patient tested positive for gp100-specific CD8+ T cells; this patient also had high numbers of circulating MART-1-specific T cells (18% of total CD8+ T-cell pool). In the vast majority of samples that contained tyrosinase-specific T cells, these CD8+ T cells were clearly detectable when counterstained using the panel of lineage marker antibodies but were difficult to detect when costaining with anti-CD8 antibody was performed. In 27
1
ON THE ROLE OF MELANOMA-SPECIFIC CD8+ T-CELL IMMUNITY IN PROGRESSION OF ADVANCED-STAGE MELANOMA
line with this, the avidity of tyrosinase-specific T cells appears lower than that of MART-1- or gp100-specific T cells, as judged from the intensity of MHC tetramer staining. As a control, influenza A virus-specific CD8+ T cells could be detected in 70% of patients analyzed (n = 36). In the majority of patients [27 (64%) of 42] with MHC tetramer+ T cells, the frequencies of melanoma-specific T cells in peripheral blood were above 1:1000 CD8+ T cells, ranging from 1:6 to 1:900. MART-1-specific CD8+ T cells have been detected in the peripheral blood of healthy individuals in frequencies up to 1 of 1,000 (20). Data from Dutoit et al. (21) indicate that these populations are a direct reflection of a high thymic output of T cells with this specificity, and, in line with this, these MART-1-specific T cells display a naïve phenotype. To establish whether the MART-1-specific T-cell populations detected in our cohort arose through antigen-driven expansion, the MHC tetramer-positive CD8+ T cell subset was analyzed directly ex vivo for cell surface expression of naïve-, effector-, and memory-type-associated markers. In humans, the CD45RA and CD45RO surface antigens have been used to identify naïve and memory T cells, respectively, and costaining for CD27 expression may be used to distinguish between effector cells and naïve or memory T cells (22). Using the Wilcoxon test, we found two statistically distinct patterns of cell surface expression (Table 2) . In patients with low but detectable frequencies (< 0.2%) of MART-1-specific T cells, expression of CD45RA and CD27 were high (P = 0.033 and P = 0.0015, respectively); and expression of CD45RO was low (P = 0.0004). This pattern of cell surface expression reflects a naïve phenotype. In contrast, in patients with high frequencies (> 0.5%) of MART-1-specific T cells CD4RO expression was high, whereas expression of CD45RA and CD27 was often reduced as compared with naïve T cells. In patients with 0.2–0.5% MART-1-specific T cells, both naïve and antigen-experienced phenotypes were found. In summary, in line with prior data (20), two groups of patients can be distinguished: one group [15 (45%) of 33 with low numbers of MART-1-specific CD8+ T cells in peripheral blood and a second group [18 (55%) of 33] with high numbers of MART-1-specific T cells. The phenotype of the MART-1-specific T cells of the majority of these patients has been analyzed; the analysis confirms that in patients with low numbers of MART-1-specific T cells, these cells display a naïve phenotype, whereas those cells in patients having high numbers are phenotypically antigen experienced. Melanoma-Specific T cells and Survival. Although the presence of melanocyte differentiation antigen-specific T cell immunity in patients with stage IV disease is well documented, the relevance of these T-cell responses in terms of response to treatment or survival remain unknown. We, therefore, attempted to relate the presence of melanoma-specific T cells to survival in this large cohort of HLA-A*0201+ stage IV melanoma patients. Median follow-up was 46 months. At the time of analysis, 55 patients had died and the median survival of all patients was 7.8 months (Fig. 3 ; 95% confidence 28
interval, 6.0–9.7 months). This is in close agreement with the median survival observed in several randomized clinical trials using dimethyltriazeno-imidazolecarboxamide- or temozolomide-based treatment regimens for advanced-stage melanoma patients (23 , 24). To test whether the presence of circulating melanoma-specific CD8+ T cells was related to clinical outcome, survival of patients that had detectable numbers of circulating melanoma-specific T cells was compared with those without these cells (Fig. 4A). This analysis revealed that the presence of T cells specific for melanocyte lineage antigens is not associated with a survival benefit (log-rank, P = 0.354). Survival curves of patients with either MART-1 or tyrosinase-specific T cells compared with patients without any melanoma-specific T cells, likewise, do not show any detectable survival advantage (P = 0.721 and P = 0.346 respectively; Fig. 4, B and C). To test whether a possible antitumor effect would manifest itself in patients with MART-1-specific T cells displaying an effector/memory phenotype, survival was also analyzed for this group (n = 8; Fig. 4D). No statistically significant difference between survival of patients with effector/memory T cells and those without circulating T cells was observed (log-rank, P = 0.224). In summary, the presence of melanoma-specific T cell populations in peripheral blood does not lead to a survival benefit for patients with advanced-stage melanoma. We considered the possibility that the melanoma-specific T cells we observed could have a limited capacity to kill tumor cells, for instance, because of a low avidity. In such a scenario, a high T-cell response would not predict a subsequent antitumor effect but would merely be a reflection of a growing tumor mass. To test this hypothesis, patients were divided into three groups: those with disease progression within 2 months after blood sampling, those with stable disease, and those with partial/complete response within this time period. When comparing these groups for the presence of circulating MART-1-specific T cells, we found a statistically significant (two-sided P = 0.0133, Cochran-Armitage trend test) predominance of high numbers (≥0.1%) MART-1 specific T cells in the group of patients with progressive disease in the first 2 months after blood sampling (Table 3) Immune Escape through Loss of HLA-A Expression. Tumors may evade the action of tumor-specific T cells by down-regulation of either MHC class I cell surface expression or by reducing antigen expression. To address this issue, we examined antigen and HLA-A expression by immunohistochemistry in tumor material that was available from only 17 of the 62 patients (data not shown). All 17 melanoma metastases expressed MART-1 and tyrosinase, whereas gp100 expression was homogeneously expressed in tumor material of 13 of 17 patients. Twelve of these 17 patients had circulating MART-1-specific T cells, 3 patients had tyrosinase-specific T cells, and 1 patient had gp100-specific T cells. In 7 of 17 metastases, HLA-A expression was lost. Lack of HLA-A expression was observed in 5 of 12 metastases from patients with detectable melanoma-specific T 29
1
ON THE ROLE OF MELANOMA-SPECIFIC CD8+ T-CELL IMMUNITY IN PROGRESSION OF ADVANCED-STAGE MELANOMA
cells, and 2 of 5 patients without detectable melanoma-specific T cells. These data demonstrate that although MHC class I loss does not directly correlate with the presence of detectable tumor-specific T-cell responses, the frequent occurrence of class I loss suffices to explain the lack of antitumor effect in a substantial fraction of the advanced-stage melanoma patients. Discussion Several groups have reported the presence of naturally occurring melanoma-specific T-cell reactivity in melanoma patients (7 , 9 , 25, 26, 27, 28, 29, 30, 31), but little is known on the time frame of appearance and even less on the possible role of these cells in limiting tumor outgrowth. To address this issue, we have examined spontaneous cytotoxic T-cell immunity against melanosomal antigens, MART-1, tyrosinase, and gp100 in a large group of advanced-stage (IV), HLA-A*0201-positive melanoma patients and have correlated these responses with patient survival. T-cell responses were analyzed before the initiation of treatment and should, therefore, reflect the natural response of the immune system to the growing tumor mass. These data demonstrate that, in accordance with other studies, substantial expansions of CD8+ T cells specific for melanosomal antigens can occur in patients with metastatic disease, amounting up to 18% of the total CD8+ T-cell population in peripheral blood. Importantly, these high numbers of tumor-specific CD8+ T cells do not have a measurable positive influence on patient survival. Although the primary goal of this study was to evaluate a postulated benefit of spontaneous melanocyte lineage-specific T-cell responses, it may be useful to discuss these data in relation to immunotherapeutic approaches in this patient group. Several mechanisms may account for the fact that the T-cell populations specific for these melanoma antigens do not measurably impinge on tumor growth, and, depending on the mechanism involved, immunotherapeutic approaches that aim to strengthen these responses may or may not be worth pursuing. We considered two mechanisms: “immune escape” and “insufficient number or activity of tumorspecific T cells. With regard to immune escape, tumors may evade the action of tumor-specific T cells by down-regulation of either MHC class I cell surface expression or by reducing antigen expression. In the tumor samples available for this study, we found loss of HLA-A expression in a substantial portion of the cases, whereas in all of the cases, antigen expression was maintained. This frequent occurrence of class I loss suffices to explain the lack of antitumor effect in a substantial fraction of the advanced-stage melanoma patients. In addition, it cannot be excluded that in part of the remaining cases, escape from T-cell attack is achieved by other means. With regard to insufficient number or activity of tumor-specific T cells, in 7 of 12 patients with detectable tumor-specific T-cell reactivity, there was no concomitant loss of HLA-A or antigen expression. It is tempting to speculate that in these patients the number and/or activity (be it either homing capacity or cytolytic
30
function) of the tumor-specific T cells was insufficient to mediate a significant antitumor effect. Because vitiligo is associated with successful immunotherapy (1, 2, 3, 4) , further indirect evidence for this notion is provided by the observation that none of the 42 patients with detectable T-cell responses against melanosomal antigens showed signs of vitiligo. Lack of vitiligo despite substantial numbers of T cells in some of the patients may point to functional differences between T cells induced by immunotherapy and the spontaneous responses observed in this study. MHC tetramers used in this study are indicative of the T-cell receptor specificity of the different melanoma-associated antigen-specific T-cell populations. It has been demonstrated that MHC tetramer binding does not always fully correlate with functionality (32). However, in one patient the available blood samples did allow simultaneous MHC tetramer and intracellular IFN-γ staining. In this patient with high numbers of both MART-1 and gp100-specific T cells, intracellular IFN-γ was produced against peptide-pulsed target cells directly ex vivo. This result, however, does not preclude the possibility that for other patients the detected melanomaspecific T-cell populations were nonfunctional or tolerized. On the basis of the current data, it seems that spontaneously occurring melanocyte lineage-specific immunity in stage IV melanoma patients is largely a reflection of increasing tumor mass and spreading, and stage IV melanoma patients are unlikely to benefit from these immune responses other than in exceptional cases. These melanoma-specific T cells may be functionally distinct from T cells that play a role in situations of spontaneous regression that rarely occur in these patients or from T cells that are detected on successful immunotherapy. Escape from immunity by loss of expression of target molecules either by immune selection or genetic instability of tumor cells forms one major factor that limits the effect of tumor-specific CTLs in part of this advanced-stage patient group. In the remaining patients, the limited efficacy of the tumor-specific T-cell response may well be related to the poor quality and suboptimal numbers of tumor-specific T cells at early time points during the disease. Efforts to enhance these responses by vaccination strategies or through adoptive therapy are, therefore, worth pursuing and are meeting increasing success (4 , 33 , 34). Acknowledgements We thank Colette van den Bogaard and Willeke van den Kasteele for their help with sample preparation, and Esther Kueter and Raquel Gomez for preparing MHCtetramers.
31
1
ON THE ROLE OF MELANOMA-SPECIFIC CD8+ T-CELL IMMUNITY IN PROGRESSION OF ADVANCED-STAGE MELANOMA
Table 1 Patient characteristics NO. OF PATIENTS
62
MEDIAN AGE
48 YR (RANGE, 21–79 YR)
SEX FEMALE
24 (39%)
MALE
38 (61%)
METASTATIC SITES, N (MEDIAN)
2 (1–5)
VISCERAL
42 (68%)
NONVISCERAL
20 (32%)
SERUM LACTATE DEHYDROGENASE (MEDIAN)
370
TREATMENT
a
DTICA
5 (5.1%)
TEMOZOLOMIDE
10 (6.2%)
CHEMOBIOTHERAPY
47 (76%)
DTIC, dimethyltriazeno imidazole carboxamide.
Fig. 1. Peripheral blood cells were stained with HLA-A2.1 tetramers containing the tyrosinase368–376 peptide followed by staining with a panel of lineage antibodies, as described in “Patients and Methods.” Cells were analyzed by flow cytometry (left panel), and lineage antigen negative and MHC tetramer-positive cells were sorted by fluorescence-activated cell sorting (Cell sorting). Enriched MHC tetramerpositive, lineage antigens negative cells (middle panel) were briefly incubated at 37°C, stained with CD8 monoclonal antibody, and reanalyzed by flow cytometry (right panel). Typically >95% of these cells are CD8+. 32
1
Fig. 2. A, representative dot plots of MHC-tetramer staining versus lineage antibodies. Top three panels, HLA-A*0201 melanoma patients (HLA-A2+) with detectable MART-126–35 (27L)-, tyrosinase368–376 (370D)-, and gp100280–288 (288V)specific CD8+ T cells in peripheral blood. Bottom three panels, control staining of HLA-A2.1-negative melanoma patients (HLA-A2–) with HLA-A2.1/MART-1, tyrosinase, and gp100 tetramer.
33
ON THE ROLE OF MELANOMA-SPECIFIC CD8+ T-CELL IMMUNITY IN PROGRESSION OF ADVANCED-STAGE MELANOMA
B, scatter diagram of all patients with circulating melanoma-specific T cells, separated by antigen-specificity (MART-1, tyrosinase, or gp100). On the basis of the literature, presence of naïve MART-1-specific T cells (MART-sp. T cells) amounts to about 0.1% (20) . Tm-positive, MHC tetramer-positive.
34
PATIENT NO.
CD45RA (%)
CD27 (%)
CD45RO (%)
CCR7 (%)
MART1-SPEC.A T CELLS (%)
1
GROUP A NKI9822
92
93
17
NT
0.5
NKI9834
100
100
45
NT
0.2
NKI9829
85
80
15
NT
0.2
NKI9852
75
100
30
NT
0.2
NKI9839
65
83
28
NT
0.1
NKI9843
91
100
0
NT
0.1
NKI9818
75
100
13
NT
0.1
NKI0032
91
79
9
82
0.1
NKI0057
94
95
23
76
0.05
34
64
76
NT
17.3
GROUP B NKI9835
a
NKI9836
51
31
52
NT
2.8
NKI9828
81
26
51
NT
2.6
NKI9849
14
97
98
NT
0.9
NKI9810
30
38
87
NT
0.9
NKI0002
99
31
64
1
0.5
NKI0043
3
27
93
3
0.2
NKI0021
84
54
28
7
0.2
MART-1-spec., MART-1-specific; nt, not tested.
Table 2 Naïve versus effector/memory phenotype of MART-1-specific T cells. Panel antibody negative and MHC tetramer-positive T cells were costained with the indicated antibodies. Numbers represent percentages surface antigen positive cells from total MART-1-specific CD8+ T cells. On the basis of differential expression of CD45RA, CD45RO, and CD27, patients with circulating MART-1-specific T cells can be divided into two statistically distinct groups, A and B. Group A, MART1-specific T cells displaying a naïve phenotype with high CD45RA (Wilcoxon test, 2-sided P = 0.033), high CD27 (Wilcoxon test, 2-sided P = 0.0015), and low CD45RO (Wilcoxon test: 2-sided P = 0.0004). When tested, CCR7 expression was high, supporting a naïve phenotype. The frequency of these T cells in peripheral blood is low, mostly <0.2% of peripheral CD8+ T-cell pool. Group B, MART-1specific T cells present at higher frequency in peripheral blood (mostly <0.5% of CD8+ T-cell pool) display an effector/memory phenotype with mostly high expression of CD45RO and low CCR7 expression (when tested). 35
ON THE ROLE OF MELANOMA-SPECIFIC CD8+ T-CELL IMMUNITY IN PROGRESSION OF ADVANCED-STAGE MELANOMA
Fig. 3. Kaplan-Meier survival curve of total cohort of HLA-A*0201+ stage IV melanoma patients. The median survival is 7.8 months (95% confidence interval, 6.0–9.7 months) with 55 deaths.
Fig. 4. Kaplan-Meier survival curves of HLA-A*0201+ patients with and without circulating melanoma-specific T cells. A, MART-1, tyrosinase- and gp100-specific CD8+ T cells; log-rank P = 0.354; hazard ratio, 1.32 [95% confidence interval (CI), 0.74–2.33]. 36
1
B, tyrosinase-specific T cells only; log-rank P = 0.721; hazard ratio, 0.89 (95% CI, 0.48–1.63). C, all MART-1-specific T cells; log-rank P = 0.346; hazard ratio, 1.29 (95% CI, 0.75–2.21). D, established effector/memory MART-1-specific T cells (n = 8); log-rank P = 0.224. 37
ON THE ROLE OF MELANOMA-SPECIFIC CD8+ T-CELL IMMUNITY IN PROGRESSION OF ADVANCED-STAGE MELANOMA
Table 3 Expanded MART-1-specific T cells preferentially in patients with rapid progressive diseasea. Patients were divided into three groups depending on their evaluated response to two courses of treatment. Treatment consisted of chemotherapy ± biotherapy (see Table 1 ). Responses were evaluated according to RECISTb criteria (35). The presence of expanded (>0.1%) MART-1-specific cells was compared with the presence of no or few MART-1-specific T cells (<0.1%).
MART-1-SPECIFIC CD8+ T CELLS
a
RESPONSE
<0.1%
>0.1%
TOTAL
CR/PR
12
2
14
SD
11
0
11
PD
21
16
37
TOTAL
44
18
62
Statistical analysis was performed with Cochran-Armitage trend test.
RECIST, response evaluation criteria in solid tumors; CR/PR, complete or partial response; SD, stable disease; PD, progressive disease. b
38
1. Yee, C., Thompson, J. A., Roche, P., Byrd, D. R., Lee, P. P., Piepkorn, M., Kenyon, K., Davis, M. M., Riddell, S. R., and Greenberg, P. D. Melanocyte destruction after antigen-specific immunotherapy of melanoma: direct evidence of t cell-mediated vitiligo. J Exp Med, 192: 1637-1644, 2000. 2. Phan, G. Q., Attia, P., Steinberg, S. M., White, D. E., and Rosenberg, S. A. Factors associated with response to high-dose interleukin-2 in patients with metastatic melanoma. J Clin Oncol, 19: 3477-3482, 2001. 3. Phan, G. Q., Yang, J. C., Sherry, R. M., Hwu, P., Topalian, S. L., Schwartzentruber, D. J., Restifo, N. P., Haworth, L. R., Seipp, C. A., Freezer, L. J., Morton, K. E., Mavroukakis, S. A., Duray, P. H., Steinberg, S. M., Allison, J. P., Davis, T. A., and Rosenberg, S. A. Cancer regression and autoimmunity induced by cytotoxic T lymphocyte-associated antigen 4 blockade in patients with metastatic melanoma. Proc Natl Acad Sci U S A, 100: 8372-8377, 2003. 4. Dudley, M. E., Wunderlich, J. R., Robbins, P. F., Yang, J. C., Hwu, P., Schwartzentruber, D. J., Topalian, S. L., Sherry, R., Restifo, N. P., Hubicki, A. M., Robinson, M. R., Raffeld, M., Duray, P., Seipp, C. A., Rogers-Freezer, L., Morton, K. E., Mavroukakis, S. A., White, D. E., and Rosenberg, S. A. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science, 298: 850-854, 2002. 5. Clemente, C. G., Mihm, M. C., Jr., Bufalino, R., Zurrida, S., Collini, P., and Cascinelli, N. Prognostic value of tumor infiltrating lymphocytes in the vertical growth phase of primary cutaneous melanoma. Cancer, 77: 1303-1310, 1996. 6. Mihm, M. C., Jr., Clemente, C. G., and Cascinelli, N. Tumor infiltrating lymphocytes in lymph node melanoma metastases: a histopathologic prognostic indicator and an expression of local immune response. Lab Invest, 74: 43-47, 1996. 7. Boon, T., Cerottini, J. C., Van den Eynde, B., van der Bruggen, P., and Van Pel, A. Tumor antigens recognized by T lymphocytes. Annu Rev Immunol, 12: 337-365, 1994. 8. Kawakami, Y., Robbins, P. F., Wang, R. F., Parkhurst, M., Kang, X., and Rosenberg, S. A. The use of melanosomal proteins in the immunotherapy of melanoma. J Immunother, 21: 237-246, 1998. 9. Lee, P. P., Yee, C., Savage, P. A., Fong, L., Brockstedt, D., Weber, J. S., Johnson, D., Swetter, S., Thompson, J., Greenberg, P. D., Roederer, M., and Davis, M. M. Characterization of circulating T cells specific for tumor-associated antigens in melanoma patients. Nat Med, 5: 677-685, 1999. 10. Palermo, B., Campanelli, R., Mantovani, S., Lantelme, E., Manganoni, A. M., Carella, G., Da Prada, G., della Cuna, G. R., Romagne, F., Gauthier, L., Necker, A., and Giachino, C. Diverse expansion potential and heterogeneous avidity in
39
1
ON THE ROLE OF MELANOMA-SPECIFIC CD8+ T-CELL IMMUNITY IN PROGRESSION OF ADVANCED-STAGE MELANOMA
tumor-associated antigen-specific T lymphocytes from primary melanoma patients. Eur J Immunol, 31: 412-420, 2001. 11. de Gast, G. C., Klumpen, H. J., Vyth-Dreese, F. A., Kersten, M. J., Verra, N. C., Sein, J., Batchelor, D., Nooijen, W. J., and Schornagel, J. H. Phase I trial of combined immunotherapy with subcutaneous granulocyte macrophage colonystimulating factor, low-dose interleukin 2, and interferon alpha in progressive metastatic melanoma and renal cell carcinoma. Clin Cancer Res, 6: 1267-1272, 2000. 12. Altman, J. D., Moss, P. A., Goulder, P. J., Barouch, D. H., McHeyzerWilliams, M. G., Bell, J. I., McMichael, A. J., and Davis, M. M. Phenotypic analysis of antigen-specific T lymphocytes. Science, 274: 94-96, 1996. 13. Valmori, D., Fonteneau, J. F., Lizana, C. M., Gervois, N., Lienard, D., Rimoldi, D., Jongeneel, V., Jotereau, F., Cerottini, J. C., and Romero, P. Enhanced generation of specific tumor-reactive CTL in vitro by selected Melan-A/MART-1 immunodominant peptide analogues. J Immunol, 160: 1750-1758, 1998. 14. Mosse, C. A., Meadows, L., Luckey, C. J., Kittlesen, D. J., Huczko, E. L., Slingluff, C. L., Shabanowitz, J., Hunt, D. F., and Engelhard, V. H. The class I antigen-processing pathway for the membrane protein tyrosinase involves translation in the endoplasmic reticulum and processing in the cytosol. J Exp Med, 187: 37-48, 1998. 15. Zarour, H., De Smet, C., Lehmann, F., Marchand, M., Lethe, B., Romero, P., Boon, T., and Renauld, J. C. The majority of autologous cytolytic T-lymphocyte clones derived from peripheral blood lymphocytes of a melanoma patient recognize an antigenic peptide derived from gene Pmel17/gp100. J Invest Dermatol, 107: 6367, 1996. 16. Daniels, M. A. and Jameson, S. C. Critical role for CD8 in T cell receptor binding and activation by peptide/major histocompatibility complex multimers. J Exp Med, 191: 335-346, 2000. 17. Campanelli, R., Palermo, B., Garbelli, S., Mantovani, S., Lucchi, P., Necker, A., Lantelme, E., and Giachino, C. Human CD8 co-receptor is strictly involved in MHC-peptide tetramer-TCR binding and T cell activation. Int Immunol, 14: 39-44, 2002. 18. Denkberg, G., Cohen, C. J., and Reiter, Y. Critical role for CD8 in binding of MHC tetramers to TCR: CD8 antibodies block specific binding of human tumorspecific MHC-peptide tetramers to TCR. J Immunol, 167: 270-276, 2001. 19. Whelan, J. A., Dunbar, P. R., Price, D. A., Purbhoo, M. A., Lechner, F., Ogg, G. S., Griffiths, G., Phillips, R. E., Cerundolo, V., and Sewell, A. K. Specificity of CTL interactions with peptide-MHC class I tetrameric complexes is temperature dependent. J Immunol, 163: 4342-4348, 1999.
40
20. Pittet, M. J., Valmori, D., Dunbar, P. R., Speiser, D. E., Lienard, D., Lejeune, F., Fleischhauer, K., Cerundolo, V., Cerottini, J. C., and Romero, P. High frequencies of naive Melan-A/MART-1-specific CD8(+) T cells in a large proportion of human histocompatibility leukocyte antigen (HLA)-A2 individuals. J Exp Med, 190: 705-715, 1999. 21. Dutoit, V., Rubio-Godoy, V., Pittet, M. J., Zippelius, A., Dietrich, P. Y., Legal, F. A., Guillaume, P., Romero, P., Cerottini, J. C., Houghten, R. A., Pinilla, C., and Valmori, D. Degeneracy of antigen recognition as the molecular basis for the high frequency of naive A2/Melan-a peptide multimer(+) CD8(+) T cells in humans. J Exp Med, 196: 207-216, 2002. 22. Hamann, D., Baars, P. A., Rep, M. H., Hooibrink, B., Kerkhof-Garde, S. R., Klein, M. R., and van Lier, R. A. Phenotypic and functional separation of memory and effector human CD8+ T cells. J Exp Med, 186: 1407-1418, 1997. 23. Chapman, P. B., Einhorn, L. H., Meyers, M. L., Saxman, S., Destro, A. N., Panageas, K. S., Begg, C. B., Agarwala, S. S., Schuchter, L. M., Ernstoff, M. S., Houghton, A. N., and Kirkwood, J. M. Phase III multicenter randomized trial of the Dartmouth regimen versus dacarbazine in patients with metastatic melanoma. J Clin Oncol, 17: 2745-2751, 1999. 24. Middleton, M. R., Grob, J. J., Aaronson, N., Fierlbeck, G., Tilgen, W., Seiter, S., Gore, M., Aamdal, S., Cebon, J., Coates, A., Dreno, B., Henz, M., Schadendorf, D., Kapp, A., Weiss, J., Fraass, U., Statkevich, P., Muller, M., and Thatcher, N. Randomized phase III study of temozolomide versus dacarbazine in the treatment of patients with advanced metastatic malignant melanoma. J Clin Oncol, 18: 158166, 2000. 25. Kawakami, Y., Nishimura, M. I., Restifo, N. P., Topalian, S. L., O’Neil, B. H., Shilyansky, J., Yannelli, J. R., and Rosenberg, S. A. T-cell recognition of human melanoma antigens. J Immunother, 14: 88-93, 1993. 26. Topalian, S. L., Hom, S. S., Kawakami, Y., Mancini, M., Schwartzentruber, D. J., Zakut, R., and Rosenberg, S. A. Recognition of shared melanoma antigens by human tumor-infiltrating lymphocytes. J Immunother, 12: 203-206, 1992. 27. D’Souza, S., Rimoldi, D., Lienard, D., Lejeune, F., Cerottini, J. C., and Romero, P. Circulating Melan-A/Mart-1 specific cytolytic T lymphocyte precursors in HLA-A2+ melanoma patients have a memory phenotype. Int J Cancer, 78: 699706, 1998. 28. Valmori, D., Dutoit, V., Lienard, D., Rimoldi, D., Pittet, M. J., Champagne, P., Ellefsen, K., Sahin, U., Speiser, D., Lejeune, F., Cerottini, J. C., and Romero, P. Naturally occurring human lymphocyte antigen-A2 restricted CD8+ T-cell response to the cancer testis antigen NY-ESO-1 in melanoma patients. Cancer Res, 60: 4499-4506, 2000.
41
1
ON THE ROLE OF MELANOMA-SPECIFIC CD8+ T-CELL IMMUNITY IN PROGRESSION OF ADVANCED-STAGE MELANOMA
29. Jager, E., Ringhoffer, M., Arand, M., Karbach, J., Jager, D., Ilsemann, C., Hagedorn, M., Oesch, F., and Knuth, A. Cytolytic T cell reactivity against melanoma-associated differentiation antigens in peripheral blood of melanoma patients and healthy individuals. Melanoma Res, 6: 419-425, 1996. 30. Jager, E., Nagata, Y., Gnjatic, S., Wada, H., Stockert, E., Karbach, J., Dunbar, P. R., Lee, S. Y., Jungbluth, A., Jager, D., Arand, M., Ritter, G., Cerundolo, V., Dupont, B., Chen, Y. T., Old, L. J., and Knuth, A. Monitoring CD8 T cell responses to NY-ESO-1: correlation of humoral and cellular immune responses. Proc Natl Acad Sci U S A, 97: 4760-4765, 2000. 31. Yamshchikov, G., Thompson, L., Ross, W. G., Galavotti, H., Aquila, W., Deacon, D., Caldwell, J., Patterson, J. W., Hunt, D. F., and Slingluff, C. L., Jr. Analysis of a natural immune response against tumor antigens in a melanoma survivor: lessons applicable to clinical trial evaluations. Clin Cancer Res, 7: 909s916s, 2001. 32. Rubio-Godoy, V., Dutoit, V., Rimoldi, D., Lienard, D., Lejeune, F., Speiser, D., Guillaume, P., Cerottini, J. C., Romero, P., and Valmori, D. Discrepancy between ELISPOT IFN-gamma secretion and binding of A2/peptide multimers to TCR reveals interclonal dissociation of CTL effector function from TCR-peptide/ MHC complexes half-life. Proc Natl Acad Sci U S A, 98: 10302-10307, 2001. 33. Yee, C., Thompson, J. A., Byrd, D., Riddell, S. R., Roche, P., Celis, E., and Greenberg, P. D. Adoptive T cell therapy using antigen-specific CD8+ T cell clones for the treatment of patients with metastatic melanoma: in vivo persistence, migration, and antitumor effect of transferred T cells. Proc Natl Acad Sci U S A, 99: 16168-16173, 2002. 34. Valmori, D., Dutoit, V., Schnuriger, V., Quiquerez, A. L., Pittet, M. J., Guillaume, P., Rubio-Godoy, V., Walker, P. R., Rimoldi, D., Lienard, D., Cerottini, J. C., Romero, P., and Dietrich, P. Y. Vaccination with a Melan-A peptide selects an oligoclonal T cell population with increased functional avidity and tumor reactivity. J Immunol, 168: 4231-4240, 2002.
42
C H A P T E R
2
Phase I clinical study with multiple peptide vaccines in combination with tetanus toxoid and GM-CSF in advanced-stage HLA-A*0201positive melanoma patients. A. Bins, H. Mallo, J. Sein, F. Vyth-Dreese, B. Nuijen, G. de Gast and J. Haanen Division of Immunology; Division of Medical Oncology; Division of Clinical Chemistry, The Netherlands Cancer Institute, Plesmanlaan 121, 1066CX Amsterdam, the Netherlands.
Abstract Successful induction of functional tumor specific T cells by peptide vaccination in animal models has resulted in many clinical trials to test this approach in advancedstage melanoma patients. In this phase I clinical trial, 11 end-stage melanoma patients were vaccinated intradermally with three peptides: MART-1 (26-35) E27L (ELAGIGILTV), tyrosinase (368-376) N375Q (YMDGTMSQV) and gp100(209-217) T210M (IMQVPFSV), admixed with Tetanus Toxoid and GM-CSF. The peptide vaccine was well tolerated at all tested doses, and led to grade 1-2 toxicity only. Although all patients did show a rise in anti-tetanus IgG titers, in only three patients peptide-specific CD8+ T cells were induced. In two cases the response was directed against MART-1(26-35) and consisted of 0.2% and 3.3% of the CD8+ population; however, in both instances these cells did not produce IFN-γ on stimulation with the unmodified peptide. The third patient mounted a small (0.1%) respons against GP100. In a fourth patient a non-functional tyrosinase specific response (0.6%) was found that was present prior to vaccination, but was not affected in size nor in function by the vaccine. None of the 11 patients responded clinically according to RECIST criteria. Although this study is a small scale phase I clinical trial, the efficacy that was observed was disappointingly low. In accordance with previously published peptide vaccination studies, these results add to the increasing evidence that peptide vaccination in itself is not potent enough as an effective melanoma immunotherapy in advanced-stage patients.
43
2
PHASE I CLINICAL STUDY WITH MULTIPLE PEPTIDE VACCINES IN COMBINATION WITH TETANUS TOXOID AND GM-CSF
Introduction Malignant melanomas express a range of tumor associated antigens (TAAs) (1) that can potentially serve as targets for T cell attack. Although tumor specific T cells can be detected in many end stage melanoma patients (2), the spontaneous occurrence of effective T cell responses, leading to tumor regression, is very rare. Therefore, the successful induction of functional tumor specific T cells in animal models using peptide vaccination (3;4), has resulted in various clinical trials to test this approach in melanoma patients. In this phase I study, we tested a peptide vaccination strategy based on a novel combination of features to increase vaccine efficacy (5-10). 1) The vaccine consisted of three separate melanocyte differentiation antigen derived peptide vaccines, in order to widen the tumor specific T cell repertoire stimulated by the vaccine (6;10) : MART -1 (26-35) E27L (ELAGIGILTV)(11), Tyrosinase (368-376) N375Q (YMDGTMSQV) (12) and GP100 (209-217) T210M (IMQVPFSV)(12). The anchor residues in these peptide sequences are modified to improve HLA-A*0201 binding (7) . 2) As it has been shown to exert immunostimulatory functions in the treatment of end stage melanoma, granulocyte-monocyte colony stimulating factor (GMCSF) was added to the vaccine (13). 3) Additionally, in order to facilitate T cell help (14) , Tetanus Toxoid was admixed as well. Tetanus Toxoid consists of formaldehyde treated, inactivated tetanus toxin and contains pan-DR epitopes (PADREs) (15) that can be presented in the context of multiple MHC class II molecules. 4) All injections were given subcutaneously (8) in adjuvant Montanide ISA-51 (16). Furthermore, each of the peptide vaccines was injected on a separate location, to avoid competition for HLA-A*0201 binding during antigen presentation by antigen presenting cells (APCs) (5). Considering the lack of correlation between disease prognosis and the presence of tumor specific cells after peptide vaccination (17;18), we assessed the effect of the vaccination by analyzing both the expansion and the functionality of the induced T cell populations. For this purpose, we performed tetramer and ELIspot analyses of blood samples collected before and after vaccination. In addition, we tested peptide specific delayed type hypersensivity (DTH) responses. Material and Methods Stage IV melanoma patients that were HLA-A*0201 positive, above 18 years of age, having progressive disease after chemotherapy with or without non-specific immunotherapy, were eligible for the study. Additional criteria were a WHO performance score of 0 or 1, adequate bone marrow, renal and liver function (WBC>3.0/nl, platelets>100/nl, creatinine clearance > 50 ml/min, bilirubin < 1.5xULN), a life expectancy of 3 months or more and signed informed consent. Excluded from the study were patients with brain metastases larger than 2 cm in diameter or a second malignant disease. Patients that were pregnant or lactating, taking systemic corticosteroids or requiring antibiotics for severe infections were also excluded, as were those with severe cardiac, respiratory or metabolic disease, or HIV seropositivity. 44
All patients received 4 vaccinations at 1 week intervals, each vaccination consisting of three separate subcutaneous injections of MART-1(26-35), tyrosinase(368-376) and gp100(209-217), mixed with 10 EI Tetanus Toxoid and 100 µg GM-CSF in Montanide ISA-51. To prepare the mixtures, the peptides were dissolved in a 0.9% NaCl water solution, and 100ug GM-CSF and 10IE Tetanus Toxoid were added, to a total of 1 ml. To this solution, 1.2 ml montanide ISA-51 was mixed by vigorous shaking, resulting in a homogeneous emulsion. The injections were given subcutaneously, in the proximal part of a limb, in an area near draining lymph nodes (DLN). Patients that had undergone a lymph node dissection in the axilla region were vaccinated near the inguinal region, and vice versa. Two subsequent vaccinations were never placed in the same limb. The peptide dose was increased from 250μg in patients 1 to 4, to 500μg in patients 5 to 8, to a maximum dose of 1000μg in patients 9 to 11. Before start of the vaccination (week 0) and 5 weeks after the start of the study, blood samples were collected for analysis. The samples were analyzed for the presence of tumor specific T cells by tetramer staining as described previously (2) . In short, thawed PBMC samples were incubated overnight at 37°C in Iscove’ s medium supplemented with 10% FCS to recover PBMCs and to eliminate apoptotic cells. After washing, the cells were incubated for 5 min in cold PBS with 0.5% BSA and 1% normal mouse serum to block Fc receptors. Two million cells per sample were incubated for 10 min with 2 µg/ml phycoerythrin labeled MHC-tetramer at 37°C and CD8pos T cells were negatively selected by staining with a large set of FITC-labeled lineage marker antibodies (CD4, CD13, CD14, CD16, CD19; Becton-Dickinson). Cells were stained with propidium iodide to be able to gate out dead cells. Samples were analyzed by flow cytometry using a FACScalibur and Cell Quest software (Becton Dickinson). Additionally, ELIspot assays were done, according to the instructions of the manufacturer (U-cytech biosciences, Utrecht, the Netherlands). Briefly, 96-well plates (Costar 3799) were incubated with coating antibody (U-cytech) overnight at 370 C., washed with phosphate buffered saline (PBS) and coated with coating buffer for 2 hours. Meanwhile, the patient PBL were thawed, washed and incubated 1 hour in Iscoves Modified Dulbecco’s Medium (IMDM) supplemented with 5% Fetal Calf Serum (FCS) at 2x106 cells/ml. Next, the coating buffer was removed from the plate, 100µl cell suspension was brought in each well, peptides were added at concentrations of 1, 0.5, 0.25 or 0.1 µg/ml and the plate was incubated overnight at 370C. After removal of the cell suspension the plate was washed with PBS and developed according to the instructions of the manufacturer. The plates were read using bioreader 4000 pro-X ELIspot reader (Bio-sys, Karben, Germany). Elispot and tetramer T cell counts higher than 3x the standard deviation of a panel of negative controls were considered positive. Furthermore, also at week 0 and week 5, an ELISA was performed to determine the Tetanus Toxoid specific antibody titer, together with a DTH test for reactivity against any of the three peptides. The DTH tests were read 48 hours after 45
2
PHASE I CLINICAL STUDY WITH MULTIPLE PEPTIDE VACCINES IN COMBINATION WITH TETANUS TOXOID AND GM-CSF
application, and an induration of more than 5 mm was scored as positive. In the first 2 of the 11 patients, TT was admixed to the peptides solutions used in the DTH tests. However, this caused considerable discomfort and was left out in the rest of the tests. At 5 and 13 weeks after the vaccination, tumor progression was analyzed by Computer Assisted Tomography (CAT), Magnetic Resonance Imaging (MRI) or Positron Emission Tomography (PET) scans, and scored according to RECIST criteria. The protocol was reviewed and approved by the Institutional Review Board of The Netherlands Cancer Institute / Antoni van Leeuwenhoek Ziekenhuis. All the patients signed for informed consent before participating in the clinical trial. Results 11 patients were included in the study (table 1). All patients had unresectable stage IV disease, and had been treated with standard chemotherapy prior to enrollment. Four patients had received prior aspecific immunotherapy with GM-CSF, IFN-α and low dose IL-2 in combination with temozolomide. At all tested doses the peptide vaccines were tolerated well, and led to grade 1-2 toxicity only. All patients developed a local redness at the site of injection, leading to some discomfort. Four patients experienced systemic side effects, in two cases leading to some difficulty in performing activities of daily living (ADL). (Table 2). On vaccination, all patients except one showed a significant increase in tetanus toxoid specific antibody titers (Figure 1). On the contrary, peptide specific cellular T cell responses induced by the vaccines were much less consistent. At 5 weeks after the start of the vaccination, 3 of the 11 patients (27%) showed a significant increase in T cells specific for one of the peptides in the vaccine. In 2 of these patients the response was present prior to vaccination and directed against MART-1(26-35) E27L (0.1% in patient I and 2.1% in patient II). In both patients, approximately a third of these cells were CD45RA negative, indicating recent antigen encounter. In patient I none of these pre-existing T cells could be detected in the ELIspot assay, but patient II showed some ELIspot reactivity prior to vaccination. The size of the tetramer positive populations in both patients increased upon vaccination with the MART-1(26-35) E27L peptide (0.2% and 3.3%), and in both patients the responses became clearly detectable in the ELIspot assay, indicating a change in the capacity to secrete IFN-γ (Fig. 2a/b). The third patient that showed an increase in tumor specific T cells upon vaccination, did not have any circulating melanoma specific T cells prior to vaccination, but mounted a small gp100(209-217) specific response (0.1%, Fig. 2c) and a MART-1(26-35) E27L specific response on vaccination (0.2%) , detectable by tetramer stain and ELIspot assay.
46
Interestingly, in a fourth patient a sizable pre-existing tyrosinase(368-376) specific response was detected by tetramer staining (0.8%) that decreased on vaccination (0.7%) and could not be detected by ELIspot neither before nor after vaccination (Fig. 2d). This suggested that these cells resided in a non-functional state that could not be reversed by the vaccination, contrary to the MART-1(26-35) E27L specific T cells in patient I and II. As an explanation for this difference we considered the fact that, in contrast to the modified MART-1(26-35) E27L epitope, the posttranslationally modified tyrosinase(368-376) N375Q epitope is naturally expressed by the tumor. Therefore, the tyrosinase(368-376) specific T cells could be more liable to tolerization by the tumor then the MART-1(26-35) E27L specific T cells. In order to assess the tumor crossreactivity of the MART-1(26-35) E27L specific T cells, we repeated the MART-1(26-35) specific ELIspot and tetramer analyses, using the unmodified MART-1(26-35) peptide (EAAGIGILTV) and cognate tetramers, produced by “exchange” technology (19). Similar pre-existing MART-1(26-35) specific T cell numbers were detected using the unmodified MART-1(26-35) tetramers (0.1% in patient I and 1.6% in patient II). The same applied for the MART-1(26-35) specific T cell counts after vaccination (0.3% in patient I and 2.1% in patient II). Hence, compared to the modified version the same pattern emerged using unmodified MART-1(26-35) tetramers. Strikingly however, using the unmodified peptide in the ELIspot assay no MART-1(26-35) specific T cell reactivity could be detected in both of the patients after vaccination (Fig. 3a/b), in analogy to the tyrosinase(368-376) response in patient IV. Furthermore, none of the four patients scored positive in the DTH tests against any of the three peptides. Only the first two patients enrolled in the study, that received skin tests with Tetanus Toxoid admixed to the peptide, scored highly positive. As mentioned previously, this component was subsequently removed. Finally, none of the detected immune responses coincided with any clinical response. Although in one patient of the lowest dosing group, the tumor evaluation at week 5 and week 13 showed stabilization of the disease, no tumor specific T cells could be detected in the blood of this patient. Though it may be that the disease stabilization was caused by the vaccination, it could well be part of the natural disease course in this patient. The median survival of the patients after vaccination was 4 month (table 3), comparable to the survival in untreated stage IV melanoma cohorts (20). Discussion The results of the phase I peptide vaccination trial in end-stage melanoma patients described in this report are in accordance with previously published peptide vaccination studies. In short, the treatment may induce vaccine specific T cell responses in a minority of these patients (approx. 10-50%), but clinical responses are rare (approx. 2.5-5%)(21).
47
2
PHASE I CLINICAL STUDY WITH MULTIPLE PEPTIDE VACCINES IN COMBINATION WITH TETANUS TOXOID AND GM-CSF
The meager clinical responses obtained in therapeutic tumor vaccination trials in general, and in this melanoma specific peptide vaccination trial in particular, are partially explained by the high tumor burden in the treated patients, which grows fast and may tolerize the tumor specific T cell repertoire (17). This, in combination with the suboptimal affinity of these inherently autoreactive cells, makes active immunization in these patients a difficult feat. The presence of vaccine induced T cells in the circulation and the occurrence of clinical responses hardly correlate (22). Our data seem to suggest that nonfunctionality of tumor specific T cells may in part explain this paradox. In summary, after the vaccination three of eleven patients had a tumor specific response of a magnitude comparable to responses seen in viral infections. One directed against Tyr (0.7%) and the other two against (unmodified) MART-1(26-35) (0.3% and 2.1%). In all these cases however, the cognate ELIspot counts were negligible. Interestingly, in the 2 patients with MART-1(26-35) specific T cells after vaccination, these cells did produce IFN-γ on stimulation with the altered MART-1(26-35) E27L peptide ligand. The high MART-1(26-35) specific precursor frequency (23) in humans may facilitate the emergence of these tumor non-cross reactive T cells upon vaccination with the modified peptide. Although in this study the T cell responses after peptide vaccination neither induced DTH responses nor tumor regression, it is clearly established that T cells can induce tumor regression, when present in sufficiently high numbers and in the right activation status (24). From this perspective, the induction of tumor specific T cells using current peptide vaccination strategies simply falls short. Strategies to improve this success rate include co-injection of immunostimulatory adjuvants, such as TLR agonists and cytokines (25-27) or vaccination with peptideloaded Dendritic Cells (DC) (28;29). Another interesting approach is the use of peptides that are not constricted to a minimal epitope, but contain a larger part of the antigen sequence, spanning the entire protein as a pool (30). In addition, recent developments in DNA vaccination hold a promise (31). However, the gap between T cell numbers induced by any form of vaccination, and those necessary for tumor regression in end-stage melanoma patients (24) is large. Although abovementioned improvements could bridge this gap, it seems worthwhile to explore other means of cancer immunotherapy. The transfer of autologous tumor infiltrating lymphocytes (TILs), propagated ex vivo, is the most effective current experimental immunotherapy in melanoma patients (23). Moreover, the adoptive transfer of autologous PBL transduced with TCR genes, may lead to an equally potent, yet more versatile treatment. In view of the significantly greater success of adoptive therapies, it seems wise to focus on adoptive therapy for the treatment of melanoma. Possibly, the mechanisms revealed in such studies could lead to the design of more effective vaccination approaches in the distant future.
48
Table1, population characteristics Demography Total number
11
Caucasian
11
Median age
(range)
Female / Male
45
(30-63)
3/8
2
Prior treatment No prior treatment
0
Prior radiation
3
Prior chemotherapy
10
Prior non-specific immunotherapy
4
Site of primary tumor Skin
10
Eye
1
Site of metastasis Lung
4
Skin
4
Bone
2
Liver
1
Mediastinum
1
Table 2, Toxicity according to CTC criteria Local Grade 1: itching; erythema
6
Grade 2: Pain or swelling
11
Systemic Grade 1: Mild fatigue over baseline
5
Grade 1: Fever 38.0 - 39.0 degrees C
2
Grade 2: Moderate or causing difficulty performing ADL
2
49
PHASE I CLINICAL STUDY WITH MULTIPLE PEPTIDE VACCINES IN COMBINATION WITH TETANUS TOXOID AND GM-CSF
Table 3, Median survival and blood levels of S100 and LDH before and 2 month after vaccination Median
Range
Survival
4 month
1 - >36+
S100 (µg/l)
before: 0.3 after 2.1
before: 0.1 - 0.7 after: 0.1 - 9.3
LDH (U/l)
before: 374 after 461
before:104 -1819 after: 100 - 2540
Survival, S100 and LDH levels in bloodsamples taken before and 5 weeks after the start of the vaccination.
����������������������������
����
���
��
�
������
������
Figure 1 Tetanus toxin specific IgG titers of the enrolled patients before and after vaccination. One of the eleven patients could not be tested at week 5. 50
A
Before vaccination
1,00 µg/ml
2
0.50 µg/ml
1
0.25 µg/ml
1
0.10 µg/ml
0
1,00 µg/ml
20
0.50 µg/ml
9
0.25 µg/ml
12
0.10 µg/ml
14
1,00 µg/ml
25
0.50 µg/ml
22
0.25 µg/ml
9
0.10 µg/ml
18
1,00 µg/ml
22
0.50 µg/ml
22
0.25 µg/ml
24
0.10 µg/ml
21
0.1%
HLA-A*0201 / MART-1(26-35) E27L tetramer - PE
!" # $
After vaccination
2
0.2%
!" # $ % %
CD4, 13, 14, 16, 19 - fitc
B
Before vaccination
2.1%
HLA-A*0201 / MART-1(26-35) E27L tetramer - PE
!" # $ % % %
After vaccination
3.3%
!" # $ % %
CD4, 13, 14, 16, 19 - fitc
Figure 2
Analysis of vaccine specific T cell responses before and after vaccination. (a) MART-1 (26-35) E27L specific Tetramer staining (right) and Elispot assay (left), before (above the black line) and 5 weeks after start of vaccination (below the black line) in patient I. (b) The same display for patient II. 51
PHASE I CLINICAL STUDY WITH MULTIPLE PEPTIDE VACCINES IN COMBINATION WITH TETANUS TOXOID AND GM-CSF
C
Before vaccination
0.0%
1,00 µg/ml
2
0.50 µg/ml
1
0.25 µg/ml
1
0.10 µg/ml
2
1,00 µg/ml
6
0.50 µg/ml
6
0.25 µg/ml
3
0.10 µg/ml
10
HLA-A*0201 / gp100(209-217) T210M tetramer - PE
!
" "
After vaccination
0.1%
!
"
CD4, 13, 14, 16, 19 - fitc
D
Before vaccination
0.7%
1,00 µg/ml
1
0.50 µg/ml
0
0.25 µg/ml
0
0.10 µg/ml
0
1,00 µg/ml
0
0.50 µg/ml
1
0.25 µg/ml
0
0.10 µg/ml
1
tetramer - PE
HLA-A*0201 / tyrosinase(368-376) N375Q
!
0.6%
After vaccination
! "" " " ""
CD4, 13, 14, 16, 19 - fitc
(c) Gp100(209-217) specific T cell response in patient III and (d) tyrosinase(368-376) specific responses in patient IV. The percentage in the upper left corner of the scatterplot represents the fraction of tetramer positive cells in the CD8 positive lymphocyte population. In the ELIspot assay 4 peptide concentrations were tested in triplicate or duplicate. The average number of spots per well for each tested pepide concentration is depicted at the right side of the figure.
52
0.3%
HLA-A*0201 / MART-1(26-35) (unmod.) tetramer - PE
After vaccination Patient I
1,00 µg/ml
1
0.50 µg/ml
0
0.25 µg/ml
1
0.10 µg/ml
1
1,00 µg/ml
4
0.50 µg/ml
4
0.25 µg/ml
2
0.10 µg/ml
4
!" ! # $ "
## ## $ # # ##% ##
& # ##%
'' #&%#
$ ##%
$ # # # # # ##% ##
##%
$%
& ##%
% & # # # # '" ##% ##
$' ##%'
$& %
$ ##%#
""
% $ $ # ##%'# ##'
#" ' ##%#
" '%#
"& ##%
# % $'
$ " $ '" ##%'# ##
& # ##%
# #%
# ##%
After vaccination patient II
2.1%
2
!" # $ $%
&& && #" ) $ % % % & & %%
(( % &
)
!
! !" ! " & & &&' &&
% (% &&'
$% %'
$( &&'
& & &&' &&
% &&')$
& )'
&&'
& && )% &&' &&
(( &&'))
$ '
( &&'
& &&')& &&)
% &&'&
( &'&
% &&'
) )& )% &&')& &&
% (% &&'
(( $'&
() &&'
CD4, 13, 14, 16, 19 - fitc
Figure 3
MART-1 (26-35) specific responses after vaccination in patient I and II, measured using the unmodified MART-1(26-35) (EAAGIGILTV) peptide for ELIspot and tetramer analysis. The percentage in the upper left corner of the scatterplot represents the fraction of tetramer positive cells in the CD8 positive lymphocyte population. In the ELIspot assay 4 peptide concentrations were tested in triplicate. The average number of spots per well for each tested pepide concentration is depicted at the right side of the figure.
53
PHASE I CLINICAL STUDY WITH MULTIPLE PEPTIDE VACCINES IN COMBINATION WITH TETANUS TOXOID AND GM-CSF
1. Nagorsen D, Scheibenbogen C, Marincola FM, et al. Natural T cell immunity against cancer. Clin Cancer Res. 2003;9:4296-4303. 2. van Oijen M, Bins A, Elias S, et al. On the role of melanoma-specific CD8+ T-cell immunity in disease progression of advanced-stage melanoma patients. Clin Cancer Res. 2004;10:4754-4760. 3. Bloom MB, Perry-Lalley D, Robbins PF, et al. Identification of tyrosinaserelated protein 2 as a tumor rejection antigen for the B16 melanoma. J Exp Med. 1997;185:453-459. 4. Mandelboim O, Vadai E, Fridkin M, et al. Regression of established murine carcinoma metastases following vaccination with tumour-associated antigen peptides. Nat Med. 1995;1:1179-1183. 5. Kedl RM, Rees WA, Hildeman DA, et al. T cells compete for access to antigen-bearing antigen-presenting cells. J Exp Med. 2000;192:1105-1113. 6. Lee P, Wang F, Kuniyoshi J, et al. Effects of interleukin-12 on the immune response to a multipeptide vaccine for resected metastatic melanoma. J Clin Oncol. 2001;19:3836-3847. 7. Parkhurst MR, Salgaller ML, Southwood S, et al. Improved induction of melanoma-reactive CTL with peptides from the melanoma antigen gp100 modified at HLA-A*0201-binding residues. J Immunol. 1996;157:2539-2548. 8. Puri N, Weyand EH, Abdel-Rahman SM, et al. An investigation of the intradermal route as an effective means of immunization for microparticulate vaccine delivery systems. Vaccine. 2000;18:2600-2612. 9. Slingluff CL, Jr., Yamshchikov G, Neese P, et al. Phase I trial of a melanoma vaccine with gp100(280-288) peptide and tetanus helper peptide in adjuvant: immunologic and clinical outcomes. Clin Cancer Res. 2001;7:3012-3024. 10. Weber J, Sondak VK, Scotland R, et al. Granulocyte-macrophage-colonystimulating factor added to a multipeptide vaccine for resected Stage II melanoma. Cancer. 2003;97:186-200. 11. Schneider J, Brichard V, Boon T, et al. Overlapping peptides of melanocyte differentiation antigen Melan-A/MART-1 recognized by autologous cytolytic T lymphocytes in association with HLA-B45.1 and HLA-A2.1. Int J Cancer. 1998;75:451-458. 12. Kawakami Y, Eliyahu S, Jennings C, et al. Recognition of multiple epitopes in the human melanoma antigen gp100 by tumor-infiltrating T lymphocytes associated with in vivo tumor regression. J Immunol. 1995;154:3961-3968. 13. Dranoff G. GM-CSF-secreting 2003;%19;22:3188-3192.
54
melanoma
vaccines.
Oncogene.
14. Schaed SG, Klimek VM, Panageas KS, et al. T-cell responses against tyrosinase 368-376(370D) peptide in HLA*A0201+ melanoma patients: randomized trial comparing incomplete Freund’s adjuvant, granulocyte macrophage colony-stimulating factor, and QS-21 as immunological adjuvants. Clin Cancer Res. 2002;8:967-972. 15. Alexander J, Sidney J, Southwood S, et al. Development of high potency universal DR-restricted helper epitopes by modification of high affinity DRblocking peptides. Immunity. 1994;1:751-761. 16. Yamshchikov GV, Barnd DL, Eastham S, et al. Evaluation of peptide vaccine immunogenicity in draining lymph nodes and peripheral blood of melanoma patients. Int J Cancer. 2001;92:703-711. 17. Lee PP, Yee C, Savage PA, et al. Characterization of circulating T cells specific for tumor-associated antigens in melanoma patients. Nat Med. 1999;5: 677-685. 18. Walker EB, Haley D, Miller W, et al. gp100(209-2M) peptide immunization of human lymphocyte antigen-A2+ stage I-III melanoma patients induces significant increase in antigen-specific effector and long-term memory CD8+ T cells. Clin Cancer Res. 2004;10:668-680. 19. Toebes M, Coccoris M, Bins A, et al. Design and use of conditional MHC class I ligands. Nat Med. 2006;12:246-251. 20. Wong SL, Coit DG. Role of surgery in patients with stage IV melanoma. Curr Opin Oncol. 2004;16:155-160. 21. Rosenberg SA, Yang JC, Restifo NP. Cancer immunotherapy: moving beyond current vaccines. Nat Med. 2004;10:909-915. 22. Rosenberg SA, Yang JC, Schwartzentruber DJ, et al. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. Nat Med. 1998;4:321-327. 23. Dudley ME, Wunderlich JR, Robbins PF, et al. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science. 2002;298:850-854. 24. Rosenberg SA, Dudley ME. Cancer regression in patients with metastatic melanoma after the transfer of autologous antitumor lymphocytes. Proc Natl Acad Sci U S A. 2004;101 Suppl 2:14639-14645. 25. Disis ML, Rinn K, Knutson KL, et al. Flt3 ligand as a vaccine adjuvant in association with HER-2/neu peptide-based vaccines in patients with HER-2/neuoverexpressing cancers. Blood. 2002;99:2845-2850.
55
2
PHASE I CLINICAL STUDY WITH MULTIPLE PEPTIDE VACCINES IN COMBINATION WITH TETANUS TOXOID AND GM-CSF
26. Fong L, Hou Y, Rivas A, et al. Altered peptide ligand vaccination with Flt3 ligand expanded dendritic cells for tumor immunotherapy. Proc Natl Acad Sci U S A. 2001;98:8809-8814. 27. Lee P, Wang F, Kuniyoshi J, et al. Effects of interleukin-12 on the immune response to a multipeptide vaccine for resected metastatic melanoma. J Clin Oncol. 2001;19:3836-3847. 28. Kono K, Takahashi A, Sugai H, et al. Dendritic cells pulsed with HER-2/ neu-derived peptides can induce specific T-cell responses in patients with gastric cancer. Clin Cancer Res. 2002;8:3394-3400. 29. Brossart P, Wirths S, Stuhler G, et al. Induction of cytotoxic T-lymphocyte responses in vivo after vaccinations with peptide-pulsed dendritic cells. Blood. 2000;96:3102-3108. 30. Zwaveling S, Ferreira Mota SC, Nouta J, et al. Established human papillomavirus type 16-expressing tumors are effectively eradicated following vaccination with long peptides. J Immunol. 2002;169:350-358. 31. Bins AD, Jorritsma A, Wolkers MC, et al. A rapid and potent DNA vaccination strategy defined by in vivo monitoring of antigen expression. Nat Med. 2005;11:899-904.
56
C H A P T E R
3
A rapid and potent DNA vaccination strategy defined by in vivo monitoring of antigen expression Adriaan D Bins1, Annelies Jorritsma1, Monika C Wolkers1, Chien-Fu Hung2, T-C Wu2, Ton N M Schumacher1 & John B A G Haanen1
Department of Immunology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands. 1
Department of Pathology, The Johns Hopkins Medical Institution, The Johns Hopkins University School of Medicine, Ross Research Building, Room 512, 720 Rutland Avenue, Baltimore, Maryland 21205, USA. 2
Correspondence should be addressed to John B A G Haanen
[email protected]
Abstract Induction of immunity after DNA vaccination is generally considered a slow process. Here we show that DNA delivery to the skin results in a highly transient pulse of antigen expression. Based on this information, we developed a new rapid and potent intradermal DNA vaccination method. By short-interval intradermal DNA delivery, robust T-cell responses, of a magnitude sufficient to reject established subcutaneous tumors, are generated within 12 d. Moreover, this vaccination strategy confers protecting humoral immunity against influenza A infection within 2 weeks after the start of vaccination. The strength and speed of this newly developed strategy will be beneficial in situations in which immunity is required in the shortest possible time.
57
3
A Rapid And Potent DNA Vaccination Strategy Defined By In Vivo Monitoring Of Antigen Expression
Introduction Over the past decade, DNA vaccines have emerged as a promising approach for the induction of immune responses. Generally, current DNA vaccination strategies use a regimen of multiple intramuscular or intradermal administrations, at intervals of 2 weeks or more, and require at least 1 month to achieve immunity1, 2, 3, 4, 5, 6. The slow development of T-cell responses after DNA vaccination contrasts sharply with immune responses induced by a physiological antigen encounter, such as viral infection, that build up rapidly and often peak within 10 d7. The reason for this slow development of immune responses has remained unclear. It has been postulated that DNA vaccination leads to transfection of few cells and to the expression of relatively small amounts of antigen8, 9, therefore requiring a time-consuming prime-boost strategy. But direct evidence for the proposed low antigen expression is scarce and its role in immune induction has not been addressed. To examine whether the slow induction of immune responses could be causally related to the kinetics of antigen expression induced by DNA vaccines, we analyzed in vivo levels of antigen expression after intramuscular and intradermal DNA delivery, and correlated these with the induction of T-cell immunity. Based on these data, we developed a short-interval DNA vaccination method that generates functional T- and B-cell responses within a minimal time frame. Results Intradermal DNA vaccination by skin tattooing To be able to administer DNA to the skin in a controlled manner, over a large surface and only in the upper nonvascularized layers, we made use of a simple tattoo device10 for intradermal DNA delivery. Histochemical analysis of ‘DNA-tattooed’ skin showed that transfected cells were distributed over the upper layers of the dermis and the epidermis. (Fig. 1a,b). To test the immunogenicity of this method, we vaccinated two groups of mice with a DNA vaccine encoding the influenza A nucleoprotein epitope (amino acids 366−374; NP366) fused to the carboxy terminus of a tetanus toxin fragment (d1TTFC-NP)11, either by intramuscular injection or by skin tattoo, following the conventional DNA vaccination regimen of three administrations at 2-week intervals. This comparison shows that the intradermal delivery of a DNA vaccine using a tattoo device is an efficient strategy for the induction of T-cell immunity (Fig. 1c). Imaging of in vivo antigen expression To monitor antigen expression upon DNA vaccination, we constructed a plasmid encoding the NP366 epitope fused to the carboxy terminus of firefly luciferase (LucNP). The Luc-NP vaccine elicits potent NP366-specific T-cell responses (Fig. 1d). After tattoo or intramuscular administration of the Luc-NP vaccine, we used a lightsensitive camera to determine longitudinal in vivo antigen expression. Notably, a single intramuscular injection of DNA resulted in high levels of luciferase activity, peaking after 1 week and remaining detectable up to 1 month after injection. The
58
antigen expression kinetics induced by DNA tattooing were markedly different. First, peak values of antigen expression were at least ten times lower. Second, luciferase activity in the skin peaked after 6 h (data not shown) and disappeared over the next 4 d (Fig. 1e). We subsequently determined the capacity of both methods to present the vaccineencoded NP366 epitope to naive, lymph node−resident T cells. For this purpose, 5 million carboxy-fluorescein diacetate succinimidyl ester (CSFE)-labeled splenocytes from F5 TCR transgenic mice12 were injected into mice at different time points after a single intramuscular DNA injection or DNA intradermal tattoo. Luciferase activity was measured on the day of cell transfer, and 3 d later the animals were killed, lymphoid organs excised and the CFSE signal of the F5 cells assessed by flow cytometry. These experiments showed that although antigen production is markedly greater upon intramuscular delivery, presentation of this antigen to naive T cells is markedly more efficient upon intradermal DNA delivery. Specifically, in spite of the high antigen expression in the muscle after intramuscular injection, the fraction of F5 cells in the draining lymph node that had undergone proliferation was marginal in all of the three tested time windows (days 1−4, days 8−11 and days 22−25; Fig. 2a−c). In marked contrast, the priming of F5 T cells after tattooing was very efficient (Fig. 2a). Independent of the number of F5 T cells that were infused (5 million or 1 million; latter not shown), over 80% of the draining lymph node−resident F5 cells had lost CFSE signal on day 4, despite the low antigen level expressed in skin (Fig. 2d). At later time points, NP366-specific T-cell activation became substantially less (Fig. 2b,c), consistent with the loss of luciferase signal observed by in vivo imaging. Similar results were obtained with spleen-resident F5 cells. No F5 T-cell division could be detected in nondraining lymph nodes (data not shown). Induction of T-cell immunity by short-interval DNA tattooing Based on the transient antigen expression and presentation induced by DNA tattooing as compared to intramuscular injection, we speculated that in the case of dermal DNA vaccination, shortening of the conventional 2-week interval could result in faster T-cell induction. To test this hypothesis, we reduced the interval between consecutive vaccinations to 3 d, resulting in a day 0, 3 and 6 regimen. Notably, this compact vaccination protocol induced profound T-cell responses of 4−8% of total CD8+ T cells within 12 d after the start of vaccination (Fig. 3a). This regimen of short-interval vaccination does not lead to detectable T-cell responses when the vaccine is administered intramuscularly, consistent with the observation that the duration of antigen expression is not a limiting factor in intramuscular DNA vaccination. To test the value of this new vaccination regimen for other methods of intradermal DNA delivery, we compared T-cell induction by short-interval gene-gun and shortinterval tattoo vaccination, using the previously described Hsp70-HPV-E7 DNA vaccine13. Using the short-interval regimen, both methods generated strong T-cell
59
3
A Rapid And Potent DNA Vaccination Strategy Defined By In Vivo Monitoring Of Antigen Expression
responses within 12 d against the immunodominant human papillomavirus (HPV) E749−57 cytotoxic T lymphocyte (CTL) epitope (E749) (Fig. 3b). Comparison of T-cell responses induced by short-interval DNA tattooing with other previously established vaccination strategies showed that short-interval DNA tattooing yields markedly higher T-cell responses (tenfold or more) than peptide−incomplete Freund adjuvant−CD40-specific monoclonal antibody14 or peptide-synthetic CpG oligodeoxynucleotide15 vaccines (Fig. 3c). Furthermore, T-cell induction is efficient for both internal epitopes and those fused at the carboxy terminus (Supplementary Fig. 1 online). The resulting T cells are capable of direct effector function, as indicated by antigen-induced interferon (IFN)-γ production (Supplementary Fig. 2 online). Repetitive application of DNA may conceivably boost T-cell responses by enhancing the absolute amount of antigen expression, or by prolonging the duration of antigen expression. To address this issue, we compared antigen expression levels and T-cell responses induced by three consecutive 4 s tattoos with those induced by one single 16 s tattoo. The cumulative antigen produced by one 16 s application exceeds that of three 4 s applications (Fig. 3d). Whereas three consecutive 4 s applications induce a T-cell response that is readily detectable, a single 16 s application is essentially without effect (Fig. 3e). To further assess whether the observed requirement for repetition reflects a need for continued antigen presence, or is relat ed to the prolongation of nonspecific inflammatory signals, groups of mice were vaccinated at day 0, 3 and 6 with different combinations of a mock vaccine and the Luc-NP vaccine. Replacing either the first or the last Luc-NP vaccine with a mock vaccine lead to a significant drop in the size of the ensuing T-cell response (Student t-test, P = 0.002 and P = 0.003, respectively; Fig. 3f). Together these results show that for the induction of primary T-cell responses by intradermal DNA vaccination, prolonged antigen expression is crucial Functional immunity induced by short-interval DNA tattooing To assess the ability of short-interval tattoo DNA vaccination to induce therapeutic amounts of tumor-specific T cells, we used the transplantable HPV E6/E7transformed TC-1 tumor cell model16. Three days after subcutaneous injection of 105 TC-1 cells, we vaccinated B6 mice using the short-interval tattoo method with a plasmid encoding E749 attached to the carboxy terminus of green fluorescent protein (GFP-E7; Fig. 4a). Control mice were either given an intramuscular injection in the same regimen with the same plasmid (Fig. 4b), or were tattooed with a control plasmid encoding green fluorescent protein (GFP) only (Fig. 4c). In both control groups, no T-cell responses could be detected after vaccination. In contrast, mice that had been tattooed with GFP-E7 mounted a sizeable T-cell response, up to 15% of the circulating CD8+ T-cell pool. Notably, the onset of E749-specific T-cell responses coincided with rejection of established subcutaneous tumors, whereas in both control groups tumors grew out rapidly (Fig. 4d). Furthermore, median survival of GFP-E7−tattooed animals was 50 d compared to 17 d in control animals (Fig. 4e). 60
To assess the value of short-interval DNA vaccination in providing antibodymediated protection against acute infections, we evaluated its ability to confer protection against influenza A virus infection. Given that protection against reinfection with homotypic influenza A strains is primarily mediated by antibodies17, 18, 19, 20 , we generated a plasmid that encodes the gene for hemagglutinin of influenza A/HK/2/68, a protein that forms a major target for neutralizing antibodies. Using this DNA vaccine, mice were vaccinated by short-interval DNA tattooing, whereas control groups were either tattooed with a GFP-encoding plasmid, or given intramuscular injection with the hemagglutinin-encoding DNA vaccine. Two weeks after vaccination, we intranasally infected mice with a sublethal dose of influenza A, and determined virus-induced morbidity by measuring body weight loss1. In the week after infection, mice that had previously been exposed to influenza A virus showed a minimal weight loss (1% at day 4 after infection). In contrast, mice given tattoo vaccination with the control plasmid or given an intramuscular injection of the hemagglutinin construct showed a sizable (10% and 14%, respectively) drop in body weight. Notably, mice that had received a short-interval hemagglutinin tattoo were largely protecte d from influenza A−induced morbidity (maximal weight loss of 4%; Student t-test, P = 0.039 versus control vaccine; Fig. 5a). The protection correlated with the induction of neutralizing antibodies (Fig. 5b). Furthermore, intradermal DNA tattooing also conferred long-term (4 months) protection (Supplementary Fig. 3 online). Collectively, these experiments indicate that shortinterval intradermal DNA vaccination leads to the rapid and sustained development of both T- and B-cell responses and that such responses can mediate the regression of established tumors and can prevent virus-induced morbidity. Discussion In this study we established a short-interval regimen that induces T- and B-cell immunity within 12 d. Analysis of antigen expression levels showed that after intradermal tattoo, 1/10 to 1/100 of the amount of antigen is produced than after intramuscular injection, and that compared to intramuscular administration, tattooinduced antigen production occurs over a limited timespan. In spite of this, the presentation of the vaccine-encoded epitope is markedly better upon intradermal tattoo vaccination. The efficiency with which dermally expressed antigens are presented to T cells probably results from the high numbers of antigen-presenting cells (APCs) present in this tissue. Also, the infliction of thousands of perforations could conceivably serve as a potent adjuvant. Contrary to murine skin dendritic cells (DCs), human skin DCs do not express Toll-like receptor 9 (TLR9), and this may partly explain why DNA vaccines have performed poorly in human trials as compared to mouse model systems21, 22. Encouragingly, a comparison of the immunogenicity of tattoo DNA vaccination in wild-type and Tlr9-/- mice showed that the absence of TLR9 does not measurably affect the immunogenicity of this method (Supplementary Fig. 4 online).
61
3
A Rapid And Potent DNA Vaccination Strategy Defined By In Vivo Monitoring Of Antigen Expression
The short-interval repetitive schedule developed here leads to rapid induction of T-cell responses upon either intradermal tattoo or gene-gun vaccination, and may well work for all intradermal vaccination techniques. Because of its simplicity and because the required equipment is orders of magnitude less expensive, DNA tattooing may become the preferred method, in particular in countries with a less developed healthcare system. In the past years, outbreaks of Ebola virus, severe acute respiratory syndrome and influenza A have been causes for concern. To contain such outbreaks conventional DNA vaccination strategies appear too slow at inducing immunity23, 24. The vaccination regimen described here retains the rapid production and the safety of DNA vaccines, but gains the speed of immune induction that is the characteristic of physiological antigen encounters. Acknowledgements The authors would like to acknowledge Gideon Schory for supplying tattoo accessories. This research was funded by the Dutch Cancer Society, project grant 2001-2417. The authors declare no competing financial interests. Methods Animals. We obtained C57BL/6 mice and Balb/c mice from the experimental animal department of The Netherlands Cancer Institute. The Tlr9-/- mice were a gift of H. Wagner ((with permission of S. Akira) Institute for Microbiology, Immunology and Hygiene, University of Munich, Germany). We purchased from the US National Cancer Institute the C57BL/6 mice used for the experiments involving a gene gun and kept them in the oncology animal facility of the Johns Hopkins Hospital. We performed all animal procedures according to approved protocols and in accordance with recommendations for the proper use and care of laboratory animals. All animal experiments were approved by the Dutch Animal Research Committee. DNA vaccines. We generated DNA vaccines by the introduction of fusion genes in pcDNA3.1. The GFP-E7 fusion gene has been described previously25, all other constructs were generated following a similar design. We included the four naturally flanking amino acid residues of NP366−374 (amino acids GVQI) as a linker at the amino terminus of each epitope in all constructs. We constructed the d1TTFC-NP DNA vaccine with optimized codon usage, in a template-free PCR using overlapping oligonucleotides of 100 bp, spanning the entire first domain of the TTFC gene and the preceding p2 epitope (amino acids 831−1,315 26). We generated the Luc-NP vaccine, encoding influenza A NP366−374 epitope fused to the carboxy terminus of firefly luciferase, by genetic linkage of the NP366−374 fragment to the carboxy terminus of the gene encoding full-length firefly luciferase. The gene encoding influenza A/HK/2/68
62
hemagglutinin was obtained by RT-PCR of RNA isolated from virus particles, and inserted in pcDNA3.1. to generate the hemagglutinin DNA vaccine. We purchased the construct used for histochemical detection of transfected cells (pVAX:LacZ) from Invitrogen. Replacement of the lacZ gene with chicken ovalbumin generated the construct encoding whole ovalbumin. The generation of the sigE7hsp DNA vaccine was described previously13. Sequences were confirmed by sequence analysis. All DNA batches were purified using EndoFree Plasmid kit (Qiagen). DNA vaccination. For intramuscular DNA vaccination, we shaved the hind leg of the mouse and injected it with 100 μg of DNA in 50 μl HBSS (Life Technologies). For intradermal DNA vaccination, we shaved the left hind leg of the mouse, applied a droplet of 20 μg DNA in 10 μl HBSS to the skin and used a sterile disposable 11-needle bar (Radical Clean Magnum11, Eurl Toupera) mounted on a rotary tattoo device (Cold skin, B&A Trading) to apply the vaccine. Needle depth was adjusted to 0.5 mm, and the needle bar oscillated at 100 Hz. DNA vaccines were applied to the skin by a 16 s tattoo. Gene-gun particle-mediated DNA vaccination was performed using a helium-driven gene gun (Bio-Rad) according to the manufacturer’s protocol. At various time points after immunization, we drew approximately 20 μl of peripheral blood for analysis of T-cell responses. Detection of NP- and HPV-specific T cells in peripheral blood. Peripheral blood lymphocytes were stained with phycoerythrin-conjugated antibody to CD8β (BD Pharmingen) plus allophycocyanin-conjugated H-2Db/ NP366−374-tetramers or allophycocyanin-conjugated H-2Db/E749−47 tetramers, at 20° C. for 15 min in FACS buffer (PBS, 0.5% BSA and 0.02% sodium azide). We washed cells three times in PBA and analyzed them using flow cytometry. Live cells were selected based on 7-AAD exclusion. We performed IFN-γ assays using the BD Cytofix/Cytoperm kit (BD Biosciences) according to the manufacturer’s protocol. Peripheral blood lymphocytes were stimulated for 4 h at a 100 nM peptide concentration. Intravital imaging. We anesthetized mice with isofluorane (Abbott Laboratories). We intraperitoneally injected an aqueous solution of the substrate luciferin (150 mg/kg, Xenogen) and 18 min later the luminescence produced by active luciferase was acquired during 30 s in an IVIS system100 CCD camera (Xenogen). Signal intensity was quantified as the sum of all detected light within the region of interest, after subtraction of background luminescence. Influenza A infection and detection of influenza A−specific antibodies. Purified influenza A/HK/2/68 virus was provided by G. Rimmelzwaan (Department of Virology, Erasmus University). We infected mice intranasally with 20 hemagglutinating units (HAU) of virus in PBS, under general anesthesia. Influenza
63
3
A Rapid And Potent DNA Vaccination Strategy Defined By In Vivo Monitoring Of Antigen Expression
A/HK/2/68-specific antibodies were detected in a hemagglutination inhibition (HAI) assay. Serum samples (50 μl) were serially 1:2 diluted with PBS (1% BSA) in roundbottomed polystyrene microtiter plates, in duplicate, and 50 μl of A/HK/2/68 influenza virus corresponding to 4 HAU was added to each well. After 45 min incubation at room temperature, we added 50 μl of 0.5% chicken red blood cells to each well. After 1 h, we determined the maximal serum dilution that fully inhibited the hemagglutination caused by the virus in the duplicate wells. The HAI titer was expressed as the reciprocal of this serum dilution.
B
5
0 tat. i.m. d1TTFC-NP
C
D
20
100000000 Expression level (rlu)
10
% NP spec. CD8+ T cells
% NP spec. CD8+ T cells
A
15 10 5 0
tat. i.m. Luc-NP
E
10000000 1000000 100000 10000 1000 100 0
2
4
6
8
Days after DNA administration
Figure 1. DNA tattooing results in transfection of epidermal and dermal cells and leads to efficient T-cell induction. (a) The abdominal skin of a mouse was tattooed with a β-galactosidase (lacZ)-encoding construct. Six hours after treatment, transfected cells were shown in a skin biopsy, using the X-gal substrate to generate a blue precipitate in cells expressing the transgene (arrows). (b) The abdominal skin of a mouse tattooed with empty vector and processed as in a. (c) NP366-specific CD8+ T-cell responses induced by tattoo (tat., circles, n = 5) or intramuscular (i.m., triangles, n = 5) DNA vaccination. The d1TTFC-NP DNA vaccine was administered three times at 2-week intervals. T-cell responses were measured 7 d after the last DNA administration, by staining peripheral blood lymphocytes with H-2Db/ NP366−374 tetramers. Horizontal bars depict averages. (d) NP366-specific T-cell responses induced by tattoo DNA vaccination (circles, n = 5) and i.m. DNA vaccination (triangles, n = 5) with the Luc-NP DNA vaccine. (e) Kinetics of Luc-NP antigen expression after a single intradermal DNA tattoo (circle) and after a single intramuscular DNA injection (triangle). 64
A
B
3
C
D
Figure 2. Analysis of antigen production and presentation after intradermal and intramuscular DNA vaccination. (a−c) Cohorts of mice (n = 7) were vaccinated once with the Luc-NP vaccine, either by intradermal DNA tattooing (filled circles) or by intramuscular DNA injection (filled triangles). Control groups (n = 3) received a plasmid encoding luciferase only (intradermal, open circles; intramuscular, open triangles). CFSElabeled naive F5 T cells were transferred on day 1 (a), day 8 (b) or day 22 (c) after DNA administration, and were recovered from the draining lymph nodes 3 d after transfer. Proliferation was assessed by analysis of CFSE loss of H-2Db/NP366−374 tetramer−positive CD8+ cells. (d) Representative draining lymph node sample from the tattooed cohort on day 4.
65
A Rapid And Potent DNA Vaccination Strategy Defined By In Vivo Monitoring Of Antigen Expression
0 5
A
10
15
20
Days after start of vaccination
D
100000 10000 1000 100 16
8
4
1
Tattoo duration (seconds)
5
0 5
E
10
15
Days after start vaccination
B % NP spec. CD8+ T cells
Expression level (rlu)
1000000
% E7 spec. CD8+ T cells
% NP spec. CD8+ T cells
5
% NP spec. CD8+ T cells
10
10
C
10
5
0 5 10 15 20 Days after start of vaccination
10
5
0 5
10
15
20
Days after start of vaccination
F
Figure 3. Features of tattoo DNA vaccination. (a) NP366-specific T-cell responses in cohorts of mice (n = 5) upon vaccination with d1TTFC-NP on day 0, 3 and 6 either by tattoo (filled circles) or intramuscular injection (open triangles). Control mice (open circles) were tattooed with a TTFC mock vaccine. NP366-specific T-cell responses were determined at indicated time points by tetramer staining of peripheral blood lymphocytes (n = 5). (b) The E749-specific T-cell response after intradermal application of the sigE7hsp DNA vaccine at day 0, 3 and 6 in cohorts of mice (n = 6), either by DNA tattooing on the left leg (filled circles) or by gene gun on both flanks of the abdomen (filled triangles). Control groups of mice (n = 3) were vaccinated with empty vector (open triangles and open circles). (c) NP366-specific T-cell responses in cohorts of mice (n = 5) upon subcutaneous vaccination with the NP366 peptide in incomplete Freund adjuvant at day 0 (open triangles), subcutaneous injection of NP366 peptide with CpG in PBS at day 0 (open circles) or at day 0, 3 and 6 (open squares), NP366 peptide with CpG in PBS tattoo vaccination at day 0, 3 and 6 (open diamonds) and DNA tattoo vaccination with d1TTFC-NP on day 0, 3 and 6 (filled circles). (d) Antigen expression level after single tattoos of 16, 8, 4 and 1 s, as determined by a light-sensitive camera 1 d after tattooing with the Luc-NP vaccine. (e) NP366specific T-cell responses (n = 5) induced by 16 s (open circles), 8 s (open triangles) and 4 s (open diamonds) tattoos on day 0, 3 and 6, or a single 16 s tattoo on day 0 (open squares). (f) In mice vaccinated with the Luc-NP DNA vaccine at day 0, 3 and 6, the replacement of Luc-NP with a mock DNA vaccine on day 0 (open squares; C-2x) or day 6 (open diamonds; 2x-C) reduces T-cell responses from levels induced by three consecutive tattoos (open circles; 3x), to levels induced by two consecutive tattoos (open triangles; 2x). 66
% E7 spec. CD8+ T cells
% E7 spec. CD8+ T cells
15
10
5
0
5
% E7 spec. CD8+ T cells
A
Tumor diameter (mm)
15
Days after start of vaccination
10
5
0
5
B
10
15
Days after start of vaccination
15
10
3
5
0
5 10 15 Days after start of vaccination
C 20 15 10 5 0
5
E
10
15
10
15
Days after start of vaccination
F
Figure 4. Tattoo DNA vaccination leads to rapid induction of functional tumorspecific T cells. (a−c) Mice were injected subcutaneously with 105 TC-1 cells. Three days later, mice were vaccinated three times at 3-d intervals either by GFP-E7 tattooing (open circles; a), by intramuscular injection (open triangles; b), or by tattooing with a mock vaccine (open diamonds; c). E749-specific CD8+ T-cell responses were determined by major histocompatibility complex tetramer staining of peripheral blood cells at indicated time points. Horizontal bars depict averages. (d) Tumor outgrowth in the cohorts of mice depicted in a−c; symbols are as in a−c. (e) Longterm survival of the mice depicted in a−c; symbols are as in a−c. Values represent mean +/- s.d. 67
A Rapid And Potent DNA Vaccination Strategy Defined By In Vivo Monitoring Of Antigen Expression
A
B
Figure 5. Tattoo DNA vaccination induces B-cell immunity against influenza A within 2 weeks.
% ova. spec. CD8+ T cells
(a) Cohorts of Balb/c mice (n = 6) were vaccinated at day 0, 3 and 6 either by intramuscular injection (filled triangles) or by tattooing (filled circles) with a construct encoding hemagglutinin. A positive control group was infected intranasally with influenza (open diamonds) and a negative control group tattooed using a mock vaccine (open circles). All mice were challenged with a sublethal dose of the influenza A virus at day 14 (indicated by the arrow), and weighed daily. The hemagglutinin-tattooed mice showed a significantly smaller drop in weight than the mock-tattooed mice (Student t-test, P = 0.039). Both in the mock-vaccinated and the intramuscularly vaccinated group, one mouse died from influenza-induced pneumonia. These animals were not included in the analysis. Values represent the weight relative to the weight at time of challenge. Error bars, s.e.m. (b) Influenza A/HK/2/68-specific antibodies in the sera of the mice depicted in a, at various time points after the start of vaccination and the influenza A challenge (symbols as in a). The presence of antibodies was assessed in an HAI assay. Values represent mean +/- s.d.
15 10
68
5 0 5 10 15 Days after start of vaccination
Supplementary Figure 1. DNA tattoo vaccination efficiently induces T cells against internal epitopes. Responses specific for the Chicken Ovalbumin257-264 epitope, in mice (n=5) that were short-interval tattoo vaccinated with a construct encoding whole chicken ovalbumin. Values represent the mean +/- standard deviation.
�
CD8beta-PE
CD8alpha-APC ������ ���� �����
��
������
3 CD8alpha-APC
% NP spec. CD8+ cells
Weight
Supplementary Figure 3. Tattoo ���� ����� DNA vaccination induces longterm immunity against influenza A. Cohorts of Balb/c mice (n=6) were vaccinated with a construct encoding HA, either at day 0, 14 and 28 by i.m. injection (▲) or CD8beta-PE at day 0,3 and 6 by tattooing (●). A negative control group was tattooed using a mock vaccine (○). All mice were intranasally infected with a sublethal dose of the influenza A virus 109 days after 110% the last DNA administration and weighed daily after the challenge. 100% The HA tattooed mice showed a significantly smaller drop in weight than the mock tattooed mice 90% (student t-test: P=0.038). Values represent the weight relative to the 80% weight at time of challenge. Error bars depict the standard error of the mean.
������ ���� ����� ��
B
������ ���� �����
Interferron-gamma-FITC
A
Interferron-gamma-FITC
H-2Db/NP366-374-tetramer-APC
Supplementary Figure 2. T cells induced by DNA tattoo vaccination are capable of direct effector function. (a) Tetramer staining of PBLs of a representative mouse, at day 11 after vaccination with the Luc-NP DNA vaccine. (b) Intracellular interferon-γ staining � of the same blood sample as in (a).
�
H-2Db/NP366-374-tetramer-APC
�
5
HA tattoo GFP tattoo HA intramuscular
0
2
4
6
8
10
12
14
Days after Influenza A challenge
B6 TLR9KO
Supplementary Figure 4. Absence of TLR9 does not measurably affect the immunogenicity of DNA tattoo vaccination. Peak NP366-specific T cell response induced by Luc-NP short interval tattoo DNA vaccination, in wildtype C57BL/6 mice (black bar) compared to TLR9 -/- mice (gray bar). Values represent the mean +/- standard deviation.
0 Peak response
69
A Rapid And Potent DNA Vaccination Strategy Defined By In Vivo Monitoring Of Antigen Expression
1. Ulmer,J.B. et al. Heterologous protection against influenza by injection of DNA encoding a viral protein. Science 259, 1745-1749 (1993). 2. Kirman,J.R. et al. Enhanced immunogenicity to Mycobacterium tuberculosis by vaccination with an alphavirus plasmid replicon expressing antigen 85A. Infect. Immun. 71, 575-579 (2003). 3. Yang,Z.Y. et al. A DNA vaccine induces SARS coronavirus neutralization and protective immunity in mice. Nature 428, 561-564 (2004). 4. Gurunathan,S., Klinman,D.M., & Seder,R.A. DNA vaccines: immunology, application, and optimization. Annu. Rev. Immunol. 18, 927-974 (2000). 5. Cappello,P. et al. LAG-3 enables DNA vaccination to persistently prevent mammary carcinogenesis in HER-2/neu transgenic BALB/c mice. Cancer Res. 63, 2518-2525 (2003). 6. De Marco,F. et al. DNA vaccines against HPV-16 E7-expressing tumour cells. Anticancer Res. 23, 1449-1454 (2003). 7. Murali-Krishna,K. et al. Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity 8, 177-187 (1998). 8. Kalat,M. et al. In vivo plasmid electroporation induces tumor antigenspecific CD8+ T-cell responses and delays tumor growth in a syngeneic mouse melanoma model. Cancer Res. 62, 5489-5494 (2002). 9. Smerdou,C. & Liljestrom,P. Non-viral amplification systems for gene transfer: vectors based on alphaviruses. Curr. Opin. Mol. Ther. 1, 244-251 (1999). 10. Ciernik,I.F., Krayenbuhl,B.H., & Carbone,D.P. Puncture-mediated gene transfer to the skin. Hum. Gene Ther. 7, 893-899 (1996). 11. Rice,J., Buchan,S., & Stevenson,F.K. Critical components of a DNA fusion vaccine able to induce protective cytotoxic T cells against a single epitope of a tumor antigen. J. Immunol. 169, 3908-3913 (2002). 12. Mamalaki,C. et al. Positive and negative selection in transgenic mice expressing a T-cell receptor specific for influenza nucleoprotein and endogenous superantigen. Dev. Immunol. 3, 159-174 (1993). 13. Chen,C. et al. Enhancement of DNA vaccine potency by linkage of antigen gene to an HSP70 gene. Cancer Res 60, 1035-1042 (2000). 14. Diehl,L. et al. CD40 activation in vivo overcomes peptide-induced peripheral cytotoxic T-lymphocyte tolerance and augments anti-tumor vaccine efficacy. Nat. Med. 5, 774-779 (1999).
70
15. Lipford,G.B. et al. CpG-containing synthetic oligonucleotides promote B and cytotoxic T cell responses to protein antigen: a new class of vaccine adjuvants. Eur. J. Immunol. 27, 2340-2344 (1997). 16. Lin,K.Y. et al. Treatment of established tumors with a novel vaccine that enhances major histocompatibility class II presentation of tumor antigen. Cancer Res. 56, 21-26 (1996). 17. Virelizier,J.L. Host defenses against influenza virus: the role of antihemagglutinin antibody. J. Immunol. 115, 434-439 (1975). 18. Eichelberger,M., Allan,W., Zijlstra,M., Jaenisch,R., & Doherty,P.C. Clearance of influenza virus respiratory infection in mice lacking class I major histocompatibility complex-restricted CD8+ T cells. J. Exp. Med. 174, 875-880 (1991). 19. Epstein,S.L. et al. Beta 2-microglobulin-deficient mice can be protected against influenza A infection by vaccination with vaccinia-influenza recombinants expressing hemagglutinin and neuraminidase. J. Immunol. 150, 5484-5493 (1993). 20. Scherle,P.A., Palladino,G., & Gerhard,W. Mice can recover from pulmonary influenza virus infection in the absence of class I-restricted cytotoxic T cells. J. Immunol. 148, 212-217 (1992). 21. Wang,R. et al. Induction of antigen-specific cytotoxic T lymphocytes in humans by a malaria DNA vaccine. Science 282, 476-480 (1998). 22. Wang,R. et al. Induction of CD4(+) T cell-dependent CD8(+) type 1 responses in humans by a malaria DNA vaccine. Proc. Natl. Acad. Sci. U. S. A 98, 10817-10822 (2001). 23. Sullivan,N.J., Sanchez,A., Rollin,P.E., Yang,Z.Y., & Nabel,G.J. Development of a preventive vaccine for Ebola virus infection in primates. Nature 408, 605-609 (2000). 24. Sullivan,N.J. et al. Accelerated vaccination for Ebola virus haemorrhagic fever in non-human primates. Nature 424, 681-684 (2003). 25. Wolkers,M.C., Toebes,M., Okabe,M., Haanen,J.B.A.G., & Schumacher,T.N.M. Optimizing the Efficacy of Epitope-Directed DNA Vaccination. J. Immunol. 168, 4998-5004 (2002). 26. Panina-Bordignon,P. et al. Universally immunogenic T cell epitopes: promiscuous binding to human MHC class II and promiscuous recognition by T cells. Eur. J. Immunol. 19, 2237-2242 (1989).
71
3
C H A P T E R
4
Superior immunogenicity of DNA tattooing as compared to intramuscular DNA vaccination in non-human primates. Adriaan Bins#~, Babs Verstrepen*~, Christine Rollier*, Petra Mooij*, Gerrit Koopman*, Cynthia Robson@, Neil Sheppard@, Ralf Wagner$, Hans Wolf$, Ton Schumacher#, Jonathan Heeney* and John Haanen#.
Dept of Immunology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX, Amsterdam, The Netherlands. #
Dept of Virology, Biomedical Primate Research Centre, Rijswijk, The Netherlands, *
Sir William Dunn School of Pathology, University of Oxford, Oxford, UK
@
Institute of Medical Microbiology and Hygiene, Regensburg, Germany.
$
~
Both authors contributed equally
Correspondence should be addressed to J.B.A.G.. (
[email protected])
72
Introduction The development of heterologous prime-boost vaccination regimens has accentuated the need for potent priming methods, to increase immune responses upon boosting with adeno- or pox-based immunogens. DNA vaccines are an attractive component for the priming phase of such regimens, as T cell induction is unhindered by vector specific immunity and can occur in a flexible, safe and simple manner1. Unfortunately, the high efficacy of DNA vaccines seen in early and more recent mouse studies has not been mirrored in humans and in non-human primate DNA vaccination trials. The poor track record of DNA vaccines in primates has been a major impediment for the development of clinical DNA vaccines. The low immunogenicity of intradermal DNA vaccines in primates has been suggested to be a consequence of the low TLR9 expression on skin-resident APCs2 of primates. Furthermore, in relation to body mass the typical DNA doses used in large animals are generally 5-10 fold lower than those used in rodents. Reflecting an incomplete understanding of the factors involved, the issue has been termed ‘the simian barrier’3. Recently, a new intradermal DNA vaccination method was developed and optimized by us in mouse models. In this strategy -termed DNA tattooing- vaccination is performed by applying a droplet of DNA to the skin followed by exposure of the skin to the needle head of a rapidly oscillating tattoo device, resulting in the creation of thousands of punctures in the DNA-coated area. In mice, DNA tattooing leads to a significantly higher level of antigen presentation in the draining lymph node as compared to intramuscular injection and to the induction of strong T cell responses within a timespan of weeks4. Furthermore, this novel method of immunization makes it possible to scale DNA dose, directly linear to the total body surface across different species. Based on the poor track record of DNA vaccination approaches in primates it seemed of importance to assess the effectiveness of the newly developed approach in a non-human primate model. Here we report on the efficacy of intradermal DNA tattoo vaccination in Macaca mulatta. Experimental setup We vaccinated 5 outbred macaques with a 1mg/ml DNA vaccine containing the plasmids pORT-GPN and pORT-gp120 at equimolar concentration. pORT-GPN encodes a biologically inactive gag-pol-nef fusion protein of the clade C HIV-1 strain 97cn54. pORT-gp120 encodes a secreted version of the Env protein from the same HIV strain (for a detailed description of the plasmids see supplementary information). The vaccination regimen consisted of 3 tattoo applications at 3 day intervals. This short-interval vaccination regimen was identical to that previously used in mice, where it induces robust cytotoxic T cell responses within 2 weeks4. In addition, to allow a comparison with retrospective data sets (see below), animals received an identical set of booster vaccinations after 4 weeks. During each tattoo an area of approximately 5x6 cm (30 cm2) on each leg was treated. On two legs together, each treatment covers a fraction of the body surface 73
4
INTRADERMAL DNA TATTOOING PRIMES ROBUST T CELL RESPONSES IN MACCACA MULATTA.
that is similar to the fraction treated in previous rodent experiments (5-10% of the body surface for each application). The first of the three tattoos was placed on the inner thighs, the second on the upper thighs and the third on the outer thighs. With each tattoo, 0.65 ml of a 1mg/ml DNA solution was applied on each leg and the area was treated using a disposable 27-needle bar (each creating skin perforations with a diameter of approximately 0.3 mm) mounted on a rotary tattoo machine oscillating at 100 Hz. Using this setup, approximately 300,000 skin perforations are inflicted during the 2 minute treatment of each leg. A total of approximately 8 mg DNA was applied over the course of the whole prime-boost regimen. Kidney and liver function parameters, white and red blood cell counts and inflammatory parameters, were determined once every week and 3 days after each tattoo. Blood was drawn for enumeration of IFN-γ and IL-4 producing peripheral blood mononuclear cells (PBMC) by enzyme-linked immunospot (ELIspot) assays at week 4 and 8 post vaccination, and by intracellular cytokine staining (ICS) at week 2 post vaccination. Additionally, at week 4 and 8 blood was drawn for measurement of Envelope specific IgG antibody titers by enzyme linked immunosorbent assays (ELISA). ELIspot and ELISA data generated in the experiment were compared to those of 10 macaques that previously received a prime-boost vaccination by bolus intramuscular (im) injection of 2 ml of the same DNA vaccine batch at week 0 and week 4 (8 mg of DNA in total). As a second comparison, responses were compared to those in 10 macaques vaccinated with a NYVAC (New York Vaccinia) modified pox vector expressing the same HIV gene sequences (1 ml i.m. injection in each arm, at a dose of 5x107 pfu/ml). The study design is schematically represented in Figure 1. Analyses of immune responses in both control groups had been performed at the same time points post vaccination and using exactly the same methods and reagents as in the 5 animals that were vaccinated by DNA tattoo. The study protocol and experimental procedures were approved by the institutional ‘animal ethical care and use’ committee. Results None of the tattooed animals developed adverse side-effects upon DNA tattooing and all animals gained weight normally during the course of the experiment. To a variable extent, in all animals a thin yellowish crust of coagulated transudate appeared on the treated sites the day after tattooing (Fig 2). However, these crusts disappeared within 2 - 5 days. The animals did not scratch nor otherwise manipulate the treated skin areas. Strikingly, at 2 weeks after the start of vaccination, CD8+ IFN-γ secreting PBMC directed against Pol or Nef could be detected in 4 of the 5 tattooed animals by ICS (suppl. Fig. 1). In accordance with this finding, the ELIspot analysis at 4 weeks after the start of vaccination showed that 5 of the 5 tattooed animals had developed a robust response of antigen-specific IFN-γ secreting PBMC. These responses were predominantly directed against Pol, and consisted of on average 1678 SFC (Spot Forming Cells) per million PBMC (ranging from 487 to 5398 SFC). The consistency of these vaccine-induced cytotoxic responses in all 5 animals contrasted sharply with i.m. or NYVAC vaccination (p=0.001 and p=0.001), both 74
of which induced low-level HIV-specific T cell responses in a fraction of animals at this time point (average 39 SFC, range 0-130, fig 3a). Four weeks after the second vaccination (8 weeks after the first vaccination), the antigen specific IFN-γ responses in the tattooed group were 702 SFC on average, compared to 31 SFC after i.m (p=0.001) and 1 SFC after NYVAC (p=0.001, Fig 3a). Notably, only for the group treated by DNA tattoo these values are significantly higher than the background reactivity against the used HIV antigens that was present before the onset of vaccination (Mann-Whitney test: p=0.037 for tattoo, p=0.21 for i.m. and p=1 for NYVAC). Interestingly, at 8 weeks after vaccination, in addition to IFN-γ producing PBMC, a population of IL-4 secreting PBMC was also detected in 3 out of 5 tattooed animals (Fig 3b). These cells were mainly directed against epitopes in Pol and Nef, and consisted of 727 SFC on average. (see suppl. Fig. 2 for detailed graphs). In contrast, no significant increase in IL-4 secreting PBMC populations was observed upon i.m. DNA or NYVAC vaccination (2 SFC for i.m. DNA and 16 SFC for NYVAC) (suppl. Fig 1). In spite of the presence of IL-4 secreting PBMC, no significant increase in titers of gp140 specific IgG or IgM antibodies were observed upon DNA tattoo (data not shown). Discussion In the current study we have analyzed the immunogenicity of DNA tattooing in non-human primates. A strong induction of IFN-γ T cell responses was observed within weeks after priming and the observed induction of cytotoxic responses by DNA vaccination in a primate model only upon priming has to our knowledge not been observed previously using other methodologies. Also after booster immunization, the frequencies of vaccine-specific IFN-γ-secreting cells remained 10-100 fold higher than those observed upon i.m. DNA vaccination or vaccination with a NYVAC vector expressing the same sequences. Possibly, the fact that the tattooed area can be scaled-up linear to an increase in body surface, leading to the availability of the encoded antigens to a potentially greater population of dermal APC, contributes to the efficacy of DNA tattooing in larger animals. These data indicate that further exploration of intradermal DNA vaccination by tattoo device or equivalent means in clinical trials is mandated. Such trials should focus on the use of this technology either in therapeutic settings, or on the use of DNA tattoo to induce high level T cell responses against major pathogens in high risk populations. Materials and Methods DNA vaccine. Within the EC Eurovacc vaccine development program (QLK2-CT-1999-01321) gene-optimised GagPolNef and Env gene vector inserts were designed by GENEART GmbH using the GeneOptimizerTM software package. Clade C/B GagPolNef (GPN) and Env (gp120) sequences were designed based on sequence information derived from a 97cn54 provirus clone5 (sequence submitted to GenBank). The gene-optimized 97cn54 Env construct comprises 1500 nucleotides, 75
4
INTRADERMAL DNA TATTOOING PRIMES ROBUST T CELL RESPONSES IN MACCACA MULATTA.
encoding an artificial signal peptide followed by gp120 97cn54. The 5’ part of the CN54 GPN polygene construct encodes the modified (myristylation deficient) group specific antigen. The Gag coding sequence is followed by a 952 bp fragment encoding the 5’ part of pol including a mutation leading to inactivation of the viral protease. A 618 bp fragment encoding a scrambled Nef variant was fused to the 3’ end of pol coding sequence replacing the active site of the reverse transcriptase. The 3’ pol reading frame lacking the integrase gene was extended by the 3’ end of the scrambled Nef gene. The sequence stretch encoding the active site of the reverse transcriptase was translocated to the 3’ end of the polygene construct, resulting in an open reading frame encoding the 160kDa non-glycosylated artificial GPN polyprotein. The C-clade-synGagPolNef gene was cloned into the pCRScript amp(+) cloning vector (Stratagene, La Jolla, USA) resulting in the vector pCRScript-C-synGagPolNef. The cladeC-syngp120 gene was cloned into the pCR-Script amp(+) cloning vector resulting in the vector pCRScript-C-syngp120. The pCRScript based vectors were used to generate the DNA vaccine candidate plasmids pORT1aGPN and pORT1agp120. pORT1aGPN and pORT1agp120 were selected and produced in accordance to GMP standards using the E. coli strain DH1lacdapD (Cobra Therapeutics Inc). Upscaling and production of cGMP grade material was performed by Cobra Biomanufacturing Plc, Keele, Staffordshire, UK. Therapeutics Inc. NYVAC vector The NYVAC vector expressing Gag/Pol/Nef and Env of clade C HIV-1 97cn5 was provided by Dr. MJ Frachette (Sanofi Pasteur, Lyon, France). Functional expression of the donor genes was demonstrated by Western blot analysis of the proteins produced by the pre-master seed lot. The clinical batch of the vaccine was manufactured by Sanofi Pasteur using a process previously developed for the manufacture of other ALVAC- and NYVAC-based vaccines. IFN-γ and IL-4 ELISPOT assays. Quantification of antigen specific cytokine secreting cells was performed on freshly isolated PBMC by IFN-γ and IL-4 enzyme-linked immunospot (ELIspot) assays, according to the manufacturer’s instructions (U-Cytech, Utrecht, The Netherlands). Positive control was SEB 1µg/ml, negative control was medium alone. Antigen specific responses were measured against 5µg/ml peptide pools: 15-mers with 11 aa overlap spanning the entire Gag/Pol/Nef polygene, and the Env clade C of HIV-1 97cn54 (Synpep Corporation, Dublin, CA, USA). Eight peptide pools were made as follows: Gag1, 60 peptides (Cg1-240); Gag2, 61 peptides (Cg244); Pol1, 60 peptides (Cgp485-721); Pol2, 61 peptides (Cp725-817, Cp1017-1161); Pol3, 61 486 peptides (Cp1165-1403); Nef 49, peptides (Cn838-1030); Env1, 49 peptides (CN9-249); Env2, 63 peptides (CN253-485). Results are expressed as the mean number of spot forming cells (SFC) per 1000000 PBMC from triplicate samples minus background values (mean number of SFC plus 2xSD of triplicate samples with medium alone). Mean SFC of each animal (dots) are expressed per antigen (responses to 2 Gag pools, 3 Pol pools and 2 Env pools were combined). 76
Intracellular cytokine analysis. The phenotype of responding T-cells was analyzed by ICS and flow cytometric analysis as described elsewhere6 with minor modifications. PBMC were O/N incubated, in the presence of stimulatory antibodies specific for CD28 and CD49d (2µg/ml each), with medium alone (negative control), 1µg/ml SEB (positive control) or stimulated with 5µg/ml of the same 8 peptide pools as used in the ELISPOT assay as described above. Cells were stained with directly conjugated antibodies (Becton Dickinson, Pharmingen, CA, USA) to CD3 (peridininchlorophyll-protein complex (PerCP)-labeled), CD4 (phycoerythrin-cyanin 7 (PE-CY7) labeled), CD8 (allophycocyanin (APC)-CY7 labeled), and to CD14 and CD20 (Energy Coupled Dye (ECD) labeled, Beckman Coulter, Marseille, France) for exclusion of monocytes and B-cells. The cells were fixed and permeabilised with Cytofix/Cytoperm buffer (Becton Dickinson) and stained intracellularly with antibodies specific for IFN-γ (APC-labeled), IL-2 (PE-labeled), TNF-α (fluorescein isothiocyanate (FITC) labeled) in Perm/wash solution (Becton Dickinson). 100,000 – 200,000 events were acquired on a FACS Aria (Becton Dickinson, CA, USA). FACS Diva (Beckton Dickinson) and Summit (Dako, Colorado, USA) software was used for analysis. ELISA. Antibodies to HIV-1 Env (gp140 97cn54) in serum were measured by ELISA. Plates were coated overnight with Env in 100mM NaHCO3 and were blocked for 1 hour with 1% non-fat milk before application of serum, serially diluted in 1% BSA PBS buffer. After 1 hour 1µg/ml anti-human IgG or IgM-HRP conjugate was added for 1 hour before addition of ultra-TMB ELISA development reagent. The reaction was stopped by addition of 0.5M H2SO4. Results were expressed as IgG and IgM endpoint dilution titers. Fig. 1
I.m. NYVAC day 0
Boost, day 28
retrospective group
I.m. DNA day 0
Boost, day 28
retrospective group
Tattoo DNA day 0-3-6
Boost, day 28-31-34
new data group
n=10
n=10
n=5
0 1 weeks
2
ICS
3
4
ELIspot ELISA
5
6
7
8
ELIspot ELISA
Fig 1. Schematic representation of the experimental setup. Arrows indicate time points of vaccination, either by DNA tattooing, i.m. DNA injection or NYVAC injection. 77
4
Fig. Fig. 22
INTRADERMAL DNA TATTOOING PRIMES ROBUST T CELL RESPONSES IN MACCACA MULATTA.
directly after tattooing Directly after tattooing
Fig. 3Fig. 3
3 days after tattooing tattooing
Three days after tattooing
IFN-gamma IFN-gamma
IL-4
10000
10000
1000
1000
100
100
10
10
1
1 0
0
4
4
8
8
IL-4 SFC / 106 PBMC
IFN-γ
IL-4 SFC / 106 PBMC
SFC / 1 million PBMC
IFN-gamma SFC / 106 PBMC
IFN-gamma SFC / 106 PBMC
Fig 2. Photographs of the skin areas treated by DNA tattooing of 3 representative animals, taken either directly after (upper row) or three days after (lower row) the treatment. The lower and upper rows display the same animals.
10000
10000
1000
1000
100
100
10
10
1
0
IL-4 IL-4
1 0
4
4
8
8
weeks start of the vaccination after after first afterimmunization first immunization Weeks Weeks after first afterimmunization first immunization B AWeeksWeeks
Fig 3. Quantification of antigen specific cytokine secreting cells by ELIspot at week 4 and week 8 post vaccination in the DNA tattoo vaccinated (n=5) group, i.m. DNA vaccinated (n=10) group and NYVAC vaccinated group (n=10). The assays were performed on freshly isolated PBMC by A) IFN-γ, and B) IL-4 ELIspot assays, according to the manufacturer’s instructions (U-Cytech, Utrecht, The Netherlands). Cytokine responses of individual animals are plotted as diamonds (tattoo group), circles (IM group) and triangles (NYVAC group). The group averages are indicated as thick closed lines (tattoo group), dotted lines (IM .group) and thin closed lines (NYVAC group). Results are expressed as the mean number of spot forming cells (SFC) per 106 PBMC from triplicate samples minus background values (mean number of SFC plus 2xSD of triplicate samples with medium alone). The represented spot number is calculated by summation of the readout in 8 separate assays, each done upon stimulation with a distinct fraction of the complete peptide pool. For a representation of spot numbers upon stimulation with the separate fractions representing pol, gag, nef and env, see Suppl. Fig. 2. 78
Peptide-pool Stimulated
TNF-α
Unstimulated Control
Animal #1
104
104
0.12%
10
103
102
102
3
101 101
Animal #2
0.17% 102
103
102
102
0.26%
10
102
103
102
103
104
Nef
101
1.12% 101
104
104
0.03%
Pol-2
0.62% 101
103
104
102
0.34%
103
104
Pol-3
103
3
102
102
101
101
0.17% 101
102
103
1.78% 101
104
104
104
10
3
103
102
102
102
0.59% 101
102
103
104
104
10
3
103
102
102
101
0.43% 101
102
103
104
104
4
3.25% 101
104
103
Nef
101
101
Animal #5
104
104
101
Animal #4
101
104
101
Animal #3
103
1.56%
102
103
104
Pol-2
101
2.97% 101
102
103
104
IFN-γ
Suppl. Fig 1. Flow cytometric analysis of peptide stimulated PBMC for TNF-α and IFN-γ secretion. In each animal T cell responses against each of 8 fractions of the Suppl Fig 1 complete peptide pool were tested, together covering the whole vaccine. For each animal, the plot in the right column displays cytokine secretion upon stimulation with the most stimulant peptide fraction of the pool. All plots are gated on CD8+ T-cells. The name of this fraction is printed inside each plot, indicating the HIV antigen that it (partly) represents. The plots in the left column displays cytokine secretion without peptide stimulation. See supplementary data for technical details of the assay. 79
INTRADERMAL DNA TATTOOING PRIMES ROBUST T CELL RESPONSES IN MACCACA MULATTA.
Intradermal DNA Tattoo
Peptide Pools:
Pol 1+2+3
SFC / 106 PBMC
Gag1+2
IFN-γ
Env 1+2
Nef
Intramuscular NYVAC injection
Intramuscular DNA injection
Weeks post-vaccination
Gag1+2
SFC / 106 PBMC
Env 1+2
Nef
IL-4
Pol 1+2+3
Weeks post-vaccination Suppl Fig 2
80
Suppl. Fig 2. Upper panel: IFN-γ production by peripheral blood mononuclear cells (PBMC) to Env, Gag, Pol and Nef peptides as measured by ELIspot assay. Background responses (mean number of SFC plus 2xSD of triplicate samples with medium alone) were subtracted. The solid line represents the mean response of all animals within one group. Lower panel: the same display but for IL-4 production.
References 1. Gurunathan, S., D. M. Klinman, and R. A. Seder. 2000. DNA vaccines: immunology, application, and optimization*. Annu Rev Immunol. 18:927-74.:927974. 2. Rothenfusser, S., E. Tuma, M. Wagner, S. Endres, and G. Hartmann. 2003. Recent advances in immunostimulatory CpG oligonucleotides. Curr Opin Mol Ther. 5:98-106. 3. Johnston, S. A., A. M. Talaat, and M. J. McGuire. 2002. Genetic immunization: what’s in a name? Arch Med Res. 33:325-329. 4. Bins, A. D., A. Jorritsma, M. C. Wolkers, C. F. Hung, T. C. Wu, T. N. Schumacher, and J. B. Haanen. 2005. A rapid and potent DNA vaccination strategy defined by in vivo monitoring of antigen expression. Nat Med. 11:899-904. 5. Su, L., M. Graf, Y. Zhang, H. von Briesen, H. Xing, J. Kostler, H. Melzl, H. Wolf, Y. Shao, and R. Wagner. 2000. Characterization of a virtually full-length human immunodeficiency virus type 1 genome of a prevalent intersubtype (C/B’) recombinant strain in China. J Virol. 74:11367-11376. 6. Balla-Jhagjhoorsingh, S. S., G. Koopman, P. Mooij, W. Koornstra, S. McCormack, J. Weber, G. Pantaleo, and J. L. Heeney. 2004. Long-term persistence of HIV-1 vaccine-induced CD4+CD45RA-CD62L-CCR7- memory T-helper cells. AIDS. 18:837-848. 81
4
C H A P T E R
5
In vivo antigen stability affects vaccine efficacy
Adriaan D. Bins#&, Monika C. Wolkers#&*, Marly D. van den Boom#, John B.A.G. Haanen#, and Ton N.M. Schumacher#
# Division of Immunology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX, Amsterdam, The Netherlands. * Present address: La Jolla Institute for Allergy and Immunology, 9420 Athena Circle, La Jolla, CA 92037, USA & These authors contributed equally Correspondence should be addressed to T.N.M.S. (
[email protected])
Abstract The factors that determine efficient presentation of antigens encoded by viral vaccines or DNA vaccines to cytotoxic T cells in vivo are largely unknown. Prior work has suggested that cross-presentation of cytotoxic T cell antigens would either involve transfer of peptide antigens from antigen-producing cells to antigenpresenting cells, or would involve the transfer of protein antigen between the two cell types. While these models make predictions on the requirement for in vivo accumulation of vaccine-encoded antigens, the influence of antigen stability on CD8+ T cell induction has not been addressed in clinically relevant vaccine models, nor has the accumulation of vaccine-encoded antigens been monitored in vivo. Here we describe the relationship between in vivo antigen stability and immunogenicity of DNA vaccine-encoded antigens. We show that in vivo accumulation of DNA vaccine-encoded antigens is required for the efficient induction of CD8+ T cell responses. These data suggest that many of the currently used transgene designs in DNA vaccination trials may be suboptimal, and that one should either use pathogen-derived or tumor-associated antigens that are intrinsically stable, or should increase the stability of vaccine-encoded antigens by genetic engineering. 82
Introduction Vaccines that aim to induce cytotoxic T cell responses should achieve two goals: First, vaccination should lead to the presentation of sufficient densities of the relevant peptide/MHC (pMHC) complexes and second, such presentation should occur by mature professional antigen presenting cells (APC). While the signals that lead to maturation of professional antigen-presenting cells (APCs) have been studied in considerable detail (1), our understanding of the factors that lead to efficient MHC class I-restricted presentation of vaccine-encoded antigens to naïve T cells is incomplete. Currently, there is solid evidence that peptides can be presented by MHC class I molecules by two distinct pathways (2). The classical endogenous antigen presentation pathway that processes antigens that are generated within the APC itself (3-6) and the cross-presentation pathway, in which professional APCs present T cell antigens that have been acquired from other cell types (3). Importantly, for a number of viral vectors that are employed in vaccination strategies such as vaccinia and Adenovirus group A, C, D, E and F, T cell induction depends to a large extent on cross-presentation of the vaccine-encoded antigens (7,8). Likewise, following intradermal (9) or intramuscular DNA vaccination (10,11), crosspresentation of the DNA vaccine-encoded antigens is a dominant mechanism for T cell induction. Several studies have shown that cross-presentation of MHC class I-restricted epitopes involves the transfer of intact antigens from the antigen-donating cell to the APC (12-15). Conversely, other groups have provided evidence for a role of proteasomal products and intermediates, possibly in the form of heat shock protein (HSP)-bound peptides, as a main antigen source for cross-presentation (16-18) . Finally, a recent study has indicated that gap junctions can allow transfer of peptide fragments between cells, a pathway that could potentially also be employed by pAPCs to acquire antigens (19). While the use of in vitro experiments (19) transplantable tumor models (12,15,19), the injection of apoptotic tumor cells (17) or tumor cell extracts (14,16) in these studies has provided useful information on the possible mechanisms underlying cross-presentation, neither of these experimental settings mimics clinically accepted vaccination strategies. Consequently, it remains unclear whether vaccine development should focus on designs that result in accumulation of full-length antigen, or whether production of peptide epitopes would be preferred. This issue is particularly relevant for ongoing DNA vaccination trials, as the stability of the vaccine-encoded antigen is generally not taken into account in vaccine design, thereby potentially limiting vaccination efficiency. To unravel the parameters that determine efficient priming of cytotoxic T cell responses to vaccine-encoded antigens, we have developed a vaccination model that allows us to directly follow vaccine-encoded antigen stability in vivo. We have used this model to determine the effect of antigen stability on T cell induction. and demonstrate that in vivo antigen accumulation profoundly impacts the efficacy of in vivo antigen cross-priming. These data suggest that antigen stability should form an important consideration in vaccine design. 83
5
IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY
Results Epitope location affects DNA vaccine efficacy We have previously shown that tumor-associated T cell epitopes that are derived from signal peptides are cross-presented inefficiently (15). To determine whether for vaccine-encoded antigens a similar bias occurs, we made use of a recently developed multineedle intradermal DNA vaccination strategy (20), hereafter referred as DNA tattoo. As is the case for gene gun DNA vaccination (9) induction of T cell responses by DNA tattooing occurs at least partially through cross-presentation (Suppl. Fig. 1, see below). To assess the effect of epitope location on vaccine immunogenicity, we generated a pcDNA3.1-based vector that encodes a secreted GFP-based fusion protein, sNP-GFP-E7. This model antigen contains two CD8+ T cell epitopes; the influenza A nucleoprotein-derived NP366 epitope located within the hydrophilic segment of the signal peptide, and the Human Papilloma virus-derived E749 epitope located within the COOH-terminal part of the GFP protein (Fig 1A). The rationale for this design is that it allows the simultaneous measurement of T cell responses against two T cell epitopes, of which one is incorporated in the stable mature GFP protein and the other within the signal peptide that is thought to be short lived (21,22). To be able to correct for possible intrinsic differences in immunogenicity between the two T cell epitopes, a second fusion gene (sE7-GFP-NP) containing the same two model antigens but in reverse order was used in parallel. Mice were vaccinated with the sNP-GFP-E7 or the sE7-GFP-NP DNA vaccines, and T cell responses were monitored directly ex vivo by MHC tetramer staining (Fig.1B). Vaccination with these DNA vaccines resulted in considerable frequencies of antigen-specific T cells specific for the antigen expressed within the mature protein, be it either the NP366 epitope, or the E749 epitope (Fig. 1C). In striking contrast, when the NP366 epitope, or the E749 epitope was encoded within the signal peptide, DNA vaccination failed to induce antigen-specific T cell responses above background (Fig. 1C). These data establish that, analogous to tumor cell-associated antigens (15), the immunogenicity of DNA vaccine-encoded antigens is biased towards epitopes in mature proteins, and against epitopes in signal peptides. Antigen stability determines the efficiency of DNA vaccination Signal peptide-derived epitopes may inefficiently enter the cross-presentation pathway as a consequence of their limited life-span (15,21). However, no measurements exist of the in vivo half-life of signal peptides and it is clearly possible that other characteristics of signal peptides adversely affect T cell priming via cross-presentation. Consequently, while the data show that signal peptideencoded epitopes are strongly disfavored, the broader implications of this finding are unclear. To be able to directly assess whether the half-life of a vaccine-encoded antigen would influence its immunogenicity, we generated DNA vaccines encoding an antigen that could be destabilized to a variable degree by a minor modification in its sequence. This antigen consists of three basic components (Fig. 2A). At the COOH-terminus of the fusion protein the NP366 T cell epitope is present to be able
84
to monitor immunogenicity of the antigen. This epitope is preceded by the Firefly luciferase (luc) that is used to follow in vivo expression of the antigen by intravital imaging. In addition, the luciferase gene product carries a variable residue at its amino-terminus. Finally, a ubiquitin (Ub) moiety is encoded amino-terminal to the luciferase gene product, to control antigen half-life. This NH2-terminal ubiquitin moiety is cotranslationally removed by ubiquitin-specific proteases (USPs), thereby exposing the variable amino-terminal residue of the luciferase gene product. As previously established by Varshavsky and colleagues, the nature of this amino acid can determine protein half-life in a hierarchical manner known as the N-end rule (23). Thus, the Ub-X-lucNP fusion proteins described here are designed to display a different half-life, depending on the nature of the X-lucNP NH2-terminal residue. To determine if the intracellular stability of X-Luc-NP fusion proteins is controlled by the N-end rule pathway, we transfected COS cells with plasmids encoding Ub-MlucNP or Ub-R-lucNP (in which M and R are stabilizing and destabilizing residues respectively) and determined lucNP expression level by measuring luciferase activity (Fig. 2B). Both X-lucNP variants were enzymatically active, as judged by bioluminesence measurements after addition of the substrate luciferin (Fig. 2B). To determine whether the half-life of the Luc-NP fusion proteins is dependent on the nature of the NH2-terminal amino acid, proteasomal activity was inhibited by addition of MG132 (z-Leu-Leu-Leu-al.) A 4 h inhibition of proteasomal activity did not measurably alter bioluminescence of cells transfected with Ub-M-lucNP DNA. In contrast, luciferase activity of cells transfected with Ub-R-lucNP DNA was increased approximately 4-fold upon inhibition of proteasomal activity, indicating that the resulting R-lucNP protein has a significantly shorter half-life than M-LucNP. In line with this, immunoprecipitation of luciferase from pulselabeled cells transfected with Ub-R-lucNP or Ub-M-lucNP showed a substantial difference in half-life between the two proteins (Fig 2C, D). The activity of the E3-ligases that recognize different NH2-terminal residues may be cell type dependent (24). Therefore, it was essential to determine the in vivo halflife of vaccine-encoded Ub-M-lucNP and Ub-R-lucNP fusion proteins. To this end, mice received a single intradermal DNA vaccination with either Ub-M-lucNP or Ub-R-lucNP DNA, and in vivo protein stability was subsequently determined by measuring luciferase activity at several time points post DNA application (Fig. 3A). Within the first 12 h post-immunization, luciferase activity increased rapidly and (as expected, see legend to Fig. 3) did not substantially differ between UbM-lucNP and Ub-R-lucNP vaccinated animals. However, at later time points the luciferase activity in the Ub-R-lucNP group decreased much more rapidly than in the Ub-M-lucNP group. Specifically, at 72 h post-vaccination, luciferase activity in Ub-M-lucNP vaccinated animals was approximately 100-fold higher than that in Ub-R-lucNP vaccinated animals. Based on the rate of decay of luciferase activity from 24 h onwards, the in vivo half-life of R-lucNP and M-lucNP are 5 and 11 h, respectively. It is noted that these numbers may underestimate the actual difference
85
5
IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY
in antigen half-life of the two proteins, because it assumes that LucNP synthesis has ceased after 24h while low level de novo production may still occur at this timepoint. To examine whether the in vivo difference in intracellular antigen half-life affects the immunogenicity of these DNA vaccines, we measured NP-specific CD8+ T cell responses by ex vivo MHC tetramer staining of peripheral blood samples upon vaccination with either Ub-M-lucNP or Ub-R-lucNP. Whereas vaccination with Ub-M-lucNP DNA yielded NP366-specific CD8+ T cell responses with a peak average of 10.4%, (Fig. 3B), vaccination with the destabilized variant Ub-R-lucNP resulted in a 4.7 fold lower antigen-specific T cell response (2.2% on average at the peak; p = 0.008). These data suggest that antigen half-life is a major determinant of the efficacy of intradermal DNA vaccines. Antigen stability influences cross-presentation of MHC class I-restricted epitopes in the DLN. The above data demonstrate the effect of in vivo antigen stability on the immunogenicity of intradermal DNA vaccines, but do not reveal whether this effect is due to enhanced cross-presentation or due to enhanced endogenous presentation by transfected APCs. To address this issue, mice were vaccinated with DNA vectors encoding the Ub-M-lucNP and Ub-R-lucNP variants under control of the K14 promoter. As opposed to the CMV-IE promoter employed in Fig. 1 and 3 (and in the majority of human vaccination trials (25)), the K14 promoter is solely active in epidermal cells (9), thereby eliminating a contribution of directly transfected professional APCs to antigen presentation. Mice that were vaccinated with the (K14)-Ub-M-lucNP DNA vaccine mounted peak NP366-specific T cell responses of on average 1.4% (Fig. 4A). In contrast, after vaccination with the (K14)-Ub-R-lucNP DNA vaccine, T cell responses remained close to background levels (0.1% on average, p = 0.014). These data indicate that antigen stability substantially affects the immunogenicity of DNA vaccines in a setting where T cell induction can only occur through epitope presentation by APCs that have obtained antigen from other cells. Of note, the magnitude of T cell responses induced by K14-driven DNA vaccines is close to 10-fold lower as compared to those induced by CMV IE-driven vaccines (Fig. 3C versus 4A). This difference is likely to be a consequence of the 10-50 fold lower level of protein production from K14-driven vaccines (Suppl. Fig. 1B). In addition, it is possible that endogenous antigen presentation by transfected APCs enhances T cell induction (26) in the case of CMVIE driven vaccines. Because CD4+ T cell help is required for efficient DNA vaccination (27), we set out to determine whether the higher immunogenicity of stable antigens is due to enhanced MHC class I-restricted cross-presentation, or can be explained by more effective CD4+ T cell help. To this end, CFSE-labeled NP366-specific F5 TCR transgenic CD8+ T cells were transferred into TCRβ-deficient mice that lack both CD4+ and CD8+ T cells. Subsequently, the mice received DNA vaccines encoding
86
Ub-M-lucNP or Ub-R-lucNP, and at day 7 post vaccination the draining lymph nodes were excised and analyzed for division of F5 T cells. In mice that were vaccinated with Ub-M-lucNP, on average 7.0% of the F5 T cells had divided (Fig. 4B), whereas vaccination with Ub-R-lucNP led to considerably less T cell division (2.8% on average, p = 0.03). As a second approach, we determined the induction of NP366 specific T cell responses in MHC class II-deficient mice that lack CD4+ T cells. Consistent with prior data on the requirement for CD4+ T cell help in DNA vaccination (9,27,28), overall CD8+ T cell responses in MHC class II-deficient mice are very low. However, Ub-M-lucNP vaccination still induced significantly higher CD8+ T cell responses than vaccination with Ub-R-lucNP (0.7% versus 0.2%, p = 0.0003, Fig 4C). These results demonstrate that for intradermal DNA vaccines, the stability of the antigen directly affects the efficiency of cross-presentation of MHC class I-restricted epitopes in the draining lymph node (DLN). To further dissect the effect of antigen stability on T cell induction, we generated 3 more variants of lucNP that encoded glycine (G), serine (S) or phenylalanine (F) as the NH2-terminal amino acid that is created upon Ub removal (Ub-G-lucNP, Ub-S-lucNP and Ub-F-lucNP). These residues were chosen because they were previously (29) found to yield fusion proteins with a range of half-lifes in mouse cells in vitro. Specifically, a serine-based fusion protein displayed an identical or marginally higher stability in L cells as compared to the wild type Ub-M-lucNP in a one hour chase period, and glycine and phenylanaline-based fusion proteins have an intermediate and low half-life relative to M and R based fusion proteins. First, we assessed the in vivo stability of these variants by intravital bioluminescence imaging. Interestingly, comparison of the in vivo half-life revealed that luciferase activity upon Ub-S-lucNP DNA vaccination persisted substantially longer than after Ub-M-lucNP vaccination (Fig. 5A, half life of 15 h versus 11 h, based on decay in luciferase activity from 24 h onwards), whereas the in vivo half-life of UbG-lucNP and Ub-F-lucNP fell in between that of Ub-M-lucNP and Ub-R-lucNP (9 h and 7 h, respectively). We then tested the immunogenicity of DNA vaccines encoding these three additional stability variants in parallel with DNA vaccines encoding Ub-M-lucNP and Ub-R-lucNP. In accordance with the hypothesis that antigen half-life controls vaccine efficacy, T cell responses induced by Ub-S-lucNP were highly robust and exceeded that of Ub-M-lucNP. Furthermore, the magnitude of the T cell responses induced by Ub-G-lucNP and Ub-F-lucNP ranked in between Ub-M-lucNP and Ub-R-lucNP, and correlated with their respective in vivo halflifes (Fig. 5B,C). Effect of antigen stability on intramuscular DNA vaccination To investigate the possible role of antigen stability upon administration of DNA vaccines via other routes, we tested the Ub-M-lucNP, and Ub-R-lucNP DNA vaccines for their efficiency in the induction of T cell responses upon intramuscular DNA vaccination. Analogous to the results obtained after intradermal tattoo vaccination, the magnitude of the CD8+ T cell responses induced by intramuscular
87
5
IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY
vaccination with the stable Ub-M-lucNP variant exceeded that induced by the unstable Ub-R-lucNP variant (average CD8+ T cell responses of 1.2% and 0.2%, respectively, p=0.07), with detectable NP366-specific CD8+ T cell responses in 4 out of 5, and 1 out of 5 mice, respectively (Fig. 5D). This finding suggests that the mechanisms that govern the efficiency of presentation of intramuscularly expressed antigens are similar to those involved in intradermal antigen presentation. The effect of transgene design on vaccine immunogenicity. Most DNA vaccines that enter clinical trials encode a modified version of the antigen of interest. Depending on native biological function, viral antigens have been shuffled to prevent cellular transformation (e.g. HPV E6 and E7) (30), or to prevent other protein functions such as reverse transcriptase activity or protease activity (e.g. HIV) (31). Alternatively, fusion proteins have been generated with the aim to incorporate additional immunostimulatory signals (32), or strings of T cell epitopes (33). These types of manipulations that alter the 3-dimensional structure of the resulting proteins are likely to result in aberrant folding and thereby affect protein halflife. Based on the data presented above, vaccines encoding such rearranged gene products may be inferior in T cell induction. To directly test this hypothesis we constructed a shuffled version of lucNP (hereafter referred to as ulcNP), in which the NH2-terminal 50 amino acids (a.a.) have been removed from the aminoterminus and reincorporated into the protein 114 a.a. downstream within the sequence. Based on the luciferase X-ray structure, this change is expected to disrupt luciferase domain architecture and thereby affect protein stability. Consistent with this, immunoprecipitation of ulcNP and lucNP after pulse-chase labeling indicates that the in vitro half-life of the ‘scrambled’ ulcNP antigen is decreased as compared to the wild type lucNP (Fig. 6A, B). DNA vaccines encoding lucNP and ulcNP were then compared with regard to immunogenicity upon intradermal vaccination. Consistent with the notion that manipulations that affect antigen half-life influence vaccine immunogenicity, NP366-specific CD8+ T cell responses induced by vaccination with the scrambled ulcNP variant were inferior to those induced by vaccination with the unaltered lucNP vaccine (1.0% versus 4.0%, p=0.016). This indicates that modification of vaccine-encoded antigens by gene-scrambling can substantially affect the potency of DNA vaccines (Fig. 6C). Discussion Here, we have generated DNA vaccine variants by three types of design to assess whether antigen half-life affects the immunogenicity of DNA vaccines. Using these constructs for DNA vaccination we have shown that 1) CD8+ T cell responses against two signal peptide-encoded epitopes were highly reduced as compared to those against the same epitopes contained within the mature part of the same carrier protein, 2) 5 different ubiquitin-luciferase fusion proteins induced CD8+ T cell responses that correlated fully with their relative in vivo stability, and this correlation was observed for both i.d. and i.m. DNA vaccination, and 3) CD8+ 88
T cell responses against a shuffled and as a result instable variant of luciferase were substantially reduced as compared to those against the unmodified luciferase gene product. Of these experiments, the first and third line of experiments provide indirect evidence for an effect of antigen half-life on immunogenicity, and the second line of experiments provides direct proof for this link. In addition, we show that antigen stability directly affects cross-presentation of MHC class I-restricted epitopes, and that the bias towards stable proteins is observed in settings where an effect of antigen stability on the availability of CD4+ T cell help is excluded. Elegant work by Trombetta and colleagues has recently demonstrated how the induction of helper T cell responses by protein vaccination is correlated with resistance of the antigen to degradation within the lysosomal pathway of the APC (34). Here we complement this study by demonstrating that the induction of cytotoxic T cell responses by live vaccines is correlated with antigen stability within the cytosol. Given the fact that efficient T cell priming induced by DNA vaccines depends on CD4 T cell help (8,9,15,27), it seems plausible that optimal immunogens may in fact require a high stability in both the cytosol and within the lysosomal pathway, a notion that may be tested in further studies. While these experiments were primarily intended to examine how DNA vaccines may best be designed for the induction of cytotoxic T cell responses, it is interesting to consider how the results from these experiments fit with predictions made on the basis of existing models for antigen transfer from the antigen donating cell to APC (10-16) , and in particular whether the transferred material consists of peptides or full length protein. For such an analysis it is more informative to use the data from i.m. DNA vaccination than that of i.d. DNA vaccination for the following reason. It has previously been shown by Neefjes and coworkers that peptide fragments have an extremely short half-life in cells (22). Consequently, in cells where de novo production of antigen is transient, as is the case for intradermal DNA vaccination (19) , a rapid decay in protein synthesis of an instable antigen will be mirrored in an equally rapid decay in peptide levels. Inefficient induction of antigen-specific T cell responses in this setting could therefore either be explained by a low pool of protein or a low pool of peptides in the antigen-donating cell. However, upon i.m. DNA vaccination de novo production of proteins is highly constant (20). In this setting, a short half-life of the vaccine-encoded protein will affect steady state protein levels but is not expected to affect steady state peptide levels, as these will correlate with the level of steady state protein synthesis. Consequently, a beneficial effect of antigen stability on the immunogenicity of i.m. DNA vaccines (Fig. 5D) would not be expected if transfer from the antigen donating cell to the APC would involve the transport of peptide antigens. In contrast, if cross-presentation involves the transfer of proteasomal substrates from the antigen donating cell to the APC, a beneficial effect of antigen stability is expected, because an increased protein half-life would lead to a proportional increase in the pool of antigen available for transfer. Thus, the current data seem most consistent with a model where protein antigens rather than peptides are 89
5
IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY
transferred between antigen donating cell and APC following DNA vaccination. However, the evidence for this is clearly very indirect and a final resolution of the issue will likely only be provided by the direct in vivo visualization of the process of cross-presentation. The effect of antigen stability on vaccine immunogenicity that is shown here for i.d. and i.m. DNA vaccination is likely to also hold true for other vaccination strategies. Specifically, clinical strategies for vaccination with Modified Vaccinia Ankara (MVA)-based vectors by intramuscular injection or skin scarification have been shown to be partially dependent on cross-presentation (7). Likewise, the induction of T cell responses upon vaccination with recombinant adenoviral vectors is expected to occur through cross-presentation, since the CAR receptor that is required for efficient infection of cells by commonly used adenoviral vectors such as Ad5 is absent on dendritic cells (8). Thus, T cell induction in both the priming as well as the boosting phase in DNA-MVA and DNA-Ad5 prime boost regimens that are currently in clinical trials (35,36) are likely to benefit from vaccine designs that take antigen stability into account. In this respect it is noteworthy that the type of fusion gene vaccines used in current and planned large-scale vaccination trials, are expected or have in fact been shown to yield instable protein products (37) and will therefore likely form suboptimal immunogens. The current data imply that vaccine designs that disrupt protein folding should be avoided in future vaccine development, and that the in vivo stability of antigens should form a consideration when multiple candidate antigens are available. In addition, it seems attractive to develop more generally applicable strategies to promote the accumulation of otherwise instable antigens. For instance, judicious removal of surface exposed lysine residues may increase protein half-life by preventing ubiquitin conjugation and thus proteasomal degradation of the antigen (38,39) . Secondly, incorporation of specific protein domains has previously been shown to lead to high level protein accumulation, either as ER aggregates (40) or as cytosolic proteins, both with the possibility of pharmacological control (41). Based on the data presented here, evaluation of such strategies appears useful and may provide novel formats for vaccine transgene design. Materials and Methods Mice C57BL/6 mice, MHC class II-/- (42), TCRβ-/- mice (43), and B10F5 mice (44) were obtained from the experimental animal department of The Netherlands Cancer Institute (Amsterdam, The Netherlands). All animal procedures were performed after approval of the Experimental Animal Committee and in accordance with institutional and national guidelines. DNA vaccines DNA vaccines were generated by the introduction of fusion genes into pVAX, or pcDNA3.1 (Invitrogen, San Diego, CA). The fusion genes sNP-GFP-E7 and sE790
GFP-NP have been described elsewhere (15) and were cloned into the BamH I and Not I sites of pcDNA3.1. The fusion gene Ub-M-lucNP was generated by addition of the ubiquitin gene to the 5’ end of a lucNP-encoding vaccine (20). Briefly, two intermediate products were generated with the primers Ub-top: 5’GGGGGATCC TAAGCCACCATGCAGATC3’, and Ub-M-bottom: 5’CTTTATGTTTTTGGCGT CTTCCATACCCCCCCTCAAGCGC3’, and luc-M-top: 5’GCGCTTGAGGGGG GGTATGGAAGACGCCAAAAACATAAAG3’ with luc-NP-bottom: 5’CCCTT TGCGGCCGCTTACATGGCGTCCATGTTCTCGTTGCTGGCGATCTGCACG CCCACGGCGATCTTTCCGCCCTTC3’. A second round of amplification was performed with Ub-top and Luc-NP bottom primer. To generate Ub-R-lucNP, Ub-top was used together with Ub-R-bottom: 5’CTTTATGTTTTTGGCGTCTTCGCGAC CCCCCCTCAAGCGC3’, and luc-R-top: 5’GCGCTTGAGGGGGGGTCGCGA AGACGCCAAAAACATAAAG3’ with lucNP-bottom. The variants, Ub-F-lucNP, Ub-G-lucNP and Ub-S-lucNP were obtained by changing the first amino acid of the luciferase moiety by exchanging the Ub-X part with Ub-top and Ub-F-bottom: 5’AAACCCCCCCTCAAGCG3’, Ub-S-bottom 5’CTACCCCCCCTCAAGCG3’, and Ub-G bottom: 5’CCACCCCCCCTCAAGCG3’. The PCR products Ub-F, UbG and Ub-S were digested with BamH I, and ligated into pVAX-Ub-R-lucNP that had been digested with BamH I and Nru I. K14 driven Ub-M-lucNP and Ub-R-lucNP plasmids were constructed by replacing the CMV-IE promoter of the pVAX backbone with the K14 promoter (generously provided by J.H. Cho, Pohang University of Science and Technology, Korea). UlcNP was generated by relocating the first 150 nucleotides of the lucNP open reading frame into the SSpB1 restriction site at position 492. The fragment was generated with ulcNP1-50 top: 5’ATATTGTACAAAGACGCCAAAAACATAAAG AAAGG3’ and ulcNP1-50 bottom: 5’ATATATGTACACGTCCACCTCGATATGTG CATC3’. A Kozak sequence and start codon were inserted at the 5’ end of UlcNP by PCR reaction with ulcNP top: 5’ATATATGTACACGTCCACCTCGATATGT GCATC3’ and ulcNP bottom: 5’ATATTGTACAAAGACGCCAAAAACATAAA GAAAGG3’. All sequences were confirmed by sequence analysis. DNA was grown in DH5α and isolated with endotoxin-free DNA purification kit (Qiagen, Hilden, Germany). DNA vaccines for intradermal application were dissolved in water, DNA vaccines for intramuscular administration were dissolved in PBS (Gibco). DNA vaccination and intravital imaging Intradermal vaccination by DNA tattoo and intravital analysis of protein expression was performed as described previously (20). Briefly, mice were anaesthetized and a volume of 10 µl of a 2 µg/µl DNA solution in water was applied to the shaved skin of the hind leg. The DNA vaccine was applied to the skin by a 20 s tattoo with a sterile disposable 11-needle bar oscillating at a frequency of 100Hz. For intravital imaging, mice were anaesthetized with isofluorane (Abbott Laboratories, Queensborough, UK). An aqueous solution of the substrate luciferin (150 mg/kg,
91
5
IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY
Xenogen, Alameda, CA) was injected intraperitoneally and 18 min. later the luminescence produced by active luciferase was acquired during 30s in an IVIS system200 CCD camera (Xenogen). Signal intensity was quantified as the sum of all detected light within the region of interest. Detection of NP366 and E749-specific T cells in peripheral blood Peripheral blood lymphocytes (PBL) were stained with phycoerythrin-conjugated anti-CD8β (BD Pharmingen, San Diego, CA) plus allophycocyanin-conjugated H2Db/NP366-tetramers or allophycocyanin-conjugated H-2Db/E749-tetramers, at room temperature for 15 min in PBA (PBS, 0.5% BSA and 0.02% sodium azide). Cells were washed three times in PBA and analyzed by flow cytometry. Live cells were selected based on 7-AAD exclusion. Analysis of in vitro protein stability by proteasome inhibition. COS cells seeded in 10mm dishes with a confluency of 80% were transfected with 15 µg pVAX encoding Ub-M-lucNP, Ub-R-lucNP, together with DEAE dextran, for 30 min. at 37°C, and 80µM chloroquine was added for 2.5 h. DNA-chloroquine containing medium was aspirated and cells were treated for 2.5 min with 10% DMSO in DMEM culture medium (GIBCO) containing 10% FBS. The DMSO/ medium mixture was aspirated and cells were cultured for 24 h at 37°C. To ensure equal transfection efficiencies for samples, cells were harvested and seeded into 6 well plates in triplicates. 48 h post transfection, culture medium was aspirated and fresh medium with or without 10 µM MG132 (z-Leu-Leu-Leu-al, Sigma) was added. After 4 h at 37°C, cells were trypsinized, harvested, washed twice with PBS and resuspended in 20 µl lysis buffer (Promega). Following a 20 min incubation at RT, 5 µl of cell lysate was harvested, 20 μl substrate buffer (Promega) was added, and luciferase activity was measured. Analysis of in vitro protein stability by pulse-chase analysis COS cells at 80% confluency were transfected with 10µg DNA encoding Ub-M-lucNP, Ub-R-lucNP, lucNP or ulcNP using FuGene6, according to the manufacturer’s protocol. The cells were starved in Met/Cys-free DMEM medium (Gibco) for 60 minutes prior to a 1h pulse labeling with 35S-labeled Met/Cys in DMEM. Labeled cells were washed and subsequently cultured in Met/Cyscontaining DMEM for the indicated time points. After cell lysis, luciferase variants were immunoprecipitated, and analyzed by SDS-PAGE. Protein quantification was performed by phospho-imager analysis (FLA3000, Fujifilm). The images were quantitated using TINA 2.09 analysis software.
92
Bins et al. Figure 1
1a
sE7-GFP-NP
sNP-GFP-E7
Signal GFP mature peptide protein sE7-GFP-NP
Signal peptide
E7 CTL Signal epitope
1b A CD8
peptide
epitope
1b
CD8
NP CTL Signal epitope
NP CTL
peptide C
sE7-GFP-NP
104 NP CTL
0.2%
epitope
3.2%
NP CTL epitope
sNP-GFP-E7
104
104
0.2%
100
sE7-GFP-NP
104
100
3.2%
100
104
Db-NP tetramer 104 100
CD8
E7 CTL GFP mature epitope protein 16
104
5.3% 100
104
100
8
100
5.3% 104
100
100
100
104
104
100
100
sE7-GFP-NP
4
0
sNP-GFP-E7
sE7-GFP-NP
1 5
104
D -E7 tetramer 100
sNP-GFP-E7
6
2 6
b
100
2 14 0 8
3
0.1%
100
4 16
5 2
0.1%
104
6
6 4
Db-NP tetramer 104
E7 CTL epitope
14
C
100
CD8
sNP-GFP-E7
GFP mature epitope protein
sNP-GFP-E7
104 E7 CTL
Bins et al. Figure 1
GFP mature protein
% NP-specific CD8 T cells % E7-specific CD8 T cells % NP-specific CD8 T cells % E7-specific CD8 T cells
1a
104
0 4
sNP-GFP-E7 sE7-GFP-NP
3 2 1 0
sNP-GFP-E7 sE7-GFP-NP
D -E7 tetramer b
B
C
Figure 1. Antigen location affects the immunogenicity of DNA vaccines. C57BL/6 mice were immunized at day 0, 3, and 6 by means of i.d. DNA tattoo with 20 µg pcDNA3.1-sNP-GFP-E7 or pcDNA3.1-sE7-GFP-NP. (A) At day 12 post-vaccination, peripheral blood was drawn and antigen-specific CD8+ T cell responses were measured ex vivo by combined anti-CD8 and Db-NP366 tetramer (top), or Db-E749 tetramer (bottom) staining. Schematic representation of sNPGFP-E7 and sE7-GFP-NP vaccine insert designs (B) Flow cytometric analysis of peripheral blood cells of representative mice for the two groups. (C) Symbols represent antigen-specific CD8+ T cell responses of individual mice, bars indicate averages of 5 mice. Result is representative of two independent experiments.
93
5
IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY Bins et al. Figure 2 Bins et al. Figure Bins 2et al. Figure 2
A
250000 250000
250000
250000 250000
200000
200000 200000
200000
200000 200000
150000 150000
150000
150000 150000
100000 100000
100000
100000 100000
100000
0
C
2c 2c
50000
-
0
50000 0
+ -
50000
-
+
0
50000
+
-
0
0 + -
D
Ub-MUb-M- Ub-M2d LucNP LucNP LucNP Ub-M-LucNP
0 15 30 1h 2h 4h 6h 0 15 30 1h 0 15 2h 4h 306h 1h 2h 4h 6h
NP366
NP366
50000
relative activity (% of max.)
150000
Luciferase activity
250000
50000
2c
Luciferase Luciferase Luciferase NP366
Ub-RUb-R-Ub-R-LucNP Ub-RLucNP LucNP LucNP
-
+
100 2d 2d 10
1
0
+
relative activity (% of max.)
B 2b 2b
Ubiquitin
relative activity (% of max.)
Ubiquitin
Luciferase activity
2b
2a 2a
Luciferase activity
2a
UbMlucNP M UbMlucNPUbMlucNP M M R UbRlucNP R R UbRlucNPUbRlucNP Ubiquitin
100
10
1 20
100
10
1 0
2 4 64 6 hours post labelling hours post labelling hours post labelling 42
Figure 2. Structure and in vitro stability of Ub-M-lucNP and Ub-R-lucNP fusion proteins. (A) Schematic representation of the design of the Ub-M-lucNP and Ub-RlucNP fusion gene constructs. The arrowhead indicates the cotranslational cleavage site of ubiquitin by Ub-specific proteases (USPs). M and R indicate the methionine and arginine starting amino acids. (B) COS cells were transiently transfected with Ub-M-lucNP (left) or Ub-R-lucNP (right), and incubated with the proteasomal inhibitor MG132 (z-Leu-Leu-Leu-al.) for 4 h (+), or left untreated (-). After cell lysis, luciferase activity of the lysate was determined. Results are representative of three independent experiments. (C) COS cells that were transiently transfected with Ub-M-lucNP or Ub-R-lucNP were starved in Met/Cys-free DMEM medium for 60 min prior to a 1h pulse labeling with 35S-labeled Met/Cys in DMEM. Labeled cells were washed and subsequently cultured in Met/Cys-containing DMEM for the indicated time points. After cell lysis, luciferase variants were immunoprecipitated, and samples were analyzed by SDS-PAGE. Based on the molecular weight of the observed proteins (62 kD), removal of the Ub moiety is complete at all time points examined (data not shown). (D) Graphical representation of an automated quantification of C. 94
6
et al. Figure 3 Bins etBins al. Figure 3
3a 3aA 100000100000 Luc. activity (rlu)
10000 10000 1000 1000 100
10
10 024
0
20
20
15
15
10
10
5
5
0
0
% NP spec. CD8 T cells
% NP spec. CD8 T cells
3b 3bB
24 48
48 72 96 72 time (hrs.) time (hrs.)
96
C
9 11 15 13 15 7 11 13 daysvaccination post vaccination days post 97
20
20
15
15
10
10
5
5
0
0
% NP spec. CD8 T cells
100
% NP spec. CD8 T cells
Luc. activity (rlu)
1000000 1000000
7
97
9 11 15 13 15 11 13 daysvaccination post vaccination days post
Figure 3. Antigen stability promotes vaccine immunogenicity (A) In vivo kinetics of luciferase activity upon intradermal DNA vaccination with Ub-M-lucNP or Ub-R-lucNP. Mice were injected once by DNA tattoo with 20 µg Ub-M-lucNP (♦), or Ub-R-lucNP (▲). At the indicated time points post vaccination, luciferase expression was determined by intravital imaging. It is noted that the difference in Iuciferase activity resulting from a differential half life is expected to be maximal at late timepoints at which de novo antigen production is low or has ceased, and will be suppressed at early timepoints during which antigen production is still rising (B) Immunogenicity of Ub-M-lucNP and Ub-R-lucNP DNA vaccines. Mice were vaccinated with 20 µg Ub-M-lucNP (♦) or Ub-R-lucNP (▲) at day 0, 3, and 6. At the indicated time points, peripheral blood was drawn and antigen-specific CD8+ T cell responses were determined ex vivo by Db-NP366 tetramer staining. Symbols represent CD8+ T cell responses of individual mice, bars indicate group averages. Result is representative of 3 independent experiments.
95
5
IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY
5 5 4 4
% NP spec. CD8 T cells % NP spec. CD8 T cells
5 5 4 4
% NP spec. CD8 T cells % NP spec. CD8 T cells
4a A4a
3 3 2 2
3 3 2 2 1 1 0 0
8 8 10 10 12 12 14 14 16 16 days after vaccination days after vaccination
1 1 0 0
8 8 10 10 12 1214 14 16 16 days after vaccination days after vaccination
15 15
% F5 cells with CFSE loss % F5 cells with CFSE loss
4b B4b
Bins et Figure al. Figure Bins et al. 4 4
10 10 5 5 0 0 0,5Ub-MUb-R- control control 0,5Ub-MUb-RLucNP LucNP LucNP plasmid plasmid LucNP
% NP spec. CD8 T cells % NP spec. CD8 T cells
C4c 4c
r
1 1 0.80.8 0.60.6 0.40.4
0.20.2 0 0
8 8 10 10 12 12 days after vaccination days after vaccination
% NP spec. CD8 T cells % NP spec. CD8 T cells
r
m
m
1 1 0.80.8 0.60.6 0.40.4
0.20.2 0 0
8 8 10 10 12 12 days after vaccination days after vaccination
Figure 4. Antigen stability promotes vaccine immunogenicity through enhanced cross-presentation. (A) Mice were vaccinated at day 0, 3, and 6 by with 20 µg pVAX(K14)-Ub-M-lucNP, or pVAX(K14)-Ub-R-lucNP, and induction of NP366specific CD8+ T cell responses was measured by Db-NP366 tetramer staining. The percentage NP366-specific CD8+ T cells at various time points post vaccination is plotted. (B) Antigen stability affects MHC class I-restricted antigen presentation in the draining lymph node. Mice were vaccinated with 20 µg Ub-M-lucNP, Ub-RlucNP or a control plasmid. At day 7 post vaccination, 2x106 NP366-specific CFSE labeled F5 T cells that were depleted for CD4+ cells by magnetic bead separation with a CD4-specific antibody were injected into TCRβ-/- mice. Four days later, draining lymph nodes (♦/ ▲ / ✴) and non-draining lymph nodes (◊ / ∆ / x) were excised and analyzed for dividing F5 T cells as determined by the percentage of CFSE-dull Db-NP366 tetramer positive cells. (C) MHC-II -/- mice were vaccinated at day 0, 3, and 6 by DNA tattooing with 20 µg of Ub-M-lucNP (♦) or Ub-R-lucNP (▲). At day 8, 10 and 11 peripheral blood was drawn and antigen-specific CD8+ T cell responses were determined ex vivo by Db-NP366 tetramer staining. 96
Bins et al. Bi
C
5 0
10 5
3
15 20 10 15
S
S
M M S 105R F R G F G M 05 F G 6 3 9R 6 129 1512 15 0 antigen antigen half-lifehalf-life (hrs.) (hrs.) 3 6 9 12 15 antigen half-life (hrs.)
5
5d
25 30 20 25 15 20 10 15 5 10 0 5
R F R G FM GS M S 0
5
R F G M S 5
NP CD8 spec.TCD8 % NP % spec. cellsT cells
10
15
0
NP CD8 spec.TCD8 % NP % spec. cellsT cells
15
20
% NP spec. CD8 T cells
% NP spec. CD8 T cells
5c 20
20
B 5d 5d
NP CD8 spec.TCD8 % NP % spec. cellsT cells
5c A5c
25
30
% NP spec. CD8 T cells
100000 100000 100001000000 10000 1000 100000 1000 100 10000 100 1000 10 10 100 1 1 0 10 24 0 48 2472 4896 7212096144120 144 time (hrs.) time (hrs.) 1 0 24 48 72 96 120 144 time (hrs.)
NP CD8 spec.TCD8 % NP % spec. cellsT cells
% NP spec. CD8 T cells
5b 5b 5b30
Luc. activity Luc. activity (rlu) (rlu)
Luc. activity (rlu)
5a 5a 1000000 5a1000000
Bi
5
* * 4 4 4 5 5 * * 3 3 3 3 4 4 2 2 2 2 3 3 1 1 1 1 2 2 0 0 0 0 4 51 4 6 5 7 6 8 7 9 8 9 4 51 4 6 5 7 6 8 DNA lastinjec DN 0 lastafter day after lastafter DNA lastinjection DNA injectionday afterday 0day 4 5 6 4 5 6 7 8 9 day after last DN day after last DNA injection
4
*
*
Figure 5. Antigen immunogenicity correlates with protein half-life. (A) In vivo kinetics of luciferase activity upon vaccination with Ub-X-lucNP variants. Mice were tattooed once with 20 µg DNA vaccine, encoding Ub-S-lucNP (●),Ub-MlucNP (♦), Ub-G-lucNP (-), Ub-F-lucNP (■) or Ub-R-lucNP (▲). At indicated time points post vaccination, luciferase expression was determined by intravital imaging. (B) CD8+ T cell responses upon vaccination with Ub-X-lucNP variants. Mice were vaccinated at day 0, 3, and 6 by DNA tattoo with 20 µg of the UbS-lucNP (●),Ub-M-lucNP (♦), Ub-G-lucNP (x), Ub-F-lucNP (■) or Ub-R-lucNP (▲) DNA vaccines. Peripheral blood was drawn and antigen-specific CD8+ T cell responses at the peak of the response were determined ex vivo by Db-NP366 tetramer staining. The percentage of NP366 specific CD8+ T cells in each mouse at the peak of the response is plotted, bars depict group averages. (C) The in vivo half-life calculated from the data presented in (A) and the average peak Db-NP366 CD8+ T cell response for each stability variant plotted against each other. Error bars represent the standard deviation of antigen-specific CD8+ T cell responses. 97
5
% NP spec. C
72 96 120 144 time (hrs.)
D S R
F G
6
9
M
12
15
15 10 5 0
R F G M S IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY
5d 5
% NP spec. CD8 T cells
48
% NP spec. CD8 T cells
24
20
*
4 3 2 1 0
4 5 6 7 8 9 day after last DNA injection
antigen half-life (hrs.)
5
*
4 3 2 1 0
4 5 6 7 8 9 day after last DNA injection
(D) Mice were vaccinated at day 0, 14, and 28 by i.m. injection of 50 µl (100 µg) DNA, encoding Ub-M-lucNP (◊) or Ub-R-lucNP (∆). Because antigen expression upon i.m. vaccination is highly variable, presumably due to variability in delivery of the DNA vaccine to the target site, luciferase expression was determined by intravital imaging at day 7 post vaccination, and three mice within each group with the lowest luciferase expression were excluded from the experiment. At day 7 post second vaccination one additional mouse was excluded from each group in a similar fashion. Of the 5 remaining mice in each group NP366-specific CD8+ T cell responses were determined at day 5, 6, 7, and 8 after the third vaccination by Db-NP366 tetramer staining.
6b
lucNP LucNP
0 15 30 1h 2h0 4h15 30 1h 2h 4h
ulcNP
A
6 4 2 0
5
% NP spec. CD8 T cells
6c
8
ulcNP
6b
10 0
10
2
hours post labelling
B
4 2
4 2
0 4
9
11
0 135 157
9
11
2
hours post labelling
4 2
0 0 9 11 135 157 9 11 13 15 13 5 15 7 days post vaccination days post vaccination days post vaccination days post vaccination
98
7
Bins et al. Figu
100
Figure 6. Adverse effect of gene shuffling on vaccine immunogenicity. (A) SDS10 of a anti-luciferase 10 immunoprecipitates PAGE analysis 10 of COS cells that were transiently transfected with pVAX encoding either lucNP or ulcNP. (B) Graphical 8 8 8 representation of an automated quantification of lucNP (♦) and ulcNP (▲) from 6 A. 6 6 % NP spec. CD8 T cells
6c
% NP spec. CD8 T cells
10
UlcNP
100
relative activity (% of max.)
lucNP
6a
% NP spec. CD8 T cells
6a
relative activity (% of max.)
Bins et al. Figure 6
4
relative activity (
0 15 30 1h 2h 4h
ulcNP
% NP spec. CD8 T cells
% NP spec. CD8 T cells
8 6 4 2 0
C
5
7
9
11
13
0
2
4
hours post labelling
10
6c
10
10 8 6 4 2 0 5
15
7
9
11
13
15
days post vaccination
days post vaccination
A
0 5
10
15
20
10
15 10
5 0
5
Day after start of vaccination
B
5 0 5
10
10
15
15
20
Day after start of vaccination
Day after start of vaccination
Luciferase activity (rlu)
Suppl. 1b
20
B
% NP spec. CD8+ T cells
5
% NP spec. CD8+ T cells
10
15
Supplementary Figure 1. A K14 driven 1000000 1000000 lucNP DNA vaccine induces NP366specific CD8+ T cell responses. (A) NP366 100000 specific100000 CD8+ T cell responses induced by vaccination with a CMV-IE driven lucNP 10000 DNA vaccine 10000 (right panel) and K14 driven lucNP DNA vaccine (left panel). Mice received 1-minute DNA tattoos at day 0, 3 1000 1000 and 6 and NP366-specific CD8+ T cells were enumerated directly ex vivo at the indicated 100 100 time points. Symbols represent CD8+ T cell cmvk14cmv- mice, k14- bars indicate responses of individual LucNP LucNP LucNP LucNP group averages. (B) Luciferase expression induced by K14 driven DNA vaccines is 10 to 50 fold lower than that induced by CMV-IE driven DNA vaccines. One day after DNA tattooing with either CMVIE or K14 driven lucNP vaccines, luciferase activity was measured by intravital imaging. Error bars indicate standard deviation (n=5).
Luciferase activity (rlu)
Suppl. 1b
Suppl. 1a
15
% NP spec. CD8+ T cells
Suppl. 1a
% NP spec. CD8+ T cells
(C) Immunogenicity of lucNP and ulcNP DNA vaccines. Mice were vaccinated with 20 µg lucNP (●) or ulcNP (♦) at day 0, 3, and 6. At the indicated time points, peripheral blood was drawn and antigen-specific CD8+ T cell responses were + determined ex vivo by Db-NP366 tetramer staining. Symbols represent T cell Figure 1 Bins et al.CD8 Supplementary responses of individual mice, bars indicate group averages.
99
5
15 10 5 0 5
1
Day after
IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY
1. Gallucci,S. and P.Matzinger. 2001. Danger signals: SOS to the immune system. Curr Opin Immunol. 13:114-119. 2. Rock,K.L. and L.Shen. 2005. Cross-presentation: underlying mechanisms and role in immune surveillance. Immunol Rev. 207:166-83.:166-183. 3. Bevan,M.J. 1976. Minor H antigens introduced on H-2 different stimulating cells cross-react at the cytotoxic T cell level during in vivo priming. J Immunol. 117:2233-2238. 4. Sant,A.J. 1994. Endogenous antigen presentation by MHC class II molecules. Immunol Res. 13:253-267. 5. Townsend,A.R., F.M.Gotch, and J.Davey. 1985. Cytotoxic T cells recognize fragments of the influenza nucleoprotein. Cell. 42:457-467. 6. Whitton,J.L. and M.B.Oldstone. 1989. Class I MHC can present an endogenous peptide to cytotoxic T lymphocytes. J Exp Med. 170:1033-1038. 7. Shen,X., S.B.Wong, C.B.Buck, J.Zhang, and R.F.Siliciano. 2002. Direct priming and cross-priming contribute differentially to the induction of CD8+ CTL following exposure to vaccinia virus via different routes. J Immunol. 169:42224229. 8. Tillman,B.W., T.D.de Gruijl, S.A.Luykx-de Bakker, R.J.Scheper, H.M.Pinedo, T.J.Curiel, W.R.Gerritsen, and D.T.Curiel. 1999. Maturation of dendritic cells accompanies high-efficiency gene transfer by a CD40-targeted adenoviral vector. J Immunol. 162:6378-6383. 9. Cho,J.H., J.W.Youn, and Y.C.Sung. 2001. Cross-priming as a predominant mechanism for inducing CD8(+) T cell responses in gene gun DNA immunization. J Immunol. 167:5549-5557. 10. Doe,B., M.Selby, S.Barnett, J.Baenziger, and C.M.Walker. 1996. Induction of cytotoxic T lymphocytes by intramuscular immunization with plasmid DNA is facilitated by bone marrow-derived cells. Proc Natl Acad Sci U S A. 93:85788583. 11. Corr,M., A.von Damm, D.J.Lee, and H.Tighe. 1999. In vivo priming by DNA injection occurs predominantly by antigen transfer. J Immunol. 163:47214727. 12. Basta,S., R.Stoessel, M.Basler, M.van den Broek, and M.Groettrup. 2005. Cross-presentation of the long-lived lymphocytic choriomeningitis virus nucleoprotein does not require neosynthesis and is enhanced via heat shock proteins. J Immunol. 175:796-805. 13. Norbury,C.C., S.Basta, K.B.Donohue, D.C.Tscharke, M.F.Princiotta, P.Berglund, J.Gibbs, J.R.Bennink, and J.W.Yewdell. 2004. CD8+ T cell crosspriming via transfer of proteasome substrates. Science. 304:1318-1321.
100
14. Shen,L. and K.L.Rock. 2004. Cellular protein is the source of cross-priming antigen in vivo. Proc Natl Acad Sci U S A. 101:3035-3040. 15. Wolkers,M.C., N.Brouwenstijn, A.H.Bakker, M.Toebes, and T.N.Schumacher. 2004. Antigen bias in T cell cross-priming. Science. 304:13141317. 16. Binder,R.J. and P.K.Srivastava. 2005. Peptides chaperoned by heat-shock proteins are a necessary and sufficient source of antigen in the cross-priming of CD8+ T cells. Nat Immunol. 6:593-599. 17. Blachere,N.E., R.B.Darnell, and M.L.Albert. 2005. Apoptotic cells deliver processed antigen to dendritic cells for cross-presentation. PLoS Biol. 3:e185. 18. Serna,A., M.C.Ramirez, A.Soukhanova, and L.J.Sigal. 2003. Cutting edge: efficient MHC class I cross-presentation during early vaccinia infection requires the transfer of proteasomal intermediates between antigen donor and presenting cells. J Immunol. 171:5668-5672. 19. Neijssen,J., C.Herberts, J.W.Drijfhout, E.Reits, L.Janssen, and J.Neefjes. 2005. Cross-presentation by intercellular peptide transfer through gap junctions. Nature. 434:83-88. 20. Bins,A.D., A.Jorritsma, M.C.Wolkers, C.F.Hung, T.C.Wu, T.N.Schumacher, and J.B.Haanen. 2005. A rapid and potent DNA vaccination strategy defined by in vivo monitoring of antigen expression. Nat Med. 11:899-904. 21. Lyko,F., B.Martoglio, B.Jungnickel, T.A.Rapoport, and B.Dobberstein. 1995. Signal sequence processing in rough microsomes. J Biol Chem. 270:1987319878. 22. Reits,E., J.Neijssen, C.Herberts, W.Benckhuijsen, L.Janssen, J.W.Drijfhout, and J.Neefjes. 2004. A major role for TPPII in trimming proteasomal degradation products for MHC class I antigen presentation. Immunity. 20:495-506. 23. Bachmair,A., D.Finley, and A.Varshavsky. 1986. In vivo half-life of a protein is a function of its amino-terminal residue. Science. 234:179-186. 24. Tasaki,T., L.C.Mulder, A.Iwamatsu, M.J.Lee, I.V.Davydov, A.Varshavsky, M.Muesing, and Y.T.Kwon. 2005. A family of mammalian E3 ubiquitin ligases that contain the UBR box motif and recognize N-degrons. Mol Cell Biol. 25:71207136. 25. Donnelly,J.J., B.Wahren, and M.A.Liu. 2005. DNA vaccines: progress and challenges. J Immunol. 175:633-639. 26. Heath,W.R., G.T.Belz, G.M.Behrens, C.M.Smith, S.P.Forehan, I.A.Parish, G.M.Davey, N.S.Wilson, F.R.Carbone, and J.A.Villadangos. 2004. Crosspresentation, dendritic cell subsets, and the generation of immunity to cellular antigens. Immunol Rev. 199:9-26.:9-26.
101
5
IN VIVO ANTIGEN STABILITY AFFECTS VACCINE EFFICACY
27. Maecker,H.T., D.T.Umetsu, R.H.DeKruyff, and S.Levy. 1998. Cytotoxic T cell responses to DNA vaccination: dependence on antigen presentation via class II MHC. J Immunol. 161:6532-6536. 28. Wang,J.C. and A.M.Livingstone. 2003. Cutting edge: CD4+ T cell help can be essential for primary CD8+ T cell responses in vivo. J Immunol. 171:63396343. 29. Levy,F., N.Johnsson, T.Rumenapf, and A.Varshavsky. 1996. Using ubiquitin to follow the metabolic fate of a protein. Proc Natl Acad Sci U S A. 93: 4907-4912. 30. Smahel,M., P.Sima, V.Ludvikova, and V.Vonka. 2001. Modified HPV16 E7 Genes as DNA Vaccine against E7-Containing Oncogenic Cells. Virology. 281: 231-238. 31. Burgers,W.A., J.H.van Harmelen, E.Shephard, C.Adams, T.Mgwebi, W.Bourn, T.Hanke, A.L.Williamson, and C.Williamson. 2006. Design and preclinical evaluation of a multigene human immunodeficiency virus type 1 subtype C DNA vaccine for clinical trial. J Gen Virol. 87:399-410. 32. Locher,C.P., N.W.Soong, R.G.Whalen, and J.Punnonen. 2004. Development of novel vaccines using DNA shuffling and screening strategies. Curr Opin Mol Ther. 6:34-39. 33. Goonetilleke,N., S.Moore, L.Dally, N.Winstone, I.Cebere, A.Mahmoud, S.Pinheiro, G.Gillespie, D.Brown, V.Loach, J.Roberts, A.Guimaraes-Walker, P.Hayes, K.Loughran, C.Smith, J.De Bont, C.Verlinde, D.Vooijs, C.Schmidt, M.Boaz, J.Gilmour, P.Fast, L.Dorrell, T.Hanke, and A.J.McMichael. 2006. Induction of multifunctional human immunodeficiency virus type 1 (HIV-1)specific T cells capable of proliferation in healthy subjects by using a primeboost regimen of DNA- and modified vaccinia virus Ankara-vectored vaccines expressing HIV-1 Gag coupled to CD8+ T-cell epitopes. J Virol. 80:4717-4728. 34. Delamarre,L., R.Couture, I.Mellman, and E.S.Trombetta. 2006. Enhancing immunogenicity by limiting susceptibility to lysosomal proteolysis. J Exp Med. 203:2049-2055. 35. McShane,H., A.A.Pathan, C.R.Sander, S.M.Keating, S.C.Gilbert, K.Huygen, H.A.Fletcher, and A.V.Hill. 2004. Recombinant modified vaccinia virus Ankara expressing antigen 85A boosts BCG-primed and naturally acquired antimycobacterial immunity in humans. Nat Med. 10:1240-1244. 36. Xin,K.Q., N.Jounai, K.Someya, K.Honma, H.Mizuguchi, S.Naganawa, K.Kitamura, T.Hayakawa, S.Saha, F.Takeshita, K.Okuda, M.Honda, D.M.Klinman, and K.Okuda. 2005. Prime-boost vaccination with plasmid DNA and a chimeric adenovirus type 5 vector with type 35 fiber induces protective immunity against HIV. Gene Ther. 12:1769-1777.
102
37. Nkolola,J.P., E.G.Wee, E.J.Im, C.P.Jewell, N.Chen, X.N.Xu, A.J.McMichael, and T.Hanke. 2004. Engineering RENTA, a DNA prime-MVA boost HIV vaccine tailored for Eastern and Central Africa. Gene Ther. 11:1068-1080. 38. Rodriguez,M.S., J.M.Desterro, S.Lain, D.P.Lane, and R.T.Hay. 2000. Multiple C-terminal lysine residues target p53 for ubiquitin-proteasome-mediated degradation. Mol Cell Biol. 20:8458-8467. 39. Khamsri,B., M.Fujita, K.Kamada, A.Piroozmand, T.Yamashita, T.Uchiyama, and A.Adachi. 2006. Effects of lysine to arginine mutations in HIV-1 Vif on its expression and viral infectivity. Int J Mol Med. 18:679-683. 40. Rivera,V.M., X.Wang, S.Wardwell, N.L.Courage, A.Volchuk, T.Keenan, D.A.Holt, M.Gilman, L.Orci, F.Cerasoli, Jr., J.E.Rothman, and T.Clackson. 2000. Regulation of protein secretion through controlled aggregation in the endoplasmic reticulum. Science. 287:826-830. 41. Banaszynski,L.A., L.C.Chen, L.A.Maynard-Smith, A.G.Ooi, and T.J.Wandless. 2006. A rapid, reversible, and tunable method to regulate protein function in living cells using synthetic small molecules. Cell. 126:995-1004. 42. Grusby,M.J., R.S.Johnson, V.E.Papaioannou, and L.H.Glimcher. 1991. Depletion of CD4+ T cells in major histocompatibility complex class II-deficient mice. Science. %20;253:1417-1420. 43. Mombaerts,P., J.Arnoldi, F.Russ, S.Tonegawa, and S.H.Kaufmann. 1993. Different roles of alpha beta and gamma delta T cells in immunity against an intracellular bacterial pathogen. Nature. 365:53-56. 44. Mamalaki,C., J.Elliott, T.Norton, N.Yannoutsos, A.R.Townsend, P.Chandler, E.Simpson, and D.Kioussis. 1993. Positive and negative selection in transgenic mice expressing a T-cell receptor specific for influenza nucleoprotein and endogenous superantigen. Dev Immunol. 3:159-174.
103
5
C H A P T E R
6
Intravital imaging of fluorescent markers and FRET probes by DNA tattooing. Adriaan D. Bins1*, Jacco van Rheenen1,2*§, Kees Jalink3, Jonathan R. Halstead4, Nullin Divecha4, David M. Spencer5, John B.A.G. Haanen1 & Ton N.M. Schumacher1§ Department of Immunology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands 1
Department of Anatomy and Structural Biology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461, USA 2
Department of Cell Biology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands 3
Department of Cellular Biochemistry, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands 4
Department of Immunology, Baylor College of Medicine, Texas Medical Center, One Baylor Plaza, BCMM-M929, Houston, TX 77030, USA 5
These authors contributed equally to this work
*
Corresponding authors
§
Abstract Advances in fluorescence microscopy and mouse transgenesis have made it possible to image molecular events in living animals. However, the generation of transgenic mice is a lengthy process and intravital imaging requires specialized knowledge and equipment. Here, we report a rapid and undemanding intravital imaging method using generally available equipment. By DNA tattooing we transfect keratinocytes of living mice with DNA encoding fluorescent biosensors. Subsequently, the behavior of individual cells expressing these biosensors can be visualized within hours and using conventional microscopy equipment. Using this “instant transgenic” model in combination with a corrected coordinate system, we followed the in vivo behavior of individual cells expressing either FRETor location-based biosensors for several days. The utility of this approach was demonstrated by assessment of in vivo caspase-3 activation upon induction of apoptosis. This “instant skin transgenic” model can be used to follow the in vivo behavior of individual cells expressing either FRET- or location-based probes for several days after tattooing and provides a rapid and inexpensive method for intravital imaging in murine skin. 104
Background The introduction of fluorescent proteins (FP) and the ability to visualize molecular events using fluorescence microscopy have been of substantial value for understanding intracellular signaling events. FP-based biosensors, and especially those based on Fluorescent Resonance Energy Transfer (FRET), can report on cellular processes such as apoptosis or cell division, the levels of second messengers such as Ca2+, PtdIns(4,5)P2 and cAMP, or the activation status of proteins such as RhoA [1]. To date, the vast majority of analyses utilizing FP-based biosensors have been carried out on cells in tissue culture. However, for cellular processes that are controlled by cellular interactions or by the microenvironment, it would clearly be preferable to perform such analyses in vivo. Indeed, the small number of studies that have analyzed the behavior of single cells in vivo, have been informative with regard to the intracellular interactions of immune cells and early events in tumors formation [2-5]. Intravital images of whole mice or tumors can readily be acquired with macroscopical resolution using simple instrumentations (e.g. LED flashlight, filters and a digital camera [6]). However, for imaging with subcellular resolution, sophisticated equipment and expertise (e.g. two-photon microscopes [7] or wholemouse imaging systems [8]) are required, which are only available in few specialized laboratories. Furthermore, the generation of biosensor-transgenic mice is a lengthy process and thereby hampers the use of this technique in settings where a number of different biosensors are tested or combinations of biosensors are required. A few studies have previously described fast and simple techniques to target selective genes to specific sites. For example, Li and Hoffman have targeted reporter genes to hair follicles in mice using liposomes [9]. However, it will be important to extend such approaches to other cell types. Here we present a rapid and inexpensive DNA tattoo method to target FP-genes to keratinocytes in the skin of mice. The behavior of these keratinocytes expressing fluorescent or FRET markers were followed for multiple days with subcellular resolution using a standard confocal microscope. The utility of this tattoo-approach was demonstrated by in vivo imaging of caspase3 activation upon apoptosis induction. Results and Discussion Prior experiments have demonstrated the feasibility of introducing transgenes in murine skin cells by use of a rapidly oscillating tattoo machine [10]. In an effort to determine the number of cells that express the introduced transgene upon this type of skin application (see Methods), we tattooed a small area (10x15 mm) of the abdominal skin of mice with DNA encoding a GFP reporter protein driven by the CMV immediate early (CMV-IE) promoter. 3 hours thereafter, mice were anesthetized and analyzed for DNA tattoo-induced GFP expression by intravital imaging using a conventional confocal microscope. Expression of the GFP reporter gene was observed in approximately 160 cells per 10 mm2 tattooed skin, showing that the transfection efficiency of DNA tattooing was sufficient for imaging purposes. The majority of fluorescent cells obtained after a GFP-encoding DNA tattoo were keratinocytes (see Additional file 1). 105
6
INTRAVITAL IMAGING OF FLUORESCENT MARKERS AND FRET PROBES BY DNA TATTOOING.
The uncorrected GFP images showed considerable autofluorescence of skin structures, in particular hair follicles, which was characterized by a substantially wider excitation and emission bandwidth. To discriminate the cells expressing GFP from these autofluorescent structures, autofluorescence was detected by imaging at 405nm excitation and 410-440 nm emission. The resulting autofluorescence image was then subtracted from the GFP image, resulting in an autofluorescence-free GFP image (Figure 1A). Recent studies showed that in addition to GFP, multiple FPs can be simultaneously imaged in vivo [11-14]. In line with this, the DNA tattoo strategy also allowed (the simultaneous) detection of cells expressing a broad range of fluorescent reporters (FP), including cyan fluorescent protein (CFP), yellow fluorescent protein (YFP) and DsRed. As expected, the in vivo emission spectra of the tested FPs were similar to those reported in vitro (Figure 1B). When performing intravital imaging at several time points post-tattoo, the first fluorescent cells could be detected as early as 2 hours after DNA application. Although the number of transfected cells varies per tattoo (ranging from 100 to 300 cells per 10 mm2), the kinetics of expression are the same; within 24 hours the number of detectable cells increases approximately 2-fold and gradually declines over 4 days, at which point all expression has disappeared (this was observed for all tested FPs, and a representative GFP example is shown in Figure 1C). Adding to the flexibility of the DNA tattoo approach, tattoo application of a mixture of plasmids encoding different FPs (i.e. H2B-GFP and DsRed), resulted in detectable expression of both reporter genes in approximately a third of the transfected cells (Figure 1D). This number may be an underestimate because the wide range of observed intensities suggests that in many transfected cells one of the probes may be expressed below the level of detection. Having established the feasibility of monitoring biosensor expression in DNA tattoo-treated skin, we set out to determine the versatility of this approach to report on cellular processes and its ability to follow the fate of individual cells through time. Using a 63x 1.2NA water objective, captured skin images showed subcellular-resolution with details as small as ~300 nm (FWHM). Two GFP fusion proteins, GFP-actin and the PtdIns(4,5)P2 reporter GFP-PH, displayed their expected subcellular distribution: at the periphery of the cell in the cortical cytoskeleton and at the plasma membrane, respectively (Figure 2A). Importantly, we could reproduce the previously made in vitro observation that UV exposure of cells results in decreased PtdIns(4,5)P2 levels [15], as revealed by translocation of GFP-PH to the cytosol in cells irradiated with UV but not in control cells (Figure 2A, lower panel). While many intracellular processes can be followed by monitoring the distribution of location-based probes, certain processes, such as conformational changes, are better revealed by (kinetic changes in) FRET. To assess the feasibility of in vivo FRET measurements and of the longitudinal follow-up of individual cells, we first applied the FRET-based Ca2+ sensor ‘Yellow chameleon’ (Ycam) and its fluorescent constituents, CFP and YFP on separate skin patches of the same animal. In Ycam-tattooed skin areas FRET was readily detectable either by sensitized YFP 106
emission (as described in [16], Figure 3A, left panel) or by acceptor photobleaching (Figure 3B, right panel, for control experiments, see Figure 3B). In order to follow individual cells for multiple days, we subsequently set up a coordinate system that allows one to relocate the same cell in subsequent imaging sessions. To this purpose, ink reference points are applied at the edge of the tattooed area and the position of cells of interest relative to these marker points is recorded during the first imaging session (Figure 4A). In subsequent imaging sessions, the coordinates of these cells combined with overview images at low (5x) magnification (Figure 4B) are then used to retrieve the position of defined cells (for details, see methods). To determine the potential of such longitudinal analysis, we performed a timelapse experiment to study the involvement of caspase-3 in experimentally-induced apoptosis. To this purpose, a FRET-based sensor of caspase-3 activity, consisting of CFP and YFP moieties separated by a caspase-3 substrate site (modified from [17] ) was utilized. DNA encoding this ‘apoptosis-sensor’ was mixed with DNA encoding an ‘apoptosis-switch’, consisting of a modified FK506- (tacrolimus) binding domain fused to caspase-9 [18], in a bicistronic configuration with tdtomato. This mixture was then applied to the abdominal skin of a mouse by DNA tattoo. In parallel, as a control for background apoptosis, a separate skin patch was tattooed with the apoptosis-sensor mixed with a vector encoding tdtomato only (pVax-tdtomato). To follow the consequences of apoptosis induction in vivo, autofluorescence, CFP and YFP images were recorded in the same set of cells expressing the apoptosis sensor, before and 24 hours after intraperitoneal injection of the suicide switch activator AP20187 (Figure 5A). Remarkably, after AP20187 administration the activation of caspase-3 was readily detected by a drop in YFP/ CFP fluorescence ratio, as shown in the higher magnification images in Figure 5B. In line with this, after AP20187 administration, the clear peak of sensitized YFP emission disappeared (Figure 5C). Importantly, cells in skin areas of the same animal treated with the apoptosis-sensor plus control DNA did not show a change in emission spectrum upon AP20187 administration (Figure 5D). Collectively, the data from this experiment demonstrate that this technique allows one to induce a defined process in individual cells and to subsequently monitor the effects of this process (in this case caspase-3 activation) in vivo by an appropriate fluorescent sensor. Conclusions Mouse transgenesis in combination with intravital imaging is time-consuming and can only be performed in specialized laboratories. Here we have presented a tattoo-based intravital imaging assay using techniques that are generally available. Clearly, this new approach that is based on the intradermal delivery of fluorescent reporters by DNA tattoo has its limitations. First, transgene expression obtained by the ‘instant-transgenic’ method as used here is confined to the epidermis and second, the expression of fluorescent reporters is restricted to a few days. In future studies, the duration of expression may be prolonged by enabling integration [19] or episomal maintenance of the plasmid used for transfection [20]. Furthermore, in vivo 107
6
INTRAVITAL IMAGING OF FLUORESCENT MARKERS AND FRET PROBES BY DNA TATTOOING.
transfection using naked DNA can be achieved by various means in other tissues, including hepatocytes [21], myocytes [22] and lymphocytes [23]. More importantly, for many applications the abovementioned drawbacks are balanced by the substantial advantages provided by intravital DNA tattoo imaging. Firstly, mice expressing multiple cellular reporters can be generated in a time span of hours instead of months to years. Secondly, because it is possible to restrict expression of transgenes and reporters to a subset of cells amidst a population of unmodified cells, it is feasible to follow the behavior of gene-modified cells in isolation. In addition, this allows the retracing of individual cells in order to study these cells over a period of days. Thirdly, several tattoos with different combinations of transgenes and reporters can be compared within the same animal, thereby reducing experimental variation. Finally, imaging can be performed on widely available confocal or epifluorescent microscopes. These features suggest that intravital DNA tattoo imaging can form a highly rapid and versatile system to analyze cellular processes such as cell cycle progression, cellular differentiation or apoptosis in an in vivo setting. Methods DNA tattoo The abdomen of the mouse was shaved and subsequently treated with depilation cream (Veet, Reckitt&Colmann, France). The nude abdominal surface was thoroughly rinsed to remove remaining cream, and a droplet of 20 µg DNA in 10 µl water was applied to the skin. Using a sterile disposable 11-needle bar (fine magnum 15, 'challenge in colors' China) mounted on a rotary tattoo/permanent make-up device (Cold skin, B&A trading, Lijnden, the Netherlands) the DNA was tattooed in the skin over a surface of approximately 30 mm2 [10]. During the 15 second tattoo, the needles oscillated at 100 Hz to a depth of 0.5 mm. Following application of the DNA, reference points were tattooed with black permanent make-up ink using a single point needle cartridge (Nouveau contour BV, Weert, Netherlands) mounted on a permanent make-up device (fine magnum 15, Aella, Medium-Tech, Berlin, Germany). All animal experiments were approved by the relevant institutional ethical committee (DEC) and performed in accordance with the local guidelines. Microscopy Anesthetized mice (GFP-MHC class II mice [24] were a kind gift of R. Offringa) were transferred to a 37˚C heated cage, mounted to an inverted TC-SP-AOBS confocal microscope (Leica, Mannheim, Germany). A 5x dry objective (0.15 HC PL Fluotar) and a 20x dry objective (0.7 HC) were used to collect overview images, and a 63x water objective (1.2 PL APO) to collect high resolution images. In vivo FRET imaging In vivo FRET was measured by three independent methods; measurement of sensitized YFP emission, measurement of emission spectra, and measurement of the gain in CFP fluorescence upon YFP photo bleaching. For sensitized emission (YFP fluorescence upon CFP excitation), three images were collected: a CFP 108
image excited at 405 nm and detected between 430 and 480 nm, a sensitized emission image excited at 405 nm and detected between 528 nm and 603 nm, and a YFP image excited at 514 nm and detected between 528 nm and 603 nm. Sensitized emission (FRET) was calculated from these images by correcting the sensitized emission image for leak-through of CFP fluorescence and fluorescence due to direct YFP excitation as described in [16]. Correction factors for leakthrough of CFP and indirect excitation of YFP were determined on-line from cells expressing CFP or YFP only at different locations on the same animal. For every collected FRET image, images of cells expressing CFP or YFP only were collected subsequently to re-determine correction factors for laser fluctuations. Emission spectra were measured by collecting series of xyλ-images at 405 nm excitation and at 10 nm bandwidth emission starting at 430 nm and ending at 650 nm. No significant YFP emission was observed in emission spectra of cells expressing only YFP. Moreover, FRET observed by spectrum imaging was confirmed by the gain of CFP fluorescence after photo-destruction of YFP. The CFP gain was not a result of photo conversion of YFP during photo bleaching as described in [25], because YFP bleaching of skin cells expressing YFP only did not result in a significant increase in fluorescence at CFP wavelengths. Coordinate system to image the same cells for several days Cells of interest were retraced using a coordinate system. The coordinate system is based on two black ink reference points, which are tattooed on the mouse abdomen: an origin and a reference. The distance L(ref) and the angle a(ref) between the origin and the reference are determined by a custom-made Visual Basic (v6.0) program by calling commands from the Leica macro tool package. For every cell coordinate, the distance L(cell) and the angle a(cell) between the cell and the origin are determined and saved. During the next imaging session, the re-anaesthetized animal is re-positioned on the confocal microscope. While L(ref) and L(cell) will be similar (but not identical, see below) at subsequent time-points, the reference a(ref) and all cell angles (a(cell)) will vary because of small rotations in the position of the mouse. To correct for this, a(ref) is re-determined, and all cell angles (a(cell)) are corrected for changes in a(ref). The positions of the cells are then recalculated from the corrected a(cell) values and L(cell). Stretching of the skin does result in small deviations in L in subsequent imaging sessions and therefore can result in misalignment of the position of cells up to 0.5 mm. To avoid this issue we use an overview image at low (5x) magnification to determine the exact coordinate of each cell (Figure 4). Once a single misalignment has been resolved, all positions (within a radius of 5 mm) are corrected for this misalignment. Acknowledgements We would like to thank Drs J. Segall, W. Wang, D. Kedrin, S. Goswami and J. Wyckoff for helpful discussions, Dr. L. Oomen for assistance with confocal microscopy and Dr. G. van der Krogt for technical assistance. We would like to thank ARIAD Pharmaceuticals (Cambridge, MA) for the generous supply of AP20187, and R. Offringa for providing the MHCII-eGFP mice. JvR was supported by a fellowship from the Dutch Cancer Society. 109
6
INTRAVITAL IMAGING OF FLUORESCENT MARKERS AND FRET PROBES BY DNA TATTOOING.
References 1. Giepmans BNG, Adams SR, Ellisman MH, Tsien RY: The fluorescent toolbox for assessing protein location and function. Science 2006, 312(5771):217224. 2. Condeelis J, Singer RH, Segall JE: The great escape: When cancer cells hijack the genes for chemotaxis and motility. Annu Rev Cell Dev Biol 2005, 21(1): 695-718. 3. Sumen C, Mempel TR, Mazo IB, von Andrian UH: Intravital microscopy: visualizing immunity in context. Immunity 2004, 21(3):315-329. 4. Hoffman RM: The multiple uses of fluorescent proteins to visualize cancer in vivo. Nature Rev Cancer 2005, 5(10):796-806. 5. Yang M, Baranov E, Jiang P, Sun F-X, Li X-M, Li L, Hasegawa S, Bouvet M, Al-Tuwaijri M, Chishima T, Shimada H, Moossa AR, Penman S, Hoffman RM: Whole-body optical imaging of green fluorescent protein-expressing tumors and metastases. Proc Natl Acad Sci U S A 2000, 97(3):1206-1211. 6. Yang M, Luiken G, Baranov E, Hoffman RM: Facile whole-body imaging of internal fluorescent tumors in mice with an LED flashlight. Biotechniques 2005, 39(2):170-172. 7. Condeelis J, Segall JE: Intravital imaging of cell movement in tumours. Nature Rev Cancer 2003, 3(12):921-930. 8. Yamauchi K, Yang M, Jiang P, Xu M, Yamamoto N, Tsuchiya H, Tomita K, Moossa AR, Bouvet M, Hoffman RM: Development of real-time subcellular dynamic multicolor imaging of cancer-cell trafficking in live mice with a variablemagnification whole-mouse imaging system. Cancer Res 2006, 66(8):4208-4214. 9. Li L, Hoffman RM: The feasibility of targeted selective gene therapy of the hair follicle. Nat Med 1995, 1(7):705-706. 10. Bins AD, Jorritsma A, Wolkers MC, Hung C-F, Wu TC, Schumacher TNM, Haanen JBAG: A rapid and potent DNA vaccination strategy defined by in vivo monitoring of antigen expression. Nat Med 2005, 11(8):899-904. 11. Hoffman RM, Yang M: Subcellular imaging in the live mouse. Nat Protocols 2006, 1(2):775-782. 12. Sahai E, Wyckoff J, Philippar U, Segall J, Gertler F, Condeelis J: Simultaneous imaging of GFP, CFP and collagen in tumors in vivo using multiphoton microscopy. BMC Biotechnol 2005, 5(1):14. 13. Yamauchi K, Yang M, Jiang P, Yamamoto N, Xu M, Amoh Y, Tsuji K, Bouvet M, Tsuchiya H, Tomita K, Moossa AR, Hoffman RM: Real-time In vivo dual-color imaging of intracapillary cancer cell and nucleus deformation and migration. Cancer Res 2005, 65(10):4246-4252. 110
14. Yamamoto N, Jiang P, Yang M, Xu M, Yamauchi K, Tsuchiya H, Tomita K, Wahl GM, Moossa AR, Hoffman RM: Cellular dynamics visualized in live cells in vitro and in vivo by differential dual-color nuclear-cytoplasmic fluorescent-protein expression. Cancer Res 2004, 64(12):4251-4256. 15. Halstead JR, van Rheenen J, Snel MJ, Meeuws S, Mohammed S, D'Santos CS, Heck A, Jalink K, Divecha N: A role for PtdIns(4,5)P2 and PIP5K[alpha] in regulating stress-induced apoptosis. Curr Biol 2006, 16(18):1850-1856. 16. van Rheenen J, Langeslag M, Jalink K: Correcting confocal acquisition to optimize imaging of Fluorescence Resonance Energy Transfer by sensitized emission. Biophys J 2004, 86(4):2517-2529. 17. Kohler M, Zaitsev SV, Zaitseva II, Leibiger B, Leibiger IB, Turunen M, Kapelioukh IL, Bakkman L, Appelskog IB, Boutet de Monvel J, Imreh G, Berggren P-O: On-line monitoring of apoptosis in insulin-secreting Cells. Diabetes 2003, 52(12):2943-2950. 18. Straathof KC, Pule MA, Yotnda P, Dotti G, Vanin EF, Brenner MK, Heslop HE, Spencer DM, Rooney CM: An inducible caspase 9 safety switch for T-cell therapy. Blood 2005, 105(11):4247-4254. 19. Groth AC, Olivares EC, Thyagarajan B, Calos MP: A phage integrase directs efficient site-specific integration in human cells. Proc Natl Acad Sci U S A 2000, 97(11):5995-6000. 20. Gassmann M, Donoho G, Berg P: Maintenance of an extrachromosomal plasmid vector in mouse embryonic stem cells. Proc Natl Acad Sci U S A 1995, 92(5):1292-1296. 21. Budker V, Zhang G, Knechtle S, Wolff JA: Naked DNA delivered intraportally expresses efficiently in hepatocytes. Gene Ther 1996, 3(7):593-598. 22. Wolff JA, Malone RW, Williams P, Chong W, Acsadi G, Jani A, Felgner PL: Direct gene transfer into mouse muscle in vivo. Science 1990, 247(4949):14651468. 23. Tupin E, Poirier B, Bureau MF, Khallou-Laschet J, Vranckx R, Caligiuri G, Gaston AT, Duong Van Huyen JP, Scherman D, Bariéty J, B. MJ, A. N: Non-viral gene transfer of murine spleen cells achieved by in vivo electroporation. Gene Ther 2003, 10(7):569-579. 24. Boes M, Cerny J, Massol R, Op den Brouw M, Kirchhausen T, Chen J, Ploegh HL: T-cell engagement of dendritic cells rapidly rearranges MHC class II transport. Nature 2002, 418(6901):983-988. 25. Valentin G, Verheggen C, Piolot T, Neel H, Coppey-Moisan M, Bertrand E: Photoconversion of YFP into a CFP-like species during acceptor photobleaching FRET experiments. Nat Meth 2005, 2(11):801-801.
111
6
INTRAVITAL IMAGING OF FLUORESCENT MARKERS AND FRET PROBES BY DNA TATTOOING.
Figure 1 - Fluorescent protein expression in single cells after DNA tattooing on the skin of a living mouse A) Using a standard tattoo machine, a pVAX vector encoding GFP (upper panel) or an empty pVAX vector (lower panel) were tattooed into the skin of mice. Confocal GFP and autofluorescence images were collected as described in the text and GFP images were corrected for autofluorescence by subtracting 0.95x the autofluorescence image. B) The in vivo emission spectra of CFP, GFP, YFP or DsRed and autofluorescence, the latter at 405nm excitation. C) The number of GFP expressing cells counted within the same 10mm2 skin area at different time points post tattoo, as indicated in the graph. D) Merged GFP and DsRed images of skin tattooed with a mixture of H2B-GFP- and DsRed-encoding vectors. Inset, a cell expressing both H2B-GFP and DsRed. 112
Figure 2 - Imaging of location-based biosensors with subcellular resolution A) Images of GFP-actin (top panel) and GFP-PH (lower panel) were collected from the medial section of cells using a confocal microscope. Note that GFP-PH translocated from the membrane to the cytosol in cells following UV-A exposure, but not in non-exposed (control) cells. The post-irradiation time is indicated in the images.
113
6
INTRAVITAL IMAGING OF FLUORESCENT MARKERS AND FRET PROBES BY DNA TATTOOING.
Figure 3 - In vivo FRET A) CFP, YFP and sensitized emission (FRET) images (Left panel) and emission spectra (Right panel) were collected from skin cells expressing Ycam. The red and blue spectrum are pre- and post-YFP photo destruction respectively. Note the gain in CFP fluorescence after photo destruction of YFP. B) Control experiments for intravital FRET imaging. (Left panel) The collected FRET images in A were corrected as described in methods. FRET was not observed in cells expressing only CFP or YFP, consistent with proper correction. (Right panel) The emission spectrum acquired using CFP excitation wavelength (405 nm) of a cell expressing only YFP. The expression level of YFP is similar to that of Ycam in A. Note that YFP is inefficiently excited at 405 nm, and that photo destruction of YFP does not result in an increase of fluorescence at CFP wavelengths.
114
Figure 4 - Defined cells can be followed over multiple days A) In order to set up the coordinate system, two black ink reference points are tattooed on the abdomen of mice; origin and reference. At T0, the distance L(ref) and the angle a(ref) are measured between the two reference points. For cell coordinates, the distance L(cell) and the angle a(cell) are measured between the cell and origin. B) At a subsequent timepoint (T1), small rotations in the position of the mouse alter a(ref) and a(cell), but not L(ref) and L(cell). a(ref) is measured again, and a(cell) is corrected by the difference in a(ref) at T0 and T1. Knowing L(cell) and a(cell), the positions of the previously recorded cells can be re-calculated. (Right panels) Merge of GFP (green) and background (cyan) overview images of the skin of a mouse at T=0 and T=24 hours, imaged with a low-magnification objective (5x). The same cells are found back using the coordinate system. Insets represent the boxed regions, which have been imaged with a higher magnification objective (20x).
115
6
Figure 5 - Visualizing apoptosis in the skin of a living mouse by FRET A) An AP20187 sensitive 'apoptosis-switch' and FRET-based 'apoptosis-sensor' were tattooed on the skin of living mouse. Apoptosis was induced by injection of the chemical inducer of dimerization AP20187 in the peritoneal cavity of the mouse. Images were taken by confocal microscopy before and 24 hours after injection. Cells were identified by image registration at multiple time points as described in the text and supplementary data. A representative cell is boxed. Shown are merged images of the tdtomato, YFP and autofluorescence. B) CFP and YFP images of the cell as boxed in A imaged at higher resolution. Note the loss in FRET as a consequence of apoptosis induction. C) Emission spectrum of the cell boxed 116
in A before injection of AP20187 (left, blue spectrum) and 24 hours after injection (right, red spectrum). D) Spectrum of a cell in the same animal as in C, double transfected with a control construct plus the FRET based apoptosis-sensor 24 hours after injection of AP20187. Additional text The fluorescent cells observed after DNA tattooing are predominately keratinocytes The epidermal layer of mouse skin is nearly exclusively composed of keratinocytes and antigen presenting cells (Langerhans cells). Judged by morphology, a large fraction of the cells expressing FPs after DNA tattooing are keratinocytes. In line with this, application of a GFP-encoding plasmid in which transcription is driven by the keratinocyte-specific K14 promoter yields detectable expression in a high number of cells (approximately 50% as compared to the number of FP-positive cells observed upon introduction of a CMV IEdriven transgene) (additional file Figure 1A). Because the K14 promoter is 10-100 fold weaker than CMV IE-promoter, the number of GFP-expressing keratinocytes observed upon application of the K14-driven GFP vector is likely to be an underestimate. To determine whether transgene-expression can also be observed in skin-resident antigenpresenting cells, we tattooed an H2B-mRFP-encoding plasmid to transgenic mice in which all Langerhans cells are marked by GFP expression (GFP fusion to MHC class II [24]). While both GFP-marked Langerhans and mRFP-expressing cells are readily observed, co-expression of the two FPs is rare (additional file Figure 1B), indicating that only small numbers of Langerhans cells are transfected after DNA tattooing. In conclusion, although both keratinocytes and APCs are present in the epidermal layer of the skin, the fluorescent epidermal cells that are observed after DNA tattooing are predominately keratinocytes. Additional Figure DNA tattooing results in fluorescent keratinocytes. A) GFP expression observed upon application of a keratinocyte-specific K14 promoter-driven GFP transgene. Application of DNA encoding K14 or CMV IE-driven GFP transgenes resulted in approximately 8 or 16 fluorescent cells/mm9respectively. B) Transgenic mice in which all Langerhans cells express GFP were tattooed with a plasmid encoding a CMV IE-driven H2B-mRFP transgene. Depicted is a merge of a GFP and mRFP image.
A
B
117
C H A P T E R
7
Design and use of conditional MHC class I ligands Mireille Toebes1, Adriaan Bins1,4, Miriam Coccoris1,4, Boris Rodenko2, Raquel Gomez1, Nella J. Nieuwkoop3, Willeke van de Kasteele1, Guus F. Rimmelzwaan3, John B.A.G. Haanen1, Huib Ovaa2 and Ton N.M. Schumacher1 Division of Immunology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX, Amsterdam, The Netherlands. 1
Division of Cellular Biochemistry, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX, Amsterdam, The Netherlands. 2
Department of Virology and WHO National Influenza Center, Erasmus Medical Center, PO Box 1738, 3000 DR Rotterdam, The Netherlands. 3
4
These authors contributed equally
Correspondence should be addressed to T.N.M.S. (
[email protected]) or H.O (
[email protected])
Abstract Major histocompatibility complex (MHC) class I molecules associate with a large variety of peptide ligands during biosynthesis and present these ligands on the cell surface for recognition by cytotoxic T cells. We have designed conditional MHC ligands that form stable complexes with MHC molecules, but that degrade on command, by exposure to a defined photostimulus. “Empty MHC molecules” generated in this manner can be loaded with arrays of peptide ligands to determine MHC binding properties and to monitor antigen-specific T cell responses in a highthroughput manner. The value of this approach is documented by the identification of cytotoxic T cell epitopes within the H5N1 influenza A/Vietnam/1194/04 genome.
118
Introduction MHC class I molecules are heterotrimers that consist of an invariant light chain, a polymorphic heavy chain and an eight to ten amino acid peptide ligand. The peptide forms an essential subunit of the MHC class I complex, as MHC class I molecules that fail to associate with peptide ligand are unstable1,2. Association of peptides with MHC class I molecules is in large part based on shape and electrostatic complementarity between two amino acid side chains at the anchor positions of the peptide and MHC allele-specific pockets3,4. In addition, binding of peptide ligands depends on the interaction of the MHC molecule with the terminal α-NH2 and carboxyl groups of the peptide5,6. Because the sequence requirements for binding to MHC are largely restricted to two dominant anchor residues7, the average protein contains several dozen potential T cell epitopes and, as an example, the approximately 100 open reading frames of a member of the herpesviridae family contain an estimated 4,000 potential T cell epitopes. Definition of tumor and pathogen-encoded MHC ligands and detection of T cell responses specific for such ligands remains a major challenge. The visualization of antigen-specific T cell responses was first made possible with the development of tetrameric MHC reagents by Altman et al.8. In this strategy, soluble MHC monomers complexed with a peptide of interest are biotinylated and converted to tetravalent structures by binding to fluorochrome-conjugated streptavidin or avidin. The resulting MHC tetramers have become essential reagents for the detection of antigen-specific CD4+ and CD8+ T cells by flow cytometry. More recent work indicates that definition of antigen-specific T cell responses by MHC microarray-based strategies is also feasible9,10. Furthermore, MHC-based selection of antigen-specific T cells has been proposed as a strategy to boost melanoma-specific T cell responses in melanoma patients11 and to provide defined minor histocompatibility antigen-specific and virus-specific T cells in recipients of allogeneic stem cell transplants and other immunocompromised individuals12-17. At present, the major limitation of MHC tetramer-based technologies is formed by the involved nature of the refolding and purification steps that are required for every individual peptide/MHC complex. Consequently, it has not been feasible to apply MHC tetramer technology for highthroughput applications. To address this issue, we have explored the possibility of identifying conditional MHC ligands that can be used to generate peptide-receptive MHC class I molecules at will. Results Design of conditional MHC class I ligands We set out to create MHC class I ligands that would disintegrate on command, while bound to the MHC complex. As a building block for such conditional ligands, we synthesized a 9-fluorenylmethyloxycarbonyl (Fmoc)-derivative of 3amino-3-(2-nitro)phenyl-propionic acid (J). Subsequently, we used this building block to generate variants of the human leukocyte antigen (HLA)-A2.1-restricted influenza A matrix (M1)58-66 epitope (sequence GILGFVFTL). An example of such
119
7
DESIGN AND USE OF CONDITIONAL MHC CLASS I LIGANDS
a variant, in which the Thr8 is replaced by 3-amino-3-(2-nitro)phenyl-propionic acid (resulting in compound I; GILGFVFJL), is given in Fig. 1a. Consistent with prior literature18,19, this compound disintegrates upon exposure to UV light (Fig. 1a), and following UV exposure, a molecule with a mass of 751.4 becomes apparent, corresponding to the mass of the heptameric peptide acetamide fragment (Fig. 1a, compound II, Fig. 1b). The small, carboxy-terminal peptide fragment that is also generated during this cleavage reaction was not further studied. To establish whether UV-sensitive ligands can be used to generate peptide/MHC complexes, we used two M158-66 variants with 3-amino-3-(2-nitro)phenyl-propionic acid incorporated at position four (compound III; GILJFVFTL, Suppl. Fig. 1) or position eight (compound I, Fig. 1a) in MHC class I refolding reactions20. MHC refolding reactions with either GILJFVFTL or GILGFVFJL produced high yields of HLA-A2.1 complexes. As a first crude test of the sensitivity of the non-natural MHC-bound ligands towards UV-mediated cleavage, we exposed HLA-A2.1 complexes containing either the unmodified influenza A M158-66 epitope, or the GILJFVFTL or GILGFVFJL variant to UV and analyzed these reactions by gelfiltration chromatography. Whereas the conventional HLA-A2.1/peptide complex is not affected by exposure to 366 nm light, exposure of either HLA-A2.1GILJFVFTL (not shown) or HLA-A2.1-GILGFVFJL (Suppl. Fig. 2a) leads to a substantial reduction in MHC recovery. Furthermore, when HLA-A2.1 complexes are prepared with an influenza A M158-66 derivative in which the UV-resistant building block 3-amino-3-phenyl-propionic acid21 is incorporated, the resulting MHC complexes are insensitive to UV exposure, demonstrating that the UVinduced decay of HLA-A2.1-GILJFVFTL and HLA-A2.1-GILGFVFJL requires the presence of the nitrophenyl moiety. Notably, the observed loss of folded MHC upon UV exposure of the HLA-A2.1-(GILGFVFJL) complex can be prevented by inclusion of HLA-A2.1 binding peptides during the cleavage reaction, but not by inclusion of a control HLA-A3 binding peptide (Suppl. Fig. 2b), suggesting efficient peptide exchange. While these data demonstrate the UV-sensitivity of HLA complexes containing 3-amino-3-(2-nitro)phenyl-propionic acid-based peptide ligands, it is difficult to establish the efficiency of this cleavage reaction by gel-filtration chromatography (see Legend to Suppl. Fig. 2a). As a more stringent test for replacement of the conditional ligand by the newly added peptide, we exposed HLA-A2.1-GILGFVFJL complexes to UV light in the presence of the cytomegalovirus (CMV) pp65495-503 epitope, and following this reaction, purified the peptide/MHC complexes and analyzed these complexes by mass spectrometry. The major peptide mass that is visible prior to UV exposure corresponds to the Na+ ion of GILGFVFJL, providing formal proof that this ligand forms a stable complex with HLA-A2.1 (Fig. 2). Following UV-mediated cleavage, no detectable amount of GILGFVFJL remains associated with HLA-A2.1. Furthermore, also the levels of cleavage product II complexed with HLA-A2.1 are below background, suggesting that dissociation of this heptameric peptide fragment is essentially complete. Instead, upon UV-
120
mediated cleavage the sole peptide mass visibly associated with HLA-A2.1 corresponds to the mass of the CMV pp65495-503 epitope (Fig. 2). Collectively, these experiments indicate that conditional MHC ligands can be defined that are released from MHC molecules at will and that MHC molecules generated in this process can be loaded with epitopes of choice. To extend these data to other conditional HLA A2.1 ligands, we synthesized a peptide that is predicted to bind avidly to HLA-A2.122, with the UV-sensitive building block J incorporated at position eight (ILAETVAJV). Refolding reactions with this conditional ligand gave high yields of folded HLA-A2.1 complexes and, analogous to the data obtained with HLA-A2.1-GILGFVFJL complexes, these complexes disintegrate following UV exposure (data not shown). MHC exchange tetramers To test the potential value of this MHC exchange technique for the visualization of antigen-specific T cells we performed MHC exchange reactions with biotinylated HLA-A.2.1- GILGFVFJL or HLA-A.2.1-ILAETVAJV complexes. Subsequently, phycoerythrin (PE)-streptavidin was added, and the resulting MHC tetramers (hereafter referred to as MHC exchange tetramers) were used to detect antigenspecific T cells by flow cytometry. MHC exchange tetramers containing the high affinity (A2L) variant of the Melan-A/MART-126-35 epitope23 stain a MART-1specific CTL clone as efficiently as conventional MHC class I tetramers (Fig. 3a). Likewise, MHC exchange tetramers can be used to detect low magnitude T cell responses in peripheral blood samples (Fig. 3b, Suppl. Fig. 3a). UV-induced cleavage of MHC-bound ligands also allows the synthesis and use of MHC tetramers that contain the naturally occurring MART-126-35 peptide that binds to MHC class I with low affinity, due to the absence of a leucine/ methionine residue at the P2 anchor position (Fig. 3a). This indicates that UV-induced peptide exchange is sufficiently robust to screen collections of putative T cell epitopes without a priori knowledge of MHC binding affinities. Importantly, MHC tetramers prepared from HLA-A2.1-GILGFVFJL complexes or A2Kb-GILGFVFJL complexes (see further) that have not been exposed to UV light, and that are therefore uniformly occupied by the UV-sensitive M158-66 variant, do not stain polyclonal influenza A M158-66-specific T cells (Suppl. Fig. 3b and 3c). This indicates that either the alteration in the peptide backbone due to the introduction of an unnatural β-amino acid, or the replacement of the threonine side chain is incompatible with T cell recognition by M158-66–specific T cells. Consequently, even in settings where release of conditional ligand would not be optimal, MHC exchange reagents are not expected to display background reactivity due to the presence of residual conditional ligand. It is noted that if conditional ligand-MHC complexes for other alleles would display background reactivity, it should be straightforward to prevent such reactivity by modification of T cell receptor-exposed sidechains. Notably, MHC tetramers generated by UV exchange compare favorably to MHC-Ig dimers generated by passive peptide exchange24,25, both with respect to signal intensity and signal-to-noise ratios (Suppl. Fig. 4a, 4b).
121
7
DESIGN AND USE OF CONDITIONAL MHC CLASS I LIGANDS
To test whether conditional ligands may readily be identified for other MHC class I alleles we synthesized four variants of the H2-Db-restricted ASNENMETM influenza A NP366-374 epitope. Two of those variants, IV and V (Suppl. Fig. 1), fulfilled both criteria in that H2-Db complexes could be generated with these conditional ligands and that these ligands could be cleaved in the MHC-bound state. Consistent with the data obtained for HLA-A2.1, H-2Db exchange tetramers prepared from UV-sensitive H-2Db-ASNENJETM complexes stain antigenspecific T cells with high specificity (Fig. 3c). Furthermore, as is the case for HLAA2.1-GILGFVFJL tetramers, H-2Db-ASNENJETM tetramers that are uniformly occupied by the UV-sensitive variant of NP366-374, do not stain influenza A NP366-374specific T cells (Fig. 3c). High-throughput screening with MHC exchange reagents As a first test of the potential of MHC exchange for high-throughput epitope mapping, we cloned and sequenced the four genes encoding the immunodominant proteins of A/Vietnam/1194/04, an influenza A H5N1 strain isolated from a fatal human case in Vietnam, and scanned the encoded proteins for potential HLA-A2.1 binding peptides. Within the hemagglutinin (HA), neuraminidase (NA), matrix-1 (M1) and nucleoprotein (NP) gene products, we identified 132 potential epitopes with a score for predicted MHC binding of ≥2022, and these peptides were produced by micro-scale synthesis. In parallel, we used the same set of genes to prepare vectors for DNA vaccination and groups of HLA-A2.1 transgenic mice26 were vaccinated by DNA tattoo27. At the peak of the vaccination-induced T cell response, we generated a collection of MHC exchange tetramers by performing 132 parallel UV-mediated exchange reactions on A2Kb-GILGFVFJL complexes28 in microtiter format, and the resulting MHC tetramer collection was used to screen peripheral blood samples of vaccinated mice. This analysis demonstrated the presence of two T cell epitopes within these four gene products of A/Vietnam/1194/04. Specifically, this screen confirmed the immunogenicity of the known influenza A M158-66 epitope that is conserved between the majority of influenza A strains. In addition, this scan revealed the presence of a previously unknown HLA-A2.1-restricted T cell epitope located in the A/Vietnam/1194/04 nucleoprotein (Fig. 4). Interestingly, in this HLA-A2.1 transgenic mouse model, the immunogenicity of this epitope (NP373-381) is substantially higher than that of the classical M158-66 epitope (M158-66: one out of five responding mice after primary vaccination, three out of five responding mice after secondary vaccination; NP373-381: five out of five responding mice after primary vaccination). This novel T cell epitope is shared between H5N1 strains of the past years but is distinct in older influenza A strains. Discussion Here we have described conditional MHC ligands that can disintegrate in the MHC bound state under conditions that do not affect the integrity of the MHC molecule, thereby permitting the reloading of assembled MHC molecules with epitopes of choice. Importantly, the peptide ligands that are bound to the MHC
122
in these reactions are unmodified, nor is the MHC backbone altered. In line with this, MHC complexes generated by this exchange technology display the predicted binding specificity in all cases tested. This strategy and related chemical cleavage strategies should be of substantial use in the high-throughput identification of both MHC ligands and cytotoxic T cell responses. In this strategy, a single batch of UV-sensitive MHC complex is prepared by the classical in vitro MHC class I refolding and purification protocols, and this UV-sensitive MHC complex is subsequently used to generate large arrays of desired pMHC complexes in 1 h exchange reactions. MHC-Ig dimers purified from eukaryotic cells have previously been used to generate peptide/MHC reagents, by performing exchange reactions with exogenously added peptide24,25. While the overall goal of this technology is similar to that of UV-induced peptide exchange, the technologies differ at essential points. Specifically, while the MHC-Ig dimer technology depends on the slow release of a pool of unknown endogenous peptides, and is facilitated by conditions (e.g. low or high pH) that also destabilize the MHC molecule, UV-induced peptide exchange is based on the release of a single ligand, by exposure to a defined trigger that does not affect the integrity of the MHC molecule. Furthermore, the capacity of MHC-Ig based reagents to reveal antigen-specific T cell responses is limited as compared to MHC exchange tetramers (Suppl. Fig. 4). The observation that ligands that disintegrate on command could readily be identified for both tested MHC alleles suggests that it will be straightforward to identify 2-nitrophenyl-based conditional ligands for other MHC class I alleles. Conditional ligands can be designed by replacement of amino acids in either a known peptide ligand, or in a predicted high affinity ligand based on the peptide binding motif for this allele, and for HLA A2.1 both approaches proved successful. For MHC alleles for which structural information is available, the wateraccessibility of sidechains may also be used as a criterion to select positions at which the UV-sensitive building block can be incorporated. Various types of functional assays for antigen-specific T cell detection such as intracellular cytokine staining and cytokine capture have been developed in the past few years and these assays may be utilized for high-throughput analysis of T cells that display a given effector function. MHC exchange multimer technology should complement these technologies by allowing high-throughput analysis of T cell responses, irrespective of the capacity of T cells to produce a given cytokine. Also the combination of the two technologies, in which high complexity MHC multimer arrays are used to probe T cell reactivity by monitoring cytokine production may be particularly useful. On-command cleavage of MHC ligands also seems attractive for the production of sets of clinical grade MHC reagents for adoptive T cell therapy13. Specifically, the generation of a single large batch of clinical grade MHC molecules complexed with conditional ligand should allow the straightforward assembly of an MHC reagent desired for clinical use, by performing MHC exchange reactions with the relevant peptide ligands. Such high-grade MHC 123
7
DESIGN AND USE OF CONDITIONAL MHC CLASS I LIGANDS
reagents should be particularly attractive for the isolation of melanoma-specific T cells29, and for the isolation of defined minor histocompatibility antigen-specific T cells12. Finally, in addition to the use of conditional MHC class I ligands for highthroughput diagnostic screening and for adoptive T cell therapy, we speculate that cleavage of ligands bound to MHC molecules in the crystalline state30 might be used to obtain the elusive structure of empty MHC class I. Methods Peptide synthesis and preparation of recombinant MHC. We obtained the UV-sensitive building block for peptide synthesis (N-fluorenylmethyloxycar bonyl 3-amino-3-(2-nitro)phenyl-propionic acid) by protection of 3-amino-3(2-nitro)phenyl-propionic acid (Lancaster) with fluorenylmethyl chloroformate (Sigma-Aldrich) in dioxane/ 10% aqueous Na2CO3 3/2 v/v according to a published procedure21. We synthesized naturally occurring peptides and UVsensitive peptide variants by standard Fmoc synthesis. We performed MHC class I refolding reactions as described20 and we purified refolded MHC class I molecules by gel-filtration chromatography on a Phenomenex Biosep SEC S3000 column (Phenomenex) in 20 mM Tris-HCl pH 7.0/ 150 mM NaCl. Purified MHC class I complexes are stored at -20 °C in 20 mM Tris-HCl pH 7.0/ 150 mM NaCl and 16% glycerol. MHC exchange reactions. To produce single or small sets of MHC reagents by MHC exchange, we exposed biotinylated HLA-A2.1- GILGFVFJL, HLA-A2.1ILAETVAJV or H-2Db-ASNENJETM complexes (0.5 µM in 20 mM Tris-HCl pH 7.0/ 150 mM NaCl/ 0.5 mM dithiothreitol (DTT)) to UV (366 nm UV lamp, CAMAG) in the presence of 50 µM of the indicated peptides for 1-2 h on ice. Following exchange, samples were spun at 16,000g for 5 min, (PE)-streptavidin was added and we used the resulting MHC exchange tetramers for T cell staining without further purification. To generate the collection of H5N1 A2Kb tetramers, we predicted potential peptide epitopes within four gene segments of influenza A/Vietnam/1194/04 using the SYFPEITHI prediction program22 and all peptides with a SYFPEITHI score ≥20 were produced by micro-scale (60 nmole) synthesis (JPT Peptide Technologies GmbH). Of the 132 potential epitopes, 116 terminated in either a valine, leucine or isoleucine residue and these were synthesized with the naturally occurring carboxyterminal amino acid. The remaining 16 peptides, terminating in various nonaliphatic amino acids, were all synthesized with a carboxy-terminal isoleucine to facilitate peptide production. We prepared MHC exchange tetramers by performing parallel small scale exchange reactions on biotinylated A2Kb-GILGFVFJL class I complexes (0.5 µM)28 with the 132 candidate influenza A/Vietnam/1194/04 epitopes and a set of control peptides (all peptides at 50 µM), by exposure to UV (366 nm UV lamp, CAMAG) for 1 h on ice in 20 mM Tris-HCl pH 7.0/ 150 mM NaCl/ 0.5 mM DTT in 96 well tissue culture V-bottom polystyrene plates (NUNC). Subsequently, samples were spun at 3,300g for 5 min, (PE)-streptavidin (10 µg/ml
124
final concentration) was added, and we used the resulting MHC exchange tetramers for T cell staining without further purification. Mice and vaccinations. We obtained C57BL/6 mice and mice transgenic for the HLA-A2Kb fusion gene26 from the animal department of the Netherlands Cancer Institute. All animal experiments were carried out in accordance with institutional and national guidelines and were approved by the Experimental Animal Committee of the Netherlands Cancer Institute (DEC). For live influenza A infections, we intranasally administered 50 µl of HEPESbuffered saline solution (Life Technologies) containing 200 hemagglutinating units influenza A/HKx31 virus to anesthetized mice. For vaccination of HLAA2 transgenic mice with H5N1 gene segments, we obtained the indicated gene segments from influenza A/Vietnam/1194/04, isolated from a human case in Vietnam and cloned these into pVAX. Groups of four to six mice were vaccinated by DNA tattoo27 on day 0, 3 and 6 with 20 µg of pVAX expressing either the influenza A/Vietnam/1194/04 M1, HA, NA or NP gene under control of the CMV promoter. Cells and Flow cytometry. For analysis of MHC multimer binding and T cell responses in mouse samples, we obtained peripheral blood and erythrocytes were removed by incubation in erythrocyte lysis buffer (155 mM NH4Cl, 10 mM KHCO3, 0.1 mM EDTA, pH 7.4) on ice. Cells were stained with antibody to CD8α (BD Biosciences) and the indicated MHC tetramers for 10–15 min at room temperature. For analysis of MHC multimer binding and T cell responses in human samples, we obtained peripheral blood mononuclear cells of healthy volunteers by Ficoll gradient separation. The MART-126-35 specific CTL clone was obtained by repetitive stimulation and cloning of tumor-infiltrating lymphocytes of a melanoma patient. Cells were stained with the indicated MHC tetramers for 5 min at 37 °C. Subsequently, CD8-specific antibody (BD Biosciences) was added and cells were incubated for 10–15 min at room temperature. Data acquisition and analysis was carried out on a FacsCalibur (Becton Dickinson) using CellQuest software. Acknowledgments We would like to thank V. Cerundolo (Weatherall Institute of Molecular Medicine, Oxford, UK) for the A2Kb expression construct and F. Lemonnier (Institut Pasteur, Paris, France) for HLA-A2.1-transgenic mice. We would like to thank A. Pfauth and F. van Diepen for flow cytometry assistance, I. Blonk for help in generation of H5N1 DNA vaccines and H. Hilkmann for peptide synthesis. This work was funded by NWO Pioneer (T.S.) and VIDI (H.O.) grants.
125
7
DESIGN AND USE OF CONDITIONAL MHC CLASS I LIGANDS
Toebes et al. Fig. 1b
1,057.6 [I-H+]
100
1,058.6
Percent
I
TOF MS ES+ 1.84e4
m/z
0 751.5 [II-H+]
100
TOF MS ES+ 1.02e4
II
A
B
Percent 752.5
m/z
0 750
800
850
900
950
1,000
1,050
1,100
Figure 1: Photocleavage strategy. (a). Top: Structure of I, the photocleavable analog of the influenza A M158-66 epitope obtained as an approximate one to one mixture of diastereoisomers. Single amino acid sequence: GILGFVFJL. Bottom: Structure of II, the heptameric peptide fragment generated upon UV-induced degradation of I. Asterisks in I and II indicate the preferred cleavage site. (b). Liquid chromatography-mass spectrometry (LC-MS) of I prior to and following Percent Percent exposure to 366 nm light (CAMAG) for 60 min, in the presence of 0.5 mM DTT. Expected masses: single-protonated I: 1,057.6; single-protonated II: 751.4. 1,079.6 [I-Na+]
100
TOF MS ES+ 73.3
1,079.6 [I-Na+]
100
TOF MS ES+ 73.3
1,080.6
1,080.6
1,057.6
1,057.6
0
0
100 100
1,079.6 [I-Na+] 965.6 [CMV pp65495-503-Na+]
TOF TOF MS MS ES+ ES+ 73.3 89.8
100
m/z
m/z 965.6 [CMV pp65495-503-Na+]
TOF MS ES+ 89.8
1,080.6
Percent
Percent Percent
943.6
1,057.6
943.6
0
00
700 100
966.6
966.6
750
800
850
900
950
1,000
965.6 [CMV
TOF MS ES+ 89.8
1,050
1,100
m/z
m/z m/z
700 1,150
pp65495-503-Na+]
750
800
850
900
950
1,000
1,050
1,100
Figure 2: UV-mediated peptide exchange. Mass spectrometry of HLA-A2.1associated peptide of HLA-A2.1-GILGFVFJL complexes prior to (left panel) and following (right panel) UV exposure in the presence of CMV pp65495-503 peptide. Percent Observed masses and assignments are indicated. Expected masses: Na+ ion of I: 1,079.6; Na+ ion of CMV pp65495-503: 965.5. Note the absence of detectable I-Na+ m/z following UV-induce peptide exchange (arrow). 25 µM HLA-A2.1-GILGFVFJL in 20 mM Tris-HCl pH 7.0/ 150 mM NaCl and 0.5 mM DTT was exposed to 366 nm light for 60 min on ice in the presence of 500 µM CMV pp65495-503 peptide. Subsequently, HLA-A2.1-peptide complexes were purified by gel-filtration chromatography on a Biosep SEC S-3000 column in 25 mM NH4OAc pH 7.0 and directly used for analysis on a Waters LCT ESI mass spectrometer by direct infusion under optimized conditions (160 V cone voltage, 160 °C desolvation temperature). Note that the use of high cone voltage/ high temperature results in the detection of predominantly sodiated species. 966.6
943.6
0
700
126
750
800
850
900
950
1,000
1,050
1,100
1,150
1,150
A anti-CD8-FITC A2.1-MART-1 26-35(A2L)-PE
A2.1-(I) UV MART-1 26-35(A2L)-PE
anti-CD8-FITC 0.94%
0.25%
A2.1-(I) UV HY311-319 -PE
A2.1-(I) UV MART-1 26-35(WT)-PE
*
B
A2.1-pp65 495-503-PE 0.97%
1-(I) UV pp65495-503-PE
0.94%
0.25%
Peptide + UV
A2.1-M1 58-66-PE
*
+ strep-pe
0.26%
Peptide + UV
A2.1-pp65 495-503-PE anti-CD8-FITC
0.02%
0.97%
A2.1-M1 58-66-PE
+ strep-pe
0.26%
A2.1-(I) UV M158-66-PE
A2.1-(I) UV HY311-319 -PE
A2.1-(I) UV pp65495-503-PE
A2.1-(I) UV M158-66-PE
Figure 3: T cell staining with MHC exchange tetramers. (a). Flow cytometric analysis of MHC tetramer staining of a MART-126-35-specific CTL clone with classical MHC tetramers containing the MART-126-35(A2L) epitope (top left panel), or MHC exchange tetramers containing the MART-126-35(A2L) epitope (top right panel), a control peptide (HY311-319, bottom left panel), or the naturally occurring low affinity MART-126-35 epitope (bottom right panel). (b). Flow cytometric analysis of peripheral blood mononuclear cells from an HLA-A2.1 positive individual stained with classical MHC tetramers containing the CMV pp65495-503 epitope (top left panel) or influenza A M158-66 epitope (top right panel), or with MHC exchange tetramers containing a control peptide (HY311-319, bottom left panel), the CMV pp65495-503 epitope (bottom middle panel) or the influenza A M158-66 epitope (bottom right panel). 127
7
DESIGN AND USE OF CONDITIONAL MHC CLASS I LIGANDS
0.0%
Control
Db-(IV) UV NP366-374-PE anti-CD8-APC
1.87%
0.07%
0.07%
A/HKx31 infected
C
Db-(IV) UV NP366-374-PE
Db-(IV) UV SMCY738-746-PE
Db-(IV)-PE
(c). Flow cytometric analysis of MHC tetramer staining of peripheral blood cells of a C57BL/6 mouse (top panel) or a C57BL/6 mouse eight days post intranasal infection with influenza A/HKx31, encoding the A/PR8/34 NP366-374 ASNENMETM epitope (bottom panels). Analysis was performed with H-2Db exchange tetramers containing the A/PR8/34 NP366-374 epitope (left panels), a control peptide (SMCY738bottom middle panel), or H-2Db tetramers prepared from biotinylated H-2Db746 ASNENJETM monomers that had not undergone exchange reactions (bottom right panel).
0.03%
9.9%
anti-CD8-APC
A
A2Kb-(I) UV NP373-381-PE
A2Kb-(I) UV NP373-381-PE
Figure 4: High-throughput screen of H5N1 T cell epitopes. (a). Flow cytometric analysis of mouse peripheral blood mononuclear cells stained with A2Kb-NP373-381 exchange tetramers of either a non-vaccinated HLA-A2.1 transgenic mouse (left) or of an HLA-A2.1 transgenic mouse vaccinated at days 0, 3, 6 with 20 µg of A/ Vietnam/1194/04 NP-encoding DNA. 128
A2Kb-(I) UV NP373-381
A2Kb-(I) UV M158-66 15
15 12
MHC tetramer+ CD8+ cells (%)
12 +
9 6
CD8 cells 9 6
3
3
Percent MHC tetramer
0
0
0
B
1
15
+
9 6
–1 3
+2
4
3
5
4
–
6 8
–
6 8
6
5 7
4 6
5 7
+
7 9
8
9
+
7 9
8
9
12 CD8 cells 9 6
3
C
2
15
MHC tetramer+ CD8+ cells (%)
12
0
3
Percent MHC tetramer 0
0 0
1
0 2
–1 3
+2 4
3 5
Vaccination
(b). Top: T cell responses of individual mice (closed squares) and average T cell responses (stripes) of non-vaccinated mice (–) and mice vaccinated with the influenza A/Vietnam/1194/04 M1 (left, +) and NP (right, +) genes, analyzed with the indicated A2Kb exchange tetramers at day 13 post primary vaccination. At the peak of the vaccination-induced T cell response, peripheral blood was drawn. Blood of mice in each of the four groups was pooled and analyzed by MHC tetramer staining. MHC tetramers that scored positive were used for re-analysis of individual mice. Bottom: T cell responses of individual mice (closed squares) and average T cell responses (stripes) of non-vaccinated mice (–) and mice vaccinated with the influenza A/Vietnam/1194/04 M1 (left, +) and NP (right, +) genes, analyzed with the indicated A2Kb exchange tetramers at day 11 post secondary vaccination. The identity of the NP373-381 epitope was confirmed by synthesis of this sequence on a preparative scale followed by HPLC purification and screening of vaccinated mice by MHC exchange tetramer staining and intracellular interferon-γ staining (data not shown). 129
7
III DESIGN AND USE OF CONDITIONAL MHC CLASS I LIGANDS III III
IV IV IV
V V V
Supplementary Figure 1: Conditional ligands for HLA-A2.1 and H-2Db. Top: Structure of III, the photocleavable analog of the influenza A M158-66 epitope in which Gly4 has been replaced by photolabile 3-amino-3-(2-nitro)phenylpropionic acid (single letter sequence: GILJFVFTL). Middle: Structure of IV, the photocleavable analog of the influenza A/PR8/34 NP366-374 epitope in which Met6 has been replaced by 3-amino-3-(2-nitro)phenyl-propionic acid (single letter sequence: ASNENJETM). Bottom: Structure of V, the photocleavable analog of the influenza A/PR8/34 NP366-374 in which Glu7 has been replaced by 3-amino-3-(2nitro)phenyl-propionic acid (single letter sequence: ASNENMJTM). 0.06 0.05
A2.1-M158-66
0.04 0.03
A2.1-M158-66 UV
0.02 0.01 0.00 0.05 0.04
Extinction 230 nm
A2.1-(I)
0.03 0.02
A2.1-(I) UV
0.01 0.00
A
5
10
15
Time (min.)
Supplementary Figure 2: Conditional peptide/MHC complexes are sensitive to UV. (a). Gel-filtration chromatography of HLA-A2.1 complexes occupied with the parental M158-66 epitope (top) or HLA-A2.1 complexes occupied with the UVsensitive compound GILGFVFJL (bottom), with (green) or without (red) prior exposure to UV for 1 h on ice in 20 mM Tris-HCl pH 7.0/ 150 mM NaCl and 0.5 mM DTT. 130
0.09 0.08
A2.1-(I) UV
0.07 0.06
A2.1-(I) UV + MART-1 26-35(A2L) (A2.1 binder)
0.05 0.04
A2.1-(I) UV + HY311-319 (A2.1 binder)
0.03 0.02
Extinction 230 nm
A2.1-(I) UV + gp100614-622 (A3 binder)
0.01 0.00
5
10
B
Time (min.)
(b). Gel-filtration chromatography of HLA-A2.1 complexes occupied with the UVsensitive peptide GILGFVFJL exposed to UV for 1 h on ice in 20 mM Tris-HCl pH 7.0/ 150 mM NaCl and 0.5 mM DTT without peptide (red), or in the presence of 100 µM of the HLA-A2.1-binding peptides MART-126-35(A2L) (green) or HY311(blue), or the HLA-A3-binding peptide gp100614-62 (black). Note: Free MHC 319 class I heavy chains have a propensity to aggregate and this aggregation of free MHC class I heavy chains results in a loss of material eluting with a retention time of 11.8-12 min. This analysis underestimates the extent of UV-induced 1000as aggregation of MHC class I heavy chains is incomplete and MHC destruction, separation of MHC class I complexes and free MHC class I heavy chains by gelfiltration chromatography is inefficient. Series1 MHC
100
Series2 MHC
Influenza A M1 58-66-specific T cells 1,000
exchange tetramers tetramers
CMV pp65495-503-specific T cells 1,000
10 0.00
2.00
4.00
100
6.00
100
(arbitrary units) Mean fluorescence intensity 10 0.00
A
1.00
2.00
3.00
4.00
5.00
MHC tetramer concentration (µg/ml)
6.00
10 0.00
1.00
2.00
3.00
4.00
5.00
6.00
MHC tetramer concentration (µg/ml)
Supplementary Figure 3: Sensitivity and specificity of MHC exchange tetramers. (a). Peripheral blood mononuclear cells of an HLA-A2.1 positive individual were stained with the indicated concentrations of conventional HLA-A2.1 tetramers or with HLA-A2.1 exchange tetramers containing either the influenza A M158-66 epitope or the CMV pp65495-503 epitope. Mean fluorescence intensity of the MHC tetramer+ cells was determined and is plotted as a function of MHC tetramer concentration. 131
7
DESIGN AND USE OF CONDITIONAL MHC CLASS I LIGANDS
0.02%
0.36%
anti-CD8-FITC
A2.1-(I) UV M158-66-PE
A2.1-(I)-PE
0.03%
3.0%
anti-CD8-APC
A2K b-(I)-PE
A2K b-(I) UV M158-66-PE
B (b). Flow cytometric analysis of peripheral blood mononuclear cells of an HLAA2.1 positive individual with HLA-A2.1 tetramers containing the UV-sensitive peptide GILGFVFJL (left panel), or the parental M158-66 epitope (right panel). (c). Flow cytometric analysis of peripheral blood mononuclear cells of an HLA-A2.1 transgenic mouse vaccinated with the influenza A/Vietnam/1194/04 M1-encoding expression plasmid with HLA-A2Kb tetramers containing the UV-sensitive peptide GILGFVFJL (left panel), or the parental M158-66 epitope (right panel).
132
control
CMV
Anti-CD8-FITC
control
CMV
Anti-CD8-FITC
DimerX HLA-A2:Ig
UV exchange HLA-A2 Tetramers
A Supplementary Figure 4: Detection of antigen specific T cells with MHC-Ig dimers and MHC exchange tetramers. (a). Flow cytometric analysis of peripheral blood mononuclear cells of four CMV-positive HLA-A2.1 positive individuals with MHC-Ig (DimerX) reagents (left) prepared with a control peptide (HY311) or the CMV pp65495-503 peptide, or with HLA-A2.1 exchange tetramers (right) 319 prepared with a control peptide (HY311-319) or the CMV pp65495-503 peptide.
7 133
DESIGN AND USE OF CONDITIONAL MHC CLASS I LIGANDS
B 0.6 0.5
+
0.4 cells CD8 0.3
0.2 CD8 cells 0.1
% MHC multimer
0 Percent MHC multimer+ control
CMV
control
CMV
control
CMV
control
CMV
control
CMV
0.3 0.25
+
0.2 cells CD8 0.15
0.1 CD8 cells 0.05
% MHC multimer
0 MHC multimer+ Percent control
CMV
0.7 0.6 0.5 +
CD8 cells 0.4 0.3
CD8 0.2 cells 0.1
% MHC multimer
0 Percent MHC multimer+ control
CMV
0.7 0.6 0.5 +
CD8 cells 0.4 0.3
(b). Plots of the background and specific staining obtained with MHC-Ig reagents and MHC exchange tetramers in four donors. Note that the intensity of staining of CMV pp65495-503 specific T cells is higher with HLA-A2.1 exchange tetramers than with MHC-Ig dimers and that the background observed with control MHC multimers is substantially lower than the background observed with MHC-Ig dimers. MHC-Ig dimers for HLA-A2.1 (BD Biosciences) were loaded with the HY311-319 or CMV pp65495-503 peptide per manufacturer’s protocol. In brief, DimerX reagents were prepared overnight at 37 °C, using a 640-fold molar excess of peptide over MHC-Ig (protocol 1, peptide concentration of 750 µM). Cells were stained with 1.5 µg of DimerX reagent for 1 h at 4 °C, followed by staining with PE-labeled anti-mouse IgG1 (clone A851, BD Biosciences) plus FITC-labeled CD8-specific antibody (clone G42-8 BD Biosciences) for 30 min at room temperature (protocol 2).
0.2 cells CD8 0.1
% MHC multimer 0 Percent MHC multimer+ control CMV
Dimer X HLA-A2:Ig
UV exchange HLA-A2 Tetramers
MHC exchange tetramers with the HY311-319 or CMV pp65495-503 peptide were prepared by performing 1 h UV-exchange reactions on HLA-A2.1-GILGFVFJL monomers in the presence of 50 µM of the indicated peptide on ice. Subsequently, samples were spun at 3,300g for 5 min, (PE)-streptavidin (10 µg/ml final concentration) was added, and the resulting MHC exchange tetramers were used for T cell staining without further purification.
134
1. Ljunggren, H.G. et al. Empty MHC class I molecules come out in the cold. Nature 346, 476-80 (1990). 2. Schumacher, T.N. et al. Direct binding of peptide to empty MHC class I molecules on intact cells and in vitro. Cell 62, 563-7 (1990). 3. Fremont, D.H., Matsumura, M., Stura, E.A., Peterson, P.A. & Wilson, I.A. Crystal structures of two viral peptides in complex with murine MHC class I H2Kb. Science 257, 919-27 (1992). 4. Silver, M.L., Guo, H.C., Strominger, J.L. & Wiley, D.C. Atomic structure of a human MHC molecule presenting an influenza virus peptide. Nature 360, 367-9 (1992). 5. Bouvier, M. & Wiley, D.C. Importance of peptide amino and carboxyl termini to the stability of MHC class I molecules. Science 265, 398-402 (1994). 6. Schumacher, T.N. et al. Peptide selection by MHC class I molecules. Nature 350, 703-6 (1991). 7. Falk, K., Rotzschke, O., Stevanovic, S., Jung, G. & Rammensee, H.G. Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules. Nature 351, 290-6 (1991). 8. Altman, J.D. et al. Phenotypic analysis of antigen-specific T lymphocytes. Science 274, 94-6 (1996). 9. Soen, Y., Chen, D.S., Kraft, D.L., Davis, M.M. & Brown, P.O. Detection and characterization of cellular immune responses using peptide-MHC microarrays. PLoS Biol 1, E65 (2003). 10. Stone, J.D., Demkowicz, W.E., Jr. & Stern, L.J. HLA-restricted epitope identification and detection of functional T cell responses by using MHC-peptide and costimulatory microarrays. Proc Natl Acad Sci U S A 102, 3744-9 (2005). 11. Yee, C., Savage, P.A., Lee, P.P., Davis, M.M. & Greenberg, P.D. Isolation of high avidity melanoma-reactive CTL from heterogeneous populations using peptide-MHC tetramers. J Immunol 162, 2227-34 (1999). 12. Falkenburg, J.H., van de Corput, L., Marijt, E.W. & Willemze, R. Minor histocompatibility antigens in human stem cell transplantation. Exp Hematol 31, 743-51 (2003). 13. Moss, P. & Rickinson, A. Cellular immunotherapy for viral infection after HSC transplantation. Nat Rev Immunol 5, 9-20 (2005). 14. Peggs, K.S. et al. Adoptive cellular therapy for early cytomegalovirus infection after allogeneic stem-cell transplantation with virus-specific T-cell lines. Lancet 362, 1375-7 (2003).
135
7
DESIGN AND USE OF CONDITIONAL MHC CLASS I LIGANDS
15. Rooney, C.M. et al. Use of gene-modified virus-specific T lymphocytes to control Epstein-Barr-virus-related lymphoproliferation. Lancet 345, 9-13 (1995). 16. Walter, E.A. et al. Reconstitution of cellular immunity against cytomegalovirus in recipients of allogeneic bone marrow by transfer of T-cell clones from the donor. N Engl J Med 333, 1038-44 (1995). 17. Cobbold, M. et al. Adoptive transfer of cytomegalovirus-specific CTL to stem cell transplant patients after selection by HLA-peptide tetramers. J Exp Med 202, 379-86 (2005). 18. Bosques, C.J. & Imperiali, B. Photolytic control of peptide self-assembly. J Am Chem Soc 125, 7530-1 (2003). 19. Wieboldt, R. et al. Synthesis and characterization of photolabile onitrobenzyl derivatives of urea. J Org Chem 67, 8827-31 (2002). 20. Garboczi, D.N., Hung, D.T. & Wiley, D.C. HLA-A2-peptide complexes: refolding and crystallization of molecules expressed in Escherichia coli and complexed with single antigenic peptides. Proc Natl Acad Sci U S A 89, 3429-33 (1992). 21. Sulyok, G.A. et al. Solid-phase synthesis of a nonpeptide RGD mimetic library: new selective alphavbeta3 integrin antagonists. J Med Chem 44, 1938-50 (2001). 22. Rammensee, H.G., Bachmann, J., Emmerich, N.N., Bachor, O.A. & Stevanovic, S. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics 50, 213-9 (access via: www.syfpeithi.de) (1999). 23. Valmori, D. et al. Enhanced generation of specific tumor-reactive CTL in vitro by selected Melan-A/MART-1 immunodominant peptide analogues. J Immunol 160, 1750-8 (1998). 24. Greten, T.F. et al. Direct visualization of antigen-specific T cells: HTLV-1 Tax11-19- specific CD8+ T cells are activated in peripheral blood and accumulate in cerebrospinal fluid from HAM/TSP patients. Proc Natl Acad Sci U S A 95, 756873 (1998). 25. Oelke, M. & Schneck, J.P. HLA-Ig-based artificial antigen-presenting cells: setting the terms of engagement. Clin Immunol 110, 243-51 (2004). 26. Firat, H. et al. H-2 class I knockout, HLA-A2.1-transgenic mice: a versatile animal model for preclinical evaluation of antitumor immunotherapeutic strategies. Eur J Immunol 29, 3112-21 (1999). 27. Bins, A. et al. A rapid and potent DNA vaccination strategy defined by in vivo monitoring of antigen expression. Nat Med 11, 899-904 (2005).
136
28. Choi, E.M., Palmowski, M., Chen, J. & Cerundolo, V. The use of chimeric A2K(b) tetramers to monitor HLA A2 immune responses in HLA A2 transgenic mice. J Immunol Methods 268, 35-41 (2002). 29. Dudley, M.E. et al. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science 298, 850-4 (2002). 30. Schlichting, I. et al. Time-resolved X-ray crystallographic study of the conformational change in Ha-Ras p21 protein on GTP hydrolysis. Nature 345, 309-15 (1990).
7 137
DISCUSSION
Discussion A short summary of the data described in this thesis This thesis deals with the problems posed by active tumor immunization in cancer patients. It describes the failure of spontaneous T cell responses to sort a clinical effect in end-stage melanoma patients and studies the effect of a peptide-based vaccine in such patients, detecting no impact on clinical status nor on T cell function. In search for means to improve current active immunization strategies, the intradermal administration of DNA with a tattoo device is explored in mice and macaques, and found to be highly immunogenic in both species. With regard to “DNA tattooing” it is not realistic to expect this novel immunization method to have a clinical effect in end-stage melanoma patients. In these patients the high tumor load and the tolerized state of the tumor specific T cell repertoire decimate the chances on successful immune mediated tumor eradication. Nevertheless, it may prove beneficial in earlier stages of this disease and in the treatment of virally induced tumors. Additionally, the method may be suitable for immunization against infectious diseases in remote, underdeveloped areas, due to the excellent keeping qualities of DNA and the low-tech application method. Subsequently, this thesis presents evidence that antigens encoded by DNA vaccines should be stable in order to maximize the efficacy of the vaccine, both upon intradermal and intramuscular injection. Besides this clear message, the finding that antigen stability affects the efficacy of cross presentation contributes to the evolving insight in the mechanisms that govern cross-presentation: it suggests that upon DNA vaccination intact proteins are transferred form the antigen donor cell to the antigen presenting cell. Finally, this thesis reports on two tools, one that facilitates the monitoring of vaccine induced T cell responses by peptide-MHC tetramers and another to further study the mechanism of dermal DNA vaccines, which may be helpful for their improvement. I’ll finalize this thesis by discussing the alleged steps of the mechanism of tattoo DNA vaccination. Doing so, I will try to point out where the bottlenecks may be and make suggestions on how these bottlenecks may be alleviated in order to improve the technology. Suggestions for the improvement of tattoo DNA vaccination Transfection efficiency After DNA tattooing, skin resident cells take up the DNA. In some of these cells the plasmid DNA passes through the cytosol and enters the 138
nucleus. Once the nuclei of a sufficient number of cells contain a plasmid, the first important bottleneck has been passed; when production starts approximately a few hours after the tattoo, a maximal number of antigen donor cells will help to generate sufficient antigen to induce a T cell response. To increase the amount of transfected cells after DNA application, the DNA vaccine may be formulated in lipoplexes. This is a routine procedure for in vitro transfection. Also, some laboratories have shown that the transfection efficiency upon intradermal DNA administration increases by the addition of certain substances (e.g. aurintricarboxylic acid41, ATA) to the DNA. An equally simple method leading to the same result is to increase the concentration of the DNA vaccine. However, constraints imposed by the DNA production method limit the maximal concentration. Alternative and technologically more advanced approaches to optimize the transfection efficiency include the addition of a protein group to the DNA, that helps to target the plasmid to the nucleus42;43. Asynchronicity of antigen and danger signal: early events During the DNA tattoo many perforations are made in the dermis. These microscopic wounds will be infected with the skin-resident bacterial flora and injected with the CpG containing DNA vaccine. The danger signals generated this way will lead to the rapid maturation of skin resident APCs and in a later stage to the recruitment and infiltration of immune cells from the circulation. Upon strong maturation signals, most skin resident APCs migrate away from the skin within a few hours44;45. Hence, a substantial part may leave before antigen production starts. Such a ‘timing-mismatch’ may compromise the efficacy of the method. The preliminary finding that the addition of oligomeric CpG sequences, a very strong adjuvant, to the DNA vaccine abolishes T cell priming fits with this idea. It should be noted that, before making an effort to compensate this timing issue, more experimental data are required; for example an analysis of tattoo induced APC influx in the DLN, combined with a characterization of the dominant APC type. Furthermore, it goes without saying that this issue only affects cross-presentation mediated T cell priming. Nonetheless, the presence of danger signals before production of the antigen and not during its production, is an unphysiological aspect of tattoo DNA vaccination that needs attention. It may be difficult to tackle this issue by accelerating the onset of antigen production. Maybe the addition of mRNA encoding the antigen to the DNA vaccine, or the addition of the antigen itself will prove to be beneficial. Although RNA vaccines are known to be functional
139
d
DISCUSSION
intramuscularly , their half life in the dermis is likely to be very short due to the abundant presence of RNAases. Regarding the addition of proteins or peptides it should be noted that ‘peptide tattooing’ per se does not lead to a T cell response (unpublished data). Nevertheless, mixing RNA or protein to DNA tattoo vaccines may have a synergistic effect. Both additions compromise the easy production of the vaccine though. It may be more feasible to delay the APC maturation in order to compensate the timing-mismatch. Firstly, simple measures like disinfection of the skin before applying the tattoo may have a positive effect. Secondly, the effect of the elimination of CpG sequences in the DNA vaccine, or of the addition of ‘tolerizing’46 instead of activating CpG sequences, is definitively worth testing. Asynchronicity of antigen and danger signal: late events A similar timing issue may come about when, at a later stage, systemic APCs infiltrate the tattooed area. At this time the tattoo induced antigen expression may have waned to suboptimal levels. The current regimen of 3 sequential tattoos may partly compensate this issue. However, in the current regimen each tattoo is placed on a fresh skin patch; hence outside the area where the cellular infiltrate of the previous tattoo (with the recruitment of APCs) is emerging. If further experiments confirm that the peak of the APC recruitment in the tattooed area occurs after the ‘antigen production stop’, than one may either accelerate the first or prolong the latter. With regard to prolongation of the antigen expression, exchanging the immediate early cytomegalovirus promoter (IE-CMV) for one that is less liable to silencing (i.e. the EF1-alpha promoter or the CAGGs expression cassette47;48) may further prolong antigen expression. On the other hand, the acceleration of APC infiltration may be achieved by the inclusion in the DNA vaccine of genes encoding cytokines that accelerate the recruitment of APCs. The addition of adjuvants is further discussed below. Hopefully, a combination of these measures may eventually lead to a single tattoo regimen. Partly related to this timing issue, another approach that has been shown to augment the vaccine induced responses is the combination of dermal and intramuscular administration of the DNA vaccine. Intramuscular injection results in expression lasting for months, and preliminary results suggest a synergistic effect with intradermal tattoo administration (unpublished data). The mechanism behind this effect is yet unclear, however.
140
Quantity and quality of APCs: the impact of adjuvants In light of the alleged late timing issue, the addition of adjuvants that accelerate the recruitment of APCs may be beneficial. In the past, the addition of adjuvants to protein vaccines49 and intramuscular DNA vaccines50 has lead to dramatic improvements in vaccine potency, either by augmenting APC recruitment or by accelerating APC maturation. However, as illustrated by the negative effect of the addition of CpGs to DNA tattoo vaccines, finding the right adjuvant for DNA tattooing is not trivial. So far, the addition of DNA encoding GM-CSF, MIP3a, soluble CD70 multimers, and IL-23 have been tested without clear success (unpublished data). Recently, a strong adjuvant effect of flagellin on intradermal DNA vaccination by gene gun was reported51. This adjuvant will probably be equally effective in DNA vaccines administered by tattooing. Therefore, it should be the first on the list of candidate adjuvants scheduled for testing. Other issues As described in chapter 5, the stability of the encoded antigen has an important impact on the efficacy of T cell induction (as described in chapter 5). Therefore, future vaccines should encode antigens that are inherently stable or antigens that have been modified to increase their stability. Currently, we do not yet know the intricacies of the cross presentation process, and it is likely that besides the stability of an antigen other properties of the antigen contribute to the efficiency of the cross presentation process as well. As our insight in the presentation of skinderived antigens increases, additional modifications to the antigens encoded by tattoo DNA vaccines may lead to further improvement.
141
d
DISCUSSION
1. Rock,K.L. and L.Shen. 2005. Cross-presentation: underlying mechanisms and role in immune surveillance. Immunol Rev. 207:166-83.: 166-183. 2. Yang,F. and X.F.Yang. 2005. New concepts in tumor antigens: their significance in future immunotherapies for tumors. Cell Mol Immunol. 2: 331-341. 3. Cole,W.H. 1976. Relationship of causative factors in spontaneous regression of cancer to immunologic factors possibly effective in cancer. J Surg Oncol. 8:391-411. 4. Papac,R.J. 1998. Spontaneous regression of cancer: possible mechanisms. In Vivo. 12:571-578. 5. Sussman,J.J., S.Shu, V.K.Sondak, and A.E.Chang. 1994. Activation of T lymphocytes for the adoptive immunotherapy of cancer. Ann Surg Oncol. 1:296-306. 6. Jager,D., E.Jager, and A.Knuth. 2001. Immune responses to tumour antigens: implications for antigen specific immunotherapy of cancer. J Clin Pathol. 54:669-674. 7. Baxby,D. 1999. Edward Jenner’s Inquiry; a bicentenary analysis. Vaccine. 17:301-307. 8. Dudley,M.E., J.R.Wunderlich, P.F.Robbins, J.C.Yang, P.Hwu, D.J.Schwartzentruber, S.L.Topalian, R.Sherry, N.P.Restifo, A.M.Hubicki, M.R.Robinson, M.Raffeld, P.Duray, C.A.Seipp, L.Rogers-Freezer, K.E.Morton, S.A.Mavroukakis, D.E.White, and S.A.Rosenberg. 2002. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science. 298:850-854. 9. Kessels,H.W., M.C.Wolkers, M.D.van den Boom, M.A.van der Valk, and T.N.Schumacher. 2001. Immunotherapy through TCR gene transfer. Nat Immunol. 2:957-961. 10. Stewart,A.J. and P.M.Devlin. 2006. The history of the smallpox vaccine. J Infect. 52:329-334. 11. Arlen,P.M., H.L.Kaufman, and R.S.DiPaola. 2005. Pox viral vaccine approaches. Semin Oncol. 32:549-555. 12. Sondak,V.K., M.S.Sabel, and J.J.Mule. 2006. Allogeneic and autologous melanoma vaccines: where have we been and where are we going? Clin Cancer Res. 12:2337s-2341s. 13. Choudhury,A., S.Mosolits, P.Kokhaei, L.Hansson, M.Palma, and H.Mellstedt. 2006. Clinical results of vaccine therapy for cancer: learning
142
from history for improving the future. Adv Cancer Res. 95:147-202.:147202. 14. Pardoll,D. 2003. Does the immune system see tumors as foreign or self? Annu Rev Immunol. 21:807-39. Epub@2001 Dec@19.:807-839. 15. Dunbar,P.R., C.L.Smith, D.Chao, M.Salio, D.Shepherd, F.Mirza, M.Lipp, A.Lanzavecchia, F.Sallusto, A.Evans, R.Russell-Jones, A.L.Harris, and V.Cerundolo. 2000. A shift in the phenotype of melan-A-specific CTL identifies melanoma patients with an active tumor-specific immune response. J Immunol. 165:6644-6652. 16. Maczek,C., T.G.Berger, B.Schuler-Thurner, E.S.Schultz, A.Hamann, P.R.Dunbar, V.Cerundolo, A.Steinkasserer, and G.Schuler. 2005. Differences in phenotype and function between spontaneously occurring melan-A-, tyrosinase- and influenza matrix peptide-specific CTL in HLAA*0201 melanoma patients. Int J Cancer. 115:450-455. 17. Willimsky,G. and T.Blankenstein. 2005. Sporadic immunogenic tumours avoid destruction by inducing T-cell tolerance. Nature. 437:141146. 18. Mukherjee,P., A.R.Ginardi, C.S.Madsen, T.L.Tinder, F.Jacobs, J.Parker, B.Agrawal, B.M.Longenecker, and S.J.Gendler. 2001. MUC1-specific CTLs are non-functional within a pancreatic tumor microenvironment. Glycoconj. J. 18:931-942. 19. van Oijen,M., A.Bins, S.Elias, J.Sein, P.Weder, G.de Gast, H.Mallo, M.Gallee, H.Van Tinteren, T.Schumacher, and J.Haanen. 2004. On the role of melanoma-specific CD8+ T-cell immunity in disease progression of advanced-stage melanoma patients. Clin Cancer Res. 10:4754-4760. 20. Bubenik,J. 2004. MHC class I down-regulation: tumour escape from immune surveillance? (review). Int J Oncol. 25:487-491. 21. Dunn,G.P., A.T.Bruce, H.Ikeda, L.J.Old, and R.D.Schreiber. 2002. Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol. 3:991-998. 22. Mocellin,S., S.Mandruzzato, V.Bronte, M.Lise, and D.Nitti. 2004. Part I: Vaccines for solid tumours. Lancet Oncol. 5:681-689. 23. Rosenberg,S.A., J.C.Yang, and N.P.Restifo. 2004. Cancer immunotherapy: moving beyond current vaccines. Nat Med. 10:909-915. 24. Gurunathan,S., D.M.Klinman, and R.A.Seder. 2000. DNA vaccines: immunology, application, and optimization*. Annu Rev Immunol. 18:92774.:927-974. 25. Tang,D.C., M.DeVit, and S.A.Johnston. 1992. Genetic immunization is a simple method for eliciting an immune response. Nature. 356:152-154. 143
d
DISCUSSION
26. Ulmer,J.B., J.J.Donnelly, S.E.Parker, G.H.Rhodes, P.L.Felgner, V.J.Dwarki, S.H.Gromkowski, R.R.Deck, C.M.DeWitt, A.Friedman, and 1993. Heterologous protection against influenza by injection of DNA encoding a viral protein. Science. 259:1745-1749. 27. Johnston,S.A. and D.C.Tang. 1994. Gene gun transfection of animal cells and genetic immunization. Methods Cell Biol. 43 Pt A:353-65.:353365. 28. Ye,G.N., H.Daniell, and J.C.Sanford. 1990. Optimization of delivery of foreign DNA into higher-plant chloroplasts. Plant Mol Biol. 15:809-819. 29. Vahlsing,H.L., M.A.Yankauckas, M.Sawdey, S.H.Gromkowski, and M.Manthorpe. 1994. Immunization with plasmid DNA using a pneumatic gun. J Immunol Methods. 175:11-22. 30. Schrijver,R.S., J.P.Langedijk, G.M.Keil, W.G.Middel, M.MarisVeldhuis, J.T.Van Oirschot, and F.A.Rijsewijk. 1998. Comparison of DNA application methods to reduce BRSV shedding in cattle. Vaccine. 16:130134. 31. Davis,H.L., M.L.Michel, M.Mancini, M.Schleef, and R.G.Whalen. 1994. Direct gene transfer in skeletal muscle: plasmid DNA-based immunization against the hepatitis B virus surface antigen. Vaccine. 12: 1503-1509. 32. Cherbonnel,M., J.Rousset, and V.Jestin. 2003. Strategies to improve protection against low-pathogenicity H7 avian influenza virus infection using DNA vaccines. Avian Dis. 47:1181-1186. 33. Cui,Z., A.Dierling, and M.Foldvari. 2006. Non-invasive immunization on the skin using DNA vaccine. Curr Drug Deliv. 3:29-35. 34. Qin,Y.J., J.F.Zhang, Y.J.Wei, J.F.Ding, K.H.Chen, and J.Tang. 1999. Gene suture--a novel method for intramuscular gene transfer and its application in hypertension therapy. Life Sci. 65:2193-2203. 35. Lundholm,P., Y.Asakura, J.Hinkula, E.Lucht, and B.Wahren. 1999. Induction of mucosal IgA by a novel jet delivery technique for HIV-1 DNA. Vaccine. 17:2036-2042. 36. Manoj,S., L.A.Babiuk, and van Drunen Littel-van den Hurk. 2004. Approaches to enhance the efficacy of DNA vaccines. Crit Rev Clin Lab Sci. 41:1-39. 37. Cho,J.H., J.W.Youn, and Y.C.Sung. 2001. Cross-priming as a predominant mechanism for inducing CD8(+) T cell responses in gene gun DNA immunization. J Immunol. 167:5549-5557.
144
38. Wolkers,M.C., N.Brouwenstijn, A.H.Bakker, M.Toebes, and T.N.Schumacher. 2004. Antigen bias in T cell cross-priming. Science. 304: 1314-1317. 39. Schepers,K., M.Toebes, G.Sotthewes, F.A.Vyth-Dreese,T.A.Dellemijn, C.J.Melief, F.Ossendorp, and T.N.Schumacher. 2002. Differential kinetics of antigen-specific CD4+ and CD8+ T cell responses in the regression of retrovirus-induced sarcomas. J Immunol. 169:3191-3199. 40. Soen,Y., D.S.Chen, D.L.Kraft, M.M.Davis, and P.O.Brown. 2003. Detection and characterization of cellular immune responses using peptide-MHC microarrays. PLoS Biol. 1:E65. 41. Glasspool-Malone,J., S.Somiari, J.J.Drabick, and R.W.Malone. 2000. Efficient nonviral cutaneous transfection. Mol Ther. 2:140-146. 42. Vaysse,L., L.G.Gregory, R.P.Harbottle, E.Perouzel, O.Tolmachov, and C.Coutelle. 2006. Nuclear-targeted minicircle to enhance gene transfer with non-viral vectors in vitro and in vivo. J Gene Med. 8:754-763. 43. Young,J.L., J.N.Benoit, and D.A.Dean. 2003. Effect of a DNA nuclear targeting sequence on gene transfer and expression of plasmids in the intact vasculature. Gene Ther. 10:1465-1470. 44. Falo,L.D., Jr. 1999. Targeting the skin for genetic immunization. Proc Assoc Am Physicians. 111:211-219. 45. Klinman,D.M., J.M.Sechler, J.Conover, M.Gu, and A.S.Rosenberg. 1998. Contribution of cells at the site of DNA vaccination to the generation of antigen-specific immunity and memory. J Immunol. 160:2388-2392. 46. Ho,P.P., P.Fontoura, M.Platten, R.A.Sobel, J.J.DeVoss, L.Y.Lee, B.A.Kidd, B.H.Tomooka, J.Capers, A.Agrawal, R.Gupta, J.Zernik, M.K.Yee, B.J.Lee, H.Garren, W.H.Robinson, and L.Steinman. 2005. A suppressive oligodeoxynucleotide enhances the efficacy of myelin cocktail/IL-4tolerizing DNA vaccination and treats autoimmune disease. J Immunol. 175:6226-6234. 47. Chen,Z., Y.Sahashi, K.Matsuo, H.Asanuma, H.Takahashi, T.Iwasaki, Y.Suzuki, C.Aizawa, T.Kurata, and S.Tamura. 1998. Comparison of the ability of viral protein-expressing plasmid DNAs to protect against influenza. Vaccine. 16:1544-1549. 48. Nishimura,Y., A.Kamei, S.Uno-Furuta, S.Tamaki, G.Kim, Y.Adachi, K.Kuribayashi, Y.Matsuura, T.Miyamura, and Y.Yasutomi. 1999. A single immunization with a plasmid encoding hepatitis C virus (HCV) structural proteins under the elongation factor 1-alpha promoter elicits HCVspecific cytotoxic T-lymphocytes (CTL). Vaccine. 18:675-680.
145
d
DISCUSSION
49. Pietersz,G.A., D.S.Pouniotis, and V.Apostolopoulos. 2006. Design of peptide-based vaccines for cancer. Curr Med Chem. 13:1591-1607. 50. Babiuk,L.A., R.Pontarollo, S.Babiuk, B.Loehr, and van Drunen Littelvan den Hurk. 2003. Induction of immune responses by DNA vaccines in large animals. Vaccine. 21:649-658. 51. Chen,Y.S., Y.S.Hsiao, H.H.Lin, Y.Liu, and Y.L.Chen. 2006. CpGmodified plasmid DNA encoding flagellin improves immunogenicity and provides protection against Burkholderia pseudomallei infection in BALB/c mice. Infect Immun. 74:1699-1705.
146
Nederlandse samenvatting In de jaren ’70 van de vorige eeuw werd duidelijk dat in patiënten met bepaalde soorten kanker, spontane activatie van het afweersysteem soms leidt tot genezing. Het concept van immunotherapie bij melanoompatiënten is gebaseerd op dat inzicht: de bedoeling is de T cellen van het afweersysteem gericht tegen de kanker te stimuleren, zodat deze kunnen bijdragen aan het opruimen van kankercellen. Maar de methoden voor het gericht aanwakkeren van T cel immuniteit (T cel vaccinatie) staan nog in de kinderschoenen en werken tot nu toe alleen in kleine proefdieren (muizen). Bovendien is duidelijk geworden dat veel kankers over een trukendoos beschikken om aanvallen van T cellen te ontwijken. In dit proefschrift worden de resultaten van T cel metingen in melanoom patiënten gepresenteerd waaruit blijkt dat in melanoom patiënten met een vergevorderde ziekte vaak spontaan een melanoom specifieke T cel respons optreedt, maar de aanwezigheid van deze cellen geen gunstig effect heeft op het ziektebeloop. Als zulke cellen door vaccinatie beter (eerder en sterker) kunnen worden geactiveerd, biedt dat therapeutische mogelijkheden. Daarom werd vervolgens de kracht van een op eiwit fragmenten (peptides) gebaseerd vaccin getest in melanoom patiënten. Helaas wekte dit vaccin nauwelijks melanoom specifieke T cellen op. Vervolgens introduceert dit proefschrift een nieuwe experimentele vaccinatie methode, waarbij een (tumor specifiek) DNA vaccin met een tattoo apparaat in de huid wordt geïnjecteerd. In vergelijking met andere experimentele vaccinatie methoden, is de T cel inductie na tattoo DNA vaccinatie erg sterk. Zo blijkt dat met deze methode niet alleen muizen, maar ook grote proefdieren (apen) succesvol gevaccineerd kunnen worden. Toekomstige testen in melanoom patiënten moeten uitwijzen of ook in mensen T cellen door DNA tattooing aangewakkerd kunnen worden, om te helpen bij het opruimen van tumorcellen. Verder werd met proefdier modellen gebaseerd op deze nieuwe vaccinatie methode het inzicht vergroot in de manier waarop afweer cellen de huid controleren. Hieruit kwam een nieuw praktisch inzicht voort: om T cellen optimaal te stimuleren, moet een DNA vaccin coderen voor een stabiel eiwit . Tot slot worden in dit proefschrift nog twee technieken beschreven; een voor de bereiding van reagentia waarmee T cel immuniteit in patiënten sneller en beter gemeten kan worden (zogenaamde tetrameren) en een voor microscopisch onderzoek naar de immunologische processen in de huid, volgend op een infectie. Hopelijk zullen deze technieken verder bijdragen aan de ontwikkeling van een veilig en sterk tumor vaccin.
147
Acknowledgements If you’re in possession of this book, it is very likely that you’ve been involved someway or another in the process of its conception.Therefore I'd like to express my gratitude to you for your ¨mental / ¨physical ¨support / ¨hammering. And just for being thereþ: thank you! Amsterdam, 25th of january 2007
Curriculum Vitae Adriaan Dirk Bins was born in Nijmegen, the Netherlands, on the 7th of July 1973. After having passed his final exams at the Praedinius Gymnasium in Groningen, he took A-levels at Gordonstoun School in Scotland and set of to study Medicine at the Leiden University Medical Center in 1992. In addition he started studying Biology three years later and in 1999 he graduated in both, cum laude for biology. In 2001 he finished his medical internships and received the Hippocrates Studyfund Prize for research done during his studies. In that same year he started the research leading to this thesis in the research group of J.B.A.G. Haanen MD. PhD. at the Netherlands Cancer Institute / Antoni van Leeuwenhoek Hospital in Amsterdam.The work done in this period was rewarded in 2006 with the NKI/Avl Award. Having finished his thesis he started his fellowships at the Amsterdam Medical Centre in 2006, specializing in internal medicine.After his medical specialization he plans to stay active as a researcher and medical doctor in fields related to immunology. How? Insha’Allah.
148