Advances in Applied Microbiology, Volume 56

ADVANCES IN Applied Microbiology VOLUME 56 This Page Intentionally Left Blank ADVANCES IN Applied Microbiology Ed...

Author: Allen I. Laskin | Joan W. Bennett | Geoffrey M. Gadd

6 downloads 561 Views 5MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

ADVANCES IN

Applied Microbiology VOLUME 56

This Page Intentionally Left Blank

ADVANCES IN

Applied Microbiology Edited by ALLEN I. LASKIN Somerset, New Jersey

JOAN W. BENNETT New Orleans, Louisiana

GEOFFREY M. GADD Dundee, United Kingdom

VOLUME 56

Elsevier Academic Press 525 B Street, Suite 1900, San Diego, California 92101-4495, USA 84 Theobald’s Road, London WC1X 8RR, UK

This book is printed on acid-free paper. Copyright ß 2004, Elsevier Inc. All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the Publisher. The appearance of the code at the bottom of the first page of a chapter in this book indicates the Publisher’s consent that copies of the chapter may be made for personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (www.copyright.com), for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-2004 chapters are as shown on the title pages. If no fee code appears on the title page, the copy fee is the same as for current chapters. 0065-2164/2004 $35.00 Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone: (þ44) 1865 843830, fax: (þ44) 1865 853333, E-mail: [email protected]. You may also complete your request on-line via the Elsevier homepage (http://elsevier.com), by selecting ‘‘Customer Support’’ and then ‘‘Obtaining Permissions.’’ For all information on all Academic Press publications visit our Web site at www.academicpress.com ISBN: 0-12-002658-9 PRINTED IN THE UNITED STATES OF AMERICA 04 05 06 07 08 9 8 7 6 5 4 3 2 1

CONTENTS Potential and Opportunities for Use of Recombinant Lactic Acid Bacteria in Human Health SEAN HANNIFFY, URSULA WIEDERMANN, ANDREAS REPA, ANNICK MERCENIER, CATHERINE DANIEL, JEAN FIORAMONTI, HELENA TLASKOLOVA, HANA KOZAKOVA, HANS ISRAELSEN, SøREN MADSEN, ASTRID VRANG, PASCAL HOLS, JEAN DELCOUR, PETER BRON, MICHIEL KLEEREBEZEM, AND JERRY WELLS I. Introduction ............................................................................ II. Use of Recombinant LAB to Prevent Infectious Diseases ....................... III. Potential for Immune Modulation of Type I Allergy Using Recombinant LAB ..................................................................... IV. Opportunities for the Treatment of Inﬂammatory Bowel Diseases Using Recombinant LAB ............................................................. V. LAB as Cell Factories for the Manufacturing of Pharmaceutical Proteins .............................................................. VI. Engineering LAB for Their Safe Use in Humans ................................. VII. Opportunities and Potential Applications of Future Research ................. VIII. Concluding Remarks .................................................................. References ...............................................................................

2 3 14 18 23 28 33 43 45

Novel Aspects of Signaling in Streptomyces Development GILLES P. VAN WEZEL I. II. III. IV. V.

AND

ERIK VIJGENBOOM

Introduction ............................................................................ Aspects of Vegetative Growth and Liquid Cultivation of Streptomycetes .... The Switch to Development ......................................................... Novel Genes in Development ........................................................ Concluding Remarks .................................................................. References ...............................................................................

v

65 67 69 72 83 83

vi

CONTENTS

Polysaccharide Breakdown by Anaerobic Microorganisms Inhabiting the Mammalian Gut HARRY J. FLINT I. II. III. IV. V.

Introduction: Role of Gut Microbial Fermentation in Nutrition ............... Microbial Diversity and Interactions Within Gut Ecosystems .................. Strategies for Polysaccharide Utilization by Gut Anaerobes ................... Applications ........................................................................... Conclusions and Future Prospects ................................................. References ..............................................................................

89 92 94 106 109 110

Lincosamides: Chemical Structure, Biosynthesis, Mechanism of Action, Resistance, and Applications JAROSLAV SPI´ZˇEK, JITKA NOVOTNA´,

AND

TOMA´Sˇ RˇEZANKA

I. Introduction ............................................................................ II. Chemical Structure of Lincosamides and Cultivation of Production Strains .................................................................... III. Lincomycin Biosynthetic Pathway ................................................. IV. Genetic Control of Lincomycin Biosynthesis ..................................... V. Mechanism of Action ................................................................. VI. Resistance Against Lincosamides ................................................... VII. Biological Activity and Applications .............................................. VIII. Gram-Positive Bacteria ............................................................... IX. Gram-Negative Bacteria .............................................................. X. Anaerobic Bacteria .................................................................... XI. Protozoa and Other Organisms ...................................................... XII. Conclusion and Future Prospects ................................................... References ..............................................................................

121 124 130 133 135 137 138 139 141 143 144 145 146

Ribosome Engineering and Secondary Metabolite Production KOZO OCHI, SUSUMU OKAMOTO, YUZURU TOZAWA, TAKASHI INAOKA, TAKESHI HOSAKA, JUN XU, AND KAZUHIKO KUROSAWA I. II. III. IV. V. VI. VII. VIII.

Introduction ............................................................................ General Method for Obtaining Drug-Resistant Mutants ......................... Antibiotic Overproduction by rpsL (Ribosomal Protein S12) Mutations ..... Antibiotic Overproduction by rpoB (RNA Polymerase) Mutations ............ Effect of str and rpoB Mutations in Various Bacteria ............................ Increase of Chemical Tolerance in Pseudomonas ................................ Combined Drug-Resistance Mutations ............................................. Conclusion and Future Prospects ................................................... References ..............................................................................

155 156 157 164 167 171 172 175 179

vii

CONTENTS

Developments in Microbial Methods for the Treatment of Dye Efﬂuents R. C. KUHAD, N. SOOD, K. K. TRIPATHI, A. SINGH, AND O. P. WARD I. II. III. IV. V.

Introduction ............................................................................ Conventional Methods ................................................................ Microbial Methods .................................................................... Enzymatic Methods ................................................................... Conclusion .............................................................................. References ...............................................................................

185 186 191 205 206 206

Extracellular Glycosyl Hydrolases from Clostridia WOLFGANG H. SCHWARZ, VLADIMIR V. ZVERLOV, AND HUBERT BAHL I. II. III. IV. V.

Introduction ............................................................................ Modular Structure of the Enzymes ................................................. Function of Noncatalytic Modules .................................................. Characterization of Enzyme Systems ............................................... Concluding Remarks .................................................................. References ...............................................................................

215 217 218 225 251 252

Kernel Knowledge: Smut of Corn MARI´A D. GARCI´A-PEDRAJAS I. II. III. IV. V.

AND

SCOTT E. GOLD

Introduction ............................................................................ The Fungal Saprophyte ............................................................... The Fungal Pathogen .................................................................. The Host Reaction ..................................................................... Conclusions ............................................................................. References ...............................................................................

263 263 267 282 284 285

Bacterial ACC Deaminase and the Alleviation of Plant Stress BERNARD R. GLICK I. II. III. IV. V. VI.

ACC Deaminase–Containing Bacteria .............................................. Ethylene and Plant Stress ............................................................ Decreasing Plant Stress with ACC Deaminase–Containing Bacteria ........... Modulating Nodulation of Legumes ................................................ Decreasing Stress in ACC Deaminase Transgenic Plants ........................ Conclusions ............................................................................. References ...............................................................................

291 293 295 304 306 307 308

viii

CONTENTS

Uses of Trichoderma spp. to Alleviate or Remediate Soil and Water Pollution G. E. HARMAN, M. LORITO, AND J. M. LYNCH I. II. III. IV. V. VI.

Introduction ............................................................................ Trichoderma spp. Are Opportunistic Plant Symbionts ......................... Rhizosphere Competence and Co-Metabolism .................................... Root Enhancement by Trichoderma spp. .......................................... Enhanced Extraction and Biodegradation of Toxicants ......................... Conclusions and Future Prospects ................................................. References ..............................................................................

313 314 315 316 316 326 327

Bacteriophage Defense Systems and Strategies for Lactic Acid Bacteria JOSEPH M. STURINO I. II. III. IV. V. VI. VII.

AND

TODD R. KLAENHAMMER

Introduction ............................................................................ Traditional Strategies ................................................................. Molecular Strategies .................................................................. Native Defense Systems .............................................................. Recent Advancements in Genomics ................................................ Engineered Defense Systems ........................................................ Concluding Remarks .................................................................. References ..............................................................................

332 339 342 343 351 353 368 369

Current Issues in Genetic Toxicology Testing for Microbiologists KRISTIEN MORTELMANS AND DOPPALAPUDI S. RUPA I. II. III. IV. V.

Introduction ............................................................................ Genesis of Genetic Toxicology ...................................................... Regulatory Genetic Toxicology Tests ............................................... Regulatory Genetic Toxicology Guidelines ........................................ Concluding Remarks and Outlook .................................................. References ..............................................................................

379 381 382 391 396 397

INDEX ............................................................................................ CONTENTS OF PREVIOUS VOLUMES ...............................................................

403 411

Potential and Opportunities for Use of Recombinant Lactic Acid Bacteria in Human Health SEAN HANNIFFY,* URSULA WIEDERMANN,{ ANDREAS REPA,{ ANNICK MERCENIER,{ CATHERINE DANIEL,{ JEAN FIORAMONTI,} HELENA TLASKOLOVA,k HANA KOZAKOVA,k HANS ISRAELSEN,{ SØREN MADSEN,{ ASTRID VRANG,{ PASCAL HOLS,# JEAN DELCOUR,# PETER BRON,** MICHIEL KLEEREBEZEM,** AND JERRY WELLS*,{{ *Institute of Food Research, Norwich Research Park, Colney, Norwich, NR4 7UA, United Kingdom {

University of Vienna, Department of Pathophysiology A-1090 Vienna, Austria {

Institut Pasteur de Lille, Department of Microbiology of Ecosystems F59019 Lille, France }

Neurogastroenterology and Nutrition Unit, INRA, F31931 Toulouse 9, France k

Institute of Microbiology, Department of Immunology and Gnotobiology Academy of Sciences of the Czech Republic 142 20 Prague 4, Czech Republic { #

Bioneer A/S, DK-2970 Horsholm, Denmark

Universite´ Catholique de Louvain, Unite´ de Ge´ne´tique B1348 Louvain-la-Neuve, Belgium

**Wageningen Centre for Food Sciences—NIZO Food Research 6710 BA Ede, The Netherlands {{

Author for correspondence. E-mail: [email protected]

I. Introduction II. Use of Recombinant LAB to Prevent Infectious Diseases A. Recombinant LAB as Vaccine Delivery Vehicles B. Infections of the Respiratory Tract C. Infections of the Gastrointestinal Tract D. Infections of the Urogenital Tract III. Potential for Immune Modulation of Type I Allergy Using Recombinant LAB A. Role of the Indigenous Microflora B. Animal Model of Type I Allergy C. Use of LAB for Prophylaxis and Therapy of Type I Allergy IV. Opportunities for the Treatment of Inflammatory Bowel Diseases Using Recombinant LAB A. Role of LAB in Intestinal Barrier Function

2 3 4 7 9 11 14 15 16 17 18 19

1 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

2

HANNIFFY et al.

V.

VI.

VII.

VIII.

B. LAB and Intestinal Inflammation C. Recombinant LAB as a Therapeutic Option LAB as Cell Factories for the Manufacturing of Pharmaceutical Proteins A. The Choice of Microbial Host B. The Expression Vector C. Propagation, Fermentation, and Initial Downstream Processing Engineering LAB for Their Safe Use in Humans A. Food-Grade Systems in LAB for Plasmid Maintenance and Chromosomal Insertion B. Biological Containment Systems Opportunities and Potential Applications of Future Research A. Insights from Genome Sequencing and Comparative Genomics B. The Behavior of LAB in the Host C. The Host Response to LAB Concluding Remarks References

20 22 23 23 24 26 28 28 32 33 33 35 39 43 45

I. Introduction Dietary lactic acid bacteria (LAB) are mostly known for their widespread use in the production and preservation of fermented foods and as such have obtained the ‘‘generally regarded as safe’’ (GRAS) status within the food industry (Adams and Marteau, 1995). Some members of this diverse group of bacteria are components of the indigenous gut microﬂora of both animals and humans and have long been recognized for their health-promoting properties. Indeed, speciﬁc strains of LAB, and in particular lactobacilli, have been used as probiotics (Fuller, 1989; Holzapfel et al., 1998; Isolauri et al., 2001) because they are thought to play a crucial role in maintaining a healthy microﬂora as well as contributing to an expanding list of health-promoting activities (Mercenier et al., 2003). Probiotic LAB have been shown to be beneﬁcial in the treatment of gastrointestinal disorders such as lactose intolerance, travelers’ diarrhea, antibiotic-associated diarrhea, and infections caused by various bacterial and viral pathogens (Heyman, 2000). Clinical trials and animal studies indicate that LAB may also be used to treat the symptoms of atopy and may prevent/reduce the development of allergy (Cross et al., 2001; Kalliomaki and Isolauri, 2003; Kalliomaki et al., 2003; Lodinova-Zadnikova et al., 2003; Tlaskalova-Hogenova et al., 2002). Furthermore, evidence that LAB play a role in controlling intestinal microﬂora, restoring intestinal

USE OF RECOMBINANT LAB IN HUMAN HEALTH

3

barrier function, and alleviating inﬂammatory responses has led to their proposed use for therapy and management of immunopathological disorders such as Crohn’s disease, ulcerative colitis, and pouchitis (Campieri and Gionchetti, 1999; Marteau et al., 2003; Shanahan, 2001). Over the last few years, a coordinated effort involving several European laboratories and combining a number of interdisciplinary research strategies has set out to provide experimental evidence for the efﬁcacy of different prototype health products based on the mucosal administration of recombinant LAB (http://www.labdel.eu.com). This consortium also aims to further advance technology for LAB delivery and the safe containment of genetically modiﬁed organisms in order to increase the range of potential applications for recombinant LAB and to accelerate their commercial development. In this review, the consortium discusses the potential and future opportunities for the use of recombinant lactic acid bacteria in human health. II. Use of Recombinant LAB to Prevent Infectious Diseases Although vaccines against several major pathogens are in common use, morbidity and mortality from infectious disease remains a considerable burden worldwide. According to World Health Organization (WHO) estimates, infectious diseases caused 14 million deaths in 2001, accounting for 26% of global mortality. This situation is steadily becoming worse because of increasing microbial resistance to antimicrobial drugs worldwide, especially Streptococcus pneumoniae, enterococci, and Gram-negative enteric pathogens (Hakenbeck et al., 1999; Kariuki and Hart, 2001; Threlfall, 2002; Walsh, 2000). Recent evidence is also linking a growing number of infectious agents to an increased risk of cancer, blurring the traditional distinctions between chronic and communicable diseases. In addition, behavioral changes over the last two decades have seen the emergence of new pathogens as well as the re-emergence of ‘‘old’’ infectious diseases thought to be extinct (van Ginkel et al., 2000). Some of the existing vaccines are also not without problems, and more-effective versions need to be developed. In some of the poorest countries, existing vaccines are unaffordable to those most in need, a situation that might be helped by strategies to reduce the cost of vaccination (e.g., through avoidance of the use of syringes and needles). Orally administered vaccines are also an attractive goal in developed countries where combined childhood vaccination schedules generally administered via nonmucosal or percutaneous injections are becoming increasingly more complex to formulate and combine. The main advantage of mucosal immunization (e.g., oral,

4

HANNIFFY et al.

nasal, rectal) is the potential to elicit both mucosal and systemic immune responses that would enhance the efﬁcacy of many vaccines. It is not surprising therefore that the development of effective strategies for the delivery of vaccine antigens to the mucosal tissues has received considerable attention over the past decade (Michalek et al., 1994; O’Hagan, 1994; Wells and Pozzi, 1997). The use of recombinant bacteria as carrier systems has received particular attention, most vectors under development being derived from attenuated pathogenic bacteria (Bumann et al., 2000; Gicquel, 1995; Levine et al., 1996; Michalek et al., 1994; Roberts et al., 1994; Sirard et al., 1999; Stahl et al., 1997). In addition to the potential to revert to virulence with associated risk of infection and a public opinion sensitive to the use of recombinant organisms, variation in the immunogenicity of the different attenuated strains has constituted a major problem, and it has been difﬁcult to reach the right balance between the level of attenuation (i.e., lack of disease symptoms) and immunogenicity (i.e., efﬁcacy). An additional concern with the use of attenuated pathogens is that they may still be sufﬁciently virulent to cause disease in infants, the elderly, or immunocompromised individuals. Therefore the use of noninvasive and nonpathogenic lactic acid bacteria as vehicles for mucosal delivery of vaccine antigens and other therapeutic molecules is an attractive concept. In addition to their application as vaccine vehicles (discussed below), lactic acid bacteria can be used to deliver anti-infectives or antimicrobial products in situ. An example is the use of recombinant Lactobacillus strains expressing anti-idiotypic single-chain Fv antibody to a Streptococcus mutans adhesin (SAI/II) in the prevention of dental caries in a rat model (Kruger et al., 2002). Similarily, recombinant Streptococcus gordonii expressing the microbiocidal molecule H6, which is an anti-idiotypic single chain antibody mimicking a yeast killer toxin, was shown to demonstrate candidacidal activity in a rat model (Oggioni et al., 2001). A. RECOMBINANT LAB

AS

VACCINE DELIVERY VEHICLES

Many lactic acid bacteria are used by the food industry and have a history of safe use or are classiﬁed as nonpathogenic. They can therefore be given orally in relatively large doses without risk of potential side effects such as those associated with live attenuated bacterial vectors (Medina and Guzman, 2001). In addition, many LAB strains are acid resistant (Kociubinski et al., 1999) and adhere to mucosal epithelium (Morita et al., 2002), properties that may be beneﬁcial for

USE OF RECOMBINANT LAB IN HUMAN HEALTH

5

their use as mucosal delivery vehicles. Expressing vaccine antigens and other therapeutic molecules in recombinant LAB would obviate the need to purify the active component, reducing the overall production and delivery costs. Moreover, these bacteria would ideally be given orally and thus would be amenable to large-scale vaccination programs in populations at risk. In addition to providing protection, LAB-based vaccines could also potentially elicit protective mucosal immune responses against pathogens that are also present as components of the normal microﬂora in asymptomatic carriers, thus having an impact on carriage rates. While evidence suggests that LAB exhibit low intrinsic immunogenicity when administered by the mucosal route, there is no doubt that speciﬁc strains of LAB do possess immunoadjuvant properties and can enhance antigen-speciﬁc immune responses when administered in combination with antigen (Link-Amster et al., 1994). It is also clear that the immunoadjuvant properties of LAB vary for different species, which has obvious implications for the selection of speciﬁc LAB as vaccine vehicles (Maassen et al., 2000; Repa et al., 2003). In addition, it is important to consider that not all LAB are commensals in humans and that speciﬁc strains may be associated with different mucosal sites and environmental niches within the host (Wells and Mercenier, 2003). While selection of an appropriate model species is therefore critical, their inherent diversity can also be seen as an advantage. Their ability to survive at different mucosal surfaces may increase opportunities for use of recombinant LAB as vaccine vehicles against a wider range of diseases. Similarly, differences in the immunomodulatory capacities of different LAB create additional possibilities for tailoring the choice of vehicle to meet the requirements for immunity or to modulate immune outcomes in the treatment of various immunopathological diseases. A theoretical model of the immunomodulatory effects of LAB is shown in Fig. 1. Importantly, the last decade has seen the development of the genetic tools necessary for expression of heterologous proteins in an increasing number of LAB (for reviews see de Vos, 1999a; Langella and Le Loir, 1999; Nouaille et al., 2003; Pouwels et al., 2001; Reuter et al., 2003; Wells and Mercenier, 2003; Wells and Schoﬁeld, 1996). Expression systems were ﬁrst developed in the model LAB strain Lactococcus lactis, a noninvasive, noncolonizing bacterium widely used to produce cheese curds by the fermentation of milk. These included constitutive and inducible promoters that allowed efﬁcient high-level production of antigen under various conditions (de Vos, 1999a; Israelsen et al., 1995a; Kuipers et al., 1997; Wells and Schoﬁeld, 1996; Wells et al.,

6

HANNIFFY et al.

FIG. 1. Theoretical model for Th1-enhancing effects of antigen expressing lactic acid bacteria on immune responses based on a previous illustration by Cross et al. (2001). The production of pro-interferon cytokines IL-12 and IL-18 and interferon alpha (INF) by the interaction of certain strains of lactic acid bacteria with antigen presenting cells (APC) (e.g., dendritic cells and macrophages) is supported by in vitro co-culture studies and in vivo animal models. Cell wall components of the Gram-positive lactic acid bacteria may induce pro-Th1 cytokine production through Toll-like receptors (TLRs) and possibly also through other surface receptors. The interaction of LAB with other cell types, including epithelial cells, may also inﬂuence the outcome of exposure to LAB and prevent tissue inﬂammation.

1993a). In more recent times, constitutive and regulated expression systems, some of which have been adapted from Lc. lactis (Kleerebezem et al., 1997; Pavan et al., 2000), have also been used to express heterologous proteins in other LAB, including species of Lactobacillus that have been shown to colonize different niches in both animals and humans (Chang et al., 2003; Grangette et al., 2001, 2002; Hols et al., 1997b; Maassen et al., 1999; Oliveira et al., 2003; Pouwels et al., 1996; Rush et al., 1997; Scheppler et al., 2002). These systems have been further adapted such that the expressed protein can be secreted by LAB or anchored to the bacterial surface (Bernasconi et al., 2002; Dieye et al., 2001; Grangette et al., 2002; Kruger et al., 2002; Oliveira et al., 2003; Reveneau et al., 2002; Ribeiro et al., 2002; Wells et al., 1993b). These improvements have ensured that LAB can be used to effectively deliver antigen to mucosal surfaces either in particulate form or as soluble antigen secreted into the surrounding lumen (Norton et al., 1996). The development of efﬁcient expression systems for LAB has enabled several vaccine antigens derived from a variety of mucosal pathogens to be successfully expressed in different LAB (Mercenier et al., 2000; Seegers, 2002; Thole et al., 2000; Wells and Mercenier, 2003; Wells et al., 1996).

USE OF RECOMBINANT LAB IN HUMAN HEALTH

7

Pioneering research carried out by several European laboratories demonstrated that immunization with Lc. lactis and selected species of lactobacilli expressing the tetanus toxin fragment C (TTFC) antigen were capable of eliciting antigen-speciﬁc secretory IgA and serum antibodies that were protective against lethal challenge with tetanus toxin (Grangette et al., 2001, 2002; Norton et al., 1997; Robinson et al., 1997; Wells et al., 1993a). In a continuation of this work, it was also shown that Lc. lactis could express and secrete the murine cytokines IL-2 and IL-6, which enhanced antigen-speciﬁc immune responses when co-expressed with TTFC (Steidler et al., 1998). These experiments demonstrated that biologically active molecules such as cytokines, enzymes and other molecules could be effectively delivered to mucosal surfaces by using recombinant LAB. This application was further exploited by Steidler et al. (2000), who showed that Lc. lactis strains expressing and secreting murine IL-10 could be used to treat inﬂammation in two different mouse colitis models. This work has now led to the development of a biologically contained Lc. lactis strain secreting human IL-10 (Steidler et al., 2003), which has been approved by Dutch authorities for use in a small clinical trial as an experimental therapy for use in humans with inﬂammatory bowel diseases (IBD) (see also Section IV). The aforementioned discoveries demonstrate that there is considerable potential to develop health-based products based on oral delivery of vaccine and other therapeutics when using LAB. Several recent publications have now documented the use of prototype LAB-based vaccines for production and delivery of various molecules targeting a range of mucosal pathogens (Bermudez-Humaran et al., 2002, 2003; Enouf et al., 2001; Gil et al., 2001; Gilbert et al., 2000; Lee et al., 2001; Ribeiro et al., 2002; Xin et al., 2003; Zegers et al., 1999), some of which are discussed in the following sections. B. INFECTIONS OF THE RESPIRATORY TRACT As well as being a reservoir for potentially pathogenic bacteria such as Haemophilus inﬂuenzae, Morexella catarrhalis, Pseudomonas aeruginosa, Staphylococcus aureus, S. pneumoniae, and beta-hemolytic streptococci, the respiratory tract is also susceptible to serious infections caused by Bordetella pertussis, Mycobacterium tuberculosis, Mycoplasma pneumoniae, inﬂuenza virus, and respiratory syncitial virus (RSV). All these organisms initiate disease at the mucosal surface of the respiratory tract, and thus the efﬁcacy of the host’s response to these infections is dependent on optimal local immune responses at this site. However, vaccines available for diseases caused by many

8

HANNIFFY et al.

of these pathogens have limitations in accessibility or efﬁcacy, highlighting the need for improvements in approaches and products. The need for more-effective control strategies is heightened by reports of increasing antibiotic resistance among isolates from studies on carriage in healthy individuals and from clinical specimens (Metlay and Singer, 2002). Until recently there has been little evidence that probiotic LAB might help prevent respiratory infections despite numerous studies showing their beneﬁcial effects against gastrointestinal infections. However, a number of studies have now shown that the oral application of various LAB and/or their bioactive components can enhance the antimicrobial activity of pulmonary natural killer cells (NK) and macrophages, increase IgAþ cells at the bronchial level, and induce cytokine production by nasal lymphocytes (Hori et al., 2002; Matar et al., 2001; Moineau and Goulet, 1997; Perdigon et al., 1999). Moreover, a number of studies carried out in animals and in humans have shown that LAB may attenuate infections caused by respiratory pathogens. When using animal models, the oral application of different LAB strains increased phagocytic activity in macrophages and enhanced clearance of S. pneumoniae (Cangemi de Gutierrez et al., 2001) and P. aeruginosa (Alvarez et al., 2001). Similarly, LAB were also shown to enhance cellular immunity and reduce inﬂuenza virus titers in aged mice (Hori et al., 2001, 2002). These results suggest that selected strains of LAB are capable of preventing respiratory tract infections, perhaps by microbial exclusion and/or by mediating nonspeciﬁc immune responses. Interestingly, Cangemi de Gutierrez et al. (2001) also showed that protected mice had increased numbers of lymphocytes in their lamina propria, as well as higher levels of antibodies binding to S. pneumoniae, indicating that a speciﬁc immune response may have been elicited. Some studies carried out in humans have further demonstrated the potential of using probiotics to confer protection against respiratory diseases, particularly in young children. In a randomized, double-blind, placebo-controlled study, the consumption of milk containing the probiotic strain Lactobacillus rhamnosus GG modestly reduced the incidence of respiratory infections as well as their severity in young children (1–6 yr) (Hatakka et al., 2001). In a similar study, a live dietary supplement containing Lactobacillus acidophilus and Lactobacillus casei suppressed pneumonia and decreased bronchitis in 6- to 24-month-old children (Rio et al., 2002). These studies have been supported by Gluck et al. (2003), who demonstrated that nasal colonization by pathogenic bacteria (S. aureus, S. pneumoniae, and beta-hemolytic streptococci) was reduced in individuals who ingested milk supplemented with a cocktail of

USE OF RECOMBINANT LAB IN HUMAN HEALTH

9

probiotic LAB as compared with individuals fed standard yogurt (Gluck and Gebbers, 2003). While only preliminary, these studies do suggest that oral administration of probiotic LAB may help to reduce the occurrence and severity of respiratory infections in both children and adults. These ﬁndings have important implications for the development of recombinant LAB vaccines against respiratory pathogens, because they suggest that LAB themselves may have an effect on immune function that could potentially act as an adjuvant during vaccination. Previous work has already demonstrated that intranasal administration of recombinant Lc. lactis and Lb. plantarum strains expressing TTFC could elicit signiﬁcant antigen-speciﬁc IgA in broncheoalveolar ﬂuids as well as antigen-speciﬁc serum antibody and T cell responses in mice (Grangette et al., 2001; Norton et al., 1997). Moreover, these immune responses, which were dose dependent, were sufﬁcient to protect against lethal challenge with injected tetanus toxin. Since then, reports of the construction of LAB strains expressing antigens from a number of respiratory pathogens including S. pneumoniae (Gilbert et al., 2000), Bacillus anthracis (Zegers et al., 1999) and Bordetella pertussis (Lee et al., 1999, 2002) were reported but await proper testing in relevant animal models of disease. More recently, prototype vaccine strains of Lc. lactis and Lb. plantarum expressing pneumococcal PspA antigen were constructed and evaluated in a respiratory challenge model for S. pneumoniae (Hanniffy et al., 2004). PspA, a surface protein and virulence factor found on all isolates of S. pneumoniae (Briles et al., 1998; Crain et al., 1990), is highly immunogenic and is considered a vaccine candidate because it has been shown to confer protection against virulent isolates in different animal models of pneumococcal infection (Bosarge et al., 2001; Briles et al., 1996, 2000). Intranasal administration of LAB expressing PspA have now been shown to elicit mucosal and systemic immune responses that protect against lethal challenge with virulent pneumococcci (Hanniffy et al., 2004). C. INFECTIONS OF THE GASTROINTESTINAL TRACT As constituents of the normal microﬂora, LAB have long been recognized as playing a role in maintaining oral tolerance and homeostasis in the gut. Additionally, LAB have been used therapeutically to protect against gastrointestinal infections, the most notable example being the use of Lactobacillus GG to treat acute infectious diarrhea in children (Rosenfeldt et al., 2002a,b; Szajewska and Mrukowicz, 2001). The exact

10

HANNIFFY et al.

mechanisms that confer signiﬁcant clinical beneﬁt following administration of probiotic LAB are uncertain, but various intrinsic properties of the bacteria have been proposed as an explanation for the beneﬁcial effects (Heyman, 2000). In addition to restoring normal intestinal microﬂora, LAB may help to eliminate enteric pathogens by reinforcing intestinal barrier function, increasing mucosal secretory IgA and humoral immune responses, and boosting speciﬁc and nonspeciﬁc immunity. In vitro experiments with intestinal epithelial cell lines have demonstrated that LAB probiotic strains can prevent adhesion and invasion by pathogenic bacteria and enhance barrier function in naı¨ve epithelial cells (Lee et al., 2003; Resta-Lenert and Barrett, 2003). Various studies have also shown that LAB can intervene by binding to receptors on gastric and intestinal epithelium typically used by pathogens to gain entry into host cells (Mukai et al., 2002; Neeser et al., 2000). In addition, certain LAB produce relatively large amounts of organic acids (Aiba et al., 1998; Koga et al., 1998; Midolo et al., 1995) and/or bacteriocin-like substances (Kim et al., 2003; Lee et al., 2003; Strus et al., 2001) that have been implicated in the inhibition of Helicobacter pylori, Campylobacter spp., and Clostridium difﬁcile. While the extent to which these components can cause an effect in the gastrointestinal (GI) environment remains questionable, it is becoming increasingly clear that by modulating immune responses, LAB play a vital role in protecting against pathogens. Experiments carried out in animal models have demonstrated the protective capabilities of different LAB against pathogenic Escherichia coli (Ogawa et al., 2001; Shu and Gill, 2001), Salmonella typhimurium (Shu et al., 2000), Helicobacter pylori, and Clostridial species. Shu et al. (2000) demonstrated that compared with control mice, Lb. rhamnosus HN001fed mice exhibited lower morbidity and bacterial translocation rates when challenged with E. coli O157:H7 (Shu et al., 2000), which was associated with signiﬁcantly higher levels of pathogen-speciﬁc IgA and blood leukocyte phagocytic activity in the intestines of these mice. In a separate study, mice administered dietary Biﬁdobacterium lactis demonstrated increased speciﬁc and nonspeciﬁc immune responses and reduced intestinal infection by S. typhimurium (Shu and Gill, 2001). These results would suggest that LAB may be capable of enhancing local immune responses that might play an important role in protecting against infectious disease. Using genetic engineering to further enhance probiotic LAB strains holds much promise in the prevention and treatment of infectious diseases of the GI tract. Prototype recombinant Lc. lactis vaccine strains expressing antigen have already been developed against

USE OF RECOMBINANT LAB IN HUMAN HEALTH

11

H. pylori (Lee et al., 2001) and rotavirus (Enouf et al., 2001; Gil et al., 2001). Oral administration of recombinant Lc. lactis expressing the H. pylori urease subunit B antigen did stimulate low levels of antigenspeciﬁc immune responses in serum, but no protection was observed during challenge with H. pylori (Lee et al., 2001). The inﬂuence of different LAB strains and choice of antigen is likely to be critical to any vaccine delivery strategy, particularly when dealing with the GI tract, where persistence and survival of different strains can differ widely (Vesa et al., 2000). Other variables such as interaction of the strain with mucus, intestinal epithelium, and lymphoid cells, as well as the location and amount of antigen expressed in the bacteria, are also likely to inﬂuence immune outcomes. Except for recent studies on the effect of antigen quantity on level of antibody response, the impact of many of the aforementioned variables has yet to be assessed experimentally (Wells and Mercenier, 2003). D. INFECTIONS OF THE UROGENITAL TRACT LAB-based vaccine delivery systems may also be appropriate for preventing and treating bacterial vaginosis (BV) as well as speciﬁc pathogen-related infections of the urogenital tract (Reid and Bruce, 2001; Reid and Burton, 2002). These include human immunodeﬁciency virus (HIV), Chlamydia, herpes simplex virus (HSV), papillomavirus, Treponema palladium, and Trichomonas vaginalis, as well as diseases caused by bacterial pathogens such as Neisseria gonorrhoeae, group B Streptococcus, and enteropathogenic E. coli. Lactobacilli are dominant among microﬂora associated with the urogenital tract of healthy women but are almost completely absent in patients who develop most forms of urinary tract infections. Depletion or disturbance of vaginal Lactobacillus sp. has been associated with the development of bacterial vaginosis as well as increased risk of acquiring HIV and other sexually transmitted diseases (Hillier et al., 1993; Taha et al., 1998; Wiesenfeld et al., 2003). There is evidence that lactobacilli can prevent urinary tract infections by microbial exclusion; by producing bacteriocins, biosurfactants, hydrogen peroxide, and coaggregation molecules; by maintaining a low pH; or a combination of these factors (Reid, 2002). In vitro studies have shown that adhesive lactobacilli can inhibit growth and attachment of uropathogenic bacteria to uroepithelial cells in a strain-speciﬁc manner (Boris et al., 1998; McGroarty and Reid, 1988; Osset et al., 2001). Similarly in mice, an endogenous population of Lb. casei in the urinary tract prevented colonization by uropathogenic bacteria in the absence of pathogen-speciﬁc immune

12

HANNIFFY et al.

responses (Reid et al., 1985). More recently, a randomized, placebocontrolled trial of 64 healthy women showed that daily consumption of a capsule containing the probiotic strains Lb. rhamnosus GR-1 and Lactobacillus fermentum RC-14 reduced colonization of the vagina by potential pathogenic bacteria and yeast (Reid et al., 2003). While this observed probiotic effect appeared to be mediated predominantly through microbial exclusion and/or production of antagonistic byproducts, an immunological component cannot be discounted. Although further studies are required to determine if LAB can induce protective immune responses in the vagina and bladder, these ﬁndings do suggest that daily intake of probiotic LAB could provide a natural, safe, and effective means of stabilizing the continually ﬂuctuating vaginal ﬂora and lower the risk of infection in healthy individuals as well as those prone to urogenital disease. For women, direct application of a LAB-based vaccine by the intravaginal route leading to the induction of local antigen-speciﬁc immune responses could provide increased protection against urogenital tract infections, reduce pathogen carriage, and help block sexual transmission of pathogens. Previous ﬁndings, however, suggest that the urogenital tract may have a limited ability to mount immune responses to epithelial infections. This is evident from clinical observations that pathogens such as N. gonorrhoeae can be contracted repeatedly in the absence of effective immunity from previous infections (Russell, 2002). In addition, local immunization in order to induce local immune responses would also be effected by hormonal variations that are more pronounced in the urogenital tract than at other mucosal sites and are known to have profound effects on susceptibility to sexually transmitted diseases (Gallichan and Rosenthal, 1996; Gillgrass et al., 2003; Kaushic et al., 2000). While mucosal immunization usually results in higher IgA at the site where immunization occurs, there is evidence that this is not the case in the vagina and that other routes of administration may be more effective. Experiments carried out in mice by Wu et al. (2000) showed that intranasal immunization could induce substantially higher levels of IgG and IgA in vaginal ﬂuids and serum as compared with intravaginal immunization. Such compartmentalization within the common mucosal immune system occurs also in humans, where the route of administration can signiﬁcantly inﬂuence immune responses at remote mucosal surfaces (Kantele et al., 1998). This has been demonstrated by Kutteh et al. (2001), who showed that induction of antibodies (S-IgA) in the female genital tract was best achieved by oral immunization followed by a rectal administration (Kutteh et al., 2001).

USE OF RECOMBINANT LAB IN HUMAN HEALTH

13

LAB would offer greater ﬂexibility as vaccine vectors in that they can be administered by whatever route is deemed most suitable for a given clinical situation. The nasal route of administration, for example, may be the best inductive site for a common mucosal immune response and could provide a useful strategy for inducing more-potent, longer-lasting immune responses not just in saliva but in remote secretions (e.g., vaginal ﬂuids) as well as in serum (Wu et al., 2000). Recombinant Lc. lactis and S. gordonii vaccine strains have already been developed against human papillomavirus type 16 (HPV-16) (Bermudez-Humaran et al., 2002, 2003; Cortes-Perez et al., 2003) and HIV (Oggioni et al., 2001; Xin et al., 2003). Similarly, Lactobacillus vaccine strains against Chlamydia psittaci and HIV have also been engineered (Zegers et al., 1999). Xin et al. (2003) also showed that Lc. lactis expressing the envelope protein of HIV protected mice against intraperitoneal challenge by an HIV Env-expressing vaccinia virus. While oral and intranasal administration of a number of these LAB vaccine strains has been effective in inducing antigen-antibody responses in serum, their ability to induce local and remote mucosal responses has not yet been investigated (Cortes-Perez et al., 2003; Xin et al., 2003). Interestingly, vaginal colonization of mice with recombinant strains of S. gordonii, expressing HPV and HIV antigens, has been shown to induce antigen-speciﬁc vaginal IgA as well as serum IgG (Medaglini et al., 1998). These strains were also able to induce local and systemic immune responses when repeatedly administered (three inoculations) to the vagina of Cynomologus monkeys (Di Fabio et al., 1998). These ﬁndings would indicate that LAB possess adjuvant properties that may be capable of eliciting antigen-speciﬁc local as well as systemic immune responses even when applied directly to the vagina. If persistence and colonization of the vagina are essential components of a LAB-based vaccine or therapy, intravaginal immunization would therefore be optimal. Such a strategy would be particularly preferable where the aim is to provide passive immunity in situ. This has recently been addressed by Chang et al. (2003), who engineered Lactobacillus jensenii, a natural human vaginal isolate, to express and secrete the HIV binding protein CD4, which bound the HIV gp120 protein and inhibited HIV-1 entry into target cells in vitro. Lb. jensenii efﬁciently colonizes the vaginal mucosa of women where it exists as part of a natural ‘‘bioﬁlm’’ composed of bacteria and extracellular matrix materials. It is hoped that resident LAB engineered to express HIV binding proteins that are either surface associated or secreted into the surrounding ‘‘bioﬁlm’’ and mucus layer could effectively neutralize HIV particles by impeding viral access to epithelial cells and prolonging exposure to viral inactivating

14

HANNIFFY et al.

substances that are produced naturally. The potential of this approach has been demonstrated by reports detailing the development of prototype S. gordonii vaccine strains expressing molecules that are microbiocidal against candidiasis (Beninati et al., 2000) and HIV (Giomarelli et al., 2002). Such LAB vaccines, perhaps in tablet form (Bonetti et al., 2003; Maggi et al., 2000; Mastromarino et al., 2002), could be self-administered intermittently to provide ongoing protection against these pathogens. Vaccines such as this could potentially be further improved by combining different LAB strains with different antimicrobial, adhesive, and biochemical characteristics to increase their effectiveness in preventing and treating urogenital diseases. Recombinant LAB vaccines would also be suitable for maternal vaccination strategies aimed at preventing disease in neonates, such as those caused by group B Streptococcus (GBS) and selected strains of E. coli. While antibiotic therapy has proved effective in reducing the incidence of neonatal disease in developed countries, other complications such as premature rupture of the membranes, premature birth, low-birth-weight babies, or stillbirth have been attributed to GBS and other infections of the urogenital tract. A LAB-based vaccine against GBS and other pathogens could reduce the level of colonization in the mother and provide an important ﬁrst line of defense against the pathogen at mucosal surfaces. In addition, colostrum antibody and transplacentally transferred serum IgG antibodies against GBS could confer immune protection to the newborn. Recombinant strains of Lc. lactis and Lb. plantarum have already been developed that express antigen that has been shown to be protective in an invasive animal model of GBS infection (Seepersaud et al., 2004). These strains are now being tested in animal models for their ability to elicit protective mucosal and systemic responses (Hanniffy and Wells, personal communication). III. Potential for Immune Modulation of Type I Allergy Using Recombinant LAB The prevalence of type I allergy has constantly increased within recent years, leading to the fact that currently up to 25% of the population in industrialized countries suffer from allergic symptoms such as allergic rhinoconjunctivities, allergic asthma, or atopic dermatitis. Besides atopic disposition (Marsh et al., 1994) and allergenic molecules (Scheiner and Kraft, 1995), there are various environmental factors such as air pollutants (von Mutius et al., 1994), changes in nutrition

USE OF RECOMBINANT LAB IN HUMAN HEALTH

15

(Bottcher et al., 2000), or ‘‘sterile’’ lifestyles in industrialized countries without bacterial epidemics (Ring, 1997) that may have a causal relationship with the increasing prevalence of the disease. The pathway of IgE production is well described for mice and man and is basically inﬂuenced by the reciprocal relationship between Th2 cytokines, mainly IL-4, the switching factor for IgE class, and Th1 cytokines (e.g., interferon gamma [IFN-], which counteracts the activity of IL-4) (Mosmann and Coffman, 1989). In atopic individuals an imbalance favors the production of IL-4, leading to a bias of Th2 responses with an overproduction of IgE against allergenic proteins (Romagnani, 1994). These molecules lead to cross-linking of the allergen-speciﬁc IgE bound to the surface of mast cells via speciﬁc FC-receptors, thereby triggering the release of anaphylactic mediators, leading to the characteristic allergic symptoms mentioned previously. Until now, speciﬁc immunotherapy (SIT) has been the treatment of choice against type I allergy, performed by injecting increasing doses of allergens (Bousquet et al., 1998). Although SIT is effective in the majority of treated patients, there are certain drawbacks, such as frequent injections and long duration of the treatment, leading to poor compliance in patients. Moreover, aluminium salts, which are used as adjuvants in SIT, are potent inducers of Th2 responses, which might reduce the efﬁcacy of the treatment. Thus, there is increasing interest in improving immunotherapy by using Th1-promoting adjuvants (Wheeler and Woroniecki, 2001) and/or administration via a less invasive route, such as mucosal delivery (Morris, 1999). A. ROLE OF THE INDIGENOUS MICROFLORA The indigenous microﬂora plays an important role in anti-infectious resistance by competitive interaction with pathogenic bacteria but is also important for directly inﬂuencing immune responses. This has been demonstrated in animals reared under sterile conditions (germfree animals), in which it has been shown that systemic and local immune responses are more difﬁcult to establish and that, in particular, the induction of oral tolerance is unstable and short-lived (Cebra et al., 1999). Based on these ﬁndings, an imbalance of the composition of the indigenous microﬂora is believed to play a role in the development of inﬂammatory diseases, such as intestinal bowel disease (Duchmann et al., 1995) and allergies (Rautava and Isolauri, 2002; Wiedermann, 2003). Indeed, differences in the intestinal colonization pattern between children of ‘‘Western lifestyle countries’’ with a high prevalence of allergies and of economically low developed countries

16

HANNIFFY et al.

where allergies are less common has also been reported (Bottcher et al., 2000). The ‘‘hygiene theory’’ proposes that an increasingly hygienic lifestyle has contributed to the increasing prevalence of allergic disease caused by intestinal colonization, with a limited range of microbes (Strachan, 1989). In the last few years, the possible role of speciﬁc LAB strains in the prevention of allergic diseases has become more evident. In particular the relationship between the composition of the intestinal ﬂora and the prevalence of allergic diseases has been epidemiologically documented. One study showed that infants from countries with a high prevalence of allergy, such as Sweden and England, have a lower level of intestinal colonization with certain LAB strains than children from countries where allergic diseases are less prevalent, such as Estonia and Nigeria (Bjorksten et al., 1999; Sepp et al., 1997; Simhon et al., 1982). Moreover, it was recently demonstrated that oral administration of a particular LAB strain (Lb. rhamnosus GG) led to reduced atopic dermatitis in children with a positive family history of type I allergy (Kalliomaki et al., 2001), indicating that LAB can directly or indirectly exert an anti-allergic effect. B. ANIMAL MODEL OF TYPE I ALLERGY Among the numerous inhalant allergens, tree pollen of the white birch Betula verrucosa is one of the most important sources responsible for eliciting allergic symptoms. The major allergen of birch pollen is Bet v 1, a 17-kD molecule, to which 95% of birch pollen allergic patients (and 60% exclusively) display IgE binding reactivity. Bet v 1 was the ﬁrst pollen allergen to be cloned, sequenced, and produced as a recombinant protein in E. coli (Breiteneder et al., 1989) and to have its crystalline structure determined (Gajhede et al., 1996). Recombinant Bet v 1 has also been shown to possess biological properties equivalent to those exhibited by the natural Bet v 1 molecule (Ferreira et al., 1993). Moreover, immune responses to Bet v 1 have been characterized at the B and T cell level in atopic and nonatopic individuals (Ebner et al., 1995). An animal model of allergic sensitization to birch pollen and its major allergen Bet v 1 has been developed in BALB/c mice that have been identiﬁed as high IgE responders to this allergen (Bauer et al., 1997; Wiedermann et al., 1998). The established standard sensitization scheme is based on an intraperitoneal injection of recombinant (r) Bet v 1 adsorbed to aluminium hydroxide (Al(OH)3), followed by an aerosol treatment with natural birch pollen extract (BP). This sensitization

USE OF RECOMBINANT LAB IN HUMAN HEALTH

17

procedure leads to high allergen-speciﬁc IgE/IgG1 antibody levels, positive immediate type skin reactions in vivo and Th2-like immune responses in vitro, eosinophilic inﬁltration within the lungs, as well as broncheoalveolar and airway hyper-responsiveness. Thus the immune responses represent an immunological state comparable to that of human type I allergy/asthma (Wiedermann et al., 1998). Using this model, several mucosal adjuvants have been tested for their capacity to modulate an allergic immune response. In this respect it has been demonstrated that immunomodulation can be achieved by using mucosal adjuvants such as cholera toxin (Wiedermann et al., 1998), cholera toxin subunit B (Wiedermann et al., 1999), or certain bacterial components (CpG-motifs) (Jahn-Schmid et al., 1999). C. USE OF LAB

FOR

PROPHYLAXIS AND THERAPY OF TYPE I ALLERGY

Experiments carried out in mice have shown that IgG1 and IgE antibody levels could be reduced when certain LAB strains were injected or orally applied with a particular antigen/allergen (Matsuzaki et al., 1998; Shida et al., 2002). In light of these studies, experiments were carried out to investigate the capacity of two LAB strains (Lactococcus lactis MG1363 and Lactobacillus plantarum NCIMB8826) to prevent or modulate allergic immune responses. Both LAB strains induced high levels of Th1 cytokines IL-12 and IFN- in naı¨ve murine spleen cell cultures possibly via the mechanisms depicted in Fig. 1. In the murine birch pollen allergy model, intranasal or oral co-application of Lc. lactis or Lb. plantarum with recombinant Bet v 1, prior to or after allergic sensitization, led to increased levels of allergen-speciﬁc IgG2a antibodies and in vitro IFN- production, indicating a shift toward Th1 responses. Successful immunomodulation by the mucosal pretreatment was further demonstrated by suppression of allergen-induced basophil degranulation (Repa et al., 2003). These results, based on induction of counter-regulatory Th1 responses, indicated that combined mucosal application of LAB with a speciﬁc allergen could provide an effective prophylactic and therapeutic strategy against allergy. The construction of recombinant LAB for local delivery of an allergen could further enhance this protective/immunostimulatory effect and could represent a useful tool for mucosal vaccination against type I allergy. Recently it has been shown that treatment with a recombinant LAB strain expressing the house dust mite allergen Der p1 reduced levels of allergen-speciﬁc IL-5 (a Th2 cytokine) in sensitized mice (Kruisselbrink et al., 2001). Similarly, recombinant Lc. lactis and Lb. plantarum strains expressing Bet v 1 have now also been

18

HANNIFFY et al.

constructed and preliminary data indicate that mucosal administration of these strains prior to or after sensitization results in a signiﬁcant suppression of allergic immune responses (Mercenier and Wiedermann, personal communication).

IV. Opportunities for the Treatment of Inflammatory Bowel Diseases Using Recombinant LAB IBDs, as typiﬁed by ulcerative colitis and Crohn’s disease, are characterized by chronic dysregulation of inﬂammatory immune responses in the gastrointestinal tract. While the pathogenesis of IBD remains unclear, it is thought to involve complex interactions combining host genetic susceptibility, intestinal bacteria, and gut mucosal immune responses (Farrell and Peppercorn, 2002; Rath, 2003). It is now generally accepted that intestinal microﬂora provide the antigenic stimuli to deregulate mucosal immune responses in genetically susceptible hosts such that they become overly aggressive with reduced tolerance toward the indigenous microﬂora. The proposed use of probiotics including LAB for therapy and management of IBD has arisen from increasing evidence implicating indigenous bacteria in the pathogenesis of these diseases. Much of this evidence comes from transgenic animal models in which immunopathological disease is induced by the absence of immunologically important molecules but is dependent on the presence of a normal bacterial microﬂora (Blum and Schiffrin, 2003; Hudcovic et al., 2001; Strober et al., 2002). For example, mice with disrupted IL-2 and IL-10 genes or / T cell receptor mutants that normally develop chronic intestinal inﬂammation resembling ulcerative colitis in humans remain healthy in germ-free conditions (Strober et al., 2002). These studies conﬁrmed that IBD is likely to result from abnormal immune responses to normal intestinal microﬂora and demonstrated the importance of regulatory cytokines such as IL-10 in maintaining immune homeostasis at mucosal sites (Blum and Schiffrin, 2003). As a result, there has been increasing interest in using probiotics, particularly those with immunomodulatory capacities including the ability to induce immunoregulatory cytokines such as IL-10 and TGF-. This has led to suggestions that LAB and their associated anti-inﬂammatory effects may be capable of reinstating mucosal immune homeostasis and may provide an alternative strategy for intervention in IBD patients (Blum and Schiffrin, 2003).

USE OF RECOMBINANT LAB IN HUMAN HEALTH

A. ROLE OF LAB

IN INTESTINAL

19

BARRIER FUNCTION

One task of the gut is to act as a barrier between the external and internal environments to prevent the entrance of potentially harmful components. This barrier, which can be considered both as physical (paracellular permeability and protective action derived from mucus) and functional (mucosal immune system), is strongly disrupted in inﬂammatory states. A decrease in mucosal barrier function consistently occurs in experimental colitis as well as in human IBD (Rath, 2003). One of the main actions of probiotics concerns the reinforcement of the intestinal mucosal barrier, an activity which can confer intestinal anti-inﬂammatory properties to some probiotics (see review, Fioramonti et al., 2003). For example, treatment with Lb. reuteri or Lb. plantarum reduced the level of intestinal permeability in a rat model of methotrexate-induced enterocolitis (Mao et al., 1996). In addition, as well as decreasing intestinal myeloperoxidase levels, the administration of LAB re-established the intestinal microecology and reduced bacterial translocation to extra-intestinal sites. The ability of LAB to enhance barrier function and reduce bacterial translocation may play an important role in preventing subsequent activation of inﬂammatory responses. Biﬁdobacterial supplementation has been shown to reduce the incidence of necrotizing enterocolitis in mice by preventing bacterial translocation and subsequent activation of inﬂammatory mediators such as plasma endotoxin and intestinal phospholipase A2 (Caplan et al., 1999). In another study, Shiba et al. (2003) provided evidence that treatment of Biﬁdobacterium vulgatus– implanted mice with Biﬁdobacterium infantis abrogated increases in plasma B cells in the Peyer’s patch, probably by protecting the epithelium layer (including Peyer’s Patch) from invasion by B. vulgatus. LAB may also have a trophic action on colonic mucosa. This has been elegantly shown in rats, where an elemental liquid diet induced an atrophy of colonic mucosa (assessed by the rate of crypt cell production) that was signiﬁcantly improved by treatment with Lb. casei or Clostridium butyricum (Ichikawa et al., 1999). Strong interactions exist between mucus and colonic bacteria, and some actions of probiotics may involve this protective glycoprotein layer. The inability to degrade mucus by bacteria such as Lactobacillus acidophilus or Biﬁdobacterium biﬁdum could itself be seen as protective (Ruseler-van Embden et al., 1995). Indeed, in vitro experiments have shown that adherence by speciﬁc strains of Lactobacillus induces mucin gene expression and extracellular secretion of MUC3 by intestinal epithelial cells (Mack et al., 2003). Moreover, there was a direct correlation between

20

HANNIFFY et al.

increased mucin secretion and reduced adherence by enteropathogenic Escherichia coli. Similar interactions between LAB and host epithelial and other host cells may play an important role in protecting against translocation of antigenic stimuli that contribute to the pathogenesis of IBD.

B. LAB

AND INTESTINAL INFLAMMATION

A number of studies have shown that a select number of probiotic strains can reduce experimental colonic inﬂammation in animals. A study carried out by Fabia et al. (1993) was one of the ﬁrst to show that treatment with LAB could prevent the development of acetic acid– induced colitis in rats. Intracolonic administration of Lb. reuteri produced normal myeloperoxidase (MPO) activity levels and mucosal permeability and prevented the development of morphologic lesions (Fabia et al., 1993). These pioneering experiments were subsequently reproduced with other bacteria and in other models of experimental colitis. Lb. reuteri, but not Lb. rhamnosus, was shown to be effective in attenuating acetic acid–induced colitis in rats (Holma et al., 2001). Treatment with Biﬁdobacterium longum in mice (Fujiwara et al., 2003) and C. butyricum in rats (Araki et al., 2000) both reduced the severity of a colitis induced by dextran sulfate sodium (DSS). In addition, Lb. plantarum and Lb. reuteri ameliorated methotrexate-induced entercolitis in rats (Mao et al., 1996). The spontaneous development of colitis in IL-10–deﬁcient mice has also been ameliorated by treatment with Lb. plantarum (Schultz et al., 2002) or Lb. reuteri (Madsen et al., 1999a). In a recent study it has been shown that treatment with a solution of lysed E. coli ameliorated a colitis induced by DSS in mice (Konrad et al., 2003). To date, there is no evidence from animal models that LAB can adversely effect the development of colitis or augment severity of associated symptoms. While the mechanisms remain unclear, evidence from some of these studies indicates that administration of LAB may induce tolerance to bacterial antigens by down-regulating Th1 inﬂammatory cytokines (Konrad et al., 2003; Schultz et al., 2002). However, it is also possible that LAB may beneﬁcially interact with other commensal bacteria that may play a pathogenic role in IBD. A recent study carried out by Shiba et al. (2003) showed that B. infantis inhibited the growth of B. vulgatus, a putative pathogenic microbe in IBD. In addition, B. infantis also suppressed systemic immune responses to B. vulgatus in a gnotobiotic murine model (Shiba et al., 2003). This study would indicate that LAB-like B. infantis may protect

USE OF RECOMBINANT LAB IN HUMAN HEALTH

21

epithelial layers from invasion by other commensal microbes that are believed to play a pathogenic role in IBD. In recent years we have seen probiotics including LAB being used as an experimental therapy against IBD in humans. In a double-blind, placebo-controlled trial, a probiotic preparation (VSL#3), a cocktail of Biﬁdobacterium, Lactobacillus, and Streptococcus species, was shown to be effective in preventing ﬂare-ups of chronic ileal pouchitis (Gionchetti et al., 2000). The VSL#3 preparation has also been used for maintenance treatment of ulcerative colitis (Venturi et al., 1999) and in preventing postoperative recurrence of Crohn’s disease. More recently, Gionchetti et al. (2003) showed in double-blind, placebo-controlled trial that treatment with VSL#3 was effective in preventing the onset of acute pouchitis in patients with ileal pouch-anal anastomosis. Promising data have been also obtained with E. coli strain Nissle, which was found to have an equivalent to traditional treatment with mesalazine in maintaining remission of ulcerative colitis (Kruis et al., 1997; Rembacken et al., 1999). On a cautionary note, other studies carried out in humans have shown LAB/probiotic therapy to have little or no effect in the treatment of IBD. While feeding Lactobacillus GG to patients with a history of pouchitis and endoscopic inﬂammation did change the pouch bacterial ﬂora, the treatment proved ineffective as a primary therapy for a clinical or endoscopic response (Kuisma et al., 2003). Similarly, Lactobacillus GG treatment did not prevent endoscopic recurrence 1 year after curative resection for Crohn’s disease nor did it reduce the severity of recurrent lesions (Prantera et al., 2002). While some studies indicate that probiotics do have potential as therapies against IBD, they have provided little insight into the microbiological and immunological mechanisms that underlie these diseases. A recent study showed that ex vivo production of the proinﬂammatory cytokine tumor necrosis factor alpha (TNF-) by ileal– mucosal explants surgically removed from Crohn’s patients were down-regulated in the presence of Lb. casei and Lactobacillus bulgaricus but not by E. coli and Lactobacillus crispatus (Borruel et al., 2002). This would indicate that LAB interact differently with immunocompetent cells and have different capacities in modulating the production of pro-inﬂammatory cytokines such as TNF-, which play a key role in the pathogenesis of IBD. More clinical trials are therefore needed to evaluate different LAB strains as therapeutic preparations against the various manifestations of IBD. These studies must also establish the correct placement and dosage of probiotic required for any treatment regimen. Clearly more research is also needed to characterize and establish the mechanisms underlying the interactions of probiotic

22

HANNIFFY et al.

bacteria with the immune system and the role of intestinal microﬂora in homeostasis. The generated understanding will be of crucial importance for the future screening of lactic acid bacteria affecting immune functions and the development of recombinant strains with enhanced properties. C. RECOMBINANT LAB

AS A

THERAPEUTIC OPTION

While results from experiments carried out in animals are encouraging, the potential of probiotic therapy against IBD in humans remains uncertain. Various studies have now shown that in addition to proven efﬁcacy against experimental colitis in animals, probiotics including LAB also have other beneﬁcial actions on intestinal mucosa. Unfortunately, these studies have yet to establish a link between the two sets of data, and the mechanisms that underlie the anti-inﬂammatory action of probiotics remain ill-deﬁned. Until more is known about mucosaassociated microﬂora and the mechanisms that underlie inﬂammatory diseases, the use of probiotics in therapy of IBD will remain largely empirical. An alternative approach is to genetically engineer LAB to produce and deliver to the intestinal mucosa, molecules that have a therapeutic activity against IBD. Steidler and colleagues (2000) reported the construction of a recombinant Lb. lactis strain genetically engineered to secrete interleukin-10. Intravenous administration of IL10 has previously shown clinical efﬁcacy in the treatment of Crohn’s disease (van Deventer et al., 1997) but can cause side effects that prevent long-term use. In addition, IL-10 is sensitive to acid; therefore intestinal delivery is not an option. In addressing this problem Steidler et al. (2000) showed that Lb. lactis genetically engineered to secrete IL-10 could reduce DSS-induced colitis and prevent spontaneous colitis in IL-10–deﬁcient mice. Unfortunately, only a limited number of molecules or compounds display efﬁcacy against intestinal inﬂammation when infused in the colonic lumen. Nitric oxide (NO) is one such compound and was shown to reduce inﬂammation when infused into the colonic lumen of rats (Perner and Rask-Madsen, 1999). Lactobacillus farciminis, which produces NO in vitro, has also been shown to produce NO in the colon and to reduce experimental colitis when given orally to rats (Lamine et al., 2004). Another possibility would be to engineer LAB-secreting antioxidant enzymes such as catalase or superoxide dismutase (SOD), which may prove effective in removing free radicals such as superoxides and hydrogen peroxide produced by leukocytes, which are thought to play a role in amplifying inﬂammatory responses and subsequent mucosal damage in IBD patients

USE OF RECOMBINANT LAB IN HUMAN HEALTH

23

(Babbs, 1992). Indeed, some LAB possess signiﬁcant antioxidative activity and are intrinsically resistant to oxidative stress, properties that may help some isolates of LAB to serve as defensive components in the intestinal microbial ecosystem (Kullisaar et al., 2002). While use of recombinant LAB in functional foods would not be acceptable, their use as therapeutics may be foreseen where the potential beneﬁts greatly outweigh any potential risks. This is particularly true of IBD for which current treatments are unsatisfactory and the development of new and innovative therapies is urgently needed. The recent approval by Dutch authorities to use genetically engineered Lc. lactis secreting IL-10 as an experimental therapy for humans with IBD gives encouragement for the further development of therapies based on recombinant LAB (Steidler et al., 2003). V. LAB as Cell Factories for the Manufacturing of Pharmaceutical Proteins Several aspects are considered when choosing a host and a gene expression system for the production of proteins for pharmaceutical use (‘‘pharmaceutical proteins’’). These include cost of production, yield, purity, formation of biologically active molecules, and possible contamination by toxic substances either produced by the host cell or present in the growth medium. In addition, commercial production of a pharmaceutical protein and its production process (including applied tools and unit operations) requires approval from the regulatory authorities. These processes must, therefore, follow good manufacturing practices (GMP), ensuring that the production organism and associated manufacturing procedures and materials are well-characterized and documented. When considering LAB for commercial production of pharmaceutical proteins, most issues that need to be addressed are common to all biological production systems and can be divided into four categories relating to (1) the microbial host; (2) the expression vector; (3) propagation, fermentation and initial downstream processing; and (4) quality control of the product. The latter includes testing for the proper activity as well as stability and purity and will not be discussed here, because it is speciﬁc to each product. A. THE CHOICE OF MICROBIAL HOST E. coli was the ﬁrst microorganism to be utilized for the production of pharmaceutical proteins with recombinant DNA technology (Goeddel et al., 1979a,b; Villa-Komaroff et al., 1978), primarily because this bacterium had been extensively studied in laboratories throughout the world.

24

HANNIFFY et al.

E. coli demonstrates several disadvantages from a technological point of view, such as inefﬁcient protein secretion, formation of inclusion bodies, and production of endotoxin. However, because of precedent from its previous use in producing a long list of approved pharmaceutical proteins, E. coli continues to be the preferred bacterial production host for industrial applications. This tendency to revert to a proven and accepted production system seems to have hindered the adoption of LAB as a production host. Although LAB demonstrate several features that make them ideal for producing certain products, pharmaceutical proteins produced by these bacteria have yet to be approved and brought to market. However, it is expected that this bottleneck will be eliminated in the present decade, since several proteins produced in lactic acid bacteria are currently being tested in clinical trials. For regulatory approval of the recombinant product, a historical description of the host strain is required (i.e., description of how the strain was isolated, characterized, and subsequently treated and manipulated in the laboratory prior to use). This should include a risk evaluation describing known and potential application risks and documenting any record of infections and diseases caused by the same species as the selected host strain. Finally, the evaluation should also deal with host strain stability and potential for genetic exchange with other bacteria (e.g., natural competence for DNA uptake). As many LAB are already used in food production and preservation and more recently as probiotics, many of the aforementioned requirements for regulatory approval are already in place for selected strains. Genome sequences are now available for an increasing number of LAB (Klaenhammer et al., 2002), and various postgenomic studies involving DNA arrays (van de Guchte et al., 2002) and 2-D gel electrophoresis (Champomier-Verges et al., 2002; Guillot et al., 2003) have already been carried out with different LAB under different conditions. These approaches can be used to compare production strains with their corresponding source strain, thus contributing to more-robust risk assessment procedures capable of predicting an undesirable effect. Similarly, comparative genomics and other bioinformatic approaches can also be used to identify potential conjugation or mobilization genes, which can be subsequently deleted in order to minimize gene transfer and its associated risks. B. THE EXPRESSION VECTOR The vector for expression of a desired gene can be chromosomally integrated or plasmid borne. In addition to unnecessary and redundant DNA, it is essential that all genes dedicated to gene transfer are

USE OF RECOMBINANT LAB IN HUMAN HEALTH

25

eliminated. Similarly, if a plasmid is used, the origin of replication should ideally replicate in one or few species only. The entire DNA sequence of the plasmid or any integrated DNA must be determined to trace recombinant DNA during downstream processing as well as in the ﬁnal product. In addition, the segregational and structural stability of the chromosomal insert or expression plasmid in a production host must be analyzed to determine if it is stable. Usually a selection marker is required for maintenance of the plasmid carrying the desired gene. Previously the regulatory authorities have approved the use of antibiotic selection markers. However, future approvals are expected to include nonantibiotic selection markers only (see Section VI) because of the growing concerns regarding resistant bacteria whose emergence appears to correlate with the extensive use of antibiotics. The ideal promoter for gene expression should fulﬁl several requirements (Makrides, 1996). While in most cases the activity of the promoter should be as high as possible to produce the greatest amount of protein, high levels of constitutive expression can be lethal to the host cell or result in inhibition of growth, leading to loss of the expression vector or structural instability of the recombinant DNA. An inducible promoter is therefore preferable to coordinate protein production with cell growth in a way that maximizes the ratio of protein yield relative to biomass. Background transcription should be minimal during the growth phase to produce sufﬁcient biomass, at which point induction of the promoter should initiate a burst of gene expression. The last decade has seen the development of elegant and efﬁcient tools for genetic manipulation and gene expression in Lc. lactis, the most extensively studied member of the entire LAB group (reviewed in Section II). For example, the P170 expression system (Fig. 2) utilizes the regulated promoter P170, which is activated in Lc. lactis on transition from exponential to stationary phase (Israelsen et al., 1995b; Madsen et al., 1999b). Consequently, the growth phase is separated from the protein production phase that occurs only when biomass has reached a maximum. The P170 system also utilizes the lactococcal replicon (repB) and the lactococcal hom and thrB genes, thus permiting auxotrophic selection when using a hom-thrB-deﬁcient production strain (Glenting et al., 2002; Madsen et al., 1996). The system has been further adapted to include DNA encoding the SP310mut2* lactococcal secretion signal (Ravn et al., 2000, 2003). The secretion signal directs the preprotein to the translocation apparatus, where it is cleaved off thus releasing the mature protein into the growth medium. Pharmaceutical proteins secreted into the growth medium can be easily puriﬁed, thereby reducing downstream processing and potential cost of production.

26

HANNIFFY et al.

FIG. 2. Expression vector pAMJ1223. The P170 promoter, signal peptide SP310mut2*, the auxotrophic selection marker hom-thrB, and the Lc. lactis replicon repB are indicated. The multiple cloning site is located between the signal peptide and the transcription terminator.

DNA microarray and proteomic characterization of global regulatory responses to different environmental stimuli in Lc. lactis (Champomier-Verges et al., 2002; Guillot et al., 2003; van de Guchte et al., 2002) should help the development of new expression systems that are activated under industrial conditions. In addition, Lc. lactis strains are now available that lack extracellular proteases, resulting in reduced degradation of secreted proteins (Madsen, personal communication; Miyoshi et al., 2002). Combining appropriate expression systems with such mutant strains could signiﬁcantly improve production and secretion of pharmaceutical proteins, making Lc. lactis increasingly attractive as an industrial production host. C. PROPAGATION, FERMENTATION, AND INITIAL DOWNSTREAM PROCESSING The production strain must be propagated into growth medium (propagation medium) that contains no toxic or harmful substances and ensures the stability of the strain. Once the production strain has been characterized with respect to critical parameters (e.g., composition of heterologous DNA, genetic stability, productivity, and byproducts), minimally passaged cultures are used to generate master and working cell banks that are checked at regular intervals.

USE OF RECOMBINANT LAB IN HUMAN HEALTH

27

LAB produce high amounts of lactic acid with a concomitant pH decrease that inhibits growth of the bacteria. It is therefore important that the fermentation medium is properly buffered or pH controlled to achieve higher cell densities. Alternatively, strains may be developed that either produce low amounts and/or show a high tolerance to lactate. It may also be possible to use a growth medium that produces low amounts of lactate. In any biological production system, the growth medium must be optimized according to product yield and purity. For example, fully synthetic growth media optimized for the P170 controlled expression system has now been developed, comprising commercially available components that meet the standards of United States Pharmacopoeia (USP) or the European Pharmacopoeia (Ph. Eur.). Depending on the protein being produced, the pH of the medium in the fermenter can be adjusted between 4.5 and 6.5. Fermentation is carried out as a fed batch culture in which potassium hydroxide and glucose are added automatically. Following inoculation and determination of exponential growth, the fermentation process requires only limited surveillance, since the P170 system automatically induces gene expression according to the pH and growth phase. As shown in Fig. 3, the protein produced is stable when secreted into the supernatant and can therefore be isolated over an extended time range for subsequent downstream processing. Once the fermentation process is completed, cells are separated from culture supernatant containing

FIG. 3. Data from a fed batch fermentation of a recombinant Lc. lactis strain that secretes the Staphylococcus aureus nuclease. (A) Optical density during the course of fermentation. Samples for analysis of the produced nuclease were taken at the indicated time-points (1–9). In (B), 10 l crude culture supernatant from each time-point was analyzed by SDS-PAGE. The lane numbers correspond to the time-points shown in (A). The gel was stained with Coomassie brilliant blue. Molecular weights (in kilodaltons) are indicated at the right, and the triangle indicates the position of the secreted nuclease.

28

HANNIFFY et al.

the pharmaceutical protein. The supernatant can then be concentrated by dia-ﬁltration in readiness for downstream puriﬁcation and formulation. A risk evaluation of the P170 Expression System and host strains has already been carried out by the independent Danish Toxicology Centre, which is now being used as a guide by a number of commercial clients to meet with demands from regulatory authorities (Israelsen, personal communication). VI. Engineering LAB for Their Safe Use in Humans The use of genetically modiﬁed microbes (GMMs) in humans raises concerns about their survival and proliferation in the environment. Microorganisms have evolved highly efﬁcient systems for horizontal gene transfer such as transformation, conjugation, retromobilization, and transduction to improve their adaptation to changes in their ecological niche. Therefore the transfer of recombinant DNA such as antibiotic resistance markers or other genetic modiﬁcations from a well-characterized transgenic microorganism to uncharacterized indigenous species is perceived as a signiﬁcant risk that must be minimized (Gruzza et al., 1993; Netherwood et al., 1999; Ramos et al., 1995; von Wright and Bruce, 2003). In the context of recombinant live vaccines, for example, a possible scenario could be the transfer of a cloned virulence determinant from a live vaccine strain to a pathogen potentially reinforcing its ability to cause infection. Apart from the transgene of interest, the ideal GMM for use in humans should therefore contain the minimal amount of foreign DNA and must not include an antibiotic resistance marker. Furthermore, the possibilities of transgene horizontal transfer must be minimized, and GMM lethality should be achieved in an unconﬁned environment. A. FOOD-GRADE SYSTEMS IN LAB FOR PLASMID MAINTENANCE AND CHROMOSOMAL INSERTION A large number of safe and sustainable food-grade systems for genetic modiﬁcations of LAB have been developed (initially for Lc. lactis), including endogenous cloning vectors, selection markers for plasmid maintenance, inducible high-level expression systems, and chromosomal insertion strategies (for a review, see de Vos, 1999b; Renault, 2002). The transgene of interest can either be inserted into the chromosome or cloned into a multicopy plasmid, the latter being sometimes necessary to achieve very high levels of expression such as those required for intracellular expression of antigen in live recombinant

USE OF RECOMBINANT LAB IN HUMAN HEALTH

29

LAB-based vaccines (Grangette et al., 2001). Complete food-grade plasmid-based expression systems have been successfully developed in LAB that minimize the introduction of foreign DNA (self-cloned systems) (de Vos, 1999b). In these systems, all components ideally originate from the cloning host including replicons, the inducible or constitutive expression/secretion cassette, as well as a reliable selection system for plasmid introduction and maintenance. The selection system is of crucial importance to ensure the stability of the transgenic strain during the fermentation processes that precede administration. Nonantibiotic food-grade markers developed for LAB can be grouped in two categories based on the selection method. The ﬁrst category includes so-called ‘‘dominant markers’’ that do not rely on speciﬁc host genes (versatile) and can be readily compared with antibiotic resistance markers. For example, the catabolism of speciﬁc sugars (melibiose, sucrose, xylose, starch, and inulin) offers various possibilities for the development of dominant selection markers in Lc. lactis and various lactobacilli (Boucher et al., 2002; Fitzsimons et al., 1994; Hols et al., 1994; Leenhouts et al., 1998; Posno et al., 1991). A second group in this category are genes conferring resistance or immunity (e.g., immunity to bacteriocins [nisI, lafI]) and resistance to cycloserine (alr) and heavy metals (Cdr) (Allison and Klaenhammer, 1996; Bron et al., 2002; Froseth and McKay, 1991; Liu et al., 1996; Takala and Saris, 2002; von Wright et al., 1990). Although dominant markers are convenient, most cannot be used at the industrial scale for safety or economic reasons. Furthermore, their implementation sometimes requires the transfer of numerous genes that can strongly affect plasmid stability or could result in slower growing recombinant strains. The second category includes the complementation markers resulting from speciﬁc mutations in the host chromosome, thus permitting a speciﬁc plasmid-host combination. Two types of complementation markers can be distinguished in LAB based either on sugar utilization (lactose) or on a speciﬁc auxotrophy (pyrimidine, thymine, D-alanine) (Bron et al., 2002; Fu and Xu, 2000; Hashiba et al., 1992; MacCormick et al., 1995; Platteeuw et al., 1996; Ross et al., 1990; Sorensen et al., 2000; Takala et al., 2003). A complementation system for lactose utilization based on the lacF gene has been successfully applied to Lc. lactis on an industrial scale (Kleerebezem, personal communication; Platteeuw et al., 1996). This system, however, uses a growth medium that utilizes lactose as a carbon source, which is not always desirable for implementation in other LAB. Complementation markers based on pyrimidine or thymine auxotrophy have also been described

30

HANNIFFY et al.

that exploit the amber suppressor tRNA gene (supD) and thymidilate synthase gene (thyA), respectively (Fu and Xu, 2000; Sorensen et al., 2000), but again, the requirement for dedicated growth media without these speciﬁc compounds represents a major drawback. More recently the use of the alanine racemase gene (alr) as a complementation marker was reported as functional in both Lc. lactis and Lb. plantarum (Bron et al., 2002). Alanine racemase converts L-alanine (L-Ala) into D-alanine (D-Ala), which is essential in cell wall biosynthesis as it is involved in the cross-linking of peptidoglycan. Alanine racemasedeﬁcient strains of both LAB are strictly auxotrophic for D-Ala and undergo cell lysis when starved of D-Ala (Fig. 4) (Bron et al., 2002; Hols et al., 1997a, 1999). Because D-Ala is mainly present in the cell walls of bacteria and concentrations of this compound are extremely low in most growth media, the alr gene offers greater potential as a complementation marker. Signiﬁcantly, the alr gene can also be used as a dominant marker, since high alanine racemase expression confers cycloserine resistance (Bron et al., 2002). In addition to D-Ala, the peptidoglycan contains other unique compounds (D-glutamate, D-aspartate, meso-diaminopimelate) that are essential and can also be used to supplement growth medium. In the future these compounds and their genes could potentially represent a range of complementation markers that could be used for the development of new food-grade plasmid maintenance or chromosomal integration systems. Although complete food-grade systems based on plasmids have been successfully developed, they often demonstrate intrinsic structural instability and display a high level of potential horizontal gene transfer through mechanisms such as plasmid co-integration and conjugation. To minimize these problems, several food-grade chromosome delivery systems have now been developed for LAB based on homologous recombination (recA-dependent), site-speciﬁc recombination (phage-derived systems), retrotransposition (group II intron-derived systems), and transposition (insertion sequence-based systems) (de Vos, 1999b; Frazier et al., 2003; Hols et al., 1994; Maguin et al., 1996; Martin et al., 2000; Romero and Klaenhammer, 1991). With the exception of homologous recombination, all of these delivery systems are based on genetic elements that contribute to genome plasticity and/or horizontal transfer and must be avoided for safe use. Chromosomal integration by homologous recombination can be achieved by a single crossover event (Campbell integration with one homologous region) with a nonreplicating plasmid (suicide vector). Food-grade systems based on this approach have been developed for Lc. lactis by using endogenous plasmids devoid of their repA replication protein (pORIþ plasmids)

USE OF RECOMBINANT LAB IN HUMAN HEALTH

31

FIG. 4. (A) The effects of D-Ala starvation on viability of wild-type NCIMB8826 (open squares) and the alr deletion mutant MD007 (solid triangles). The zero time-point is set to the beginning of starvation after cell re-suspension in MRS broth not supplemented with D-Ala. The log number of colony forming units/ml CFU ml1 of Lb. plantarum were enumerated over time by dilution and plating on solid medium supplemented with D-Ala. Scanning electron micrographs of the alr mutant MD007 are shown at time zero (B) and after 20 hours (C) in MRS broth lacking D-Ala. The arrows in (C) indicate V-shaped cells that have broken in the middle, resulting in the release of cytoplasmic material.

and food-grade markers of the complementation type (lacF) or the dominant type (scrAB for sucrose utilization) (de Vos, 1999b; Leenhouts et al., 1996, 1998). In some cases, continuous selection for the food-grade marker can result in a chromosomal plasmid ampliﬁcation of up to 20 copies, thus improving gene dosage (Leenhouts et al., 1998). However, a possible drawback of this system is that replication of the chromosomally integrated plasmid could be reactivated subsequent to horizontal transfer of an incoming plasmid bringing repA in trans. The best strategy for reduced risk of horizontal transfer employs two crossover events whereby the transgene is safely and stably integrated into the chromosome as a single copy and in the absence of additional DNA. This double crossover strategy has recently been used to stably integrate the human interleukin 10 (hIL10) gene into the chromosomal thyA gene of Lc. lactis. Furthermore, the hIL10 gene was translationally fused to the expression signals of the thyA gene to minimize the introduction of foreign DNA. The level of hIL10 expression was reasonably high in comparison to its expression from a

32

HANNIFFY et al.

multicopy plasmid (Steidler et al., 2000, 2003). It is now likely that such a strategy will be used increasingly for the genetic engineering of LAB for other medical applications. B. BIOLOGICAL CONTAINMENT SYSTEMS Biological containment systems can be subdivided into two groups: active systems and passive systems. Active containment is based on the conditional genetic control of either a killing gene (activation) or an essential gene (repression). Various killing genes have been identiﬁed in LAB such as those encoded by phage lytic cassettes (lysin or lysin/ holin combination) and DNA restriction systems (de Ruyter et al., 1997; Djordjevic and Klaenhammer, 1996, Djordjevic et al., 1997; Sanders et al., 1997). While a large number of essential genes could potentially be exploited, peptidoglycan biosynthetic genes are particularly attractive, because their inactivation usually results in cell lysis (Curtiss III et al., 1977). Proof of the concept for lysis induction with peptidoglycan hydrolases (lysins) and biosynthetic genes (alr; D-ala-D-ala ligase, ddl) tightly regulated by the nisin-inducible system has already been achieved in the laboratory by using Lc. lactis and Lb. plantarum (Bron et al., 2002; LABDEL project; Prozzi and Fontaine, unpublished data). To be implemented in humans, these active containment systems would require expression signals that are tightly controlled by environmental parameters. These expression signals could either be activated/repressed in the host or subsequent to release of the recombinant strain into the environment. A range of expression cassettes have been isolated from Lb. plantarum that are speciﬁcally activated in the GI tract of mice (Bron et al., 2004a; LABDEL project) (see Section VII). These expression systems are now being exploited in Lb. plantarum to speciﬁcally induce lysis and consequent release of a biomolecule of interest (therapeutic molecules) into the GI tract via the control of lytic cassettes. A drawback to using active containment systems is that mutations can occur that can inactivate a killing gene/compound or result in constitutive expression of the essential gene (Hols, P., unpublished data; Molina et al., 1998; Szafranski et al., 1997). However, it may be possible to minimize these problems by combining more than one system in a deﬁned recombinant strain (Ronchel and Ramos, 2001). Passive containment systems, by contrast, are robust and very simple and based mainly on complementation of an auxotrophy by supplementing an essential metabolite. This metabolite is ideally not available or occurs in extremely low amounts in the environment. Furthermore, if the compound must be added to growth medium for

USE OF RECOMBINANT LAB IN HUMAN HEALTH

33

inoculum preparation, it is preferable that it occurs at a low cost to facilitate future industrial applications. Several mutant strains used for plasmid maintenance via complementation markers also offer potential as passive containment systems. These include two auxotrophies that have a high potential for human application since they are bactericidal. Alanine racemase mutants that demonstrate a requirement for D-Ala can be obtained in a large number of bacteria including LAB (Bron et al., 2002; Hols et al., 1997a). D-Ala starvation results in cell death via a lysis process, and experiments are currently underway for containment of S. mutans GMMs designed for anti-caries therapy (Hillman, 2002). The second bactericidal auxotrophy is based on a thyA mutant of Lc. lactis that has recently been evaluated for its efﬁcacy in a pig model (Steidler et al., 2003). Thymine starvation results in activation of the SOS repair system and DNA fragmentation, thus constituting an intrinsic suicide system (Curtiss et al., 1977). Exploiting the thyA gene as a passive containment system by replacing it with a transgene of interest offers additional advantages in terms of biosafety (Steidler et al., 2003). While any reversion of the thyA mutation through homologous recombination with a Lc. lactis thyA gene that might be acquired by horizontal transfer would restore a wild-type phenotype, it should also result in loss of the transgene. Furthermore, thyA mutants of Lc. lactis are severely impaired in phage replication, further reducing the risk of horizontal transfer by phage transduction (Pedersen et al., 2002). This elegant system has recently been applied to the delivery of hIL10 by live Lc. lactis bacteria and has now been approved for use in humans (Steidler et al., 2003).

VII. Opportunities and Potential Applications of Future Research A. INSIGHTS FROM GENOME SEQUENCING AND COMPARATIVE GENOMICS The relatively small genome size of bacteria, combined with the availability of high-throughput sequencing facilities, has stimulated the determination of many bacterial genomes. Over the past decade, the sequences of more than 90 bacterial genomes have become available in the public domain. These efforts have focused primarily on pathogenic bacteria and have included the completion of several genome sequences of food-borne pathogens (Schoolnik, 2002; Wells and Bennik, 2003). However, in more recent times, the genomes of food-grade bacteria have received considerable attention and genome sequences (including partial sequences) are now available for more than 20 LAB as well as a number of related species (Klaenhammer et al., 2002).

34

HANNIFFY et al.

FIG. 5. Phylogenetic tree constructed by using 16 S rDNA sequences of LAB and related bacteria that have been completely or partially analyzed at the genomic level. It should be noted that for a number of species the genome sequence has (partially) been determined for more than one strain of this species. The estimated genome sizes are indicated between brackets. Complete genomic sequences (and thus their exact sizes) are available for the paradigm Gram-positive bacterium Bacillus subtilis (Kunst et al., 1997) and the LAB, Lactococcus lactis subspecies lactis IL1403 (Bolotin et al., 2001) and Lactobacillus plantarum WCFS1 (Kleerebezem et al., 2003).

The ﬁrst complete LAB genome sequence to be published was that of Lc. lactis subspecies lactis strain IL1403 (Bolotin et al., 2001). More recently, a high-ﬁdelity genome sequence has also been published for Lb. plantarum strain WCFS1 (Kleerebezem et al., 2003). In addition, more LAB genomes are nearing completion (Fig. 5; for a review see Klaenhammer et al., 2002) including draft genome information for a number of LAB that was made available in the public domain in 2002 by the Joint Genome Institute (ftp://ftp.jgi-psf.org/pub/JGI_data/ Microbial/) in collaboration with the lactic acid bacteria genomics consortium. Comparative genomics will be used to determine the evolutionary relationships between different species of LAB and allow analysis of genomic diversity among different strains and closely related species. Studies on pathogenic bacteria have already shown that comparative genome analysis and microarray-based studies of genome composition can indeed reveal genetic factors linked to speciﬁc functions or

USE OF RECOMBINANT LAB IN HUMAN HEALTH

35

plasticity regions associated with lifestyle properties of speciﬁc strains (Israel et al., 2001; Pearson et al., 2003; Porwollik et al., 2002; Salama et al., 2000). In the case of LAB, it is already well established that some species are genetically diverse and that strain adaptation is rapid. Moreover, horizontal gene transfer appears to be the rule rather than the exception and is especially important for genes or functions related to mobile genetic elements such as plasmids, transposons, and bacteriophages. Recently, genomic variation within a single LAB species was exempliﬁed by a detailed genome diversity study of Lb. plantarum isolates using DNA microarrays (Molenaar et al., 2003). In addition to high levels of genome diversity, Molenaar et al. (2003) also identiﬁed speciﬁc genomic regions of relatively high plasticity that may represent lifestyle adaptation islands. These ﬁndings show that genomic diversity is a key determinant of bacterial functionality and indicate that speciﬁc approaches are required to fully exploit the collective genomic potential of LAB. B. THE BEHAVIOR OF LAB

IN THE

HOST

The human GI tract is colonized by a vast, complex, and dynamic population of commensal microorganisms including food-associated bacteria that contribute to nutrient processing, affect the host’s immune function, and stimulate a variety of other host activities. Molecular approaches based on 16 S rDNA sequencing and proﬁling of dominant GI tract microbiota have revealed that the majority of commensals belong to unknown, ‘‘novel’’ bacterial species that so far have not been studied under laboratory conditions (Zoetendal et al., 1998). In addition, these studies revealed that while the microbial composition was relatively stable in individual adults, it varies signiﬁcantly between different individuals (Zoetendal et al., 1998, 2002). Intriguingly, the composition of dominant GI tract microbiota was also shown to be affected by speciﬁc host-microbe interactions that appear to be related to the host genotype (Zoetendal et al., 2001). Similar molecular approaches as well as classical cultivation experiments have also shown that several food-grade bacteria are encountered as natural inhabitants of the human GI tract. In particular, certain Lactobacillus species are frequently found among the natural intestinal microbiota and would appear to be relatively numerous in the upper regions of the small intestine (Ahrne et al., 1998; Heilig et al., 2002; Walter et al., 2000, 2001). The increasing availability of genome sequences for several intestinal species of LAB opens new avenues to study their functionality in the gut, including interactions with host cells. For example,

36

HANNIFFY et al.

when ingested, Lb. plantarum WCFS1 was shown to effectively survive passage of the human stomach and to reach the ileum in high numbers as compared with other LAB strains (Vesa et al., 2000). In addition, the bacterium could also be detected in the colon. These characteristics and the availability of its genome sequence (Kleerebezem et al., 2003) render this strain a suitable candidate for the study of bacterial behavior in the GI tract. To date, DNA microarray and proteomic studies carried out in LAB have been conﬁned to in vitro experiments investigating global regulatory responses to different environmental stimuli (ChampomierVerges et al., 2002; Guillot et al., 2003; van de Gucht et al., 2002). Similarly, studies investigating host-LAB interaction have focused primarily on physiological aspects such as intrinsic levels of acid and bile tolerance (Chou and Weimer, 1999; Hyronimus et al., 2000) and the development of complex media to selectively enrich for LAB species that are tolerant to digestive stress (Shah, 2000). For example, a genome-wide genetic screen carried out in Lb. plantarum identiﬁed 31 open reading frames (ORFs) whose expression would appear to be induced by bile acids, including genes encoding efﬂux pumps as well as several membrane and cell wall–associated functions (Bron et al., 2004b,c). While such in vitro experiments might unravel responses by speciﬁc microorganisms to certain GI-tract conditions, they will not sufﬁce in portraying bacterial behavior in the GI tract. The full response repertoire will be triggered only in vivo, where all physicochemical conditions are combined with speciﬁc host-microbe and microbe-microbe interactions. This notion has led to development of more sophisticated in vivo approaches aimed at identifying bacterial genes that are important during residence in the GI tract. Three main strategies have been developed for the identiﬁcation of genes that are highly expressed in vivo as compared to laboratory conditions, that is, (recombination-based) in vivo expression technology ((R-)IVET), signature tagged mutagenesis (STM), and selective capture of transcribed sequences (SCOTS). The basic characteristics and relative (dis)advantages of each of these approaches have recently been reviewed (Mahan et al., 2000), but because of their recent application in LAB, (R-)IVET approaches will be explored in somewhat more detail here. To date, four variations of IVET utilizing different reporter genes have evolved that allow the selection of genomic sequences that harbor promoters that are speciﬁcally induced in situ, for example, during GI-tract passage (for a review see Angelichio and Camilli, 2002; Mahan et al., 2000). The (R-)IVET approach relies on the generation of transcriptional fusions of random genomic fragments to a promoterless

USE OF RECOMBINANT LAB IN HUMAN HEALTH

37

reporter gene encoding an enzymatic activity. In the decade following the ﬁrst report describing IVET technology, this technique has solely been used for the identiﬁcation of genes that were up-regulated during the infection processes of a number of pathogenic bacteria (Angelichio and Camilli, 2002; Mahan et al., 2000). Recently, however, two reports have described for the ﬁrst time, the application of (R-)IVET technology to food-grade LAB. An IVET strategy based on in vivo selection of an antibiotic-resistant phenotype was used to identify three in vivo induced (ivi) genes in Lb. reuteri that were expressed at induced levels during colonization of the GI tract of reconstituted Lactobacillus-free mice (Walter et al., 2003). Interestingly, one of these genes encoding a peptide methionine sulfoxide reductase has previously been identiﬁed by using IVET in the non-food-associated bacterium S. gordonii during experimental endocarditis (Kili et al., 1999). Although not noticed by the authors, this ﬁnding gives a ﬁrst glimpse into potential overlap that may exist between genetic responses triggered by both pathogenic and non-pathogenic bacteria subsequent to host interaction. A disadvantage of this approach, however, is that it required antibiotic selection pressure during host transit, which was achieved by administering antibiotic to the mice during the course of the experiment. This treatment would have been detrimental to native intestinal microbiota already present in these mice, and it is highly likely that the in vivo conditions encountered by Lb. reuteri would have been signiﬁcantly changed. More recently, Bron et al. (2004a) have reported the successful application of a second (R-)IVET approach for the identiﬁcation of ivi genes in Lb. plantarum. In contrast to other IVET systems based on selection of an antibiotic marker, this approach employs the irreversible enzymatic activity of resolvases as a reporter and screening tool. In this system, an antibiotic resistance marker ﬂanked by two resolvase-recognition sites (Fig. 6: loxP-ery-loxP cassette) is integrated into the chromosome of the bacterium of interest. In addition, a promoterless copy of the corresponding resolvase-encoding gene (Fig. 6: cre) is introduced on a plasmid and used as a reporter to trap transcriptional activation (promoter activity) by monitoring changes in the antibiotic resistance phenotype. Since promoter activation leads to irreversible excision of the antibiotic resistance marker, this strategy does not rely on selective pressure during animal experiments. This form of (R-)IVET can therefore be used to monitor induction of bacterial gene expression in a live host possessing an intact unmodiﬁed GI tract. Moreover, this strategy is the only IVET approach that functions as a genetic screen as it discards all promoters displaying activity under laboratory

38

HANNIFFY et al.

FIG. 6. Schematic representation describing the (R)-IVET screen performed with Lactobacillus plantarum (Bron et al., 2004a). A loxP-ery-loxP cassette is introduced in the Lactobacillus plantarum chromosome. A library of random Lactobacillus plantarum chromosomal fragments is constructed in pIVET-cre, upstream of a promoterless copy of the cre resolvase gene. Expression of cre leads to excision of the loxP-ery-loxP cassette and erythromycin sensitivity. An erythromycin-resistant sub-library is selected under laboratory conditions and administered to conventional mice. The complete sub-library is recovered from fecal samples by selection for the pIVET-cre encoded chloramphenicol resistance and replicated to select erythromycin-sensitive clones (CmR, EmR). These clones contain pIVET-cre derivatives that harbor a GI tract activated promoter in the Lactobacillus plantarum chromosomal fragment cloned upstream of cre, since loxP-eryloxP excision has occurred during transit through the host intestine. Sequencing of the pIVET-cre inserts is used to identify the corresponding ivi genes of Lactobacillus plantarum (Bron et al., 2004a).

conditions, since the (R-)IVET library is prepared under antibiotic selection pressure prior to administration to the host (Fig. 6). Application of this genetic screen in Lb. plantarum resulted in the identiﬁcation of 72 genes that were induced during passage of the GI tract in conventional mice (Bron et al., 2004a). A diverse range of functions was predicted for these 72 ivi-genes, including nutrient acquisition, intermediate and/or co-factor biosynthesis, stress response, cell surface proteins, and a number of (conserved) hypothetical proteins. Remarkably, one of the hypothetical proteins identiﬁed in this study displays signiﬁcant homology (32% identity) to the conserved hypothetical protein that was identiﬁed by using IVET in Lb. reuteri (Bron et al., 2004a). Moreover, a large number of the functions and pathways identiﬁed

USE OF RECOMBINANT LAB IN HUMAN HEALTH

39

by (R-)IVET in Lb. plantarum have also been described in pathogenic bacteria, where they have also been shown to play an important role during infection. The striking parallels that exist between ivi responses exhibited by pathogens and nonpathogens in vivo may well be explained by the importance of these genes for survival rather than pathogenesis during host residence (Bron et al., 2004a). These studies represent an important step toward elucidating the behavior of food-grade LAB in the complex environment they encounter after ingestion by the consumer. Approaches such as IVET will provide the genetic tools necessary, including the required promoters, that will allow the development of dedicated LAB-based delivery vehicles that will express only desired pharmaceutical proteins in situ. Moreover, these approaches should provide a more geographical and precise insight into the exact locations of ivi-gene activation in the GI tract that may allow the construction of a new generation of site-specific delivery vehicles. Eventually, ivi-promoters could be combined with certain genes (e.g., bacteriophage-derived lytic cassettes) to develop LAB delivery systems that exhibit controlled release of their cellular content (including desirable molecules) at a speciﬁc location in the GI tract. IVET and other genomic-based approaches will also provide a better understanding of LAB behavior in the GI tract as well as in other environmental niches. They should facilitate the genetic engineering of mutant strains that can be compared with their wild-type counterparts in appropriate in vitro and in vivo models. Ultimately, it is hoped that these approaches will help deﬁne potential probiotic functions and provide a molecular basis to support and explain health beneﬁts that are associated with LAB and related species. C. THE HOST RESPONSE TO LAB The ability of LAB to alter the function of the systemic and immune responses has been the subject of intense research over the last few years. Some LAB have been shown to translocate across the mucosal epithelium of the GI tract, allowing them to come into direct contact with the underlying immune cells and inﬂuence immune responses. There is also evidence that some LAB can directly stimulate the immune system at the gut mucosal surface via localized GI tract lymphoid cell loci, increasing lymphocyte populations and cell surface expression in the gut-associated lymphoid tissue (GALT) environment, and facilitate increased immunoglobulin output into the intestinal lumen. It is also becoming increasingly clear that interactions of different LAB with host cells in the mucosa can have different effects on immune

40

HANNIFFY et al.

function. Administration of both viable and/or heat-killed preparations of the probiotic Lb. acidophilus strain UFV-H2b20 has been shown to enhance phagocytic activity in germ-free mice and to improve clearance capacity against circulating E. coli as compared with controls (Neumann et al., 1998). Experiments carried out in mice have shown that oral administration of different LAB increased IgA-producing cells associated with the lamina propria (Vitini et al., 2000). However, this increase did not always correlate with an increase in CD4þ T lymphocytes, suggesting that these bacteria only induced clonal expansion of cells in the lamina propria that were already triggered to produce IgA. In addition, while some LAB did increase macrophages, neutrophils and eosinophils indicative of an inﬂammatory response, CD8þ T cell populations were diminished or not affected. Other studies carried out in animals and humans have also shown that the administration of various Lactobacilli and Biﬁdobacteria can signiﬁcantly increase total IgA as well as antigen or pathogen-speciﬁc IgA, thereby contributing to mucosal resistance to gastrointestinal pathogens (Cukrowska et al., 2002; Fang et al., 2000; Fukushima et al., 1998; Herias et al., 1999; Tejada-Simon et al., 1999). In one of these studies, co-administration of Lb. plantarum and an E. coli strain to germ-free rats resulted in signiﬁcantly increased densities of CD25þ cells in the lamina propria and decreased proliferative spleen cell responses to E. coli (Herias et al., 1999). CD 25þ cells have been shown to be involved in tolerance and down-regulation of immune responses to self and nonself antigens. While the production of IgA may be important in preventing infections by pathogens, it may also be considered anti-inﬂammatory because it involves mechanisms controlled by Th2 effector cytokines that are more associated with tolerance. These studies have been supported by work carried out by Kirjavainen et al. (1999a) that showed that administration of selected LAB strains could reduce T cell reactivity in mice in a dose-dependent fashion (Kirjavainen et al., 1999a,b). Similarily, in vitro experiments have shown that LAB or their components can inhibit mitogen- and antigen-induced T lymphocyte proliferation and down-regulate both Th1 and Th2 effector cytokines (Chen et al., 2002; Pessi et al., 1999; von der Weid et al., 2001). In addition, LAB have been reported to induce IL-10 production in gut epithelium cells (Haller et al., 2000) and in peripheral blood mononuclear cells isolated from both animals and humans (Miettinen et al., 1996; Vinderola et al., 2004). Interestingly, von der Weid et al. (2001) also identiﬁed a Tr1-like population of CD4þ T cells with low proliferative capacity that produced substantial levels of the regulatory cytokines IL-10 and TGF-. These studies would indicate that

USE OF RECOMBINANT LAB IN HUMAN HEALTH

41

mechanisms controlled by LAB may play an important role in maintaining immune homeostasis and, in particular, maintaining tolerance to the indigenous microbial population. This is supported by experiments in animals that show that select strains of LAB induce IL-10 and suppress Th1-related immunopathologies such as IBD. Conversely, LAB have also been shown to increase lymphocyte proliferation as well as the production of inﬂammatory cytokines. Experiments carried out by Aattour et al. (2002) in pathogen-free rats showed that oral ingestion of yogurt containing S. thermophilus and Lb. bulgaricus resulted in increased in vitro proliferation of lymphocytes isolated from peripheral blood, spleen, and Peyer’s patches that, for the two latter lymphoid compartments, were accompanied by increased interferon- production. Similarly, Gill et al. (2000) showed that feeding various LAB strains to mice signiﬁcantly enhanced mitogen-induced proliferation of splenocytes and production of IFN-, increased phagocytic activity of peripheral blood leukocytes and peritoneal macrophages, and enhanced serum antibody responses to antigen administered by the oral and systemic routes. Moreover, numerous studies have also shown that certain LAB or their components are potent inducers of Th1 cytokines (Cleveland et al., 1996; Kato et al., 1999; Maassen et al., 2000) and can decrease Th2 responses (Marshall et al., 1995; Murosaki et al., 1998; Repa et al., 2003; Shida et al., 2002). Similarly, studies carried out in animals and humans have demonstrated that ingestion of dietary LAB can prevent atopy and reduce the development of allergy (Bjorksten et al., 1999; Kalliomaki et al., 2001; Repa et al., 2003; Sepp et al., 1997; Simhon et al., 1982). These studies would indicate that LAB can enhance natural and acquired immunity and may be beneﬁcial in optimizing or enhancing immune responses in healthy and immunocompromised individuals (Gill et al., 2000). In addition, counter-regulatory properties based on their Th1-inducing capacities could provide a useful strategy against Type I allergy (Repa et al., 2003) (see also Section III). It has been demonstrated that different strains of LAB have substantially different capacities to induce IL-12 and TNF- production in dendritic cells (DCs) (Christensen et al., 2002). In addition to exerting less-pronounced effects on IL-6 and IL-10 production, all strains upregulated expression of surface MHC class II and B7-2 indicative of DC activation and maturation. In a separate study, signiﬁcant functional and phenotypic dichotomy was also observed in DCs exposed either to Lb. rhamnosus or Klebsiella pneumoniae (Braat et al., 2003). Interestingly, compared to K. pneumoniae, Lb. rhamnosus induced lower TNF-, IL-6, and IL-8 production in immature DCs as well as lower IL12 and IL-18 production in mature DCs that were also associated

42

HANNIFFY et al.

with the development of T cells without a typical Th phenotype. Signiﬁcantly, Christensen et al. (2002) also showed that at least one species of Lactobacillus was capable of inhibiting the activities of other Lactobacillus species with respect to DC activation and production of cytokines. These studies demonstrated that gut microﬂora, and in particular LAB, can differentially modulate immune responses by stimulating DC activation and function, which in turn may determine whether the DC population may favor a Th1 or Th2 response as well as tolerance. It is now clear that different LAB possess intrinsically different immunoadjuvant capacities that can play a signiﬁcant role in modulating immune responses. Germ-free animal models will help elucidate the interactions that occur between lactic acid bacteria and the host. This includes their ability to induce innate immunity that has not yet been appropriately compared in germ-free and conventional animals. Like other commensals, lactic acid bacteria will express various commensal associated molecular patterns (CAMPs) that are able to recognize and activate mucosal cells (including epithelial cells) through pattern recognition receptors (Funda et al., 2001; Janeway and Medzhitov, 2002). Recombinant mutants could be compared with wild-type strains to identify speciﬁc components, lipoteichoic acid and peptidoglycan for example, which may be responsible for distinct immune responses. The availability of genomic data will accelerate these efforts by facilitating the identiﬁcation of other key bioactive components such as those displayed on the bacterial surface. Combining post-genomic approaches such as microarrays and proteomics with GF animal models and selective colonization strategies could improve understanding of host-commensal interactions and bring new insights into the mechanisms of mucosal immunity. These efforts can also be combined with ex vivo approaches such as laser microdissection techniques that will allow us to examine individual cellular components as well as general cell populations from different anatomical regions of the immune system. Similarly, these tools can be applied to animal models of human diseases developed under deﬁned gnotobiotic conditions and may help elucidate the etiology of frequent disorders such as those associated with several infectious, inﬂammatory, autoimmune and neoplastic diseases. These approaches may help to explain why lactic acid bacteria and other commensals do not trigger pathological inﬂammatory responses in mucosal tissues in normal hosts and may answer intriguing questions regarding their associated beneﬁts.

USE OF RECOMBINANT LAB IN HUMAN HEALTH

43

VIII. Concluding Remarks The last decade has seen increasing interest in the use of recombinant LAB as mucosal delivery vehicles for vaccines and other therapeutic molecules. It is now clear that sufﬁcient advances in the genetics of LAB have made it possible to construct safe LAB-based recombinant vaccines that are capable of eliciting protection against lethal challenge with toxin or a human pathogen in a relevant disease model. Although multiple doses of LAB vaccines are currently needed to afford immune protection by the intranasal and oral routes of administration, this should not preclude their suitability for use in humans. Multiple doses of the injected DTP and oral polio vaccines (OPV) are already given to infants in the ﬁrst several months of life and have the potential to be combined with LAB vaccines. Moreover, this technology is relatively inexpensive and could be applied successfully to mass-vaccination programs that are particularly relevant in developing countries. There are also opportunities to enhance the efﬁcacy of LAB vaccines through increased antigen expression or through the combined delivery of multiple immunogens and speciﬁc adjuvants. Further insights may be gained through direct comparisons of LAB strains with different persistence and survival characteristics or immunostimulatory properties with different immunization routes or schedules against a selected target disease. Genetic engineering clearly has the potential to further optimize the survival characteristics of selected LAB, deﬁne optimal placement and dosage regimes in different clinical settings, and enhance their ability to deliver a pharmaceutical protein. Ultimately, we hope that LAB-based delivery technology will add a new dimension to vaccine development with respect to safety, potential for use in continual vaccination stratagems that encompass both control and eradication of a disease, and increased ﬂexibility and potential use against a wide range of diseases. Opportunities for developing more effective LAB vaccines may also arise from ongoing research on the fate of LAB in the human host and the interaction of LAB with the cells and tissues of the immune system. Genetic engineering may help elucidate the mechanisms by which probiotic strains of LAB exert positive effects on human health and facilitate the identiﬁcation of the key bioactive components responsible for conferring probiotic traits to these bacteria. These efforts will be further supported by ongoing developments in genomic and postgenomic-based approaches and use of appropriate animal models. As well as providing intriguing answers to questions regarding their therapeutic role in conditions such as allergy and IBD, these studies may help clarify the effect of nutrition and genetic background on the

44

HANNIFFY et al.

development of these diseases to establish the true role of these bacteria in human health. It is foreseen that these approaches will enable the selection of appropriate species that might prove most beneﬁcial under certain conditions and facilitate the identiﬁcation of new probiotics. In addition, it should be possible to determine optimal probiotic compositions, ensuring that they demonstrate synergistic rather than competing activities. Most importantly, these approaches will help establish proven criteria to substantiate the sometimes ‘‘vague’’ health claims of currently available probiotic preparations. In doing so, these criteria will lead to the possibility of creating custom-designed probiotics that are scientiﬁcally proven, increase their range of potential applications, and accelerate their development into health products. Currently, many probiotic strains of LAB are marketed and regulated as conventional foods. The regulatory environment surrounding these products is diverse, but national differences are being harmonized by ongoing European legislation. For example, the 1997 novel foods regulation of the EU parliament governs new food organisms and makes recommendations for assessment of foods containing genetically modiﬁed organisms (GMOs). In principle, this framework governs new recombinant strains of probiotic microorganisms and requires information on toxicology, genetic stability, potential for genetic transfer in the host, and the effect of the GMO in humans. Speciﬁc health claims for the product would also need to be substantiated by studies in humans resembling what is required by regulatory frameworks governing human pharmaceutical products. The future of recombinant LAB as novel therapies for humans can be foreseen where the potential beneﬁts are signiﬁcant, particularly where the absence of satisfactory treatments for a particular disease necessitates the development of new and innovative therapies. However, there is an urgent need for clariﬁcation to differentiate food supplements from medicines for product applications based on probiotic organisms. As well as ensuring the expansion of both these applications, appropriate legislation would allay fears within the food supplement industry and accelerate the development of health products based on recombinant LAB. It is now clear that biotechnological applications of LAB over the past decade has meant that many food-grade systems for genetic modiﬁcation are in fact already available, including methods for chromosomal integration and controlled high-level gene expression. Moreover, the recent approval of a recombinant strain of Lc. lactis expressing IL-10 for use in trials with human IBD patients provides further encouragement. We are therefore optimistic that as well as the continued development of prototype health products based on recombinant LAB, the next decade will see more of these reach the marketplace as ﬁnished products.

USE OF RECOMBINANT LAB IN HUMAN HEALTH

45

ACKNOWLEDGMENTS The Partners of the LABDEL consortium are grateful for financial support from The European Framework 5 Programme (EC contract QLK3-CT-QLK3-CT-2000-00340). The authors also wish to thank Andy Carter, Wendy Glennison, Sally M. Hoffer, Maria Marco, Esther van Mullekom, Douwe Molenaar, Willem M. de Vos, Anne Mette Wolff, Astrid Vrang, Renata Stepankova, Deborah Prozzi, Marie Deghorain, Laetitia Fontaine, Vassilia Therodorou, Heimo Breiteneder, and Michael Hisbergues for their contribution to the work described. We are also grateful to Therese Hall for help with coordination of the LABDEL project. Pascal Hols is scientific collaborator at FNRS.

REFERENCES Aattour, N., Bouras, M., Tome, D., Marcos, A., and Lemonnier, D. (2002). Oral ingestion of lactic-acid bacteria by rats increases lymphocyte proliferation and interferon-gamma production. Br. J. Nutr. 87, 367–373. Adams, M. R., and Marteau, P. (1995). On the safety of lactic acid bacteria from food. Int. J. Food Microbiol. 27, 263–264. Ahrne, S., Nobaek, S., Jeppsson, B., Adlerberth, I., Wold, A. E., and Molin, G. (1998). The normal Lactobacillus ﬂora of healthy human rectal and oral mucosa. J. Appl. Microbiol. 85, 88–94. Aiba, Y., Suzuki, N., Kabir, A. M., Takagi, A., and Koga, Y. (1998). Lactic acid-mediated suppression of Helicobacter pylori by the oral administration of Lactobacillus salivarius as a probiotic in a gnotobiotic murine model. Am. J. Gastroenterol. 93, 2097–2101. Allison, G. E., and Klaenhammer, T. R. (1996). Functional analysis of the gene encoding immunity to lactacin F, lafI, and its use as a Lactobacillus-speciﬁc, food-grade genetic marker. Appl. Environ. Microbiol. 62, 4450–4460. Alvarez, S., Herrero, C., Bru, E., and Perdigon, G. (2001). Effect of Lactobacillus casei and yogurt administration on prevention of Pseudomonas aeruginosa infection in young mice. J. Food Prot. 64, 1768–1774. Angelichio, M. J., and Camilli, A. (2002). In vivo expression technology. Infect. Immun. 70, 6518–6523. Araki, Y., Fujiyama, Y., Andoh, A., Koyama, S., Kanauchi, O., and Bamba, T. (2000). The dietary combination of germinated barley foodstuff plus Clostridium butyricum suppresses the dextran sulfate sodium-induced experimental colitis in rats. Scand. J. Gastroenterol. 35, 1060–1067. Babbs, C. F. (1992). Oxygen radicals in ulcerative colitis. Free. Radic. Biol. Med. 13, 169–181. Bauer, L., Bohle, B., Jahn-Schmid, B., Wiedermann, U., Daser, A., Renz, H., Kraft, D., and Ebner, C. (1997). Modulation of the allergic immune response in BALB/c mice by subcutaneous injection of high doses of the dominant T cell epitope from the major birch pollen allergen Bet v 1. Clin. Exp. Immunol. 107, 536–541. Beninati, C., Oggioni, M. R., Boccanera, M., Spinosa, M. R., Maggi, T., Conti, S., Magliani, W., De Bernardis, F., Teti, G., Cassone, A., Pozzi, G., and Polonelli, L. (2000). Therapy of mucosal candidiasis by expression of an anti-idiotype in human commensal bacteria. Nat. Biotechnol. 18, 1060–1064. Bermudez-Humaran, L. G., Cortes-Perez, N. G., Le Loir, Y., Gruss, A., Rodriguez-Padilla, C., Saucedo-Cardenas, O., Langella, P., and Montes de Oca-Luna, R. (2003). Fusion to

46

HANNIFFY et al.

a carrier protein and a synthetic propeptide enhances E7 HPV-16 production and secretion in Lactococcus lactis. Biotechnol. Prog. 19, 1101–1104. Bermudez-Humaran, L. G., Langella, P., Miyoshi, A., Gruss, A., Guerra, R. T., Montes de Oca-Luna, R., and Le Loir, Y. (2002). Production of human papillomavirus type 16 E7 protein in Lactococcus lactis. Appl. Environ. Microbiol. 68, 917–922. Bernasconi, E., Germond, J. E., Delley, M., Fritsche, R., and Corthesy, B. (2002). Lactobacillus bulgaricus proteinase expressed in Lactococcus lactis is a powerful carrier for cell wall-associated and secreted bovine beta-lactoglobulin fusion proteins. Appl. Environ. Microbiol. 68, 2917–2923. Bjorksten, B., Naaber, P., Sepp, E., and Mikelsaar, M. (1999). The intestinal microﬂora in allergic Estonian and Swedish 2-year-old children. Clin. Exp. Allergy 29, 342–346. Blum, S., and Schiffrin, E. J. (2003). Intestinal microﬂora and homeostasis of the mucosal immune response: Implications for probiotic bacteria? Curr. Issues Intest. Microbiol. 4, 53–60. Bolotin, A., Wincker, P., Mauger, S., Jaillon, O., Malarme, K., Weissenbach, J., Ehrlich, S. D., and Sorokin, A. (2001). The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res. 11, 731–753. Bonetti, A., Morelli, L., Campominosi, E., Ganora, E., and Sforza, F. (2003). Adherence of Lactobacillus plantarum P 17630 in soft-gel capsule formulation versus Doderlein’s bacillus in tablet formulation to vaginal epithelial cells. Minerva. Ginecol. 55, 279–284. Boris, S., Suarez, J. E., Vazquez, F., and Barbes, C. (1998). Adherence of human vaginal lactobacilli to vaginal epithelial cells and interaction with uropathogens. Infect. Immun. 66, 1985–1989. Borruel, N., Carol, M., Casellas, F., Antolin, M., de Lara, F., Espin, E., Naval, J., Guarner, F., and Malagelada, J. R. (2002). Increased mucosal tumour necrosis factor alpha production in Crohn’s disease can be downregulated ex vivo by probiotic bacteria. Gut 51, 659–664. Bosarge, J. R., Watt, J. M., McDaniel, D. O., Swiatlo, E., and McDaniel, L. S. (2001). Genetic immunization with the region encoding the alpha-helical domain of PspA elicits protective immunity against Streptococcus pneumoniae. Infect. Immun. 69, 5456–5463. Bottcher, M. F., Nordin, E. K., Sandin, A., Midtvedt, T., and Bjorksten, B. (2000). Microﬂora-associated characteristics in faeces from allergic and nonallergic infants. Clin. Exp. Allergy 30, 1590–1596. Boucher, I., Parrot, M., Gaudreau, H., Champagne, C. P., Vadeboncoeur, C., and Moineau, S. (2002). Novel food-grade plasmid vector based on melibiose fermentation for the genetic engineering of Lactococcus lactis. Appl. Environ. Microbiol. 68, 6152–6161. Bousquet, J., Lockey, R., Malling, H. J., Alvarez-Cuesta, E., Canonica, G. W., Chapman, M. D., Creticos, P. J., Dayer, J. M., Durham, S. R., Demoly, P., Goldstein, R. J., Ishikawa, T., Ito, K., Kraft, D., Lambert, P. H., Lowenstein, H., Muller, U., Norman, P. S., Reisman, R. E., Valenta, R., Valovirta, E., and Yssel, H. (1998). Allergen immunotherapy: Therapeutic vaccines for allergic diseases. World Health Organization. American Academy of Allergy, Asthma and Immunology. Ann. Allergy Asthma Immunol. 81, 401–405. Braat, H., De Jong, E. C., Van Den Brande, J. M., Kapsenberg, M. L., Peppelenbosch, M. P., Van Tol, E. A., and Van Deventer, S. J. (2003). Dichotomy between Lactobacillus rhamnosus and Klebsiella pneumoniae on dendritic cell phenotype and function. J. Mol. Med. 82(3), 197–205. Breiteneder, H., Pettenburger, K., Bito, A., Valenta, R., Kraft, D., Rumpold, H., Scheiner, O., and Breitenbach, M. (1989). The gene coding for the major birch pollen allergen

USE OF RECOMBINANT LAB IN HUMAN HEALTH

47

Betv1, is highly homologous to a pea disease resistance response gene. EMBO J. 8, 1935–1938. Briles, D. E., Hollingshead, S. K., King, J., Swift, A., Braun, P. A., Park, M. K., Ferguson, L. M., Nahm, M. H., and Nabors, G. S. (2000). Immunization of humans with recombinant pneumococcal surface protein A (rPspA) elicits antibodies that passively protect mice from fatal infection with Streptococcus pneumoniae bearing heterologous PspA. J. Infect. Dis. 182, 1694–1701. Briles, D. E., King, J. D., Gray, M. A., McDaniel, L. S., Swiatlo, E., and Benton, K. A. (1996). PspA, a protection-eliciting pneumococcal protein: Immunogenicity of isolated native PspA in mice. Vaccine 14, 858–867. Briles, D. E., Tart, R. C., Swiatlo, E., Dillard, J. P., Smith, P., Benton, K. A., Ralph, B. A., Brooks-Walter, A., Crain, M. J., Hollingshead, S. K., and McDaniel, L. S. (1998). Pneumococcal diversity: Considerations for new vaccine strategies with emphasis on pneumococcal surface protein A (PspA). Clin. Microbiol. Rev. 11, 645–657. Bron, P. A., Benchimol, M. G., Lambert, J., Palumbo, E., Deghorain, M., Delcour, J., De Vos, W. M., Kleerebezem, M., and Hols, P. (2002). Use of the alr gene as a food-grade selection marker in lactic acid bacteria. Appl. Environ. Microbiol. 68, 5663–5670. Bron, P. A., Grangette, C., Mercenier, A., de Vos, W. M., and Kleerebezem, M. (2004a). Identiﬁcation of Lactobacillus plantarum genes that are induced in the gastrointestinal tract of mice. J. Bacteriol. 186, 5721–5729. Bron, P. A., Hoffer, S. M., van Swam, I. I., de Vos, W. M., and Kleerebezem, M. (2004b). Selection and characterisation of conditionally active promoters in Lactobacillus plantarum, using alanine racemase as a promoter probe. Appl. Env. Microbiol. 70, 310–317. Bron, P. A., Marco, M., Hoffer, S. M., van Mullekom, E., de Vos, W. M., and Kleerebezem, M. (2004c). In vitro analysis of the bile salt response in Lactobacillus plantarum WCFS1 and its translation to in situ conditions in the gastrointestinal tract. Submitted for publication. Bumann, D., Hueck, C., Aebischer, T., and Meyer, T. F. (2000). Recombinant live Salmonella spp. for human vaccination against heterologous pathogens. FEMS Immunol. Med. Microbiol. 27, 357–364. Campieri, M., and Gionchetti, P. (1999). Probiotics in inﬂammatory bowel disease: New insight to pathogenesis or a possible therapeutic alternative? Gastroenterology 116, 1246–1249. Cangemi de Gutierrez, R., Santos, V., and Nader-Macias, M. E. (2001). Protective effect of intranasally inoculated Lactobacillus fermentum against Streptococcus pneumoniae challenge on the mouse respiratory tract. FEMS Immunol. Med. Microbiol. 31, 187–195. Caplan, M. S., Miller-Catchpole, R., Kaup, S., Russell, T., Lickerman, M., Amer, M., Xiao, Y., and Thomson, R., Jr. (1999). Biﬁdobacterial supplementation reduces the incidence of necrotizing enterocolitis in a neonatal rat model. Gastroenterology 117, 577–583. Cebra, J., Han-Quiang, J., Sterzl, J., and Tlaskalova, H. (1999). The role of mucosal microbiota in the development and maintenance of the mucosal immune system. In ‘‘Handbook of Mucosal Immunology’’ (P. L. Ogra, J. Mestecky, M. E. Lamm, W. Strober, J. R. McGhee, and J. Bienenstock, eds.), pp. 267–280. Academic Press, New York. Champomier-Verges, M. C., Maguin, E., Mistou, M. Y., Anglade, P., and Chich, J. F. (2002). Lactic acid bacteria and proteomics: Current knowledge and perspectives. J. Chromatogr. B. Analyt. Technol. Biomed. Life Sci. 771, 329–342. Chang, T. L., Chang, C. H., Simpson, D. A., Xu, Q., Martin, P. K., Lagenaur, L. A., Schoolnik, G. K., Ho, D. D., Hillier, S. L., Holodniy, M., Lewicki, J. A., and Lee, P. P.

48

HANNIFFY et al.

(2003). Inhibition of HIV infectivity by a natural human isolate of Lactobacillus jensenii engineered to express functional two-domain CD4. Proc. Natl. Acad. Sci. USA 100, 11672–11677. Chen, T., Toivanen, P., and Vainio, O. (2002). Suppression of antigen-speciﬁc lymphocyte proliferation by Gram-positive bacterial cell walls. Apmis 110, 490–498. Chou, L. S., and Weimer, B. (1999). Isolation and characterization of acid- and biletolerant isolates from strains of Lactobacillus acidophilus. J. Dairy Sci. 82, 23–31. Christensen, H. R., Frokiaer, H., and Pestka, J. J. (2002). Lactobacilli differentially modulate expression of cytokines and maturation surface markers in murine dendritic cells. J. Immunol. 168, 171–178. Cleveland, M. G., Gorham, J. D., Murphy, T. L., Tuomanen, E., and Murphy, K. M. (1996). Lipoteichoic acid preparations of Gram-positive bacteria induce interleukin-12 through a CD14-dependent pathway. Infect. Immun. 64, 1906–1912. Cortes-Perez, N. G., Bermudez-Humaran, L. G., Le Loir, Y., Rodriguez-Padilla, C., Gruss, A., Saucedo-Cardenas, O., Langella, P., and Montes-de-Oca-Luna, R. (2003). Mice immunization with live lactococci displaying a surface anchored HPV-16 E7 oncoprotein. FEMS Microbiol. Lett. 229, 37–42. Crain, M. J., Waltman, W. D., 2nd, Turner, J. S., Yother, J., Talkington, D. F., McDaniel, L. S., Gray, B. M., and Briles, D. E. (1990). Pneumococcal surface protein A (PspA) is serologically highly variable and is expressed by all clinically important capsular serotypes of Streptococcus pneumoniae. Infect. Immun. 58, 3293–3299. Cross, M. L., Stevenson, L. M., and Gill, H. S. (2001). Anti-allergy properties of fermented foods: An important immunoregulatory mechanism of lactic acid bacteria? Int. Immunopharmacol. 1, 891–901. Cukrowska, B., LodInova-ZadnIkova, R., Enders, C., Sonnenborn, U., Schulze, J., and Tlaskalova-Hogenova, H. (2002). Speciﬁc proliferative and antibody responses of premature infants to intestinal colonization with nonpathogenic probiotic E. coli strain Nissle 1917. Scand J. Immunol. 55, 204–209. Curtiss III, R., Pereira, D. A., Hsu, J. C., Hull, S. C., Clark, J. E., Maturin, Sr., L. J., Goldschmidt, R., Moody, R., Inoue, M., and Alexander, L. (1977). Biological containment: The subordination of Escherichia coli K12. In ‘‘Recombinant Molecules: Impact on Science and Society’’ (R. F. Bers, Jr. and E. G. Bassett, eds.), pp. 45–56. Raven Press, New York. de Ruyter, P. G., Kuipers, O. P., Meijer, W. C., and de Vos, W. M. (1997). Food-grade controlled lysis of Lactococcus lactis for accelerated cheese ripening. Nat. Biotechnol. 15, 976–979. de Vos, W. M. (1999a). Gene expression systems for lactic acid bacteria. Curr. Opin. Microbiol. 2, 289–295. de Vos, W. M. (1999b). Safe and sustainable systems for food-grade fermentations by genetically modiﬁed lactic acid bacteria. International Dairy Journal 9, 3–10. Di Fabio, S., Medaglini, D., Rush, C. M., Corrias, F., Panzini, G. L., Pace, M., Verani, P., Pozzi, G., and Titti, F. (1998). Vaginal immunization of Cynomolgus monkeys with Streptococcus gordonii expressing HIV-1 and HPV 16 antigens. Vaccine 16, 485–492. Dieye, Y., Usai, S., Clier, F., Gruss, A., and Piard, J. C. (2001). Design of a protein-targeting system for lactic acid bacteria. J. Bacteriol. 183, 4157–4166. Djordjevic, G. M., and Klaenhammer, T. R. (1996). Positive selection, cloning vectors for Gram-positive bacteria based on a restriction endonuclease cassette. Plasmid 35, 37–45. Djordjevic, G. M., O’Sullivan, D. J., Walker, S. A., Conkling, M. A., and Klaenhammer, T. R. (1997). A triggered-suicide system designed as a defense against bacteriophages. J. Bacteriol. 179, 6741–6748.

USE OF RECOMBINANT LAB IN HUMAN HEALTH

49

Duchmann, R., Kaiser, I., Hermann, E., Mayet, W., Ewe, K., and Meyer zum Buschenfelde, K. H. (1995). Tolerance exists towards resident intestinal ﬂora but is broken in active inﬂammatory bowel disease (IBD). Clin. Exp. Immunol. 102, 448–455. Ebner, C., Schenk, S., Najaﬁan, N., Siemann, U., Steiner, R., Fischer, G. W., Hoffmann, K., Szepfalusi, Z., Scheiner, O., and Kraft, D. (1995). Nonallergic individuals recognize the same T cell epitopes of Bet v 1, the major birch pollen allergen, as atopic patients. J. Immunol. 154, 1932–1940. Enouf, V., Langella, P., Commissaire, J., Cohen, J., and Corthier, G. (2001). Bovine rotavirus nonstructural protein 4 produced by Lactococcus lactis is antigenic and immunogenic. Appl. Environ. Microbiol. 67, 1423–1428. Fabia, R., Ar’Rajab, A., Johansson, M. L., Willen, R., Andersson, R., Molin, G., and Bengmark, S. (1993). The effect of exogenous administration of Lactobacillus reuteri R2LC and oat ﬁber on acetic acid-induced colitis in the rat. Scand. J. Gastroenterol. 28, 155–162. Fang, H., Elina, T., Heikki, A., and Seppo, S. (2000). Modulation of humoral immune response through probiotic intake. FEMS Immunol. Med. Microbiol. 29, 47–52. Farrell, R. J., and Peppercorn, M. A. (2002). Ulcerative colitis. Lancet 359, 331–340. Ferreira, F. D., Hoffmann-Sommergruber, K., Breiteneder, H., Pettenburger, K., Ebner, C., Sommergruber, W., Steiner, R., Bohle, B., Sperr, W. R., Valent, P., et al. (1993). Puriﬁcation and characterization of recombinant Bet v I, the major birch pollen allergen. Immunological equivalence to natural Bet v I. J. Biol. Chem. 268, 19574–19580. Fioramonti, J., Theodorou, V., and Bueno, L. (2003). Probiotics: What are they? What are their effects on gut physiology? Best Pract. Res. Clin. Gastroenterol. 17, 711–724. Fitzsimons, A., Hols, P., Jore, J., Leer, R. J., O’Connell, M., and Delcour, J. (1994). Development of an amylolytic Lactobacillus plantarum silage strain expressing the Lactobacillus amylovorus alpha-amylase gene. Appl. Environ. Microbiol. 60, 3529–3535. Frazier, C. L., San Filippo, J., Lambowitz, A. M., and Mills, D. A. (2003). Genetic manipulation of Lactococcus lactis by using targeted group II introns: Generation of stable insertions without selection. Appl. Environ. Microbiol. 69, 1121–1128. Froseth, B. R., and McKay, L. L. (1991). Molecular characterization of the nisin resistance region of Lactococcus lactis subsp. lactis biovar diacetylactis DRC3. Appl. Environ. Microbiol. 57, 804–811. Fu, X., and Xu, J. G. (2000). Development of a chromosome-plasmid balanced lethal system for Lactobacillus acidophilus with thyA gene as selective marker. Microbiol. Immunol. 44, 551–556. Fujiwara, M., Kaneko, T., Iwana, H., Taketomo, N., Tsunoo, H., Kanno, J., Ohkusa, T., and Okayasu, I. (2003). Inhibitory effects of Biﬁdobacterium longum on experimental ulcerative colitis induced in mice by synthetic dextran sulfate sodium. Digestion 67, 90–95. Fukushima, Y., Kawata, Y., Hara, H., Terada, A., and Mitsuoka, T. (1998). Effect of a probiotic formula on intestinal immunoglobulin A production in healthy children. Int. J. Food Microbiol. 42, 39–44. Fuller, R. (1989). Probiotics in man and animals. J. Appl. Bacteriol. 66, 365–378. Funda, D. P., Tuckova, L., Farre, M. A., Iwase, T., Moro, I., and Tlaskalova-Hogenova, H. (2001). CD14 is expressed and released as soluble CD14 by human intestinal epithelial cells in vitro: Lipopolysaccharide activation of epithelial cells revisited. Infect. Immun. 69, 3772–3781.

50

HANNIFFY et al.

Gajhede, M., Osmark, P., Poulsen, F. M., Ipsen, H., Larsen, J. N., Joost van Neerven, R. J., Schou, C., Lowenstein, H., and Spangfort, M. D. (1996). X-ray and NMR structure of Bet v 1, the origin of birch pollen allergy. Nat. Struct. Biol. 3, 1040–1045. Gallichan, W. S., and Rosenthal, K. L. (1996). Effects of the estrous cycle on local humoral immune responses and protection of intranasally immunized female mice against herpes simplex virus type 2 infection in the genital tract. Virology 224, 487–497. Gicquel, B. (1995). BCG as a vector for the construction of multivalent recombinant vaccines. Biologicals 23, 113–118. Gil, M. T., Perez-Arellano, I., Buesa, J., and Perez-Martinez, G. (2001). Secretion of the rotavirus VP8* protein in Lactococcus lactis. FEMS Microbiol. Lett. 203, 269–274. Gilbert, C., Robinson, K., Le Page, R. W., and Wells, J. M. (2000). Heterologous expression of an immunogenic pneumococcal type 3 capsular polysaccharide in Lactococcus lactis. Infect. Immun. 68, 3251–3260. Gill, H. S., Rutherfurd, K. J., Prasad, J., and Gopal, P. K. (2000). Enhancement of natural and acquired immunity by Lactobacillus rhamnosus (HN001), Lactobacillus acidophilus (HN017) and Biﬁdobacterium lactis (HN019). Br. J. Nutr. 83, 167–176. Gillgrass, A. E., Ashkar, A. A., Rosenthal, K. L., and Kaushic, C. (2003). Prolonged exposure to progesterone prevents induction of protective mucosal responses following intravaginal immunization with attenuated herpes simplex virus type 2. J. Virol. 77, 9845–9851. Giomarelli, B., Provvedi, R., Meacci, F., Maggi, T., Medaglini, D., Pozzi, G., Mori, T., McMahon, J. B., Gardella, R., and Boyd, M. R. (2002). The microbicide cyanovirin-N expressed on the surface of commensal bacterium Streptococcus gordonii captures HIV-1. AIDS 16, 1351–1356. Gionchetti, P., Rizzello, F., Venturi, A., Brigidi, P., Matteuzzi, D., Bazzocchi, G., Poggioli, G., Miglioli, M., and Campieri, M. (2000). Oral bacteriotherapy as maintenance treatment in patients with chronic pouchitis: A double-blind, placebo-controlled trial. Gastroenterology 119, 305–309. Gionchetti, P., Rizzello, F., Helwig, U., Venturi, A., Lammers, K. M., Brigidi, P., Vitali, P., Poggioli, G., and Campieri, M. (2003). Prophylaxis of pouchitis onset with probiotic therapy: A double-blind, placebo-controlled trial. Gastroenterology 124, 1202–1209. Glenting, J., Madsen, S. M., Vrang, A., Fomsgaard, A., and Israelsen, H. (2002). A plasmid selection system in Lactococcus lactis and its use for gene expression in L. lactis and human kidney ﬁbroblasts. Appl. Environ. Microbiol. 68, 5051–5056. Gluck, U., and Gebbers, J. O. (2003). Ingested probiotics reduce nasal colonization with pathogenic bacteria (Staphylococcus aureus, Streptococcus pneumoniae, and betahemolytic streptococci). Am. J. Clin. Nutr. 77, 517–520. Goeddel, D. V., Heyneker, H. L., Hozumi, T., Arentzen, R., Itakura, K., Yansura, D. G., Ross, M. J., Miozzari, G., Crea, R., and Seeburg, P. H. (1979a). Direct expression in Escherichia coli of a DNA sequence coding for human growth hormone. Nature 281, 544–548. Goeddel, D. V., Kleid, D. G., Bolivar, F., Heyneker, H. L., Yansura, D. G., Crea, R., Hirose, T., Kraszewski, A., Itakura, K., and Riggs, A. D. (1979b). Expression in Escherichia coli of chemically synthesized genes for human insulin. Proc. Natl. Acad. Sci. USA 76, 106–110. Grangette, C., Muller-Alouf, H., Geoffroy, M., Goudercourt, D., Turneer, M., and Mercenier, A. (2002). Protection against tetanus toxin after intragastric administration of two recombinant lactic acid bacteria: Impact of strain viability and in vivo persistence. Vaccine 20, 3304–3309.

USE OF RECOMBINANT LAB IN HUMAN HEALTH

51

Grangette, C., Muller-Alouf, H., Goudercourt, D., Geoffroy, M. C., Turneer, M., and Mercenier, A. (2001). Mucosal immune responses and protection against tetanus toxin after intranasal immunization with recombinant Lactobacillus plantarum. Infect. Immun. 69, 1547–1553. Gruzza, M., Langella, P., Duval-Iﬂah, Y., and Ducluzeau, R. (1993). Gene transfer from engineered Lactococcus lactis strains to Enterococcus faecalis in the digestive tract of gnotobiotic mice. Microb. Releases 2, 121–125. Guillot, A., Gitton, C., Anglade, P., and Mistou, M. Y. (2003). Proteomic analysis of Lactococcus lactis, a lactic acid bacterium. Proteomics 3, 337–354. Hakenbeck, R., Grebe, T., Zahner, D., and Stock, J. B. (1999). Beta-lactam resistance in Streptococcus pneumoniae: Penicillin-binding proteins and non-penicillin-binding proteins. Mol. Microbiol. 33, 673–678. Haller, D., Bode, C., Hammes, W. P., Pfeifer, A. M., Schiffrin, E. J., and Blum, S. (2000). Non-pathogenic bacteria elicit a differential cytokine response by intestinal epithelial cell/leucocyte co-cultures. Gut 47, 79–87. Hanniffy, S., Carter, A. T., and Wells, J. M. (2004). Protection from respiratory infection with Streptococcus pneumoniae by intranasal vaccination with Lc. lactis expressing PspA. Submitted for publication. Hashiba, H., Takiguchi, R., Jyoho, K., and Aoyama, K. (1992). Establishment of a hostvector system in Lactobacillus helveticus with beta-galactosidase activity as a selection marker. Biosci. Biotechnol. Biochem. 56, 190–194. Hatakka, K., Savilahti, E., Ponka, A., Meurman, J. H., Poussa, T., Nase, L., Saxelin, M., and Korpela, R. (2001). Effect of long term consumption of probiotic milk on infections in children attending day care centres: Double blind, randomised trial. BMJ 322, 1327. Heilig, H. G., Zoetendal, E. G., Vaughan, E. E., Marteau, P., Akkermans, A. D., and de Vos, W. M. (2002). Molecular diversity of Lactobacillus spp. and other lactic acid bacteria in the human intestine as determined by speciﬁc ampliﬁcation of 16S ribosomal DNA. Appl. Environ. Microbiol. 68, 114–123. Herias, M. V., Hessle, C., Telemo, E., Midtvedt, T., Hanson, L. A., and Wold, A. E. (1999). Immunomodulatory effects of Lactobacillus plantarum colonizing the intestine of gnotobiotic rats. Clin. Exp. Immunol. 116, 283–290. Heyman, M. (2000). Effect of lactic acid bacteria on diarrheal diseases. J. Am. Coll. Nutr. 19, 137S–146S. Hillier, S. L., Krohn, M. A., Rabe, L. K., Klebanoff, S. J., and Eschenbach, D. A. (1993). The normal vaginal ﬂora, H2O2-producing lactobacilli, and bacterial vaginosis in pregnant women. Clin. Infect. Dis. 16, S273–S281. Hillman, J. D. (2002). Genetically modiﬁed Streptococcus mutans for the prevention of dental caries. Antonie Van Leeuwenhoek 82, 361–366. Holma, R., Salmenpera, P., Lohi, J., Vapaatalo, H., and Korpela, R. (2001). Effects of Lactobacillus rhamnosus GG and Lactobacillus reuteri R2LC on acetic acid-induced colitis in rats. Scand. J. Gastroenterol. 36, 630–635. Hols, P., Defrenne, C., Ferain, T., Derzelle, S., Delplace, B., and Delcour, J. (1997a). The alanine racemase gene is essential for growth of Lactobacillus plantarum. J. Bacteriol. 179, 3804–3807. Hols, P., Ferain, T., Garmyn, D., Bernard, N., and Delcour, J. (1994). Use of homologous expression-secretion signals and vector-free stable chromosomal integration in engineering of Lactobacillus plantarum for alpha-amylase and levanase expression. Appl. Environ. Microbiol. 60, 1401–1413.

52

HANNIFFY et al.

Hols, P., Kleerebezem, M., Schanck, A. N., Ferain, T., Hugenholtz, J., Delcour, J., and de Vos, W. M. (1999). Conversion of Lactococcus lactis from homolactic to homoalanine fermentation through metabolic engineering. Nat. Biotechnol. 17, 588–592. Hols, P., Slos, P., Dutot, P., Reymund, J., Chabot, P., Delplace, B., Delcour, J., and Mercenier, A. (1997b). Efﬁcient secretion of the model antigen M6-gp41E in Lactobacillus plantarum NCIMB 8826. Microbiology 143, 2733–2741. Holzapfel, W. H., Haberer, P., Snel, J., Schillinger, U., and Huis in’t Veld, J. H. (1998). Overview of gut ﬂora and probiotics. Int. J. Food Microbiol. 41, 85–101. Hori, T., Kiyoshima, J., Shida, K., and Yasui, H. (2001). Effect of intranasal administration of Lactobacillus casei Shirota on inﬂuenza virus infection of upper respiratory tract in mice. Clin. Diagn. Lab. Immunol. 8, 593–597. Hori, T., Kiyoshima, J., Shida, K., and Yasui, H. (2002). Augmentation of cellular immunity and reduction of inﬂuenza virus titer in aged mice fed Lactobacillus casei strain Shirota. Clin. Diagn. Lab. Immunol. 9, 105–108. Hudcovic, T., Stepankova, R., Cebra, J., and Tlaskalova-Hogenova, H. (2001). The role of microﬂora in the development of intestinal inﬂammation: Acute and chronic colitis induced by dextran sulfate in germ-free and conventionally reared immunocompetent and immunodeﬁcient mice. Folia Microbiol (Praha) 46, 565–572. Hyronimus, B., Le Marrec, C., Sassi, A. H., and Deschamps, A. (2000). Acid and bile tolerance of spore-forming lactic acid bacteria. Int. J. Food Microbiol. 61, 193–197. Ichikawa, H., Kuroiwa, T., Inagaki, A., Shineha, R., Nishihira, T., Satomi, S., and Sakata, T. (1999). Probiotic bacteria stimulate gut epithelial cell proliferation in rat. Dig. Dis. Sci. 44, 2119–2123. Isolauri, E., Sutas, Y., Kankaanpaa, P., Arvilommi, H., and Salminen, S. (2001). Probiotics: Effects on immunity. Am. J. Clin. Nutr. 73, 444S–450S. Israel, D. A., Salama, N., Krishna, U., Rieger, U. M., Atherton, J. C., Falkow, S., and Peek, R. M., Jr. (2001). Helicobacter pylori genetic diversity within the gastric niche of a single human host. Proc. Natl. Acad. Sci. USA 98, 14625–14630. Israelsen, H., Madsen, S. M., Johansen, E., Vrang, A., and Hansen, E. B. (1995a). Environmentally regulated promoters in lactococci. Dev. Biol. Stand. 85, 443–448. Israelsen, H., Madsen, S. M., Vrang, A., Hansen, E. B., and Johansen, E. (1995b). Cloning and partial characterization of regulated promoters from Lactococcus lactis Tn917lacZ integrants with the new promoter probe vector, pAK80. Appl. Environ. Microbiol. 61, 2540–2547. Jahn-Schmid, B., Wiedermann, U., Bohle, B., Repa, A., Kraft, D., and Ebner, C. (1999). Oligodeoxynucleotides containing CpG motifs modulate the allergic TH2 response of BALB/c mice to Bet v 1, the major birch pollen allergen. J. Allergy Clin. Immunol. 104, 1015–1023. Janeway, C. A., Jr., and Medzhitov, R. (2002). Innate immune recognition. Annu. Rev. Immunol. 20, 197–216. Epub 2001 Oct 2004. Kalliomaki, M., and Isolauri, E. (2003). Role of intestinal ﬂora in the development of allergy. Curr. Opin. Allergy Clin. Immunol. 3, 15–20. Kalliomaki, M., Salminen, S., Arvilommi, H., Kero, P., Koskinen, P., and Isolauri, E. (2001). Probiotics in primary prevention of atopic disease: A randomised placebocontrolled trial. Lancet 357, 1076–1079. Kalliomaki, M., Salminen, S., Poussa, T., Arvilommi, H., and Isolauri, E. (2003). Probiotics and prevention of atopic disease: 4-year follow-up of a randomised placebo-controlled trial. Lancet 361, 1869–1871. Kantele, A., Hakkinen, M., Moldoveanu, Z., Lu, A., Savilahti, E., Alvarez, R. D., Michalek, S., and Mestecky, J. (1998). Differences in immune responses induced by oral and

USE OF RECOMBINANT LAB IN HUMAN HEALTH

53

rectal immunizations with Salmonella typhi Ty21a: Evidence for compartmentalization within the common mucosal immune system in humans. Infect. Immun. 66, 5630–5635. Kariuki, S., and Hart, C. A. (2001). Global aspects of antimicrobial-resistant enteric bacteria. Curr. Opin. Infect. Dis. 14, 579–586. Kato, I., Tanaka, K., and Yokokura, T. (1999). Lactic acid bacterium potently induces the production of interleukin-12 and interferon-gamma by mouse splenocytes. Int. J. Immunopharmacol. 21, 121–131. Kaushic, C., Zhou, F., Murdin, A. D., and Wira, C. R. (2000). Effects of estradiol and progesterone on susceptibility and early immune responses to Chlamydia trachomatis infection in the female reproductive tract. Infect. Immun. 68, 4207–4216. Kili, A. O., Herzberg, M. C., Meyer, M. W., Zhao, X., and Tao, L. (1999). Streptococcal reporter gene-fusion vector for identiﬁcation of in vivo expressed genes. Plasmid 42, 67–72. Kim, T. S., Hur, J. W., Yu, M. A., Cheigh, C. I., Kim, K. N., Hwang, J. K., and Pyun, Y. R. (2003). Antagonism of Helicobacter pylori by bacteriocins of lactic acid bacteria. J. Food Prot. 66, 3–12. Kirjavainen, P. V., El-Nezami, H. S., Salminen, S. J., Ahokas, J. T., and Wright, P. F. (1999a). Effects of orally administered viable Lactobacillus rhamnosus GG and Propionibacterium freudenreichii subsp. shermanii JS on mouse lymphocyte proliferation. Clin. Diagn. Lab. Immunol. 6, 799–802. Kirjavainen, P. V., El-Nezami, H. S., Salminen, S. J., Ahokas, J. T., and Wright, P. F. (1999b). The effect of orally administered viable probiotic and dairy lactobacilli on mouse lymphocyte proliferation. FEMS Immunol. Med. Microbiol. 26, 131–135. Klaenhammer, T., Altermann, E., Arigoni, F., Bolotin, A., Breidt, F., Broadbent, J., Cano, R., Chaillou, S., Deutscher, J., Gasson, M., van de Guchte, M., Guzzo, J., Hartke, A., Hawkins, T., Hols, P., Hutkins, R., Kleerebezem, M., Kok, J., Kuipers, O., Lubbers, M., Maguin, E., McKay, L., Mills, D., Nauta, A., Overbeek, R., Pel, H., Pridmore, D., Saier, M., van Sinderen, D., Sorokin, A., Steele, J., O’Sullivan, D., de Vos, W., Weimer, B., Zagorec, M., and Siezen, R. (2002). Discovering lactic acid bacteria by genomics. Antonie Van Leeuwenhoek 82, 29–58. Kleerebezem, M., Beerthuyzen, M. M., Vaughan, E. E., de Vos, W. M., and Kuipers, O. P. (1997). Controlled gene expression systems for lactic acid bacteria: Transferable nisin-inducible expression cassettes for Lactococcus, Leuconostoc, and Lactobacillus spp. Appl. Environ. Microbiol. 63, 4581–4584. Kleerebezem, M., Boekhorst, J., van Kranenburg, R., Molenaar, D., Kuipers, O. P., Leer, R., Tarchini, R., Peters, S. A., Sandbrink, H. M., Fiers, M. W., Stiekema, W., Lankhorst, R. M., Bron, P. A., Hoffer, S. M., Groot, M. N., Kerkhoven, R., de Vries, M., Ursing, B., de Vos, W. M., and Siezen, R. J. (2003). Complete genome sequence of Lactobacillus plantarum WCFS1. Proc. Natl. Acad. Sci. USA 100, 1990–1995. Kociubinski, G., Perez, P., and De Antoni, G. (1999). Screening of bile resistance and bile precipitation in lactic acid bacteria and biﬁdobacteria. J. Food Prot. 62, 905–912. Koga, T., Mizobe, T., and Takumi, K. (1998). Antibacterial activity of Lactobacillus species against Vibrio species. Microbiol. Res. 153, 271–275. Konrad, A., Mahler, M., Flogerzi, B., Kalousek, M. B., Lange, J., Varga, L., and Seibold, F. (2003). Amelioration of murine colitis by feeding a solution of lysed Escherichia coli. Scand. J. Gastroenterol. 38, 172–179. Kruger, C., Hu, Y., Pan, Q., Marcotte, H., Hultberg, A., Delwar, D., van Dalen, P. J., Pouwels, P. H., Leer, R. J., Kelly, C. G., van Dollenweerd, C., Ma, J. K., and Hammarstrom, L.

54

HANNIFFY et al.

(2002). In situ delivery of passive immunity by lactobacilli producing single-chain antibodies. Nat. Biotechnol. 20, 702–706. Kruis, W., Schutz, E., Fric, P., Fixa, B., Judmaier, G., and Stolte, M. (1997). Doubleblind comparison of an oral Escherichia coli preparation and mesalazine in maintaining remission of ulcerative colitis. Aliment. Pharmacol Ther. 11, 853–858. Kruisselbrink, A., Heijne Den Bak-Glashouwer, M. J., Havenith, C. E., Thole, J. E., and Janssen, R. (2001). Recombinant Lactobacillus plantarum inhibits house dust mitespeciﬁc T-cell responses. Clin. Exp. Immunol. 126, 2–8. Kuipers, O. P., de Ruyter, P. G., Kleerebezem, M., and de Vos, W. M. (1997). Controlled over-production of proteins by lactic acid bacteria. Trends Biotechnol. 15, 135–140. Kuisma, J., Mentula, S., Jarvinen, H., Kahri, A., Saxelin, M., and Farkkila, M. (2003). Effect of Lactobacillus rhamnosus GG on ileal pouch inﬂammation and microbial ﬂora. Aliment. Pharmacol. Ther. 17, 509–515. Kullisaar, T., Zilmer, M., Mikelsaar, M., Vihalemm, T., Annuk, H., Kairane, C., and Kilk, A. (2002). Two antioxidative lactobacilli strains as promising probiotics. Int. J. Food Microbiol. 72, 215–224. Kunst, F., Ogasawara, N., Moszer, I., Albertini, A. M., Alloni, G., Azevedo, V., Bertero, M. G., Bessieres, P., Bolotin, A., Borchert, S., Borriss, R., Boursier, L., Brans, A., Braun, M., Brignell, S. C., Bron, S., Brouillet, S., Bruschi, C. V., Caldwell, B., Capuano, V., Carter, N. M., Choi, S. K., Codani, J. J., Connerton, I. F., Danchin, A., et al. (1997). The complete genome sequence of the Gram-positive bacterium Bacillus subtilis. Nature 390, 249–256. Kutteh, W. H., Kantele, A., Moldoveanu, Z., Crowley-Nowick, P. A., and Mestecky, J. (2001). Induction of speciﬁc immune responses in the genital tract of women after oral or rectal immunization and rectal boosting with Salmonella typhi Ty 21a vaccine. J. Reprod. Immunol. 52, 61–75. Lamine, F., Fioramonti, J., Bueno, L., Nepveu, F., Cauquil, E., Lobysheva, I., Eutamene, H., and Theodorou, V. (2004). Nitric oxide released by Lactobacillus farciminis improves TNBS-induced colitis in rats. Scand. J. Gastroenterol. 39, 37–45. Langella, P., and Le Loir, Y. (1999). Heterologous protein secretion in Lactococcus lactis: A novel antigen delivery system. Braz. J. Med. Biol. Res. 32, 191–198. Lee, M. H., Roussel, Y., Wilks, M., and Tabaqchali, S. (2001). Expression of Helicobacter pylori urease subunit B gene in Lactococcus lactis MG1363 and its use as a vaccine delivery system against H. pylori infection in mice. Vaccine 19, 3927–3935. Lee, S. F., Halperin, S. A., Knight, J. B., and Tait, A. (2002). Puriﬁcation and immunogenicity of a recombinant Bordetella pertussis S1S3FHA fusion protein expressed by Streptococcus gordonii. Appl. Environ. Microbiol. 68, 4253–4258. Lee, S. F., March, R. J., Halperin, S. A., Faulkner, G., and Gao, L. (1999). Surface expression of a protective recombinant pertussis toxin S1 subunit fragment in Streptococcus gordonii. Infect. Immun. 67, 1511–1516. Lee, Y. K., Puong, K. Y., Ouwehand, A. C., and Salminen, S. (2003). Displacement of bacterial pathogens from mucus and Caco-2 cell surface by lactobacilli. J. Med. Microbiol. 52, 925–930. Leenhouts, K., Bolhuis, A., Venema, G., and Kok, J. (1998). Construction of a food-grade multiple-copy integration system for Lactococcus lactis. Appl. Microbiol. Biotechnol. 49, 417–423. Leenhouts, K., Buist, G., Bolhuis, A., ten Berge, A., Kiel, J., Mierau, I., Dabrowska, M., Venema, G., and Kok, J. (1996). A general system for generating unlabelled gene replacements in bacterial chromosomes. Mol. Gen. Genet. 253, 217–224.

USE OF RECOMBINANT LAB IN HUMAN HEALTH

55

Levine, M. M., Galen, J., Barry, E., Noriega, F., Chatﬁeld, S., Sztein, M., Dougan, G., and Tacket, C. (1996). Attenuated Salmonella as live oral vaccines against typhoid fever and as live vectors. J. Biotechnol. 44, 193–196. Link-Amster, H., Rochat, F., Saudan, K. Y., Mignot, O., and Aeschlimann, J. M. (1994). Modulation of a speciﬁc humoral immune response and changes in intestinal ﬂora mediated through fermented milk intake. FEMS Immunol. Med. Microbiol. 10, 55–63. Liu, C. Q., Leelawatcharamas, V., Harvey, M. L., and Dunn, N. W. (1996). Cloning vectors for lactococci based on a plasmid encoding resistance to cadmium. Curr. Microbiol. 33, 35–39. Lodinova-Zadnikova, R., Cukrowska, B., and Tlaskalova-Hogenova, H. (2003). Oral administration of probiotic Escherichia coli after birth reduces frequency of allergies and repeated infections later in life (after 10 and 20 years). Int. Arch. Allergy Immunol. 131, 209–211. Maassen, C. B., Laman, J. D., den Bak-Glashouwer, M. J., Tielen, F. J., van Holten-Neelen, J. C., Hoogteijling, L., Antonissen, C., Leer, R. J., Pouwels, P. H., Boersma, W. J., and Shaw, D. M. (1999). Instruments for oral disease-intervention strategies: Recombinant Lactobacillus casei expressing tetanus toxin fragment C for vaccination or myelin proteins for oral tolerance induction in multiple sclerosis. Vaccine 17, 2117–2128. Maassen, C. B., van Holten-Neelen, C., Balk, F., den Bak-Glashouwer, M. J., Leer, R. J., Laman, J. D., Boersma, W. J., and Claassen, E. (2000). Strain-dependent induction of cytokine proﬁles in the gut by orally administered Lactobacillus strains. Vaccine 18, 2613–2623. MacCormick, C. A., Grifﬁn, H. G., and Gasson, M. J. (1995). Construction of a food-grade host/vector system for Lactococcus lactis based on the lactose operon. FEMS Microbiol. Lett. 127, 105–109. Mack, D. R., Ahrne, S., Hyde, L., Wei, S., and Hollingsworth, M. A. (2003). Extracellular MUC3 mucin secretion follows adherence of Lactobacillus strains to intestinal epithelial cells in vitro. Gut 52, 827–833. Madsen, K. L., Doyle, J. S., Jewell, L. D., Tavernini, M. M., and Fedorak, R. N. (1999a). Lactobacillus species prevents colitis in interleukin 10 gene-deﬁcient mice. Gastroenterology 116, 1107–1114. Madsen, S. M., Albrechtsen, B., Hansen, E. B., and Israelsen, H. (1996). Cloning and transcriptional analysis of two threonine biosynthetic genes from Lactococcus lactis MG1614. J. Bacteriol. 178, 3689–3694. Madsen, S. M., Arnau, J., Vrang, A., Givskov, M., and Israelsen, H. (1999b). Molecular characterization of the pH-inducible and growth phase-dependent promoter P170 of Lactococcus lactis. Mol. Microbiol. 32, 75–87. Maggi, L., Mastromarino, P., Macchia, S., Brigidi, P., Pirovano, F., Matteuzzi, D., and Conte, U. (2000). Technological and biological evaluation of tablets containing different strains of lactobacilli for vaginal administration. Eur. J. Pharm. Biopharm. 50, 389–395. Maguin, E., Prevost, H., Ehrlich, S. D., and Gruss, A. (1996). Efﬁcient insertional mutagenesis in lactococci and other Gram-positive bacteria. J. Bacteriol. 178, 931–935. Mahan, M. J., Heithoff, D. M., Sinsheimer, R. L., and Low, D. A. (2000). Assessment of bacterial pathogenesis by analysis of gene expression in the host. Annu. Rev. Genet. 34, 139–164. Makrides, S. C. (1996). Strategies for achieving high-level expression of genes in Escherichia coli. Microbiol. Rev. 60, 512–538.

56

HANNIFFY et al.

Mao, Y., Nobaek, S., Kasravi, B., Adawi, D., Stenram, U., Molin, G., and Jeppsson, B. (1996). The effects of Lactobacillus strains and oat ﬁber on methotrexate-induced enterocolitis in rats. Gastroenterology 111, 334–344. Marsh, D. G., Neely, J. D., Breazeale, D. R., Ghosh, B., Freidhoff, L. R., Ehrlich-Kautzky, E., Schou, C., Krishnaswamy, G., and Beaty, T. H. (1994). Linkage analysis of IL-4 and other chromosome 5q31.1 markers and total serum immunoglobulin E concentrations. Science 264, 1152–1156. Marshall, J. D., Secrist, H., DeKruyff, R. H., Wolf, S. F., and Umetsu, D. T. (1995). IL-12 inhibits the production of IL-4 and IL-10 in allergen-speciﬁc human CD4þ T lymphocytes. J. Immunol. 155, 111–117. Marteau, P., Seksik, P., and Shanahan, F. (2003). Manipulation of the bacterial ﬂora in inﬂammatory bowel disease. Best. Pract. Res. Clin. Gastroenterol. 17, 47–61. Martin, M. C., Alonso, J. C., Suarez, J. E., and Alvarez, M. A. (2000). Generation of foodgrade recombinant lactic acid bacterium strains by site-speciﬁc recombination. Appl. Environ. Microbiol. 66, 2599–2604. Mastromarino, P., Brigidi, P., Macchia, S., Maggi, L., Pirovano, F., Trinchieri, V., Conte, U., and Matteuzzi, D. (2002). Characterization and selection of vaginal Lactobacillus strains for the preparation of vaginal tablets. J. Appl. Microbiol. 93, 884–893. Matar, C., Valdez, J. C., Medina, M., Rachid, M., and Perdigon, G. (2001). Immunomodulating effects of milks fermented by Lactobacillus helveticus and its non-proteolytic variant. J. Dairy Res. 68, 601–609. Matsuzaki, T., Yamazaki, R., Hashimoto, S., and Yokokura, T. (1998). The effect of oral feeding of Lactobacillus casei strain Shirota on immunoglobulin E production in mice. J. Dairy Sci. 81, 48–53. McGroarty, J. A., and Reid, G. (1988). Detection of a Lactobacillus substance that inhibits Escherichia coli. Can. J. Microbiol. 34, 974–978. Medaglini, D., Oggioni, M. R., and Pozzi, G. (1998). Vaginal immunization with recombinant gram-positive bacteria. Am. J. Reprod. Immunol. 39, 199–208. Medina, E., and Guzman, C. A. (2001). Use of live bacterial vaccine vectors for antigen delivery: Potential and limitations. Vaccine 19, 1573–1580. Mercenier, A., Muller-Alouf, H., and Grangette, C. (2000). Lactic acid bacteria as live vaccines. Curr. Issues. Mol. Biol. 2, 17–25. Mercenier, A., Pavan, S., and Pot, B. (2003). Probiotics as biotherapeutic agents: Present knowledge and future prospects. Curr. Pharm. Des. 9, 175–191. Metlay, J. P., and Singer, D. E. (2002). Outcomes in lower respiratory tract infections and the impact of antimicrobial drug resistance. Clin. Microbiol. Infect. 8(Suppl. 2), 1–11. Michalek, S. M., Eldridge, J. H., Curtiss III, R., and Rosenthal, K. L. (1994). Antigen delivery systems: New approaches to mucosal immunization. In ‘‘Handbook of Mucosal Immunology’’ (P. L. Ogra, J. Mestecky, M. E. Lamm, W. Strober, J. R. McGhee, and J. Bienenstock, eds.), pp. 373–380. Academic Press, Inc., Boston. Midolo, P. D., Lambert, J. R., Hull, R., Luo, F., and Grayson, M. L. (1995). In vitro inhibition of Helicobacter pylori NCTC 11637 by organic acids and lactic acid bacteria. J. Appl. Bacteriol. 79, 475–479. Miettinen, M., Vuopio-Varkila, J., and Varkila, K. (1996). Production of human tumor necrosis factor alpha, interleukin-6, and interleukin-10 is induced by lactic acid bacteria. Infect. Immun. 64, 5403–5405. Miyoshi, A., Poquet, I., Azevedo, V., Commissaire, J., Bermudez-Humaran, L., Domakova, E., Le Loir, Y., Oliveira, S. C., Gruss, A., and Langella, P. (2002). Controlled production

USE OF RECOMBINANT LAB IN HUMAN HEALTH

57

of stable heterologous proteins in Lactococcus lactis. Appl. Environ. Microbiol. 68, 3141–3146. Moineau, S., and Goulet, J. (1997). Effect of feeding feremented milks on the pulmonary macrophage activity of mice. Milchwissenshaft 46, 551–554. Molenaar, D., Bron, P. A., Siezen, R. J., Bringel, F., de Vos, W. M., and Kleerebezem, M. (2003). Comparative analysis of 20 strains of Lactobacillus plantarum using microarrays. Manuscript in preparation. Molina, L., Ramos, C., Ronchel, M. C., Molin, S., and Ramos, J. L. (1998). Construction of an efﬁcient biologically contained Pseudomonas putida strain and its survival in outdoor assays. Appl. Environ. Microbiol. 64, 2072–2078. Morita, H., He, F., Fuse, T., Ouwehand, A. C., Hashimoto, H., Hosoda, M., Mizumachi, K., and Kurisaki, J. (2002). Adhesion of lactic acid bacteria to caco-2 cells and their effect on cytokine secretion. Microbiol. Immunol. 46, 293–297. Morris, D. L. (1999). WHO position paper on oral (sublingual) immunotherapy. Ann. Allergy Asthma Immunol. 83, 423–424. Mosmann, T. R., and Coffman, R. L. (1989). TH1 and TH2 cells: Different patterns of lymphokine secretion lead to different functional properties. Annu. Rev. Immunol. 7, 145–173. Mukai, T., Asasaka, T., Sato, E., Mori, K., Matsumoto, M., and Ohori, H. (2002). Inhibition of binding of Helicobacter pylori to the glycolipid receptors by probiotic Lactobacillus reuteri. FEMS Immunol. Med. Microbiol. 32, 105–110. Murosaki, S., Yamamoto, Y., Ito, K., Inokuchi, T., Kusaka, H., Ikeda, H., and Yoshikai, Y. (1998). Heat-killed Lactobacillus plantarum L-137 suppresses naturally fed antigenspeciﬁc IgE production by stimulation of IL-12 production in mice. J. Allergy Clin. Immunol. 102, 57–64. Neeser, J. R., Granato, D., Rouvet, M., Servin, A., Teneberg, S., and Karlsson, K. A. (2000). Lactobacillus johnsonii La1 shares carbohydrate-binding speciﬁcities with several enteropathogenic bacteria. Glycobiology 10, 1193–1199. Netherwood, T., Bowden, R., Harrison, P., O’Donnell, A. G., Parker, D. S., and Gilbert, H. J. (1999). Gene transfer in the gastrointestinal tract. Appl. Environ. Microbiol. 65, 5139–5141. Neumann, E., Oliveira, M. A., Cabral, C. M., Moura, L. N., Nicoli, J. R., Vieira, E. C., Cara, D. C., Podoprigora, G. I., and Vieira, L. Q. (1998). Monoassociation with Lactobacillus acidophilus UFV-H2b20 stimulates the immune defense mechanisms of germfree mice. Braz. J. Med. Biol. Res. 31, 1565–1573. Norton, P. M., Brown, H. W., Wells, J. M., Macpherson, A. M., Wilson, P. W., and Le Page, R. W. (1996). Factors affecting the immunogenicity of tetanus toxin fragment C expressed in Lactococcus lactis. FEMS Immunol. Med. Microbiol. 14, 167–177. Norton, P. M., Wells, J. M., Brown, H. W., Macpherson, A. M., and Le Page, R. W. (1997). Protection against tetanus toxin in mice nasally immunized with recombinant Lactococcus lactis expressing tetanus toxin fragment C. Vaccine 15, 616–619. Nouaille, S., Ribeiro, L. A., Miyoshi, A., Pontes, D., Le Loir, Y., Oliveira, S. C., Langella, P., and Azevedo, V. (2003). Heterologous protein production and delivery systems for Lactococcus lactis. Genet. Mol. Res. 2, 102–111. Ogawa, M., Shimizu, K., Nomoto, K., Takahashi, M., Watanuki, M., Tanaka, R., Tanaka, T., Hamabata, T., Yamasaki, S., and Takeda, Y. (2001). Protective effect of Lactobacillus casei strain Shirota on Shiga toxin-producing Escherichia coli O157:H7 infection in infant rabbits. Infect. Immun. 69, 1101–1108.

58

HANNIFFY et al.

Oggioni, M. R., Beninati, C., Boccanera, M., Medaglini, D., Spinosa, M. R., Maggi, T., Conti, S., Magliani, W., De Bernardis, F., Teti, G., Cassone, A., Pozzi, G., and Polonelli, L. (2001). Recombinant Streptococcus gordonii for mucosal delivery of a scFv microbicidal antibody. Int. Rev. Immunol. 20, 275–287. O’Hagan, D. T. (1994). Oral immunization and the common mucosal immune system. In ‘‘Novel Delivery Systems for Oral Vaccines’’ (D. T. O’Hagan, ed.), pp. 1–24. Boca Raton, Florida: CRC Press. Oliveira, M. L., Monedero, V., Miyaji, E. N., Leite, L. C., Lee Ho, P., and Perez-Martinez, G. (2003). Expression of Streptococcus pneumoniae antigens, PsaA (pneumococcal surface antigen A) and PspA (pneumococcal surface protein A) by Lactobacillus casei. FEMS Microbiol. Lett. 227, 25–31. Osset, J., Bartolome, R. M., Garcia, E., Andreu, A., Boris, S., Suarez, J. E., Vazquez, F., Barbes, C., Reid, G., Chan, R. C., Bruce, A. W., Costerton, J. W., McGroarty, J. A., Oggioni, M. R., Medaglini, D., Maggi, T., Pozzi, G., and Russell, M. W. (2001). Assessment of the capacity of Lactobacillus to inhibit the growth of uropathogens and block their adhesion to vaginal epithelial cells. J. Infect. Dis. 183, 485–491. Pavan, S., Hols, P., Delcour, J., Geoffroy, M. C., Grangette, C., Kleerebezem, M., and Mercenier, A. (2000). Adaptation of the nisin-controlled expression system in Lactobacillus plantarum: A tool to study in vivo biological effects. Appl. Environ. Microbiol. 66, 4427–4432. Pearson, B. M., Pin, C., Wright, J., I’Anson, K., Humphrey, T., and Wells, J. M. (2003). Comparative genome analysis of Campylobacter jejuni using whole genome DNA microarrays. FEBS Lett. 554, 224–230. Pedersen, M. B., Jensen, P. R., Janzen, T., and Nilsson, D. (2002). Bacteriophage resistance of a thyA mutant of Lactococcus lactis blocked in DNA replication. Appl. Environ. Microbiol. 68, 3010–3023. Perdigon, G., Alvarez, S., Medina, M., Vintini, E., and Roux, E. (1999). Inﬂuence of the oral administration of lactic acid bacteria on IgA producing cells associated to bronchus. Int. J. Immunopathol. Pharmacol. 12, 97–102. Perner, A., and Rask-Madsen, J. (1999). Review article: The potential role of nitric oxide in chronic inﬂammatory bowel disorders. Aliment. Pharmacol. Ther. 13, 135–144. Pessi, T., Sutas, Y., Saxelin, M., Kallioinen, H., and Isolauri, E. (1999). Antiproliferative effects of homogenates derived from ﬁve strains of candidate probiotic bacteria. Appl. Environ. Microbiol. 65, 4725–4728. Platteeuw, C., van Alen-Boerrigter, I., van Schalkwijk, S., and de Vos, W. M. (1996). Foodgrade cloning and expression system for Lactococcus lactis. Appl. Environ. Microbiol. 62, 1008–1013. Porwollik, S., Wong, R. M., and McClelland, M. (2002). Evolutionary genomics of Salmonella: Gene acquisitions revealed by microarray analysis. Proc. Natl. Acad. Sci. USA 99, 8956–8961. Posno, M., Heuvelmans, P. T., van Giezen, M. J., Lokman, B. C., Leer, R. J., and Pouwels, P. H. (1991). Complementation of the inability of Lactobacillus strains to utilize D-xylose with D-xylose catabolism-encoding genes of Lactobacillus pentosus. Appl. Environ. Microbiol. 57, 2764–2766. Pouwels, P. H., Leer, R. J., and Boersma, W. J. (1996). The potential of Lactobacillus as a carrier for oral immunization: Development and preliminary characterization of vector systems for targeted delivery of antigens. J. Biotechnol. 44, 183–192. Pouwels, P. H., Vriesema, A., Martinez, B., Tielen, F. J., Seegers, J. F., Leer, R. J., Jore, J., and Smit, E. (2001). Lactobacilli as vehicles for targeting antigens to mucosal tissues by surface exposition of foreign antigens. Methods Enzymol. 336, 369–389.

USE OF RECOMBINANT LAB IN HUMAN HEALTH

59

Prantera, C., Scribano, M. L., Falasco, G., Andreoli, A., and Luzi, C. (2002). Ineffectiveness of probiotics in preventing recurrence after curative resection for Crohn’s disease: A randomised controlled trial with Lactobacillus GG. Gut 51, 405–409. Ramos, J. L., Andersson, P., Jensen, L. B., Ramos, C., Ronchel, M. C., Diaz, E., Timmis, K. N., and Molin, S. (1995). Suicide microbes on the loose. Biotechnology (N Y) 13, 35–37. Rath, H. C. (2003). The role of endogenous bacterial ﬂora: Bystander or the necessary prerequisite? Eur. J. Gastroenterol. Hepatol. 15, 615–620. Rautava, S., and Isolauri, E. (2002). The development of gut immune responses and gut microbiota: Effects of probiotics in prevention and treatment of allergic disease. Curr. Issues Intest. Microbiol. 3, 15–22. Ravn, P., Arnau, J., Madsen, S. M., Vrang, A., and Israelsen, H. (2000). The development of TnNuc and its use for the isolation of novel secretion signals in Lactococcus lactis. Gene 242, 347–356. Ravn, P., Arnau, J., Madsen, S. M., Vrang, A., and Israelsen, H. (2003). Optimization of signal peptide SP310 for heterologous protein production in Lactococcus lactis. Microbiology 149, 2193–2201. Reid, G. (2002). The potential role of probiotics in pediatric urology. J. Urol. 168, 1512–1517. Reid, G., and Bruce, A. W. (2001). Selection of lactobacillus strains for urogenital probiotic applications. J. Infect. Dis. 183, S77–S80. Reid, G., and Burton, J. (2002). Use of Lactobacillus to prevent infection by pathogenic bacteria. Microbes Infect. 4, 319–324. Reid, G., Chan, R. C., Bruce, A. W., and Costerton, J. W. (1985). Prevention of urinary tract infection in rats with an indigenous Lactobacillus casei strain. Infect. Immun. 49, 320–324. Reid, G., Charbonneau, D., Erb, J., Kochanowski, B., Beuerman, D., Poehner, R., and Bruce, A. W. (2003). Oral use of Lactobacillus rhamnosus GR-1 and L. fermentum RC-14 signiﬁcantly alters vaginal ﬂora: Randomized, placebo-controlled trial in 64 healthy women. FEMS Immunol. Med. Microbiol. 35, 131–134. Rembacken, B. J., Snelling, A. M., Hawkey, P. M., Chalmers, D. M., and Axon, A. T. (1999). Non-pathogenic Escherichia coli versus mesalazine for the treatment of ulcerative colitis: A randomised trial. Lancet 354, 635–639. Renault, P. (2002). Genetically modiﬁed lactic acid bacteria: Applications to food or health and risk assessment. Biochimie 84, 1073–1087. Repa, A., Grangette, C., Daniel, C., Hochreiter, R., Hoffmann-Sommergruber, K., Thalhamer, J., Kraft, D., Breiteneder, H., Mercenier, A., and Wiedermann, U. (2003). Mucosal co-application of lactic acid bacteria and allergen induces counter-regulatory immune responses in a murine model of birch pollen allergy. Vaccine 22, 87–95. Resta-Lenert, S., and Barrett, K. E. (2003). Live probiotics protect intestinal epithelial cells from the effects of infection with enteroinvasive Escherichia coli (EIEC). Gut 52, 988–997. Reuter, M. A., Hanniffy, S., and Wells, J. M. (2003). Expression and delivery of heterologous antigens using lactic acid bacteria. Methods Mol. Med. 87, 101–114. Reveneau, N., Geoffroy, M. C., Locht, C., Chagnaud, P., and Mercenier, A. (2002). Comparison of the immune responses induced by local immunizations with recombinant Lactobacillus plantarum producing tetanus toxin fragment C in different cellular locations. Vaccine 20, 1769–1777. Ribeiro, L. A., Azevedo, V., Le Loir, Y., Oliveira, S. C., Dieye, Y., Piard, J. C., Gruss, A., and Langella, P. (2002). Production and targeting of the Brucella abortus antigen L7/

60

HANNIFFY et al.

L12 in Lactococcus lactis: A ﬁrst step towards food-grade live vaccines against brucellosis. Appl. Environ. Microbiol. 68, 910–916. Ring, J. (1997). Allergy and modern society: Does ‘Western life style’ promote the development of allergies? Int. Arch. Allergy. Immunol. 113, 7–10. Rio, M. E., Zago Beatriz, L., Garcia, H., and Winter, L. (2002). [The nutritional status change the effectiveness of a dietary supplement of lactic bacteria on the emerging of respiratory tract diseases in children]. Arch. Latinoam. Nutr. 52, 29–34. Roberts, M., Chatﬁeld, S. N., and Dougan, G. (1994). Salmonella as carriers of heterologous antigens. In ‘‘Novel Delivery Systems for Oral Vaccines’’ (D.T. O’Hagan, ed.), pp. 27–58. Boca Raton, Florida: CRC Press. Robinson, K., Chamberlain, L. M., Schoﬁeld, K. M., Wells, J. M., and Le Page, R. W. (1997). Oral vaccination of mice against tetanus with recombinant Lactococcus lactis. Nat. Biotechnol. 15, 653–657. Romagnani, S. (1994). Regulation of the development of type 2 T-helper cells in allergy. Curr. Opin. Immunol. 6, 838–846. Romero, D. A., and Klaenhammer, T. R. (1991). Construction of an IS946-based composite transposon in Lactococcus lactis subsp. lactis. J. Bacteriol. 173, 7599–7606. Ronchel, M. C., and Ramos, J. L. (2001). Dual system to reinforce biological containment of recombinant bacteria designed for rhizoremediation. Appl. Environ. Microbiol. 67, 2649–2656. Rosenfeldt, V., Michaelsen, K. F., Jakobsen, M., Larsen, C. N., Moller, P. L., Pedersen, P., Tvede, M., Weyrehter, H., Valerius, N. H., and Paerregaard, A. (2002a). Effect of probiotic Lactobacillus strains in young children hospitalized with acute diarrhea. Pediatr. Infect. Dis. J. 21, 411–416. Rosenfeldt, V., Michaelsen, K. F., Jakobsen, M., Larsen, C. N., Moller, P. L., Tvede, M., Weyrehter, H., Valerius, N. H., and Paerregaard, A. (2002b). Effect of probiotic Lactobacillus strains on acute diarrhea in a cohort of nonhospitalized children attending day-care centers. Pediatr. Infect. Dis. J. 21, 417–419. Ross, P., O’Gara, F., and Condon, S. (1990). Thymidylate synthase gene from Lactococcus lactis as a genetic marker: An alternative to antibiotic resistance genes. Appl. Environ. Microbiol. 56, 2164–2169. Ruseler-van Embden, J. G., van Lieshout, L. M., Gosselink, M. J., and Marteau, P. (1995). Inability of Lactobacillus casei strain GG, L. acidophilus, and Biﬁdobacterium biﬁdum to degrade intestinal mucus glycoproteins. Scand. J. Gastroenterol. 30, 675–680. Rush, C., Hafner, L., and Timms, P. (1997). Protein A as a fusion partner for the expression of heterologous proteins in Lactobacillus. Appl. Microbiol. Biotechnol. 47, 537–542. Russell, M. W. (2002). Immunization for protection of the reproductive tract: A review. Am. J. Reprod. Immunol. 47, 265–268. Salama, N., Guillemin, K., McDaniel, T. K., Sherlock, G., Tompkins, L., and Falkow, S. (2000). A whole-genome microarray reveals genetic diversity among Helicobacter pylori strains. Proc. Natl. Acad. Sci. USA 97, 14668–14673. Sanders, J. W., Venema, G., and Kok, J. (1997). A chloride-inducible gene expression cassette and its use in induced lysis of Lactococcus lactis. Appl. Environ. Microbiol. 63, 4877–4882. Scheiner, O., and Kraft, D. (1995). Basic and practical aspects of recombinant allergens. Allergy 50, 384–391. Scheppler, L., Vogel, M., Zuercher, A. W., Zuercher, M., Germond, J. E., Miescher, S. M., and Stadler, B. M. (2002). Recombinant Lactobacillus johnsonii as a mucosal vaccine delivery vehicle. Vaccine 20, 2913–2920.

USE OF RECOMBINANT LAB IN HUMAN HEALTH

61

Schoolnik, G. K. (2002). Functional and comparative genomics of pathogenic bacteria. Curr. Opin. Microbiol. 5, 20–26. Schultz, M., Veltkamp, C., Dieleman, L. A., Grenther, W. B., Wyrick, P. B., Tonkonogy, S. L., and Sartor, R. B. (2002). Lactobacillus plantarum 299V in the treatment and prevention of spontaneous colitis in interleukin-10-deﬁcient mice. Inﬂamm. Bowel Dis. 8, 71–80. Seegers, J. F. (2002). Lactobacilli as live vaccine delivery vectors: Progress and prospects. Trends Biotechnol. 20, 508–515. Seepersaud, R., Hanniffy, S. B. P., Sizer, P., LePage, R. W. F., and Wells, J. M. (2004). Characterisation of a novel leucine-rich repeat protein antigen from Streptococcus agalactiae that elicits protective immunity. Submitted for publication. Sepp, E., Julge, K., Vasar, M., Naaber, P., Bjorksten, B., and Mikelsaar, M. (1997). Intestinal microﬂora of Estonian and Swedish infants. Acta Paediatr. 86, 956–961. Shah, N. P. (2000). Probiotic bacteria: Selective enumeration and survival in dairy foods. J. Dairy Sci. 83, 894–907. Shanahan, F. (2001). Probiotics in inﬂamatory bowel disease. Gut 48, 609. Shiba, T., Aiba, Y., Ishikawa, H., Ushiyama, A., Takagi, A., Mine, T., and Koga, Y. (2003). The suppressive effect of biﬁdobacteria on Bacteroides vulgatus, a putative pathogenic microbe in inﬂammatory bowel disease. Microbiol. Immunol. 47, 371–378. Shida, K., Takahashi, R., Iwadate, E., Takamizawa, K., Yasui, H., Sato, T., Habu, S., Hachimura, S., and Kaminogawa, S. (2002). Lactobacillus casei strain Shirota suppresses serum immunoglobulin E and immunoglobulin G1 responses and systemic anaphylaxis in a food allergy model. Clin. Exp. Allergy 32, 563–570. Shu, Q., and Gill, H. S. (2001). A dietary probiotic (Biﬁdobacterium lactis HN019) reduces the severity of Escherichia coli O157:H7 infection in mice. Med. Microbiol. Immunol. (Berl) 189, 147–152. Shu, Q., Lin, H., Rutherfurd, K. J., Fenwick, S. G., Prasad, J., Gopal, P. K., and Gill, H. S. (2000). Dietary Biﬁdobacterium lactis (HN019) enhances resistance to oral Salmonella typhimurium infection in mice. Microbiol. Immunol. 44, 213–222. Simhon, A., Douglas, J. R., Drasar, B. S., and Soothill, J. F. (1982). Effect of feeding on infants’ faecal ﬂora. Arch. Dis. Child. 57, 54–58. Sirard, J. C., Niedergang, F., and Kraehenbuhl, J. P. (1999). Live attenuated Salmonella: A paradigm of mucosal vaccines. Immunol. Rev. 171, 5–26. Sorensen, K. I., Larsen, R., Kibenich, A., Junge, M. P., and Johansen, E. (2000). A foodgrade cloning system for industrial strains of Lactococcus lactis. Appl. Environ. Microbiol. 66, 1253–1258. Stahl, S., Samuelson, P., Hansson, M., Andreoni, C., Goetsch, L., Libon, C., Liljeqvist, S., Gunneriusson, E., Binz, H., Nguyen, T. N., and Uhlen, M. (1997). Development of non-pathogenic Staphylococci as vaccine delivery vehicles. In ‘‘Gram-Positive Bacteria as Vaccine Vehicles for Mucosal Immunisation’’ (G. Pozzi and J. M. Wells, eds.), pp. 61–81. Springer-Verlag, Berlin. Steidler, L., Hans, W., Schotte, L., Neirynck, S., Obermeier, F., Falk, W., Fiers, W., and Remaut, E. (2000). Treatment of murine colitis by Lactococcus lactis secreting interleukin-10. Science 289, 1352–1355. Steidler, L., Neirynck, S., Huyghebaert, N., Snoeck, V., Vermeire, A., Goddeeris, B., Cox, E., Remon, J. P., and Remaut, E. (2003). Biological containment of genetically modiﬁed Lactococcus lactis for intestinal delivery of human interleukin 10. Nat. Biotechnol. 21, 785–789. Steidler, L., Robinson, K., Chamberlain, L., Schoﬁeld, K. M., Remaut, E., Le Page, R. W., and Wells, J. M. (1998). Mucosal delivery of murine interleukin-2 (IL-2) and IL-6 by

62

HANNIFFY et al.

recombinant strains of Lactococcus lactis coexpressing antigen and cytokine. Infect. Immun. 66, 3183–3189. Strachan, D. P. (1989). Hay fever, hygiene, and household size. BMJ 299, 1259–1260. Strober, W., Fuss, I. J., and Blumberg, R. S. (2002). The immunology of mucosal models of inﬂammation. Annu. Rev. Immunol. 20, 495–549. Strus, M., Pakosz, K., Gosciniak, H., Przondo-Mordarska, A., Rozynek, E., Pituch, H., Meisel-Mikolajczyk, F., and Heczko, P. B. (2001). [Antagonistic activity of Lactobacillus bacteria strains against anaerobic gastrointestinal tract pathogens (Helicobacter pylori, Campylobacter coli, Campylobacter jejuni, Clostridium difﬁcile)]. Med. Dosw. Mikrobiol. 53, 133–142. Szafranski, P., Mello, C. M., Sano, T., Smith, C. L., Kaplan, D. L., and Cantor, C. R. (1997). A new approach for containment of microorganisms: Dual control of streptavidin expression by antisense RNA and the T7 transcription system. Proc. Natl. Acad. Sci. USA 94, 1059–1063. Szajewska, H., and Mrukowicz, J. Z. (2001). Probiotics in the treatment and prevention of acute infectious diarrhea in infants and children: A systematic review of published randomized, double-blind, placebo-controlled trials. J. Pediatr. Gastroenterol. Nutr. 33(Suppl. 2), S17–S25. Taha, T. E., Hoover, D. R., Dallabetta, G. A., Kumwenda, N. I., Mtimavalye, L. A., Yang, L. P., Liomba, G. N., Broadhead, R. L., Chiphangwi, J. D., and Miotti, P. G. (1998). Bacterial vaginosis and disturbances of vaginal ﬂora: Association with increased acquisition of HIV. AIDS 12, 1699–1706. Takala, T. M., and Saris, P. E. (2002). A food-grade cloning vector for lactic acid bacteria based on the nisin immunity gene nisI. Appl. Microbiol. Biotechnol. 59, 467–471. Takala, T. M., Saris, P. E., and Tynkkynen, S. S. (2003). Food-grade host/vector expression system for Lactobacillus casei based on complementation of plasmid-associated phospho-beta-galactosidase gene lacG. Appl. Microbiol. Biotechnol. 60, 564–570. Tejada-Simon, M. V., Lee, J. H., Ustunol, Z., and Pestka, J. J. (1999). Ingestion of yogurt containing Lactobacillus acidophilus and Biﬁdobacterium to potentiate immunoglobulin A responses to cholera toxin in mice. J. Dairy Sci. 82, 649–660. Thole, J. E., van Dalen, P. J., Havenith, C. E., Pouwels, P. H., Seegers, J. F., Tielen, F. D., van der Zee, M. D., Zegers, N. D., and Shaw, M. (2000). Live bacterial delivery systems for development of mucosal vaccines. Curr. Opin. Mol. Ther. 2, 94–99. Threlfall, E. J. (2002). Antimicrobial drug resistance in Salmonella: Problems and perspectives in food- and water-borne infections. FEMS Microbiol. Rev. 26, 141–148. Tlaskalova-Hogenova, H., Tuckova, L., Lodinova-Zadnikova, R., Stepankova, R., Cukrowska, B., Funda, D. P., Striz, I., Kozakova, H., Trebichavsky, I., Sokol, D., Rehakova, Z., Sinkora, J., Fundova, P., Horakova, D., Jelinkova, L., and Sanchez, D. (2002). Mucosal immunity: Its role in defense and allergy. Int. Arch. Allergy Immunol. 128, 77–89. van de Guchte, M., Serror, P., Chervaux, C., Smokvina, T., Ehrlich, S. D., and Maguin, E. (2002). Stress responses in lactic acid bacteria. Antonie Van Leeuwenhoek 82, 187–216. van Deventer, S. J., Elson, C. O., and Fedorak, R. N. (1997). Multiple doses of intravenous interleukin 10 in steroid-refractory Crohn’s disease. Crohn’s Disease Study Group. Gastroenterology 113, 383–389. van Ginkel, F. W., Nguyen, H. H., and McGhee, J. R. (2000). Vaccines for mucosal immunity to combat emerging infectious diseases. Emerg. Infect. Dis. 6, 123–132. Venturi, A., Gionchetti, P., Rizzello, F., Johansson, R., Zucconi, E., Brigidi, P., Matteuzzi, D., and Campieri, M. (1999). Impact on the composition of the faecal ﬂora by a new

USE OF RECOMBINANT LAB IN HUMAN HEALTH

63

probiotic preparation: Preliminary data on maintenance treatment of patients with ulcerative colitis. Aliment. Pharmacol. Ther. 13, 1103–1108. Vesa, T., Pochart, P., and Marteau, P. (2000). Pharmacokinetics of Lactobacillus plantarum NCIMB 8826, Lactobacillus fermentum KLD, and Lactococcus lactis MG 1363 in the human gastrointestinal tract. Aliment. Pharmacol. Ther. 14, 823–828. Villa-Komaroff, L., Efstratiadis, A., Broome, S., Lomedico, P., Tizard, R., Naber, S. P., Chick, W. L., and Gilbert, W. (1978). A bacterial clone synthesizing proinsulin. Proc. Natl. Acad. Sci. USA 75, 3727–3731. Vinderola, C. G., Medici, M., and Perdigon, G. (2004). Relationship between interaction sites in the gut, hydrophobicity, mucosal immunomodulating capacities and cell wall protein proﬁles in indigenous and exogenous bacteria. J. Appl. Microbiol. 96, 230–243. Vitini, E., Alvarez, S., Medina, M., Medici, M., de Budeguer, M. V., and Perdigon, G. (2000). Gut mucosal immunostimulation by lactic acid bacteria. Biocell 24, 223–232. von der Weid, T., Bulliard, C., and Schiffrin, E. J. (2001). Induction by a lactic acid bacterium of a population of CD4(þ) T cells with low proliferative capacity that produce transforming growth factor beta and interleukin-10. Clin. Diagn. Lab. Immunol. 8, 695–701. von Mutius, E., Martinez, F. D., Fritzsch, C., Nicolai, T., Roell, G., and Thiemann, H. H. (1994). Prevalence of asthma and atopy in two areas of West and East Germany. Am. J. Respir. Crit. Care. Med. 149, 358–364. von Wright, A., and Bruce, A. W. (2003). Genetically modiﬁed microorganisms and their potential effects on human health and nutrition. Trends Food Sci. Technol. 14, 264–276. von Wright, A., Wessels, S., Tynkkynen, S., and Saarela, M. (1990). Isolation of a replication region of a large lactococcal plasmid and use in cloning of a nisin resistance determinant. Appl. Environ. Microbiol. 56, 2029–2035. Walsh, C. (2000). Molecular mechanisms that confer antibacterial drug resistance. Nature 406, 775–781. Walter, J., Heng, N. C., Hammes, W. P., Loach, D. M., Tannock, G. W., and Hertel, C. (2003). Identiﬁcation of Lactobacillus reuteri genes speciﬁcally induced in the mouse gastrointestinal tract. Appl. Environ. Microbiol. 69, 2044–2051. Walter, J., Hertel, C., Tannock, G. W., Lis, C. M., Munro, K., and Hammes, W. P. (2001). Detection of Lactobacillus, Pediococcus, Leuconostoc, and Weissella species in human feces by using group-speciﬁc PCR primers and denaturing gradient gel electrophoresis. Appl. Environ. Microbiol. 67, 2578–2585. Walter, J., Tannock, G. W., Tilsala-Timisjarvi, A., Rodtong, S., Loach, D. M., Munro, K., and Alatossava, T. (2000). Detection and identiﬁcation of gastrointestinal Lactobacillus species by using denaturing gradient gel electrophoresis and species-speciﬁc PCR primers. Appl. Environ. Microbiol. 66, 297–303. Wells, J. M., and Bennik, M. H. J. (2003). Impact of genomics on foodborne pathogens. Nutri. Res. Rev. 16, 21–35. Wells, J. M., and Mercenier, A. (2003). Lactic acid bacteria as mucosal delivery vehicles. In ‘‘The Lactic Acid Bacteria,’’ Vol III, pp. 260–291. Kluwer Academic/Plenum Publishers, New York. Wells, J. M., and Pozzi, G. (1997). An overview of gram-positive bacteria as vaccine vehicles for musocal immunization. In ‘‘Gram-Positive Bacteria as Vaccine Vehicles for Mucosal Immunisation’’ (G. Pozzi and J. M. Wells, eds.), pp. 1–8. Springer-Verlag, Berlin.

64

HANNIFFY et al.

Wells, J. M., Robinson, K., Chamberlain, L. M., Schoﬁeld, K. M., and Le Page, R. W. (1996). Lactic acid bacteria as vaccine delivery vehicles. Antonie Van Leeuwenhoek 70, 317–330. Wells, J. M., and Schoﬁeld, K. M. (1996). Cloning and expression systems for lactococci. Nato Asi. Ser. H. 98, 37–63. Wells, J. M., Wilson, P. W., Norton, P. M., Gasson, M. J., and Le Page, R. W. (1993a). Lactococcus lactis: High-level expression of tetanus toxin fragment C and protection against lethal challenge. Mol. Microbiol. 8, 1155–1162. Wells, J. M., Wilson, P. W., Norton, P. M., and Le Page, R. W. (1993b). A model system for the investigation of heterologous protein secretion pathways in Lactococcus lactis. Appl. Environ. Microbiol. 59, 3954–3959. Wheeler, A. W., and Woroniecki, S. R. (2001). Immunological adjuvants in allergy vaccines: Past, present and future. Allergology International 50, 295–301. Wiedermann, U. (2003). Mucosal immunity–mucosal tolerance. A strategy for treatment of allergic diseases. Chem. Immunol. Allergy 82, 11–24. Wiedermann, U., Jahn-Schmid, B., Fritsch, R., Bauer, L., Renz, H., Kraft, D., and Ebner, C. (1998). Effects of adjuvants on the immune response to allergens in a murine model of allergen inhalation: Cholera toxin induces a Th1-like response to Bet v 1, the major birch pollen allergen. Clin. Exp. Immunol. 111, 144–151. Wiedermann, U., Jahn-Schmid, B., Lindblad, M., Rask, C., Holmgren, J., Kraft, D., and Ebner, C. (1999). Suppressive versus stimulatory effects of allergen/cholera toxoid (CTB) conjugates depending on the nature of the allergen in a murine model of type I allergy. Int. Immunol. 11, 1131–1138. Wiesenfeld, H. C., Hillier, S. L., Krohn, M. A., Landers, D. V., and Sweet, R. L. (2003). Bacterial vaginosis is a strong predictor of Neisseria gonorrhoeae and Chlamydia trachomatis infection. Clin. Infect. Dis. 36, 663–668. Wu, H. Y., Abdu, S., Stinson, D., Russell, M. W., Moldoveanu, Z., Huang, W. Q., Compans, R. W., and Mestecky, J. (2000). Generation of female genital tract antibody responses by local or central (common) mucosal immunization: Compartmentalization within the common mucosal immune system. Infect. Immun. 68, 5539–5545. Xin, K. Q., Hoshino, Y., Toda, Y., Igimi, S., Kojima, Y., Jounai, N., Ohba, K., Kushiro, A., Kiwaki, M., Hamajima, K., Klinman, D., and Okuda, K. (2003). Immunogenicity and protective efﬁcacy of orally administered recombinant Lactococcus lactis expressing surface-bound HIV Env. Blood 102, 223–228. Zegers, N. D., Kluter, E., van Der Stap, H., van Dura, E., van Dalen, P., Shaw, M., and Baillie, L. (1999). Expression of the protective antigen of Bacillus anthracis by Lactobacillus casei: Towards the development of an oral vaccine against anthrax. J. Appl. Microbiol. 87, 309–314. Zoetendal, E. G., Akkermans, A. D., and De Vos, W. M. (1998). Temperature gradient gel electrophoresis analysis of 16S rRNA from human fecal samples reveals stable and host-speciﬁc communities of active bacteria. Appl. Environ. Microbiol. 64, 3854–3859. Zoetendal, E. G., Akkermans, A. D. L., Akkermans-van Vliet, W. M., de Visser, J. A. G. M., and de Vos, W. M. (2001). The host genotype affects the bacterial community in the human gastrointestinal tract. Microbial Ecol. Health Dis. 13, 129–134. Zoetendal, E. G., von Wright, A., Vilpponen-Salmela, T., Ben-Amor, K., Akkermans, A. D., and de Vos, W. M. (2002). Mucosa-associated bacteria in the human gastrointestinal tract are uniformly distributed along the colon and differ from the community recovered from feces. Appl. Environ. Microbiol. 68, 3401–3407.

Novel Aspects of Signaling in Streptomyces Development GILLES P.

VAN

WEZEL*

AND

ERIK VIJGENBOOM

Department of Biochemistry, Leiden Institute of Chemistry 2300RA Leiden, The Netherlands *Author for correspondence. E-mail: [email protected]

I. Introduction II. Aspects of Vegetative Growth and Liquid Cultivation of Streptomycetes III. The Switch to Development A. General Considerations B. Submerged Development C. Towards an Aerial Mycelium D. Influence of Carbon Sources on Developmental Signaling IV. Novel Genes in Development A. Discovery of New Developmental Genes B. The ram and amf Gene Clusters C. Novel Regulators of Sporulation: The SsgA-Like Proteins (SALPs) V. Concluding Remarks References

65 67 69 69 70 70 71 72 72 73 78 83 83

I. Introduction Actinomycetes have an unusually complex life cycle, many aspects of which are globally similar to those observed in some lower eukaryotes, which makes them particularly interesting for the study of bacterial development and evolution (Chater and Losick, 1997). Their ability to produce a large array of biologically active natural products, including the majority of antibiotics, as well as many agents with other medical and agricultural merits, makes these organisms also highly relevant from an industrial perspective. One of the best-characterized genera among the actinomycetes is Streptomyces, the subject of this review, with Streptomyces coelicolor, Streptomyces griseus, and Streptomyces lividans being the most well-studied species. Pioneering work by David Hopwood 40 to 50 years ago established Streptomyces coelicolor as the model system for the genus (Chater, 1999; Hopwood, 1999). The latter organism has become the paradigm for the study of Streptomyces development (Chater, 1998) and antibiotic production (Bibb, 1996), which was helped by a wealth of mutants and the development of a large genetic toolbox (Kieser et al., 2000). Recently the genome sequences of S. coelicolor and S. avermitilis were completed, taking Streptomyces research into the genomics era (Bentley et al., 65 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

66

VAN WEZEL AND VIJGENBOOM

2002; Ikeda et al., 2003). The closely related species S. lividans is especially important as an excellent expression host for industrial enzymes. Much more distantly related is the streptomycin producer S. griseus, which is of particular genetic interest for two main reasons: the profound and well-studied effect of a signal molecule, the hormonelike A-factor (2-isocapryloyl-3R-hydroxymethyl--butyrolactone) on its development and antibiotic production (ﬁrst described in Khoklov et al., 1967; reviewed in Horinouchi, 2002), and its ability to sporulate in submerged cultures (Kendrick and Ensign, 1983; reviewed in Fla¨rdh and van Wezel, 2003). On solid media, a germinating spore will produce one or more hyphae, which will grow and branch to form a vegetative mycelium. Exponential growth is achieved by a combination of tip growth and branching, resulting in a complex mycelial network. At this stage, the vegetative hyphae consist of multi-nucleoid syncytial cells separated by occasional cross-walls (Wildermuth, 1970). Then, as colonies develop, an aerial mycelium is produced, with hydrophobic hyphae breaking through the moist surface, erected into the air. This is the start of the reproductive phase, initiated in response to nutrient depletion and the resulting requirement of mobilization. The substrates required for the production of the aerial hyphae are derived from reuse of material such as nucleic acids, proteins, and storage compounds from the vegetative mycelium. Eventually, sporulation-programmed hyphae are formed, producing chains of mono-nucleoid spores, which are released after a poorly understood maturation process. A typical example of colonies of streptomycetes growing on an agar plate is shown in Fig. 1A (see color insert) and a close-up of sporulating aerial hyphae in Fig. 1B. Mutants that fail to develop an aerial mycelium are called bld (bald, reﬂecting the ‘‘hairless’’ phenotype), and mutants that fail to produce mature grey-pigmented spores are called whi (white, referring to the production of a nonpigmented fuzzy aerial mycelium). Several reviews have been written on the involvement of bld and whi genes in the control of Streptomyces development and aerial hyphae formation (e.g., Chater, 1998, 2001; Kelemen and Buttner, 1998). In this review we focus on recently discovered genes that play an important role in the two main switches in Streptomyces development, namely the ram gene cluster (for transition from vegetative to aerial growth) and the ssgA-like genes (control of the sporulation process). We also discuss how these and other developmental genes allow streptomycetes to respond to changes in the nutritional state of the microenvironment. Understanding these mechanisms is important for fundamental and applied Streptomyces research.

SIGNALING IN STREPTOMYCES DEVELOPMENT

67

FIG. 1. (A) Sporulating colonies of Streptomyces ramocissimus. Clearly visible are aerial hyphae (white outer circle) and spores (grey inner circle); the vegetative mycelium lies below the aerial mycelium and is not visible. The brown pigment secreted by the colonies is melanine. (B) Scanning electron micrograph of sporulating aerial hyphae of Streptomyces coelicolor. Photograph courtesy of Dr. H. K. Koerten (Centre for Electron Microscopy, LUMC, Leiden, The Netherlands). Bar ¼ 10 m.

II. Aspects of Vegetative Growth and Liquid Cultivation of Streptomycetes Models for mycelial growth have been worked out for ﬁlamentous fungi, and particularly the penicillin producer Penicillium chrysogenum (Krabben, 1997; Nielsen, 1996; Trinchi, 1971). While at the molecular level the processes are very different in actinomycetes

68

VAN WEZEL AND VIJGENBOOM

(prokaryotes) and ﬁlamentous fungi (lower eukaryotes), it appears that at the microscopic level these organisms exhibit similar growth forms and hyphal and mycelial growth kinetics (reviewed in Prosser and Tough, 1991). Therefore several of the kinetic models for growth of streptomycetes may be derived from the better-studied fungi (Bushell, 1988). Exponential growth of the mycelium is achieved by a combination of linear (polar) growth, branching, and—particularly in submerged culture—hyphal breakage (Locci, 1980). The frequency of branching is not constant but apparently is dictated by the growth conditions; nutrient-rich conditions favor branching, to optimally profit from the available nutrients in the habitat (typically the soil), whereas under nutrient-depleted conditions branching is reduced and growth is dictated by tip extension, which favors the formation of socalled ‘‘searching hyphae’’ (Bushell, 1988). Interestingly, branching and cross-wall formation (which often coincide) markedly reduce hyphal strength (Wardell et al., 2002), a phenomenon supported by the observation that ftsZ mutants of S. coelicolor are viable and produce unbranched, long and stable hyphae in the absence of cross-walls (McCormick et al., 1994). The relationship between mycelial morphology and stability is particularly relevant for biotechnological applications, because it determines clump size and therefore indirectly also the efﬁciency of the production process (Bushell, 1988; Wardell et al., 2002). While development is mostly studied in solid-grown cultures, in the industrial production process large-scale liquid cultures are the reality. Unfortunately, it is difﬁcult to translate morphological principles of one culture type to the other, which is at least in part due to the morphological diversity of liquid-grown mycelium. In batch fermentations, variations as large as three to four orders of magnitude (m to cm scale) occur. Analysis of erythromycin biosynthesis in Saccharopolyspora erythraea showed that production took place only in hyphal fragments with a diameter larger than approximately 90 m (Martin and Bushell, 1996). Mixing problems with larger mycelial clumps also negatively affect the production process, because an oxygen and nutrient gradient exists from the surface of the mycelial clump to its center, affecting growth and production (Bushell, 1988; Huang and Bungay, 1973). Therefore, better understanding of the factors affecting growth and morphology would obviously be of advantage for biotechnological applications. Studies in streptomycetes with ﬂuorescently labeled vancomycin or radiolabelled N-acetyl glucosamine (both of which are incorporated into newly synthesized peptidoglycan) revealed that peptidoglycan

SIGNALING IN STREPTOMYCES DEVELOPMENT

69

biosynthetic activity primarily occurs at hyphal tips and at branching sites (Daniel and Errington, 2003; Gray et al., 1990; Young, 2003). Also, the addition of penicillins results in defects particularly at the apical sites of the hyphae, and less at the lateral walls, which may therefore be regarded as a relatively inert murein polymer. Several of the penicillinbinding proteins and other proteins involved in the synthesis and integrity of the cell wall become recruited to cell wall construction sites. The ﬁrst clear example of a protein associated with apical growth is DivIVA (Fla¨rdh, 2003). Considering that it is involved in driving (the initiation of) linear extension and that its overexpression results in erratic branching, it is conceivable that another important role for DivIVA is to coordinate the initiation of new branching points. The Bacillus subtilis DivIVA homolog plays a direct role in septum-site determination by interacting with the MinCD cell division inhibitor (Edwards and Errington, 1997) and was recently shown to interact with the chromosome segregation machinery to help position the oriC region of the chromosome at the cell pole, in preparation for polar division (Thomaides et al., 2001). Streptomycetes lack a homolog of MinC, and the function of the two MinD homolog is unclear, as minD disruptants have no obvious phenotype (McCormick and van Wezel, unpublished data). The high frequency of co-occurrence of septa and branches (a feature also seen in ﬁlamentous fungi) suggests coordination between cell division and branching, and it is perhaps DivIVA that may play a role in this coordinating process, although there is no evidence that DivIVA directly affects cell division (Fla¨rdh, 2003).

III. The Switch to Development A. GENERAL CONSIDERATIONS As a result of their mycelial lifestyle, streptomycetes are sessile microorganisms, and in contrast to other bacteria, the mycelium itself cannot migrate to a more favorable environment, such as by chemotaxis. When deprived of nutrients, the mycelium responds by favoring apical (linear) growth over branching (see Section II) and by the onset of morphological differentiation, resulting in the production of exospores. Under nutrient-limiting conditions, lysis of the vegetative hyphae probably provides the nutrients necessary for the construction of the aerial mycelium (Mendez et al., 1985). During this part of the life cycle, several control mechanisms come into play, such as carbon catabolite repression and stringent response, which constitute important sensors of the nutritional state of the environment and have a

70

VAN WEZEL AND VIJGENBOOM

repressing or activating effect on the onset of sporulation, respectively. These global regulatory processes and their impact on development are discussed elsewhere (Hodgson, 2000; Ingram et al., 1995; Kamionka et al., 2002; Takano and Bibb, 1994; Ueda et al., 1999). B. SUBMERGED DEVELOPMENT Typically, ‘‘development’’ refers to solid-culture differentiation (vegetative mycelium, aerial mycelium, spores) (Chater, 1972). In liquid culture, several streptomycetes, with Streptomyces griseus as the most well-known example, also form submerged spores, produced by sporogenic hyphae at the extremities of liquid-grown vegetative mycelium (recently reviewed in Fla¨rdh and van Wezel, 2003). This process is generally triggered by nutritional shift down, although some streptomycetes also produce submerged spores in nutrient-rich cultures. Comparison of the ultrastructures of submerged spores with surface spores failed to reveal signiﬁcant structural differences. Perhaps counterintuitively, no differences were found either between the sporogenic hyphae in submerged and solid-grown cultures: both were essentially unbranched and thin-walled (Rueda et al., 2001). The only signiﬁcant difference was in the sheath, which was thinner and less regularly structured in submerged sporulating hyphae. Some regulatory aspects of submerged sporulation are dealt with in the section on ssgA (Section IV.C). C. TOWARDS AN AERIAL MYCELIUM Early developmental (bld) mutants are not only defective in aerial hyphae formation, but also their antibiotic production is strongly affected (either negatively or positively), which links development and secondary metabolism. Most bld mutants fail to produce antibiotics, although some are antibiotic overproducers. By deﬁnition, all nonessential genes that are required for aerial hyphae formation are bld genes, and genes required for any of the developmental stages between the onset of aerial mycelium formation and the production of the spore pigment WhiE are called whi genes (Kelemen et al., 1998). There is evidence that several of the bld gene products are part of a signaling cascade. This was discovered by extracellular complementation experiments, where bld mutants were grown in close proximity to each other, without physical contact (Nodwell et al., 1996; Willey et al., 1991, 1993). A low degree of aerial hyphae formation could be restored by one bld mutant to the other, and typically in a unidirectional

SIGNALING IN STREPTOMYCES DEVELOPMENT

71

manner. A detailed overview of the possible function of the bld genes, and their role within the putative signaling cascade, has been provided elsewhere (Kelemen and Buttner, 1998; Nodwell et al., 1999). The best known of the bld mutants is bldA, resulting from mutations in the gene for a leucyl tRNA, necessary for the translation of UUAcontaining transcripts (Leskiw et al., 1991a,b). Approximately 150 genes of the S. coelicolor genome harbor a TTA codon (and are therefore dependent on BldA), many of which are involved in the regulation of development or antibiotic production (Bentley et al., 2002). Interestingly, the failure to obtain bldA deletion mutants of S. coelicolor M145 suggests that at least one bldA-dependent gene is essential for growth in this model strain. Thus, while bldA is an important control point on the way to aerial development, at least in M145 it is also required for the translation of nondevelopmental mRNAs. Recently it was discovered that the activity of several Bld and Whi proteins is not conﬁned to a speciﬁc developmental stage. Some of the so-called late bld genes—including bldD, bldM, and bldN—are active not only in a stage temporally related to the switch to aerial mycelium formation but are also required much later in the developmental program. For example, bldM and bldN mutants were originally classiﬁed as whiK and whiN, respectively, as several point mutants obtained from classical screens were blocked in later stages of aerial development. Gene disruption and expression studies later showed that a low level of WhiK and WhiN activity was required for the onset of aerial mycelium formation, after which they were renamed bldM and bldN, respectively (Bibb and Buttner, 2003; Bibb et al., 2000). D. INFLUENCE OF CARBON SOURCES ON DEVELOPMENTAL SIGNALING The relationship between the nutritional state of the environment and Streptomyces development is underlined by the medium-dependence of several of the bld mutants, which sporulate on minimal medium agar plates with mannitol but not with glucose (Merrick, 1976; Pope et al., 1996), suggesting a role for glucose repression in Streptomyces, mediated through glucose kinase (Angell et al., 1992). Indeed, glkA mutant derivatives of S. coelicolor bldA mutants do sporulate in the presence of glucose (van Wezel, unpublished data). One bld gene that is of particular interest for the link between carbon source-dependent gene regulation and development is bldB. BldB null mutants have a bald phenotype on all carbon sources and fail to produce aerial hyphae or antibiotics under any condition (Merrick, 1976). Furthermore, bldB mutants are defective in catabolite control

72

VAN WEZEL AND VIJGENBOOM

and do not fall in the hierarchy of extracellular complementation exhibited by other bld mutants (Nodwell et al., 1999; Pope et al., 1996; Willey et al., 1993). Therefore BldB possibly constitutes a key control point in the switch to development. The bldB gene encodes an 11 kDa DNA binding protein that likely functions as a homodimer (Eccleston et al., 2002; Pope et al., 1998). BldB belongs to the family of AbaA-like proteins, with six paralogs on the S. coelicolor genome and—similarly to the SsgA-like proteins (Section IV.C)—no clear homologs outside the actinomycetes. Other connections with glucose metabolism have been reported. Ectopic sporulation was observed in an S. coelicolor mutant with a 7.4 kb deletion around the glkA gene (Kelemen et al., 1995), while an S. griseus das mutant produced ectopic spores at regular intervals in the vegetative hyphae, but only on glucose-containing media, providing another example of a carbon source-dependent (conditional) requirement for a developmental gene (Seo et al., 2002). Recently attention has been directed toward the relationship between high-energy tricarboxylic acid (TCA)-cycle intermediates and development. Interruption of aerobic TCA cycle-based metabolism through mutations in citrate synthase or aconitase resulted in irreversible acidiﬁcation of the medium during growth on glucose, with obvious defects in morphological differentiation and antibiotic biosynthesis. These effects could at least in part be compensated by buffering of the medium (Viollier et al., 2001a,b). This indicates that the outcome of extracellular complementation experiments, such as for the bld mutants, should be carefully evaluated, as the signal passed on from one Streptomyces strain to another could be the result of an exported (protein) factor, but also of changes in pH or medium composition.

IV. Novel Genes in Development A. DISCOVERY OF NEW DEVELOPMENTAL GENES The quest for novel developmental genes required different strategies and deployment of new experimental approaches. Originally, the discovery of developmental mutants by selecting/screening typically resulted from random mutagenesis experiments with ultraviolet (UV)-irradiated or chemically treated cells. A large collection of bld and whi mutants was obtained, which were classiﬁed on the basis of morphological characteristics, helped by scanning electron microscopy (Chater, 1972; Hopwood, 1999; Ryding et al., 1999) or transposon mutagenesis (Gehring et al., 2000). Although these studies were extensive, not all the currently known developmental mutants were thus identiﬁed.

SIGNALING IN STREPTOMYCES DEVELOPMENT

73

The complete elucidation of the S. coelicolor genome (Bentley et al., 2002), and the concurrent advent of the genomics era to Streptomyces research, allowed a directed search for possible or likely developmental genes on the basis of a presumed homology to developmental genes from Bacillus, or on the basis of development-dependent expression proﬁles (Donadio et al., 2002; Hesketh and Chater, 2003; Huang et al., 2001). However, several important new classes of developmental genes were recently identiﬁed on the basis of physiological criteria such as the acceleration of aerial mycelium formation in S. lividans (ram genes, for rapid aerial mycelium; Ma and Kendall, 1994), restoration of development in the presence of glucose (Seo et al., 2002), or complementation of disturbed submerged sporulation of S. griseus mutants (ssgA-like genes, for sporulation of Streptomyces griseus; Kawamoto and Ensign, 1995b). In the following sections we review the complex data generated by several laboratories on two of these examples, namely the ram/amf gene clusters and the ssgA-like genes. B. THE

RAM AND AMF

GENE CLUSTERS

1. Genetic Organization of the Clusters The ram (in S. coelicolor/S. lividans) and amf (in S. griseus) gene clusters are important for the transition from vegetative to aerial growth, as well as for early stages of aerial growth. Surprisingly, different screens using complementation of certain phenotypes by genomic libraries all resulted in the identiﬁcation of the same gene clusters: accelerated aerial hyphae development in wild-type S. lividans (Ma and Kendall, 1994); complementation of aerial hyphae development in the A-factor deﬁcient S. griseus strain HH1 (Kudo et al., 1995; Ueda et al., 1998); relief of the dependence on increased copper ion levels for development in S. lividans (Keijser et al., 2000); and complementation of the bldJ mutant (Nguyen et al., 2002). The overall similarity between the ram and amf genes is not very high, but the gene organization is strictly conserved. Both clusters consist of ﬁve genes (Fig. 2): ramC/amfT for a serine/threonine (ser/thr) kinase (Hudson et al., 2002), ramS/amfS encoding a small peptide, ramAB/amfAB encoding an integral membrane ABC transporter, and a ﬁfth oppositely oriented gene ramR/amfR, encoding a transcription factor of the two-component regulator family. 2. Transcriptional Control by RamR The ramR gene is expressed from a single promoter that is already active in vegetative mycelium and displays its highest activity after approximately 48 hours of growth on solid media, corresponding with

74

VAN WEZEL AND VIJGENBOOM

FIG. 2. Model for the transcriptional control of the ram cluster by the response regulator RamR, and the function of RamCSAB in the signal transduction. On the basis of sequence analysis, the ABC transporter complex RamAB and the serine/threonine kinase RamC are integrally and C-terminally inserted in the membrane, respectively. In this model, there are two possibilities: (1) RamS is phosphorylated by membrane-bound RamC and subsequently exported by RamAB, and the external RamS then provides a developmental signal (‘‘Signal 2’’) to receiver cells; or (2) it is processed and modiﬁed to become SapB, a possibility that is strongly supported by the studies of Kodani et al. (2004). In the latter case it is not yet clear whether all processing and modiﬁcation functions are provided by RamC or whether other proteins are involved. A model for the primary structure of SapB is presented in the top right corner (based on Kodani et al., 2004). The RamS residues that are modiﬁed in SapB are indicated between brackets below the SapB sequence. The developmental signal that triggers RamR expression (‘‘Signal 1’’), and the putative target for RamS signaling, are unknown. Dha, didehydroalanine.

the initiation of aerial hyphae production (Keijser et al., 2002). The same authors showed that ramR transcripts are present throughout aerial growth and even during sporulation, although promoter probing assays with xy/E as a reporter of transcriptional activity suggested a sudden drop in ramR promoter activity upon entry of the aerial growth phase (Nguyen et al., 2002). In all screens, enhanced expression of RamR/AmfR accelerated development, without the need of the other ram/amf genes (Keijser et al., 2000;

SIGNALING IN STREPTOMYCES DEVELOPMENT

75

Nguyen et al., 2002; Ueda et al., 1993). The important function of RamR was conﬁrmed by mutational experiments, which showed that S. coelicolor ramR disruption mutants were severely delayed in development, with sparse and signiﬁcantly delayed aerial mycelium formation. S. lividans ramR mutants were even more delayed in aerial hyphae formation than those of S. coelicolor, in line with the idea that the ram cluster is more important for the development of S. lividans than for the development of S. coelicolor (Keijser et al., 2000). Similarly to most other bld mutants, ram mutants have a conditionally nonsporulating phenotype, with normal development on solid media containing mannitol, and a bld phenotype in the presence of glucose. This once more stresses the important role of carbon catabolite control in the decision to switch to development. Transcriptional analysis showed that expression of ramC is dependent on RamR. The target of RamR was identiﬁed in studies by O’Connor et al. (2002) and Nguyen et al. (2002), who showed that RamR has at least one binding site in the ramC promoter region. The ability of recombinant RamR to bind in vitro to the sequence upstream of the ramC promoter demonstrated that phosphorylation of D53, which is essential for the in vivo function of RamR, is not required for its DNA binding activity (Nguyen et al., 2002). Similarly, the corresponding amino acid residue in AmfR, D54, was shown to be critical for restoration by AmfR of sporulation to bld mutant HH1 of S. griseus (Ueda et al., 1993). In S. griseus, expression of amfR is under control of Afactor through the A-factor-dependent activator protein AdpA (Chater and Horinouchi, 2003; Ohnishi et al., 2002; Ueda et al., 1998). The factors controlling ramR expression in S. coelicolor and S. lividans are unknown, but considering the relatively normal phenotype of A-factor null-mutants (Takano et al., 2001), it is unlikely that AdpA also plays an important role in the control of ramR. 3. RamC Is Essential for Development The ramC gene did not show up in the screens as an essential gene for the acceleration of aerial hyphae formation. However, constructed ramC null mutants were bald (O’Connor et al., 2002) or severely delayed in aerial hyphae production (Nguyen et al., 2002), demonstrating the importance of the gene for development. Analysis of the ramC gene product predicted that the protein consists of an N-terminal serine/ threonine protein kinase domain and a C-terminal domain with several membrane spanning regions. In line with this prediction, amino acid residues that are invariant and essential in similar kinases were also essential for RamC function, as shown by complementation

76

VAN WEZEL AND VIJGENBOOM

experiments in which a ramC disruption mutant was transformed with constructs expressing single amino acid substitution mutants of RamC (Hudson et al., 2002). It is attractive to speculate that RamC is associated with the membrane through its C-terminal sequence and functions in close proximity to RamAB (Fig. 2 and Section IV.B.4). Several studies demonstrated that ramC expression occurred at the onset of aerial growth and was dependent on RamR (Keijser et al., 2002; Nguyen et al., 2002; O’Connor et al., 2002). The promoter upstream of ramC also directs transcription of ramS and ramAB, suggesting an operon-like organization. However, the transcription of the ramCSAB gene cluster is complex. The ramAB genes are transcribed much less frequently because of a transcriptional attenuator between ramS and ramA (Keijser et al., 2002). Furthermore, the strong accumulation of ramS transcripts and the low level of ramC transcripts, as detected by Northern hybridization experiments, suggests that after processing of the full-length transcript, the ramC mRNA is rapidly degraded, because there is no promoter in the intergenic region between ramC and ramS (Keijser, 2002). 4. RamS and RamAB What could be the target for the RamC/AmfT kinase activity? The ‘S-peptides’ (RamS/AmfS) are promising candidates for three reasons. The peptides contain several conserved serine and threonine residues as potential phosphorylation sites (Fig. 3). Ueda et al. (2002) showed that the S-peptide has a signaling function and the genes are cotranscribed with the kinase genes. A phosphorylation-dependent function of the S-peptide would ﬁt with a role as signaling molecule. The signal is switched on at a time corresponding to the onset of ramC expression, which in turn is activated by RamR, and switched off again as soon as RamC levels drop. Therefore, the signal presumably does not relate to the half-life of RamS. In a hypothetical model, as depicted in Fig. 2, a cascade of events starting with the induction of ramR

FIG. 3. Alignment of sequences predicted for the three known ‘‘S-peptides.’’ Amino acids shared by at least two proteins are boxed. Residues that form potential target sites for phosphorylation are indicated with asterisks. The accession numbers for the protein sequences are: AAD33774 (RamS S. lividans), BAA33539 (AmfS S. griseus), and BAC75213 (AmfS S. avermitilis). The sequences of S. coelicolor and S. lividans RamS are identical.

SIGNALING IN STREPTOMYCES DEVELOPMENT

77

transcription by an unknown signal ﬁnally results in the transport of the phosphorylated S-peptide across the membrane. Once outside, the S-peptide will signal the onset of development to neighboring hyphae. Support for this model is provided by several experiments. Disruption of ramS in S. coelicolor resulted in a severe delay of its aerial hyphae development (Nguyen et al., 2002), while an amfS disruption mutant of S. griseus had a bld phenotype (Ueda et al., 2002). Extracellular complementation of the amfS mutant was observed when wild-type S. griseus was grown in close proximity. However, neither amfR disruption mutants nor amfAB disruption mutants were capable of extracellular complementation of the amfS mutant. This is supportive of a model in which the extracellular complementing activity is AmfS, which is not produced by the amfR mutant and not exported by the amfAB mutant (Fig. 2) (Ueda et al., 2002). The observation that the amfAB mutant itself shows normal development suggests that sufﬁcient amounts of AmfS are exported (through other transporters) to provide the developmental signal. However, this contradicts the observation that an amfAB mutant cannot complement an amfS mutant. Therefore, generation of the developmental signal by accumulated AmfS in the cytoplasm, in its native or processed form, is more likely. The role of RamAB/AmfAB as transporters seems to be crucial for RamS/AmfS activity, although this is not supported by the phenotype of the corresponding disruption mutants. The observation that a ramB mutant has a bld phenotype (Ma and Kendall, 1994) could not be reproduced by others. The importance of ramAB is demonstrated in S. lividans by the observation that a triple disruption, ramABR, has a bld phenotype, while the ramR mutant in the end does produce some aerial hyphae (Keijser et al., 2000). Interestingly, a synthetic full-length AmfS peptide was not capable of inducing aerial hyphae formation in the amfS mutant, but a synthetic C-terminal octapeptide did induce aerial growth in the mutant (Ueda et al., 2002). 5. Are RamS and SapB the Same Protein? Another small peptide implicated in the onset of aerial hyphae formation is SapB (Willey et al., 1991), which has been suggested to be identical to RamS (Chater and Horinouchi, 2003). This hypothesis was supported by the observation that SapB levels are signiﬁcantly higher in strains carrying multiple copies of ramR or ramSABR (Keijser et al., 2002; Nguyen et al., 2002). The N-terminal amino acid sequence of SapB, TG(S/G)RR, is 4/5 identical to residues 22 to 26 of RamS (Keijser et al., 2002; Willey et al., 1991). Assuming that SapB consists of the C-terminal 21 amino acids of RamS, a mass of 2099 Da would be

78

VAN WEZEL AND VIJGENBOOM

expected. However, the reported mass of SapB is 2027 Da (Keijser et al., 2003; Willey et al., 1991, ). In an elegant new study by Kodani et al. (2004), it was shown that the mass of 2027 Da is in agreement with a posttranslational modiﬁed peptide consisting of the C-terminal 21 amino acids of RamS, with four out of ﬁve serine residues dehydrated resulting in didehydroalanine residues. Two of these residues then react with the two cysteine residues, producing two eight-membered rings with thioether lanthionine bridges (Fig. 2). The latter modiﬁcations are suggested to be introduced by the C-terminal domain of RamC that shows signiﬁcant similarity to lantibiotic modifying enzymes. Whether the N-terminal domain of RamC, having similarity to ser/thr kinases, plays a role in the RamS processing remains to be elucidated. Another question that remains relates to the initial observation that SapB reacts with Schiff’s reagent (Willey et al., 1991), indicating the presence of a sugar residue or another molecule with vicinal hydroxyl groups, which is not explained by the current structure. C. NOVEL REGULATORS OF SPORULATION: THE SSGA-LIKE PROTEINS (SALPS) 1. Occurrence of SALPs Another novel family of developmental regulators ﬁrst identiﬁed in streptomycetes, and later also in other Actinomyces species such as Thermobiﬁda and Streptoverticillium, is that of the SsgA-like proteins (SALPs; Pfam PF04686). The surprising ﬁnding that enhanced expression of SsgA directly stimulates sporulation-speciﬁc cell division indicates that SsgA is an important control point in the onset of sporulation. This is supported indirectly by phylogenetic evidence, because SALPs are apparently unique to sporulating actinomycetes and are absent from the nonsporulating actinomycetes Corynebacterium glutamicum, Mycobacterium leprae, and Mycobacterium tuberculosis (van Wezel et al., 2000a). The recently completed genome sequences of S. avermitilis and S. coelicolor revealed six and seven ssgA-like genes, respectively (Bentley et al., 2002; Ikeda et al., 2003). The genes encode relatively small (130–140 aa) proteins, which share 30–50% amino acid identity (Keijser et al., 2003; van Wezel et al., 2000a). Homologs of S. coelicolor ssgA (Sco3926), ssgB (Sco1541), ssgD (Sco7622), and ssgE (Sco3158) are found on the S. avermitilis genome (Sav3926, 6810, 1687, and 3605, respectively), with high conservation in these otherwise distantly related species: The ssgB gene products differ in only one amino acid residue. The highest conservation is found in two sections

SIGNALING IN STREPTOMYCES DEVELOPMENT

79

of the proteins, corresponding to amino acid residues 13–30 and 40–65 of SsgA. In total, 20 amino acid residues (approximately 15% of the protein) are fully conserved among all 19 SALP proteins identiﬁed so far. However, there are no sequences in these proteins that resemble known functional motifs. 2. SsgA Triggers Sporulation-Specific Cell Division Studies on ssgA strongly suggest that it is an activator of sporulationspeciﬁc cell division. ssgA was originally identiﬁed as a suppressor of the hyper-sporulating mutant SY1 of Streptomyces griseus B2682 and shown to be involved in the regulation of submerged sporulation (Kawamoto and Ensign, 1995a; Kawamoto et al., 1997). The gene is of particular interest for both applied and fundamental aspects of Streptomyces research, as its expression level apparently controls morphology and development. Increased expression of SsgA alters the phenotype of liquid-grown mycelium of S. coelicolor, which normally forms large clumps but produces open mycelial structures (so-called mycelial mats) when expression is increased and shows fragmentation and submerged sporulation at high levels (van Wezel et al., 2000a,b). At these high expression levels, thick and amorphous septa are formed at regular intervals, forming spore-like compartments (Fig. 4). The stimulation by SsgA is apparently speciﬁc to sporulation-speciﬁc cell division, as ssgA mutants are defective in sporulation but form normal vegetative septa (Jiang and Kendrick, 2000; van Wezel et al., 2000a). Interestingly, ssgA mutants produce viable spores on mannitol-containing media, making it the only known ‘‘conditional’’ whi mutant. Transcriptional analysis showed that ssgA is transcribed from two developmentally regulated promoters in both S. coelicolor and in S. griseus (Traag et al., 2004; Yamazaki et al., 2003). One of these promoters is species-speciﬁc, suggesting that ssgA is regulated differently in these two organisms. This is indeed the case. In S. griseus, transcriptional activation of ssgA (further designated ssgAsg to discriminate it from ssgA from S. coelicolor, referred to as ssgAsc) is dependent on the y-butyrolactone A-factor (Horinouchi, 2002; Horinouchi and Beppu, 1995) and involves at least three different regulatory proteins. First, its transcription requires activation by AdpA, a central regulator of the A-factor pathway, which was shown to bind to three distinct sites upstream of ssgA (Yamazaki et al., 2003). The same authors showed that ssgAsg transcription is (probably indirectly) dependent on AdsA, the homolog of the developmental -factor BldN of S. coelicolor. Moreover, ssgAsg is most likely activated by the upstream located ssfR (ssgR in S. coelicolor), which encodes an IcIR-type transcriptional

80

VAN WEZEL AND VIJGENBOOM

FIG. 4. Effect of enhanced expression of SsgA on the morphology of submerged hyphae and on septation of S. coelicolor A3(2). Transmission electron micrographs show

SIGNALING IN STREPTOMYCES DEVELOPMENT

81

regulator. Although ssgA is still transcribed in an ssfR mutant, one of its promoters is strongly down-regulated. Interestingly, while transcription of ssgA is fully dependent on A-factor in S. griseus, it most probably is not in S. coelicolor. Here, the upstream located ssgR gene is essential for its activation, and both promoters are silent in an ssgR mutant (Traag et al., 2004). Considering the dominant role of SsgA in triggering Streptomyces cell division, this could explain why A-factor plays such an important role in development of S. griseus but not in S. coelicolor. The apparent requirement for tight control of ssgA transcription may be necessitated by the profound changes in Streptomyces morphology brought about by ﬂuctuations in the ssgA expression level. While transcription of ssgA and ssgR is strongly up-regulated during the onset of sporulation in S. coelicolor, it is not signiﬁcantly affected in any of the so-called early whi mutants of S. coelicolor (whiA, whiB, whiG, whiH, whiI, and whiJ ) (Traag et al., 2004). This places these genes outside the generally accepted regulatory cascade leading to solid-culture sporulation. This is the ﬁrst clear example of a sporulation gene that is expressed in a whi-independent manner. The physiological reason for this may be to provide a way to bypass the whi cascade under conditions where aerial hyphae formation is not desired, such as during ectopic or submerged sporulation. It is unclear what morphological changes occur during liquid-culture differentiation, but the fact that it does not require an aerial mycelium implies that there are two routes towards sporulation: one via the traditional whi cascade and one via the whi-independent route. SsgA is essential for submerged sporulation, and if ssgA transcription would be fully dependent on (some of) the whi genes, this process would probably be impossible because it is likely that several of the whi genes are not expressed under submerged conditions. This working hypothesis requires further testing in S. griseus. 3. ssgB Is Essential for Sporulation Expression proﬁling studies with DNA microarrays by the Cohen laboratory revealed two other SALPs as possibly developmentally regulated, namely ssgB and ssgD (Huang et al., 2001). Of these, ssgB has

S. coelicolor M145 (A–B) and S. coelicolor GSA2 overexpressing SsgA (C–F). (A) image showing vegetative hyphae and cross-walls; (B) Magniﬁcation of wild-type vegetative septum; (C) Image showing submerged hypha forming pre-spore-like compartments as the result of the overexpression of SsgA; (D–F) examples of abnormal septa in GSA2. Magniﬁcations: (A) and (C), bar ¼ 1 m; (B, D–F): bar ¼ 0.2 m. Figure reproduced from van Wezel et al., 2000a, with permission from the American Society for Microbiology.

82

VAN WEZEL AND VIJGENBOOM

been subject of more detailed transcriptional and mutational analysis. Like SsgA, SsgB is involved in the regulation of Streptomyces development but acts in an earlier phase of the sporulation process. Detailed studies with confocal ﬂuorescence microscopy and electron microscopy showed that ssgB deletion mutants fail to produce sporulation septa, and genome segregation and condensation were not observed. In contrast to ssgA mutants, the nonsporulating phenotype could not be rescued by growth on mannitol-containing media. These ssgB mutants produce colonies that are larger than those of the parental (wildtype) strain (Keijser et al., 2003), possibly linking SsgB to the process of growth cessation that occurs prior to sporulation-speciﬁc cell division (Chater, 1989, 2001; Fla¨rdh et al., 1999). The developmental role of ssgB is underlined by the observation that transcription of ssgB coincides with aerial mycelium formation and depends on the developmental H (Kormanec and Sevcikova, 2002), a factor that itself is developmentally controlled at the transcriptional and post-translational level and plays a role in stress responses (Kelemen et al., 2001; Viollier et al., 2003). However, while sigH mutants are still able to produce spores, ssgB mutants are not (Keijser et al., 2003). This suggests that ssgB is transcribed by at least one other factor, active earlier in the developmental program. Interestingly, the BldD protein is involved in the repression of the sigHp2 promoter, and therefore indirectly of ssgB. BldD is a repressor protein that becomes active at the end of the bld signaling cascade and controls transcription of the developmental factor genes whiG and bldN/whiN. These genes play crucial roles during several stages of aerial mycelium formation. This is a clear example of links that exist between the regulation of the switches to aerial mycelium formation (by the bld genes) and to sporulation (by whi genes). 4. What Is the Function of the SALPs? The mode of action of the SsgA-like proteins is as yet unknown. The relatively highly conserved region corresponding approximately to amino acid residues 20–70 possibly lends a common function to the SALPs, such as interaction with the same protein or protein complex, while the highly variable N- and C-terminal parts are likely to provide functional speciﬁcity to the individual SALP proteins. Recent data suggest that they are expressed during distinct phases in the Streptomyces life cycle and may play a role in the coordination of cell division and DNA segregation (Noens, Koerten, and van Wezel, unpublished data). In accordance with this idea, when the amino acid sequence most conserved among SALPs is used in a database screen, the only

SIGNALING IN STREPTOMYCES DEVELOPMENT

83

non-SALP hit is MukB, a protein also involved in DNA segregation in bacteria (van Wezel and Vijgenboom, unpublished). Structural and physiological studies are required to elucidate the role of this interesting new family of proteins. V. Concluding Remarks In recent years several new developmental genes have surfaced. The discovery and characterization of genes such as ram/amf and the members of the family of ssgA-like genes have shed new light on the complex morphological development in Streptomyces. Also, the ﬁrst insights were provided into the relationship between carbon metabolism and development. However, the picture is still far from complete. Important missing links are the signaling molecules, the signal receptors, and the signal transducers, postulated for connecting the individual parts of the developmental machinery, and the exact timing of their expression. Designing new approaches to disclose the identity of many more components of the developmental system is the challenge for the years to come. Helped by the recent elucidation of the complete genome sequences of S. coelicolor and S. avermitilis, and with that of many others soon to follow, signiﬁcant research efforts employing functional genomics, classical biochemistry, and protein chemistry are needed to identify and characterize the components involved in the morphological development of streptomycetes. ACKNOWLEDGMENTS We are grateful to B. Kraal, B. F. Keijser, and G. Kelemen for very useful comments on the manuscript and to H. K. Koerten for Fig. 1B. We also express our sincere thanks to the many colleagues for the stimulating discussions on the subjects discussed in this review. This work was supported by a grant from the Dutch Royal Academy of Sciences to GPvW.

REFERENCES Angell, S., Schwarz, E., and Bibb, M. J. (1992). The glucose kinase gene of Streptomyces coelicolor A3(2): Its nucleotide sequence, transcriptional analysis and role in glucose repression. Mol. Microbiol. 6, 2833–2844. Bentley, S. D., Chater, K. F., Cerdeno-Tarraga, A. M., Challis, G. L., Thomson, N. R., James, K. D., Harris, D. E., Quail, M. A., Kieser, H., Harper, D., Bateman, A., Brown, S., Chandra, G., Chen, C. W., Collins, M., Cronin, A., Fraser, A., Goble, A., Hidalgo, J., Hornsby, T., Howarth, S., Huang, C. H., Kieser, T., Larke, L., Murphy, L., Oliver, K., O’Neil, S., Rabbinowitsch, E., Rajandream, M. A., Rutherford, K., Rutter, S., Seeger, K., Saunders, D., Sharp, S., Squares, R., Squares, S., Taylor, K., Warren, T., Wietzorrek, A., Woodward, J., Barrell, B. G., Parkhill, J., and Hopwood, D. A. (2002). Complete genome

84

VAN WEZEL AND VIJGENBOOM

sequence of the model actinomycete Streptomyces coelicolor A3(2). Nature 417, 141–147. Bibb, M. (1996). 1995 Colworth Prize Lecture. The regulation of antibiotic production in Streptomyces coelicolor A3(2). Microbiology 142, 1335–1344. Bibb, M. J., and Buttner, M. J. (2003). The Streptomyces coelicolor developmental transcription factor sigma(BldN) is synthesized as a proprotein. J. Bacteriol. 185, 2338–2345. Bibb, M. J., Molle, V., and Buttner, M. J. (2000). sigma(BldN), an extracytoplasmic function RNA polymerase sigma factor required for aerial mycelium formation in Streptomyces coelicolor A3(2). J. Bacteriol. 182, 4606–4616. Bushell, M. E. (1988). Growth, product formation and fermentation technology. In ‘‘Actinomycetes in Biotechnology,’’ pp. 185–217. Academic Press, London. Chater, K. (1999). David Hopwood and the emergence of Streptomyces genetics. Int. Microbiol. 2, 61–68. Chater, K. F. (1972). A morphological and genetic mapping study of white colony mutants of Streptomyces coelicolor. J. Gen. Microbiol. 72, 9–28. Chater, K. F. (1989). Sporulation in Streptomyces. In ‘‘Regulation of Prokaryotic Development’’ (I. Smith, R. A. Slepecky, and P. Setlow, eds.), pp. 277–299. American Society for Microbiology, Washington DC. Chater, K. F., and Losick, R. (1997). Mycelial life style of Streptomyces coelicolor A3(2) and itrs relatives. In ‘‘Bacteria as Multicellular Organisms’’ (M. Dworkin, ed.), pp. 149–182. Oxford University Press, New York. Chater, K. F. (1998). Taking a genetic scalpel to the Streptomyces colony. Microbiology 144, 1465–1478. Chater, K. F. (2001). Regulation of sporulation in Streptomyces coelicolor A3(2): A checkpoint multiplex? Curr. Opin. Microbiol. 4, 667–673. Chater, K. F., and Horinouchi, S. (2003). Signalling early developmental events in two highly diverged Streptomyces species. Mol. Microbiol. 48, 9–15. Daniel, R. A., and Errington, J. (2003). Control of cell morphogenesis in bacteria: Two distinct ways to make a rod-shaped cell. Cell 113, 767–776. Donadio, S., Monciardini, P., Alduina, R., Mazza, P., Chiocchini, C., Cavaletti, L., Sosio, M., and Puglia, A. M. (2002). Microbial technologies for the discovery of novel bioactive metabolites. J. Biotechnol. 99, 187–198. Eccleston, M., Ali, R. A., Seyler, R., Westpheling, J., and Nodwell, J. (2002). Structural and genetic analysis of the BldB protein of Streptomyces coelicolor. J. Bacteriol. 184, 4270–4276. Edwards, D. H., and Errington, J. (1997). The Bacillus subtilis DivIVA protein targets to the division septum and controls the site speciﬁcity of cell division. Mol. Microbiol. 24, 905–915. Fla¨rdh, K. (2003). Essential role of DivIVA in polar growth and morphogenesis in Streptomyces coelicolor A3(2). Mol. Microbiol. 49, 1523–1536. Fla¨rdh, K., and van Wezel, G. P. (2003). Cell division during growth and development of Streptomyces. In ‘‘Recent Research Developments in Bacteriology’’ (S. G. Pandalai, ed.), pp. 71–90. Transworld Research Network, Trivandrum, India. Fla¨rdh, K., Findlay, K. C., and Chater, K. F. (1999). Association of early sporulation genes with suggested developmental decision points in Streptomyces coelicolor A3(2). Microbiology 145, 2229–2243. Gehring, A. M., Nodwell, J. R., Beverley, S. M., and Losick, R. (2000). Genomewide insertional mutagenesis in Streptomyces coelicolor reveals additional genes involved in morphological differentiation. Proc. Natl. Acad. Sci. USA 97, 9642–9647.

SIGNALING IN STREPTOMYCES DEVELOPMENT

85

Gray, D. I., Gooday, G. W., and Prosser, J. I. (1990). Apical hyphal extension in Streptomyces coelicolor A3(2). J. Gen. Microbiol. 136, 1077–1084. Hesketh, A., and Chater, K. F. (2003). Evidence from proteomics that some of the enzymes of actinorhodin biosynthesis have more than one form and may occupy distinctive cellular locations. J. Ind. Microbiol. Biotechnol. 30, 523–529. Hodgson, D. A. (2000). Primary metabolism and its control in streptomycetes: A most unusual group of bacteria. Adv. Microb. Physiol. 42, 47–238. Hopwood, D. A. (1999). Forty years of genetics with Streptomyces: From in vivo through in vitro to in silico. Microbiology 145, 2183–2202. Horinouchi, S., and Beppu, T. (1995). Autoregulators. Biotechnology 28, 103–119. Horinouchi, S. (2002). A microbial hormone, A-factor, as a master switch for morphological differentiation and secondary metabolism in Streptomyces griseus. Front. Biosci. 7, d2045–d2057. Huang, J., Lih, C. J., Pan, K. H., and Cohen, S. N. (2001). Global analysis of growth phase responsive gene expression and regulation of antibiotic biosynthetic pathways in Streptomyces coelicolor using DNA microarrays. Genes Dev. 15, 3183–3192. Huang, M. Y., and Bungay, H. R. (1973). Microprobe measurements of oxygen concentrations in mycelial pellets. Biotechnol. Bioeng. 15, 1193–1197. Hudson, M. E., Zhang, D., and Nodwell, J. R. (2002). Membrane association and kinaselike motifs of the RamC protein of Streptomyces coelicolor. J. Bacteriol. 184, 4920–4924. Ikeda, H., Ishikawa, J., Hanamoto, A., Shinose, M., Kikuchi, H., Shiba, T., Sakaki, Y., Hattori, M., and Omura, S. (2003). Complete genome sequence and comparative analysis of the industrial microorganism Streptomyces avermitilis. Nature Biotechnol. 14, 14. Ingram, C., Delic, I., and Westpheling, J. (1995). ccrA1: A mutation in Streptomyces coelicolor that affects the control of catabolite repression. J. Bacteriol. 177, 3579–3586. Jiang, H., and Kendrick, K. E. (2000). Characterization of ssfR and ssgA, two genes involved in sporulation of Streptomyces griseus. J. Bacteriol. 182, 5521–5529. Kamionka, A., Parche, S., Nothaft, H., Siepelmeyer, J., Jahreis, K., and Titgemeyer, F. (2002). The phosphotransferase system of Streptomyces coelicolor. Eur. J. Biochem. 269, 2143–2150. Kawamoto, S., and Ensign, J. C. (1995a). Cloning and characterization of a gene involved in regulation of sporulation and cell division in Streptomyces griseus. Actinomycetologica 9, 136–151. Kawamoto, S., and Ensign, J. C. (1995b). Isolation of mutants of Streptomyces griseus that sporulate in nutrient rich media: Cloning of DNA fragments that suppress the mutations. Actinomycetologica 9, 124–135. Kawamoto, S., Watanabe, H., Hesketh, A., Ensign, J. C., and Ochi, K. (1997). Expression analysis of the ssgA gene product, associated with sporulation and cell division in Streptomyces griseus. Microbiology 143, 1077–1086. Keijser, B. J. (2002). ‘‘The ram Genes and the Copper-Dependent Development of Streptomyces lividans,’’ p. 169. Leiden University, Leiden, The Netherlands. Keijser, B. J., Noens, E. E., Kraal, B., Koerten, H. K., and van Wezel, G. P. (2003). The Streptomyces coelicolor ssgB gene is required for early stages of sporulation. FEMS Microbiol. Lett. 225, 59–67. Keijser, B. J., van Wezel, G. P., Canters, G. W., Kieser, T., and Vijgenboom, E. (2000). The ram-dependence of Streptomyces lividans differentiation is bypassed by copper. J. Mol. Microbiol. Biotechnol. 2, 565–574.

86

VAN WEZEL AND VIJGENBOOM

Keijser, B. J., van Wezel, G. P., Canters, G. W., and Vijgenboom, E. (2002). Developmental regulation of the Streptomyces lividans ram genes: Involvement of RamR in regulation of the ramCSAB operon. J. Bacteriol. 184, 4420–4429. Kelemen, G. H., and Buttner, M. J. (1998). Initiation of aerial mycelium formation in Streptomyces. Curr. Op. Microbiol. 1, 656–662. Kelemen, G. H., Brian, P., Fla¨rdh, K., Chamberlin, L., Chater, K. F., and Buttner, M. J. (1998). Developmental regulation of transcription of whiE, a locus specifying the polyketide spore pigment in Streptomyces coelicolor A3(2). J. Bacteriol. 180, 2515–2521. Kelemen, G. H., Plaskitt, K. A., Lewis, C. G., Findlay, K. C., and Buttner, M. J. (1995). Deletion of DNA lying close to the glkA locus induces ectopic sporulation in Streptomyces coelicolor A3(2). Mol. Microbiol. 17, 221–230. Kelemen, G. H., Viollier, P. H., Tenor, J., Marri, L., Buttner, M. J., and Thompson, C. J. (2001). A connection between stress and development in the multicellular prokaryote Streptomyces coelicolor A3(2). Mol. Microbiol. 40, 804–814. Kendrick, K. E., and Ensign, J. C. (1983). Sporulation of Streptomyces griseus in submerged culture. J. Bacteriol. 155, 357–366. Khoklov, A. S., Tovarova, I. I., Borisova, N., Pliner, S. A., Schevchenko, L. A., Kornitskaya, N. S., Ivkina, N. S., and Rapoport, I. A. (1967). A-factor responsible for the biosynthesis of streptomycin by mutant strain of Actinomyces streptomycini. Dokl. Akad. Nauk. SSSR 177, 232–235. Kieser, T., Bibb, M. J., Buttner, M. J., Chater, K. F., and Hopwood, D. A. (2000). Practical Streptomyces genetics. John Innes Foundation, Norwich, U.K. Kodani, S., Hudson, M. E., Durrant, M. C., Buttner, M. J., Nodwell, J. R., and Willey, J. M. (2004). The SapB morphogen is a lantibiotic-like peptide derived from the product of the developmental gene ramS in Streptomyces coelicolor. Proc. Natl. Acad. Sci USA 101, 11448–11453. Kormanec, J., and Sevcikova, B. (2002). The stress-response sigma factor sigma(H) controls the expression of ssgB, a homologue of the sporulation-speciﬁc cell division gene ssgA, in Streptomyces coelicolor A3(2). Mol. Genet. Genomics 267, 536–543. Krabben, P. (1997). Morphology of Penicillium chrysogenum. Technical University of Denmark, Lyngby, Denmark. Kudo, N., Kimura, M., Beppu, T., and Horinouchi, S. (1995). Cloning and characterization of a gene involved in aerial mycelium formation in Streptomyces griseus. J. Bacteriol. 177, 6401–6410. Leskiw, B. K., Bibb, M. J., and Chater, K. F. (1991a). The use of a rare codon speciﬁcally during development. Mol. Microbiol. 5, 2861–2867. Leskiw, B. K., Lawlor, E. J., Fernandez-Abalos, J. M., and Chater, K. F. (1991b). TTA codons in some genes prevent their expression in a class of developmental, antibioticnegative, Streptomyces mutants. Proc. Natl. Acad. Sci. USA 88, 2461–2465. Locci, R. (1980). Response of developing branched bacteria to adverse environments. II. Micromorphological effects of lysozyme on some aerobic actinomycetes. Zentralbl. Bakteriol. Acta 247, 374–382. Ma, H., and Kendall, K. (1994). Cloning and analysis of a gene cluster from Streptomyces coelicolor that causes accelerated aerial mycelium formation in Streptomyces lividans. J. Bacteriol. 176, 3800–3811. Martin, S. M., and Bushell, M. E. (1996). Effect of hyphal micromorphology on bioreactor performance of antibiotic-producing saccharopolyspora erythraea cultures. Microbiology 142, 1783–1788.

SIGNALING IN STREPTOMYCES DEVELOPMENT

87

McCormick, J. R., Su, E. P., Driks, A., and Losick, R. (1994). Growth and viability of Streptomyces coelicolor mutant for the cell division gene ftsZ. Mol. Microbiol. 14, 243–254. Mendez, C., Brana, A. F., Manzanal, M. B., and Hardisson, C. (1985). Role of substrate mycelium in colony development in Streptomyces. Can. J. Microbiol. 31, 446–450. Merrick, M. J. (1976). A morphological and genetic mapping study of bald colony mutants of Streptomyces coelicolor. J. Gen. Microbiol. 96, 299–315. Nguyen, K. T., Willey, J. M., Nguyen, L. D., Nguyen, L. T., Viollier, P. H., and Thompson, C. J. (2002). A central regulator of morphological differentiation in the multicellular bacterium Streptomyces coelicolor. Mol. Microbiol. 46, 1223–1238. Nielsen, J. (1996). Modelling the morphology of ﬁlamentous microorganisms. Trends Biotechnol. 14, 438–443. Nodwell, J. R., Mcgovern, K., and Losick, R. (1996). An oligopeptide permease responsible for the import of an extracellular signal governing aerial mycelium formation in streptomyces coelicolor. Mol. Microbiol. 22, 881–893. Nodwell, J. R., Yang, M., Kuo, D., and Losick, R. (1999). Extracellular complementation and the identiﬁcation of additional genes involved in aerial mycelium formation in Streptomyces coelicolor. Genetics 151, 569–584. O’Connor, T. J., Kanellis, P., and Nodwell, J. R. (2002). The ramC gene is required for morphogenesis in Streptomyces coelicolor and expressed in a cell type-speciﬁc manner under the direct control of RamR. Mol. Microbiol. 45, 45–57. Ohnishi, Y., Seo, J. W., and Horinouchi, S. (2002). Deprogrammed sporulation in Streptomyces. FEMS Microbiol. Lett. 216, 1–7. Pope, M. K., Green, B. D., and Westpheling, J. (1996). The bld mutants of Streptomyces coelicolor are defective in the regulation of carbon utilization, morphogenesis and cell-cell signalling. Mol. Microbiol. 19, 747–756. Pope, M. K., Green, B., and Westpheling, J. R. W., IV (1998). The bldb gene encodes a small protein required for morphogenesis, antibiotic production, and catabolite control in Streptomyces coelicolor. J. Bacteriol. 180, 1556–1562. Prosser, J. I., and Tough, A. J. (1991). Growth mechanisms and growth kinetics of ﬁlamentous microorganisms. Crit. Rev. Biotechnol. 10, 253–274. Rueda, B., Miguelez, E. M., Hardisson, C., and Manzanal, M. B. (2001). Mycelial differentiation and spore formation by Streptomyces brasiliensis in submerged culture. Can. J. Microbiol. 47, 1042–1047. Ryding, N. J., Bibb, M. J., Molle, V., Findlay, K. C., Chater, K. F., and Buttner, M. J. (1999). New sporulation loci in Streptomyces coelicolor A3(2). J. Bacteriol. 181, 5419–5425. Seo, J. W., Ohnishi, Y., Hirata, A., and Horinouchi, S. (2002). ATP-binding cassette transport system involved in regulation of morphological differentiation in response to glucose in Streptomyces griseus. J. Bacteriol. 184, 91–103. Takano, E., and Bibb, M. J. (1994). The stringent response, ppGpp and antibiotic production in Streptomyces coelicolor A3(2). Actinomycetologica 8, 1–8. Takano, E., Chakraburtty, R., Nihira, T., Yamada, Y., and Bibb, M. J. (2001). A complex role for the gamma-butyrolactone SCB1 in regulating antibiotic production in Streptomyces coelicolor A3(2). Mol. Microbiol. 41, 1015–1028. Thomaides, H. B., Freeman, M., El Karoui, M., and Errington, J. (2001). Division site selection protein DivIVA of Bacillus subtilis has a second distinct function in chromosome segregation during sporulation. Genes Dev. 15, 1662–1673. Traag, B., Kelemen, G. H., and van Wezel, G. P. (2004). Transcription of the sporulation gene ssgA is activated by the IclR-type regulator SsgR in a whi-independent manner in Streptomyces coelicolor A3(2). Mol. Microbiol. 53, 985–1000.

88

VAN WEZEL AND VIJGENBOOM

Trinchi, A. P. J. (1971). A study of kinetics of hyphal extension and branch initiation of fungal mycelia. J. Gen. Microbiol. 81, 225–236. Ueda, K., Hsheh, C. W., Tosaki, T., Shinkawa, H., Beppu, T., and Horinouchi, S. (1998). Characterization of an A-factor-responsive repressor for amfR essential for onset of aerial mycelium formation in Streptomyces griseus. J. Bacteriol. 180, 5085–5093. Ueda, K., Matsuda, K., Takano, H., and Beppu, T. (1999). A putative regulatory element for carbon-source-dependent differentiation in Streptomyces griseus. Microbiology 145, 2265–2271. Ueda, K., Miyake, K., Horinouchi, S., and Beppu, T. (1993). A gene cluster involved in aerial mycelium formation in Streptomyces griseus encodes proteins similar to the response regulators of two-component regulatory systems and membrane translocators. J. Bacteriol. 175, 2006–2016. Ueda, K., Oinuma, K., Ikeda, G., Hosono, K., Ohnishi, Y., Horinouchi, S., and Beppu, T. (2002). AmfS, an extracellular peptidic morphogen in Streptomyces griseus. J. Bacteriol. 184, 1488–1492. van Wezel, G. P., van der Meulen, J., Kawamoto, S., Luiten, R. G. M., Koerten, H. K., and Kraal, B. (2000a). ssgA is essential for sporulation of Streptomyces coelicolor A3(2) and affects hyphal development by stimulating septum formation. J. Bacteriol. 182, 5653–5662. van Wezel, G. P., van der Meulen, J., Taal, E., Koerten, H., and Kraal, B. (2000b). Effects of increased and deregulated expression of cell division genes on the morphology and on antibiotic production of streptomycetes. Antonie Van Leeuwenhoek 78, 269–276. Viollier, P. H., Minas, W., Dale, G. E., Folcher, M., and Thompson, C. J. (2001a). Role of acid metabolism in Streptomyces coelicolor morphological differentiation and antibiotic biosynthesis. J. Bacteriol. 183, 3184–3192. Viollier, P. H., Nguyen, K. T., Minas, W., Folcher, M., Dale, G. E., and Thompson, C. J. (2001b). Roles of aconitase in growth, metabolism, and morphological differentiation of Streptomyces coelicolor. J. Bacteriol. 183, 3193–3203. Viollier, P. H., Weihofen, A., Folcher, M., and Thompson, C. J. (2003). Post-transcriptional regulation of the Streptomyces coelicolor stress responsive sigma factor, SigH, involves translational control, proteolytic processing, and an anti-sigma factor homolog. J. Mol. Biol. 325, 637–649. Wardell, J. N., Stocks, S. M., Thomas, C. R., and Bushell, M. E. (2002). Decreasing the hyphal branching rate of Saccharopolyspora erythraea NRRL 2338 leads to increased resistance to breakage and increased antibiotic production. Biotechnol. Bioeng. 78, 141–146. Wildermuth, H. (1970). Development and organization of the aerial mycelium in Streptomyces coelicolor. J. Gen. Microbiol. 60, 43–50. Willey, J., Santamaria, R., Guijarro, J., Geistlich, M., and Losick, R. (1991). Extracellular complementation of a developmental mutation implicates a small sporulation protein in aerial mycelium formation by S. coelicolor. Cell 65, 641–650. Willey, J., Schwedock, J., and Losick, R. (1993). Multiple extracellular signals govern the production of a morphogenetic protein involved in aerial mycelium formation by Streptomyces coelicolor. Genes Dev. 7, 895–903. Yamazaki, H., Ohnishi, Y., and Horinouchi, S. (2003). Transcriptional switch on of ssgA by A-factor, which is essential for spore septum formation in Streptomyces griseus. J. Bacteriol. 185, 1273–1283. Young, K. D. (2003). Bacterial shape. Mol. Microbiol. 49, 571–580.

Polysaccharide Breakdown by Anaerobic Microorganisms Inhabiting the Mammalian Gut HARRY J. FLINT Microbial Genetics Group Rowett Research Institute Bucksburn, Aberdeen, AB21 9SB, United Kingdom E-mail: [email protected]

I. Introduction: Role of Gut Microbial Fermentation in Nutrition II. Microbial Diversity and Interactions Within Gut Ecosystems A. Diversity, Functional Groups B. Nutritional Interactions and Cross-Feeding III. Strategies for Polysaccharide Utilization by Gut Anaerobes A. Stages in Polysaccharide Utilization B. CFB (Cytophaga–Flavobacterium–Bacteroides) Phylum C. Fibrobacter D. Cellulolytic Ruminococcus Species E. Clostridial Cluster XIVa (C. coccoides/E. rectale) Group F. Bifidobacteria G. Eukaryotes IV. Applications A. Manipulation of Gut Metabolism with Probiotics, Prebiotics, and Enzymes B. Biotechnology V. Conclusions and Future Prospects References

89 92 92 92 94 94 98 101 101 104 105 106 106 106 108 109 110

I. Introduction: Role of Gut Microbial Fermentation in Nutrition A high proportion of the solar energy trapped by plants through photosynthesis is used in the synthesis of polysaccharides. The structural polysaccharides that comprise the plant cell wall are synthesized in vast amounts, approximately 40 Gt of cellulose (Coughlan, 1985) and 30 Gt of hemicellulose globally per annum, together with smaller amounts of energy reserve polysaccharides such as starch and inulin. Herbivorous and omnivorous animals have a variety of strategies for taking advantage of this constantly replenished store of energy. While almost all animals produce digestive amylases, only some invertebrates, such as mollusks and termites (Watanabe et al., 1998; Xu et al., 2001), secrete enzymes capable of degrading structural plant polysaccharides, such as cellulose. Herbivorous mammals therefore rely entirely on the remarkable degradative activities of microorganisms 89 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

90

HARRY J. FLINT

that colonize their gastrointestinal (GI) tracts for their ability to gain energy from these substrates. The anatomy of the digestive tract in herbivores reﬂects dramatic evolutionary changes that promote microbial fermentation in either the foregut (reticulo-rumen) or the hindgut (colon and/or cecum) (van Soest, 1984). In ruminants, some 60–70% of the total GI tract volume is given over to this fermentative activity, while in hindgut fermentors, such as the horse, the colon and cecum occupy a similarly large proportion of the total gut volume (Parra, 1978). Depending on the diet, short-chain fatty acids (SCFA) arising from gut microbial fermentation account for 70% of the total energy dietary supply in ruminants and up to 30% in pigs (Bergman, 1990; Fonty and Gouet, 1989). Even in man, where the colon accounts for only 17% of total gut volume, large intestinal fermentation can account for 10% of daily energy supply (Bergman, 1990). Because of its economic importance and accessibility, the rumen has played a key role in developing our understanding of the microbial ecology of anaerobic gut ecosystems. Microbial growth in the rumen provides not only energy sources in the form of SCFA but also microbial protein that can be assimilated by the ruminant further down the GI tract (van Soest, 1984). Optimizing ruminant nutrition and the utilization of poor quality plant material by ruminants for human food production remains an important goal of research. There is also increasing emphasis on reducing pollution caused by nitrogenous waste and methane production by ruminants and on reducing reliance on chemical feed additives (Flint, 1997; Russell and Rychlik, 2001). Whereas rumen microorganisms have access to all dietary carbohydrates, microorganisms in the large intestine can access only those carbohydrates that survive passage through the small intestine. These ‘‘low-’’ or ‘‘non-digestible’’ carbohydrates include the plant structural polysaccharides cellulose, xylan, and pectin and a variety of polysaccharide food additives and oligosaccharides (MacFarlane and Gibson, 1997). In addition, some portion of dietary starch (‘‘resistant starch’’) can escape small intestinal digestion. In man, such dietary carbohydrates have a wide range of claimed health beneﬁts including prevention of colorectal cancer and colitis and lowering of cholesterol (Scheppach et al., 2001; Topping and Clifton, 2001; Wollowski et al., 2000) (Table I). Short-chain fatty acids, especially butyric acid, produced by fermentation provide the major energy sources for the colonic epithelium (Csordas, 1996), and their production rates and molar ratios are inﬂuenced by the type and quantity of carbohydrate entering the large intestine (Topping and Clifton, 2001). Other consequences of

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

91

TABLE I SOME EXAMPLES OF DIETARY POLY- AND OLIGO-SACCHARIDES CONSIDERED TO INFLUENCE HUMAN HEALTH THROUGH EFFECTS ON LARGE INTESTINAL METABOLISM

Dietary carbohydrate

Claimed effects on

Reference (examples)

Pectin

Gut transit, polyamine synthesis; cancer protection, antitoxin

Hayashi et al. (2000); Noack et al. (1998); Olano-Martin et al. (2003)

Inulin

Prebiotic (biﬁdogenic); elevated butyrate; laxative

Kleessen et al. (1997)

Resistant starch

Elevated SCFA, butyrate; fecal bulking; cancer protection

Govers et al. (1999); Wolin et al. (1999)

Barley -glucans

Bile acid excretion

Dongowski et al. (2002)

Psyllium husk, whole cereal ﬂour

Reduced cholesterol, low density lipoprotein

Adam et al. (2001); Anderson et al. (2000)

Fructo-oligosaccharides

Prebiotic (biﬁdogenic); elevated butyrate

Kleessen et al. (2001)

Lactulose

Laxative, cancer protection

Rowland et al. (1996)

fermentation, such as excessive gas production, can, however, be detrimental. Carbohydrates can also act as prebiotics, selecting for particular groups of bacteria and against others within the intestinal community, resulting potentially in reduced pathogen populations or immune stimulation (Gibson, 1998). Other effects (e.g., sequestration of metabolites, pH, viscosity, and gut transit) do not depend on fermentation but can result in radical alteration of the gut environment. Intestinal fermentation of polysaccharides and oligosaccharides is also a key factor in the nutrition and health of monogastric animals including pigs, poultry, and horses. Understanding the effects of different carbohydrates on gut metabolism requires us to know not only which bacterial groups degrade different substrates but also what strategies they use to compete for energy from the available carbohydrate energy sources. This review attempts to summarize our state of knowledge with particular emphasis on the utilization of dietary polysaccharides by anaerobic gut bacteria, drawing on information from the rumen and large intestinal systems.

92

HARRY J. FLINT

II. Microbial Diversity and Interactions Within Gut Ecosystems A. DIVERSITY, FUNCTIONAL GROUPS The rumen and large intestine are the most densely colonized regions of the mammalian gut, where microbial cell densities can exceed 1011/ml. Rapid depletion of oxygen by the more oxygen-tolerant microorganisms results in communities that are dominated by oxygen-sensitive obligate anaerobes. Thus in the rumen, obligate anaerobes outnumber facultative anaerobes by at least 1000:1 (Hungate, 1966). These are among the most diverse and complex microbial communities found in nature. Analysis of the pig gut microbiota, based on ampliﬁed small subunit ribosomal gene sequences, detected 375 bacterial phylotypes (provisional species), of which 309 showed limited resemblance to previously described species (Leser et al., 2002). Similar analyses of the rumen (Tajima et al., 1999; Whitford et al., 1998) and of human fecal (Hayashi et al., 2002; Suau et al., 1999) and human (Hold et al., 2002) and equine (Daly et al., 2001) large intestinal bacteria have produced similar conclusions. The most abundant bacterial groups in all of these anaerobic gut communities appear to be low GþC content Gram-positive bacteria, followed by Gram-negative bacteria belonging to the CFB phylum (Table II). The wealth of available ribosomal sequence data has allowed the development of techniques for microbial detection independent of cultivation that are based on oligonucleotide probes (Fig. 1), quantitative polymerase chain reaction (QPCR), and molecular proﬁling (e.g., Franks et al., 1998; Tajima et al., 2001; Zoetendal et al., 1998). B. NUTRITIONAL INTERACTIONS AND CROSS-FEEDING The fate of hydrogen is a key factor in anaerobic ecosystems. Methanogenic, acetogenic, sulphate-reducing, and nitrate-reducing microorganisms are all potentially capable of utilizing hydrogen produced by other anaerobes, producing methane, acetate, and hydrogen sulphide and reduced nitrogenous compounds, respectively. Methane formation due to methanogenic Archaea dominates in the rumen. Methanogenesis, acetogenesis, and sulphate reduction all occur in the human large intestine, with their relative contributions apparently depending on the individual, the site within the colon, and the availability of substrates (reviewed by MacFarlane and Gibson, 1997). Reductive acetogenesis was estimated to account for 30% of acetate formation in incubations with human feces (Miller and Wolin, 1996) and is also detectable in

93

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES TABLE II

APPROXIMATE PROPORTIONS OF DIFFERENT PHYLOGENETIC GROUPS OF BACTERIA IN THE MAMMALIAN GUT, BASED ON 16S rRNA SEQUENCE ANALYSESa Rumenb

Humanc,d

Pige

Horsef

CFB phylum (Bacteroides/Prevotella spp.)

32

26, 31

11

21

Clostridial cluster XIVa (C. coccoides group)

31

46, 44

25

37

Clostridial cluster IV (C. leptum group)

4

15, 20

19

8

Other low G þ C Gram-positives

18

5, 2

33

26

Others

11

7, 3

8

1

a

It should be noted that PCR biases in the construction of 16S rRNA gene libraries may lead to over-representation of some groups and absence of others. b From data of Tajima et al. (1999), cow rumen liquor. c From data of Hold et al. (2002), human colon, three individuals. d From data of Suau et al. (1999), human feces, one individual. e From data of Leser et al. (2002), pig intestine. f From data of Daly et al. (2001), horse intestine.

FIG. 1. Fluorescent in situ recognition of a cellulolytic coccus (Ruminococcus ﬂavefaciens) by using a speciﬁc Cy3-labelled 16S rRNA probe. Courtesy of Alan Walker.

the pig large intestine (De Graeve et al., 1994) and in the rumen (Morvan et al., 1994). Hydrogen consumption has the effect of shifting fermentation in hydrogen producers from reduced products such as

94

HARRY J. FLINT

ethanol towards acetate, which can enhance energy metabolism and polysaccharide breakdown by cellulolytic microorganisms (Morvan et al., 1996; Wolin et al., 1997). Hydrogen transfer is only one of many types of nutritional interactions that occur in gut ecosystems. Lactate and succinate do not generally appear as signiﬁcant fermentation products because of consumption by other species (Fig. 2). Acetate reaches substantial concentrations in most regions of the gut, despite consumption by butyrate producers in the human colon (Barcenilla et al., 2000). A small number of species appear to act as the primary degraders of plant cell wall material in gut ecosystems, but many others beneﬁt through cross-feeding of breakdown products (Fig. 2). Many cellulolytic rumen bacteria, such as Fibrobacter succinogenes and strains of Ruminococcus ﬂavefaciens, release xylo-oligosaccharides that they cannot themselves utilize, and these are efﬁciently utilized by noncellulolytic bacteria such as Prevotella spp. (Dehority, 1991; Dehority and Scott, 1967; Osborne and Dehority, 1989). In turn, some cellulolytic bacteria depend on other members of the gut microbial community for certain vitamins and precursors for amino acid synthesis (Hungate and Stack, 1982; Scott and Dehority, 1965). While such interactions are best documented for ﬁber breakdown in the rumen, similar interactions undoubtedly occur in the mammalian large intestine and with other complex substrates. Antagonistic interactions (e.g., bacteriocin formation [Kalmakoff and Teather, 1997]), may also be signiﬁcant factors in interstrain/species competition.

III. Strategies for Polysaccharide Utilization by Gut Anaerobes A. STAGES IN POLYSACCHARIDE UTILIZATION The main steps involved in the utilization of polysaccharides by individual microorganisms can be summarized as (1) attachment to the substrate; (2) disruption and enzymatic degradation of the substrate; (3) transport of breakdown products into the cell, accompanied by further degradation, metabolism and energy generation. 1. Attachment In the rumen, plant fragments develop a microbiota distinct from that of the lumen that may considered as a bioﬁlm (McAllister et al., 1994). Rather little is known, however, about the initial stages of colonization

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

95

FIG. 2. Nutritional interactions between polysaccharide degrading microorganisms in the rumen. This ﬁgure has been modiﬁed from Flint (1997), with permission.

of insoluble substrates by obligately anaerobic gut microorganisms. Motility and chemotaxis may assist some bacterial species, fungal zoospores, and protozoa in reaching their substrates. Possible quorum-sensing molecules have been detected in culture ﬂuid from some rumen bacteria (Mitsumori et al., 2003). Initial attachment of cellulolytic bacterial cells is often likely to involve the extensive glycocalyx (Latham et al., 1978; Miron et al., 2001; Roger et al., 1990). In addition, substrate-binding modules in microbial enzymes and

96

HARRY J. FLINT

cell-associated structural proteins mediate speciﬁc binding to polysaccharides including cellulose, xylan, and starch. It has been proposed in R. albus that special appendages resembling type IV pili have a role in binding to cellulose (Pegden et al., 1998). In Salmonella typhimurium, bioﬁlm formation appears to involve the synthesis of thin, aggregative ﬁmbriae and of an exopolysaccharide, bacterial cellulose (Romling et al., 2000). 2. Degradation Detailed treatment of the biochemistry and structural biology of the >1,000 individual microbial polysaccharide-degrading enzymes so far studied is beyond the scope of this review, but this information can be accessed in specialist reviews (Bayer et al., 1998; Bourne and Henrissat, 2001; Warren, 1996) and through the excellent CAZY website (http:// afmb.cnrs-mrs.fr/CAZY/index.html). The great majority of such microbial enzymes are multi-modular in their organization. Single polypeptides may consist of one or more catalytic domains, which can differ in their enzymatic speciﬁcity (e.g., Flint et al., 1993; Fontes et al., 1995), together with modules responsible for substrate binding and protein: protein interactions, linker regions between domains and modules, and a variety of other domains whose functions have yet to be elucidated. Catalytic domains for glycoside hydrolases that cleave polysaccharide chains have been classiﬁed into 91 different families based on their primary amino acid sequences. As far as is known, enzymes belonging to the same family show the same mechanism of hydrolysis (inverting versus retaining) and have similar catalytic sites and threedimensional structures; families with related 3D structures can be further grouped into superfamilies (Henrissat et al., 1995). A variety of substrate speciﬁcities, and both exo- and endo-acting enzymes, can be found within the same family. In addition, 13 families of carbohydrate lyase are known, of which seven include enzymes involved in pectin breakdown. Lyases are also involved in the degradation of hostderived polysaccharides such as heparin and chondroitin (Guthrie et al., 1985). Another important group of catalytic domains are those involved in cleaving ester bonds that link substituents such as acetyl groups and phenolic acid residues present in xylans and acetyl and methyl groups in pectins (e.g., Aurilia et al., 2000; Dalrymple et al., 1997; Dongowski et al., 2000; McSweeney et al., 1998). Thirteen different families of carbohydrate esterase are currently described. Similar sequence diversity is encountered among substrate binding modules, with 33 different families so far described. Functionally

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

97

these modules are classiﬁed by the size, solubility, composition, and organization of the substrate molecules that they bind (e.g., crystalline cellulose, amorphous cellulose, large or small cello-oligosaccharides) (reviewed in Tomme et al., 1995, 1998). Some binding modules enhance catalytic activity of the enzymes or complexes carrying them against insoluble substrates (e.g., Din et al., 1994). Linkers typically consist of repeated threonines and/or serines and provide ﬂexible regions between different functional domains in large polypeptides as well as sites for glycosylation (Warren, 1996). Dockerin modules are regions of up to 80 amino acids occurring typically at the C-terminus of the polypeptide that contain two inexact repeats of an EF hand-type Ca2þ binding motif. They were initially reported from Clostridium spp. and were shown to interact with cohesin modules present in noncatalytic structural proteins located on the cell surface (Beguin and Lemaire, 1996). Their role is in the assembly of cellulosomes, extremely large (2–6.5 MDa in C. thermocellum) surface–bound complexes of plant cell wall degrading enzymes (Bayer et al., 1998; Beguin and Lemaire 1996). The breakdown of crystalline cellulose may involve processive endocellulases (Gilad et al., 2003) and/or synergy between endo- and one or more exo-acting cellulases (Barr et al., 1996). The degradation of branched polysaccharide substrates, including xylans and pectins, as well as amylopectin starch, relies on synergy between enzymes that cleave the main chain and debranching enzymes (e.g., Biely et al., 1986). Cooperation between many different enzyme speciﬁcities is therefore required for the efﬁcient degradation of plant cell walls, in which cellulose ﬁbrils are embedded in a matrix of hemicellulose and pectin. In the cellulosomes of Gram-positive bacteria most of these speciﬁcities are concentrated together into high-molecular-weight enzyme complexes (Bayer et al., 1998). 3. Transport Sugars resulting from polysaccharide breakdown can potentially be taken up as monosaccharides, disaccharides, or oligosaccharides by several widespread mechanisms. These include ATP driven, binding protein-dependent systems known as ABC transporters; ion-linked transport (symport, antiport, uniport) via members of the major facilitator superfamily; and PEP-dependent phosphotransferase transport (Saier, 2000). Although ﬁrst described in Gram-negative bacteria where they involve periplasmic binding proteins, ABC transporters are also abundant in Gram-positive bacteria where binding proteins

98

HARRY J. FLINT

are surface-bound lipoproteins and include systems that can take up relatively large molecules. An ABC transporter for cellobiose/triose has been described in Streptomyces reticuli (Schlosser et al., 1999). The ability to take up oligosaccharides, as the immediate products of polysaccharide breakdown, may be particularly important for survival in the highly competitive environment of the gut, and many anaerobic bacteria show poorer growth on monosaccharides compared with disaccharides or oligosaccharides (e.g., Thurston et al., 1993). Oligosaccharides may be cleaved by hydrolysis before or after transport or by phosphorylysis during transport. A wide variety of di/oligosaccharide phosphorylases is known from various bacteria (Kitaoka and Hayashi, 2002). Cellobiose phosphorylase occurs in many rumen bacteria (Ayers, 1959; Lou et al., 1996), and cellodextrin phosphorylase is reported in Clostridium stercorarium (Reichenbecher et al., 1997). Most of the hexose released by polysaccharide breakdown is metabolized via the EMP pathway in the rumen (Russell and Wallace, 1997) and in the human large intestine (Miller and Wolin, 1996). In biﬁdobacteria, however, hexoses are metabolized via the fructose6-phosphate shunt, while heterofermentative lactobacilli use the hexose monophosphate shunt pathway (Gottschalk, 1979; MacFarlane and Gibson, 1997). B. CFB (CYTOPHAGA–FLAVOBACTERIUM–BACTEROIDES) PHYLUM The most abundant Gram-negative bacteria in the GI tract belong to the Cytophaga–Flavobacterium–Bacteroides (CFB) phylum. Many species play important roles in polysaccharide breakdown in the gut, although few if any appear to be cellulolytic; the species referred to as ‘‘B. cellulosolvens’’ is in fact related to Gram-positive bacteria. Rumen Prevotella spp. form a phylogenetically distinct subgroup of CFB bacteria that accounts for at least 30% of rumen bacterial diversity (Ramsak et al., 2000; Tajima et al., 1999; Whitford et al., 1998; Wood et al., 1998). Many species are xylanolytic and pectinolytic and utilize these polymers efﬁciently when in coculture with cellulolytic bacteria (Fondevila and Dehority, 1996; Osborne and Dehority, 1989). Of 188 human colonic Bacteroides strains surveyed by Salyers et al. (1977a,b), most (72%) were able to ferment amylose and amylopectin. Only 25% of strains (mainly B. ovatus and B. eggerthii) fermented xylan, while almost half (including B. ovatus and B. thetaiotaomicron) fermented pectin and polygalacturonate. Although none could ferment pig gastric mucin, some, including B. thetaiotaomicron and B. ovatus

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

99

FIG. 3. Probable location of the major polysaccharide-hydrolyzing activities in different gut groups of anaerobe. Examples are provided by starch utilization in Bacteroides thetaiotaomicron, cell surface enzyme complexes in R. ﬂavefaciens and rumen fungi, and xylanases in Polyplastron multivesiculatum (see text).

strains, could ferment individual host–derived polysaccharides including heparin, chondroitin sulphate, and hyaluronate. In B. thetaiotaomicron, the hydrolytic enzymes responsible for starch breakdown (amylases, pullulanases) are primarily cellassociated and fractionate with periplasmic markers (Anderson and Salyers, 1989) (Fig. 3). Biochemical and subsequent genetic evidence indicated that starch molecules larger than maltohexaose bound to outer membrane proteins, with subsequent transportation and processing by periplasmic enzymes. This ability to sequester large polysaccharide fragments is thought to be important in the ability of B. thetaiotaomicron to compete for nutrients in the large intestine. Four outer membrane proteins encoded by a cluster of starch utilization (sus) genes appear to be involved in starch binding, with susC and susD playing the major roles (Reeves et al., 1997; Shipman et al., 2000). The linked gene susG encodes a low-afﬁnity outer membrane amylase that is thought to hydrolyze starch molecules bound to the cell surface (Shipman et al., 1999). A major starch-degrading enzyme encoded by susA, is a neopullulanase, a periplasmic enzyme able to hydrolyse –(1,4) linkages in amylose, amylopectin, or pullulan (D’Elia and Salyers, 1996). Disruption of susA reduced the growth rate on starch by 30% (D’Elia and Salyers, 1996).

100

HARRY J. FLINT

The complete genome sequence of Bacteroides thetaiotaomicron 5482 (6.26 Mb) is now available. This is one of the largest bacterial genomes and contains at least 172 genes encoding enzymes that are involved in the breakdown of polysaccharides. In addition, there are 71 glycosyltransferase genes that are assumed to be involved in capsular polysaccharide formation (Xu et al., 2003). This gene multiplicity reﬂects considerable redundancy, including, for example, 31 -galactosidase genes and 8 amylase genes. In addition, the genome contains 163 copies of susC- or susD-related genes, many of them linked to polysaccharide utilization genes, that are presumed to encode outer membrane binding proteins. The abundance of sus genes suggests that the sequestration mechanism described previously for starch may apply quite generally to the utilization of macromolecules in this group of organisms. The apparently remarkable redundancy in polysaccharide utilization genes may simply be required to ensure high enough production of the hydrolytic enzymes, although subtle differences in regulation and speciﬁcity may also occur between related enzymes. Curiously, although 11 xylanase genes were apparently identiﬁed in its genome, B. thetaiotaomicron does not grow on xylan. These genes, however, belong to glycoside hydrolase family 43 rather than to families 10 or 11, which account for the majority of microbial xylanases. Since many family 43 enzymes function as xylosidases, their real role could be in scavenging oligosaccharides. Xylan degradation clusters identiﬁed in the xylan-utilizing species P. bryantii from the rumen (Gasparic et al., 1995a; Miyamoto et al., 2003) and B. ovatus from the human colon (Weaver et al., 1992) in both cases reveal a family 10 xylanase linked to a family 43 enzyme. P. bryantii (formerly P. ruminicola) possesses at least one other family 10 xylanase, with an unusual primary structure (Flint et al., 1997). Xylanase activity appears to be largely cell-associated both in Bacteroides (Hespell and Whitehead, 1990) and in rumen Prevotella bryantii, where assayable xylanase activity is increased ﬁvefold by sonication (Miyazaki et al., 1997). Although P. bryantii possesses a carboxymethylcellulase, its role appears to be in the degradation of mixed link –glucans (Fields et al., 1998) and the cloned enzyme has 1000-fold higher activity against barley –glucan than CM-cellulose (Gasparic et al., 1995b). P. bryantii (formerly P. ruminicola) has the ability to store excess carbon as glycogen, which can account for 60% of cell dry weight (Lou et al., 1997). Under low N conditions, excess carbohydrate can result in severe loss in viability due to the accumulation of methyl glyoxal (Russell, 1998).

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

101

C. FIBROBACTER Recent estimates of the combined populations of the three major species of cellulolytic bacteria in the rumen (F. succinogenes, R. albus, and R. ﬂavefaciens) by molecular probing are around 4% (Krause et al., 1999), 0.3–3.9% (Weimer et al., 1999), and 4.5% (Michalet-Doreau et al., 2002). In the latter study, R. ﬂavefaciens was found to be the most abundant of the three groups in the ruminant cecum and F. succinogenes in the rumen, while Weimer et al. (1999) reported R. albus as the most abundant in the rumen. These estimates are in broad agreement with previous work based on anaerobic cultivation. A second species F. intestinalis, has been described and as yet undescribed Fibrobacter species appear to be abundant in the cecum of the horse (Lin and Stahl, 1995). F. succinogenes is Gram-negative, but like the CFB group, shows evidence of early evolutionary divergence from proteobacteria (Grifﬁths and Gupta, 2001). The organization of polysaccharidase enzymes in this species remains unclear, but complete genome sequencing of strain S85 should soon produce the ﬁrst full picture of the enzyme complement of a cellulolytic gut species. A preliminary report on the F. succinogenes S85 genome indicates that it contains at least 24 endoglucanase and cellodextrinase genes and at least 23 genes concerned with hemicellulose breakdown (Nelson et al., 2002). Sixteen cell surface cellulose binding proteins, including six endoglucanases, have been identiﬁed, of which 13 are glycosylated, and it is postulated from the behavior of mutant strains that glycosylation may play an important role in substrate attachment (Miron and Forsberg, 1999). Despite possessing multiple xylanases, F. succinogenes fails to grow on xylan breakdown products and lacks xylose isomerase activity (Matte et al., 1992). This presumably indicates that this bacterium gains a more than sufﬁcient energy supply from cellulose, with xylanases necessary for gaining access to the cellulose in plant cell walls. F. succinogenes is observed to secrete glucose and cellotriose in the presence of excess cellobiose (Russell, 1998; Wells et al., 1995). D. CELLULOLYTIC RUMINOCOCCUS SPECIES As discussed above, two species of ruminococci (R. ﬂavefaciens and R. albus) are among the most abundant cellulose-degrading bacteria in the rumen and may also make a major contribution to plant cell wall breakdown in the large intestine (e.g., of horses [Julliand et al., 1999]

102

HARRY J. FLINT

FIG. 4. (A) A model for the organization of plant cell wall degrading enzymes in the cellulolytic bacterium Ruminococcus ﬂavefaciens (based on Ding et al., 2001, and Rincon et al., 2003). Structural proteins carrying repeat cohesin domains (ScaA and ScaB) together form a scaffold for the assembly of enzyme subunits that include a variety of catalytic domains (CD1,2,3-eg cellulase, xylanase, esterase) and carbohydrate-binding modules (CBMs). Two types of sequence that pair with the cohesins (dockerins) are indicated by different hatchings. A divergent cohesin present in the Sca protein binds a

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

103

and pigs [Varel and Yen, 1997]). Both species are detectable in human feces (Wang et al., 1997), although their noncellulolytic relatives R. bromii and R. callidus have been isolated more frequently (Moore and Moore, 1995). R. albus, R. ﬂavefaciens. R. bromii, and R. callidus belong to the clostridial cluster IV (Collins et al., 1994), whose members may account for up to 20% of bacteria present in human feces (Franks et al., 1998; Hold et al., 2002; Suau et al., 1999). In R. ﬂavefaciens many polysaccharidases appear to be organized into high-molecular-weight enzyme complexes. Dockerin modules have been found in six of the seven plant cell wall degrading enzymes so far described from R. ﬂavefaciens 17, which include cellulases, mixed link -glucanases, xylanases, and esterases (Aurilia et al., 2000). The xylanase XynA does not contain a dockerin (Zhang and Flint, 1992) and is presumably separate from the complex. A gene cluster has now been identiﬁed that encodes three proteins—ScaA, ScaB, and ScaC—that carry repeated cohesin-like domains. The three cohesins in ScaA can interact with one group of enzyme dockerins, while ScaA interacts via its C-terminal dockerin with the seven cohesins found in ScaB (Ding et al., 2001; Rincon et al., 2003) (Fig. 4). The more recently described ScaC has a dockerin that reacts with ScaA and a novel type of cohesin with a distinct binding speciﬁcity and may act as a type of adaptor broadening the binding speciﬁcity of the complex. Some features of this organization are reminiscent of the cellulosome of Clostridium thermocellum (Bayer et al., 1998), except that the scaffolding protein ScaA has not been found to carry a cellulose-binding module. Mechanisms for binding to cellulose may reside with individual enzymes (Rincon et al., 2001) or with other proteins yet to be identiﬁed. A related strain of R. ﬂavefaciens was previously shown to have lost its ability to degrade cotton, but not other forms of cellulose (Stewart et al., 1990), but the molecular basis for this is not yet clear. A distinctive feature of R. ﬂavefaciens, for which the species is named, is the production of a yellow pigment (Kopecny and Hodrova, 1997) during growth on cellulose. Such pigment production is found in other species, such as C. thermocellum (Ljungdahl et al., 1983), and

further set of as yet unidentiﬁed proteins. (B) Detection of the scaffoldin protein ScaA by immunogold labelling in cellulose-grown culture of Ruminococcus ﬂavefaciens 17. The protein is seen to be associated with the cell surface, and with substrate particles. This ﬁgure is from Rincon et al. (2003), and is reproduced by permission.

104

HARRY J. FLINT

is proposed to play a role in promoting adhesion and/or colonization of the substrate. Evidence also exists for cellulosome-associated enzymes in R. albus F40 (Ohara et al., 2000), and it is also proposed in R. albus that piluslike cell surface appendages are involved in binding the cells to the substrate (Pegden et al., 1998; Rakotoarivivina et al., 2002). Optimal growth and expression of cellulolytic activity in R. albus depends on the availability of phenylacetic and phenylpropionic acids (Stack and Hungate, 1982). R. albus and R. ﬂavefaciens possess cellobiose phosphorylase (Ayers, 1959; Helazcek and White, 1991; Thurston et al., 1993). As noted earlier for Fibrobacter, not all cellulolytic R. ﬂavefaciens strains can grow on the breakdown products of xylans (Dehority and Scott, 1967). A cluster that includes genes for xylose isomerase, -xylosidase genes and components of an oligosaccharide ABC transport system was detected in the xylan-utilizing strain R. ﬂavefaciens 17 but was not detected by DNA hybridization or activity in xylan non utilizing strains of R. ﬂavefaciens (Aurilia et al., 2001). E. CLOSTRIDIAL CLUSTER XIVa (C.

COCCOIDES/E. RECTALE)

GROUP

One of the most numerous groups of anaerobic gut bacteria is the clostridial cluster XIVa, also known as the C. coccoides/E. rectale cluster (Table I). This cluster is represented in the rumen mainly by relatives of Butyrivibrio ﬁbrisolvens, which possess hemicellulase, pectinase, and amylase activities (Dalrymple et al., 1999; Hespell and Cotta, 1995; Stewart et al., 1997). Some strains of B. ﬁbrisolvens are reported to be weakly cellulolytic, and the related cellulolytic species Eubacterium cellulosolvens occurs in the rumen (Stewart et al., 1997). Many cluster XIVa bacteria are butyrate producers, and together with F. prausnitzii, which belongs to cluster IV, they account for the majority of human gut bacteria that are able to produce butyric acid (Barcenilla et al., 2000; Pryde et al., 2002). Butyrate provides the preferred energy source for the colonic epithelium (Csordas, 1996), and its formation can be stimulated by ‘‘low digestible’’ dietary carbohydrates (Kanauchi et al., 1998; McIntyre et al., 1993; Perrin et al., 2001; Pryde et al., 2002). The plant polysaccharide-degrading activities of human fecal representatives of this group have received little attention, however, largely because of a lack of cultured representatives. Abundant butyrate-producing genera found in the human gut, Eubacterium and Roseburia spp., nevertheless, include strains able to utilize xylan, inulin, and starch (Duncan et al., 2002a, 2003).

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

105

Among nonbutyrate producing species found in cluster XIVa are several Ruminococcus species that are abundant in the human large intestine, notably R. torques, R. obeum, and R. gnavus. R. torques strains are among the few human gut bacteria that have been shown to be able degrade pig gastric mucin (Bayliss and Houston, 1984; Hoskins, 1993; Salyers et al., 1977a,b). Rather little is known about the role of gut bacteria belonging to other clostridial clusters in polysaccharide breakdown, although E. cylindroides (cluster XVI) relatives were stimulated by inulin in a human colonic fermentor simulation (Duncan et al., 2003). F. BIFIDOBACTERIA Although not well represented in 16S rRNA clone libraries (Fig. 1), biﬁdobacteria can account for 3% or more of human fecal bacteria in adults (Franks et al., 1998) and are particularly abundant in the feces of breastfed infants (Gibson et al., 1995). Biﬁdobacteria accounted for 70 of 120 colonies of human fecal bacteria able to form clear zones in soluble starch plates (MacFarlane and Englyst, 1986), while in a more recent study only biﬁdobacterial species and C. butyricum among the human fecal isolates tested were able to form clear zones in high amylose starch (Wang et al., 1999). This suggests a signiﬁcant role for biﬁdobacteria in starch breakdown in the human colon, although clear zone formation may not always equate to growth on starch in vivo, and it is not ruled out that other, less readily detected starch-degrading species play a major role. Degradation of xylan by strains of B. adolescentis and B. infantis and of arabinogalactan by B. longum was reported by Salyers et al. (1977). A genome sequence for Biﬁdobacterium longum (Schell et al., 2002) reveals far fewer polysaccharidase genes than found in B. thetaiotaomicron. Nevertheless, genes concerned with poly/oligosaccharide metabolism still account for 8% of the genome. Despite the absence of pectinases, or amylases, and cellulases, the genome includes around 40 glycoside hydrolases. The genome includes eight high-afﬁnity MalEFG type ABC systems involved with oligosaccharide transport but apparently only one PTS transporter, which it is suggested may reﬂect selection for rapid uptake of a wide variety of oligosaccharides under gut conditions (Schell et al., 2002). The MalEFG transport system in E. coli takes up maltose and maltodextrins, which then become substrates for amylomaltase, which releases glucose and a larger maltodextrin via a glycosyltransferase reaction. This process is complemented by maltodextrin phosphorylase, which releases

106

HARRY J. FLINT

glucose-1-phosphate from the nonreducing end of maltodextrins (Boos and Shuman, 1998). G. EUKARYOTES Anaerobic chytrid fungi have only relatively recently been recognized as important components of the rumen ecosystem and are now known to occur in the large intestine of other herbivores including horses and camelids (reviewed in Orpin and Joblin, 1997; Trinci et al., 1994). The cellulolytic activity of these fungi can be exceptionally high, as are the speciﬁc activities of certain isolated enzymes (Selinger et al., 1996; Wood et al., 1986). There is evidence that their enzymes, which show the type of modular organization characteristic of cellulolytic bacteria (e.g., Dalrymple et al., 1997) may also be organized into cellulosome-like complexes via noncatalytic docking sequences (Fanutti et al., 1995; Steenbakkers et al., 2001). Certain anaerobic protozoa have long been considered to play a role in plant cell wall breakdown in the rumen. Until recently, however, it was unclear whether cellulase or xylanase activity associated with protozoa was due to ingested bacterial or fungal enzymes. Sequences from cDNAs have now resolved this issue at least for Polyplastron multivesiculatum, establishing that it produces its own xylanases (Devillard et al., 1999, 2003). Plant cell wall breakdown by protozoa is likely to differ markedly from that in bacteria and fungi, since protozoa are capable of engulﬁng large particles and digestion is assumed to occur within food vacuoles (Fig. 3). Since protozoa can account for 50% of the rumen microbial biomass, their contribution to a range of polysaccharide degrading activities may be considerable. IV. Applications A. MANIPULATION OF GUT METABOLISM WITH PROBIOTICS, PREBIOTICS, AND ENZYMES Enzymatic treatment or supplementation of animal feed with polysaccharidases is of considerable commercial importance in pig and poultry production and, increasingly, in ruminant production. In ruminants the main aim is to complement the action of rumen microbes and thus to enhance the degradation of plant ﬁber (Kung et al., 2002; Wang and McAllister, 2002) (Fig. 5). The potential effects of feed enzymes in monogastric animals are more complex, including alterations in viscosity and passage rates, as well as increasing the release of

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

107

FIG. 5. Alternative approaches for optimizing the breakdown of plant ﬁber in the rumen. Degradability can potentially be improved by plant breeding, by pretreatment and formulation of feed, or the manipulation of rumen microbial activity.

breakdown products and modifying hind gut fermentation (Bedford and Schultze, 1998; Chesson, 1993). There must be considerable scope for improving the efﬁcacy and formulation of such products through better understanding of interactions with polysaccharide-degrading gut microorganisms. Dietary additives that are used in an attempt to modify gut function or enhance health by acting as substrates to stimulate particular groups of microorganisms are referred to as prebiotics (Gibson and Roberfroid, 1995). Almost all prebiotics are oligosaccharides or polysaccharides, and candidates include inulin, fructooligosaccharides, galactooligosaccharides, isomaltooligosaccharides, xylooligosaccharides, and pectic oligosaccharides (Gibson, 1998). Dietary resistant starch, although mainly present as a normal food component, is increasingly being described as a prebiotic in light of evidence for its beneﬁcial health effects (Topping and Clifton, 2001). The best studied prebiotics are inulin and fructooligosaccharides (FOS). In some cases fecal populations of biﬁdobacteria have been shown to increase tenfold in humans in response to inulin or FOS in the diet (Gibson et al., 1995; Kleessen et al., 1997; Kruse et al., 1999; Tuohy et al., 2001). It is also clear, however, that inulin and FOS are likely to affect other components of the microbial ecosystem. In rats with an associated human fecal ﬂora, inulin increased members of the E. rectale/C. coccoides group (Kleesen et al., 2001) and stimulation of E. cylindroides, ruminococci and an added Roseburia strain has been demonstrated in an anaerobic fermentor system with mixed human fecal inoculum (Duncan et al., 2003).

108

HARRY J. FLINT

Thus it seems important to understand the effects of prebiotics on the whole microbial community and not just the intended target groups, and to use knowledge of carbohydrate utilization in representative gut microbes to predict responses to novel prebiotics. Lactobacilli and biﬁdobacteria present in live yogurts are widely used as probiotics and have a range of claimed health beneﬁts (Fooks and Gibson, 2002). The combination of a prebiotic with a probiotic strain able to utilize it has been termed a synbiotic. The ability of biﬁdobacteria to attach to starch creates the possibility of encapsulation in starch granules as a method of delivery to the hindgut (Crittenden et al., 2001). Commercial interest in probiotics has so far concentrated almost entirely on relatively oxygen-tolerant lactic acid bacteria with a history of food use. There is also considerable potential, however, for exploiting some of the more numerous oxygen-sensitive gut bacteria, as evidenced by successful pilot experiments with Oxalobacter formigenes to degrade oxalate in the human GI tract (Duncan et al., 2002b). Specially selected or modiﬁed strains of strict anaerobes have also been introduced into the rumen. Perhaps not surprising given the highly complex nature of the native community, enhancing ﬁber degradation by introducing particularly active cellulolytic bacterial strains has proved difﬁcult (Krause et al., 2001). The use of added strains to restore rumen conditions altered by intensive feeding regimes (e.g., pH adjustment) so as to favor cellulose breakdown may deserve further attention (Russell and Rychlik, 2001). Perhaps the most promising application, however, has been to degrade plant toxins, such as ﬂuoroacetate (Gregg et al., 1994), offering the potential to extend the range of edible forage. B. BIOTECHNOLOGY Microbial polysaccharidases already have a long history of use and thus a wealth of applications ranging from textile manufacture, washing powders, food and drink production, and animal feed pretreatment to waste recycling and paper making (e.g., Beg et al., 2001; van Bielen and Li, 2002; van den Maarel et al., 2002). While mesophilic enzymes from gut microorganisms have usually not been the ﬁrst choice, they include some of the highest speciﬁc activity enzymes known (Selinger et al., 1996) and represent a hugely diverse reservoir of enzymes that has not yet been fully explored. In addition to single cloned enzymes, the possibility of constructing deﬁned complexes tailored for particular applications (Bayer et al., 1994) is being actively researched in

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

109

nongut Clostridium species (e.g., Fierobe et al., 2001). Not only the catalytic properties of these proteins are of value. Recombinant microbial cellulose binding domains present a range of possibilities ranging from the manipulation of plant growth (Sphigel et al., 1998) to afﬁnity chromatography (Tomme et al., 1998). There is also potential to exploit polysaccharide-degrading gut microorganisms themselves, not only as probiotics (discussed previously) but also in in vitro fermentations. Anaerobic digestors have been tested by using ruminococci to digest cellulosic material (Lynd et al., 2002) to produce potentially valuable chemical feedstocks such as succinic acid (e.g., Golkarn et al., 1997) and for the production of methane (Miller et al., 2000) and hydrogen (Nandi and Sengupta, 1998). V. Conclusions and Future Prospects Recent developments in molecular ecology have emphasized our limited knowledge of diversity and function, particularly among anaerobic bacteria that inhabit the GI tract. If we do not have cultured representatives for, perhaps, 70–90% of microbial colonizers (Daly et al., 2001; Suau et al., 1999), then how can we expect to be able to manipulate the system in any targeted manner, or to explain the selective effects of different substrates and dietary additives? Fortunately, prospects for further progress appear very good. First, it seems likely that most gut anaerobes are not intrinsically unculturable. The extreme situation found in soils and marine environments, where far less than 1% of microbial diversity may be cultured, does not apply to gut microbes, which require a minimum growth rate for survival because of the relatively rapid gut transit. Early anaerobic cultivation methods (Hungate, 1966) appear to have been very successful, and 16S rRNA sequences from cultures obtained in this way should gradually ﬁll in many gaps in the phylogenetic tree (e.g., Ramsak et al., 2000), although the possibility of obligate syntrophs remains. Second, the availability of molecular detection of speciﬁc groups by ﬂuorescence in situ hybridization (FISH) and real-time PCR (e.g., Franks et al., 1998; Tajima et al., 2001) and community proﬁling techniques such as terminal restriction fragment length polymorphism (T-RFLP), denaturing gradient gel electrophoresis (DGGE), and microarrays provides unprecedented power of analysis in vivo. Other techniques such as isotopic labelling (Gray and Head, 2001; Radajewski et al., 2000) can also be used to identify groups that use particular substrates in vivo but have yet to be applied to the gut. Third, genome sequencing of further representative polysaccharide-degrading microorganisms, allied to

110

HARRY J. FLINT

gene expression studies, will help to reveal the transport systems, enzyme systems, and regulatory circuits that allow them to utilize polysaccharides. Ultimately, the ecological niche occupied by a given microbe in the gut should be entirely predictable from its genome sequence, but we have some way to go. Particularly important for functional studies is the ability to do in vivo genetics on a wider range of anaerobic microorganisms, as illustrated by elegant work on hostbacterial interactions in B. thetaiotaomicron (e.g., Bry et al., 1996; Hwa et al., 1992).

ACKNOWLEDGMENTS The author is supported by the Scottish Executive Environment and Rural Affairs Department. Thanks are due to Sylvia Duncan and Karen Scott for proofreading the manuscript. REFERENCES Adam, A., Levrat-Verny, M. A., Lopez, H. W., Leuillet, M., Demigne, C., and Remesy, C. (2001). Whole wheat and triticale ﬂours with differing viscosities stimulate cecal fermentations and lower plasma and hepatic lipids in rats. J. Nutr. 131, 1770–1776. Anderson, J. W., Allgood, L. D., Lawrence, A., Altringer, L. A., Jerdack, G. R., Hengehold, D. A., and Morel, J. G. (2000). Cholesterol-lowering effects of psyllium intake adjunctive to diet therapy in men and women with hypercholesterolemia: Meta-analysis of 8 controlled trials. Amer. J. Clin. Nutr. 71, 472–479. Anderson, K. L., and Salyers, A. A. (1989). Biochemical evidence that starch breakdown by Bacteroides thetaiotaomicron involves outer-membrane starch-binding sites and periplasmic starch-degrading enzymes. J. Bacteriol. 171, 3192–3198. Aurilia, V., Martin, J. C., McCrae, S. I., Scott, K. P., Rincon, M. T., and Flint, H. J. (2000). Three multidomain esterases from the rumen cellulolytic anaerobe Ruminococcus ﬂavefaciens 17 that carry divergent dockerin sequences. Microbiology 146, 1391–1397. Aurilia, V., Martin, J. C., Munro, C. A., Mercer, D. K., and Flint, H. J. (2001). Organisation and strain distribution of genes responsible for the utilization of xylans by the rumen cellulolytic bacterium Ruminococcus ﬂavefaciens. Anaerobe 6, 333–340. Ayers, W. A. (1959). Phosphorolysis and synthesis of cellobiose by cell extracts of Ruminococcus ﬂavefaciens. J. Biol. Chem. 234, 2819–2822. Barcenilla, A., Pryde, S. E., Martin, J. C., Duncan, S. H., Stewart, C. S., Henderson, C., and Flint, H. J. (2000). Phylogenetic relationships of butyrate-producing bacteria from the human gut. Appl. Environ. Microbiol. 66, 1654–1661. Barr, B. K., Hsieh, Y.-L., Ganem, B., and Wilson, D. B. (1996). Identiﬁcation of two functionally different subclasses of exocellulases. Biochemistry 35, 586–592. Bayer, E. A., Morag, E., and Lamed, R. (1994). The cellulosome—a treasure trove for biotechnology. Trends Biotech. 12, 379–386. Bayer, E. A., Shimon, L. J., Shoham, Y., and Lamed, R. (1998). Cellulosomes—structure and ultrastructure. J. Struct. Biol. 124, 221–234.

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

111

Bayliss, C., and Houston, A. (1984). Characterisation of plant polysaccharide and mucinfermenting anaerobic bacteria from human feces. Appl. Environ. Microbiol. 48, 626–632. Bedford, M. R., and Schulze, H. (1998). Exogenous enzymes for pigs and poultry. Nutr. Res. Reviews 11, 91–114. Beg, Q. K., Kapoor, M., Mahajan, L., and Hoondahl, G. S. (2001). Microbial xylanases and their industrial applications: A review. Appl. Microbiol. Biotechnol. 56, 323–338. Beguin, P., and Lemaire, M. (1996). The cellulosome: An exocellular, multiprotein complex specialised in cellulose degradation. Crit. Rev. Biochem. Mol. 31, 201–236. Bergman, E. N. (1990). Energy contributions of volatile fatty acids from the gastrointestinal tract in various species. Physiol. Rev. 70, 567–590. Biely, P., MacKenzie, C. R., Puls, J., and Schneider, H. (1986). Cooperativity of esterases and xylanases in enzymatic degradation of acetylxylan. Bio/Technology 4, 731–733. Boos, W., and Shuman, H. (1998). Maltose/Maltodextrin system of Escherichia coli: Transport, metabolism and regulation. Microbiol. Mol. Biol. Revs. 62, 204–229. Bourne, Y., and Henrissat, B. (2001). Glycoside hydrolases and glycosyltransferases: Families and functional modules. Curr. Opin. Struct. Biol. 11:, 593–600. Bry, L., Falk, P. G., Midvedt, T., and Gordon, J. I. (1996). A model of host-microbial interactions in an open mammalian ecosystem. Science 273, 1380–1383. Chesson, A. (1993). Feed enzymes. Anim. Feed Sci. Technol. 45, 65–79. Collins, M. D., Lawson, P. A., Willems, A., Cordoba, J. J., Fernendez-Garayzabal, J., Garcia, P., Cai, J., Hippe, H., and Farrow, J. A. E. (1994). The phylogeny of the genus Clostridium: Proposal of ﬁve new genera and eleven new species combinations. Int. J. Syst. Bacteriol. 44, 812–826. Coughlan, M. P. (1985). The properties of fungal and bacterial cellulases with comment on their production and application. Biotechnol. Genet. Eng. Rev. 3, 39–109. Crittenden, R., Laitila, A., Forssell, P., Matto, M., Saarela, M., Mattila-Sandholm, T., and Myllarinen, P. (2001). Adhesion of Biﬁdobacteria to granular starch and its implications in probiotic technology. Appl. Environ. Microbiol. 67, 3469–3475. Csordas, A. (1996). Butyrate, aspirin and colorectal cancer. Eur. J. Cancer Prevent. 5, 221–231. D’Elia, J. N., and Salyers, A. A. (1996). Contribution of a neopullulanase, a pullulanase and an -glucosidase to growth of Bacteroides thetaiotaomicron on starch. J. Bacteriol. 178, 7173–7179. Dalrymple, B. P., Cybinski, D. H., Layton, I., McSweeney, C. S., Xue, G. P., Swadling, Y. J., and Lowry, J. B. (1997). Three Neocallimastix patriciarum esterases associated with the degradation of complex polysacchraides are members of a new family of hydrolases. Microbiology 143, 2605–2614. Dalrymple, B. P., Swadling, Y., Layton, I., Gobius, K. S., and Xue, G. P. (1999). Distribution and evolution of the xylanase genes xynA and xynB and their homologues in strains of Butyrivibrio ﬁbrisolvens. Appl. Environ. Microbiol. 65, 3660–3667. Daly, K., Stewart, C. S., Flint, H. J., and Shirazi-Beechey, S. P. (2001). Bacterial diversity within the equine large intestine as revealed by molecular analysis of cloned 16S rRNA genes. FEMS Microbiol. Ecol. 38, 141–151. De Graeve, K. G., Grivet, J. P., Durand, M., Beaumatin, P., Cordelet, C., Hannequart, G., and Demeyer, D. (1994). Competition between reductive accetogenesis and methanogenesis in the pig large intestine. J. Appl. Bacteriol. 76, 55–61. Dehority, B. A. (1991). Effects of microbial synergism on ﬁber digestion in the rumen. Proc. Nutr. Soc. 50, 149–159.

112

HARRY J. FLINT

Dehority, B. A., and Scott, H. W. (1967). Extent of cellulose and hemicellulose digestion in various forages by pure cultures of rumen bacteria. J. Dairy Sci. 50, 1136–1141. Devillard, E., Bera-Maillet, C., Flint, H. J., Scott, K. P., Newbold, C. J., Wallace, R. J., Jouany, J. P., and Forano, E. (2003). Characterisation of XynB, a modular xylanase from the ruminal protozoan Polyplastron multivesiculatum, bearing a family 22 carbohydrate binding module that binds to cellulose. Biochem. J. 373, 495–503. Devillard, E., Newbold, C. J., Scott, K. P., Forano, E., Wallace, R. J., Jouany, J. P., and Flint, H. J. (1999). A family 11 xylanase from the ruminal protozoan Polyplastron multivesiculatum. FEMS Microbiol. Letts. 181, 145–152. Din, N., Damude, H. G., Gilkes, N. R., Miller, R. C., Warren, R. A. J., and Kilburn, D. G. (1994). C-1 CX revisited—intramolecular synergism in a cellulase. PNAS 91, 11383–11387. Ding, S. Y., Rincon, M. T., Lamed, R., Martin, J. C., McCrae, S. I., Aurilia, V., Shoham, Y., Bayer, E. A., and Flint, H. J. (2001). Cellulosomal scaffoldin-like proteins from Ruminococcus ﬂavefaciens. J. Bacteriology 183, 1945–1953. Dongowski, G., Lorenz, A., and Anger, H. (2000). Degradation of pectins with different degrees of esteriﬁcation by Bacteroides thetaiotaomicron isolated from the human gut ﬂora. Appl. Environ. Microbiol. 66, 1321–1327. Dongowski, G., Huth, M., Gebhardt, E., and Flamme, W. (2002). Dietary ﬁber-rich barley products beneﬁcially affect the intestinal tract of rats. J. Nutr. 132, 3704–3714. Duncan, S. H., Hold, G. L., Barcenilla, A., Stewart, C. S., and Flint, H. J. (2002a). Roseburia intestinalis sp. nov., a novel saccharolytic, butyrate-producing bacterium from human faeces. Int. J. System. Evol. Microbiol. 52, 1615–1620. Duncan, S. H., Richardson, A. J., Kaul, R. P., Allison, M. J., and Stewart, C. S. (2002b). Oxalobacter formigenes and its potential role in human health. Appl. Envrion. Microbiol. 68, 3841–3847. Duncan, S. H., Scott, K. P., Ramsay, A. G., Harmsen, H. J. M., Welling, G. W., Stewart, C. S., and Flint, H. J. (2003). Effects of alternative dietary substrates on competition between human colonic bacteria in an anaerobic fermentor system. Appl. Environ. Microbiol. 69, 1136–1142. Fanutti, C., Ponyi, T., Black, G. W., Hazlewood, G. P., and Gilbert, H. J. (1995). The conserved noncatalytic 40-residue sequence in cellulases and hemicellulases from anaerobic fungi functions as a protein docking domain. J. Biol. Chem. 270, 29314–29322. Fields, M. W., Russell, J. B., and Wilson, D. B. (1998). The role of ruminal carboxymethyl cellulases in the degradation of beta-glucans from cereal grain. FEMS Microbiol. Ecol. 27, 261–268. Fierobe, H.-P., Mechaly, A., Tardif, C., Belaich, A., Lamed, R., Shoham, Y., Belaich, J.-P., and Bayer, E. A. (2001). Design and production of active cellulosome chimeras: Selective incorporation of dockerin-containing enzymes into deﬁned functional complexes. J. Biol. Chem. 276, 21257–21261. Flint, H. J. (1997). The rumen microbial ecosystem—some recent developments. Trends Microbiol. 5, 483–488. Flint, H. J., Martin, J. C., McPherson, C. A., Daniel, A., and Zhang, J.-X. (1993). A bifunctional enzyme having separate xylanase and (1,3-1,4). glucanase domains, encoded by the xynD gene of Ruminococcus ﬂavefaciens. J. Bacteriol. 175, 2943–2951. Flint, H. J., Whitehead, T. R., Martin, J. C., and Gasparic, A. (1997). Interrupted domain structures in xylanases from two distantly related strains of Prevotella ruminicola. Biochim. Biophys. Acta 1337, 161–165.

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

113

Fondevila, M., and Dehority, B. A. (1996). Interactions between Fibrobacter succinogenes, Prevotella ruminicola and Ruminococcus ﬂavefaciens in the digestion of cellulose from forages. J. Anim. Sci. 74, 678–684. Fontes, C. M., Hazlewood, G. P., Morag, E., Hall, J., Hirst, B. H., and Gilbert, H. J. (1995). Evidence for a general role for non-catalytic thermostabilising domains in xylanases from thermophilic bacteria. Biochem. J. 307, 151–158. Fonty, G., and Gouet, P. (1989). Fibre-degrading microorganisms in the monogastric digestive tract. Anim. Feed Sci. Technol. 23, 91–107. Fooks, L. J., and Gibson, G. R. (2002). Probiotics as modulators of the gut ﬂora. Br. J. Nutr. 88, S39–S49. Franks, A. H., Harmsen, H. J. M., Rangs, G. C., Jaansen, G. J., and Welling, G. W. (1998). Variations of bacterial populations in human feces quantiﬁed by ﬂuorescent in situ hybridization with group-speciﬁc 16S rRNA-targetted oligonucleotide probes. Appl. Environ. Microbiol. 64, 3336–3345. Gasparic, A., Daniel, A., Martin, J., and Flint, H. J. (1995a). A xylan hydrolase gene cluster from Prevotella ruminicola: Sequence relationships, oxygen sensitivity and synergistic interactions of a novel exoxylanase. Appl. Env. Microbiol. 61, 2958–2964. Gasparic, A., Marinsek-Logar, R., Martin, J., Wallace, R. J., Nekrep, F. V., and Flint, H. J. (1995b). Isolation of genes encoding xylanase, -D-xylosidase and -L-arabinofuranosidase activities from the rumen bacterium Prevotella ruminicola B14. FEMS Microbiol. Letts. 125, 135–142. Gibson, G. R. (1998). Dietary modulation of the human gut microﬂora using prebiotics. Br. J. Nutr. 80, S209–S212. Gibson, G. R., and Roberfroid, M. B. (1995). Dietary modulation of the human colonic microbiota—introducing the concept of prebiotics. J. Nutr. 125, 1401–1412. Gibson, G. R., Beatty, E. R., Wang, X., and Cummings, J. H. (1995). Selective stimulation of biﬁdobacteria in the human colon by oligofructose and inulin. Gastroenterology 108, 975–982. Gilad, R., Rabinovitch, L., Yaron, S., Bayer, E. A., Lamed, R., Gilbert, H. J., and Shoham, Y. (2003). CelI, a noncellulosomal family 9 enzyme from Clostridium thermocellum, is a processive endoglucanase that degrades crystalline cellulose. J. Bacteriol. 185, 391–398. Golkarn, R. R., Eiteman, M. A., Martin, S. A., and Eriksson, K. E. L. (1997). Production of succinate from glucose, cellobiose and various cellulosic materials by the rumen anaerobic bacteria Fibrobacter succinogenes and Ruminococcus ﬂavefaciens. Appl. Biochem. Biotechnol. 68, 69–80. Gottschalk, G. (1979). ‘‘Bacterial Metabolism.’’ Springer Verlag. Govers, M. J. A. P., Gannon, N. J., Dunshea, F. R., Gibson, P. R., and Muir, J. G. (1999). Wheat bran affects the site of fermentation of resistant starch and luminal indexes related to colon cancer risk: A study in pigs. Gut 45, 840–847. Gray, N. D., and Head, I. M. (2001). Linking genetic identity and function in communities of uncultured bacteria. Environ. Microbiol. 3, 481–492. Gregg, K., Cooper, C. L., Schafer, R. J., Sharp, H., Beard, C. E., Allen, G., and Xu, J. W. (1994). Detoxiﬁcation of the plant toxin ﬂuoroacetate by a genetically modiﬁed rumen bacterium. Biotechnology 12, 1361–1365. Grifﬁths, E., and Gupta, R. S. (2001). The use of signature sequences in different proteins to determine the relative branching order of bacterial division: Evidence that Fibrobacter diverged at a similar time to Chlamydia and the Cytophaga-FlavobacteriumBacteroides division. Microbiology 147, 2611–2622.

114

HARRY J. FLINT

Guthrie, E. P., Shoemaker, N. B., and Salyers, A. A. (1985). Cloning and expression in Escherichia coli of a gene coding for chondroitin lyase from Bacteroides thetaiotaomicron. J. Bacteriol. 164, 510–515. Hayashi, A., Gillen, A. C., and Lott, J. R. (2000). Effects of daily oral administration of quercitin chalcone and modiﬁed citrus pectin on implanted colon-25 tumour growth in Balb-C mice. Altern. Med. Rev. 5, 546–552. Hayashi, H., Sakamoto, M., and Benno, Y. (2002). Fecal microbial diversity in a strict vegetarian as determined by molecular analysis and cultivation. Microbiol. Immunol. 46, 819–831. Helaszek, C. T., and White, B. A. (1991). Cellobiose uptake and metabolism by Ruminococcus albus. Appl. Environ. Microbiol. 57, 64–68. Henrissat, B., Callebaut, I., Fabrega, S., Lehn, P., Mornon, J.-P., and Davies, G. (1995). Conserved catalytic machinery and the prediction of a common fold for several families of glycosyl hydrolase. Proc. Natl. Acad. Sci. 92, 7090–7094. Hespell, R. B., and Cotta, M. A. (1995). Degradation and utilization by Butyrivibrio ﬁbrisolvens H17c of xylans with different chemical and physical properties. Appl. Environ. Microbiol. 61, 3042–3050. Hespell, R. B., and Whitehead, T. R. (1990). Physiology and genetics of xylan degradation by gastrointestinal-tract bacteria. J. Dairy Sci. 73, 3013–3022. Hold, G. L., Pryde, S. E., Russell, V. J., Furrie, E., and Flint, H. J. (2002). Assessment of microbial diversity in human colonic samples by 16S rDNA sequence analysis. FEMS Microbiol. Ecol. 39, 33–39. Hoskins, L. C. (1993). Mucin degradation in the human gastrointestinal tract and its signiﬁcance to enteric microbial ecology. Eur. J. Gastroenterol. Hepatol. 5, 205–213. Hungate, R. E. (1966). ‘‘The Rumen and Its Microbes.’’ Academic Press, New York and London. Hungate, R. E., and Stack, R. J. (1982). Phenylpropanoic acid: A growth factor for Ruminococcus albus. Appl. Environ. Microbiol. 44, 79–83. Hwa, V., and Salyers, A. A. (1992). Analysis of two chondroitin sulphate utilization mutants of Bacteroides thetaiotaomicron that differ in their abilities to compete with the wild type in the gastrointestinal tracts of germ-free mice. Appl. Environ. Microbiol. 58, 869–876. Julliand, V., de Vaux, A., Millet, L., and Fonty, G. (1999). Identiﬁcation of Ruminococcus ﬂavefaciens as the predominant cellulolytic bacterial species of the equine cecum. Appl. Environ. Microbiol. 65, 3738–3741. Kalmakoff, M. L., and Teather, R. M. (1997). Isolation and characterisation of a bacteriocin (butyrivibriocin AR10). from the ruminal anaerobe Butyrivibrio ﬁbrisolvens AR10: Evidence in support of the widespread occurrence of bacteriocin-like activity among rumen isolates of B. ﬁbrisolvens. Appl. Environ. Microbiol. 63, 394–402. Kanauchi, O., Mitsuyama, K., Saiki, T., Fushikia, T., and Iwanaga, T. (1998). Germinated barley foodstuff increases fecal volume and butyrate production in humans. Int. J. Mol. Medicine 1, 937–941. Kitaoka, M., and Hayashi, K. (2002). Carbohydrate-processing phosphorylytic enzymes. Trends Glycosci. Glyc. 14, 35–50. Kleessen, B., Hartmann, L., and Blaut, M. (2001). Oligofructose and long chain inulin: Inﬂuence on the gut microbial ecology of rats associated with a human faecal ﬂora. Br. J. Nutr. 86, 291–300. Kleessen, B., Sykura, B., Zunft, H. J., and Blaut, M. (1997). Effects of inulin and lactose on fecal microﬂora, microbial activity and bowel habit in elderly constipated persons. Am. J. Clin. Nutr. 65, 1397–1402.

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

115

Kopecny, J., and Hodrova, B. (1997). The effect of yellow afﬁnity substance on cellulases of Ruminococcus ﬂavefaciens. Letts. Appl. Microbiol. 25, 191–196. Krause, D. O., Dalrymple, B. P., Smith, W. J., Mackie, R. I., and McSweeney, C. S. (1999). 16S rDNA sequencing of Ruminococcus albus and Ruminococcus ﬂavefaciens: Design of a signature probe and its application in adult sheep. Microbiology 145, 1797–1807. Krause, D. O., Bunch, R. J., Conlan, L. L., Kennedy, D. M., Smith, W. J., Mackie, R. M., and McSweeney, C. S. (2001). Repeated dosing of Ruminococcus spp. does not result in persistence, but changes in other microbial populations occur that can be measured with quantitative 16S based probes. Microbiology 147, 1719–1729. Kruse, H. P., Kleessen, B., and Blaut, M. (1999). Effects of inulin on faecal biﬁdobacteria in human subjects. Br. J. Nutr. 82, 375–382. Kung, L., Cohen, M. A., Rode, L. M., and Treacher, R. J. (2002). The effect of ﬁbrolytic enzymes sprayed onto forages and fed in a total mixed ration to lactating dairy cows. J. Dairy Sci. 85, 2396–2402. Latham, M. J., Brooker, B. J., Pettipher, G. L., and Harris, P. J. (1978). Ruminococcus ﬂavefaciens cell coat and adhesion to cotton celulose and to cell walls in leaves of perennial ryegrass. Appl. Environ. Microbiol. 35, 156–165. Leser, T. D., Amenuvor, J. Z., Jensen, T. K., Lindecrona, R. H., Boye, M., and Moller, K. (2002). Culture independent analysis of gut bacteria: The pig gastrointestinal tract revisited. Appl. Environ. Microbiol. 68, 673–690. Lin, C. Z., and Stahl, D. A. (1995). Taxon-speciﬁc probes for the cellulolytic genus Fibrobacter reveal abundant and novel equine-associated populations. Appl. Environ. Microbiol. 61, 1348–1351. Ljungdahl, L. G., Petterson, B., Ericksson, K.-E., and Weigel, J. (1983). A yellow afﬁnity substance involved in the cellulolytic system of Clostridium thermocellum. Curr. Microbiol. 9, 195–200. Lou, J., Dawson, K., and Strobel, H. J. (1996). Role of phosphorylytic cleavage in cellobiose and cellodextrin metabolism in Prevotella ruminicola. Appl. Environ. Microbiol. 62, 1770–1773. Lou, J., Dawson, K. A., and Strobel, H. J. (1997). Glycogen formation by the ruminal bacterium Prevotella ruminicola. Appl. Environ. Microbiol. 63, 1483–1488. Lynd, L. R., Weimer, P. J., van Zyl, W. H., and Pretorius, I. S. (2002). Microbial cellulose utilization: Fundamentals and biotechnology. Microbiol. Mol. Biol. Revs. 66, 506–577. MacFarlane, G. T., and Englyst, H. N. (1986). Starch utilization by the human large intestinal microﬂora. J. Appl. Bacteriol. 60, 195–201. MacFarlane, G. T., and Gibson, G. R. (1997). Carbohydrate fermentation, energy transduction and gas metabolism in the human large intestine. In ‘‘Gastrointestinal Microbiology’’ vol. 1, pp. 269–318. Kluwer Academic Publishers. Matte, A., Forsberg, C. W., and Gibbins, A. (1992). Enzymes associated with metabolism of xylose and other pentoses by Prevotella (Bacteroides) ruminicola strains, Selenomonas ruminantium D and Fibrobacter succinogenes S85. Can. J. Microbiol. 38, 370–376. McAllister, T. A., Bae, H. D., Jones, G. A., and Cheng, K. J. (1994). Microbial attachment and feed digestrion in the rumen. J. Anim. Sci. 72, 3004–3018. McIntyre, A., Gibson, P. R., and Young, G. P. (1993). Butyrate production from dietary ﬁber and protection against large bowel cancer in a rat model. Gut 34, 386–391.

116

HARRY J. FLINT

McSweeney, C. S., Dulieu, A., and Bunch, R. (1998). Butyrivibrio and other xylanolytic organisms from the rumen have cinnamoyl esterase activity. Anaerobe 4, 57–65. Michalet-Doreau, B., Fernandez, I., and Fonty, G. (2002). A comparison of enzymatic and molecular approaches to characterise the cellulolytic microbial ecosystems of the rumen and cecum. J. Anim. Sci. 80, 790–796. Miller, T. L., and Wolin, M. J. (1996). Pathways of acetate, propionate and butyrate formation by the human fecal microbial ﬂora. Appl. Environ. Microbiol. 62, 1589–1592. Miller, T. L., Currenti, E., and Wolin, M. J. (2000). Anaerobic bioconversion of cellulose by Ruminococcus albus, Methanobrevibacter smithii and Methanosarcina barkeri. Appl. Environ. Microbiol. 54, 494–498. Miron, J., and Forsberg, C. W. (1999). Characterisation of cellulose-binding proteins that are involved in the adhesion mechanism of Fibrobacter intestinalis DR7. Appl. Environ. Microbiol. 51, 491–497. Miron, J., Ben-Ghedalia, D., and Morrison, M. (2001). Adhesion mechanisms of rumen cellulolytic bacteria. J. Dairy Sci. 84, 1294–1309. Mitsumori, M., Xu, L. M., Kajikawa, M., Kurihara, M., Tajima, K., Hai, K., and Takanaka, A. (2003). Possible quorum-sensing in the rumen microbial community: Detection of quorum-sensing molecules from rumen bacteria. FEMS Microbiol. Letts. 219, 47–52. Miyamoto, H., Hirase, T., Mercer, D. K., Miyazaki, K., Martin, J. C., Ushida, K., Kojima, Y., and Flint, H. J. (2003). Involvement of the two component regulatory protein XynR in positive control of xylanase gene expression in the ruminal anaerobe Prevotella bryantii B14. J. Bacteriol. 185, 2219–2226. Miyazaki, K., Martin, J. C., Marinsek-Logar, R., and Flint, H. J. (1997). Degradation and utilization of xylans by the rumen anaerobe Prevotella bryantii (formerly P. ruminicola subsp. brevis B14). Anaerobe 3, 373–381. Moore, W. E. C., and Moore, L. H. (1995). Intestinal ﬂoras of populations that have a high risk of colon cancer. Appl. Environ. Microbiol. 61, 3202–3207. Morvan, B., Dore, J., Rieu Lesme, F., Foucat, L., Fonty, G., and Gouet, P. (1994). Establishment of hydrogen-utilizing bacteria in the rumen of the newborn lamb. FEMS Microbiol. Letts. 117, 249–256. Morvan, B., RieuLesme, F., Fonty, G., and Gouet, P. (1996). in vitro interactions between H2-producing cellulolytic microorganisms and H2-utilizing acetogenic and sulfatereducing bacteria. Anaerobe 2, 175–180. Nandi, R., and Sengupta, S. (1998). Microbial production of hydrogen: An overview. Crit. Rev. Microbiol. 24, 61–84. Nelson, K., Aminov, R., Forsberg, C., Mackie, R. I., Russell, J. B., White, B. A., Wilson, D. B., Mulligan, S., Tran, K., Carty, H., Khouri, H., Nelson, W., Daugherty, S., Fraser, C., and Morrison, M. (2002). The Fibrobacter succinogenes S85 genome sequencing project. Repr. Nutr. Dev. 42(Suppl.1), S44. Noack, J., Kleessen, B., Proll, J., Dongowski, G., and Blaut, M. (1998). Dietary guar gum and pectin stimulate intestinal microbial polyamine synthesis in rats. J. Nutr. 128, 1385–1391. Ohara, H., Karita, S., Kimura, T., Sakka, K., and Ohmiya, K. (2000). Characterisation of the cellulolytic complex (cellulosome). from Ruminococcus albus. Biosci. Biotech Biochem. 64, 254–260. Olano-Martin, E., Williams, M. R., Gibson, G. R., and Rastall, R. A. (2003). Pectins and pectic-oligosaccharides inhibit Escherichia coli O157:H7 shiga toxin as directed towards the human colonic cell line HT29. FEMS Microbiol. Letts. 218, 101–105.

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

117

Orpin, C. G., and Joblin, K. N. (1997). The rumen anaerobic fungi. In ‘‘The Rumen Microbial Ecosystem’’ (P. N. Hobson and C. S. Stewart, eds.). Osborne, J. M., and Dehority, B. A. (1989). Synergism in degradation and utilization of intact forage cellulose, hemicellulose and pectin by three pure cultures of rumen bacteria. Appl. Environ, Microbiol. 55, 2247–2250. Parra, R. (1978). Comparison of foregut and hindgut fermentation in herbivores. In ‘‘The Ecology of Arboreal Herbivores’’ (GG. Montgomery, ed.), p. 205. Smithsonian Institution, Washington, DC. Pegden, R. S., Larson, M. A., Grant, R. J., and Morrison, M. (1998). Adherence of the grampositive bacterium Ruminococcus albus to cellulose and identiﬁcation of a novel form of cellulose-binding protein which belongs to the Pil family of proteins. J. Bacteriol. 180, 5921–5927. Perrin, P., Pierre, F., Patry, Y., Champ, M., Berreur, M., Pradal, G., Bornet, F., Meﬂah, K., and Menanteau, J. (2001). Only ﬁbres promoting a stable butyrate producing colonic ecosystem decrease the rate of aberrant crypt foci in rats. Gut 48, 53–61. Pryde, S. E., Duncan, S. H., Hold, G. L., Stewart, C. S., and Flint, H. J. (2002). The microbiology of butyrate formation in the human colon. FEMS Microbiol. Letts 217, 133–139. Radajewski, S., Ineson, P., Parekh, N. T. R., and Murrell, J. C. (2000). Stable isotope probing as a tool in microbial ecology. Nature 403, 646–649. Rakotoarivinina, H., Jubelin, G., Hebraud, M., Gaillard-Martinie, B., Forano, E., and Mosoni, P. (2002). Adhesion to cellulose of the Gram-positive bacterium Ruminococcus albus involves type IV pili. Microbiology 148, 1871–1880. Ramsak, A., Peterka, M., Tajima, K., Martin, J. C., Wood, J., Johnston, M. E. A., Aminov, R. I., Flint, H. J., and Avgustin, G. (2000). Unravelling the genetic diversity of ruminal bacteria belonging to the CFB phylum. FEMS Microbiol. Ecol. 33, 69–79. Reeves, A. R., Wang, G. R., and Salyers, A. A. (1997). Characterization of four outer membrane proteins that play a role in utilization of starch by Bacteroides thetaiotaomicron. J. Bacteriol. 179, 643–649. Reichenbecher, M., Lottspeich, F., and Bronnenmeier, K. (1997). Puriﬁcation and properties of cellobiose phosphorylase (CepA). and a cellodextrin phosphorylase (CepB). from the cellulolytic thermophile Clostridium stercorarium. Eur. J. Biochem. 247, 262–267. Rincon, M. T., Ding, S.-Y., McCrae, S. I., Martin, J. C., Aurilia, V., Lamed, R., Shoham, Y., Bayer, E. A., and Flint, H. J. (2003). Novel organisation and divergent dockerin speciﬁcities in the cellulosome system of Ruminococcus ﬂavefaciens. J. Bacteriol. 185, 703–713. Rincon, M. T., McCrae, S. I., Kirby, J., Scott, K. P., and Flint, H. J. (2001). EndB, a multidomain cellulase from the rumen cellulolytic bacterium Ruminococcus ﬂavefaciens 17, binds cellulose via a novel cellulose binding domain and to a 130 Kda R. ﬂavefaciens protein via a dockerin domain. Appl. Environ. Microbiol. 67, 4426–4431. Roger, V., Fonty, G., Komisarczuk-Bony, S., and Gouet, P. (1990). Effects of physicochemical factors on the adhesion to cellulose Avicel of ruminal bacteria Ruminococcus ﬂavefaciens and Fibrobacter succinogenes subsp. succinogenes. Appl. Environ. Microbiol. 56, 3081–3087. Romling, U., Rohde, M., Olsen, A., Normark, S., and Reinkoster, J. (2000). AgfD, the checkpoint of multicellular and aggregative behaviour in Salmonella typhimureum regulates at least two independent pathways. Mol. Microbiol. 36, 10–23. Rowland, I. R., Bearne, E. A., Fischer, R., and Pool-Zobel, B. L. (1996). The effect of lactulose on DNA damage induced by 1,2-dimethylhydrazine in the colon of human ﬂora associated rats. Nutr. Cancer 26, 37–47.

118

HARRY J. FLINT

Russell, J. B. (1998). Strategies that ruminal bacteria use to handle excess carbohydrate. J. Anim. Sci. 76, 1955–1963. Russell, J. B., and Rychlik, J. L. (2001). Factors that affect rumen microbial ecology. Science 292, 1119–1122. Russell, J. B., and Wallace, R. J. (1997). Energy-yielding and energy-consuming reactions. In ‘‘The Rumen Microbial Ecosystem,’’ ed 2 (P. N. Hobson and C. S. Stewart, eds.), pp. 246–282. Chapman and Hall, London. Saier, M. H. (2000). A functional-phylogenetic classiﬁcation system for transmembrane solute transporters. Microbiol. Mol. Biol. Revs. 64, 354. Salyers, A. A., Vercelotti, J. R., West, S. E. H., and Wilkins, T. D. (1977a). Fermentation of mucin and plant polysaccharides by strains of Bacteroides from the human colon. Appl. Environ. Microbiol. 33, 319–322. Salyers, A. A., West, S. E. H., Vercellotti, J. R., and Wilkins, T. D. (1977b). Fermentation of muicoins and plant polysacchraides by anaerobic bacteria from the human colon. Appl. Environ. Microbiol. 34, 529–533. Schell, M. A., Karmirantzou, M., Snel, B., Vilanova, D., Berger, B., Pessi, G., Zwaheln, M.-C., Desiere, F., Bork, P., Delley, M., Pridmore, R. D., and Arigoni, F. (2002). The genome sequence of Biﬁdobacterium longum reﬂects its adaptation to the human gastrointestinal tract. PNAS 99, 14422–14427. Scheppach, W., Luehrs, H., and Menzel, T. (2001). Beneﬁcial health effects of low digestible carbohydrate consumption. Br. J. Nutr. 85, S23–S30. Schlosser, A., Jantos, J., Hackmann, K., and Schrempf, H. (1999). Characterization of the binding dependent cellobiose and cellotriose transport system of the cellulose degrader Streptomyces reticuli. Appl. Environ. Microbiol. 65, 2636–2643. Scott, H. W., and Dehority, B. A. (1965). Vitamin requirement of several cellulolytic rumen bacteria. J. Bacteriol. 89, 1169–1175. Selinger, L. B., Forsberg, and Cheng, K.-J. (1996). The rumen: A unique source of enzymes for livestock production. Anaerobe 2, 263–284. Shipman, J. A., Hong Cho, K., Siegel, H. A., and Salyers, A. A. (1999). Physiological characterisation of SusG, an outer membrane protein essential for starch utlisation by Bacteroides thetaiotaomicron. J. Bacteriol. 181, 7206–7211. Shipman, J. A., Berleman, J. E., and Salyers, A. A. (2000). Characterisation of four outer membrane proteins involved in binding starch to the cell surface in Bacteroides thetaiotaomicron. J. Bacteriol. 182, 5365–5372. Shpigel, E., Roiz, L., Goren, R., and Shoseyov, O. (1998). Bacterial cellulose-binding domain modulates in vitro elongation of different plant cells. Plant Physiol. 117, 1185–1194. Steenbakkers, P. J. M., Li, X., Ximenes, E. A., Arts, J. G., Chen, H. Z., Ljungdahl, L. G., and Op den Camp, H. J. M. (2001). Noncatalytic docking domains of cellulosomes of anaerobic fungi. J. Bacteriol. 183, 5325–5333. Steer, T., Carpenter, H., Tuohy, K., and Gibson, G. R. (2000). Perspectives on the role of the human gut microbiota and its modulation by pro- and prebiotics. Nutr. Res. Revs. 13, 229–254. Stewart, C. S., Duncan, S. H., and Flint, H. J. (1990). The properties of forms of Ruminococcus ﬂavefaciens which differ in their ability to degrade cotton cellulose. FEMS Microbiol. Letts 72, 47–50. Stewart, C. S., Flint, H. J., and Bryant, M. P. (1997). The rumen bacteria. In ‘‘The Rumen Microbial Ecosystem’’ (P. N. Hobson and C. S. Stewart, eds.), 2nd ed., pp. 10–72. Chapman and Hall, London.

POLYSACCHARIDE BREAKDOWN BY GUT ANAEROBES

119

Suau, A., Bonnet, R., Stutren, M., Godon, J.-J., Gibson, G. R., Collins, M. D., and Dore, J. (1999). Direct analysis of genes encoding 16S rRNA from complex communities reveals many novel molecular species within the human gut. Appl. Environ. Microbiol. 65, 4799–4807. Tajima, K., Aminov, R. I., Nagamine, T., Matsui, H., Nakamura, M., and Benno, Y. (2001). Diet-dependent shifts in the bacterial population of the rumen revealed with real time PCR. Appl. Environ. Microbiol. 67, 2766–2774. Tajima, K., Aminov, R. I., Nagamine, T., Ogata, K., Nakamura, M., Matsui, H., and Benno, Y. (1999). Rumen bacterial diversity as determined by sequence analysis of 16S rDNA libraries. FEMS Microbiol. Ecol. 29, 159–169. Thurston, B., Dawson, K. A., and Strobel, H. J. (1993). Cellobiose versus glucose utilization by the ruminal bacterium Ruminococcus albus. Appl. Environ. Microbiol. 59, 2631–2637. Tomme, P., Boraston, A., McLean, B., Kormos, J., Creagh, A. L., Struch, K., Gilkes, N. R., Haynes, C. A., Warren, R. A. J., and Kilburn, D. G. (1998). Characterization and afﬁnity applications of cellulose-binding domains. J. Chromatography B 715, 283–296. Tomme, P., Warren, R. A. J., Miller, R. C. Jr., Kilburn, D. G., and Gilkes, N. R., (1995). Cellulose binding domains: Classiﬁcation and properties. In Enzymatic Degradation of Insoluble Carbohydrates. Topping, D. L., and Clifton, P. M. (2001). Short chain fatty acids and human colonic function: Roles of resistant starch and nonstarch polysaccharides. Physiol. Revs. 81, 1031–1064. Trinci, A. P. J., Davies, D. R., Gull, K., Lawrence, M. I., Rickers, A., and Theodorou, M. K. (1994). Anaerobic fungi of herbivorous animals. Mycol. Res. 98, 129–152. Tuohy, K. M., Finlay, R. K., Wynne, A. G., and Gibson, G. R. (2001). A human volunteer study on the prebiotic effects of HP-inulin–faecal bacteria enumerated using ﬂuorescent in situ hybridization (FISH). Anaerobe 7, 113–118. van Bielen, J. B., and Li, Z. (2002). Enzyme Technology: An overview. Curr. Op. Biotechnol. 13, 338–344. van der Maarel, M. J. E. C., van der Veen, B., Uitdehaag, J. C. M., Leemhuis, H., and Duikhuizen, L. (2002). Properties and applications of starch-converting enzymes of the alpha-amylase family. J. Biotechnol. 94, 137–155. van Soest, P. J. (1984). ‘‘Nutritional Ecology of the Ruminant’’ 2nd ed. Cornell Univ. Press. Varel, V. H., and Yen, J. T. (1997). Microbial perspective in ﬁber utilization by swine. J. Anim Sci. 75, 2715–2722. Wang, R. F., Cao, W. W., and Cerniglia, C. E. (1997). PCR detection of Ruminococcus spp. in human and animal feces. Mol. Cell. Probes 11, 259–265. Wang, X., Conway, P. L., Brown, I. L., and Evans, A. J. (1999). in vitro utilization of amylopectin and high amylose maize (amylomaize). starch granules by human colonic bacteria. Appl. Environ. Microbiol. 65, 4848–4854. Wang, Y., and McAllister, T. A. (2002). Rumen microbes, enzymes and feed digestion—a review. Asian Austr. J. Anim Sci. 15, 1659–1676. Warren, R. A. J. (1996). Microbial hydrolysis of polysaccharides. Ann. Rev. Microbiol. 50, 183–212. Watanabe, H., Noda, H., Tokuda, G., and Lo, N. (1998). A cellulase gene of termite origin. Nature 394, 330–331. Weaver, J., Whitehead, T. R., Cotta, M. A., Valentine, P. C., and Salyers, A. A. (1992). Genetic analysis of a locus on the Bacteroides ovatus chromosome which contains xylan utilization genes. Appl. Environ. Microbiol. 58, 2764–2770.

120

HARRY J. FLINT

Weimer, P. J., Waghorn, G. C., Odt, C. L., and Mertens, D. R. (1999). Effect of diet on populations of three species of ruminal cellulolytic bacteria in lactating dairy cows. J. Dairy Sci. 82, 122–134. Wells, J. E., Russell, J. B., Shi, Y., and Weimer, P. J. (1995). Cellodextrin efﬂux by the cellulolytic ruminal bacterium Fibrobacter succinogenes and its potential role in the growth of non-adherent bacteria. Appl. Environ. Microbiol. 61, 1757–1762. Whitford, M. F., Forster, R. J., Beard, C. E., Gong, J., and Teather, R. M. (1998). Phylogenetic analysis of rumen bacteria by comparative sequence analysis of cloned 16S rDNA libraries. Anaerobe 4, 153–163. Wolin, M. J., Miller, T. J., and Stewart, C. S. (1997). Microbe-microbe interactions. In ‘‘The Rumen Microbial Ecosystem’’ (P. N. Hobson and C. S. Stewart, eds.), 2nd ed., pp. 477–491. Chapman and Hall, London. Wolin, M. J., Miller, T. J., Yerry, S., Zhang, Y. C., Bank, S., and Weaver, G. A. (1999). Changes in fermentation pattern of fecal microbial communities associated with a drug treatment that increases dietary starch in the human colon. Appl. Environ. Microbiol. 65, 2807–2812. Wollowski, I., Rechkemmer, G., and Pool-Zobel, B. L. (2000). Protective role of probiotics and prebiotics in colon cancer. Am. J. Clin. Nutr. 73, 451S–455S. Wood, J., Scott, K. P., Avgustin, G., Newbold, C. J., and Flint, H. J. (1998). Estimation of the relative abundance of different Bacteroides and Prevotella ribotypes in gut samples by restriction enzyme proﬁling of PCR-ampliﬁed 16S rRNA gene sequence. Appl. Environ. Microbiol. 64, 3683–3689. Wood, T. M., Wilson, C. A., McCrae, S. I., and Joblin, K. N. (1986). A highly active extracellular cellulase from the anaerobic rumen fungus Neocallimastix frontalis. FEMS Microbiol. Lett. 34, 37–40. Xu, B. Z., Janson, J. C., and Sellos, D. (2001). Cloning and sequencing of a molluscan endo-beta-1,4-glucanase gene from the blue mussel, Mytilus edulis. Eur. J. Biochem. 268, 3718–3727. Xu, J., Bjursell, M. K., Himrod, J., Deng, S., Carmichael, L. K., Chiang, H. C., Hooper, L. V., and Gordon, J. I. (2003). A genomic view of the human-Bacteroides thetaiotaomicron symbiosis. Science 299(5615), 2074–2076. Zhang, J.-X., and Flint, H. J. (1992). A bifunctional xylanase encoded by the xynA gene of the rumen cellulolytic bacterium Ruminococcus ﬂavefaciens 17 comprises two dissimilar domains linked by an asparagine/glutamine rich sequence. Mol. Microbiol. 6, 1013–1023. Zoetendal, E. G., Akkermans, A. D. L., and de Vos, W. M. (1998). Temperature gradient gel electrophoresis analysis of 16S rRNA from human fecal smples reveals stable and host speciﬁc communities of active bacteria. Appl. Environ. Microbiol. 64, 3854–3859.

Lincosamides: Chemical Structure, Biosynthesis, Mechanism of Action, Resistance, and Applications JAROSLAV SPI´ZˇEK,* JITKA NOVOTNA´,

AND

ˇ EZANKA TOMA´Sˇ R

Institute of Microbiology Academy of Sciences of the Czech Republic 142 20 Prague 4, Czech Republic *Author for correspondence. E-mail: [email protected]

I. Introduction II. Chemical Structure of Lincosamides and Cultivation of Production Strains III. Lincomycin Biosynthetic Pathway A. Biosynthesis of the Amino Acid Moiety B. Biosynthesis of the Sugar Moiety C. Condensation and Final Modification IV. Genetic Control of Lincomycin Biosynthesis V. Mechanism of Action VI. Resistance Against Lincosamides VII. Biological Activity and Applications VIII. Gram-Positive Bacteria IX. Gram-Negative Bacteria X. Anaerobic Bacteria XI. Protozoa and Other Organisms XII. Conclusion and Future Prospects References

121 124 130 130 130 133 133 135 137 138 139 141 143 144 145 146

I. Introduction Lincosamides constitute a relatively small group of antibiotics with a chemical structure consisting of amino acid and sugar moieties. The naturally occurring members of the group are lincomycin and celesticetin, the latter exhibiting 5% of the biological activity of lincomycin in vivo. Many semi-synthetic derivatives of lincomycin have been prepared. Of these, only the chlorinated derivative clindamycin is highly biologically active and is applied practically. Natural lincosamides are produced by several Streptomyces species, mainly by Streptomyces lincolnensis, S. roseolus, and S. caelestis and by Micromonospora halophytica. Their mechanism of action is via inhibition of protein synthesis in sensitive microorganisms. Lincosamides have an unusual antimicrobial spectrum, being active against only Gram-positive and not Gram-negative aerobic bacteria but widely and potently active against anaerobic bacteria and some protozoa. 121 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

TABLE I STRUCTURE OF LINCOMYCIN AND RELATED ANTIBIOTICS Name

R1

R2

R3

R4

R5

122

1

Lincomycin A

SCH3

CH3

CH2CH2CH3

OH

H

4

Lincomycin B

SCH3

CH3

CH2CH3

OH

H

5

Lincomycin C

SCH2CH3

CH3

CH2CH2CH3

OH

H

7

Lincomycin D

SCH3

H

CH2CH2CH3

OH

H

6

Lincomycin S

SCH2CH3

CH2CH3

CH2CH2CH3

OH

H

8

Lincomycin K

SCH2CH3

H

CH2CH2CH3

OH

H

10

Lincomycin sulfoxide

CH3

CH2CH2CH3

OH

H

11

1-Demethylthio-1-hydroxylincomycin

CH3

CH2CH2CH3

OH

H

2

Celesticetin A

CH3

H

OCH3

H

14

Celesticetin B

CH3

H

OCH3

H

15

Celesticetin C

CH3

H

OCH3

H

16

Celesticetin D

SCH2CH2OOCCH3

CH3

H

OCH3

H

3

Desalicetin

SCH2CH2OH

CH3

H

OCH3

H

17

N-Demethylcelesticetin

H

H

OCH3

H

OH

SCH2CH2OOCCH2CH(CH3)2

18

Desalicetinsalicylate

9

Clindamycin

12

Clindamycin sulfoxide

13

10 -Demethylclindamycin

SCH3

SCH3

CH3

H

OCH3

H

CH3

CH2CH2CH3

H

Cl

CH3

CH2CH2CH3

H

Cl

H

CH2CH2CH3

H

Cl

123

124

SPI´ZˇEK et al.

The major route of resistance to lincosamides is modiﬁcation of the 23S rRNA in the 50S ribosomal subunit, similarly to resistance against macrolides and streptogramin B (MLSb resistance). Clindamycin is the main lincosamide antibiotic that is applied in clinical practice. II. Chemical Structure of Lincosamides and Cultivation of Production Strains The chemical structure of lincomycin was investigated by Hoeksema et al. (1964), employing both classical chemical degradation and nuclear magnetic resonance. These studies resulted in the chemical structure of lincomycin (1), (Table I). It consists of an unusual amino acid, viz. trans-N-methyl-4-n-L-proline (propylhygric acid), linked by a peptide bond with the sugar 6-amino-6, 8-dideoxy-1-thioD-erythro--D-galactopyranoside (methylthiolincosamide). Details of the chemical characterization and nuclear magnetic resonance studies of lincomycin and its degradation products were subsequently described (Herr and Slomp, 1967; Magerlein, 1971; Schroeder et al., 1967; Slomp and MacKellar, 1967). The electron-impact mass spectrum and the chemical-ionization mass spectrum of lincomycin were reported later (Horton et al., 1974; Kagan and Grostic, 1972). Lincomycin belongs to a novel class of antibiotics characterized by an alkyl 6-amino-6,8-dideoxy-1-thio-D-erythro--D-galacto-octopyranoside joined with a proline moiety by an amide linkage. The availability of this structure presented an excellent opportunity for chemical and microbiological modiﬁcation of lincomycin. Before a review of the other naturally occurring members of the lincomycin family of antibiotics, processes used for the cultivation of producing strains (Table II) and methods for separation of lincomycin A will be described. The ﬁrst patent describing lincomycin production was registered by Upjohn (Bergy et al., 1963). S. lincolnensis var. lincolnensis was cultivated in a 380-liter fermentor at 28 C for 5 days, and 210 liters of cultivation broth were harvested to give approximately 105 g of dried lincomycin A at a purity of 232 biounits/mg. The actinomycete S. vellosus var. vellosus isolated from Arizona soil was used according to Bergy et al. (1981) for the production of lincomycin A without the concomitant production of lincomycin B (see subsequent discussion). The novel actinomycete described in the patent of Argoudelis et al. (1972) produced only lincomycin A. S. espinosus produced 50 g/ml of the desirable lincomycin A during a 4-h cultivation at 28 C in

125

LINCOSAMIDES TABLE II STRAINS PRODUCING LINCOMYCIN GROUP OF ANTIBIOTICS Organisms

Literature

S. lincolnensis 78–11

Peschke et al. (1995), Zhang et al. (1992)

S. lincolnensis NRRL 2936

Bergy et al. (1963), Brahme et al. (1984a), Peschke et al. (1995)

S. lincolnensis RIA 1246

Neusser et al. (1998)

S. lincolnensis UC 5124

Chung and Crose (1990)

S. espinosus NRRL 5729

Peschke et al. (1995)

S. espinosus NRRL 3890

Peschke et al. (1995)

S. espinosus

Argoudelis et al. (1972)

S. espinosus

Reusser and Argoudelis (1974)

S. pseudogriseolus NRRL 3985

Peschke et al. (1995)

S. variabilis var. liniabilis

Argoudelis and Coats (1974)

S. vellosus NRRL 8037

Bergy et al. (1963), Peschke et al. (1995)

shaken ﬂasks by using a classical medium. According to the U.S. patent (Argoudelis and Coats, 1973), S. pseudogriseolus var. linmyceticus cultivated for 2 days at 28 C produces 22 g/ml of lincomycin. Another patent on production of lincomycin A only (Argoudelis and Coats, 1974), without contamination with lincomycin B, was registered by The Upjohn Company. Streptomyces variabilis var. liniabilis produces 26 g/ml of lincomycin A after the 120-h cultivation at 28 C in shaken 500-ml Erlenmeyer ﬂasks. The strain S. lincolnensis used by Zhang (1993) is a phageresistant, lincomycin-overproducing mutant synthesizing approximately 2.5 g of lincomycin A per liter. In the last 5 years, Chinese investigators have become interested in the cultivation of lincomycin-producing strains, especially in the isolation of lincomycin with a simultaneous decrease or complete removal of the undesirable lincomycin B (Dong and Dong, 2001; Li et al., 2001; Wang and Qi, 1999, 2000, 2001). In addition to lincomycin, a very closely related antibiotic was isolated from fermentations of Streptomyces lincolnensis var. lincolnensis. Structural studies indicated this compound to be lincomycin B (i.e., 40 -depropyl-4-ethyllincomycin (2). It had already been

126

SPI´ZˇEK et al.

described in 1965 as a minority component in S. lincolnensis cultures. It exhibits only a 25% antibiotic activity as compared with lincomycin A. As shown in Table I, lincomycin B contains one fewer methylene group in the side chain of propylhygric acid. Several patents (Reusser and Argoudelis, 1974; Visser, 1972; Witz, 1972) were published in which procedures were described to minimize the amount of lincomycin B during the fermentation. The addition of propylproline to the cultivation medium was found to decrease the amount of lincomycin B. It was assumed that propylproline is a likely precursor of propylhygric acid in lincomycin A, whereas ethylproline is a tentative precursor of ethylhygric acid of lincomycin B. Aromatic amino acids, primarily L-tyrosine, L-dihydroxyphenylalanine, and other compounds similar to tyrosine, are by S. lincolnensis incorporated into lincomycin A. It was found that L-tyrosine or L-dihydroxyphenylalanine added to the culture medium induces accumulation of propylproline and also, although to a lesser extent, that of ethylproline. The mechanism by which the addition of propylproline or L-tyrosine and similar compounds inﬂuences the biosynthesis of lincomycin has not yet been clariﬁed; however, the results obtained so far indicate that N-demethyllincomycin synthetase, the enzyme catalyzing the linkage between the amino acid and sugar moiety of lincomycin, preferentially utilizes propylproline. Lincomycin A was the ﬁrst member of its family whose structure had been completely elucidated; however; celesticetin, a related antibiotic, was reported earlier in the culture broth of Streptomyces caelestis, a new actinomycete species. This antibiotic was also isolated by Hoeksema et al. (1955). Employing some of the techniques used in the structural studies on lincomycin, Hoeksema (1968) assigned structure 3 to celesticetin. Desalicetin, the alkaline hydrolysis product of celesticetin, was shown to possess structure 4. Celesticetin and desalicetin proved to be less effective than lincomycin against a number of microorganisms, both in vitro and in vivo (Mason and Lewis, 1964). Celesticetin exhibits a broad antibacterial spectrum, particularly against Gram-positive bacteria. Other new antibiotics, viz. N-demethyl7-O-demethylcelesticetin and N-demethylcelesticetin, were found to be produced in cultures of a strain of S. caelestis. The introduction of various chemicals to the fermentation of S. lincolnensis var. lincolnensis was found to induce the formation of lincomycin-related antibiotics. Thus the addition of DL-ethionine led to the formation of 1-demethylthio-1-ethylthiolincomycin (5) (Argoudelis and Mason, 1965) and 1-demethyl-1-demethylthio-10 -

LINCOSAMIDES

127

ethyl-1-ethylthiolincomycin (6) (Argoudelis et al., 1970); methylthiolincosaminide induced formation of 10 -demethyllincomycin (7); the addition of ethyl thiolinconsaminide yielded 10 -demethyl-1demethylthio-1-ethylthiolincomycin (8). Sulfonamides, sulfanilamide in particular, inhibit N-methylation rather selectively, resulting in the production of 10 -demethyllincomycin (7). Chemical modiﬁcation (Argoudelis and Stroman, 1984; Birkenmeyer, 1981, 1982a,b; Patt et al., 1984) of lincomycin derivatives consisted mainly in substitution of hydrogen atoms bound to a heteroatom. A number of lincomycin esters, with either organic (from acetate to stearate) or inorganic (phosphoric, carbonic) acids or of lincomycin alkyl derivatives (ethers), were prepared. Substitutions were performed primarily in positions 2, 3, 4, and 7. Salts of lincomycin with inorganic acids (i.e., hydrochlorides and derivatives of sulfamic acid) were also synthesized. Replacement of hydroxyl in the side chain (i.e., C-7 of octose) is another modiﬁcation of lincomycin. The commercially used derivative with the generic name clindamycin (9) was produced by introduction of chlorine. Additional thio-analogs, of which, for example, the C-1 -anomer exhibits only one tenth of the activity of the natural -anomer, were synthesized. The changed conﬁguration of 1-methyl-4-proline is also very important. It was demonstrated that the derivative containing the D amino acid exhibits only 50% of the antibacterial activity. Although hundreds of lincomycin derivatives were prepared, among them derivatives produced totally by chemical synthesis, clindamycin is the only drug that has been used successfully in clinical practice. Detailed synthetic procedures leading to numbers of lincomycin derivatives and their biological activity were described by Magerlein (1971). A new and stereoselective route to the aminoglycoside components of the antibiotics lincomycin and clindamycin has been described. The key step involves diastereoselective introduction of the amino group at C-6 of D-galactose by (3,3)-sigmatropic rearrangements of the corresponding allylic (Z)-triﬂuoroacetimidate and (Z)- and (E)-allylic thiocyanates. Epoxidation of the resulting triﬂuoroacetamide with m-chloroperbenzoic acid led to the epoxide with a high threo-selectivity (Gonda et al., 2000). In another paper (Bowden and Stevens, 2000) a novel synthesis of clindamycin from lincomycin by using N-chlorosuccinimide and triphenylphosphine was reported, resulting in high yields and avoiding the use of tetrachloromethane employed in the currently used manufacturing process. Microbial modiﬁcation of lincomycin and clindamycin does not lead to the formation of the number and diversity of transformation

128

SPI´ZˇEK et al.

products obtained by chemical modiﬁcation. However, several types of interesting transformations have been reported. Thus, for instance, lincomycin was glycosylated by using jack bean -mannosidase to produce 7-O--D-mannopyranosyl-lincomycin (Weignerova et al., 2001). The addition of lincomycin to a fermentation of S. lincolnensis var. lincolnensis resulted in the formation of lincomycin sulfoxide (10). This sulfoxide was also prepared by oxidation of lincomycin hydrochloride with hydrogen peroxide (Pospisil et al., 2001). As compared with lincomycin, it exhibits only a 0.01 activity against Sarcina lutea. A lower yield of 1-demethylthio-l-hydroxylincomycin (11) was also isolated from the fermentation broth. Similarly, clindamycin sulfoxide (12) was the major transformation product when clindamycin was added to fermentations of S. armentorus. Trace amounts of sulfoxide were detected in similar fermentations of other Streptomyces species, particularly S. punipalus. Clindamycin sulfoxide is about equally active as lincomycin against Sarcina lutea. In addition to sulfoxide formation, several species of streptomycetes, including those mentioned above, perform a partial 1-demethylation (Argoudelis et al., 1969). S. punipalus was found to demethylate clindamycin (9) to 10 -demethylclindamycin (13), which had previously been prepared by chemical modiﬁcation. Oxidation of lincomycin with H2O2 in alkaline media leads to N-oxides and the conversion of thiomethyl group to sulfoxides and sulfones. NH4OH favors formation of the S-isomer; both R- and S-isomers of N-oxide form in the presence of NaOH (Pospisil et al., 2004). Argoudelis and Coats (1969) observed that S. rochei grown in a synthetic medium converted lincomycin and lincomycin-related antibiotics to their 3-phosphate esters. Lincomycin 3-phosphate was inactive in vitro against several organisms. In addition to ‘‘classical methods’’ for the analysis of lincomycin and clindamycin, in the last 20 years methods of instrumental analysis based on liquid chromatography (LC) as a separation method and soft ionization techniques of mass spectrometry as an identiﬁcation method were developed. Reverse-phase (RP) high-performance liquid chromatography (HPLC) for preparation of lincomycin A was patented (Hofstetter, 1982). This process yielded a highly puriﬁed lincomycin A with less than 0.5% lincomycin B. Lincomycin salvage mother liquor, containing about 30% lincomycin B, was used as the starting material for preparation of lincomycin B.

LINCOSAMIDES

129

Procedures for the determination of clindamycin in human plasma using HPLC with ultraviolet (UV) detection have been described (Fieger-Bu¨schges et al., 1999; La Follette et al., 1988; Liu et al., 1997). These methods are either not very sensitive or are time-consuming and require laborious sample preparation. More recently, methods using HPLC coupled with MS detection have been published for the quantiﬁcation of clindamycin in human plasma. Yu et al. (1999) used an LC-MS/MS/ESI method, with acceptable linearity in the 0.05–20 l/ml range and a quantiﬁcation limit of 0.050 g/ml. Martens-Lobenhoffer and Banditt (2001) used an LC-MS/ APCI method for both human plasma and bone. The method showed good linearity in the 0.1–4.0 g/ml range for plasma with a limit of quantiﬁcation of 0.1 g/ml. A method for the quantiﬁcation of clindamycin in animal plasma by using LC-MS/MS/ESI has also been published (Cherlet et al., 2002). Lincomycin was used as the internal standard. Good linearity was observed within the range of 0–10 g/ml. The limit of quantiﬁcation of the method is 50 g/ml, and the detection limit is 1.3 ng/ml. The method was used for pharmacokinetic studies of clindamycin formulations in dogs. Lincomycin and related antibiotics were analyzed by an MS/MS/CID technique in bovine milk extract. Lincomycin gave a limit of detection of 0.83 pg on-column (Crellin et al., 2003). For a further analysis of lincomycin and spectinomycin, an RP ionpair LC method with a base-deactivated column and pulsed electrochemical detection by means of a gold electrode was used (Szunyog et al., 2002). Capillary electrophoresis was used for a simultaneous determination of ﬁve aminoglycoside antibiotics (netilmicin, tobramycin, lincomycin, kanamycin, and amikacin). Under the optimum separation conditions, the aminoglycoside antibiotics were baseline separated within 20 min and the detection limit was below 6.7 M for lincomycin (Yang et al., 2001). HPLC-integrated pulsed amperometric detection was found to be a good primary detection method to complement or replace the absorbance detection method in the case of nonchromophoric sulfurcontaining antibiotics and their sulfur-containing impurities and decomposition products (Hanko et al., 2001). A thin-layer chromatography/densitometric method was developed for the identiﬁcation and quantitation of oxytetracycline, tiamulin, lincomycin, and spectinomycin in veterinary preparations (Krzek et al., 2000).

130

SPI´ZˇEK et al.

III. Lincomycin Biosynthetic Pathway The lincomycin biosynthetic pathway proceeds via a heterogeneously rooting biphasic pathway, giving rise to propylproline and methylthiolincosamide (Fig. 1). These basic precursors become condensed to N-demethyllincomycin, which is ﬁnally methylated to yield lincomycin. A. BIOSYNTHESIS OF THE AMINO ACID MOIETY Tyrosine seems to be the principal precursor of the amino acid moiety. Feeding studies using L-[1-14C]tyrosine and L-[U-14C]tyrosine performed by Witz et al. (1971) suggested that seven of the nine carbons in the propylhygric acid come from tyrosine. The other two carbons, viz. the N-methyl and terminal aliphatic side chain methyl groups, come from S-adenosylmethionine (Argoudelis et al., 1969; Brahme et al., 1984a,b; Witz et al., 1971). Biosynthetic experiments with D-(13C6) glucose suggest that glucose is converted via glycolysis and the hexose monophosphate shunt to phosphoenolpyruvate and erythrose-4-phosphate, respectively, which are in turn converted via the shikimic acid pathway to tyrosine and then to dihydroxyphenylalanine. The pathway probably continues through the 2,3-extradiol cleavage of the aromatic ring of dihydroxyphenylalanine, followed by condensation to form a pyrrolo ring (Brahme, 1984a). A similar biosynthetic route leading from tyrosine through dihydroxyphenylalanine and pyrrolo ring formation lincomycin probably shares with the pyrrolo[1,4]benzodiazepine antibiotics anthramycin, sibiromycin, and tomaymycin (Hurley, 1980; Hurley et al., 1979). The multistep conversion of dihydroxyphenylalanine to the propylproline remains unknown, but it appears to lead to 1,2,3,6-tetradehydro-propylproline, which was shown to accumulate in mutants lacking a reductase that requires lincomycin cosynthetic factor (Kuo et al., 1992) identiﬁed as 7,8-didemethyl-8-hydroxy-5deazariboﬂavin (Kuo et al., 1989). Based on this ﬁnding, Kuo and coworkers modiﬁed the scheme of Brahme et al. (1984a,b) and postulated the remaining steps leading to propylproline, as shown in Fig. 1. B. BIOSYNTHESIS OF THE SUGAR MOIETY No intermediates of the aminooctose moiety methylthiolincosamide biosynthesis have so far been identiﬁed.

LINCOSAMIDES

131

FIG. 1. Hypothetical biosynthetic pathway for lincomycin A. (A) Amino acid branch. (B) Sugar branch. Two alternative routes are proposed for methylthiolincosamide biosynthesis. (B1) From intermediates of the pentose phosphate cycle via formation of a C8 sugar precursor. (B2) From dTDP-glucose via a 6-deoxyhexose pathway and a later extension of the carbon chain. The possible involvement of three of the gene products of the putative methylthio-lincosamide biosynthetic genes is indicated. (C) Condensation and ﬁnal methylation.

132

SPI´ZˇEK et al.

A likely pathway leading from glucose to methylthiolincosamide was proposed by Brahme et al. (1984b), who studied the metabolic origin of methythiolincosamide by analyzing 13C-13C spin coupling patterns in methythiolincosamide and lincomycin-derived from biosynthetic experiments with D-(13C6)glucose and the speciﬁc 13C enrichments in methylthiolincosamide synthesized from speciﬁcally labeled substrates. The data obtained by means of both these approaches indicated that the C8 carbon skeleton of methythiolincosamide arises from condensation of a pentose unit (C5) and a C3 unit. The pentose unit could either be derived from glucose through the hexose monophosphate shunt as an intact unit or result from condensation of glyceraldehyde-3-phosphate with a C2 donor such as sedoheptulose7-phosphate via transketolase reaction. The C3 unit probably arises from a suitable donor molecule such as sedoheptulose-7-phosphate and is added via a transaldolase reaction. According to the C3 unit donor origin, the unit could consist either of an intact C3 unit or a C2 unit combined with a C1 unit. The C3 and C5 units are then condensed, giving rise to octose, which is then converted to methylthiolincosamide. The ﬁnal conversion of the C8-skeleton to methylthiolincosamide was predicted to involve isomerization of the octulose to octose, dephosphorylation and reduction of the C8 carbon, transamination of the precursor 6-ketooctose, and ﬁnal thiomethylation of the C1 carbon. First attempts to ﬁnd out the metabolic origin of the sugar moiety were by means of a combination of radioactive labeling and mass spectroscopy. It was determined that the S-methyl group of the methylthiolincosamide subunit is derived from C1 fragments at the oxidation state of methyl groups. Sequencing of the lincomycin gene cluster, and amino acid sequence analysis of the putative protein products, led Peschke et al. (1995) to modify the aforementioned design of the biosynthetic pathway. Eight genes, lmbL through lmbQ, form a subcluster, which probably codes for a set of enzymes involved in sugar metabolism (Fig. 1). Putative proteins LmbO, LmbM, and LmbS show similarity to enzymes involved in the central steps of many NDP-6-deoxyhexose pathways, including sugar-activating pyrophosphorylases (LmbO), NDP-hexose dehydratases (LmbM), and (NDP-) ketosugar (or cyclitol) aminotransferases/ dehydratases (LmbS). The corresponding genes are found mostly in clusters in both Gram-positive and Gram-negative bacteria, and their protein products are involved in the formation of other actinomycete secondary metabolites (Piepersberg, 1994; Pissowotzki et al., 1991; Retzlaff et al., 1993; Stockmann and Piepersberg, 1992) or of extracellular polysaccharides (Thorson et al., 1993).

LINCOSAMIDES

133

The genetic record led Peschke et al. (1995) to the conclusion that methylthiolincosamide biosynthesis involves a nucleotide (probably dTDP) activation step and a series of modiﬁcation steps on dNTPactivated sugar intermediates. Thus, two totally different pathways, starting from either D-glucose or NDP-activation and modiﬁcation of a C8 sugar intermediate, both represent routes leading to the production of methylthiolincosamide. The authors assume that the 13C-labelling pattern in methylthiolincosamide published by Brahme et al. (1984b), suggesting that an octulose-phosphate intermediate is formed ﬁrst could support glucose as a starter unit if there is rapid and efﬁcient equilibrium in the hexose phosphate pool mediated by a hexose monophosphate shunt during the production phase. This type of rearrangement of labelling has been described for other streptomycete products directly derived from glucose (Rinehart et al., 1992). The detection of a gene (lmbR) encoding a transaldolase-like enzyme in the ‘‘sugar subcluster’’ could support an alternative route via a pyranosidic octose intermediate that is NDP-activated and further modiﬁed in this form. Finally, a thiomethyl unit at the C1 position of the postulated (NDP-)6-amino-6,8-deoxyoctose intermediate would be added, which could be transferred from 50 -thiomethyladenosine, a side product of polyamine biosynthesis from S-adenosyl-methionine (Piepersberg and Distler, 1997). C. CONDENSATION AND FINAL MODIFICATION Formation of the amide bond between the carboxyl group of the aglycone propylproline and the amino group of the methylthiolincosamide yielding N-demethyllincomycin is apparently the penultimate step in lincomycin biosynthesis. N-demethyllincomycin-synthetase, catalyzing this reaction, is a complex of readily dissociable subunits with nonidentical molecular weights (Hausknecht and Wolf, 1986). The ﬁnal step in the pathway is N-methylation of N-demethyllincomycin, catalyzed by S-adenosylmethionine: N-demethyllincomycin methyl transferase (Patt and Horvath, 1985). IV. Genetic Control of Lincomycin Biosynthesis Similarly to other antibiotics (Martin and Liras, 1989) produced by actinomycetes, genes coding for lincomycin biosynthesis are clustered together in a single genomic region and closely linked to the corresponding resistance determinants (Chung and Crose, 1990; Wovcha et al., 1986; Zhang et al., 1992).

134

SPI´ZˇEK et al.

FIG. 2. The lincomycin gene cluster. Grey arrows designated A-Z indicate putative biosynthetic genes and black arrows designated rA-rC indicate genes coding for functions that impart lincomycin resistance phenotype.

The 35-kpb-long chromosomal region bearing a lincomycin-production gene cluster from the overproducing industrial strain Streptomyces lincolnensis 78–11 was cloned and sequenced by Peschke et al. (1995). The cluster contains 27 open reading frames with putative biosynthetic or regulatory functions (lmb genes) and three resistance (lmr) genes (Fig. 2). The genes designated lmrA and lmr C ﬂank the cluster and appear to code for proteins probably involved in lincomycin export (Peschke et al., 1995; Zhang et al., 1992). The lmrB gene codes for a protein very similar to several 23S RNA methyltransferases (Zhang et al., 1992). Four other lincomycin producers probably share the overall cluster organization; however, the clusters are embedded in nonhomologous chromosomal surroundings. DNA sequence analysis suggests a complicated transcription pattern with at least 12 possible transcription units. On the basis of experiments using transposon 4560-mediated mutagenesis of the lincomycin production cluster of the strain derived from Streptomyces lincolnensis UC 5124, it is assumed that genes coding for enzymes involved in propylproline synthesis are located close to lmrA gene, whereas those controlling methylthiolincosamide synthesis are located further from it (Chung and Crose, 1990). Thus, biosynthetic genes seem to be organized according to functional pathways within the cluster region; however, the same experiments suggest that genes coding for subunits of the NDL synthetase, the key enzyme catalyzing propylproline and methylthiolincosamide condensation, are located at three widely separated loci (Chung et al., 1997). The functions coded for by only a few of the aforementioned open reading frames were demonstrated experimentally. In addition to resistance genes, only the function of the genes lmbB1 and lmbB2, coding for enzymes converting L-tyrosine and L-dihydroxyphenylalanine, was clariﬁed (Neusser et al., 1998). Gene lmbJ apparently codes for a speciﬁc N-demethyllincomycin methyltransferase (Janata et al., 1999). Functions of other proteins coded for by lincomycin cluster genes can only be deduced.

LINCOSAMIDES

135

V. Mechanism of Action Lincosamides belong to antibiotics that block microbial protein synthesis. Protein synthesis includes a large number of steps from activation of amino acid monomers by aminocayl-tRNA synthetases to many steps of chain initiation, elongation, and chain termination of the grown polypeptides on the ribosome. Antibiotics interrupt the timing and speciﬁcity of any of these steps, and such disruptions decelerate the growth or are lethal to the microorganism. A molecular mechanism by which clindamycin inhibits ribosomal protein biosynthesis in prokaryotic microorganisms has therefore been unclear. However, recently it was shown that clindamycin’s three-dimensional structure closely resembles the L-Pro-Met and the D-ribosyl ring of adenosine (Fitzhugh, 1998) biomolecules, which occur proximate to one another at the 30 -ends of L-Pro-Met-tRNA and deacylated-tRNA for a brief interval following the formation of a peptide bond between L-Pro-tRNA and L-Met-tRNA. This ﬁnding strongly suggests that clindamycin and other lincosamides act as structural analogs of the 30 -ends of L-Pro-Met-tRNA and deacylated-tRNA as they are positioned during an initial phase of pretranslocation in the peptide elongation cycle. Clindamycin thus constitutes a structural analog of an intermediate state in protein biosynthesis. Although the chemical structure of macrolides (e.g., erythromycin), lincosamides (e.g., lincomycin, clindamycin, and celesticetin), and streptogramins is very different, their mechanism of action is similar. Erythromycin binding with the 23S rRNA blocks polypeptide translation, resulting in a release of peptidyl-tRNA intermediates prematurely by blocking the approach to the elongating peptide’s exit tunnel (Schlunzen et al., 2001). Erythromycin also blocks assembly of 50S subunits, probably through its interaction with 23S rRNA. Although the macrolides in general do not directly block the peptide bond-forming step at the peptidyltransferase center of the 50S subunits, it is known that they compete with lincosamide antibiotics that are direct peptidyltransferase inhibitors. A point mutation at A2058 of 23S rRNA induces a triple resistance to macrolides, lincosamides, and streptogramin B. Schlunzen et al. (2001) provided a direct proof with the cocrystal structure of the lincosamide antibiotic clindamycin, in which the 20 - and 30 -hydroxyl groups of the antibiotic sugar moiety form hydrogen bonds with the same exocyclic N6 amino group of A2058. Clindamycin and erythromycin binding show a partial physical overlap. Clindamycin has separately been known to interact with both the A site and the P site of the peptidyltransferase center. In accordance with their mechanism of action, bacteria quite often

136

SPI´ZˇEK et al.

develop cross-resistance to macrolides, lincosamides, and streptogramin B. Lincomycin and clindamycin share a common mechanism in sensitive microorganisms and also exhibit a similar antibacterial spectrum. Nevertheless, slight differences in their antimicrobial activity exist as clindamycin also affects some protozoa—for example, Toxoplasma gondii, Plasmodium falciparum, and Pneumocystis carinii. Administration of 450–600 mg of clindamycin every 6 h and 15–30 mg of primaquine base applied once a day were used successfully for the treatment of infections caused by Pneumocystis carinii (Fishman, 1998), whereas lincomycin did not exhibit any activity. Remington (1990) demonstrated that a combination of clindamycin with primaquine is highly effective in the treatment of pneumonia caused by Pneumocystis carinii, even in AIDS patients. Clindamycin inhibits bacterial protein synthesis and acts speciﬁcally on the 50S subunit of the bacterial ribosome, most likely by affecting the process of peptide chain initiation. It may also stimulate dissociation of peptidyl-tRNA from ribosomes (Menninger and Coleman, 1993). Escherichia coli exposed to subminimal inhibitory concentrations of clindamycin shows a decreased adherence to buccal mucosal cells, which may be due to protein synthesis inhibition. Presumably it was suppression of protein synthesis that also enabled Sanders et al. (1983) to show that clindamycin is an effective in vitro inhibitor of the derepression of bacterial -lactamases that would otherwise have been produced by certain nonfastidious Gram-negative bacilli exposed to various -lactam antibiotics. Similarly, Schlievert and Kelly (1984) showed inhibition of toxin production in toxic shock syndrome-producing strains of Staphylococcus aureus by concentrations of clindamycin that do not inhibit bacterial growth. Macrolides and lincosamides are ﬁrst-choice bacteriostatic antibiotics used in veterinary dermatology. The main antibiotics of these classes are erythromycin, lincomycin, clindamycin, and tylosin. They are well absorbed if administered orally and are able to penetrate well into infected skin. Their spectrum of action comprises bacteria commonly associated with skin infections, including staphylococci. Their main disadvantage is a rapid development of bacterial resistance and occasional gastrointestinal upset (Noli and Boothe, 1999). Macrolides, ﬂuoroquinolones, rifamycins, tetracyclines, trimethoprimsulfamethoxazole, and clindamycin were described as antimicrobial agents of preference for the dermatologist (Epstein et al., 1997). Clindamycin administration results in changes in intestinal microﬂora. Numbers of enterococcal species increase and those of all anaerobes decrease (Nord and Heimdahl, 1986).

LINCOSAMIDES

137

Some antibiotics also exhibit immunomodulatory effects, clindamycin among them (VanVlem et al., 1996). VI. Resistance Against Lincosamides In general, basic mechanisms of antibiotic resistance include microbial cell impermeability, target site modiﬁcation, and enzymatic modiﬁcation or destruction of the antibiotic and its increased efﬂux. The main type of resistance to lincomycin and clindamycin is the socalled MLSb resistance that renders sensitive microorganisms resistant to macrolides, lincosamides, and streptogramin B (hence MLSb resistance). It is monomethylation or dimethylation of the N6 exocyclic amino group of A2058 by speciﬁc ribosome methylation modiﬁcation enzymes. This type of resistance is associated with genes encoding methyltransferases modifying the common target site of macrolides and lincosamides (i.e., 23S ribosomal RNA—e.g. genes ermA and ermC). A speciﬁc gene was also described whose protein product modiﬁes and thus inactivates lincosamide antibiotics (linA). Increased efﬂux of lincosamides was detected in some microorganisms resistant to them. The occurrence of some genes involved in MLSB resistance in methicillin-resistant strains of Staphylocococcus aureus was investigated, and it was shown that, in the Czech Republic, genes ermC and ermA are the most frequent determinants of MLS resistance (up to 90% cases) and that gene msrA, encoding the protein responsible for the active excretion by the resistant cells of macrolides and streptogramins but not of lincosamides, is less common. Gene linA, whose protein product modiﬁes and thus inactivates lincosamide antibiotics only, is an additional resistance gene that is less frequent (Novotna et al., 2002). Clindamycin and lincomycin share a common or overlapping binding site on the ribosome. As already mentioned, a ribosomal mutation makes the ribosome insensitive to clindamycin. Expression of clindamycin-resistance in Gram-positive cocci may be constitutive or inducible. Staphylococci can also owe their resistance to clindamycin because of enzymatic inactivation of the drug, the enzyme production being speciﬁed by small nonconjugative plasmids. The third resistance mechanism to clindamycin involves active efﬂux of the antibiotic from the periplasmic space. This mainly occurs in Gram-negative bacteria (Leclercq and Courvalin, 1991). Lincomycin resistance in clinical isolates of staphylococci and streptococci has been recognized for several decades. This resistance is plasmid mediated and is encoded on transposons. The resistance

138

SPI´ZˇEK et al.

results from the induction of an enzyme that is normally repressed. The methylated RNA binds lincomycin-type drugs less well than does the nonmethylated RNA. The induction of resistance varies by species, and in most Gram-positive species, erythromycin is a moreeffective inducer of resistance than clindamycin. The plasmids that mediate lincomycin resistance in streptococci and staphylococci are highly similar structurally, indicating that they could have been readily transferred among strains of these species. Inactivation (resistance) of lincosamides by the products of the linA (encoding 3-lincomycin 4-clindamycin O-nucleotidyltransferase) genes of Staphylococcus aureus is one of the resistance mechanism in this bacterium (Matsuoka, 2000). VII. Biological Activity and Applications When usual doses are applied, both lincomycin and clindamycin exhibit bacteriostatic activity (Table III). At higher concentrations that can still be reached in vivo, their effect may be even bactericidal. However, the onset of the bactericidal effect is delayed and is less complete than that of, for example, -lactams. It increases generally in parallel with an increasing concentration of the antibiotic, and even in this respect clindamycin is more effective. On the other hand, the main advantage of lincomycin is that it can be applied at a substantially wider concentration range of clinical therapeutical doses. Macrolides and lincosamides are ﬁrst-choice bacteriostatic antibiotics used in veterinary dermatology. The main antibiotics of these classes are erythromycin, lincomycin, clindamycin, and tylosin. They are well absorbed if administered orally and are able to penetrate well into infected skin. Their spectrum of action comprises bacteria commonly associated with skin infections, including staphylococci. Their main disadvantage is the rapid development of bacterial resistance and occasional gastrointestinal upset (Noli and Boothe, 1999). Macrolides, ﬂuoroquinolones, rifamycins, tetracyclines, trimethoprim-sulfamethoxazole, and clindamycin have been described as antimicrobial agents of preference for the dermatologist (Epstein et al., 1997). Clindamycin administration results in changes in intestinal microﬂora. Numbers of enterococcal species increase and those of all anaerobes decrease (Nord and Heimdahl, 1986). Lincomycin was also used as an inhibitor of protein synthesis in experiments concerning photoinactivation in plants and algae (Kato et al., 2002; Sicora et al., 2003).

139

LINCOSAMIDES TABLE III EFFECT OF LINCOMYCIN AND CLINDAMYCIN ON SOME COMMON PATHOGENIC BACTERIA MIC (g/ml) Test organism

Lincomycin

Clindamycin

Gram-positive

Generally sensitive

Generally sensitive

Bacillus anthracis

0.25–8.0

0.25–5.0

Staphylococcus aureus

0.2–3.2

0.04–1.6

Staphylococcus epidermidis

0.4–1.8

0.1–0.2

Streptococcus agalactiae

0.1–0.2

0.02–0.1

Streptococcus pneumoniae

0.01–0.8

0.002–0.1

Streptococcus pyogenes

0.04–0.8

0.01–0.2

Streptococcus viridans

0.02–1.0

0.005–0.2

Gram-negative

Generally resistant

Generally resistant

Escherichia coli

1000

64

Haemophilus inﬂuenzae

4–16

0.5–16.0

Klebsiella pneumoniae

8

125

Neisseria gonorhoeae

8–64

0.5–4.0

Neisseria meningitis

>32

4

Proteus vulgaris

1000

250

Pseudomonas aeruginosa

>1000

1000

Salmonella schottmuelleri

125

64

The values in Table III can serve for a general orientation only. They were obtained from a number of experimental papers in which different strains were used and the data often varied substantially.

VIII. Gram-Positive Bacteria Clindamycin is active against most of the following bacteria: Staphylococcus aureus, Streptococcus pyogenes, S. pneumoniae, S. viridans and S. bovis, Corynebacterium diphtheriae, Enterococcus durans, Bacillus anthracis, B. cereus, and the Nocardia spp., but unfortunately, it is inactive against Enterococcus faecalis and E. faecium (Gigantelli et al., 1991). In stomatology, clindamycin can be used for the treatment of infections caused by Bacillus melaninogenicus and B. fragilis. Bacterial skin and skin structure infections are caused by aerobic staphylococci and streptococci (Streptococcus pyogenes and Staphylococcus aureus), with aerobic Gram-negative bacilli and anaerobes being involved in more complicated infections. Systemic therapy with lincosamides (clindamycin) has been the cornerstone of the treatment of these infections for many years (Guay, 2003).

140

SPI´ZˇEK et al.

Staphylococcus aureus and Streptococcus pyogenes cause a number of serious infections, such as necrotizing fascitis and toxic shock syndrome, which are associated with the release of bacterial toxins. Animal studies showed that clindamycin is more effective in treating these severe infections than are other drugs (Coyle, 2003). Bacillus anthracis infection can occur in three forms: cutaneous, gastrointestinal, and inhalational, depending on the mode of infection. Anthrax is a zoonotic disease but, unfortunately, the inhalation form can also be used as a biological warfare agent. In addition to other antibiotics, clindamycin was used for treatment (Brook, 2002a). Severe infections caused by Streptococcus pyogenes (group A streptococci) can be treated with clindamycin, acting as an inhibitor of the synthesis of protein M and of extracellular proteins (Bouvet, 1996). Pneumonia caused by Bacillus cereus could be treated with clindamycin (Bastian et al., 1997). Although tests of clindamycin resistance in streptococci are not generally performed in clinical practice, resistance to clindamycin has already been detected in strains of the following species: Streptococcus pyogenes, strains of group B streptococci, and strains of S. pneumoniae. Pneumococci that are multiply resistant to many antibiotics, including clindamycin, were reported from South Africa, but in most areas in the United States they are still clindamycin-sensitive. Staphylococci resistant to clindamycin are more common, and clindamycin sensitivity tests are always performed (Reeves et al., 1991). Type 3 pneumococci produce a capsule composed of cellobiuronic acid units connected in a (1 ! 3) linkage. The genes implicated in the biosynthesis of the type 3 capsule (cap3 genes) were cloned, expressed, and biochemically characterized. The three cap3 genes designated as cap3ABC form an operon. The Cap3A, Cap3B, and Cap3C proteins were biochemically characterized as a UDP-glucose dehydrogenase, the type 3 polysaccharide synthase, and a glucose-1-P uridyltransferase, respectively. The Cap3B protein was expressed in E. coli, and pneumococcal type 3 polysaccharide was synthesized in this heterologous system. When a recombinant plasmid containing cap3B (pLSE3B) was introduced into encapsulated pneumococci of types 1, 2, 5, or 8, the lincomycin-resistant transformants showed a type 3 capsule in addition to that of the recipient type. Unencapsulated (S2) laboratory strains of S. pneumoniae also synthesized a type 3 capsule when transformed with the pLSE3B plasmid (Garcia et al., 1997).

LINCOSAMIDES

141

The increasing prevalence of clindamycin-resistant bacteria (resistant Streptococcus pneumoniae in particular) is leading to new approaches to the management of common respiratory infections in the outpatient setting (Green and Wald, 1996). Clindamycin is an effective therapy for community-acquired, methicillin-resistant Staphylococcus aureus, but there is a risk of development of clindamycin resistance during the treatment of these bacteria (Marcinak and Frank, 2003). Emerging treatments for streptococcal toxic shock syndrome (caused by, for example, Streptococcus pyogenes) include administration of clindamycin and intravenous -globulin (Stevens, 2000). Streptococcus pyogenes, particularly the capsule and protein M, as well as streptococcal toxins, cause severe septic and toxic syndromes. Clindamycin should be used in case of septic shock (Veyssier-Belot et al., 1999). The efﬁcacy of clindamycin and the failure of penicillin to treat a severe group A streptococcal infection and streptococcal toxic shock syndrome were described (Stevens, 1996). An enhanced bactericidal response against -hemolytic streptococci has been found with a combination of penicillin and clindamycin (Seal, 2001). Resistance of enterococci and staphylococci to many antibiotics including clindamycin was described by McManus (1997). Antibiotic resistance of different strains of Bacteroides, Prevotella, and Porphyromonas species and in vitro antimicrobial susceptibility to many antibiotics, including clindamycin, were reviewed by Falagas and Siakavellas (2000). The true fungi and acid-fast and poorly Gram-stainable Mycobacterium tuberculosis are clindamycin-resistant, but clindamycin has some activity against M. leprae. IX. Gram-Negative Bacteria In general, aerobic Gram-negative bacteria are resistant to clindamycin. It was described that in vitro clindamycin is more active against H. inﬂuenzae than lincomycin. Campylobacter jejuni is sensitive to clindamycin, but E. coli is much more resistant (128 g/ml) compared with C. jejuni (8 g/ml). Capnocytophaga canimorosus causing bacteremic illness and dog sickness or other animal diseases is clindamycinsensitive. Flavobacteria may also be clindamycin-sensitive (Sheridan et al., 1993).

142

SPI´ZˇEK et al.

Clindamycin has good activity against the Bacteroides fragilis group of anaerobic bacteria (2 g/ml). Unfortunately, the number of clindamycin-resistant strains increases with time, but B. fragilis still remains one of the most sensitive bacteria as compared with B. thetaiotaomicron, B. ovatus, B. vulgatus, and B. distasonis (Tanaka-Bandoh et al., 1995). For instance, in the United States, the resistance is very low, varying to up to 25% of strains. Unfortunately, as usually, it increases proportionally with time. Thus in the B. fragilis group it increased from 4% to 38% in 10 years. A similar situation has also been observed in other parts of the world (Patey et al., 1994; Turgeon et al., 1994). Clindamycin is quite active against other Gram-negative anaerobes such as Prevotella disiens and P. melaninogenica and the Fusobacterium spp. Bacteroides gracilis may be clindamycin-sensitive, but some strains are resistant. Additional Gram-negative bacteria comprising strains of Butyrivibrio, Succinimonas, and Anaerovibrio can be sensitive to clindamycin. Necrotizing fascitis continues to occur due to -hemolytic streptococci but is now also recognized as being due to Vibrio spp. in ﬁshermen and those in contact with warm water in the Gulf of Mexico and Southeast Asia, including Hong Kong. The mechanism of Bacteroides resistance to clindamycin is usually as in Gram-positive bacteria (Jimenez-Diaz et al., 1992; Reig et al., 1992a). The resistance gene in Bacteroides spp. can be located on plasmids or on the chromosome; it can be transferred between species by a plasmid or transposon. Leng et al. (1975) used combinations of clindamycin with gentamicin against Enterobacteriaceae and Pseudomonas aeruginosa and demonstrated synergism. Klastersky and Husson (1977) showed that gentamicin did not interfere with the activity of clindamycin against B. fragilis, and that clindamycin did not inﬂuence the activity of gentamicin against E. coli. Adjunctive use of clindamycin, along with mechanical debridement is recommended for the treatment of Actinobacillus actinomycetemcomitans (Gram-negative, facultative anaerobic bacterium)-associated periodontitis as an acceptable therapeutic regimen (Walker and Karpinia, 2002). Laboratory results could be used in clinical practice by comparing the MIC50 of the germs regularly encountered in bone infections (staphylococci, streptococci including enterococci, Gram-negative bacilli, P. aeruginosa, and H. inﬂuenzae) with concentrations obtained in the different studies. In these studies it was described that clindamycin has a moderate bone diffusion (Boselli and Allaouchiche, 1999). Mycoplasma and Ureaplasma urealyticum are clindamycin-resistant;

LINCOSAMIDES

143

on the contrary, clindamycin has some activity against Chlamydia trachomatis (Rice et al., 1995) or Coxiella burnetii. X. Anaerobic Bacteria Clostridium tetani and C. perfringens are sensitive to clindamycin, but some C. perfringens strains and strains of C. sporogenes, C. tertium, C. bifermentans, C. novyi, C. ramosum, and C. sordelli may be clindamycin-resistant. A systematic review of studies that investigated the association of clindamycin-like antibiotics with hospital-acquired Clostridium difﬁcile diarrhea was undertaken to summarize the strength of the evidence for this relationship (Thomas et al., 2003). Clostridium difﬁcile is responsible for 300,000 to 3,000,000 cases of diarrhea and colitis in the United States every year. Antibiotics most frequently indicated for this infection are clindamycin, ampicillin, amoxicillin, and cephalosporins (Mylonakis et al., 2001). Clindamycin was an effective drug in the treatment of Gram-positive anaerobic infections (e.g., Clostridium perfringens). Very rapidly, the anti-anaerobic armamentarium was extended with clindamycin, cefoxitin, imipenem, and co-amoxyclavin or piperacillintazobactam. The resistance rate to metronidazole and imipenem remains low, but clindamycin has seen an important decrease in bacterial susceptibility (Bryskier, 2001). Clostridium difﬁcile is now established as a major nosocomial pathogen. C. difﬁcile infection is seen almost exclusively as a complication of antibiotic therapy and is particularly associated with clindamycin and third-generation cephalosporins (Freeman and Wilcox, 1999). Agents with a high potential to induce Clostridium difﬁcile– associated disease include aminopenicillins, cephalosporins, and clindamycin (Job and Jacobs, 1997). Other anaerobic Gram-positive organisms such as Peptococcus, Peptostreptococcus, Eubacterium, Propionibacterium, Biﬁdobacterium and Lactobacillus spp., Actinomyces israelii or Biﬁdobacterium, and Eubacterium spp. (Brook and Frazier, 1993) are usually sensitive to clindamycin. Naturally, even here, resistant strains have been described—for example, in Peptostreptococcus spp. (Reig et al., 1992b) or Lactobacillus spp. Vaginal bacterial infections are usually caused by mixed bacterial populations, including Peptostreptococcus sp., Peptococcus sp., B. fragilis, and of the aerobic bacteria by Streptococcus viridans, S. agalactiae, and less by S. pyogenes and other enterobacteria. The mixed

144

SPI´ZˇEK et al.

population can also contain enterococci and staphylococci. With the exception of enterococci, the population is to a considerable extent sensitive to a combination of clindamycin with an aminoglycosidic antibiotic. Clindamycin was also found to inhibit Chlamydia trachomatis. Bacterial vaginosis is an alteration of the vaginal ﬂora, where the normally predominant lactobacilli are replaced by a mixture of organisms including Gardnerella vaginalis and anaerobes and is treated with metronidazole or clindamycin (Priestley and Kinghorn, 1996). In bacterial vaginosis the normal hydrogen peroxide-producing Lactobacillus sp. in the vagina is replaced with high concentrations of characteristic sets of aerobic and anaerobic bacteria. It also occurs in women treated, for example, by orally administered clindamycin (McGregor and French, 2000). Increased doses of clindamycin and lincomycin (at least 8 g per day in adults) have to be administered for the treatment of B. fragilis infections, and even then the effect is not guaranteed. In cases of anaerobic sepsis, usually caused by B. fragilis or Peptostreptococcus sp., the application of clindamycin as the ﬁrst choice antibiotic is fully justiﬁed. In many patients with acne, caused by resistant Propionibacterium acnes, continued treatment with antibiotics such as clindamycin can be inappropriate or ineffective (Cooper, 1998). Clostridium difﬁcile may be clindamycin-sensitive or -resistant, and the proportion of sensitive strains has varied from 10% to 90% in different studies. During outbreaks of diarrhea associated with C. difﬁcile, the strains are usually clindamycin-resistant, and they contain a plasmid, probably located on the chromosome. This codes for transferable macrolide-lincosamide-streptogramin B (MLSB) resistance. This resistance can be transferred from C. difﬁcile to Staphylococcus aureus. In one study, almost all of 161 isolates of C. difﬁcile of serogroups A, F, G, H, and X were susceptible to clindamycin and other antibiotics, but 32 toxigenic isolates of serogroup C were clindamycin resistant. The microbiology, diagnosis, and management of bacteremia caused by anaerobic bacteria (Bacteroides fragilis, Peptostreptococcus sp., Clostridium sp., and Fusobacterium sp.) in children were reviewed by Brook (2002b). XI. Protozoa and Other Organisms Clindamycin has been used as an antimalarial drug (Lell and Kremsner, 2002). It was found effective in animals infected with chloroquine-resistant and chloroquine-sensitive Plasmodium falcipar-

LINCOSAMIDES

145

um. It is also effective against P. vivax but not against the exo-erythrocytic parasites. In P. falciparum clindamycin appears to be quite effective in the treatment of semi-immune subjects, and it enhances quinine activity (Patenotte et al., 1995). Malaria caused by chloroquine-resistant strains can be treated successfully with combinations of clindamycin with a ‘‘classical’’ antimalarial drug (Mordmuller et al., 1998). Clindamycin is effective in experimental toxoplasmosis in mice. In cultured mammalian cells, clindamycin reduces the level of replication of Toxoplasma gondii affecting protein synthesis of free parasites and also impairs the ability of the parasite to infect host cells. Infections caused by Toxoplasma gondii can be treated with clindamycin (Fung and Kirschenbaum, 1996). T. gondii clindamycin-resistant mutants can be selected that usually exhibit cross-resistance to spiromycin and azithromycin (Fichera et al., 1995). The inﬂuence of antimicrobial agents on replication and stage conversion of Toxoplasma gondii was described by Gross and Pohl (1996). Molecular genetic tools for the identiﬁcation and analysis of drug targets in Toxoplasma gondii were reviewed by Roos (1996). Human babesiosis is an important emerging tick-borne disease. Babesia divergens, a parasite of cattle, has been implicated as the most common agent of human babesiosis in Europe and/or Babesia microti in the United States, causing severe disease in splenectomized individuals. Human babesiosis can be treated by clindamycin administered intravenously (Uguen et al., 1997). Current treatment for babesiosis is focused on a regimen of clindamycin and quinine (Kjemtrup and Conrad, 2000). XII. Conclusion and Future Prospects Lincomycin and clindamycin are clinically important antibiotics. According to its worldwide production, the semisynthetic lincosamide derivative clindamycin is one of the 20 most important antibiotic compounds. They are active against most Gram-positive bacteria and against the genera Staphylococcus and Streptococcus in particular. They do not affect Gram-negative bacteria but exhibit a signiﬁcant antibiotic activity against some anaerobic bacteria. They are used therapeutically, especially in cases where synergistic effects of a mixed anaerobic and aerobic microﬂora are anticipated (prevention of intraabdominal infection after surgery, stomatological infections, anaerobic sepsis, skin and mucosa infections, and especially infections of bone

146

SPI´ZˇEK et al.

and articular infections). Lincomycin and clindamycin are also useful alternatives to penicillin and its derivatives in the treatment of upper respiratory tract infections in patients with allergy to penicillin. As compared with lincomycin, clindamycin is highly effective in the treatment of toxoplasmosis and pneumocystosis in AIDS patients. Clindamycin and some of its derivatives seem to be promising for the treatment of malaria caused by Plasmodium falciparum, even of strains that developed resistance to chloroquine, sulfonamides, and pyrimethamine. By means of chemical modiﬁcation of lincomycin, a number of derivatives with improved properties were obtained, whereas biotransformation has so far been less successful. However, with the current knowledge of gene clusters specifying biosynthesis of lincomycin and the related antibiotic celesticetin, one can easily imagine production of new derivatives of lincosamides by genetic engineering, namely by combination of parts of the lincomycin cluster with those of the celesticetin cluster, resulting in production of new derivatives of lincosamides. The knowledge of genomes of more than 100 pathogenic bacteria will make it possible to direct the search for new antibiotics speciﬁc for new targets in the pathogens. In addition, the recent completion of the genomes of Streptomyces coelicolor and Streptomyces avermitilis revealed that new secondary metabolites that had not yet been described are coded for by speciﬁc genes. Thus, not only hybrid antibiotics but also completely new compounds can still be discovered in the near future. In this respect it appears to the authors that lincosamides are not only useful antibiotics, namely for the treatment of anaerobic and protozoal infections, but that they also may serve as a model for the future development of new derivatives of antibiotics by means of genetic engineering.

ACKNOWLEDGMENTS This work was supported by the Grant Agency of the Czech Republic (grant no. 204/01/ 1004) and by the Institutional Research Concept no. AV 0Z 5020 903. The authors wish to express their thanks to Mrs G. Broucˇkova´ for administrative help. Excellent technical ˇ ezanka (Institute of Chemical Technology and Faculty of Science of assistance of Mr. P. R the Charles University, Prague) is gratefully acknowledged. REFERENCES Argoudelis, A. D., and Coats, J. H. (1969). Microbial transformation of antibiotics. II. Phosphorylation of lincomycin by Streptomyces species. J. Antibiot. 22, 341–343. Argoudelis, A. D., and Coats, J. H. (1973). Process for producing linkomycin. US Patent 3,726,766.

LINCOSAMIDES

147

Argoudelis, A. D., and Coats, J. H. (1974). Process for preparing lincomycin. US Patent 3,812,014. Argoudelis, A. D., Coats, J. H., and Pyke, T. R. (1972). Lincomycin production. US Patent 3,697,380. Argoudelis, A. D., Eble, T. E., Fox, J. A., and Mason, D. J. (1969). Studies on the biosynthesis of lincomycin. IV. The origin of methyl groups. Biochemistry 8, 3408–3411. Argoudelis, A. D., Eble, T. E., and Mason, D. J. (1970). J. Antibiot. 23, 1–8. Argoudelis, A. D., and Mason, D. J. (1965). Biochemistry 4, 704–709. Argoudelis, A. D., and Stroman, D. W. (1984). Process of producing lincomycin nucleotides. US Patent 4,464,466. Bastian, L., Weber, S., and Regel, G. (1997). Bacillus cereus pneumonia after severe thoracic trauma, case report and review of the literature. Anesthesiol. Intensivmedizin Otfallmedizin Schmerztherapie 32, 124–129. Bergy, M. E., Coats, J. H., and Malik, V. S. (1981). Process for preparing lincomycin. US Patent 4,271,266. Bergy, M. E., Herr, R. R., and Mason, D. J. (1963). Antibiotic lincolnensin and method of production. US Patent 3,086,912. Birkenmeyer, R. D. (1981). Lincomycin compounds. Novel useful analogs of the well known antibiotics lincomycin and clindamycin. US Patent 4,278,789. Birkenmeyer, R. D. (1982a). Lincomycin compounds. Novel useful analogs of the well known antibiotics lincomycin and clindamycin. US Patent 4,309,533. Birkenmeyer, R. D. (1982b). Lincomycin compounds. Novel useful analogs of the well known antibiotics lincomycin and clindamycin. US Patent 4,310,660. Boselli, E., and Allaouchiche, B. (1999). Bone tissue diffusion of antibiotics. Presse Medicale 28, 2265–2276. Bouvet, A. (1996). Severe infections due to Streptococcus pyogenes (group a streptococci). Medecine Maladies Infectieuses 6, 803–808. Bowden, K., and Stevens, G. P. (2000). An alternative synthesis of clindamycin. J. Serbian Chem. Soc. 65, 691–693. Brahme, N. M., Gonzalez, J. E., Mizsak, S., Rolls, J. R., Hessler, E. J., and Hurley, L. H. (1984a). Biosynthesis of the lincomycins. 2. Studies using stable isotopes on the biosynthesis of methylthiolincosaminide moiety of lincomycin A. J. Am. Chem. Soc. 106, 7878–7883. Brahme, N. M., Gonzalez, J. E., Rolls, J. R., Hessler, E. J., Mizsak, S., and Hurley, L. H. (1984b). Biosynthesis of the lincomycins. 1. Studies using stable isotopes on the biosynthesis of the propyl- and ethyl-L-hygric acid moieties of lincomycin A and B. J. Am. Chem. Soc. 106, 7873–7878. Brook, I. (2002a). Clinical review: Bacteremia caused by anaerobic bacteria in children. Crit. Care 6, 205–211. Brook, I. (2002b). The prophylaxis and treatment of anthrax. Int. J. Antimicrob. Agent, 320–325. Brook, I., and Frazier, E. H. (1993). Signiﬁcant recovery of nonsporulating anaerobic rods from clinical specimens. Clin. Infect. Dis. 16, 476–480. Bryskier, A. (2001). Anti-anaerobic activity of antibacterial agents. Exp. Opin. Invest. Drugs 10, 239–267. Cherlet, M., Croubels, S., and De Backer, P. (2002). Determination of clindamycin in animal plasma by high-performance liquid chromatography combined with electrospray ionization mass spectrometry. J. Mass Spectrometry 37, 848–853.

148

SPI´ZˇEK et al.

Chung, S. T., and Crose, L. L. (1990) Transposon Tn4556-mediated DNA insertion and site-directed mutagenesis. In ‘‘Proceedings of the 6th International Symposium on Genetics of Industrial Microorganisms’’ (H. Heslot, J. Davies, J. Florent, J. Bobichon, G. Durand, and L. Peneasse, eds.) pp. 207–218. Societe Francaise de Microbiologie, Strasbourg. Chung, S. T., Manis, J. J., McWethy, S. J., Patt, T. E., Witz, D. F., Wolf, H. J., and Wovcha, M. G. (1997). Fermentation, biosynthesis, and molecular genetics of lincomycin. In ‘‘Biotechnology of Industrial Antibiotics’’ (W. R. Strohl, ed.), 2nd ed., pp. 165–186. Dekker, New York. Cooper, A. J. (1998). Systematic review of Propionibacterium acnes resistance to systemic antibiotics. Medical J. Australia 169, 259–261. Coyle, E. A. (2003). Targeting bacterial virulence: The role of protein synthesis inhibitors in severe infections—Insights from the Society of Infectious Diseases Pharmacists. Pharmacotherapy 23, 638–642. Crellin, K. C., Sible, E., and Van Antwerp, J. (2003). Quantiﬁcation and conﬁrmation of identity of analytes in various matrices with in-source collision-induced dissociation on a single quadrupole mass spectrometer. Int. J. Mass Spectrom. 222, 281–311. Dong, B., and Dong, M. (2001). Extracting technology for lincomycin. China Patent 1,301,703. Epstein, M. E., AmodioGroton, M., and Sadick, N. S. (1997). Antimicrobial agents for the dermatologist. 2. Macrolides, ﬂuoroquinolones, rifamycins, tetracyclines, trimethoprim-sulfamethoxazole, and clindamycin. J. Amer. Acad. Dermatol. 37, 365–381. Falagas, M. E., and Siakavellas, E. (2000). Bacteroides, Prevotella, and Porphyromonas species: A review of antibiotic resistance and therapeutic options. Int. J. Antimicrob. Agents 15, 1–9. Fieger-Bu¨schges, H., Schussler, G., Larsimont, V., and Blume, H. (1999). Determination of clindamycin in human plasma by high-performance liquid chromatography using coupled columns. J. Chromatogr. B 724, 281–286. Fichera, M. E., Bhopale, M. K., and Roos, D. S. (1995). In vitro assays elucidate peculiar kinetics of clindamycin action against Toxoplasma gondii. Antimicrob. Agents. Chemother. 39, 1530–1537. Fishman, J. A. (1998). Treatment of infection due to Pneumocystis carinii. Antimicrob. Agents. Chemother. 42, 1309–1314. Fitzhugh, A. L. (1998). Antibiotic inhibitors of the peptidyl transferase center. 1. Clindamycin as a composite analogue of the transfer RNA fragments 1-Pro-Met and the D-ribosyl ring of adenosine. Bioorg. Med. Chem. Lett. 8, 87–92. Freeman, J., and Wilcox, M. H. (1999). Antibiotics and Clostridium difﬁcile. Microbes Infection. 1, 377–384. Fung, H. B., and Kirschenbaum, H. L. (1996). Treatment regimens for patients with toxoplasmic encephalitis. Clin. Therapeut. 18, 1037–1056. Garcia, E., Arrecubieta, C., Munoz, R., Mollerach, M., and Lopez, R. (1997). A functional analysis of the Streptococcus pneumoniae genes involved in the synthesis of type 1 and type 3 capsular polysaccharides. Microb. Drug Resist. Mech. Epidem. Dis. 3, 73–88. Gigantelli, J. W., Gomez, J. T., and Osato, M. S. (1991). In Vitro susceptibilities of ocular Bacillus cereus isolates to clindamycin, gentamicin, and vancomycin alone or in combination. Antimicrob. Agents Chemother. 35, 201–202. Gonda, J., Zavacka, E., Budesinsky, M., Cisarova, I., and Podlaha, J. (2000). Stereocontrolled introduction of an amino group at C-6 of D-galactose via (3,3)-sigmatropic

LINCOSAMIDES

149

rearrangements—novel synthesis of lincosamine and 7-epi-lincosamine precursors. Tetrahedron Lett. 41, 525–529. Green, M., and Wald, E. R. (1996). Emerging resistance to antibiotics: Impact on respiratory infections in the outpatient setting. Annals Allergy Asthma Immunol. 77, 167–173. Gross, U., and Pohl, F. (1996). Inﬂuence of antimicrobial agents on replication and stage conversion of Toxoplasma gondii. Current Topics Microbiol. Immunol. 219, 235–245. Guay, D. R. P. (2003). Treatment of bacterial skin and skin structure infections. Expert Opinion Pharmacotherapy 4, 1259–1275. Hanko, V. P., LaCourse, W. R., Dasenbrock, C. O., and Rohrer, J. S. (2001). Determination of sulfur-containing antibiotics using high-performance liquid chromatography with integrated pulsed amperometric detection. Drug Development Res. 53, 268–280. Hausknecht, E. C., and Wolf, H. J. (1986). Radiometric assay for N-demethyllincomycin synthetase. American Society of Microbiology, Annual Meeting. Herr, R. R., and Slomp, G. (1967). Lincomycin. II. Characterization and gross structure. J. Am. Chem. Soc. 89, 2444–2447. Hoeksema, H. (1968). Celesticetin. V. Structure of celesticetin. J. Am. Chem. Soc. 90, 755–757. Hoeksema, H., Bannister, B., Birkenmeyer, R. D., Kagan, F., Magerlein, B. J., MacKellar, F. A., Schroeder, W., Slomp, G., and Herr, R. R. (1964). Chemical studies on lincomycin. 1. The structure of lincomycin. J. Am. Chem. Soc. 86, 4223–4224. Hoeksema, H., Crum, G. F., and DeVries, W. H. (1955). Isolation and puriﬁcation of celesticetin. Antibiot. Ann. 2, 837–841. Hofstetter, J. R. (1982). Process for the puriﬁcation of lincomycin. US Patent 4,317,903. Horton, D., Wander, J. D., and Foltz, R. L. (1974). Chemical-ionization mass spectrometry of Lincomycin and Clindamycin. Anal. Biochem. 59, 452–461. Hurley, L. H. (1980). Elucidation and formulation of novel biosynthetic pathways leading to the pyrrolo-benzodiazepine antibiotics anthramycin, tomaymycin, and sibiromycin. Acc. Chem. Res. 13, 263–269. Hurley, L. H., Laswell, W. L., Ostrander, J. M., and Parry, R. (1979). Pyrrolo antibiotics. Biosynthetic conversion of tyrosine to the C2- and C3-proline moieties of anthramycin, tomaymycin and sibiromycin. Biochemistry 18, 4230–4237. Janata, J., Kadlec, J., Gakh, O., Licha, I., Kopecky, J., Tichy, P., and Spizek, J. (1999). LmbJ protein alone catalyzes conversion of N-demethyllincomycin to lincomycin 199. In ‘‘11th International Symposium on the Biology of Actinomycetes,’’ Abstr Book, October 24–28, p. 188, Sissi-Heraklion, Crete-Greece. Jimenez-Diaz, A., Reig, M., Baquero, F., and Ballesta, J. P. (1992). Antibiotic sensitivity of ribosomes from wild-type and clindamycin resistant Bacteroides vulgatus strains. J. Antimicrob. Chemother. 30, 295–301. Job, M. L., and Jacobs, N. F. (1997). Drug-induced Clostridium difﬁcile-associated disease. Drug Safety 17, 37–46. Kagan, F., and Trostic, M. F. (1972). Mass spectra of lincomycin, lincomycin analogs, and degradation products. Org. Mass Spectrom. 6, 1217–1223. Kato, M. C., Hikosaka, K., and Hirose, T. (2002). Photoinactivation and recovery of photosystem II in Chenopodium album leaves grown at different levels of irradiance and nitrogen availability. Funct. Plant Biol. 29, 787–795. Kjemtrup, A. M., and Conrad, P. A. (2000). Human babesiosis: An emerging tick-borne disease. Int. J. Parasitol. 30, 1323–1337.

150

SPI´ZˇEK et al.

Klastersky, J., and Husson, M. (1977). Bactericidal activity of the combinations of gentamicin with clindamycin or chloramphenicol against species of Escherichia coli and Bacteroides fragilis. Antimicrob. Agents Chemother. 12, 135–138. Krzek, J., Kwiecien, A., Starek, M., Kierszniewska, A., and Rzeszutko, W. (2000). Identiﬁcation and determination of oxytetracycline, tiamulin, lincomycin, and spectinomycin in veterinary preparations by thin-layer chromatography/densitometry. J. AOAC Int. 83, 1502–1506. Kuo, M. S., Yurek, D. A., Coats, J. H., and Li, G. P. (1989). Isolation and identiﬁcation of 7,8-didemethyl-8-hydroxy-5-deazariboﬂavin, an unusual cosynthetic factor in streptomycetes, from Streptomyces lincolnensis. J. Antibiot. 42, 475–478. Kuo, M. S., Yurek, D. A., Coats, J. H., Chung, S. T., and Li, G. P. (1992). Isolation and identiﬁcation of 3-propylidiene acid, a biosynthetic precursor of lincomycin. J. Antibiot. 45, 1773–1777. La Follette, G., Gambertoglio, J., White, J. A., Knuth, D. W., and Lin, E. T. (1988). Determination of clindamycin in plasma or serum by high-performance liquid chromatography with ultraviolet detection. J. Chromatogr. 431, 379–388. Leclercq, R., and Courvalin, P. (1991). Bacterial resistance to macrolide, lincosamide, and streptogramin antibiotics by target modiﬁcation. Antimicrob. Agents Chemother. 35, 1267–1272. Lell, B., and Kremsner, P. G. (2002). Clindamycin as an antimalarial drug: Review of clinical trials. Antimicrob. Agents Chemother. 46, 2315–2320. Leng, B., Meyers, B. R., Hirschman, S. Z., and Keusch, G. T. (1975). Susceptibilities of gram-negative bacteria to combinations of antimicrobial agents in vitro. Antimicrob. Agents Chemother. 8, 164–171. Li, C., Yu, W., and Li, B. (2001). Producing technology for lowering lincomycin B content in lincomycin by vibrative sieve plate extraction tower. China Patent 1,280,133. Liu, C. M., Chen, Y. K., Yang, T. H., Hsieh, S. Y., Hung, M. H., and Lin, E. T. (1997). Highperformance liquid chromatographic determination of clindamycin in human plasma or serum: Application to the bioequivalency study of clindamycin phosphate injections. J. Chromatogr. B 696, 298–302. Magerlein, B. J. (1971). Modiﬁcation of lincomycin. Adv. Appl. Microbiol. 14, 185–229. Marcinak, J. F., and Frank, A. L. (2003). Treatment of community-acquired methicillinresistant Staphylococcus aureus in children. Curr. Opin. Infectious Dis. 16, 265–269. Martens-Lobenhoffer, J., and Banditt, P. (2001). Sensitive and speciﬁc determination of clindamycin in human serum and bone tissue applying liquid chromatographyatmospheric pressure chemical ionization-mass spectrometry. J. Chromatogr. B 755, 143–149. Martin, J. F., and Liras, P. (1989). Organization and expression of genes involved in the biosynthesis of antibiotics and other secondary metabolites. Annu. Rev. Microbiol. 43, 173–206. Mason, D. J., and Lewis, C. (1964). Biological activity of the lincomycin related antibiotics. Antimicrob. Agents Chemother. 10, 7–12. Matsuoka, M. (2000). Study of macrolide, lincosamide, and streptogramin B antibiotics resistance in Staphylococcus aureus. Yakugaku Zasshi. 120, 374–386. McGregor, J. A., and French, J. I. (2000). Bacterial vaginosis in pregnancy. Obst. Gynecol. Survey 55 (Suppl.), S1–S19. McManus, M. C. (1997). Mechanisms of bacterial resistance to antimicrobial agents. Amer. J. Health-System Pharmacy. 54, 1420–1433.

LINCOSAMIDES

151

Menninger, J. R., and Coleman, R. A. (1993). Lincosamide antibiotics stimulate dissociation of peptidyl-transfer RNA from ribosomes. Antimicrob. Agents. Chemother. 37, 2027–2029. Mordmuller, B., Graninger, W., and Kremsner, P. G. (1998). Malaria chemotherapy in the time of chloroquine resistance. Wiener Klinische Wochenschrift. 110, 321–325. Mylonakis, E., Ryan, E. T., and Calderwood, S. B. (2001). Clostridium difﬁcile-associated diarrhea – A review. Archiv Internal. Medicine 161, 525–533. Neusser, D., Schmidt, H., Spizek, J., Novotna, J., Peschke, U., Kaschabeck, S., Tichy, P., and Piepersberg, W. (1998). The genes lmbB1 and lmbB2 of Streptomyces lincolnensis encode enzymes involved in the conversion of L-tyrosine to propylproline during the biosynthesis of the antibiotic lincomycin A. Arch. Microbiol. 169, 322–332. Noli, C., and Boothe, D. (1999). Macrolides and lincosamides. Veterinary Dermatology 10, 217–223. Nord, C. E., and Heimdahl, A. (1986). Impact of orally-administered antimicrobial agents on human oropharyngeal and colonic microﬂora. J. Antimicrob. Chemother. 18, 159–164. Novotna, G., Janata, J., Kopecky, J., Najmanova, L., and Spizek, J. (2002). Genes controlling MLS resistance in methicillin-resistant Staphylococcus aureus. In ‘‘Confronting infections, antibiotic resistance, and bioterrorism around the world,’’ Inter Academy Medical Panel (IAMP), pp. 169–170, Elsevier. Patenotte, A., Dupuy, O., Peyron, F., Delolme, H., and Flechaire, A. (1995). Antibiotics and malaria – Review of the literature. Medecine Maladies Infectieuses 25, 970–975. Patey, O., Varon, E., Prazuck, T., Podglajen, I., Dublanchet, A., Dubreuil, L., and Breuil, J. (1994). Multicentre survey in France of the antimicrobial susceptibilities of 416 blood culture isolates of the Bacteroides fragilis group. J. Antimicrob. Chemother. 33, 1029–1034. Patt, T. E., Argoudells, A. D., and Marshall, V. P. (1984). Process for preparing lincomycin and clindamycin ribonucleotides. US Patent 4,430,495. Patt, T. E., and Horvath, B. A. (1985). Isolation and characterization of N-demethyllincomycin methyltransferase. In 13th International Congress of Biochemistry, Amsterdam, The Netherlands, August 25–30. Peschke, U., Schmidt, H., Zhang, H. Z., and Piepersberg, W. (1995). Molecular characterization of the lincomycin production gene cluster of Streptomyces lincolnensis 78–11. Mol. Microbiol. 16, 1137–1156. Piepersberg, W. (1994). Streptomycin and related aminoglycoside antibiotics. In ‘‘Biochemistry and Genetics of Antibiotic Biosynthesis’’ (L Vining and C. Stuttard, eds.), pp. 531–570. Butterworth-Heinemann, Newton. Piepersberg, W., and Distler, J. (1997). Aminoglycosides and sugar components in other secondary metabolites in biotechnology. In ‘‘Biotechnology, 2nd edn, vol 7. Products of secondary metabolism’’ (H. Kleinkauf and H. von Do¨ren, eds.), pp. 397–488. Wiley-VCH, Weinheim. Pissowotzki, K., Mansouri, K., and Piepersberg, W. (1991). Genetics of streptomycin production in Streptomyces griseus: Molecular structure and putative function of genes strELMB2N. Mol. Gen. Genet. 231, 113–123. Pospisil, S., Sedmera, P., Halada, P., and Spizek, J. (2001). Oxidation of lincomycin by hydrogen peroxide restricts its potential biotransformation with haloperoxidases. Folia Microbiol. 46, 376–378. Pospisil, S., Sedmera, P., Halada, P., Havlicek, L., and Spizek, J. (2004). Access to lincomycin N-oxide isomers controlled by reaction conditions. Tetrahedron Lett. 45, 2943–2945.

152

SPI´ZˇEK et al.

Priestley, C. J. F., and Kinghorn, G. R. (1996). Bacterial vaginosis. Brit. J. Clinic. Practice 50, 331–334. Reeves, D. S., Holt, H. A., Phillips, I., King, A., Miles, R. S., Paton, R., Wise, R., and Andrews, J. M. (1991). Activity of clindamycin against Staphylococcus aureus and Staphylococcus epidermidis from 4 UK centers. J. Antimicrob. Chemother. 27, 469–474. Reig, M., Fernandez, M. C., Ballesta, J. P., and Baquero, F. (1992a). Inducible expression of ribosomal clindamycin resistance in Bacteroides vulgatus. Antimicrob. Agents Chemother. 36, 639–642. Reig, M., Moreno, A., and Baquero, F. (1992b). Resistance of Peptostreptococcus spp. to macrolides and lincosamides: Inducible and constitutive phenotypes. Antimicrob. Agents Chemother. 36, 662–664. Remington J. S. (ed.) (1990). The role of clindamycin alone or in combination with another agent for the treatment of toxoplasmosis and pneumocystosis in AIDS. AM Medica Communications, New York. Retzlaff, L., Mayer, G., Beyer, G., Ahlert, J., Verseck, S., Distler, J., and Piepersberg, W. (1993). Streptomycin production in streptomycetes: A progress report. In ‘‘Industrial Microorganisms: Basic and Applied Molecular Genetics’’ (G. D. Hegeman, R. H. Baltz, and P. L. Skatrud, eds.), pp. 183–194. American Society for Microbiology, Washington, DC. Reusser, F., and Argoudelis, A. D. (1974). Process for preparing lincomycin. US Patent 3,833,475. Rice, R. J., Bhullar, V., Mitchell, S. H., Bullard, J., and Knapp, J. S. (1995). Susceptibilities of Chlamydia trachomatis isolates causing uncomplicated female genital-tract infections and pelvic inﬂammatory disease. Antimicrob. Agents Chemother. 39, 760–762. Rinehart, K. L., Snyder, W. C., Staley, A. L., and Lau, R. C. M. (1992). Biosynthetic studies on antibiotics. In ‘‘Secondary Metabolite Biosynthesis and Metabolism’’ (R. J. Petroski and S. P. McCormick, eds.), pp. 41–60. Plenum Press, New York. Roos, D. S. (1996). Molecular genetic tools for the identiﬁcation and analysis of drug targets in Toxoplasma gondii. Curr. Topics Microbiol. Immunol. 219, 247–259. Sanders, C. C., Sanders, W. E., Jr., and Goering, R. V. (1983). Effects of clindamycin on derepression of beta-lactamases in gram-negative bacteria. J. Antimicrob. Chemother. 12(Suppl C), 97–104. Seal, D. V. (2001). Necrotizing fasciitis. Curr. Opin. Infectious Dis. 14, 127–132. Sheridan, R. L., Ryan, C. M., Pasternack, M. S., Weber, J. M., and Tompkins, R. G. (1993). Flavobacterial sepsis in massively burned pediatric patients. Clin. Infect. Dis. 17, 185–187. Schlievert, P. M., and Kelly, J. A. (1984). Clindamycin-induced suppression of toxicshock syndrome associated exotoxin production. J. Infect. Dis. 149, 471. Schlunzen, F., Zarivach, R., Harms, R., Bashan, A., Tocilj, A., Albrecht, R., Yonath, A., and Franceschi, F. (2001). Structural basis for the interaction of antibiotics with the peptidyl transferase centre in eubacteria. Nature 413, 814–821. Schroeder, W., Bannister, B., and Hoeksema, H. (1967). Lincomycin. The structure and stereochemistry of the carbohydrate moiety. J. Am. Chem. Soc. 89, 2448–2453. Sicora, C., Mate, Z., and Vass, I. (2003). The interaction of visible and UV-B light during photodamage and repair of photosystem II. Photosynthesis Res. 75, 127–137. Slomp, G., and MacKellar, F. A. (1967). Lincomycin. IV. Nuclear magnetic resonance studies on the structure of lincomycin, its degradation products, and some analogs. J. Am. Chem. Soc. 89, 2454–2459.

LINCOSAMIDES

153

Stevens, D. L. (1996). Invasive group A streptococcal disease. Infectious Agents DiseaseReviews Issues Commentary 5, 157–166. Stevens, D. L. (2000). Streptococcal toxic shock syndrome associated with necrotizing fasciitis. Ann. Rev. Med. 51, 271–288. Stockmann, M., and Piepersberg, W. (1992). Gene probes for the detection of 6-deoxyhexose metabolism in secondary metabolite-producing streptomycetes. FEMS Microbiol. Lett. 90, 185–190. Szunyog, J., Adams, E., Liekens, K., Roets, E., and Hoogmartens, J. (2002). Analysis of a formulation containing lincomycin and spectinomycin by liquid chromatography with pulsed electrochemical detection. J. Pharm. Biomed. Anal. 29, 213–220. Tanaka-Bandoh, K., Kato, N., Watanabe, K., and Ueno, K. (1995). Antibiotic susceptibility proﬁles of Bacteroides fragilis and Bacteroides thetaiotaomicron in Japan from 1990 to 1992. Clin. Infect. Dis. 20(Suppl 2), S352–S355. Thomas, C., Stevenson, M., and Riley, T. V. (2003). Antibiotics and hospital-acquired Clostridium difﬁcile-associated diarrhoea: A systematic review. J. Antimicrob. Chemother. 51, 1339–1350. Thorson, J. S., Lo, S. F., Liu, H. W., and Hutichinson, C. R. (1993). Biosynthesis of 3,6dideoxyhexoses: New mechanistic reﬂections upon 2,6-dideoxy, 4,6-dideoxy, and amino sugar construction. J. Am. Chem. Soc. 115, 6963–6994. Turgeon, P., Turgeon, V., Gourdeau, M., Dubois, J., and Lamothe, F. (1994). Longitudinal study of susceptibilities of species of the Bacteroides fragilis group to ﬁve antimicrobial agents in three medical centers. Antimicrob. Agents Chemother. 38, 2276–2279. Uguen, C., Girard, L., Brasseur, P., and Leblay, R. (1997). Human babesiosis in 1997. Revue Medecine Int. 18, 945–951. VanVlem, B., Vanholder, R., DePaepe, P., Vogelaers, D., and Ringoir, S. (1996). Immunomodulating effects of antibiotics: Literature review. Infection 24, 275–291. Veyssier-Belot, C., Lejoyeux-Chartier, F., and Bouvet, A. (1999). Erysipela, cellulitis and other Streptococcus pyogenes skin infections. Presse Medicale 28, 1959–1965. Visser, J. (1972). Lincomycin production. US Patent 3,676,302. Walker, C., and Karpinia, K. (2002). Rationale for use of antibiotics in periodontics. J. Periodontol. 73, 1188–1196. Wang, B., and Qi, P. (1999). Process for recovering lincomycin from ﬁltered dregs of its fermented liquid. China Patent 1,239,100. Wang, B., and Qi, P. (2000). Process for extracting lincomycin by using organic acid. China Patent 1,266,063. Wang, B., and Qi, P. (2001). Lincomycin extracting process. China Patent 1,288,895. Weignerova, L., Spizek, J., Najmanova´, L., and Kren, V. (2001). Enzymatic glycosylation of lincomycin. Biosci. Biotechnol. Biochem. 65, 1897–1899. Witz, D. F. (1972). Lincomycin production. US Patent 3,687,814. Witz, D. F., Hessler, E. J., and Miller, T. L. (1971). Bioconversion of tyrosine into the propylhygric acid moiety of lincomycin. Biochemistry 10, 1128–1133. Wovcha, M. G., Steuerwald, D. C., and Manis, J. J. (1986). Cloning and characterization of a lincomycin resistance gene from Streptomyces lincolnensis. In ‘‘Abstracts of the 5th International Symposium on the Genetics of Industrial Microorganisms.’’ Abstract W4-P4, September 14–20, Split, Yugoslavia. Yang, W. C., Yu, A. M., and Chen, H. Y. (2001). Applications of a copper microparticlemodiﬁed carbon ﬁber microdisk array electrode for the simultaneous determination of aminoglycoside antibiotics by capillary electrophoresis. J. Chromatogr. A 905, 309–318.

154

SPI´ZˇEK et al.

Yu, L. L., Chao, C. K., Liao, W. J., Twu, T. Y., Liu, C. M., Yang, T. H., and Lin, E. T. (1999). Determination of clindamycin in human plasma by liquid chromatography electrospray tandem mass spectrometry: Application to the bioequivalence study of clindamycin. J. Chromatogr. B 724, 287–294. Zhang, H. Z. (1993). Genetik und Biochemie der Lincomycin-Biosynthese in Streptomyces lincolnensis 78–11. Klonierung and Analyse der Produktions und Resistenzgene. Wuppertal, Dissertation, Bugh. Zhang, H. Z., Schmidt, H., and Piepersberg, W. (1992). Molecular cloning and characterization of two lincomycin-resistance genes, lmrA and lmrB, from Streptomyces lincolnensis 78–11. Mol. Microbiol. 6, 2147–2157.

Ribosome Engineering and Secondary Metabolite Production KOZO OCHI,* SUSUMU OKAMOTO, YUZURU TOZAWA, TAKASHI INAOKA, TAKESHI HOSAKA, JUN XU, AND KAZUHIKO KUROSAWA National Food Research Institute Ibaraki 305-8642, Japan *Author for correspondence. E-mail: [email protected]

I. Introduction II. General Method for Obtaining Drug-Resistant Mutants III. Antibiotic Overproduction by rpsL (Ribosomal Protein S12) Mutations A. Activation of Actinorhodin Production B. Mechanism of Activation C. rpsL Mutations by Site-Directed Mutagenesis IV. Antibiotic Overproduction by rpoB (RNA Polymerase) Mutations V. Effect of str and rpoB Mutations in Various Bacteria A. Antibiotic Overproduction B. Enzyme Overproduction C. Neotrehalosadiamine VI. Increase of Chemical Tolerance in Pseudomonas VII. Combined Drug-Resistance Mutations A. Model Experiment B. Applicability to Industrial Strain VIII. Conclusion and Future Prospects A. Future Prospects References

155 156 157 157 161 162 164 167 167 169 169 171 172 172 173 175 176 179

I. Introduction Improvement of the productivity of commercially viable microbial strains is an important ﬁeld in microbiology, especially since wildtype strains isolated from nature usually produce only a low level (1–100 g/ml) of antibiotics. Therefore, a great deal of effort and resources have been committed to improving antibiotic-producing strains to meet commercial requirements. Current methods of improving the productivity of industrial microorganisms range from the classical random approach to using highly rational methods—for example, metabolic engineering. Although classical methods are still effective even without using genomic information or genetic tools to obtain highly productive strains, these methods are always time and resource intensive (Vinci and Byng, 1999; Zhang et al., 2002). One of the current topics is to use microorganisms for bioremediation. Environmental protection efforts have been focused on the development of more effective 155 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

156

OCHI et al.

processes for the treatment of toxic wastes. Soil bacteria have a wide range of metabolic abilities that make them useful tools for mineralization of toxic compounds (Timmis et al., 1994; Van der Meer et al., 1992). The discovery of microorganisms capable of tolerating, or growing on, high concentrations of organic solvents provides a potentially interesting avenue for development of genetically engineered organisms for treating hazardous wastes. Thus, strain improvement is crucially important to fully exploit the cell’s ability. Since we found a dramatic activation of antibiotic production by a certain ribosomal mutation (a mutation in rpsL gene encoding the ribosomal protein S12) (Shima et al., 1996), we had an idea that bacterial gene expression may be changed dramatically by modulating the ribosomal proteins or rRNA, eventually leading to activation of inactive (silent) genes. Thus, our ultimate aim was to develop ‘‘ribosome engineering’’ for a rational approach to fully elicit the bacterial abilities. In bacteria, the ribosome plays a special role for their own gene expression by synthesis of a bacterial alarmone, ppGpp. Namely, one of the most important adaptation systems for bacteria is the stringent response, which leads to the repression of stable RNA synthesis in response to nutrient limitation (Cashel et al., 1996). The stringent response depends on the transient increase of hyperphosphorylated guanosine nucleotide ppGpp, which is synthesized from GDP and ATP by the relA gene product (ppGpp synthetase) in response to binding of uncharged tRNA to the ribosomal A site. Since bacterial secondary metabolism is often triggered by ppGpp when cells enter into stationary phase, it is important to take the stringent response into consideration in activating or enhancing the bacterial secondary metabolism. In this review, we outline our ribosome engineering and its applicability, especially focusing on strain improvement for antibiotic overproduction in Streptomyces and Bacillus and for enhancement of tolerance to organic chemicals in Pseudomonas. II. General Method for Obtaining Drug-Resistant Mutants One of the most conventional ways to modulate the ribosome is the introduction of mutations conferring resistance to drugs that attack the ribosome. Such drugs include streptomycin, gentamicin, paromomycin, thiostrepton, fusidic acid, kanamycin, chloramphenicol, lincomycin, spectinomycin, and neomycin. The mutants resistant to these drugs frequently possess a point mutation or a deletion mutation within a ribosomal component (ribosomal protein, rRNA, or translation

RIBOSOME ENGINEERING

157

factor). An advantage of obtaining the drug-resistant mutants is selectability on the drug-containing plates, even if the incidence of resistant mutants is as low as 10–8–10–10. It is important to use several concentrations of drugs (e.g., 3-, 30-, and 100-fold of MIC) for obtaining a wide variety of mutants. (MIC: minimum inhibitory concentration to suppress cell’s growth). The MIC value varies depending on the species used. It is also recommended to use both the cells and spores grown in plate culture or liquid culture, since mutant forms emerging from spores or cells are often different. Since the frequency of spontaneous mutation is normally 106–108, abundant (107–109) cells or spores may be plated, followed by 1–3 days of incubation (in typical bacteria) or 5–14 days (in Streptomyces) to allow the development of resistant colonies. If the frequency of mutation is very low, more (1010–1011) cells or spores may be spread on a plate. Once mutant colonies develop, it is preferable to carry out single colony isolation before further testing. The mutants resistant to a high level of drugs were stable in their phenotype, whereas mutants resistant to a low level of drugs were unstable in general, as represented by gentamicin-resistant mutants. If a number of resistant colonies develop on the plate, one can select a variety of mutant forms on the basis of colony size, colony morphology, sporulation, pigment formation, etc. To test the ability of each mutant to produce antibiotics (or other metabolites), several kinds of media may be employed because the efﬁcacy of mutation on productivity often depends on the medium used. In this way, one can ﬁnd overproducing mutants among drug-resistant mutants at a high frequency of 2–40%. To determine the mutated gene, DNA sequence analysis with PCR can be done, focusing on the most probable mutated genes (e.g., ribosomal proteins S12 and L11 genes in streptomycin-resistant and thiostrepton-resistant mutants, respectively).

III. Antibiotic Overproduction by rpsL (Ribosomal Protein S12) Mutations A. ACTIVATION OF ACTINORHODIN PRODUCTION Members of the genus Streptomyces produce a wide variety of secondary metabolites that include about half of the known microbial antibiotics. Advances in understanding the regulation of secondary metabolism in this genus have come from the studies of antibiotic production in Streptomyces coelicolor A3(2) and its close relative Streptomyces lividans 66 (Bibb, 1996; Hopwood et al., 1995; Kieser

158

OCHI et al.

et al., 2000). S. coelicolor produces at least four antibiotics, including the blue-pigmented polyketide antibiotic actinorhodin (Act). S. lividans normally does not produce Act, although the strain has a complete set of Act biosynthetic genes. However, Act production in this organism can be activated by the introduction of certain regulatory genes (Floriano and Bibb, 1996; Horinouchi et al., 1990; Martı´nezCosta et al., 1996; Vogtli et al., 1994) or by cultivation under speciﬁc conditions (Kim et al., 2001). A strain of S. lividans, TK24, has been found to produce a large amount of Act under normal culture conditions (Shima et al., 1996) (Fig. 1, see color insert). Genetic analyses revealed that a streptomycinresistant mutation, str-6, in TK24 is responsible for activation of Act synthesis and that str-6 is a point mutation in the rpsL gene encoding ribosomal protein S12, changing Lys-88 to Glu (K88E mutation). It was also shown that introduction of streptomycin-resistant mutations improves Act production in wild-type S. coelicolor (Hesketh and Ochi, 1997) (Fig. 1) and circumvents the detrimental effects on Act

FIG. 1. Activation of antibiotic production by rpsL (encoding ribosomal protein S12) mutations in Streptomyces lividans 66 and Streptomyces coelicolor A3(2). Blue color represents an antibiotic, actinorhodin. K88E means a mutation at lysine-88 altering glutamate.

RIBOSOME ENGINEERING

159

production in certain developmental mutants (relA, relC, and brgA) of S. coeliolor (Ochi and Hosoya, 1998; Ochi et al., 1997; Shima et al., 1996). These streptomycin-resistant mutations result in the alteration of the Lys-88 to Glu (K88E) or Arg (K88R) and Arg-86 to His (R86H) in the rpsL gene. In addition to these streptomycin-resistant rpsL mutations, a paromomycin-resistant rpsL mutation (P91S) also can activate Act production in S. coelicolor (Okamoto-Hosoya et al., 2000). These ﬁndings indicate that the antibiotic production (secondary metabolism) in streptomycetes is signiﬁcantly controlled by the translational machinery, that is, the ‘‘ribosome.’’ Much progress has been made in elucidating the organization of antibiotic biosynthesis gene clusters in several Streptomyces species, and a number of pathway-speciﬁc regulatory genes have been identiﬁed, which are required for the activation of their cognate biosynthetic genes (Wietzorrek and Bibb, 1997). In the Act biosynthetic gene cluster, actII-ORF4 plays such a pathway-speciﬁc regulatory role, and the expression level of this gene directly determines the productivity of Act (Arias et al., 1999; Gramajo et al., 1993). Western blot analysis using anti-ActII-ORF4 antibody showed that the expression of ActIIORF4 protein was strongly enhanced in the Act-high-producing rpsL mutant strains (Hu and Ochi, 2001; Okamoto et al., unpublished). Furthermore, RT-PCR experiments revealed that the increase of this regulatory protein can be attributed to the enhanced expression of actII-ORF4 mRNA (Okamoto et al., unpublished). Thus, certain rpsL mutations enhance expression of the actII-ORF4 gene, leading to massive production of Act. In addition to the rpsL mutations that confer a high level of resistance to streptomycin, another type of mutation (ND mutation) conferring a low-level resistance to streptomycin also gave rise to a similar increase in Act production (Hesketh and Ochi, 1997; Shima et al., 1996). The appearance of ND mutants is relatively frequent (105–107), and their ability to produce Act is noticeably higher than that of rpsL mutants. By using proteomic analysis, Okamoto et al. (2003) found that a 46-kDa protein is highly expressed in an S. coelicolor mutant KO-179, a representative strain of such ND mutants. This protein was identiﬁed as S-adenosylmethionine (SAM) synthetase, which is a product of the metK gene. These ﬁndings imply that SAM synthetase may be involved in the Act-overproducing phenotype observed in these ND mutants. In fact, the introduction of a high-copy-number plasmid containing the metK gene into wild-type strain resulted in an extensive hyperproduction of Act (Fig. 2A, see color insert). Furthermore, addition of SAM to the culture medium activated Act production in wild-type

160

OCHI et al.

FIG. 2. Activation of antibiotic (actinorhodin) production by introducing multi-copy metK gene encoding S-adenosylmethionine synthetase (A), or by directly adding S-adenosylmethionine (B) in S. coelicolor. (Adapted from Okamoto et al., 2003.)

cells (Fig. 2B). Therefore, enhanced Act production in strain KO-179 can be ascribed, at least in part, to the overexpression of SAM synthetase, which leads to an elevation of intracellular SAM level. Consistent with this conclusion, Kim et al. (2003) found that introduction of a multicopy plasmid containing the Streptomyces spectabilis metK gene into S. lividans can induce Act production in this organism. SAM is known to be the methyl donor for the methylation of the various biological substances (DNA, RNA, proteins, and other small molecules). However, the Act biosynthetic pathway does not contain any steps that require SAM as a methyl donor. Interestingly, overexpression of the metK gene stimulated the expression of a pathway-speciﬁc regulatory gene

RIBOSOME ENGINEERING

161

actII-ORF4 (Okamoto et al., 2003). Furthermore, the addition of SAM also caused overproduction of streptomycin (by Streptomyces griseus) and bicozamycin (by Streptomyces griseoﬂavus) (Saito et al., 2003a). These ﬁndings highlight the signiﬁcance of SAM as a common intracellular signal molecule for onset of secondary metabolism in widespread Streptomyces species. B. MECHANISM OF ACTIVATION The ribosomal protein S12, a component of the 30S subunit in bacteria, is best characterized with respect to its role in the selection efﬁciency of cognate tRNAs for an accuracy enhancement (Carter et al., 2000; Kurland et al., 1990). Most of the mutations in S12 protein associated with streptomycin resistance lead to an error restrictive phenotype because of changes in the kinetic properties of the tRNAribosome interaction (Kurland et al., 1996). As already mentioned, antibiotic production by bacteria, including Streptomyces spp., is activated or enhanced by introducing certain mutations into the rpsL gene (encoding the ribosomal protein S12) that confer resistance to streptomycin (Hosoya et al., 1998; Okamoto-Hosoya et al., 2003b; Tamehiro et al., 2003). Recently we found that K88E (which corresponds to position 87 in Escherichia coli S12 protein) rpsL mutant of S. coelicolor A3(2), with an enhanced Act production, exhibits an aberrant protein synthesis activity. To clarify the presence or absence of the causal relationship between this aberrant protein synthesis activity and the observed antibiotic overproduction, we have discovered characteristic properties of the S. coelicolor K88E mutant to synthesize protein in vivo and in vitro (Hosaka, Xu, and Ochi, unpublished; Okamoto-Hosoya et al., 2003a). The results demonstrated that (1) the K88E mutation, like classic S12 mutations (K42N and K42T) in E. coli, confers a restrictive phenotype in addition to resistance to streptomycin; (2) the K88E mutant exhibits a high level of protein synthesis activity in vivo at the late growth phases as examined by measuring the incorporation of labeled leucine; (3) the K88E mutant ribosomes from the latestationary-phase cells have a high capacity for translating both synthetic polynucleotide [poly(U)] and natural mRNA; (4) S150 solution from K88E mutant cells grown to late-stationary-phase supports a higher level of protein synthesis activity in vitro; and (5) the K88E mutant ribosomes are structurally more stable under stress conditions such as amino acid starvation and low concentration of magnesium. We concluded that the increased stability of the 70S complex and the levels of speciﬁc translation-associated factor(s) are responsible for the aberrant

162

OCHI et al.

FIG. 3. Outline of mechanism by which mutation in S12 protein activates antibiotic production.

activation of protein synthesis in the S. coelicolor K88E mutant (Fig. 3). This observation is in agreement with our recent ﬁndings that the E. coli K87E mutant also shows an aberrant protein synthesis activity at late growth phase (Fig. 4) (Hosaka et al., 2004). Our ﬁndings lead to the suggestion that a change from Lys to Glu at position 87 (according to the E. coli numbering) in S12 protein renders cells potentially more active for protein synthesis under the starvation conditions represented by the late growth phase. Such a characteristic would be highly advantageous for the production of proteins from newly transcribed genes (such as those involved in antibiotic production) at the late growth phase. Thus, the aberrant protein synthesis found in the S. coelicolor K88E rpsL mutant could be the cause, at least in major part, of remarkably activated antibiotic production in this mutant strain. C. rpsL MUTATIONS BY SITE-DIRECTED MUTAGENESIS The rpsL mutations found so far in S. coelicolor A3(2) and S. lividans 66 are K43N, K43R, K43T, K88E, K88R, and P91S. Of these, only two mutations (K88E and P91S) effectively activated antibiotic production.

RIBOSOME ENGINEERING

163

FIG. 4. In vitro translation of GFP mRNA with ribosomes from E. coli wild-type strain W3110 (s) and K87E mutant (d) grown to late stationary phase (26 h) in LB medium. Upper panel shows ﬂuorographs of GFP that was synthesized in vitro.

It thus appears that certain mutations around the Lys88 region may distinctively affect antibiotic production and that isolation of such mutants could be effective for developing the antibiotic-overproducing strains. However, since most of those mutations are not likely to confer resistance to streptomycin, it would be impossible to identify such mutations by virtue of their ability to resist the drug. To circumvent this difﬁculty, we used site-directed mutagenesis to create seven novel rpsL mutations (R86L, V87K, K88G, D89R, L90K, G92D, and R94G), and introduced these mutant genes into wild-type S. lividans cells by using a single-copy-number plasmid (Okamoto-Hosoya et al., 2003b). Of these mutations, two (L90K and R94G) activated production of a redpigmented antibiotic (undecylprodigiosin) by S. lividans much more potently than the streptomycin-resistant K88E mutation (Fig. 5, see color insert). Neither the L90K nor the R94G mutation conferred an increase in the level of resistance to streptomycin and paromomycin, indicating nonavailability of these mutant alleles among the resistant isolates. Since

164

OCHI et al.

FIG. 5. Mutations that were constructed by site-directed mutagenesis (upper panel) and effect of a single copy of the mutant rpsL gene on antibiotic (undecylprodigiosin) production. Red color represents undecylprodigiosin. (Adapted from Okamoto-Hosoya et al., 2003b.)

the experimental system chosen in this study to evaluate the new rpsL mutations involves expression of the mutant S12 protein with a plasmid and does not require any special technique for replacing the wild-type rpsL gene on the chromosome with the created mutant genes, this approach could be applicable to a variety of strains (even if no genetic information is available) for improvement of antibiotic productivity. IV. Antibiotic Overproduction by rpoB (RNA Polymerase) Mutations Antibiotic biosynthesis pathways and their genetic regulatory cascades comprise one of the most attractive ﬁelds in Streptomyces genetics and are important in considering strain improvement. Onset of the morphological differentiation and the secondary metabolism, including antibiotic production, are thought to be coupled and inﬂuenced by a variety of physiological and environmental factors (Chater and Bibb, 1997). Antibiotic production in streptomycetes is generally growth phase–dependent. Thus, the signal molecule for growth rate control, ppGpp, is suggested to play a central role in triggering the onset of antibiotic production in Streptomyces. Namely, the ribosomes play an essential role in adjusting gene expression levels by synthesizing ppGpp in response to nutrient limitation. There is a positive correlation between ppGpp and antibiotic biosynthesis: disruption of the

RIBOSOME ENGINEERING

165

FIG. 6. Functional map of E. coli RNA polymerase -subunit and location of rif cluster I within RNA polymerase -subunit in relation to the previously suggested ppGpp-binding site (Chatterji et al., 1998) in E. coli. Positions of the rif mutations found in the present study are designated by arrows. Numbering begins at the start codon of the open reading frame. ML, Mycobacterium leprae; SC, Streptomyces coelicolor A3(2); BS, Bacillus subtilis; EC, Escherichia coli. (Adapted from Ishihama et al., 1990, and Xu et al., 2002.)

ppGpp synthetase gene, relA, or a deletion mutation (designated as relC) in the ribosomal L11 protein gene has been shown to lead to a deﬁciency in ppGpp accumulation after amino acid depletion (socalled ‘‘relaxed’’ phenotype) accompanied by impairment in antibiotic production (Chakraburtty et al., 1996; Hoyt and Jones, 1999; Jin et al., 2004; Kelly et al., 1991; Ochi, 1987, 1990a,b). The expression level of many genes are regulated by ppGpp, either positively or negatively. Many genetic studies in E. coli suggested that RNA polymerase (RNAP) is the target for ppGpp regulation. Genetic analysis reveals that four major functional domains exist in the RNAP -subunit (Fig. 6). The ppGpp-sensitivity domain is close to another important domain of the RNAP -subunit, the rifampicin (Rif)-binding domain (Ishihama et al., 1990). The crystal structure clearly revealed that Rif-cluster I is involved in the E. coli RNAP active center. Therefore it is reasonable to consider that certain mutations in the Rif-binding domain could affect the activity of RNAP and then may affect the function of the adjacent ppGpp-binding domain. We postulated that the impaired ability to produce antibiotic due to the relA or relC mutation may be circumvented by introducing certain Rif-resistant (rif) mutations into the RNAP -subunit. This hypothesis is based on a notion that the mutated RNAPs may behave like ‘‘stringent’’ RNAP without ppGpp binding. The results from rel mutants of S. coelicolor A3(2) and S. lividans strongly supported this hypothesis (Lai et al., 2002; Xu et al., 2002). The Rif-resistant isolates from the rel mutants regained the ability to produce the colored antibiotic

166

OCHI et al.

actinorhodin, and various types of point mutation were mapped in the so-called Rif-cluster I in the rpoB gene that encodes the RNAP -subunit (Fig. 6). More impressively, gene expression analysis revealed that the restoration of actinorhodin production in the rel rif double mutant strains is accompanied by increased expression of the pathway-speciﬁc regulatory gene actII-ORF4, which normally decreased in the rel mutants. Accompanying the restoration of antibiotic production, the rel rif mutants also exhibited a lower rate of RNA synthesis compared to the parental strain when grown in a nutritionally rich medium. Since the dependence of S. coelicolor A3(2) on ppGpp to initiate antibiotic production can apparently be bypassed by certain mutations in the RNAP, the mutant RNAP may function by mimicking the ppGpp-bound form (Fig. 7). This proposal can be supported by the fact that the mutant RNAP behaved like ‘‘stringent’’ RNAP with respect to RNA synthesis, as demonstrated using cells growing in a nutritionally rich medium. To test the feasibility of rif mutation on the breeding of an antibiotic producing strain, Hu et al. (2002) attempted to activate the antibiotic biosynthetic gene cluster in S. lividans. The results demonstrated

FIG. 7. Hypothesis for ppGpp-independent antibiotic (actinorhodin) production. actIIORF4 is the gene encoding a pathway speciﬁc regulatory protein ActII-ORF4. rif represents mutated -subunit. relA and relC are mutations that block the synthesis of ppGpp.

RIBOSOME ENGINEERING

167

that the biosynthesis of actinorhodin, undecylprodigiosin, and calcium-dependent antibiotic all could be remarkably activated by introducing speciﬁc types of rif mutations into the rpoB gene. In most cases, the spontaneously arising rif mutations are mapped in the Rif-cluster I of the -subunit (Jin and Gross, 1988; Severinov et al., 1993, 1994; Singer et al., 1993), and cluster I is only a few angstroms away from the active center of RNAP, as demonstrated in E. coli (Severinov et al., 1995). It is reasonable to consider that the RNAP containing a rif-type -subunit may be structurally similar to a wild-type RNA polymerase modiﬁed by ppGpp, because numerous genetic analyses in E. coli have revealed that rif mutations frequently circumvent the ppGpp0 phenotype (Barker et al., 2001a,b). Recent developments in clariﬁcation of the ternary structure of RNA polymerase–ppGpp complex by X-ray analysis (Artsimovich et al., 2004) are helpful in assessing the foregoing consideration. V. Effect of str and rpoB Mutations in Various Bacteria A. ANTIBIOTIC OVERPRODUCTION Members of the genera Streptomyces, Bacillus, and Pseudomonas are soil bacteria that produce a high number of agriculturally and medically important antibiotics. The development of rational approaches to improve the production of antibiotics from these organisms is therefore of considerable industrial and economic importance. The impairment in antibiotic production resulting from a relA or relC mutation (that causes a failure to synthesize ppGpp) could be completely restored by introducing mutations conferring resistance to streptomycin (str) (Ochi et al., 1997; Shima et al., 1996). No accompanying restoration of ppGpp synthesis was detected in these relA str or relC str mutants. It is therefore apparent that acquisition of certain str mutations allows antibiotic production to be initiated without the requirement for ppGpp. This offers a possible strategy for improving the antibiotic productivity. Indeed, in addition to actinorhodin production by S. coelicolor and S. lividans, introduction of a str mutation was effective in enhancing antibiotic production by various bacteria. 1. Effect of str Mutation on Streptomyces spp. The effect of an str mutation on antibiotic production in three Streptomyces spp., S. chattanoogensis, S. antibioticus, and S. lavendulae was examined (Table I; Hosoya et al., 1998). When the spores of Streptomyces spp. were spread and incubated on a plate containing 5 or

168

OCHI et al. TABLE I ANTIBIOTIC PRODUCTIVITY OF BACTERIAL STREPTOMYCIN-RESISTANT MUTANTS Frequency of str mutants producing increased antibiotic

Microorganism

Antibiotic

Production in parental (¼ wild-type) strain (g/ml)

Streptomyces chattanoogensis

Fredericamycin

10

46%

260

Streptomyces antibioticus

Actinomycin

12

4%

63

Streptomyces lavendulae

Formycin

25

3%

130

Pseudomonas pyrrocinia

Pyrrolnitrin

30%

15

Bacillus cereus

FR900493

7%

550

Bacillus subtilis

Unidentiﬁed antibiotic

1.5 76 8a

19%

Highest productivity detected (g/ml)

80a

a unit/ml. Adapted from Hosoya et al., 1998.

30 g of streptomycin per ml, streptomycin resistant (str) mutants developed after 7–14 days at a frequency of 106 to 108. These spontaneous str mutants were characterized from a wide variety of colonies by size, morphology, and pigment formation. In S. chattanoogensis nearly half of the str mutants tested exhibited a signiﬁcantly increased ability (greater than ﬁvefold) to produce fredericamycin. The highest productivity detected was 26-fold higher than that of the wild-type strain. Similarly, strains producing high levels of actinomycin and formycin could be detected at a relatively high frequency (3%–4%) among str mutants of S. antibioticus and S. lavendulae, respectively. Thus, like actinorhodin production by S. coelicolor A3(2), introduction of mutations conferring resistance to streptomycin was effective for improving the antibiotic productivity of the Streptomyces spp. 2. Effect of str Mutations on Bacillus and Pseudomonas Introduction of the str mutation also improved antibiotic productivity of bacteria such as Bacillus spp. and Pseudomonas spp. (Table I; Hosoya et al., 1998). In B. cereus and P. pyrrocinia, the frequency of antibiotic overproducing strains among str mutants ranged from 7% to 30%. str mutants of B. subtilis developed on GYM agar containing

RIBOSOME ENGINEERING

169

either a low (5 g/ml) or a high (400 g/ml) concentration of streptomycin were examined. Antibiotic overproducing strains were detected at a higher frequency among str mutants selected at a high concentration, rather than a low concentration, of streptomycin (19% versus 3%). In contrast to str mutants, none of the mutants resistant to chloramphenicol, tetracycline, lincomycin, or spectinomycin exhibited increased antibiotic production. Unlike the case of S. coelicolor, the str mutations found in B. subtilis all fell into amino acid position K56 (corresponding to K42 in E. coli) of ribosomal protein S12. Mutations K56R, K56T, and K56Q were effective in increasing antibiotic productivity, whereas mutations K56I and K56N were ineffective. B. ENZYME OVERPRODUCTION Introduction of drug-resistant mutations has also been veriﬁed to be effective in improving enzyme productivity. Several str mutants of B. subtilis were shown to produce an increased amount (20–30%) of -amylase and protease (Kurosawa and Ochi, unpublished). Jorgensen et al. (personal communication) showed that rpoB mutations are also effective for overproduction (1.5-fold to twofold) of extracellular enzymes such as amylase and protease. Thus these methods may be applicable for overproduction of other enzymes produced by various microorganisms, especially at late growth phase. C. NEOTREHALOSADIAMINE Neotrehalosadiamine (3,30 -diamino-3,30 -dideoxy-,-trehalose; NTD), which is an aminosugar antibiotic produced by Bacillus pumilus and Bacillus circulans, inhibits growth of Staphylococcus aureus and Klebsiella pneumoniae. In contrast, Bacillus subtilis normally does not produce this antibiotic. However, introduction of a certain rifampicinresistant rpoB mutation (rif ) enables cells to activate the dormant ability to produce NTD in B. subtilis (Inaoka et al., 2004). A polycistronic gene, ntdABC, and a monocistronic gene, ntdR, were identiﬁed as the NTD biosynthesis operon and a positive regulator for ntdABC, respectively (Fig. 8). Surprisingly, NTD acts as autoinducer for its own biosynthetic process. The mechanism of autoinduction of signaling molecules has been studied extensively in several bacteria in relation to quorumsensing systems that are important for various physiological processes, such as acquisition of competence, sporulation, motility, bioﬁlm formation, bioluminescence, and virulence (Fuqua et al., 2001; Kaiser and Losick, 1993; Kleerebezem and Quadri, 2001). In Gram-positive

170

OCHI et al.

FIG. 8. Scheme of mechanism by which an antibiotic neotrehalosadiamine (NTD) synthesis is regulated via an autoinduction system in Bacillus subtilis. NtdR protein represents a positive regulator for NTD synthesis. In the absence of NTD, NtdR protein cannot function as an activator for ntdABC expression, although NtdR itself can bind to promoter region (upper panel).

RIBOSOME ENGINEERING

171

bacteria, most of these signaling molecules are peptides or modiﬁed peptides including subtilin and nisin; these antibiotics induce the transcription of their own biosynthesis genes via a two-component signal transduction system. In contrast, NTD acts as an autoinducer for its own biosynthesis operon by directly interacting with NtdR protein (Fig. 8). In our model, the NtdR protein dramatically stimulates ntdABC transcription when the promoter region is occupied by NTD-bound NtdR protein. In line with this notion, NTD-unbound NtdR protein results in a failure to express ntdABC. In addition, this rif mutation was also shown to cause a twofold increase in the activity of the ntdABC promoter recognized by house-keeping sigma factor (A). Therefore this mutant RNAP likely enhances the A-dependent promoters, resulting in a dramatic activation of the NTD biosynthesis pathway by an autoinduction mechanism. On the basis of these ﬁndings, together with the results from Streptomyces spp., improvement of RNAP by introduction of a rif mutation could be a useful approach to elicit the bacterial ability. VI. Increase of Chemical Tolerance in Pseudomonas Solvent resistance of bacteria is inheritable and can be generated by random mutagenesis and thus appears to be a useful phenotype for waste processing (Timmis et al., 1994). Therefore development of genetically enhanced microorganisms, through the use of genetic engineering, is important for developing better biological waste processing technologies. The ribosome engineering approach was effective not only in overproduction of useful metabolites but also in improvement of tolerance by bacteria of aromatic compounds, a property that could be useful for bioremediation. Certain str, gen, or rif mutants derived from Pseudomonas putida, which are resistant to streptomycin, gentamicin or rifampicin, respectively, were tolerant to the aromatic compound 4-hydroxybenzoate (4HBA) (Hosokawa et al., 2002). The minimum inhibitory concentration (MIC) of 4HBA for the wild-type strain was 1%, whereas the MIC for mutants was 1.7%. Frequency of 4HBA-tolerant mutants among spontaneous str, gen, and rif mutants was 5–15%, 3–5%, and 3% respectively. These 4HBA-tolerant mutants also tolerated a variety of organic chemicals such as 3-hydroxybenzoate, aliphatic and heterocyclic compounds, chlorobenzoates, as well as the organic solvents toluene and m-xylene. The str mutants had a point mutation in the rpsL gene. Interestingly, str, gen, and rif-phenotypes occurred in spontaneous 4HBA-tolerant mutants that had been selected by successively increasing concentrations (from 0.8% to 5%) of 4HBA, implying that

172

OCHI et al.

breeding approaches by traditional mutagenesis may often involve mutations in the ribosomal component or RNA polymerase. Uptake experiments using [14C]-4HBA revealed that the apparent ability of 4HBA to be taken up by the membrane transport system was reduced twofold to threefold in the mutants compared to the wild-type strain, accounting at least partly for enhanced tolerance to 4HBA (Hosokawa et al., 2002). As the apparent uptake rate is the sum of the actual uptake (inﬂux) and efﬂux, it is possible that these drug-resistant mutants have acquired a capacity to overproduce wide-speciﬁcity efﬂux pumps by somehow enhancing the expression of the key genes (such as marA, mexAB) that play a role in the efﬂux system. These ﬁndings might help in elucidating the mechanisms of tolerance to toxic organic chemicals, particularly its relationship to transcription and translation machinery. The efﬁcacy of str and rif mutations for elevating the cells’ tolerance to organic solvents such as toluene and xylene suggests a widespread applicability of ribosome engineering in improving the ability of microorganisms to process chemical wastes. VII. Combined Drug-Resistance Mutations A. MODEL EXPERIMENT Introduction of combined drug-resistant mutations was found to be quite effective in increasing the productivity of antibiotics in a hierarchical order (Hu and Ochi, 2001). The increased productivity of actinorhodin by sequential introduction of str, gen, and rif in S. coelicolor A3(2) is shown in Fig. 9A. Mutants with enhanced (1.6-fold to threefold higher) actinorhodin production were detected

FIG. 9. Hierarchical increase of antibiotic production by introducing combined drugresistance mutations in Streptomyces coelicolor wild-type strain 1147 (A) and Streptomyces albus industrial strain SAM-X (B), which produces a high amount (10 mg/ml) of salinomycin. (Adapted from Hu and Ochi, 2001, and Tamehiro et al., 2003.)

RIBOSOME ENGINEERING

173

at a high frequency (5–10%) among isolates resistant to streptomycin (str), gentamicin (gen), or rifampin (rif ), which developed spontaneously on agar plates that contained one of the three drugs. Construction of double mutants (str gen and str rif ) by introducing gentamicin or rifampin resistance into an str mutant resulted in further increased (1.7- to 2.5-fold-higher) actinorhodin productivity. Likewise, triple mutants (str gen rif ) thus constructed were found to have an even greater ability for producing the antibiotic, eventually generating a mutant able to produce 48 times more actinorhodin than the wild-type strain. Although analysis of str mutants revealed that a point mutation occurred within the rpsL gene, mutation points in gen mutants still remain unknown. These single, double, and triple mutants displayed in hierarchical order a remarkable increase in the production of Act II-ORF4, a pathway-speciﬁc regulatory protein (Hu and Ochi, 2001). The superior ability of the triple mutants was demonstrated by physiological analyses under various cultural conditions. Thus, using combined drug-resistant mutations, we can continuously increase the production of an antibiotic in a stepwise manner. Although much progress has been made in improving antibiotic producers (Chater, 1990; Lai et al., 1996; Lee et al., 1999), our method is characterized by the host cell’s amenability (generation of spontaneous drug-resistant mutation) and the method’s applicability to a number of microorganisms. It should also be emphasized that combined resistant mutations (triple mutations) demonstrated no signiﬁcant impairment in growth or sporulation. Antibiotic production is in general subjected to the suppressive effects caused by an excess of nutrients such as carbon, nitrogen, and phosphate sources. In particular, ammonium and phosphate both appear to be major regulators of antibiotic production, and their control systems may be interrelated in some way. Consistent with this notion, actinorhodin production in wild-type and mutant strains is more or less medium dependent; it is greater in R4 medium (containing less yeast extract and phosphate) than in R3 medium. The triple (str gen rif ) mutants revealed less sensitivity to such suppressive effects. B. APPLICABILITY TO INDUSTRIAL STRAIN Unlike the wild-type strains discussed earlier, improvement of industrial strains is, in general, much more difﬁcult, as productivity has already been raised by various genetic and physiological approaches. Nevertheless, demonstration of the efﬁcacy of drug-resistant mutation in industrial strains is intriguing, because improvement of industrial

174

OCHI et al.

strains is linked directly with economic aspects. Working with a Streptomyces albus strain that had previously been bred to produce industrial amounts (10 mg/ml) of salinomycin, the efﬁcacy of introducing drug-resistant mutations for further strain improvement has been demonstrated (Tamehiro et al., 2003). Mutants with enhanced salinomycin production were detected at a high incidence (7–12%) among spontaneous isolates resistant to streptomycin, gentamicin, or rifampicin. Finally, we demonstrated improvement of the salinomycin productivity of the industrial strain by 2.3-fold by introducing a triple mutation (Fig. 9B). The str mutant was shown to have a point mutation within the rpsL gene (encoding ribosomal protein S12). Likewise, the rif mutant possessed a mutation in the rpoB gene (encoding the RNA polymerase subunit). Combined drug-resistant mutations (triple mutation) caused no impairment of growth and sporulation. Rather, the ability to produce aerial mycelium and spores was, in fact, enhanced. Strikingly, the str mutant with increased salinomycin production exhibited high translation activity at the stationary phase as determined with the in vitro translation assay system (Tamehiro et al., 2003) (Fig. 10). This high-translation activity could be a reason why the str mutant is capable of producing a greater amount of salinomycin. The aberrant protein synthesis ability of the str mutant resulted presumably

FIG. 10. In vitro translation activities of ribosomes prepared from S. albus cells at various growth phase. Translation activities were determined with poly(U)-directed cell-free translation system. (Adapted from Tamehiro et al., 2003.)

RIBOSOME ENGINEERING

175

from a more stable ribosome structure, as demonstrated in the presence of a low Mg2þ concentration. Antibiotic production, including salinomycin production, usually commences at the late growth phase (i.e., transition phase or stationary phase). Therefore the enhanced protein synthesis ability at late growth phase would promote the initiation processes (i.e., formation of positive regulatory proteins for antibiotic gene expression) or biosynthetic processes (or both).

VIII. Conclusion and Future Prospects In this review, we demonstrated that a cell’s function can be altered dramatically by modulating the ribosome using a drug-resistance mutation technique. Our approach is characterized by focusing on ribosomal function at late growth phase (i.e., stationary phase). The importance of this phase has been largely overlooked in studies of ribosomes, with a few exceptions (such as a 1990 work by Wada et al.). In summary, our novel breeding approach is based on two different aspects, modulation of the translational apparatus by induction of str and gen mutations, and modulation of the transcriptional apparatus by induction of a rif mutation (Fig. 11, see color insert). Modulation of these two mechanisms may function cooperatively to increase antibiotic productivity. Introduction of mutations conferring resistance to fusidic acid (fus) or thiostrepton (tsp), though not yet published, also causes activation of antibiotic production as well as str mutation. Moreover, these fus and tsp mutations were found to give rise to an aberrant protein synthesis activity, as did the str mutant ribosome (T. Hosaka and K. Ochi, unpublished). Resistance to fusidic acid and thiostrepton is known to come frequently from a mutation in elongation factor G and ribosomal protein L11, respectively. However, no mutations were found within the genes encoding elongation factor G or ribosomal protein L11. It is therefore highly likely that these fus and tsp mutations are located on the genes encoding rRNAs. This is important because it implies the existence of a new way to modulate ribosomal function, in addition to ribosomal protein mutations. In the study of current topics, we have found several important facts (or phenomena), which might be useful in eliciting the cell’s ability. These facts, together with the aforementioned studies, encourage us to construct more elegantly designed and more widely applicable ribosome engineering in the near future.

176

OCHI et al.

FIG. 11. Scheme of ‘‘ribosome engineering’’ to activate cell’s ability.

A. FUTURE PROSPECTS 1. EshA Protein That Affects Developmental Processes Kwak et al. (2001) and our laboratory (Kawamoto et al., 2001; Saito et al., 2003b) independently found a novel 52-kDa protein that is produced during the late growth phase. The disruption of the gene (eshA), which codes for this 52-kDa protein (EshA), was shown to abolish antibiotic production in S. coelicolor A3(2) and S. griseus.

RIBOSOME ENGINEERING

177

Conversely, propagation of wild-type eshA with low-copy-number plasmids caused overproduction of antibiotics in both S. griseus and S. coelicolor (Kawamoto et al., 2001; Saito et al., 2003a). Thus it is evident that EshA plays an important role as a positive regulator of antibiotic production. It is also notable that our eshA null mutants show considerable similarity in phenotype to the relA mutant, as characterized by severely impaired ability to produce antibiotics. In fact, eshA null mutant showed a lower level of ppGpp compared to wild-type strain, and the impaired ability to produce antibiotic was completely restored by propagating the relA gene, accompanied by an increase in ppGpp (Saito and Ochi, unpublished). The EshA protein was found to exist as a multimer (20-mers) creating a cubiclike structure with a diameter of 27 nm, possibly forming an icosahedron. EshA may offer a feasible target for strain improvement, because genes homologous to eshA appear to be widely distributed among streptomycetes. 2. ppGpp-GTP Dual Control in B. subtilis Bacilysin is one of the simplest peptide antibiotics produced by B. subtilis. Recently it has been reported that CodY protein regulates the expression of various stationary-phase genes by sensing the intracellular GTP level (Inaoka and Ochi, 2002; Ratnayake-Lecamwasam et al., 2001) (Fig. 12). In fact, sporulation and genetic competence development can be initiated by addition of decoyinine (a GMP synthetase inhibitor) or by inactivation of CodY. Likewise, bacilysin production is apparently controlled by CodY protein, since codY disruption increased the transcription of the bacilysin biosynthesis cluster (Inaoka et al., 2003). However, a codY relA double mutant does not produce bacilysin, indicating that ppGpp plays a pivotal role as a positive regulator even in B. subtilis, and that GTP functions as a negative regulator, producing a synergistic effect on antibiotic production when its level declined (Fig. 12). Thus, unlike antibiotic production in Streptomyces spp., bacilysin production in B. subtilis is controlled by a ‘‘dual regulation’’ system composed of the guanine nucleotides ppGpp and GTP. The codY gene may be a feasible target for activation of secondary metabolism, since many low GþC Gram-positive bacteria contain the CodY homologue. 3. Existence of ppGpp in Plants Plants have a complex signal transduction network activated in response to such stressful conditions as pathogenic infection, wounding, heat shock, drought, and high salinity (Ryan and Moura, 2002;

178

OCHI et al.

FIG. 12. Scheme of a dual regulation system by ppGpp and GTP in bacilysin biosynthesis. ppGpp promotes bacilysin production, while GTP inhibits it via CodY protein.

Xiong et al., 2002). Despite the apparent signiﬁcance of ppGpp in bacterial gene expression, the importance of ppGpp in plant biology has been largely overlooked. Recently, we unambiguously demonstrated that ppGpp is produced in the chloroplasts of plant cells in response to stressful conditions (Takahashi et al., 2004). Levels of ppGpp increased markedly when plants were subjected to such biotic and abiotic stresses as wounding, heat shock, high salinity, acidity, heavy metal, drought, and ultraviolet irradiation. Abrupt changes from light to dark also caused a substantial elevation in ppGpp levels. Elevation of ppGpp levels was also elicited by treatment with the plant horomones jasmonic acid, abscisic acid and ethylene. In vitro, chloroplast RNA polymerase activity was inhibited in the presence of ppGpp, demonstrating the existence of a bacterial-type stringent response in plants. Given the signiﬁcance of ppGpp in bacterial physiology, ppGpp apparently plays a critical role in adjusting plant physiology. Our ﬁnding is consistent with recent work demonstrating the existence of RelA homologs (At-RSH1 and Cr-RSH) in Arabidopsis and Chlamydomonas (Kasai et al., 2002; van der Biezen et al., 2000). Understanding

RIBOSOME ENGINEERING

179

in detail the functions of ppGpp and the signaling cascades it activates is clearly an important challenge that should provide important insight into plant evolution and adaptation to environmental changes, and should also make a contribution in the area of plant breeding. 4. Applicability of Ribosome Engineering to Cell-Free Translation Systems In contrast to the utilization of whole cell, cell-free translation systems have been shown to achieve high-throughput production of a wide variety of foreign-originated proteins. Studies to improve the productivity of protein synthesis in vitro have employed several strategies, such as (1) the development of continuous ﬂow in vitro protein synthesis (Spirin et al., 1988), (2) a source for preparing cell-free extracts (Baranov and Spirin, 1993; Endo et al., 1992), and (3) optimization of translation components, reaction condition, and generation and consumption of an energy source (Kim and Swartz, 2001; Kim et al., 1996; Sawasaki et al., 2002). However, there have been no reports focusing on the ribosome in terms of high-throughput production. Recently, to examine the effects of ribosomal protein S12 mutations on the efﬁciency of cell-free protein synthesis, we isolated a wide variety of str mutants from E. coli and found that a mutant replacing Lys-42 with Thr (K42T) in the S12 protein shows higher (1.3-fold to twofold) protein production than the wild-type (Chumpolkulwong et al., 2004). Therefore, the method of ribosome engineering may be effective as one of the strategies to enhance protein production in E. coli-based (and even in wheat germ-based) cell-free translation systems.

ACKNOWLEDGMENTS This work was supported by a grant to K. Ochi (for the project ‘‘Construction of Ribosome Engineering’’) from the Organized Research Combination System of Education, Culture, Sports, Science and Technology of Japan. The authors are grateful to S. Kawamoto, Y. Hosoya, H. Hu, K. Matsubara, and K. Hosokawa for their contributions in promoting the project, and to Prof. S. Yokoyama (RIKEN Institute) for valuable discussions and Prof. T. Fukui for encouragement through the project.

REFERENCES Arias, P., Ferna´ndez-Moreno, M. A., and Malpartida, F. (1999). Characterization of the pathway-speciﬁc positive transcriptional regulator for actinorhodin biosynthesis in Streptomyces coelicolor A3(2) as a DNA-binding protein. J. Bacteriol. 181, 6958–6968.

180

OCHI et al.

Artsimovich, I., Patlan, V., Sekine, S., Vassyleva, M. N., Hosaka, T., Ochi, K., Yokoyama, S., and Vassylev, D. G. (2004). Structural basis for transcription regulation by alarmone ppGpp. Cell 117, 299–310. Baranov, V. I., and Spirin, A. S. (1993). Gene expression in cell-free systems on preparative scale. Methods Enzymol. 217, 123–142. Barker, M. M., Gaal, T., and Gourse, R. L. (2001a). Mechanism of regulation of transcription initiation by ppGpp. II. Models for positive control based on properties of RNAP mutants and competition for RNAP. J. Mol. Biol. 305, 689–702. Barker, M. M., Gaal, T., Josaitis, C. A., and Gourse, R. L. (2001b). Mechanism of regulation of transcription initiation by ppGpp. I. Effects of ppGpp on transcription initiation in vivo and in vitro. J. Mol. Biol. 305, 673–688. Bibb, M. (1996). The regulation of antibiotic production in Streptomyces coelicolor A3(2). Microbiology 142, 1335–1344. Carter, A. P., Clemons, W. M., Brodersen, D. E., Morgan-Warren, R. J., Wimberly, B. T., and Ramakrishnan, V. (2000). Functional insights from the structure of the 30S ribosomal subunit and its interactions with antibiotics. Nature 407, 340–348. Cashel, M., Gentry, D. R., Hernandez, V. J., and Vinella, D. (1996). The stringent response. In ‘‘Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology’’ (F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger, eds.), 2nd ed., Vol. 1, pp. 1458–1496. American Society for Microbiology, Washington, D.C. Chakraburtty, R., White, J., Takano, E., and Bibb, M. J. (1996). Cloning, characterization and disruption of a (p)ppGpp synthetase gene (relA) of Streptomyces coelicolor A3(2). Mol. Microbiol. 19, 357–368. Chater, K. F. (1990). The improving prospects for yield increase by genetic engineering in antibiotic-producing streptomycetes. Bio/Technology 8, 115–121. Chater, K. F and Bibb, M. J. (1997). Regulation of bacterial antibiotic production. In ‘‘Biotechnology, Products of Secondary Metabolism’’ (H. Kieinkauf, H. von Dohren, eds.), Vol. 7, pp. 57–105. VCH, Weinheim, Germany. Chatterji, D., Fujita, N., and Ishihama, A. (1998). The mediator for stringent control, ppGpp, binds to the -subunit of Escherichia coli RNA polymerase. Genes Cells 3, 279–287. Chumpolkulwong, N., Hori-Takemoto, T., Hosaka, T., Inaoka, T., Kigawa, T., Shirouzu, M., Ochi, K., and Yokoyama, S. (2004). Effects of Escherichia coli ribosomal protein S12 mutations on cell-free protein synthesis. Eur. J. Biochem 271, 1127–1134. Endo, Y., Otsuzuki, S., Ito, K., and Miura, K. (1992). Production of an enzymatic active protein using a continuous ﬂow cell-free translation system. J. Biotechnol. 25, 221–230. Floriano, B., and Bibb, M. (1996). afsR is a pleiotropic but conditionally required regulatory gene for antibiotic production in Streptomyces coelicolor A3(2). Mol. Microbiol. 21, 385–396. Fuqua, C., Parsek, M. R., and Greenberg, E. P. (2001). Regulation of gene expression by cell-to-cell communication: Acyl-homoserine lactone quorum sensing. Annu. Rev. Genet. 35, 439–468. Gramajo, H. C., Takano, E., and Bibb, M. J. (1993). Stationary-phase production of the antibiotic actinorhodin in Streptomyces coelicolor A3(2) is transcriptionally regulated. Mol. Microbiol. 7, 837–845. Hesketh, A., and Ochi, K. (1997). A novel method for improving Streptomyces coelicolor A3(2) for production of actinorhodin by introduction of rpsL (encoding ribosomal protein S12) mutations conferring resistance to streptomycin. J. Antibiot. 50, 532–535.

RIBOSOME ENGINEERING

181

Hopwood, D. A., Chater, K. F., and Bibb, M. J. (1995). Genetics of antibiotic production in Streptomyces coelicolor A3(2), a model streptomycete. Bio/Technology 28, 65–102. Horinouchi, S., Kito, M., Nishiyama, M., Furuya, K., Hong, S. K., Miyake, K., and Beppu, T. (1990). Primary structure of AfsR, a global regulatory protein for secondary metabolite formation in Streptomyces coelicolor A3(2). Gene 95, 49–56. Hosaka, T., Tamehiro, N., Chumpolkulwong, N., Hori-Takemoto, C., Shirouzu, M., Yokoyama, S., and Ochi, K. (2004). The novel mutation K87E in ribosomal protein S12 enhances protein synthesis activity during the late growth phase in Escherichia coli. Mol. Genet. Genomics. (in press). Hosokawa, K., Park, N. H., Inaoka, T., Itoh, Y., and Ochi, K. (2002). Streptomycinresistant (rpsL) or rifampicin-resistant (rpoB) mutation in Pseudomonas putida KH146-2 confers enhanced tolerance to organic chemicals. Environ. Microbiol. 4, 703–712. Hosoya, Y., Okamoto, S., Muramatsu, H., and Ochi, K. (1998). Acquisition of certain streptomycin-resistant (str) mutations enhances antibiotic production in bacteria. Antimicrob. Agents Chemother. 42, 2041–2074. Hoyt, S., and Jones, G. H. (1999). relA is required for actinomycin production in Streptomyces antibioticus. J. Bacteriol. 181, 3824–3829. Hu, H., and Ochi, K. (2001). Novel approach for improving the productivity of antibioticproducing strains by inducing combined resistant mutations. Appl. Environ. Microbiol. 67, 1885–1892. Hu, H., Zhang, Q., and Ochi, K. (2002). Activation of antibiotic biosynthesis by speciﬁed mutations in the rpoB gene (encoding RNA polymerase -subunit) of Streptomyces lividans. J. Bacteriol. 184, 3984–3991. Inaoka, T., and Ochi, K. (2002). RelA protein is involved in induction of genetic competence in certain Bacillus subtilis strains by moderating the level of intracellular GTP. J. Bacteriol. 184, 3923–3930. Inaoka, T., Takahashi, K., Ohnishi-Kameyama, M., Yoshida, M., and Ochi, K. (2003). Guanine nucleotides guanosine 50 -diphosphate 30 -diphosphate and GTP co-operatively regulate the production of an antibiotic bacilysin in Bacillus subtilis. J. Biol. Chem. 278, 2169–2176. Inaoka, T., Takahashi, K., Yada, H., Yoshida, M., and Ochi, K. (2004). RNA polymerase mutation activates the production of a dormant antibiotic 3, 30 -neotrehalosadiamine via an autoinduction mechanism in Bacillus subtilis. J. Biol. Chem. 279, 3885–3892. Ishihama, A., Fujita, N., Igarashi, K., and Ueshima, R. (1990). Structural and functional modulation of Escherichia coli RNA polymerase. In ‘‘Structure and Function of Nucleic Acids and Proteins’’ (Y. Felicia, H. Wu, and W. W. Cheng, eds.), pp. 153–159. Raven Press, New York. Jin, D. J., and Gross, C. A. (1988). Mapping and sequencing of mutations in the Escherichia coli rpoB gene that lead to rifampicin resistance. J. Mol. Biol. 202, 45–58. Jin, W., Ryu, Y. G., Kang, S. G., Kim, S. K., Saito, N., Ochi, K., Lee, S. H., and Lee, K. J. (2004). Two relA/spoT homologous genes are involved in the morphological and physiological differentiation of Streptomyces clavuligerus. Microbiology 150, 1485–1493. Kaiser, D., and Losick, R. (1993). How and why bacteria talk to each other. Cell 73, 873–885. Kasai, K., Usami, S., Yamada, T., Endo, Y., Ochi, K., and Tozawa, Y. (2002). A RelA-SpoT homolog (Cr-RSH) identiﬁed in Chlamydomonas reinhardtii generates stringent factor in vivo and localizes to chloroplasts in vitro. Nucl. Acids Res. 30, 4985–4992.

182

OCHI et al.

Kawamoto, S., Watanabe, M., Saito, N., Hesketh, A., Vachalova, K., Matsubara, K., and Ochi, K. (2001). Molecular and functional analyses of the gene (eshA) encoding the 52-kilodalton protein of Streptomyces coelicolor A3(2) required for antibiotic production. J. Bacteriol. 183, 6009–6016. Kelly, K. S., Ochi, K., and Jones, G. H. (1991). Pleiotropic effects of a relC mutation in Streptomyces antibioticus. J. Bacteriol. 173, 2297–2300. Kieser, T., Bibb, M. J., Buttner, M. J., Chater, K. F., and Hopwood, D. A. (2000). Practical Streptomyces genetics. John Innes Foundation, Norwich. Kigawa, T., Yabuki, T., Yoshida, Y., Tsutsui, M., Ito, Y., Shibara, T., and Yokoyama, S. (1999). Cell-free production and stable-isotope labeling of milligram quantities of proteins. FEBS Lett. 442, 15–19. Kim, D. J., Huh, J. H., Yang, Y. Y., Kang, C. M., Lee, I. H., Hyun, C. G., Oh, S. Y., Hong, S. K., and Suh, J. W. (2003). The accumulation of S-adenosyl-L-methionine enhances the production of actinorhodin but inhibits sporulation in Streptomyces lividans TK23. J. Bacteriol. 185, 592–600. Kim, D. M., and Swartz, J. R. (2001). Regeneration of adenosine triphosphate from glycolytic intermediates for cell-free protein synthesis. Biotech. and Bioeng. 74, 309–316. Kim, D. M., Kigawa, T., Choi, C. Y., and Yokoyama, S. (1996). A highly efﬁcient cell-free protein synthesis system from Escherichia coli. Eur. J. Biochem. 239, 881–886. Kim, E. S., Hong, H. J., Choi, C. Y., and Cohen, S. N. (2001). Modulation of actinorhodin biosynthesis in Streptomyces lividans by glucose repression of afsR2 gene transcription. J. Bacteriol. 183, 2198–2203. Kleerebezem, M., and Quadri, L. E. (2001). Peptide pheromone-dependent regulation of antimicrobial peptide production in Gram-positive bacteria: A case of multicellular behavior. Peptides 22, 1579–1596. Kurland, C. G., Jorgensen, F., Richter, A., Ehrenberg, M., Bilgin, N., and Rojas, A. M. (1990). Through the accuracy window. In ‘‘The Ribosome; Structure, Function, & Evolution’’ (J. R. Warner, ed.), pp. 513–526. ASM Press, Washington, D. C. Kurland, C. G., Hughes, D., and Ehrenberg, M. (1996). Limitations of translational accuracy. In ‘‘Escherichia coli and Salmonella, Cellular and Molecular Biology’’ (C. N. Frederick, ed.), pp. 979–1004. ASM Press, Washington, D.C. Kwak, J., McCue, L. A., Trczianka, K., and Kendrick, K. E. (2001). Identiﬁcation and characterization of a developmentally regulated protein, EshA, required for sporogenic hyphal branches in Streptomyces griseus. J. Bacteriol. 183, 3004–3015. Lai, C., Xu, J., Tozawa, Y., Okamoto-Hosoya, Y., Yao, X., and Ochi, K. (2002). Genetic and physiological characterization of rpoB mutations that activate antibiotic production in Streptomyces lividans. Microbiology 148, 3365–3373. Lai, R., Khanna, R., Kaur, H., Khanna, M., Dhingra, N., Lal, S., Gartemann, K. H., Eichenlaub, R., and Ghosh, P. K. (1996). Engineering antibiotic producers to overcome the limitations of classical strain improvement programs. Crit. Rev. Microbiol. 22, 201–255. Lee, S. H., and Rho, Y. T. (1999). Improvement of tylosin fermentation by mutation and medium optimization. Lett. Appl. Microbiol. 28, 142–144. Martı´nez-Costa, O. H., Arias, P., Romero, N. M., Parro, V., Mellado, R. P., and Malpartida, F. (1996). A relA/spoT homologous gene from Streptomyces coelicolor A3(2) controls antibiotic biosynthetic genes. J. Biol. Chem. 271, 10627–10634. Ochi, K. (1987). Metabolic initiation of differentiation and secondary metabolism by Streptomyces griseus: Signiﬁcance of the stringent response (ppGpp) and GTP content in relation to A-factor. J. Bacteriol. 169, 3608–3616.

RIBOSOME ENGINEERING

183

Ochi, K. (1990a). A relaxed (rel ) mutant of Streptomyces coelicolor A3(2) with a missing ribosomal protein lacks the ability to accumulate ppGpp, A-factor and prodigiosin. J. Gen. Microbiol. 136, 2405–2412. Ochi, K. (1990b). Streptomyces relC mutants with an altered ribosomal protein ST-L11 and genetic analysis of a Streptomyces griseus relC mutant. J. Bacteriol. 172, 4008–4016. Ochi, K., and Hosoya, Y. (1998). Genetic mapping and characterization of novel mutations which suppress the effect of a relC mutation on antibiotic production in Streptomyces coelicolor A3(2). J. Antibiot. 51, 592–595. Ochi, K., Zhang, D., Kawamoto, S., and Hesketh, A. (1997). Molecular and functional analysis of the ribosomal L11 and S12 protein genes (rplK and rpsL) of Streptomyces coelicolor A3(2). Mol. Gen. Genet. 256, 488–498. Okamoto, S., Lezhava, A., Hosaka, T., Okamoto-Hosoya, Y., and Ochi, K. (2003). Enhanced expression of S-adenosylmethionine synthetase causes overproduction of actinorhodin in Streptomyces coelicolor A3(2). J. Bacteriol. 185, 601–609. Okamoto-Hosoya, Y., Sato, T., and Ochi, K. (2000). Resistance to paromomycin is conferred by rpsL mutations, accompanied by an enhanced antibiotic production in Streptomyces coelicolor A3(2). J. Antibiot. 53, 1424–1427. Okamoto-Hosoya, Y., Hosaka, T., and Ochi, K. (2003a). An aberrant protein synthesis activity is linked with antibiotic overproduction in rpsL mutants of Streptomyces coelicolor A3(2). Microbiology 149, 3299–3309. Okamoto-Hosoya, Y., Okamoto, S., and Ochi, K. (2003b). Development of antibioticoverproducing strains by site-directed mutagenesis of the rpsL gene in Streptomyces lividans. Appl. Environ. Microbiol. 69, 4256–4259. Ratnayake-Lecamwasam, M., Serror, P., Wong, K. W., and Sonenshein, A. L. (2001). Bacillus subtilis CodY represses early-stationary-phase genes by sensing GTP levels. Genes. Dev. 15, 1093–1103. Ryan, C. A., and Moura, D. S. (2002). Systemic wound signaling in plants: A new perception. Proc. Natl. Acad. Sci. USA 99, 6519–6520. Saito, N., Kurosawa, K., Xu, J., Okamoto, S., and Ochi, K. (2003a). Effect of S-adenosylmethionine on antibiotic production in Streptomyces griseus and Streptomyces griseoﬂavus. Actinomycetologica 17, 47–49. Saito, N., Matsubara, K., Watanabe, M., Kato, F., and Ochi, K. (2003b). Genetic and biochemical characterization of EshA, a protein that forms large multimers and affects developmental processes in Streptomyces griseus. J. Biol. Chem. 21:278, 5902–5911. Sawasaki, T., Ogasawara, T., Morishita, R., and Endo, Y. (2002). A cell-free protein synthesis system for high-throughput proteomics. Proc. Natl. Acad. Sci. USA 99, 14652–14657. Severinov, K., Mustaev, A., Severinova, E., Kozlov, M., Darst, S. A., and Goldfarb, A. (1995). The subunit Rif-cluster I is only angstroms away from the active center of Escherichia coli RNA polymerase. J. Biol. Chem. 270, 29428–29432. Severinov, K., Soushko, M., Goldfarb, A., and Nikiforov, V. (1993). Rifampicin region revisited: new rifampicin-resistant and streptolydigin-resistant mutants in the subunit of Escherichia coli RNA polymerase. J. Biol. Chem. 268, 14820–14825. Severinov, K., Soushko, M., Goldfarb, A., and Nikiforov, V. (1994). RifR mutations in the beginning of the Escherichia coli rpoB gene. Mol. Gen. Genet. 244, 120–126. Shima, J., Hesketh, A., Okamoto, S., Kawamoto, S., and Ochi, K. (1996). Induction of actinorhodin production by rpsL (encoding ribosomal protein S12) mutations that

184

OCHI et al.

confer streptomycin resistance in Streptomyces lividans and Streptomyces coelicolor A3(2). J. Bacteriol. 178, 7276–7284. Singer, M., Jin, D. J., Walter, W. A., and Gross, C. A. (1993). Genetic evidence for the interaction between cluster I and cluster III rifampicin resistant mutations. J. Mol. Biol. 231, 1–5. Spirin, A. S., Baranov, V. I., Ryabova, L. A., Ovodov, S. Y., and Alakhov, Y. B. (1988). A continuous in vitro translation system capable of producing polypeptides in high yield. Science 242, 1162–1164. Takahashi, K., Kasai, K., and Ochi, K. (2004). Identiﬁcation of the bacterial alarmone guanosine 50 -diphosphate 30 -diphosphate (ppGpp) in plants. Proc. Natl. Acad. Sci. USA 101, 4320–4324. Tamehiro, N., Hosaka, T., Xu, J., Hu, H., Otake, N., and Ochi, K. (2003). Innovative approach for improvement of an antibiotic-overproducing industrial strain of Streptomyces albus. Appl. Environ. Microbiol. 69, 6412–6417. Timmis, K. N., Steffan, R. J., and Unterman, R. (1994). Designing microorganisms for the treatment of toxic wastes. Annu. Rev. Microbiol. 48, 525–557. van der Biezen, E. A., Sun, J., Coleman, M. J., Bibb, M. J., and Jones, J. D. (2000). Arabidopsis RelA/SpoT homologs implicate (p)ppGpp in plant signaling. Proc. Natl. Acad. Sci. USA 97, 3747–3752. Van der Meer, J. R., de Vos, W. M., Harayama, S., and Zehnder, A. J. (1992). Molecular mechanisms of genetic adaptation to xenobiotic compounds. Microbiol. Rev. 56, 677–694. Vinci, V. A., and Byng, G. (1999). ‘‘Manual of Industrial Microbiology and Biotechnology.’’ ASM Press, Washington, D. C. Vogtli, M., Chang, P. C., and Cohen, S. N. (1994). afsR2: A previously undetected gene encoding a 63-amino-acid protein that stimulates antibiotic production in Streptomyces lividans. Mol. Microbiol. 14, 643–653. Wada, A., Igarashi, K., Yoshimura, S., Aimoto, S., and Ishihama, A. (1995). Ribosome modulation factor: Stationary growth phase-speciﬁc inhibitor of ribosome functions from Escherichia coli. Biochem. Biophys. Res. Commun. 214, 410–417. Wietzorrek, A., and Bibb, M. (1997). A novel family of proteins that regulates antibiotic production in streptomycetes appears to contain an OmpR-like DNA-binding fold. Mol. Microbiol. 25, 1181–1184. Xiong, L., Schumaker, K. S., and Zhu, J. K. (2002). Cell signaling during cold, drought, and salt stress. Plant Cell 14(Suppl.), S165–S183. Xu, J., Tozawa, Y., Lai, C., Hayashi, H., and Ochi, K. (2002). A rifampicin resistance mutation in the rpoB gene confers ppGpp-independent antibiotic production in Streptomyces coelicolor A3(2). Mol. Genet. Genomics 268, 179–189. Zhang, Y. X., Perry, K., Vinci, V. A., Powell, K., Stemmer, W. P., and del Cardayre, S. B. (2002). Genome shufﬂing leads to rapid phenotypic improvement in bacteria. Nature 415, 644–646.

Developments in Microbial Methods for the Treatment of Dye Effluents R. C. KUHAD,* N. SOOD,* K. K. TRIPATHI,{ A. SINGH,{,}

AND

O. P. WARD{

*

Department of Microbiology, University of Delhi New Delhi–110 021, India

{

Department of Biotechnology, Ministry of Science and Technology New Delhi–110 003, India {

}

Department of Biology, University of Waterloo Waterloo, Ontario N2L 3G1, Canada

Author for correspondence. E-mail: [email protected]

I. Introduction II. Conventional Methods A. Physical B. Chemical III. Microbial Methods A. Biosorption B. Aerobic Biodegradation C. Anaerobic Biodegradation D. Combined Anaerobic/Aerobic Biodegradation IV. Enzymatic Methods V. Conclusion References

185 186 188 190 191 191 193 201 203 205 206 206

I. Introduction Synthetic chemical dyes are extensively used for textile, paper printing, color photography, cosmetic, pharmaceutical, and leather industries. Dyes are of environmental interest because of their potential of forming toxic aromatic amines. Over the last two decades there has been a tremendous increase in awareness of the toxic and carcinogenic effects of many polluting chemicals that were not considered hazardous in the past (King, 1997). More than 10,000 different dyes and pigments are used industrially with a production of 7 105 tons of these dyes per year (Zollinger, 1987). It is estimated that 2–50% of these dyes are lost into wastewaters, depending on the class of dye used (O’Neill et al., 1999). Since most of these dyes are persistent environmental pollutants, they are not removed from industrial efﬂuents by conventional wastewater treatments (Cripps et al., 1990; Moran et al., 1997; Willmott et al., 1998). Dyes may be toxic and mutagenic, and if they are discharged directly into the environment, they contaminate not only the environment but also traverse through the entire food chain, leading to biomagniﬁcation. Through environmental legislation, tougher controls are being applied regarding the requirement to remove dyes from industrial efﬂuents. 185 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

186

KUHAD et al.

Industrial dyes are classiﬁed as anionic (direct, acid, and reactive dyes), cationic (basic dyes), and nonionic (disperse dyes) (Dawson, 1981; Mishra and Tripathy, 1993). Anionic and nonionic dyes mostly contain azo or anthraquinone type chromophores. Azo dyes are the most widely used, accounting for more than 60% of the total number of dyes known to be manufactured, and toxic amines in the industrial efﬂuents are the result of reductive cleavage of azo linkages. Anthraquinone dyes are more resistant to degradation because of the fused aromatic structure. High brilliance and intensity of colors make basic dyes highly visible even at low concentration. Chromium-containing metal complex dyes are carcinogenic in nature. More than 90% of about 4,000 dyes tested in a survey had LD50 values >2000 mg/kg with highest toxicities being found among basic and diazo dyes (Shore, 1996). Table I shows examples of some common azo dyes used for microbial dye decolorization studies. Textile industry efﬂuents are characterized as having the high visible color (3000–4500 units), chemical oxygen demand (800–1600 mg/L), alkaline pH range of 9–11, and total solids (6000–7000 mg/L) (Manu and Chaudhari, 2002). A variety of effective physical and chemical treatment methods are available (Nigam et al., 2000; Robinson et al., 2001a). There has been considerable interest in development of biological methods (microbial and enzymatic), because these methods are considered attractive because of their potential low-cost, environmental compatibility, and public acceptability (Dubin and Wright, 1975; Paszczynski and Crawford, 1991). A wide variety of microorganisms, including bacteria, fungi, and algae, are capable of decolorizing a diverse range of dyes (McMullan et al., 2001). Many bacteria are able to degrade dyes both aerobically and anaerobically. Biodegradation of azo dyes by bacteria is often initiated by azoreductase-driven cleavage of azo bonds, followed by aerobic or anaerobic degradation of resulting amines (Stolz, 2001). On the other hand, fungal degradation typically originates from the lignolytic activity to degrade azo dyes aerobically with the aid of lignin peroxidase (Fu and Viraraghavan, 2001). In this chapter, various dye decolorization methods and developments in biological treatment methods for dye are critically reviewed. II. Conventional Methods Physical and chemical methods such as ﬂocculation, electrochemistry, ozonation, bleaching, membrane ﬁltration, irradiation, and adsorption to activated carbon are commonly used for the treatment of

DEVELOPMENTS IN MICROBIAL METHODS

187

TABLE I EXAMPLES OF SOME AZO DYES COMMONLY USED IN MICROBIAL DYE DECOLORIZATION STUDIES

188

KUHAD et al.

industrial efﬂuents. Most of these methods are highly speciﬁc and uneconomical but very effective. Table II shows the merits and demerits of various methods for dye decolorization of industrial efﬂuents. A. PHYSICAL Adsorption is one of the most common physical methods used for dye removal that is economically feasible but is inﬂuenced by a number of factors such as surface area of the sorbent, particle size, dye/ sorbent interaction, pH, and temperature (Kumar et al., 1998). Activated carbon is generally very effective for cationic and acid dyes and less effective for dispersed, direct, and reactive dyes (Raghavacharya, 1997). However, the activated carbon adsorption process depends on the type of carbon used, regeneration capacity, and the characteristic of wastewater. A mixture of ﬂy ash and coal can be substituted for activated carbon. Silica gel is another effective adsorbent for removing dyes, but the commercial use is uneconomical because of side reactions such as air binding and fouling with particulate matter. Some naturally occurring material such as peat, wood chips, and agricultural lignocellulosic residues (straws, wood chips, etc.) are potentially economical adsorbents (Nigam et al., 2000; Robinson et al., 2002a,b,c). Dye color removal of up to 90% has been achieved by using steam or chemically pretreated wheat straw, corncobs, and barley husk (Robinson et al., 2002a) in static or continuous packed bed reactor (Robinson et al., 2002b). Widespread availability of agricultural residues offers an economical alternative to activated carbon for dye removal from the industrial efﬂuents. Unlike activated carbon, regeneration is not required when agricultural residues are used, and further potential exists for fermenting dye-adsorbed waste into useful products. Solid state fermentation of the dye-adsorbed residue with white-rot fungi can simultaneously degrade dyes and enrich the nutritional value of the substrate for animal feed or for use of the fermented product used as soil conditioner (Nigam et al., 2000). Membrane ﬁltration can concentrate and separate most of the dyes continuously from the efﬂuent. This approach is typically suitable for low concentrations of dye and water recycling within the plant (Xu and Lebrun, 1999). Although the system is generally resistant to temperature and microbial attack, high capital cost, disposal of concentrated dye, and possibility of clogging are serious disadvantages of membrane technology. Ion exchange is a very effective method for removing both cationic and anionic dyes but less effective for disperse dyes and hence is not widely used (Slokar and Le Marechal, 1997). Usually there is no

TABLE II VARIOUS METHODS FOR DYE DECOLORIZATION OF INDUSTRIAL EFFLUENTS Method Physical

Chemical

Biological

Specific method

Comments

Adsorption

Excellent removal of various dyes; activated carbon is expensive; material loss on regeneration; side reactions with silica gel are undesirable; natural cellulosic material may be cost effective, but speciﬁc surface area is comparatively lower

Membrane ﬁltration

Effective in removing all dye types from wastewater, but concentrated dye sludge needs to be disposed of properly

Ion exchange

Both cation and anion dyes can be removed from the efﬂuents; regeneration is possible without loss of material; may not be applicable to all type of dyes; cost is prohibitive

Irradiation

Efﬁcient oxidation of various dyes at lab scale; high volumes of oxygen needed, which makes the system unattractive

Oxidation

Effective decolorization of various kinds of dyes; problems associated with by-product formation; sludge problem may be associated with Fenton’s reagent treatment; ozonation has short 20-min half-life; release of aromatic amines is a concern with NaOCl

Electrochemical

Relatively new method for effective removal; nonhazardous breakdown of products; electricity costs are high

Coagulation

Excellent removal of direct dyes using ferrous sulfate and ferric chloride; poor removal of acid dyes; high volumes of sludge formation; high disposal costs

Biosorption

Microbial biomass to sorb and remove dyes from wastewater is still in the research stage; may not be practical to treat large volumes of industrial efﬂuents; problems associated with disposal of the dye-adsorbed biomass; may be regenerated using chemicals

Biodegradation

Mixed culture consortium in a combined anaerobic/aerobic continuous system for complete removal of dye compounds shows potential; immobilized cell systems appeared to be more practical than free bacterial cells; practical uses of bacterial processes are not well documented; more understanding physiological/genetic information is required

Enzymatic

Laccase and peroxidase preparations offer rapid method for decolorization of dye wastewater; detailed analysis of reaction by-products, scale-up studies and careful economic evaluation are required for commercial applications

190

KUHAD et al.

loss of adsorbent on regeneration, but the organic solvents used in this technique are expensive. The irradiation process to treat dye-containing efﬂuent in a dual-tube bubbling reactor requires large volumes of dissolved oxygen in the system, which makes the process economically less attractive (Kumar et al., 1998). Sonophotocatalytic technology, a hybrid technology utilizing photocatalytic and sonochemical processing, has recently been proposed for the treatment of reactive dye-containing wastewater (An et al., 2003). However, more data from scale-up studies are required before realization of a commercial process. B. CHEMICAL Because of the simplicity of the process, chemical oxidation is the most commonly used method for dye decolorization. Chemical oxidation involves the removal of dye resulting from aromatic ring cleavage of the dye molecule. Fenton’s reagent (H2O2/FeSO4) is a suitable method for treating wastewater that is resistant to biological treatment or toxic to the microorganisms. This method can be used for the treatment of both soluble and insoluble dyes (Pak and Chang, 1999). However, ﬂocculation of the reagent and residual dye results in sludge generation containing concentrated impurities, which still requires disposal. The performance is also dependent on the ﬁnal ﬂoc formation and its settling quality. While cationic dyes usually do not coagulate well, acid, direct, mordant, and reactive dyes result in poor quality ﬂocs that do not settle well. Sodium hypochlorite (NaOCl) is effective in azo bond cleavage. However, it is not suitable for disperse dyes. Increased chemical use may have a negative impact in waterways because of the presence of chloride ions and the release of aromatic amines, which are both toxic and carcinogenic. A cyclic polymer of glycoluril and formaldehyde, cucurbituril shows good adsorption capacity for various types of textile dyes, but use of this chemical is cost prohibitive (Karcher et al., 1999). Coagulation with ferrous sulfate and ferric chloride may be another feasible method for removing direct dyes from wastewater. However, poor results with acid dyes and the high disposal cost because of the production of large volumes of ﬂocculated sludge were prohibitive in widespread use of this method (Kumar et al., 1998; Slokar and Le Marechal, 1997). Photochemical methods (UV/H2O2) can be used to degrade organic molecules to CO2 and H2O in a batch or continuous system (Yang et al., 1998). The structure of dye, intensity of UV radiation, and pH affect the

DEVELOPMENTS IN MICROBIAL METHODS

191

rate of dye removal. The advantages of this method include the reduction of foul odors and no sludge generation in the process. However, depending on the initial substrate and the extent of treatment, byproducts such as halides, metals, inorganic and organic acids, and aldehydes may be produced. As compared with chlorine (oxidation potential, 1.36) and H2O2 (1.78), ozone (2.08) is a stronger oxidizing agent capable of degrading phenols and chlorinated and aromatic hydrocarbons (Peralto-Zamora et al., 1999). Ozonation leaves dye-containing efﬂuent with no color on toxic byproduct with little or no sludge formation and a low COD. Since the ozone is supplied in its gaseous state, volume of wastewater does not increase. However, because of its short half-life (20 min), the stability of ozone can be strongly affected by the presence of salts, pH, and temperature. Ozonation is a costly process and has been recommended to use in combination with irradiation or membrane ﬁltration. Electrochemical removal of dyes from wastewater is a relatively new process exhibiting efﬁcient color removal and degradation of recalcitrant pollutants (Pelegrini et al., 1999). The method effectively and economically degrades dyes without using chemicals or generating toxic byproducts and sludge build-up. Although a variety of effective physical and chemical treatment methods are commercially available, most of them are either expensive, not adaptable to a wide range of dyes, or do not completely solve the problem of complete decolorization of dye-containing industrial efﬂuents. This has resulted in considerable interest in alternative methods such as microbial, enzymatic, and a combination of physico-chemical and biological methods. III. Microbial Methods Biological methods are currently viewed as effective, speciﬁc, less energy intensive, and environmentally benign, since they result in partial or complete bioconversion of organic pollutants to stable nontoxic end products (Baker and Herson, 1994). Both biosorption and biodegradation have been explored as methods of biological treatments of dye-contaminated efﬂuents. A. BIOSORPTION Biosorption includes both adsorption and absorption. The idea of dye removal through absorption originated from the reasonable success achieved in removal of heavy metal contamination from wastewater.

192

KUHAD et al.

Both bacterial and fungal cells are reported for their capability of partial or complete removal of industrial dyes by using adsorption process. Not all dyes adsorb to a particular type of biomass. However, with some fungi, adsorption is the only decolorization mechanism, but with white-rot fungi both adsorption and degradation can occur simultaneously or sequentially (Knapp et al., 2001). The potential of 10 actinomycetes isolates to decolorize the polymeric dye polyvinyl amine sulfonate anthrapyridone, Poly R-478, has been studied in our laboratory (Vasdev and Kuhad, 1994). Almost 100% decolorization was observed in 8 d. We also studied the capability of three actinomycetes to decolorize azo dyes (Congo red and Remazol brilliant blue R–RBBR), triphenylmethane dyes (crystal violet and malachite green), the heterocylic dye methylene blue, the polymeric dye Poly R, and xylidine (unpublished results). Decolorization caused by sorption was seen in the case of RBBR, Congo red, and Poly R. In the case of triphenylmethane dyes (0.001%), there was no growth and hence no decolorization was observed, whereas in the case of xylidine no decolorization was observed in spite of good growth. Proteus mirabilis isolated from acclimated sludge from a dyeing wastewater treatment plant rapidly decolorized >95% of a deepred azo dye solution (red RBN) within 20 h of incubation, but a part of decolorization (13–17%) was found to be due to biosorption as conﬁrmed by inactivated bacterial cells (Chen et al., 1999). Although decolorization of dye wastewater by live or dead fungal biomass has been a subject of various studies, only limited information is available on interactions between biomass and molecular structure of dyes (Fu and Viraraghavan, 2001). Dead cells remove dyes through the mechanism of biosorption, which involves physicochemical interactions such as adsorption, deposition, and ion exchange. The extent of dye biosorption depends on the chemical structure of dyes and the functional group of the dye molecules. In the case of Aspergillus niger, different functional groups in the fungal biomass play different roles in biosorption of different dyes (Fu and Viraraghavan, 2002). Electrostatic attractions could be the primary mechanism, and the amino, carboxylic acid, phosphate groups, and lipid fractions could be important binding sites depending on the type of the dye used. Integrity of the cell is also important for the binding capacity. Both Freundlich and Langmuir isotherm models ﬁt well for the biosorption of Reactive brilliant red to Rhizopus oryzae biomass (Gallaghar et al., 1997). The kinetics of activated sludge biomass adsorption for the removal of basic dyes from wastewater follows ﬁrstorder processes, controlled by ﬁlm diffusion (Chu and Chen, 2002).

DEVELOPMENTS IN MICROBIAL METHODS

193

The activated sludge biomass process was exothermic in nature, with an activation energy of 3.27 kcal mol1. Yesilada et al. (2002) investigated decolorization of textile dyes by using pellets of the white-rot fungus Funalia trogii. The decolorization activity was signiﬁcantly affected by dye concentration, amount of pellet, temperature, and agitation of the media. With live cells, the decolorization involved adsorption of the dye compound by the fungal pellet at the initial stage, followed by fungal catabolism (Yesilada et al., 2003). On the other hand, decolorization of media containing the azo reactive dyes Procion and Drimarene with greater than 90% removal under aerobic conditions with Aspergillus foetidus has been reported (Sumathi and Manju, 2000, 2001). Adsorption does not appear to be the principal mechanism of decolorization in white-rot fungi (Knapp et al., 1997). It is likely that degradation occurs after initial adsorption. Prior adsorption to fungal mycelium may serve to bring chromophores into closer contact with the degradative enzymes (Wang and Yu, 1998). The method of using fungal biomass to sorb and remove dyes from wastewater is still in the research stage. Biosorption may not be a practical approach for treating large volumes of dye-contaminated industrial efﬂuents because of the problems associated with disposal of the large volumes of biomass after biosorption of dyes from industrial efﬂuents.

B. AEROBIC BIODEGRADATION Decolorization and degradation of dyes by mixed as well as pure cultures of bacteria and fungi have been studied under aerobic and anaerobic conditions. In most studies, the microbial consortia have been found more effective than pure cultures. In addition to chemical structure, several environmental and nutritional factors such as pH, temperature, amount of oxygen, and co-metabolic carbon sources inﬂuence aerobic biodegradation processes. 1. Bacteria Actinomycetes are known to produce extracellular peroxidases that participate in the initial oxidation of lignin to produce various watersoluble polymeric compounds and have also been shown to catalyze hydroxylation, oxidation, and dealkylation reactions against various xenobiotic compounds (Ball et al., 1989; Goszczynski et al., 1994). Species of Streptomyces and Thermomonospora are good examples of actinomycetes capable of effective dye decolorization (Table III). In a

194

KUHAD et al.

screening program designed to use actual textile efﬂuents, 83 positive isolates were found (Zhou and Zimmerman, 1993). Pasti-Grigsby and Crawford (1991) investigated the ability of ligninolytic microbes (white-rot fungi and Streptomyces) to mineralize and decolorize textile dyes and found a strong correlation between dye decolorization ability and ligninolytic ability. Decolorization of mono sulphonated monoazo dye derivatives of azo benzene by the Streptomyces spp. was observed with ﬁve azo dyes having the common structural pattern of a hydroxy group in the para position relative to the azo linkage and at least one methoxy and/or one alkyl group in an ortho position relative to the hydroxy group. While Streptomyces chromofuscus was unable to mineralize aromatics with sulpho groups and both sulpho and azo groups, it mediated the mineralization of modiﬁed dyes containing lignin-like substitution patterns. This work showed

TABLE III BACTERIA AND THEIR MECHANISM OF DYE DECOLORIZATION Bacteria Actinomycetes

Anaerobic

Aerobic

Species

Mechanism of action

Nocardia corallina

Peroxidase

Nocardia globerulla

Peroxidase

Streptomyces badius

Peroxidase

Streptomyces chromofuscus

Peroxidase

Streptomyces viridosporus

Peroxidase

Thermomonospora fusca

Peroxidase

Thermomonospora mesophila

Peroxidase

Clostridium paraputriﬁcum

Anaerobic reduction

Clostridium perfringens

Azoreductase

Proteus vulgaris

Anaerobic reduction

Rhodococcus sp.

Azoreductase

Sphingomonas xenophaga

Anaerobic reduction

Streptococcus faecalis

Anaerobic reduction

Bacillus subtilis

Aerobic biodegradation

Citrobacter sp.

Adsorption and biodegradation

Pseudomonas luteola

Aerobic biodegradation

Pseudomonas mendocina

Aerobic biodegradation

Pseudomonas pseudonallei

Aerobic biodegradation

DEVELOPMENTS IN MICROBIAL METHODS

195

that lignocelluloytic bacteria could be used for the biodegradation of anionic azo dyes (Paszczynski et al., 1992). Ball and Cotton (1996) have also studied three well-characterized lignocellulose-degrading actinomycetes, Streptomyces viridosporus, Streptomyces badius, and Thermomonospora mesophila and showed that they decolorized the polymeric dye Poly-R with a maximum decolorization rate of 0.1 unit/day. The potential of different Nocardia species such as N. corallina and N. globerulla for their ability to degrade triphenylmethane dyes has also been demonstrated (Yatome et al., 1991). Several bacterial strains aerobically decolorize azo dyes by reductive mechanisms. Most of these bacteria do not use azo dyes as carbon or energy source and decolorize azo dyes only in the presence of other carbon sources. Govindaswami et al. (1993) reported a Gram-negative rod capable of oxygen-insensitive azo bond cleavage of dyes (such as Acid orange-7 and Acid red-151) during aerobic growth, in glucoseenriched minimal medium that they considered as a potential candidate for incorporation into experimental bioreactors operated for azo dye degradation. Coughlin et al. (1999) isolated a Sphingomonas strain from a wastewater treatment plant that was capable of aerobically degrading a suite of azo dyes by using them as a sole source of carbon and nitrogen. After an analysis of the structures of dyes, they suggested that there were certain positions and types of substituents on the azo dye that determined the degradation of the dye. Their strain decolorized dye with either 1-amino-2-naphthol or 2-amino-1-naphthol in their structure, and the decolorization appeared to be through reductive cleavage of the azo bond. On the other hand, a Proteus mirabilis strain decolorized RRBN by a combination of biodegradative and biosorptive processes. This organism displayed good growth on the contaminant in shake culture, but color removal was best in anoxic static culture (Chen et al., 1999). Sarnaik and Kanekar (1999) described the aerobic mineralization of the triphenyl methane dye methyl violet by a strain of Pseudomonas mendocina MCMB 402. P. mendocina degraded the dye via a number of unidentiﬁed metabolites to phenol that then entered the -ketoadipic acid pathway. An et al. (2002) recently reported optimum decolorization of several recalcitrant triphenylmethane and azo dyes by Citrobacter sp. at pH 7–9 and temperature 35–40 C. Color removal by Citrobacter sp. was both by adsorption to cells and enzymatic, as evidenced by the experiments with extracellular culture ﬁltrate. P. luteola cells growing under shaking conditions for 24 h were capable of removing 59–99% of the color of seven azo dyes in static

196

KUHAD et al.

conditions (Hu, 2001). There are other reports on the ‘‘aerobic’’ metabolism of azo dyes, where the bacterial strains (e.g., Aeromonas sp., Bacillus subtilis, Proteus mirabilis, P. pseudomallei BNA, P. luteola) were grown aerobically with complex media or sugars, then incubated (often with high cell densities) without shaking in the presence of different azo dyes (Chang and Lin, 2000; Chen et al., 1999; Hayase et al., 2000; Horitsu et al., 1977). However, resting cell cultures presumably become rapidly oxygen depleted, and the reactions observed should therefore be viewed as an anaerobic degradation of azo dyes (Stolz, 2001). To develop novel decolorization processes for practical use, Chang et al. (2001) attempted an immobilized-cell system of P. leuteola with a view to enhance the stability, mechanical strength, and reusability of the biocatalyst. Cell immobilization by entrapment within natural or synthetic matrices is particularly suitable for bacterial dye decolorization since it creates a local anaerobic environment favorable to dye metabolism (Stolz, 2001). P. leuteola cells entrapped in natural and synthetic polymeric matrices efﬁciently decolorized azo dyes enzymatically. Immobilized cells were less sensitive to dissolved oxygen levels and pH as compared with suspended cells, while the effect of temperature was similar for both suspended and immobilized cells. After four repeated experiments, the decolorization rate of the free cells decreased by nearly 45%, while immobilized cells retained 75–85% of their original activity in different matrices. It is generally recognized that azoreductases play an important role in bacterial dye decolorization. However, only a limited number of studies have attempted molecular characterization of dye decolorization. The genes encoding for azoreductase and other possible proteins involved in decolorization have not been clearly identiﬁed. Recently the gene coding for an aerobic azoreductase was cloned from Xenophilus azovorans KF46F (formerly Pseudomonas sp. KF46F), a strain able to grow with carboxylated azo compound 1-(40 -0 arboxyphenylazo)-2-naphthol (carboxy-Orange II) as the sole carbon and energy source (Blu¨mel et al., 2002). The enzyme was heterologously expressed in E. coli. A presumed NAD(P)H-binding site was identiﬁed in the amino-terminal region of the azoreductase. While the cell extracts from the recombinant strain demonstrated the turnover of several industrially relevant azo dyes, the whole cells of the recombinant E. coli were unable to take up sulfonated azo dyes and did not show in vivo azoreductase activity. A recombinant E. coli strain NO3 containing genomic DNA fragments from azo-reducing wild-type P. luteola effectively decolorized an azo dye Reactive red 22 at the rate of about 17 mg/g cells/h (Chang

DEVELOPMENTS IN MICROBIAL METHODS

197

et al., 2000). In another study from the same group (Chang and Lin, 2001), a 6.3 kb fragment from genome DNA of Rhodococcus sp. containing genes responsible for azo dye decolorization were cloned and expressed in E. coli. The recombinant strain E. coli CY1 decolorized Reactive red 22 at the rate of 8.2 mg/g cells/h with performance of excellent stability during repeated batch operations. Although encouraging laboratory results have been obtained indicating the potential of aerobic bacteria for dye removal, practical uses of bacterial processes for color removal have not been well documented. Immobilized cell systems appeared to be more effective than free bacterial cells. 2. Fungi White-rot fungi are the most widely studied microorganisms for dye decolorization/degradation (Table IV). This group of microorganisms is central to the global carbon cycle as a result of their ability to mineralize the complex polymeric woody plant material lignin. In addition to their natural substrate, white-rot fungi have been found to be capable of mineralizing a diverse range of persistent organic pollutants, which distinguishes them from biodegradative bacteria that tend to be rather substrate speciﬁc (Reddy, 1995). The ability of these fungi to degrade such a range of organic compounds results from the relatively nonspeciﬁc nature of their ligninolytic enzymes, such as lignin peroxidase (LiP), manganese peroxidase (MnP), and laccase. LiP catalyzes the oxidation of nonphenolic aromatic compounds such as veratryl alcohol, while MnP oxidizes Mnþþ to Mnþþþ, which is able to oxidize many phenolic compounds (Glenn and Gold, 1986). Laccase is a copper-containing enzyme that catalyzes the oxidation of phenolic substrates by coupling the reduction of oxygen to water (Edens et al., 1999). Earlier it was assumed that laccases and peroxidases can only convert a limited type of azo dyes with preferential conversion of dyes carrying a phenolic substituent in paraposition to the azo bond and additional methyl- or methoxy-substituents in 2- or 2,6-position in relation to the hydroxy group (Chivukula and Renganathan, 1995). However, it has been later demonstrated that certain laccases and peroxidase are able to decolorize certain complex azo dyes such as Reactive black 5 (Schliephake et al., 2000). The decolorization of dyes by white-rot fungi was ﬁrst reported by Glenn and Gold (1983), who developed a method to measure ligninolytic activity of Phanerochaete chrysosporium based on the decolorization of a number of sulphonated polymeric dyes. Subsequently, the decolorization of dyes has also been used to rapidly assess the biodegradative capabilities of diverse white-rot fungi (Chivukula and

198

KUHAD et al. TABLE IV FUNGI AND THEIR MECHANISM OF DYE DECOLORIZATION Fungi

White-rot

Other ﬁlamentous

Yeast

Species

Mechanism of action

Bjerkandera adusta

Mn-peroxidase

Cyathus bulleri

Laccase

Funalia trogii

Adsorption, biodegradation

Lentinula edodus

Laccase

Phanerochaete chrysosporium

Lignin peroxidase

Phlebia radiata

Peroxidase

Pleurotus ostreatus

Peroxidase

Pycnoporus cinnbarinus

Laccase

Trametes versicolor

Biosorption, ligninase

Trametes hispida

Laccase

Aspergillus foetidus

Biosorption, biodegradation

Aspergillus niger

Adsorption, biodegradation

Aspergillus sojae

Biosorption

Botrytis cineria

Adsorption

Halosarpheia ratnagiriensis

Ligninases

Myrothecum verucaria

Adsorption

Neurospora crassa

Biosorption

Rhizopus arrhizus

Biodegradation

Trichoderma sp.

Biosorption, biodegradation

Candida rugosa

Adsorption

Cryptococcus heveanensis

Adsorption

Dekkera bruxellensis

Adsorption

Klyveromyces maxianus

Adsorption

Klyveromyces waltii

Adsorption

Pichia carsonii

Adsorption

Rhodotorula rubra

Biodegradation

Saccharomyces cerevisiae

Biosorption

Renganathan, 1995; Cripps et al., 1990; Field et al., 1993; Goszczynski et al., 1994; Itoh et al., 1998; Wunch et al., 1997). Wunch et al. (1997) suggested a screening method for selecting fungi capable of removing benzo[a]pyrene based on their ability to decolorize the polymeric dye R-478. For most of the 17 ﬁlamentous fungi tested, the disappearance of benzo[a]pyrene was correlated with the ability to decolorize R-478.

DEVELOPMENTS IN MICROBIAL METHODS

199

Paszczynski et al. (1991) reported an approach based on modifying the chemical structure of commercial dyes by linking selected substituents into the dyes’ chemical structure to enhance aerobic azo dye transformation by P. chrysosporium. Guaiacyl or syringyl fragments introduced into the dye molecule seem to provide an access point for the fungal ligninolytic enzymes (Martins et al., 2001). The percentage of biological decolorization of Poly-R-478 was higher than 95%. Nearly complete decolorization (95%) of azo dye Orange-II was achieved by immobilized P. chrysosporium in a continuous packed bed reactor for periods longer than 30 days (Mielgo et al., 2001). Other white-rot fungi such as Cyathus, Trametes versicolour, Bjerkandera adusta, Pleurotus, Phlebia, and Thelephora species have also been screened for their dye decolorizing activity (Conneely et al., 1999; Heinﬂing et al., 1998; Kirby et al., 2000; Pointing et al., 2000; Selvam et al., 2003; Swamy and Ramsay, 1999; Vasdev and Kuhad, 1994). Vasdev et al. (1995) observed effective decolorization of three triphenylmethane dyes by Cyathus bulleri, C. stercorues, and C. straitus. C. Irpex lacteus and Pleurotus ostreatus were selected from 103 wood-rotting fungi for degradation of six different groups of dyes (azo, diazo, anthraquinone based, heterocyclic, triphenylmethane dyes, pthalocyanine) (Novotny et al., 2001). Decolorization of crystal violet and brilliant green by white-rot fungi Coriolus versicolor, Funalia trogii, and P. chrysosporium and one brown-rot fungus, Lacciporus sulphureus, has been reported. Trametes versicolor is capable of efﬁcient decolorization of azo dyes (Toh et al., 2003). Immobilized cultures tend to have a higher level of activity and be more resilient to environmental perturbations than suspension cultures. The decolorization ability of T. versicolor ATCC 20869 was evaluated by using amaranth after immobilization in several natural and synthetic materials such as wheat straw, jute, hemp, maple wood chips, nylon, and polyethylene tetraphthalate ﬁbers (Shin et al., 2002). Jute was found to be the best support material, facilitating good growth of T. versicolor without the loss of jute’s integrity over a 4-week period. Besides white-rot fungi, other ﬁlamentous fungi such as A. niger, Trichoderma viride, and A. foetidus have been found to be efﬁcient in decolorizing textile dyes such as scarlet direct red, fast greenish blue, and brilliant direct violet (Fu and Viraraghavan, 2001; Kousar et al., 2000; Sumathi and Manju, 2000). Decolorization of Poly R-478 and Poly S-119 by Penicillium appeared to involve initial adsorption and followed by biodegradation (Zheng et al., 1999). Geotrichum candidum

200

KUHAD et al.

Dec 1 was found to decolorize 21 kinds of synthetic dyes, and an extracellular enzyme dye-decolorizing peroxidase (DyP) was responsible for the decolorization of the dyes (Kim and Shoda, 1999). To produce large amounts of DyP with dye-decolorizing activity, Sugano et al. (2000) achieved efﬁcient heterologous expression of DyP from Geotrichum candidum Dec 1 in A. oryzae, and by fusing mature cDNA encoding dyp with the A. oryzae -amylase promotor (amyB). A. oryzae is a safe host with higher growth rate and with the capacity to secrete gram-per-liter quantities of heterologous proteins. Studies on dye decolorization with yeast species are scarce. Biodegradation of crystal violet by oxidative yeast, Rhodotorula rubra (Kwasniewska, 1985), and a number of simple azo dyes by Candida zeylanoides have also been reported (Martins et al., 1999). With live cells, fungal nutrition has been shown to be of signiﬁcant importance in effective fungal decolorization systems. Inﬂuence of various carbon and nitrogen sources, micronutrients, vitamins, and amino acids on fungal decolorization has been investigated by various research groups. Majority of decolorization studies used glucose as the carbon source; however, glycerol, xylose, sucrose, maltose, fructose, cellobiose, starch, ethanol, and xylan have also been used (reviewed by Fu and Viraraghavan, 2001; Knapp et al., 2001; Robinson et al., 2001a). Presence of an added carbon source was found to be essential, particularly in the case of fungal mycelia recycling and reuse. Responses of white-rot fungi P. chrysosporium and Coriolopis gallica were found to be different in N-rich and N-limited artiﬁcial textile efﬂuent (Robinson et al., 2001b). Nitrogen supplementation improved enzyme activities and dye decolorization for P. chrysosporium, whereas the additional nitrogen increased enzyme activities for C. gallica but did not improve decolorization. However, in the case of efﬂuent treatment plants, the addition of supplementary carbon or nitrogen sources will depend on the nature of the dye wastewater. Efﬂuents from distilling and paper pulping contain high levels of usable carbohydrates, whereas efﬂuents from dyeing or chemical plants are unlikely to have a sufﬁcient amount of usable carbon sources. Similarly, the presence of usable nitrogen sources in the efﬂuent should be considered for designing medium for industrial efﬂuent treatment. White-rot fungi appear to utilize a wide range of inorganic and organic nitrogen sources. Fungal decolorization offers a promising alternative method to replace or supplement current physicochemical methods. White-rot fungi have the capacity to decolorize a wide range of dyes and colored efﬂuents from the pulp and paper industry, olive milling, cotton

DEVELOPMENTS IN MICROBIAL METHODS

201

bleaching, etc., and may be repeatedly used in continuous and fedbatch cultures over a prolonged period. Fungal mycelia can be repeatedly used or stored for several months at 4 C with 100% retention of activity (Zhang et al., 1999). However, there is a further need to understand the mechanism of decolorization by non-white-rot fungi, establish a relationship between dye structure and decolorization, and develop fungal strains that can grow on simple, inexpensive medium and have a high production rate. 3. Bioreactor Processes Decolorization research has been carried out in batch, fed-batch, or semi-continuous and continuous cultures with a range of reactor conﬁgurations. The preferred bioreactors were air-lift reactor types, but trickling ﬁlters, packed beds, ﬂuidized beds, and stirred tank reactors have also been used in decolorization studies (Bajpai et al., 1993; Zhang et al., 1999). Immobilization on rotating biological contactors (RBC), semi-permeable hollow ﬁber membrane reactors, and rope has also been proven successful (Marwaha et al., 1998; Yin et al., 1990). Most of the reactors were designed to retain a high biomass in the reactor. White-rot fungi, P. chrysosporium, and T. versicolor have been shown to work effectively as immobilized pellet (Pallerla and Chambers, 1996, 1997). The ability of white-rot fungi to treat efﬂuents from pulp and paper, cotton bleaching, olive mills, and distilleries are now established. However, more scale-up studies with white-rot fungi are required before a commercial process can be realized. C. ANAEROBIC BIODEGRADATION Under anaerobic conditions, many bacteria have been reported to readily decolorize azo dyes by bringing about the reductive cleavage of the azo linkage, which results in dye decolorization and the production of colorless aromatic amines (Chung et al., 1992). The initial step in bacterial azo dye metabolism under anaerobic conditions involves the reductive cleavage of the azo linkage by azoreductases, although the in vivo role of such cytoplasmic enzymes is uncertain (Raﬁi et al., 1990). It appears that under anaerobic conditions, speciﬁc azoreductases are probably only of limited importance for the reduction of azo dyes. This is in contrast to the requirement for true azoreductases under aerobic conditions and readily explains the ubiquitous range of microorganisms that reduce azo compounds under anaerobic conditions. In vitro experiments with a recombinant ﬂavin reductase have

202

KUHAD et al.

demonstrated that cytosolic ﬂavin reductases are able to act like azoreductases and may be responsible for unspeciﬁc reactions of azo dyes (Russ et al., 2000). Under strict anaerobic conditions, decolorization of dyes can be enhanced in the presence of redox mediators such as benzyl viologen or quinones (Zee et al., 2000). As compared with resting cells, cell extracts show much higher rates of anaerobic reduction of azo dyes, probably because of the low permeability of the cell membranes for highly polar sulfonated azo compounds. Extracellular reduction of azo dyes by microorganisms may also be due to the action of reduced inorganic compounds such as Fe2þ and H2S, which are formed as anaerobic bacterial metabolic reaction end products. The H2S produced by sulfate-reducing bacteria can reduce azo dye Reactive orange 96 (Libra et al., 1997). Decolorization of dyes with pure culture is impractical, as the isolated culture would be dye speciﬁc, and their application in largescale wastewater treatment plants with a variety of contaminant dyes is not feasible. Efﬁcient biodegradation of dyes can be accomplished when catabolic activities complement each other in a mixed culture community (Nigam et al., 1996). However, Clostridium paraputriﬁcum was found capable of reducing seven commercially available, structurally related azo dyes (Moir et al., 2001). The rates of reduction of these dyes varied between 24 and 74 nmoles/mg protein/h. Beughmann and Weber (1994) demonstrated that, in anoxic sediment environments, nonionic azo dyes readily undergo biologically mediated reduction to the corresponding amines. Generally, during the anaerobic process, 60–70% reduction in COD can be achieved. The presence of a competitive electron acceptor may be a rate-limiting factor. Bras et al. (2001) have described the behavior of methanogenic and mixed bacterial cultures on color removal of a commercial azo dye, Acid orange 7. Low redox conditions maintained by the methanogenic cultures are supposed to be responsible for color removal (Beydilli et al., 1998). However, Chinwetkitvanich et al. (2000) did not ﬁnd any relationship between oxidation-reduction potential and color removal. Manu and Chaudhary (2002) investigated anaerobic decolorization of textile wastewater containing azo dyes, Acid orange 7, and Reactive black 3HN. Color removal of >99% and COD removal of 92–95% was achieved, and salts present in textile wastewater inhibited methanogenesis to a limited extent. Recently, Kapdan and Oztekin (2003) reported over 90% decolorization efﬁciency up to 350 mg/L dyestuff concentration and 200 ml/h feeding rate of textile dyestuff Reactive orange 16 in fed-batch culture

DEVELOPMENTS IN MICROBIAL METHODS

203

when using a facultative anaerobic bacterium consortium called PDW. The addition of 3 g/L yeast extract improved the color removal efﬁciency to around 95%. In anaerobic treatment azo reduction is achieved, but mineralization does not occur. Toxic amines are produced in the environment, and therefore, careful monitoring of the wastewater efﬂuent is required before release into waterways. An advantage of this system could be the simultaneous generation of biogas, which could be recovered to provide heat and power to reduce energy costs. To overcome the problem of the relative recalcitrance of azo dye breakdown products under anaerobic conditions, a sequential or simultaneous two-stage anaerobic/aerobic system could be used. A bioﬁlm-based reactor possibly may be able to completely mineralize contaminant dyes in industrial efﬂuents. D. COMBINED ANAEROBIC/AEROBIC BIODEGRADATION It has been repeatedly suggested that aromatic amines formed during anaerobic cleavage of the azo dyes could be further degraded in an aerobic treatment system. The feasibility of this strategy was ﬁrst demonstrated for the sulfonated azo dye mordant yellow-3 (Glasser et al., 1992). Haug et al. (1991) showed that under anaerobic conditions, mordant yellow was reduced by the biomass of a bacterial consortium grown aerobically with 6-amino napthalene-2-sulfonic acid. After reaeration, these amines were completely mineralized by the culture. This system was believed to be useful for the treatment of azo dyes containing wastewater because under anaerobic conditions it reduces a wide range of azo dyes and aerobically oxidizes many different amino napthalene sulfonic acids. The anaerobic/aerobic treatment can be carried out either sequentially or simultaneously. Sequential processes may combine the anaerobic and the aerobic step, either alternately in the same reaction vessel or in a continuous system in separate vessels. The simultaneous treatment systems utilize anaerobic zones within basically aerobic bulk phases, such as observed in bioﬁlms, granular sludge, or biomass immobilized in various matrices (Field et al., 1995; Jiang and Bishop, 1994; Kudlich et al., 1996; Tan et al., 1999). In the sequential and simultaneous treatment systems, auxiliary substrates are required that supply the bacteria in the anaerobic zones with a source of carbon and energy and a source of reduction equivalents for the cleavage of the azo bond. Chang and Lin (2000) studied a

204

KUHAD et al.

P. luteola strain possessing azoreductase activity to decolorize azo dye, Reactive red-22, with fed-batch processes consisting of an aerobic cell growth stage and an anaerobic fed batch decolorization stage. Tan et al. (1999) studied the degradation of two azo dyes in batch experiments where anaerobic and aerobic conditions were integrated by exposing anaerobic granular sludge to oxygen. The study indicates that aerobic enrichment cultures developed on aromatic amines combined with oxygen tolerant anaerobic granular sludge can potentially be used to completely degrade azo dyes under integrated anaerobic and aerobic conditions. A sequential anaerobic-aerobic treatment process based on mixed cultures of bacteria isolated from textile dye efﬂuent-contaminated soil was used for degradation of sulfonated azo dyes Orange G, Amido black 10B, Direct red 4BS, and Congo red (Rajaguru et al., 2000). In a ﬁxed-bed column using glucose as co-substrate, amines produced by reduction of azo dyes were completely mineralized in a subsequent aerobic treatment. The degradation rates achieved for different dyes ranged from 60.9 mg/d to 571 mg/d. In another anaerobic-aerobic sequencing batch reactor, 60–70% removal of azo-reactive dye by polyphosphate- and glycogen-accumulating organisms was obtained within the ﬁrst 2 h of the anaerobic stage (Panswad et al., 2001). Different reactor designs have been proposed for effective anaerobic/ aerobic treatment systems for azo dyes, including a system of anaerobic and aerobic rotating biological contactors, anaerobic ﬁxed-ﬁlm ﬂuidized bed reactors and aerobic activated sludge reactors, a system of anaerobic and aerobic rotating drum reactors, and anaerobic up-ﬂow ﬁx bed columns and aerobic agitated reactors (O’Neill et al., 2000; Rajaguru et al., 2000; Stolz, 2001). However, it is difﬁcult to compare these systems because of differences in the dyes and conditions, the presence of auxiliary carbon sources, and the difﬁculty of analysis of biological or spontaneous reactions. In continuous anaerobic/aerobic systems, fed with high BOD/COD substrates and low concentrations of dye, a complete decolorization of dyes and signiﬁcant reduction of BOD and COD can be achieved in the anaerobic stage (Stolz, 2001). Encouraging results have also been obtained in laboratory experiments that demonstrated that the anaerobic breakdown of azo dyes results in products that are signiﬁcantly more available for subsequent aerobic processes. This observation has formed a basis for a fullscale anaerobic/aerobic treatment plant for the treatment of more than 1000 m3 of dye-containing wastewater per day from the textile processing industry (Krull et al., 2000).

DEVELOPMENTS IN MICROBIAL METHODS

205

IV. Enzymatic Methods Oxidative enzymes represent an attractive option for wastewater treatment, have been used for efﬂuents from pulp and paper mills, and have potential application in treatment of dye-contaminated wastewater (Duran and Esposito, 2000; Karam and Nicell, 1997). Laccases (EC 1.10.3.2) are copper-containing glycoproteins that require O2 to oxidize phenols and aromatic amines as well as nonphenolic organic substrates by one-electron abstractions resulting in the formation of H2O and reactive radicals that undergo further depolymerization, repolymerization, demethylation, dehalogenation, or quinone formation (Alexandre and Zhulin, 2000; Thurston, 1994). Decolorization and detoxiﬁcation of azo, triphenylmethane, and anthraquinonic dyes by laccases from Pyricularia oryzae, Pycnoporous sanguineus, Trametes hirsuta, and Sclerotium rolfsii have been reported (Abadulla et al., 2000; Muralikrishna and Renganathan, 1995; Pointing and Vrijmoed, 2000; Ryan et al., 2003). The broad substrate speciﬁcity of laccases can be further extended by addition of redox mediators (Claus et al., 2002). Laccases from the lignin-degrading basidiomycetes T. versicolor and Polyporus pinisitus and the ascomycete Myceliophthora thermophila were found to decolorize synthetic dyes to different extents. The addition of the redox mediator 1-hydroxybenzotriazole further improved the decolorization activity of laccase. In the presence of bentonite or by immobilized system, laccase decolorized both individually and in complex mixture. A commercial laccase formulation containing laccase, a redox mediator, and a nonionic surfactant was used for decolorization of Remazol brilliant blue R (RBBR), an important class of recalcitrant anthraquinone-type dye (Soares et al., 2001). Interestingly, laccase alone did not decolorize RBBR and a small molecular weight redox mediator was necessary for decolorization to occur. Puriﬁed laccase has been used to transform novel synthetic disazo dyes (Soares et al., 2002). A laccase isolated and puriﬁed from the culture ﬁltrate of edible mushroom Lentinula edodes, was effective in decolorizing various chemically different dyes such as EBBR, bromophenol blue, methyl red, and naphthol blue black without any redox mediator, whereas Reactive orange 16 and red poly(vinylamine) sulfonate anthrapyridone dyes did require some redox mediators (Nagai et al., 2002). A catalase-peroxidase from the alkalothermophilic Bacillus sp. SF effectively treated the textile bleaching efﬂuent both in free and immobilized form (Fruhwirth et al., 2002). An extracellular peroxidase from Streptomyces chromofuscus A11, capable of oxidizing azo dyes, showed

206

KUHAD et al.

substrate speciﬁcity similar to Mn-peroxidase from P. chrysosporium and horseradish peroxidase (Pasti-Grigsby et al., 1996). Enzymatic approaches to dye decolorization offer a rapid treatment method for dye wastewater. However, scale-up studies and careful economic evaluations are required before this could be applicable at the industrial scale. V. Conclusion Knowledge of physiological and genetic characteristics, biochemical capabilities, and ecology of the relevant microbial species and consortia is an essential prerequisite for successful cleaning up of dyecontaminated water bodies, irrespective of the nature of the dye or the choice of biological treatment system used. Continuous efﬂuent treatment methods using combined anaerobic/aerobic systems for complete removal of dye compounds show some potential. Mixed culture consortia that are capable of surviving in efﬂuents by utilizing the constituents as sources of carbon, energy, and nitrogen would make the process economically feasible. Advances in molecular techniques can help create microbes with improved metabolic capabilities by cloning the gene(s) coding for the decolorizing enzyme(s) into suitable expression systems under strong promoters. Although a robust laccase/ mediator enzyme system is a suitable biocatalyst for rapid treatment of efﬂuents from textile, dye, or printing industries, more studies are required to characterize the nature of reaction products. Moreover, the scale-up of enzyme-based processes needs to be further researched. Moreover, the higher volumes of efﬂuents generated by larger dye plants need to be considered where physical methods of adsorption, extraction, and concentration of dyes and other pollutants may be required before a biological method is feasible (Robinson et al., 2001a). In an integrated system, dye-adsorbed agricultural residues can be further fermented by using white-rot fungi for animal feed or as fertilizer or soil conditioner. ACKNOWLEDGMENTS We thank Manoj Kumar and Jing Ye for their help in preparing the manuscript.

REFERENCES Abadulla, E., Jzanov, T., Costa, S., Robra, K. H., Caracto Paulo, A., and Gubitz, G. M. (2000). Decolourization and detoxiﬁcation of textile dyes with a laccase from Trametes hirsutus. Appl. Environ. Microbiol. 66, 3357–3362.

DEVELOPMENTS IN MICROBIAL METHODS

207

Alexandre, G., and Zhulin, I. B. (2000). Laccases are widespread in bacteria. Trends Biotechnol. 18, 41–42. An, T., Gu, H., Ya, X., Chen, W., Zhu, X., Sheng, G., and Fu, J. (2003). Decolorization and COD removal from reactive dye-containing wastewater using sonophotocatalytic technology. J. Chem. Technol. Biotechnol. 78, 1142–1148. An, S.-Y., Min, S.-K., Cha, I.-H., Choi, Y.-K., Cho, Y.-S., Kim, C.-H., and Lee, Y.-C. (2002). Decolorization of triphenylamine and azo dyes by Citrobacter sp. Biotechnol. Lett. 24, 1037–1040. Bajpai, P., Mehna, A., and Bajpai, P. K. (1993). Decolorization of kraft bleach plant efﬂuent with the white-rot fungus Trametes Versicolor. Proc. Biochem. 28, 377–384. Baker, K. H., and Herson, D. S. (1994). Introduction and overview of bioremediation. In ‘‘Bioremediation’’ (K. H. Baker and D. S. Herson, eds.), pp. 1–7. McGraw Hill, New York. Ball, A. S., and Cotton, J. (1996). Decolourization of two polymeric dye Poly R by Streptomyces viridosporus T7A. J. Basic Microbiol. 36, 13–18. Ball, A. S., Betts, W. B., and McCarthy, A. J. (1989). Degradation of lignin related compounds by actinomycetes. Appl. Environ. Microbiol. 55, 1642–1644. Beughmann, G. L., and Weber, E. J. (1994). Transformation of dyes and related compounds in anoxic sediment: Kinetics and products. Environ. Sci. Technol. 28, 267–276. Beydilli, M. I., Pavlostathis, S. G., and Tincher, W. C. (1998). Decolorization and toxicity screening of selected reactive azo dyes under methanogenic conditions. Water Sci. Technol. 38, 225–232. Blu¨mel, S., Knackmuss, H.-J., and Stolz, A. (2002). Molecular cloning and characterization of the gene coding for the aerobic azoreductase from Xenophilus azovorans KF46F. Appl. Environ. Microbiol. 68, 3948–3955. Bras, R., Isabel, A., Ferra, M., Pinheiro, H. M., and Goncalves, I. C. (2001). Batch Tests for assessing decolourisation of azo dyes by methanogenic and mixed cultures. J. Biotechnol. 89, 155–162. Chang, J.-S., Chou, C., and Chen, S.-Y. (2001). Decolorization of azo dyes with immobilized Pseudomonas luteola. Proc. Biochem. 36, 757–763. Chang, J.S., Kuo, T.-S., Chao, Y.-P., Ho, J.-Y., and Lin, P.-J. (2000). Azo dye decolorization with a mutant Escherichia coli strain. Biotechnol. Lett. 22, 807–812. Chang, J.-S., and Lin, Y.-C. (2000). Fed-batch bioreactor strategies for microbial decolorization of azo dye using a Pseudomonas luteola strain. Biotechnol. Prog. 16, 979–985. Chang, J.-S., and Lin, Y.-C. (2001). Decolorization kinetics of a recombinant Escherichia coli strain harboring azo-dye-decolorizing determinants from Rhodococcus sp. Biotechnol. Lett. 23, 631–636. Chen, K. C., Huang, W. T., Wu, J. Y., and Houng, J. Y. (1999). Microbial decolourization of azo dyes by Proteus mirabilis. J. Ind. Microbiol. Biotechnol. 23, 686–690. Chinwetkitvanich, S., Tuntoolvest, M., and Panswad, T. (2000). Anaerobic decoloration of reactive dye bath efﬂuents by a two-stage UASB system with tapioca as cosubstrate. Water Res. 34, 2223–2232. Chivukula, M., and Renganathan, V. (1995). Phenolic azo dye oxidation by laccase from Pyricularia oryzae. Appl. Environ. Microbiol. 60, 4374–4377. Chu, H. C., and Chen, K. M. (2002). Reuse of activated sludge biomass. I. Removal of basic dyes from wastewater by biomass. Proc. Biochem. 37, 595–600. Chung, K. T., Stevens, S. E., Jr., and Cerniglia, C. R. (1992). The reduction of azo dyes by the intestinal microﬂora. Crit. Rev. Microbiol. 18, 175–190.

208

KUHAD et al.

Claus, H., Faber, G., and Ko¨nig, H. (2002). Redox-mediated decoloration of synthetic dyes by fungal laccases. Appl. Microbiol. Biotechnol. 59, 672–678. Conneely, A., Smyth, W. F., and McMullan, G. (1999). Metabolism of the phhalocyanine textile dye remazol turquoise blue by Phanerochaete chrysosporium. FEMS Microbiol. Lett. 179, 333–337. Coughlin, M. F., Kinkle, B. K., and Bishop, P. L. (1999). Degradation of azo dyes containing sacinonaphthol by Sphingomonas sp. strain 1Cx. J. Ind. Microbiol. Biotechnol. 23, 341–346. Cripps, C., Bumpus, J. A., and Aust, S. D. (1990). Biodegradation of azo and heterocyclic dyes by Phanerochaete chrysosporium. Appl. Environ. Microbiol. 56, 1114–1118. Dawson, D. J. (1981). Polymeric dyes. Aldrichim. Acta 14, 23–29. Dubin, P., and Wright, K. L. (1975). Reduction of azo food dyes in cultures of Proteus vulgaris. Xenobiotica 5, 563–571. Duran, N., and Esposito, E. (2000). Potential applications of oxidative enzymes and phenoloxidase-like compounds in wastewater and soil treatment: A review. Appl. Catal. B. Environ. 28, 83–99. Edens, W. A., Goins, Y. Q., Dooley, D., and Henosn, J. M. (1999). Puriﬁcation and characterization of secreted laccase of Gaeumannomyces graminis var. tritici. Appl. Environ. Microbiol. 45, 1741–1747. Field, J. A., Stams, A. J. M., Kato, M., and Schraa, G. (1995). Enhanced biodegradation of aromatic pollutants in cocultures of anaerobic and aerobic bacterial consortia. Anton. van Leeuwen. 67, 47–77. Fruhwirth, G. O., Paar, A., Gudelj, M., Cavaco-Paulo, A., Robra, K.-H., and Gubitz, G. M. (2002). An immobilised catalase peroxidase from the alkalothemophilic Bacillus SF for the treatment of textile-bleaching efﬂuents. Appl. Microbiol. Biotechnol. 60, 313–319. Fu, Y., and Viraraghavan, T. (2001). Fungal decolorization of dye wastewaters: A review. Biores. Technol. 79, 251–262. Fu, Y., and Viraraghavan, T. (2002). Dye biosorption sites in Aspergillus niger. Biores. Technol. 82, 139–145. Gallaghar, K. A., Healy, M. G., and Allen, S. J. (1997). Biosorption of synthetic dye and metal ions from aqueous efﬂuents using fungal biomass. In ‘‘Global Environmental Biotechnology’’ (D. L. Wise, ed.), pp. 27–50. Elsevier, Amsterdam. Glasser, A., Liebelt, U., and Hempei, D. C. (1992). Design of a two-stage process for total degradation of azo dyes. DECHEMA Biotechnol. Conf. 5B, 1085–1088. Glenn, A. L., and Gold, M. H. (1986). Mn (II) oxidation is the principal function of the extracellular Mn peroxidase from Phanerochaete chrysosporium. Arch. Biochem. Biophys. 251, 688–696. Glenn, J. K., and Gold, M. H. (1983). Decolorization of several polymeric dyes by the lignin degrading basidiomycete Phanerochaete chrysosporium. Appl. Environ. Microbiol. 45, 1741–1747. Goszczynski, S., Paszczynski, A., Pasti-Grigsby, M. B., Crawford, R. L., and Grandford, P. L. (1994). New pathway for degradation of sulphonated azo dyes by microbial peroxidases of Phanerochaete chrysosporium and Streptomyces chromofuscus. J. Bacteriol. 176, 339–347. Govindaswami, M. T., Schmidt, M., White, D. C., and Loper, J. E. (1993). Phylogenetic analysis of a bacterial aerobic degrades of azo dyes. J. Bacteriol. 175, 6062–6066. Haug, W., Schmidt, A., Nortemann, B., Hempel, D. C., Stolz, A., and Knackmuss, H. J. (1991). Mineralization of the sulfonated azo dye Mordant Yellow 3 by a

DEVELOPMENTS IN MICROBIAL METHODS

209

6-aminonaphthalene-2-sulfonate-degrading bacterial consortium. Appl. Environ. Microbiol. 57, 3144–3149. Hayase, N., Kouno, K., and Ushio, K. (2000). Isolation and characterization of Aeromonas sp. B-5 capable of decolorizing various dyes. J. Biosci. Bioeng. 90, 570–573. Heinﬂing, A., Martinez, M. J., Martinez, A. T., Bergabauer, M., and Szewyk, U. (1998). Puriﬁcation and characterization of peroxidases from the dye-decolorizing fungus Bjerkandera adusta. FEMS Microbiol. Lett. 165, 43–50. Horitsu, H., Takada, M., Idaka, E., Tomoyeda, M., and Ogawa, T. (1977). Degradation of p-aminoazobenzene by Bacillus subtilis. Eur. J. Appl. Microbiol. Biotechnol. 4, 217–224. Hu, T. L. (2001). Kinetics of azoreductase and assessment of toxicity of anaerobic products from azo dyes by Pseudomonas luteola. Water Sci. Technol. 43, 261–269. Itoh, K., Kitacle, Y., and Yatome, C. (1998). Oxidative biodegradation of an anthraquinone dye, pigment violet 1Z by Coriolus versicolor. Environ. Contamin. Toxicol. 60, 786–790. Jiang, H., and Bishop, P. L. (1994). Aerobic biodegradation of azo dyes in bioﬁlms. Water Sci. Technol. 29, 525–530. Kapdan, I. K., and Oztekin, R. (2003). Decelorisation of textile dyestuff Reactive Orange 16 in fed-batch reactor under anaerobic condition. Enz. Microb. Technol. 33, 231–235. Karam, J., and Nicell, J. A. (1997). Potential applications of enzymes in waste treatment. J. Chem. Technol. Biotechnol. 69, 141–153. Karcher, S., Kornmuller, A., and Jekel, M. (1999). Removal of reactive dyes by sorption/ complexation with cucurbituril. Water Sci. Technol. 40, 425–433. Kim, S. J., and Shoda, P. (1999). Puriﬁcation and characterization of a novel peroxidase from Geotrichum candidum dec 1 involved in decolorization of dyes. Appl. Environ. Microbiol. 65, 1029–1035. King, B. R., Long, G. M., and Sheldon, H. K. (1997). Practical environmental bioremediation: The ﬁeld guide. CRC, Boca Raton, FL. Kirby, N., Marchant, R., and McMullan, G. (2000). Decolourisation of synthetic textile dyes by Phlebia tremellosa. FEMS Microbiol. Lett. 188, 93–96. Knapp, J. S., Vantoch-Wood, E. J., and Zhang, F. (2001). Use of wood-rotting fungi for the decolorization of dyes in and industrial efﬂuent. In ‘‘Fungi in Bioremediation’’ (G. M. Gadd, ed.), pp. 242–304. Cambridge University Press, Cambridge. Knapp, J. S., Zhang, F., and Tapley, K. N. (1997). Decolourisation of Orange II by a wood rotting fungus. J. Chem. Technol. Biotechnol. 69, 289–296. Kousar, N., Seshikala, D., and Singara Charya, M. A. (2000). Decolourization of textile dyes by fungi. Ind. J. Microbiol. 40, 191–197. Krull, F., Hempel, D. C., Metzen, P., and Alfter, P. (2000). Konzept zur technislen umsetzung einer zweistuﬁgen anoxischen and aerober extilabwasserbehandlung. Chem. Ing. Tech. 72, 1113–1114. Kudlich, F. M., Biship, P., Knackmuss, H.-J., and Stolz, A. (1996). Synchronous anaerobic and aerobic degradation of the sulfonated azo dye mordant yellow 3 by immobilized cells from a naphthalensulfonate degrading mixed culture. Appl. Microbiol. Biotechnol. 46, 597–603. Kumar, M. N. V. R., Sridhar, T. R., Bhavani, K. D., and Dutta, P. K. (1998). Trends in color removal from textile mill efﬂuents. Colorage 40, 25–34. Kwasniewska, K. (1985). Biodegradation of crystal violet (hexamethyl-p-rosaniline chloride) by oxidative red yeasts. Bull. Environ. Contam. Toxicol. 34, 323–330.

210

KUHAD et al.

Libra, J., Yoo, E. S., Borchert, M., and Wiesmann, U. (1997). Mechanism of biological decolorization: Their engineering application. In ‘‘Treatment of Wastewaters from Textile Processing’’ (A. Kornmueller, ed.), pp. 245–266. Schriftenreihe Biologisch Abwasserreinigung 9, TU Berlin. Manu, B., and Chaudhari, S. (2002). Anaeobic decolorization of simulated textile wastewater containing azo dyes. Biores. Technol. 82, 225–231. Martins, M. A., Cardoso, M. H., Quciroz, M. J., Ramalho, M. T., and Campos, A. M. (1999). Biodegradation of azo dyes by the yeast Candida zeylanoides in batch aerated cultures. Chemosphere 38, 2455–2460. Martins, M. A., Ferreira, I. C., Santos, I. M., and Lima, N. (2001). Biodegradation of bioaccessible textile azo dyes by Phanerochaete chrysosporium. J. Biotechnol. 89, 91–98. Marwaha, S. S., Grover, R., Prakash, C., and Kennedy, J. F. (1998). Continuous biobleaching of black liquor from pulp and paper industry using an immobilized cell system. J. Chem. Technol. Biotechnol. 73, 292–296. McMullan, G., Meehan, C., Conneely, A., Kirby, N., Robinson, T., Nigam, P., Banat, I. M., Marchant, R., and Smyth, W. F. (2001). Microbial decolourisation and degradation of textile dyes. Appl. Microbiol Biotechnol. 56, 81–87. Mielgo, I., Moriera, M. T., Feijoo, G., and Lema, J. M. (2001). A packed bed fungal bioreactor for the continuous decolourization of azo dyes (orange II). J. Biotechnol. 89, 99–106. Mishra, G., and Tripathy, M. (1993). A critical review of the treatments for decolourization of textile efﬂuents. Colourage 40, 35–38. Moir, D., Masson, S., and Chu, I. (2001). Structure activity relationship study on the bioreduction of azo dyes by Clostridium paraputriﬁcum. Environ. Toxicol. Chem. 20, 479–484. Moran, C., Hall, M. E., and Howell, R. C. (1997). Effect of sewage treatment on textile efﬂuents. J. Soc. Dyers Colour 113, 272–274. Muralikrishna, C., and Renganathan, V. (1995). Phenolic azo dye oxidation by laccase from Pyricularia oryzae. Appl. Environ. Microbiol. 61, 4374–4377. Nagai, M., Sato, T., Watanabe, H., Saito, K., Kawata, M., and Enei, H. (2002). Puriﬁcation and characterization of an extracellular laccase from the edible mushroom Lentinula edodes, and decolorization of chemically different dyes. Appl. Microbiol. Biotechnol. 60, 327–335. Nigam, P., Armour, G., Banat, I. M., Singh, D., and Marchant, R. (2000). Physical removal of textile dyes from efﬂuents and solid-state fermentation of dye-adsorbed agricultural residues. Biores. Rechnol. 72, 219–226. Nigam, P., McMullan, G., Banat, I. M., and Merchant, R. (1996). Decolourisation of efﬂuent from the textile industry by a microbial consortium. Biotech. Lett. 18, 117–120. Novotny, C., Rawat, B., Bhatt, M., Patel, M., Sasek, V., and Molitors, H. P. (2001). Capacity of Irpex lacteus and Pleurotus ostreatus for decolorization of chemically different dyes. J. Biotechnol. 89, 113–122. O’Neill, C., Hawkes, F. R., Hawkes, D. L., Lourenco, N. D., Pinheiro, H. M., and Delee, W. (1999). Colour in textile efﬂuents—sources measurement, discharge consents and simulations: A review. J. Chem. Technol. Biotechnol. 74, 1009–1018. O’Neill, C., Lopez, A., Esteves, S., Hawkes, F. R., Hawkes, D. I., and Wilcox, S. (2000). Azo-dye degradation in an anaerobic–aerobic treatment system operating on simulated textile efﬂuent. Appl. Microbiol. Biotechnol. 53, 249–254. Pak, D., and Changm, W. (1999). Decolorizing dye wastewater with low temperature catalytic oxidation. Water Sci. Technol. 40, 115–121.

DEVELOPMENTS IN MICROBIAL METHODS

211

Pallerla, S., and Chambers, R. (1996). New urethane prpolymer immobilized fungal bioreactor for decolorization and dechlorination of kraft bleach efﬂuents. Tappi J. 79, 155–161. Pallerla, S., and Chambers, R. (1997). Characterization of a Ca-alginate-immobilized Trametes versicolor bioreactor for decolorization and AOX reduction of paper mill efﬂuents. Biores. Technol. 60, 1–8. Panswad, T., Iamsamer, K., and Anotai, J. (2001). Decoloration of azo-reactive dye by polyphosphate- and glycogen-accumulating organisms in an anaerobic-aerobic sequencing batch reactor. Biores. Technol. 76, 151–159. Pasti-Grigsby, M. B., and Crawford, D. L. (1991). Relationships between the abilities of Streptomyces to decolourize invce cathro type dyes and to degrade lignocellulose. Can. J. Microbiol. 37, 902–907. Pasti-Grigsby, M. B., Burke, N. S., Goszczynski, S., and Crawford, D. L. (1996). Transformation of azo dye isomers by Streptomyces chromofuscus A11. Appl. Environ. Microbiol. 62, 1814–1817. Paszczynski, A. M. B., Goszczynski, S., Crawford, D. L., and Crawford, R. L. (1991). New approach to improve degradation of recalcitrant azo dyes by Streptomyces spp. and Phanerochaete chrysosporium. Enz. Microb. Technol. 13, 378–384. Paszczynski, A., and Crawford, R. L. (1991). Degradation of azo compounds by ligninase from Phanerochaete chrysosporium: Involvement of veratryl alcohol. Biochem. Biophys. Res. Commun. 178, 1056–1063. Paszczynski, A., Pasti-Grigsby, M. B., Goszczynski, S., Crawford, R. L., and Crawford, D. L. (1992). Mineralization of sulfonated azo dyes and sulfanilic acid by Phanerochaete chrysosporium and Streptomyces chromofuscus. Appl. Environ. Microbiol. 58, 3598–3604. Pelegrini, R., Peralto-Zamora, P., de Andrade, A. R., Ryers, J., and Duran, N. (1999). Electrochemically assisted photocatalytic degradation of reactive dyes. Appl. Catal. B. Environ. 22, 83–90. Peralto-Zamora, P., Kunz, A., Gomez de Morales, S., Pelegrini, R., de Capos Moleiro, P., Reyes, J., and Duran, N. (1999). Degradation of reactive dyes. I. A comparative study of ozonation, enzymatic and photochemical processes. Chemosphere 38, 835–852. Pointing, S. B., Bucher, V. V. C., and Vrijmoed, L. L. P. (2000). Dye decolorization by subtropical basidiomycetous fungi and the effect of metals on decolorizing ability. World J. Microbiol. Biotechnol. 16, 199–205. Pointing, S. B., and Vrizmoed, L. L. P. (2000). Decolourization of azo and triphenyl methane dyes by Pycnoporus sanguineus producing laccase as the sole phenoloxidase. World J. Microbiol. Biotechnol. 16, 317–318. Raﬁi, F., Franklin, W., and Cerniglia, C. E. (1990). Azoreductase activity of anaerobic bacteria isolated from human intestinal microﬂora. Appl. Environ. Microbiol. 56, 146–215. Raghavacharya, C. (1997). Colour removal from industrial efﬂuents—a comparative review of available technologies. Chem. Eng. World 32, 53–54. Rajaguru, P., Kalaiselvi, K., Palanivel, M., and Subburam, V. (2000). Biodegradation of azo dyes in a sequential anaerobic-aerobic system. Appl. Microbiol. Biotechnol. 54, 268–273. Reddy, C. A. (1995). The potential of white rot fungi in the treatment of pollutants. Curr. Opin. Biotechnol. 6, 320–328. Robinson, T., McMullan, G., Marchant, R., and Nigam, P. (2001a). Remediation of dyes in textile efﬂuent: A critical review on current treatment technologies with a proposed alternative. Biores. Technol. 77, 247–255.

212

KUHAD et al.

Robinson, T., Chandran, B., and Nigam, P. (2001b). Studies on the decolorization of an artiﬁcial textile-efﬂuent by white-rot fungi in N-rich and N-limited media. Appl. Microbiol. Biotechnol. 57, 810–813. Robinson, T., Chandran, B., and Nigam, P. (2002a). Effect of pretreatment of three waste residues, wheat straw, corn cobs and barley husks on dye adsorption. Biores. Technol. 85, 119–124. Robinson, T., Chandran, B., Sathya Naidu, G., and Nigam, P. (2002b). Studies on the removal of dyes from a synthetic textile efﬂuent using barley husk in static-batch mode and in a continuous ﬂow packed-bed reactor. Biores. Technol. 85, 43–49. Robinson, T., Chandran, B., and Nigam, P. (2002c). Studies on desorption of individual textile dyes and a synthetic dye efﬂuent from dye-adsorbed agricultural residues using solvents. Biores. Technol. 84, 299–301. Russ, R., Rau, J., and Stolz, A. (2000). The function of cytoplasmic ﬂavin reductases in the bacterial reduction of azo dyes. Appl. Environ. Microbiol. 66, 1429–1434. Ryan, S., Schnitzhfer, W., Tznov, T., Cavaco-Paulo, A., and Gubitz, G. M. (2003). An acidstable laccase from Sclerotium rolfsii for wool dye decolourization. Enz. Microb. Technol. 33, 766–774. Sarnaik, S., and Kanekar, P. (1999). Biodegradation of methyl violet by Pseudomonas mendocina MCM B-402. Appl. Microbiol. Biotechnol. 52, 251–254. Schliephake, K., Maninwaring, D. E., Lonegran, G. T., Jones, I. K., and Baker, W. L. (2000). Transformation and degradation of the disazo dye Chicago Sky Blue by a puriﬁed laccase from Pycnoporus cinnbarinus. Enz. Microb. Technol. 27, 100–107. Selvam, K., Swaminathan, K., and Chae, K.-S. (2003). Decolorization of azo dyes and a dye industry efﬂuent by a white rot fungus Thelephora sp. Biores. Technol. 88, 115–119. Shin, M., Nguyen, T., and Ramsay, J. (2002). Evaluation of support materials for surface immobilization and decoloration of amaranth by Trametes versicolor. Appl. Microbiol. Biotechnol. 60, 218–223. Shore, J. (1996). Advances in direct dyes. Ind. J. Fib. Text. Res. 21, 1–29. Slokar, Y. M., and Le Marechal, A. M. (1997). Methods of decoloration of textile wastewater. Dyes Pigments 37, 335–356. Soares, G. M. B., Costa-Ferreira, M., and Pesso´a de Amorim, M. T. (2001). Decolorization of an anthraquinone-type dye using a laccase formulation. Biores. Technol. 79, 171–177. Soares, G. M. B., Hrdina, R., Pessoa de Amorim, M. T., and Costa-Ferreira, M. (2002). Studies on biotransformation of novel disazo dyes by laccase. Proc. Biochem. 37, 581–587. Stolz, A. (2001). Basic and applied aspects in the microbial degradation of azo dyes. Appl. Microbiol. Biotechnol. 56, 69–80. Sugano, Y., Nakano, R., Sasaki, K., and Shoda, M. (2000). Efﬁcient heterologous expression in Aspergillus oryzae of a unique dye-decolorizing peroxidase, DyP, of Geotrichum candidum Dec 1. Appl. Environ. Microbiol. 66, 1754–1758. Sumathi, S., and Manju, B. S. (2000). Uptake of reactive textile dyes by Aspergillus foetidus. Enz. Microb. Technol. 27, 347–355. Sumathi, S., and Manju, B. S. (2001). Fungal mediated decolorization of media containing procion dyes. Water Sci. Technol. 43, 285–290. Swamy, J., and Ramsay, J. A. (1999). The evaluation of white rot fungi in the decolorization of textile dyes. Enz. Microb. Technol. 24, 130–137. Tan, N., Prenafeta-Boldu, F. X., Opsteeg, J. L., Lettinga, G., and Field, J. A. (1999). Biodegradation of azo dyes in cocultures of anaerobic granular sludge with aerobic

DEVELOPMENTS IN MICROBIAL METHODS

213

aromatic amine degrading enrichment cultures. Appl. Microbiol. Biotechnol. 51, 865–871. Thurston, C. F. (1994). The structure and function of fungal laccases. Microbiology 140, 19–26. Toh, Y.-C., Yen, J. L., Obbard, J. P., and Ting, Y.-P. (2003). Decolourisation of azo dyes by white-rot fungi (WRF) isolated in Singapore. Enz. Microb. Technol. 33, 569–575. Vasdev, K., and Kuhad, R. C. (1994). Decolorization of poly R-478 (Polyvinylamine sulfonate Anthra Pyridone) by Cyathus bulleri. Folia Microbiol. 39, 61–64. Vasdev, K., Kuhad, R. C., and Saxena, R. K. (1995). Decolourization of Triphenylmethane dyes by the Bird’s Nest Fungus Cyathus bulleri. Curr. Microbiol. 30, 269–272. Wang, Y., and Yu, J. (1998). Adsorption and degradation of synthetic dyes on the mycelium of Trametes versicolor. Water Sci. Technol. 38, 233–238. Willmott, N., Guthrie, J., and Nelson, G. (1998). The biotechnology approach to colour removal from textile efﬂuent. J. Soc. Dyer. Colour 114, 38–41. Wunch, K. G., Feibelman, T., and Bennett, J. W. (1997). Screening for fungi capable of removing benzo[a]pyrene in culture. Appl. Microbiol. Biotechnol. 47, 620–624. Xu, Y., and Leburn, R. E. (1999). Treatment of textile dye plant efﬂuent by nanoﬁltration membrane. Separ. Sci. Technol. 34, 2501–2519. Yang, Y., Wyatt II, D. T., and Bahorski, M. (1998). Decolorization of dyes using UV/H2O2 photochemical oxidation. Text. Chem. Color 30, 27–35. Yatome, C., Ogawa, T., Hayashi, H., and Ogawa, T. (1991). Microbial reduction of azo dyes by several strains. J. Environ. Sci. Health A26, 471–485. Yesilada, O., Cing, S., and Asma, D. (2002). Decolourization of the textile dye Astrazon Red FBL by Funalia trogii pellets. Biores. Technol. 81, 155–157. Yesilada, O., Asma, D., and Cing, S. (2003). Decolorization of textile dyes by fungal pellets. Proc. Biochem. 38, 933–938. Yin, C.-F., Joyce, T. W., and Chang, H.-M. (1990). Dechlorination of conventional softwood bleaching efﬂuent by sequential biological treatment. In ‘‘Biotechnology in Pulp and Paper Manufacture’’ (T. K. Kirk and H.-M. Chang, eds.), pp. 231–244. Butterworth-Heinemann, Boston, MA. Zee, V. F. P., Lettinga, G., and Field, J. A. (2000). The role of (auto)catalysis in the mechanism of an anaerobic azo reduction. Water Sci. Technol. 42, 301–308. Zhang, F., Knapp, J. S., and Taptey, K. N. (1999). Development of bioreactor systems for decolorization of Orange II using white rot fungus. Enz. Microb Technol. 24, 48–53. Zheng, Z., Levin, R. E., Pinkham, J. L., and Shetty, K. (1999). Decolorization of polymeric dyes by a novel Penicillium isolate. Proc. Biochem. 34, 31–37. Zhou, W., and Zimmerman, W. (1993). Decolourization of industrial efﬂuents containing reactive dyes by actinomycetes. FEMS Microbiol Lett. 107, 157–162. Zollinger, H. (1987). Colour chemistry—synthesis, properties of organic dyes and pigments. VCH, New York.

This Page Intentionally Left Blank

Extracellular Glycosyl Hydrolases from Clostridia WOLFGANG H. SCHWARZ,* VLADIMIR V. ZVERLOV,{ {,} AND HUBERT BAHL *Technical University of Munich Institute of Microbiology, D-85350 Freising, Germany {

Russian Academy of Science, Institute of Molecular Genetics 123182 Moscow, Russia {

University of Rostock, Institute of Biological Sciences Department of Microbiology, D-18051 Rostock, Germany }

Author for correspondence. E-mail: [email protected]

I. Introduction II. Modular Structure of the Enzymes III. Function of Noncatalytic Modules A. Substrate Binding B. SLH Module C. Fibronectin Type III Module IV. Characterization of Enzyme Systems A. Starch Degradation B. Cellulose Degradation C. Xylan and Hemicellulose Degradation V. Concluding Remarks References

215 217 218 219 223 224 225 228 229 247 251 252

I. Introduction Anaerobic bacteria are key players in the fate of rotting biomass. They play a major role in the digestion of biomass by herbivores and insects (such as termites), possibly even as endosymbionts of ﬂagellates common in the intestinal tract of plant-feeding animals, such as the rumen of cattle. The hosts help by mechanical degradation (chewing) and by providing a favorable environment. A part of the natural rotting process of biomass in soil and compost heaps is also performed by the anaerobic bacteria when the easily degradable constituents (e.g., soluble sugars and proteins) of the biomass are already used up. Among the anaerobic bacteria are specialists for the degradation of the insoluble components of biomass that are most difﬁcult to degrade: crystalline starch, hemicellulose, and cellulose. In nature, polysaccharide-degrading bacteria thrive in symbiotic relationships with secondary microorganisms (Ljungdahl and Eriksson, 215 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

216

SCHWARZ et al.

1985). The enzymes secreted by the primary cellulose degraders break the substrate down into cellodextrins, cellobiose, and glucose, only a part of which is assimilated by the polymer-degrading strains themselves. The rest is utilized by the secondary microbial ﬂora, as are the fermentation products of the anaerobic cellulose degraders: hydrogen, carbon dioxide, alcohols, and short-chain fatty acids. Thus polysaccharide degradation is just the ﬁrst step in a food chain within a complex ecosystem. Approximately 1.8 1012 metric tons of biomass (dry weight) exist on the continents, which are continuously recycled by enzymatic processes. Polysaccharides from plant material form a major part of the biomass: They are the most important factors in the carbon cycle in nature that regulates the CO2 content of the atmosphere. An estimated 40 GT per year alone of cellulose are produced by land plants—about the same amount is degraded. The natural rotting process is catalyzed by hydrolytic enzymes produced from ubiquitous microorganisms. The energy contained in the resulting sugars drives the build-up of micro- and macrobiotic biomass. But the energy gradient from polysaccharide to CO2 can also be exploited for industrial purposes without increasing the CO2 content in the atmosphere: biomass, through burning or enzymatic hydrolysis, is a CO2-neutral source of environmentally friendly energy for the future. Anaerobic bacteria, among them primarily the clostridia, are an excellent source for hydrolytic enzymes able to hydrolyze polysaccharides in biomass to fermentable sugars. An example of special interest is the utilization of the hydrolytic extracellular enzymes of the solventogenic bacterium Clostridium acetobutylicum for the fermentation of starch to the organic solvents butanol and acetone (Du¨rre, 1998; Gapes, 2000). Although, because of economic reasons, the industrial process at present is not utilized in the Western world, it is still a very attractive alternative to the mineral oil–based production of energy and bulk chemicals, since it runs with renewable substrates, enabling sustainable energy production. Consequently, research on the bacterial solvent production process is going on; for example, a number of new strains degrading a wide range of polysaccharides have been isolated (Du¨rre, 1998; Montoya et al., 2001). Meanwhile, determination of the genomic sequence of C. acetobutylicum made a thorough analysis of its genes possible, and a complete cluster of genes for the expression of a cellulosome was detected (No¨lling et al., 2001). Unfortunately, only few of these genes are expressed, and hydrolysis of cellulosic substrates could not be achieved (Sabathe et al., 2002). Nevertheless, this opens the possibility that related strains may exist that express the whole operon and would then be able to produce solvents directly from cellulose. In

EXTRACELLULAR CLOSTRIDIUM ENZYMES

217

addition to genetic engineering of producer organisms and new fermentation and product separation technologies, this will help to make the bacterial solvent production economically feasible in the near future. So far none of the strains used for industrial production of acetone and butanol have been able to degrade cellulose as a cheap and available substrate to fermentable sugars. However, only a few of the industrial strains have survived the shutdown of the production facilities. The rest of the valuable strains are permanently lost and cannot be tested. The search for new solventogenic strains capable of efﬁcient lignocellulose hydrolysis is therefore going on, and research on the clostridial extracellular enzymes is an increasingly urgent necessity. This chapter will focus on the most prevalent polysaccharides present in biomass: starch, cellulose, and hemicellulose—the latter two of which are especially difﬁcult substrates to degrade. The unique strategies of the clostridia to cope with these substrate problems are discussed. II. Modular Structure of the Enzymes In contrast to the enzymes isolated from eukaryotic organisms (mostly fungi) and aerobic bacteria, many extracellular enzymes of the anaerobic bacteria have a modular structure—that is, they consist not only of a catalytic module but of a complex arrangement of different modules: one or even more than one catalytic module(s) and in addition, noncatalytic modules. In Fig. 1, a schematic modular structure of a hypothetical clostridial glycosyl hydrolase is depicted. The catalytic module can be accompanied by one, several, or all of the following modules: carbohydrate binding (CBM), immunoglobulin (Ig)-like, dockerin (Doc), ﬁbronectin type III (Fn3), and S-layer homology (SLH). These modules constitute independent folding units that often are covalently connected by ﬂexible linkers such as the so-called PTS boxes (irregular stretches of hydroxy amino acids). As a consequence of the presence of several modules, these enzymes are often quite large, consisting of more than 1,000 amino acids with a molecular mass above

FIG. 1. Schematic representation of the modular structure of a hypothetical clostridial extracellular enzyme. CBM: carbohydrate-binding module; Doc: dockerin module; Fn3: ﬁbronectin type III module; GH: catalytic module of glycosyl hydrolase family; Ig: immunoglobulin-like module; SLH: surface-layer homology module. Numbers indicate the relative position of the modules.

218

SCHWARZ et al.

100 kDa. The order of modules in a given enzyme does not follow strict rules. Noncatalytic modules may appear on the N- or the C-terminal end of the catalytic units. SLH or Doc modules are in most cases located near the C-terminus of the enzymes. In general, the noncatalytic modules may support or even modulate the catalytic activity. Some are stuffer proteins between a catalytic unit and a functionally important noncatalytic module; some are closely connected with a catalytic module and stabilize it against thermal denaturation. Binding modules are known for the substrate (CBM) or for the host cell (SLH). The catalytic modules of these enzymes hydrolyze the glycosidic bond between two or more carbohydrates or between a carbohydrate and a noncarbohydrate moiety. Based on amino acid sequence similarities, a classiﬁcation of glycoside hydrolases into families has been proposed (Henrissat, 1991): The updated list (October 2003) contains 91 families, GH1 to GH91 (Coutinho and Henrissat, 1999a). The catalytic mechanism is a general acid catalysis that requires two critical amino acid residues: a proton donor and nucleophile/base (Davies and Henrissat, 1995). The hydrolysis results in either retention or inversion of the conﬁguration at the anomeric C-atom. The roles of the Doc, Fn3, and SLH modules are summarized in the following section, whereas the function of the Ig module has not yet been successfully addressed. The most complex enzymes are those of the extremely thermophilic cellulose degraders Cellulosiruptor cellulolyticus and the closely related Anaerocellum thermophilum, which contain a so-called multifunctional enzyme system; these are not included in this review (Bayer et al., 2000). They also belong to the order Clostridiales. Many of their cellulases and hemicellulases are composed of more than one catalytic module, connected with binding modules and stufﬁng peptides. Functionally related and mutually synergistic catalytic components are combined in one polypeptide chain to enhance the effectiveness of enzymatic action. This seems to be an independent evolutionary way towards an enzyme complex that combines all necessary functions in a close spatial arrangement but with more ﬂexibility in structure and composition. The most advanced of these complexes is the cellulosome, which is described now. III. Function of Noncatalytic Modules The functionally most important and best-characterized noncatalytic module in the extracellular enzymes of the clostridia is the CBM. In recent years the SLH module was in the focus of functional analysis,

EXTRACELLULAR CLOSTRIDIUM ENZYMES

219

whereas only limited knowledge exists on the function of the Fn3 module. Therefore these three modules will be described in this part of the chapter. The function of the Doc module can be found within the description of the cellulosome.

A. SUBSTRATE BINDING The interaction of enzymes with polymeric substrates is severely slowed by the limited diffusion of the enzyme as well as the substrate. This difﬁculty is greatly overcome by the introduction of binding modules. These are protein modules of about 35 to less than 180 amino acid residues that target the enzyme in a noncatalytic way to suitable areas of the large substrate, a single polysaccharide molecule thread as in soluble or in amorphic parts of insoluble substrates, or a bundle of insoluble substrate molecules as in crystalline cellulose (Linder and Teeri, 1997). This increases the enzyme concentration on the substrate surface and improves substrate interaction (Bolam et al., 1998). The carbohydrate binding modules (CBM) are categorized into families according to sequence homology and the consequent three-dimensional fold (Coutinho and Henrissat, 1999b). A list of the presently known CBMs with links to nucleotide and amino acid sequences and a short compilation of general information on each family is given at the CAZY server (Coutinho and Henrissat, 1999a). Some CBMs have a ﬂat strip of aromatic amino acid residues for binding to the surface of an array of parallel substrate molecules as in crystals; others bind single substrate molecules in a pocket-like structure (reviewed in Bayer et al., 1998). The anchoring is mediated by polar residues such as asparagine or glutamine (Tormo et al., 1996). Some families can be separated into slightly different subfamilies that have the same global fold but differ in their binding abilities. An example is family CBM3, where subfamily CBM3a modules bind tightly to crystalline cellulose, whereas CBM3b modules seem to be more variable, and CBM3c modules modulate the enzymatic activity by feeding a single substrate molecule with a predeﬁned directionality through the active site pocket of the catalytic module, leading to processive cleavage (Sakon et al., 1997). An enzyme with endoglucanase activity is consequently transformed into an exo-glucanase (or a processive endo-glucanase) by the activity of a binding module. This emphasizes the important role of the noncatalytic modules for the enzyme activity, especially for the hydrolysis at different sites on crystalline or amorphic cellulose (Carrard et al., 2000).

220

SCHWARZ et al.

It has been proposed that some CBMs may be degenerated and function as thermostabilizing modules, such as the CBM3c in C. stercorarium cellulase Cel9Z or C. thermocellum cellulase Ce19I (see below). Although it is not clear if such CBMs have lost their binding capacity, they are functionally attached to the catalytic module. This seems to stabilize the structure of the catalytic core and in some cases increases the thermostability up to 30 C (Riedel et al., 1998a). The loss of activity at high temperature on deletion of the CBM may be so drastic that the function of the module was in some cases interpreted as essential for activity through binding. Similar results have been reported for the CBM22 modules that, for example, exert a thermostabilizing effect on xylanase XynA from Thermotoga maritima but at the same time bind to xylan and -1,3-1,4-glucan (Meissner et al., 2000). The authors argue that thermostabilization is a side effect of the close association of the enzyme with its substrate binding module. Despite a very tight binding to the substrate through a CBM, the enzymes seem to diffuse laterally along the substrate molecule (Jervis et al., 1997). This was shown with fungal enzymes in impressive pictures of the crystal surface through atomic force microscopy (Lee et al., 2000). Some dynamic experiments have been performed with the family CBM3a, which is important for the effectiveness of the cellulosome of C. thermocellum. In in vitro experiments, this CBM binds the scaffolding irreversibly to crystalline cellulose and allows the cellulolytic cellulosome components to be effective without a CBM and probably with a greater freedom of movement for activity around the binding site. The binding ability can be investigated primarily by two methods (Tomme et al., 1996). In equilibrium assays, the binding protein is mixed with an insoluble substrate in a suitable buffer; after equilibration, the decrease of protein in the cleared supernatant is estimated. Alternatively, a soluble binding substrate is mixed into the gel matrix of a native polyacrylamide gel; the binding protein is subjected to electrophoresis with these gels, and retardation in comparison with other (nonretarded control) proteins is determined. However, the in vivo function of the CBM can be detected only by constructing deletion mutants of enzymes and thorough characterization of the mode of the enzymatic activity, especially the processivity in the case of the cellulases. The binding speciﬁcity is not necessarily identical with the substrate speciﬁcity of the associated catalytic module. Especially with the xylanases it is common to ﬁnd cellulose binding modules. The xylanases present in the cellulosome do not have their own CBMs but

EXTRACELLULAR CLOSTRIDIUM ENZYMES

221

rely on the cellulose binding capacity of the scaffoldin-linked CBM. This may have to do with the function of these enzymes that serve as supporters in biomass degradation, where xylan is associated with cellulose (see below). In contrast, a CBM6 module was ascribed a role in hemicellulose degradation by binding to insoluble xylan and several soluble polysaccharides (Sun et al., 1998). CBMs are not randomly connected with catalytic modules. A limited number of them play a role in the extracellular enzymes in clostridia, and patterns of module structures emerge that seem to be evolutionarily successful. Examples are the ‘‘themes’’ A to D of the GH9 cellulases depicted by Bayer et al. (2000), where the GH9 cellulases form a repeatedly observed pattern with noncatalytic modules: Theme Theme Theme Theme

A: B: C: D:

GH9 GH9–CBM3c Ig–GH9 CBM4–Ig–GH9

Others are given in the next list, where the preferential binding activity of the CBM families is also indicated. In connection with cellulases: CBM3a (with scaffoldin)—crystalline cellulose CBM3b (often as GH9-CBM3c-CBM3b)—crystalline cellulose CBM3c (often as GH9-CBM3c)—as thermostabilizing module and possibly for substrate feeding (binding amorphous cellulose) CBM4 (often as CBM4-Ig-GH9)—amorphous cellulose or soluble oligosaccharides CBM6—amorphous cellulose, -1,3-glucan, or xylan CBM9 (e.g., GH5-CBM9)—cellulose In connection with xylanases: CBM22—thermostabilizing module, or xylan or -1,3-1,4-glucan CBM4 (also in -glucanase Ct-Lic16A)—amorphous cellulose or soluble oligosaccharides With esterases/glycosidases: CBM6 (with esterase CE1, PL6, PL1, PL11, GH30, GH43, GH39)— amorphous cellulose, -1,3-glucan or xylan In connection with amylases: CBM 20—binding to starch CBM 26—binding to cyclodextrins

222

SCHWARZ et al.

In connection with other enzymes: CBM2 (PL, GH74)—crystalline cellulose, chitin or xylan CBM13 (GH43)—as threefold repeat (xylan binding?) The best-characterized carbohydrate binding module in starch degrading enzymes belongs to the CBM20 family. This module is also known as starch-binding domain: It consists of approximately 100 residues and has granular starch-binding activity. CBM26 has been found in enzymes with different enzymatic activities (e.g., -amylases, amylases, glucoamylases, and especially cyclodextrin glucanotransferases). The presence in the latter enzyme is consistent with the fact that CBM26 strongly interacts with cyclodextrins. In clostridial enzymes this module has been found so far only in a few cases (Table I). Other CBMs with afﬁnity to starch have been identiﬁed in not further characterized gene products of C. acetobutylicum (Table I).

TABLE I OCCURRENCE OF CARBOHYDRATE BINDING MODULES IN STARCH DEGRADING ENZYMES FROM CLOSTRIDIAa Module CBM20

Species

Enzyme

Reference

Thermoanaerobacterium thermosulfurigenes

-amylase

Kitamoto et al., 1988

thermosulfurigenes EM1

amylase-pullulanase

Matuschek et al., 1994

cyclodextrin glucanotransferase

Wind et al., 1995

ethanolicus 39E

amylase-pullulanase

Mathupala et al., 1990

saccharolyticum

amylase-pullulanase

Ramesh et al., 1994

thermohydrosulfuricus

amylase-pullulanase

Melasniemi et al., 1990

CBM21

Clostridium acetobutylicum

CAP 0129

No¨lling et al., 2001

CBM25

Clostridium acetobutylicum

-amylase

No¨lling et al., 2001

CBM26

Clostridium acetobutylicum

CAC 2252

No¨lling et al., 2001

Thermoanaerobacter

CAC 2891

a Several thermophilic Clostridium species have been reclassified as members of the genera Thermoanaerobacterium and Thermoanaerobacter (Lee et al., 1993). Data taken from Coutinho and Henrissat (1999b).

EXTRACELLULAR CLOSTRIDIUM ENZYMES

223

B. SLH MODULE The function of SLH modules (for surface-layer homology) in the extracellular enzymes of clostridia is not immediately obvious. SLH modules are found in proteins from phylogenetically unrelated bacteria (e.g., Gram-positive and Gram-negative bacteria) and are present in three types of proteins: surface-layer (S-layer) proteins, extracellular enzymes/proteins, and outer membrane proteins (Engelhardt and Peters, 1998). In most cases the SLH module is present in three copies of about 50–60 residues each. A single module is predicted to have the following secondary structure pattern: -helix (HI)—-sheet (S)—loop (LI)—-helix (HII)—loop (LII) (Fig. 2). The overall similarity of SLH modules in proteins from different organisms is low, but they contain at least two highly conserved motifs, a FxDV motif at the N-terminus and an TRAE motif at the beginning of the second -helix. Our data indicate that at least the TRAE motif contributes to the function of SLH modules (unpublished results). In S-layer and outer membrane proteins these modules are generally located at the N-terminus and in enzymes at the C-terminus. Their role as cell-wall targeting modules was initially proposed mostly on the basis of sequence comparison (Fujino et al., 1993; Lupas et al., 1994; Matuschek et al., 1994). Now, because of several in vitro and in vivo studies, there is strong evidence that the SLH modules indeed serve as an anchor to the cell wall for the different protein types (Brechtel et al., 1999; Lemaire et al., 1995; Mesnage et al., 1999; Olabarria et al., 1996; Ries et al., 1997). Although it was initially thought that SLH modules bind to peptidoglycan, it is now clear that the adhesion component in the cell wall is not the peptidoglycan itself but a polymer covalently linked to it (Brechtel and Bahl, 1999; Ilk et al., 1999; Mesnage et al., 1999; Ries et al., 1997; Sa´ra et al., 1996). Complete structural analysis has indicated that these cell-wall associated polymers are teichuronic acids (Ilk et al., 1999). Furthermore, it has been found that they are pyruvylated and that a strong correlation between the existence of SLH modules and genes involved in the addition of pyruvate to the wallassociated polymer exists in different bacteria (Mesnage et al., 2000).

FIG. 2. Consensus sequence and highly conserved motifs of SLH modules. H: -helix; S: -sheet; L: loop.

224

SCHWARZ et al.

FIG. 3. Model of the attachment of extracellular proteins to the cell surface of clostridia. The interaction of SLH modules ( ) in S-layer proteins, enzymes, and functional proteins with polymers ( ) associated with the peptidoglycan (PG) is illustrated. Enzymes can be attached to the cell wall via SLH modules either directly, mediated by a linker protein, or as part of a multienzyme complex. CM: cytoplasmic membrane.

Thus the SLH-mediated anchoring mechanism is one of several, but highly conserved strategy bacteria have developed to display proteins on their surface. In clostridia, SLH modules have been found in S-layer proteins and in several hydrolases (e.g., cellulases, xylanases, amylase-pullulanases) (Fuchs et al., 2003; Matuschek et al., 1996). Figure 3 illustrates how SLH modules mediate the attachment of S-layer proteins, single enzymes, or multienzyme complexes to the cell wall. C. FIBRONECTIN TYPE III MODULE The Fn3 module is one of three types of internal repeats found in the plasma protein ﬁbronectin. Many animal proteins contain the Fn3 module, including extracellular, intracellular, and membranespanning proteins, and adhesion molecules. Surprisingly, Fn3-like modules are also found in bacterial glycosyl hydrolases (Little et al., 1994), including cellulases, pullulanases, and polygalacturonases from clostridia (Matuschek et al., 1996; Zverlov et al., 1998b). In

EXTRACELLULAR CLOSTRIDIUM ENZYMES

225

Thermoanaerobacterium thermosulfurigenes EM1 Fn3 modules are present in the amylase-pullulanase (AmyB) and the polygalacturonate hydrolase (PglA) (Matuschek et al., 1996). Interestingly, the Fn3 modules of PglA and the exopolygalacturonate hydrolase of Erwinia chrysanthemi (He and Collmer, 1990) share a higher degree of similarity (38% identical residues) than the module of PglA and the two modules of AmyB (27% and 29% identical residues, respectively). On the other hand, the Fn3 modules of AmyB are 64% identical to the corresponding modules in the amylase-pullulanase of Thermoanaerobacter thermohydrosulfuricus E101-69 (Melasniemi et al., 1990). Therefore the Fn3 modules appear to be clustered by protein type and not by organism. Very little information is available on the function of Fn3 modules in extracellular enzymes of bacteria. It has been postulated that they serve as spacers or linkers allowing optimal interaction between the catalytic and substrate-binding modules (Little et al., 1994). In agreement with this suggestion, Watanabe et al. (1994) reported that deletion of the Fn3 module(s) located between the catalytic and substrate-binding modules of a chitinase from Bacillus circulans did not affect binding to chitin but decreased hydrolytic activity of the enzyme to colloidal chitin. Recently the ﬁrst evidence of a function for Fn3 in a clostridial enzyme during hydrolysis of a polysaccharide was presented. It was shown that the two Fn3 modules of the multi-modular cellobiohydrolase CbhA of Clostridium thermocellum are able to change the surface of cellulose that had been loosened up and crenellated. That promoted hydrolysis by the catalytic domain (Kataeva et al., 2002). IV. Characterization of Enzyme Systems Polysaccharides are difﬁcult substrates for enzymes. They are usually larger than the enzyme itself, and quite often they are not soluble (i.e., they are not hydrated or occur in tight aggregates or even in crystalline form). Moreover, many natural polysaccharides such as hemicelluloses are extremely heterogeneous and contain many different sugar moieties with different linkage types, or they are derivatized. Others may contain only one type of sugar moiety, which, however, as in starch, are linked in different ways (1->4 or 1->6). Others are chemically homogeneous (such as cellulose: -1,4-glucosidic linkages only) but are at least partially crystalline and topologically diverse. Long substrate molecules can be shortened by a statistical hydrolytic cut in one of the many linkages between the building blocks of the substrates. This is the so-called endo-mode of action. The sites for such

226

SCHWARZ et al.

enzymatic attacks may be extremely limited because of occlusion by other molecules, by the viscosity of the polymer solution, or by the tight assembly of many molecules (e.g., by crystal formation). Only freely accessible, hydrated parts of a long molecule can be recognized by enzymes, and it could be postulated that even within the same polymer strand different enzyme types are needed to hydrolyze different topologies. The substrate size poses the problem of diffusion: It is not the substrate that tumbles around until it ﬁnds an enzyme pocket, but the enzyme must ﬁnd its substrate. The diffusion of such large molecules is slow, and hence it is an advantage to stick to the substrate and degrade it successively once a large substrate molecule is found (see the previous section on binding modules). The sequential, processive action is executed by the exo-mode enzymes, which recognize either a reducing or a nonreducing end of the substrate molecule and feed it through the activity pocket, chopping off a monomer (e.g., -glucosidase), a dimer (e.g., -amylase, cellobiohydrolase), or a multimer (several exoglucanases). A synergism between endo-glucanases that produce the open ends, and exo-glycanases and processive enzymes that widen the gap, has been observed (i.e., the sum of the single activities is smaller, as if both types of enzymes act in combination simultaneously). Another synergism exists between the exo-glycanases active from the reducing and the nonreducing end: they can be thought to act from one open cut in a long molecule into both directions, opening a hole (e.g., in the surface of a cellulose crystal). Polysaccharides are degraded by extracellular enzymes (sometimes also called ‘‘depolymerases’’). The resulting monosaccharides or oligosaccharides are either taken up by the cell or degraded extracellularly by secreted glycosidases. Examples are discussed below. Commonly, the potential of a bacterial strain to produce extracellular enzymes is evaluated by assaying the cell free culture supernatant for enzymatic activities. This was done, for example, in an effort to compare the thermophilic polysaccharide hydrolyzing bacteria C. stercorarium and C. thermocellum for enzymes in the culture supernatant (Table II). The cellulolytic activities of these two species were comparable, but C. stercorarium culture supernatant had a higher soluble activity for mixed-linkage glucan and xylan, and especially for glucoside, xyloside, and arabinoside. This can be taken as an indication for a higher activity on hemicellulose in C. stercorarium. The higher cellobiosidase activity of C. thermocellum corroborates its specialization for cellulose as substrate, since the major components of cellulases are cellobiohydrolases that are active on the cellobioside.

227

EXTRACELLULAR CLOSTRIDIUM ENZYMES TABLE II COMPARISON BETWEEN TWO THERMOPHILIC, SACCHAROLYTIC CLOSTRIDIA

Substrate

C. thermocellum (mU/ml)a

Microcrist. cellulose

2

Phosphoric acid swollen cellulose

C. stercorarium (mU/ml)a 2

13

10

140

120

1,3-1,4--glucan (lichenan)

6.500

12.000

Arabino-xylan

3.000

20.000

Carboxymethyl cellulose (CMC)

pNP--glucopyranoside pNP--cellobioside

1,3 12

7 1,7

pNP--xylopyranoside

0,3

2

pNP--arabinofuranoside

1,3

21

a

Activity in cell free culture fluid (grown on cellobiose).

However, the results of such assays have to be interpreted with great care: 1. A great portion of the exo-enzymes in C. thermocellum are located on the cell surface, such as the cellulosome complex and some single enzymes such as the -1,3-glucanase Lic16A (Fuchs et al., 2003); the same is true for one xylanase of C. stercorarium (see below); their activity escapes the assay. 2. Many enzymes are not speciﬁc (e.g., -glucosidases are also active on cellobiosides and xylosides). 3. Xylanases of GH10 have high activity on mixed-linkage glucans. Thus, information from the activity of culture supernatants on model substrates alone is not sufﬁcient to estimate the hydrolytic potential of a given bacterium. Nevertheless, C. stercorarium was identiﬁed as a thermophilic bacterium with a more general activity on polysaccharides present in biomass, whereas C. thermocellum is known as a specialist for the degradation of cellulose, despite the presence of a number of other enzymatic activities for -1,3-glucan, xylan, mannan, pectin, chitin, and probably other polysaccharides. Valuable information on a complete catabolic pathway of a bacterium comes from fermentation experiments; if a given polymeric substrate can be depolymerized into oligosaccharides and further to monomeric sugars, transported into the cell and metabolized, the complete set of at

228

SCHWARZ et al.

least the hydrolytic enzymes must be present. The presence of the genes alone as shown by gene cloning or genome sequencing is not sufﬁcient; expression and secretion of the proteins must also occur. Furthermore, the presence of a hydrolytic gene (e.g., for an endo-xylanase) is not sufﬁcient to prove xylan degrading activity, even if the enzyme is found extracellularly. To hydrolyze xylan, cellulose, raw starch, and a number of other natural polysaccharides, a network of enzymes is needed, a socalled enzyme system. Examples of such enzyme systems are discussed below. It is obvious from the synergism explained above and from the immobility of most polymeric substrates that a high local concentration of all enzyme components necessary for the substrate degradation is needed for efﬁcient hydrolysis. Two strategies are possible: 1. To increase the concentration of all enzyme components in the medium, or 2. To combine the necessary components in a complex and to add a binding module for the substrate to hold the complex on the substrate once it has found it. Both possibilities are realized by clostridia as is explained with the cellulase enzyme systems below. An intermediate possibility is found with the species Caldicellulosiruptor and Anaerocellum, where large proteins are secreted containing several catalytic and noncatalytic modules in one polypeptide chain. The cellulosome of the clostridia, a protein complex on a scaffolding protein, seems to be the more elegant and ﬂexible solution. A. STARCH DEGRADATION Starch is an abundant polymer in plant biomass and consists of two components: amylose, a linear polymer of -1,4-linked glucose residues, and amylopectin, a branched polymer in which amylose chains are connected via -1,6-linkages. The relative distribution of amylose and amylopectin in a starch molecule and the degree of branching depends on the source of starch. Complete degradation of starch is achieved by endo-acting (-amylase) and exo-acting (-amylase and glucoamylase) enzymes. Enzymes that hydrolyze the -1,6-linkages are named pullulanase or debranching enzymes. Maltose and short oligosaccharides produced during primary hydrolysis are converted to glucose by -glucosidase. Starch is a good substrate for most of the saccharolytic clostridia, and all types of starch hydrolases have been found among them (Mitchell, 2001). Nevertheless, the knowledge

EXTRACELLULAR CLOSTRIDIUM ENZYMES

229

about starch-degrading enzymes from clostridia is limited. In contrast to the many Clostridium species able to degrade starch, there are relatively few entries in the database on hydrolases of GH13 (-amylase, pullulanase, cyclodextrin glucanotransferase), GH14 (-amylase), or GH15 (glucoamylase) (Coutinho and Henrissat, 1999b). Furthermore, although starch was one of the preferred substrates for the industrial acetone-butanol fermentation by C. acetobutylicum, very little is known about its starch-degrading enzyme system (Gerischer and Du¨rre, 1988; Paquet et al., 1991). Sequencing of the genome has identiﬁed a few genes related to starch degradation in this important organism (Table I; No¨lling et al., 2001). Thermophilic species and their thermostable enzymes have attracted particular interest (Antranikian, 1990). In some of these organisms, which later were reclassiﬁed as members of the genera Thermoanaerobacter and Thermoanaerobacterium, a novel type of pullulanase, which hydrolyzes -1,4- and -1,6-linkages, was identiﬁed (Spreinat and Antranikian, 1990). Other enzymes that hydrolyze both types of linkages are the glucoamylase and the -glucosidase of C. thermosaccharolyticum and of C. beijerinckii (Albasheri and Mitchell, 1995; Ganghofner et al., 1998; Specka et al., 1991). In addition, synergistic action of pullulanase and -amylase (cyclodextrin glucanotransferase) has been observed (Spreinat and Antranikian, 1992). In Fig. 4, the action of enzymes from clostridia and related bacteria on a starch molecule is illustrated. B. CELLULOSE DEGRADATION Cellulose is a completely insoluble, partially nonhydrated, and crystalline substrate that poses special difﬁculties for enzymatic hydrolysis. Although it is a chemically homogeneous, unbranched polymer of -1,4-linked glucopyranose residues, it is structurally heterogeneous. Only a very small fraction of the substrate molecules on the surface or in amorphic regions of the crystal are susceptible to immediate enzyme attack. The current understanding of enzymatic cellulose hydrolysis is as follows: an endo-glucanase binds with its attached binding module (CBM) to the surface of a substrate bundle, opens the cellulose molecule at one of a few accessible sites and consequently produces a new reducing and a nonreducing end. The endo-glucanase stays bound near this site and may open other available cellulose chains in reachable distance. Processive glucanases (exo-glucanases) ﬁnd the open ends and walk successively along the cellulose thread either from the nonreducing or the reducing end. They produce

FIG. 4. Enzymes from clostridia and related bacteria involved in the degradation of starch. Glucose units in the starch molecule with a reducing end are drawn in black; those with a nonreducing end are in grey. Data taken from Coutinho and Henrissat (1999a).

EXTRACELLULAR CLOSTRIDIUM ENZYMES

231

cellobiose (cellobiohydrolases) or cellotetraose (processive endo-glucanases), depending on the enzyme type (Reverbel-Leroy et al., 1997), widen the gap, and expose another layer of cellulose chains on the surface of the crystal, which may in turn be attacked by endo-glucanases. The cellodextrins produced are transported into the bacterial cells, where they are hydrolyzed by -glucosidases to glucose or, energetically more favorable, cleaved phosphorolytically by phosphorylases to glucose-1-phosphate. All cellulases cleave a -1,4-glucosidic bond by a hydrolytic reaction (hence ‘‘-1,4-glucanases’’). It is the same type of chemical reaction that takes place, but the mode of attack differs: cellulases may be endoor exo-glucanases (cellobiohydrolases or processive endo-glucanases); exo-glucanases may be active from the reducing or the nonreducing end of the molecule. Only a combination of enzymes with a different mode of action works synergistically and degrades the crystalline substrate effectively (Barr et al., 1996). Some enzymes hydrolyze the glucosidic bond by an inverting mechanism, others by a retaining mechanism (Davies and Henrissat, 1995). However, this difference in the hydrolytic mechanism does not seem to play a role in action modes; neither is the basic fold of an enzyme important, which is reﬂected in the glycosyl hydrolase family (GH family) to which a catalytic module is assigned (Coutinho and Henrissat, 1999a,b). Some of the GH families contain exo- as well as endo-glucanases (e.g., GH5, GH9). The endo- or exo-mode of a given enzyme is determined by the depth and the accessibility of the active site pocket; the processivity seems to be a function of attached substrate-binding modules and their orientation towards the hydrolytic center of the catalytic module (Barr et al., 1996; Bayer et al., 2000). At least in some enzymes the direction of the processivity could be explained by the way the substrate is bound and released and not by the gross structure of the protein backbone (Parsiegla et al., 2000). An especially high synergism between bacterial cellulases has been described between enzymes of GH48 and GH9 (Riedel et al., 1997), which are present in all cellulase enzyme systems known so far in bacteria. The synergism is higher with higher enzyme concentration, (i.e., the vicinity of proteins of both types is a crucial factor). By placing two enzymes close to each other, the cellulolysis can be optimized. Many clostridia reach this goal by packing enzymes of suitable types together in a cell- or substrate-bound huge enzyme complex, the cellulosome. One species secretes the enzymes separately as a ‘‘soluble’’ enzyme system. Examples for both types are given next. Interestingly, cellulosomes have been found exclusively in clostridia and some closely related Lachnospiraceae, such as Butyrivibrio and

TABLE III BACTERIA PRODUCING THE EXTRACELLULAR PROTEIN COMPLEX, THE CELLULOSOME Taxonom. group Clostridiaceae

Lachnospiraceae

a

Genus Clostridium

Species

Temp.

acetobutylicum

M

Source Sewage

Reference Sabathe et al., 2002

cellulovorans

M

Wood fermenter

Tamaru et al., 2000

cellobioparum

M

Rumen

Lamed et al., 1987

cellulolyticum

M

Compost

Be´laich et al., 1997

josui

M

Compost

Kakiuchi et al., 1998

papyrosolvens

M

Paper mill

Pohlschro¨der et al., 1995

thermocellum

T

Sewage, soil

Lamed et al., 1987

Acetivibrioa

cellulolyticus

M

Sewage

Ding et al., 1999

Bacteroidesa

cellulosolvens

M

Sewage

Lamed et al., 1991

Butyrivibrio

ﬁbrisolvens

M

Rumen

Berger et al., 1990

Ruminococcus

albus

M

Rumen

Ohara et al., 2000

ﬂavefaciens

M

Rumen

Aurilia et al., 2000

succinogenes

M

Rumen

Fields et al., 2000

Unambiguous assignment to the Clostridiaceae due to the 16S rDNA sequence in phylogenetic tree construction by ARB. Temperature: M, mesophilic; T, thermophilic.

EXTRACELLULAR CLOSTRIDIUM ENZYMES

233

Ruminococcus strains (Table III). Other species have been sporadically reported but were so far not proven by genetic data. 1. Clostridium stercorarium: A Soluble Cellulase System Only two enzymes that hydrolyze cellulose could be isolated from the culture supernatant of C. stercorarium NCIB 11764, the type strain (Bronnenmeier et al., 1990, 1991). The search for other components in culture supernatants was not successful (Bronnenmeier, personal communication). Extensive screening of genomic libraries in cosmid and bacteriophage Lambda vectors also revealed not more than two genes involved in the production of cellulases: celZ and celY (Schwarz et al., 1989). These genes coded for the cellulases Cel9Z, an endo-glucanase with some exo-glucanase activity, and Cel48Y, an exoglucanase. The gene products were identical to the previously isolated Avicelases I and II, respectively (Bronnenmeier et al., 1997; Jauris et al., 1990). Both enzymes are modular proteins consisting of an N-terminal catalytic module, a CBM3c module, and a binding module of family 3b with high afﬁnity for crystalline cellulose (CBM3b) (Fig. 5). In both enzymes the presence of a cellulose binding module enhances the local concentration of the enzymes on the substrate surface and is necessary for the activity on the solid substrate. The arrangement CBM3c-CBM3b occurs in both enzymes. In addition, both modules in inverse order are located downstream in Cel9Z, resulting in the order CBMc-b-b-c (Jauris et al., 1990). The CBM3c module adjacent to the catalytic module of Cel9Z is essential for the enzymatic activity at elevated temperature and has no experimentally

FIG. 5. Structure of the cellulase cluster in the Clostridium stercorarium genome. The order and approximate size (numbers) of the genes celY and celZ, the direction of transcription (arrows), and the module architecture of the cellulases is indicated.

234

SCHWARZ et al.

detectable binding activity (Riedel et al., 1998a). It may function, however, in feeding the emerging cellulose molecule into the active site pocket and thus determining the orientation of the processive activity of the enzyme, as described previously. No indication for a cellulosomal structure could be found for the cellulolytic activities of C. stercorarium: (1) both isolated enzymes were ‘‘free’’ exo-enzymes and could be puriﬁed as single proteins from the culture supernatant; (2) despite the isolation of dozens of glucanase clones, no gene other than celY and celZ was obtained (unpublished observation); and (3) none of the genes for extracellular enzymes cloned so far from C. stercorarium contained the dockerin module, which is typical for all hitherto identiﬁed cellulosome components (see below). Nevertheless, the two cellulases, Cel48Y and Cel9Z, constitute a functionally complete enzyme system in which both components are essential for the hydrolysis of crystalline cellulose. The combination of exo- and endo-glucanases is typical for the soluble enzyme systems of the cellulolytic bacteria. Moreover, GH9 and GH48 enzymes are the major components in all bacterial cellulase systems. Cel48Y and Cel9Z show a distinct synergistic interaction in the degradation of microcrystalline cellulose, which is dependent on the ratio of the two enzymes and on the type of the cellulosic substrate (Riedel et al., 1997). The synergism depends on the simultaneous presence of both enzymes and is not expressed by sequential addition of the two activities. To investigate this synergism further, a bifunctional hybrid, Cel48YCel9Z, was constructed with the structure GH48-CBM3c-CBM3b-GH9CMB3c (Riedel et al., 1998b). The large fusion protein (170 kDa) was expressed in E. coli and puriﬁed. It exhibited endo- as well as exoglucanase activity, and it retained the thermostability of the parent enzymes. But its cellulolytic activity was threefold to fourfold higher than the sum of the individual enzyme activities, underscoring the effect of packing two catalytic activities physically together. A natural hybrid enzyme, CelA, with a similar structure was identiﬁed in the extremely thermophilic, nonclostridial bacterium Anaerocellum thermophilum (Zverlov et al., 1998a). It also consists of GH9 and GH48 modules connected to CBM3 modules. It is able to hydrolyze microcrystalline cellulose. Both catalytic modules showed sequence identities of about 70% to the C. stercorarium cellulases Cel48Y and Cel9Z, respectively, and were active if expressed separately as recombinant proteins. C. stercorarium so far is the only Clostridium species shown to have a soluble cellulase system comparable to that of other bacterial genera,

EXTRACELLULAR CLOSTRIDIUM ENZYMES

235

especially the actinomycetes, and does not possess a cellulosome. Other cellulolytic clostridia have not been investigated in such detail (Schwarz, 2003). Whether C. stercorarium, similar to C. acetobutylicum, lost a cellulosome during evolution and became a hemicellulose specialist or if it simply did not acquire the cellulosomal genes can only be determined by genomic sequencing. The reduction or development to the simplest known cellulase systems with only two components is astonishing. This is a singular observation among the bacteria and could be an encouraging model for an industrial cellulase preparation. However, cellulose hydrolysis in C. stercorarium is comparatively slow and incomplete, although it allows growth of the cells on pure cellulose as sole carbon source. C. stercorarium feeds the cello-oligosaccharides into its catabolism by phosphorylation through the cellobiose phosphorylase CepA and cellodextrin phosphorylase CepB, rather than by the energy-wasting hydrolytic -glucosidase action (Reichenbecher et al., 1997). These enzymes can be isolated from the cell extract and seem to be located intracellularly. 2. Clostridium thermocellum Cellulosomes The cell-free culture supernatant of C. thermocellum contains a cocktail of enzymes with high cellulolytic activity on crystalline cellulose (Fig. 6, see color insert). Dependent on the growth phase of the bacterial culture, the majority of this activity is cell bound. Whether

FIG. 6. (Left) Hydrolysis of crystalline cellulose by C. thermocellum. Cellulose powder (MN300) in a thin layer of agar-medium on top of an agar plate is completely degraded around C. thermocellum colonies: the dark background is shining through. (Right) A strip of Whatman No. 1 ﬁlter paper is decomposed by a growing culture (left to right: uninoculated, culture after 1 and 2 days).

236

SCHWARZ et al.

enriched from the cell surface or from the culture ﬂuid, there are more than 25 different extracellular enzymes visible in a denaturing electrophoresis gel. Many of these proteins are hydrolytically active on cellulose. However, the complete cellulolytic activity could not be reconstituted from single, isolated components (Bhat et al., 1994; Morag et al., 1996; Wu et al., 1987). Nevertheless, a huge multienzyme complex was isolated, which in intact form was highly active on crystalline cellulose. It was called a cellulosome and could be isolated either from the culture supernatant or from the cells (Lamed et al., 1983). A puriﬁcation method called ‘‘afﬁnity digestion’’ uses the adsorption of the cellulosomes to cellulose ﬁbers, washing the cellulose-cellulosome complex, followed by the complete hydrolysis of the cellulose. Cellulosomes are then puriﬁed by gel ﬁltration chromatography (Morag et al., 1992). This puriﬁcation scheme makes use of the basic characteristics of the cellulosomes: They not only bind to the substrate but also to the cell surface and thus form a bridge that holds the cell on its much larger substrate. This is energetically favorable because a sufﬁcient enzyme concentration can easily be reached on the cell surface without producing a high amount of extracellular protein. Furthermore, the enzymatic components necessary for optimum synergism stay in close proximity, and the products of hydrolysis are present in high concentration near the cell surface, ready for uptake by the enzyme-producing cells. They are not ‘‘wasted’’ for other competing bacteria (Lynd et al., 2002). The large protein complexes on the outer surface of the bacterial cells could be made visible by electron microscopy, but only after ﬁxation with cationized ferritin (Bayer and Lamed, 1986; Madkour and Mayer, 2003; Mayer et al., 1987). The attachment of the cells to the substrate via the cellulosomes was also observed (Bayer et al., 2000). Isolated, puriﬁed cellulosomes of C. thermocellum vary in size depending on the strain (from 2.0 to 6.5 Mda) and may even aggregate to large supercomplexes, called poly-cellulosomes (up to 100 Mda). Approximately 25 genes for cellulosomal genes have been isolated from genomic libraries by random screening for hydrolytic activity (for review see Schwarz, 2001). There is no proof yet for all gene products that they are actually present in the cellulosome. Moreover, a number of components surprisingly have a hydrolytic activity that apparently has nothing to do with cellulose degradation. These components degrade other polysaccharides in biomass—such as mixedlinkage -glucan, pectin, xylan, mannan or chitin—which in natural substrates wrap the cellulose crystals (Blum et al., 2000; Kurokawa et al., 2001; Spinnler et al., 1986; Zverlov et al., 1994, 2002a). This hydrolytic activity is in contrast to the lack of fermentation ability of

EXTRACELLULAR CLOSTRIDIUM ENZYMES

237

C. thermocellum for pentoses: evidence is accumulating that the cellodextrins derived from cellulose (not glucose or cellobiose) are the best substrate providing the most energy (Lynd et al., 2002). Hemicellulases seem to have an accessory function in providing access to the preferred substrate. However, it was shown recently that C. thermocellum also ferments -1,3-glucan (Fuchs et al., 2003). Gene cloning, together with immunological investigations, provided clues for the presence of noncatalytic proteins in the cellulosomes that are involved in structure-forming or other functions. Most important was the discovery of a scaffolding protein (CipA, ‘‘cellulosome integrating protein’’), the so-called scaffoldin, which has nine docking sites called cohesins (Fujino et al., 1992; Gerngross et al., 1993). The binding partner on the catalytic cellulosome components is a conserved twofold repeat of 24 amino acid residues, the dockerin (type I) (Tokatlidis et al., 1991). The dockerin sequences of the different cellulosomal genes are well conserved. Slight differences together with the differences in the sequence of the cohesins may lead to preferences of speciﬁc cellulosomal components to speciﬁc sites. This assumption is not in contrast to experimental results showing that a single cellulosomal component can bind to different cohesins (Fierobe et al., 2001). X-ray analysis of the cohesin structure revealed a ﬂat binding area exposing surface residues for relatively unspeciﬁc interaction with the dockerins (Lytle et al., 2000; Mechaly et al., 2000; Shimon et al., 1997). The scaffoldin CipA brings together nine catalytic components in close proximity and thus may stimulate the synergism between the enzymes. However, there are still other modules besides the cohesins in the scaffoldin: a cellulose binding module (CBM3a) on the C-terminus for tight binding to a crystalline cellulose surface and a dockerin module on the N-terminus. This dockerin type II is not as closely related to the dockerins of the catalytic components as they are to each other. Type I and II dockerins bind to their complementary cohesin types but are not cross-reacting (Fierobe et al., 2001). The binding partner for the CipA dockerin was found in other noncatalytic extracellular proteins, SdbA and OlpB (and probably others), which carry S-layer homologous modules, anchoring the proteins in the bacterial cell wall, and cohesins of type II (Leibowitz and Be´guin, 1996; Lemaire et al., 1995). The role of other outer layer proteins, which were identiﬁed as reading frames in the genome of C. thermocellum, has still to be elucidated. To date, only one of the dockerin(I)-bearing components of the cellulosome was obviously noncatalytic: CseP with homology to CotH, a spore coat forming structural protein in Bacillus subtilis (Table IV;

238

SCHWARZ et al. TABLE IV LIST OF CELLULOSOMAL COMPONENTS IN THE GENOME OF C. Reading frame/function

Structurea

THERMOCELLUM

Ref.b

Structural component 1. CipA (c) scaffoldin, Cthe1933-1930

2(Coh1)-CBM3a-7(Coh1)X2-Doc2

þ (Fujino et al., 1992; Zverlov et al., in prep.)

GH2 2. Cthe1580

GH2-CBM6-Doc1

GH5 3. CelO cellobiohydrolase, Cthe1674

CBM3b-GH5-Doc1

Zverlov et al., 2002b

4. Cthe1575

GH5-CBM6-Fn3-Doc1

5. CelB endoglucanase, Cthe0374

GH5-Doc1

Grepinet and Be´guin, 1986

6. CelG endoglucanase, Cthe0885

GH5-Doc1

þ (Lemaire and Be´guin, 1993)

7. Cthe0444

GH5-Doc1

GH8 GH8-Doc1

þ (Be´guin et al., 1985; Zverlov et al., in prep.)

CBM4-Ig-GH9-2(Fn3)CBD3b-Doc1

þ (Zverlov et al., 1998b)

10. CelK cellobiohydrolase, Cthe2598

CBM4-Ig-GH9-Doc1

þ (Zverlov et al., 1999)

11. CelD endoglucanase, Cthe0968

Ig-GH9-Doc1

Joliff et al., 1986

12. Cthe1953

GH9-CBM3c-CBM3b-Doc1

8. CelA endoglucanase, Cthe0722 GH9 9. CbhA cellobiohydrolase

13. Cthe0850

GH9-CBM3c-CBM3b-Doc1

14. CelN endoglucanase, Cthe1222

GH9-CBM3c-Doc1

þ (Zverlov et al., in prep.)

15. CelR endoglucanase, Cthe1837

GH9-CBM3c-Doc1

þ (Zverlov et al., in prep.)

16. CelQ endoglucanase, Cthe0300

GH9-CBM3c-Doc1

þ (Arai et al., 2001)

17. CelF endoglucanase, Cthe0382

GH9-CBM3c-Doc1

Navarro et al., 1991 (continued )

239

EXTRACELLULAR CLOSTRIDIUM ENZYMES TABLE IV (Continued) Structurea

Reading frame/function

Ref.b

Structural component 18. Cthe1308

GH9-CBM3c-Doc1

19. Cthe0727

GH9-Doc1

20. CelT endoglucanase

GH9-Doc1

þ (Kurokawa et al., 2002)

21. XynD xylanase, Cthe0688

CBM22-GH10–Doc1

þ (Zverlov et al., in prep.)

22. XynC xylanase, C the0626

CBM22-GH10-Doc1

þ (Hayashi et al., 1997)

23. XynA, XynU xylanase, Cthe1161

GH11-CBM4-Doc1-NodB

þ (Hayashi et al., 1999)

24. XynB, XynV xylanase

GH11-CBM4-Doc1

þ (Hayashi et al., 1997)

25. LicB lichenase

GH16-Doc1

þ (Zverlov et al., 1994a,b)

26. ChiA chitinase

GH18-Doc1

þ (Zverlov et al., 2002a)

27. ManA mannanase, Cthe0533

CBM-GH26-Doc1

þ (Halstead et al., 1999)

28. Cthe2142

GH26-Doc1

29. Cthe1127

GH30-CBM6-Doc1

30. Cthe2333

GH53-Doc1

31. Cthe0269

GH81-Doc1

Xylanases

Other hemicellulases

Putative glycosidases 32. Cthe1665

GH39-2(CBM6)-Doc1

33. Cthe1579

GH43-CBM6-Doc1

34. Cthe0268

GH43-CBM13-Doc1

35. Cthe0484

GH43-2(CBM6)-Doc1

GH48 36. CelS exoglucanase Cthe0939

GH48-Doc1

þ (Wang et al., 1993)

GH74-CBM2-Doc1

þ (Zverlov et al., in prep.)

Xyloglucanase 37. XghA xyloglucanhydrolase, Cthe2335

(continued )

240

SCHWARZ et al. TABLE IV (Continued) Structurea

Reading frame/function

Ref.b

Structural component Putative carbohydrate esterases 38. Cthe0066

Fn3-CE12-Doc1-CBM6-CE12

39. Cthe1577

CE1-CBM6-Doc1

Putative pectinases 40. Cthe2008

GH28-Doc1

41. Cthe2236

PL1-Doc1-CBM6

42. Cthe1810

<-Doc1-CBM6-PL9

43. Cthe2234

PL10-UN-Doc1

44. Cthe0702

Doc1-CBM6-PL11

Multifunctional components 45. CelJ cellulase, Cthe0301

X-Ig-GH9-GH44-Doc1-X

þ (Ahsan et al., 1996)

46. CelH endoglucanase, Cthe0837

GH26-GH5-CBD9-Doc1

Yaguee et al., 1990

47. Cthe1667

GH30-GH54-GH43-Doc1

48. Cthe1211

GH54-Doc1-GH43

49. Cthe1666

GH54-GH43-Doc1

50. XynZ xylanase, Cthe1691

CE1-CBM6-Doc1-GH10

þ (Grepinet et al., 1988; Zverlov et al., in prep.)

51. XynY xylanase, Cthe2036 CBM22-GH10-CBM22Doc1-CE1

Fontes et al., 1995

52. CelE endoglucanase, Cthe0940, Cthe2702, Cthe2514

þ (Hazlewood et al., 1990)

GH5-Doc1-CE2

Putative protease inhibitors 53. Cthe1412

Fn3-Doc1-serpin

54. Cthe1413

Doc1-serpin

Components with unknown function 55. Cthe0694

2(UN)-UN-UN(CelP 550–870)-Doc1 (continued )

241

EXTRACELLULAR CLOSTRIDIUM ENZYMES TABLE IV (Continued) Structurea

Reading frame/function

Ref.b

Structural component 56. Cthe1578

UN-CBM6-Doc1

57. CseP, Cthe1223

UN-Doc1

58. Cthe1474

Doc1-UN

59. Cthe0287

UN1-UN2-Doc1

60. Cthe0416

Doc1-UN

61. Cthe0073

UN-Doc1

62. Cthe0649

UN-Doc1

þ (Zverlov et al., 2003)

Known components and hypothetical proteins in the unfinished genome sequence. Only reading frames containing dockerin modules, a Shine-Dalgarno sequence and stop codon, and a recognizable module composition are listed (no ORF fragments). The protein designation and enzymatic activity or function (if known), and the ORF number from http://genome.ornl.gov/microbial/cthe/ (Cthe) are given. If no Cthe number is given, the gene is cloned but not yet contained in the genomic sequence. The components are sorted according to their putative function, as obvious from the catalytic module family. Components with more than one catalytic module or unknown modules are listed at the end. a Module classification according to Coutinho and Henrissat (1999a), URL: http://afmb.cnrs-mrs.fr/ CAZY/; Coh, cohesin module; Doc, dockerin module; CBM, carbohydrate binding module; X, hydrophobic module; GH, glycosyl hydrolase family; Fn3, fibronectin III module; Ig, immunoglobulin like fold; NodB, acetylxylan-esterase NodB type; CE, carbohydrate esterase; PL, pectin lyase; UN, unknown module; serpin, serine-protease inhibitor homologue. b A ‘‘þ’’ in the reference column indicates that the component was shown to be present in the cellulosome.

Zverlov et al., 2003). Its suspected role in the three-dimensional stabilization of the cellulosome or in the multimerization of different cellulosomal particles has still to be demonstrated. Other hypothetical non-catalytic proteins in the cellulosome were identiﬁed among the reading frames obvious from the genomic sequence. Two of them carry protease inhibitor modules (#53 and #54 in Table IV); others carry modules with unknown function (#55–62). These proteins, however, have not been shown to be expressed; their function can only be speculated. The CBM3a module locates the whole cellulosome ﬁrmly on the surface of the crystalline substrate. The large size of the structure and the spacer modules (X modules) may give single components some ﬂexibility to successively attack a number of sites around that location. In a cellulosome it is not necessary that each single catalytic component has its own CBM as with soluble enzyme systems. Indeed, only a few components have CBMs, many of them with a different binding

242

SCHWARZ et al.

speciﬁcity compared to the CipA-CBM3. The role of the CipA-CBM3a was shown experimentally by fusing the recombinant Cel5E endoglucanase with the CipA-CBM ([GH5-DD]CelE-CBM3CipA). The hydrolytic activity of this hybrid protein was strongly enhanced only on crystalline cellulose but not on amorphous cellulose. No stimulation was found if Cel5E and CBMCipA were added separately to the crystalline substrate (Ciruela et al., 1998). This is an indication that a structural arrangement like that given in the cellulosome is necessary for the synergistic effect. Only a few genes of cellulosomal components were identiﬁed by targeted immunological methods (e.g., cipA) or by partial protein sequencing. And although a number of laboratories have isolated clones active against polysaccharide substrates from genomic libraries, the list of cloned components is surely incomplete (albeit highly repetitive) (Guglielmi and Be´guin, 1998). Cloned genes whose presence in the cellulosome has been demonstrated are shown in Table IV. Other genes may be inactive (not expressed) or expressed only under certain circumstances (e.g., by induction through speciﬁc substrates or substances). On the other hand, some genes may be hard to screen for because of a lack of a suitable substrate or its unusual activity proﬁle, or may be hard to clone because their DNA is toxic to or not well expressed in the cloning host or is easily degraded. Using a pure cloning approach, we will never get all genes and we will not know if a cloned gene is expressed and incorporated in the cellulosome. Only with the emerging genomic sequence of the type strain ATCC 27405 did a rational approach toward a complete list of the cellulosomal components become available. The puriﬁed cellulosomes have been denatured and the components separated by proteomic methods. The protein spots were identiﬁed by comparing the MALDI-TOF data with the translated genomic sequence, which, however, is still fragmentary and full of alignment errors (Zverlov et al., in preparation). The major components of the cellulosome (as it is produced from cellobiose grown cells) were identiﬁed by this method. Surprisingly (but not unexpectedly), the ﬁrst results showed that three of the major proteins had not been cloned before. The advantage of this protein-directed approach is that only proteins show up that are translated, secreted, and correctly incorporated in the huge extracellular complex. At the same time, the distribution pattern of major and minor components can be addressed, together with an analysis of the possible shift in composition that is due to changes in the substrate. Another approach possible with the genomic sequence is the compilation of a list of the proteins that contain a dockerin or a cohesin

EXTRACELLULAR CLOSTRIDIUM ENZYMES

243

module and thus must have something to do with the cellulosome. A provisional list is shown in Table IV. This list is still incomplete, and in some cases the DNA sequence shows an unveriﬁed composition of modules that is due to the unﬁnished character of the genomic sequence. Conspicuous is the large number of hemicellulase (xylanase, -1,3/1,4- and -1,6-glucanase, chitinase, mannanase, galactanase, xyloglucanase), glycosidase (-xylosidase, -arabinofuranosidase) and esterase genes that did not show up in the genomic library screening, possibly because they have not been screened for. A complete set of genes potentially involved in pectin degradation showed up (pectate lyase, polygalacturonase, esterase) (Zverlov et al., in preparation). The reading frames with homology to protease inhibitors may have to do with the suspected proteolysis of the cellulosomes in outgrowing cultures of C. thermocellum. Interesting is the occurrence of genes containing more than one catalytic module—for example, combining -xylosidase and -arabinofuranosidase modules (GH43/54) or xylanase and arabinoxylan-esterase modules (GH10/CE1), which are functionally related. This enhances the capacity of the cellulosome with its only nine cohesin sites on the scaffoldin and reminds the module arrangement of the extracellular enzymes of the Caldicellulosiruptor-Anaerocellum group. The number of modules with unknown or unrelated sequences is relatively small compared with the average situation in reading frames obtained from genomic sequencing. This reﬂects the high intensity of research on glycosyl hydrolases and the high coverage of cloned genes. Further investigations of the composition of cellulosomes have to reveal the presence of the encoded proteins in the cellulosome of C. thermocellum. In addition to the cellulosomal proteins, membrane binding proteins are essential for the attachment of the cellulosome (e.g., SdbA, OlpA, OlpC) or single proteins to the cell wall. These may also play a role in the assembly of the huge protein complex. Many of these proteins may be detected by subproteomic approaches (e.g., with isolated bacterial cell walls). 3. Does Clostridium thermocellum Have a Soluble Cellulase System Too? In addition to the cellulosome, other soluble, noncellulosomal enzymes have been cloned from C. thermocellum (Table V). Cel5C and Lic16A seem to have a role in the hydrolysis of soluble -1,3(4)-glucans and may be restricted to substrates connected with the cellulose surface to which they bind via their CBMs. Lic16A is, in addition, able to degrade -1,3-glucan (Fuchs et al., 2003). CelM was reported to have

244

SCHWARZ et al.

endo-glucanase activity (Kobayashi et al., 1993); however, its role remains doubtful because the sequence is not homologous to any other endo-glucanase but to aminopeptidases. The soluble enzyme component Cel9I has a high degree of similarity to Cel9Z of C. stercorarium (in structure and sequence homology), which is also a soluble cellulase. Both enzymes are processive endo-glucanases (Riedel et al., 1998a; Zverlov et al., 2003). In C. stercorarium, Cel9Z together with Cel48Y is able to degrade crystalline cellulose in synergistic co-operation (Riedel et al., 1998b). Indeed, database screening of the unﬁnished C. thermocellum genome revealed a reading frame homologous to Cel48Y that also is not connected to a dockerin module (Table V). Although the expression of both genes has not been investigated so far, C. thermocellum seems to possess a second, soluble cellulase system besides the cellulosome. C. thermocellum is so far the only bacterium having more than one gene with a GH48 module. Once the genome sequence is complete, more reading frames may be identiﬁed that are not connected to a dockerin module and thus are not located in the cellulosome. The expression of such genes has to be veriﬁed—for example, by concentration of culture supernatants and analysis of the activities of their proteins. Intracellular cellulolysis related enzymes are, for example, the glucosidases BglA, BglB, and the cellobiose and cellodextrin phosphorylases CdP and CbP. The latter enzymes enhance the energetical efﬁciency of cellulose hydrolysis for the cell considerably (Lynd et al., 2002).

TABLE V NONcELLULOSOMAL EXTRACELLULAR PROTEINS OF C. Gene

Structure

THERMOCELLUM

Reference

Lic16A, Cthe0535

SLH(1–3)-GH16-CBM4a(1–4)

Fuchs et al., 2003

Cel5C, Cthe0537

GH5

Schwarz et al., 1988

Cel9I, Cthe1219

GH9-CBM3c-CBM3b

Hazlewood et al., 1993; Zverlov et al., 2003

CelM, Cthe0317

homologous to peptidases

Kobayashi et al., 1993

XynX

CBM(TSM)-GH10-CBM9-CBM9

Kim et al., 2000

Cthe0247

GH48-CBM3c-CBM3b (hom. to Cel48Y/C. stercorarium)

This chapter

Abbreviations as in Table IV. TSM, thermostabilizing module.

EXTRACELLULAR CLOSTRIDIUM ENZYMES

245

4. Cellulosomes of Mesophilic Clostridia A number of mesophilic clostridia have been found to produce cellulosomes for the hydrolysis of crystalline cellulose. Cellulolytic genes were cloned from Acetivibrio cellulolyticus (belonging to the genus Clostridium by 16S rRNA sequencing), C. acetobutylicum, C. cellulolyticum, C. cellulovorans, and C. josui. A recent review nicely summarizes the data on mesophilic cellulosomes (Doi et al., 2003). In contrast to the scattered gene location in the genome of C. thermocellum, the cellulosomal genes of the mesophilic clostridia are basically arranged as large gene clusters with nine (e.g., in C. cellulovorans) or even more genes as in C. cellulolyticum (Be´laich et al., 2002; Tamaru et al., 2000). The arrangement of the genes in the clusters is surprisingly similar: it starts with the scaffoldin gene that is followed immediately by the only GH48 enzyme gene, the major exo-glucanase component (compiled in Schwarz, 2001). These two components seem to be indispensable for the cellulolytic function of a cellulosome and probably are expressed with the highest efﬁciency. It is interesting to note that in the scaffoldins of the cellulosomes of C. cellulovorans, C. cellulolyticum, C. josui, and C. acetobutylicum (but not of C. thermocellum) hydrophobic modules were observed (up to ﬁve copies: formerly called X-modules). This module is not present in any catalytic cellulosomal components. The 3-D structure of the X2 module from CipC of C. cellulolyticum was determined to be an immunoglobulin-like fold with a remarkable conformational stability. It may have a function as a structural linker and a solubility enhancer between other modules (Mosbah et al., 2000). The majority of the enzyme genes contain GH9 and GH5 catalytic modules, both of which contain endo- and exo-glucanases. Besides the catalytic module, noncatalytic domains play an important role in deﬁning the function of an enzyme. The architecture of the cellulosomal components is summarized in Table VI. Given the similarity in the operon structures, it is not surprising that the module composition of single components is also quite similar. For almost any module architecture type present in one bacterium, a similar component is found in almost any other cellulosome. This list is not complete, as many components may not yet have been identiﬁed and cloned. In addition to the gene clusters, cellulosomal components are also encoded by unlinked monocistronic genes. It is rather improbable that a complicated enzyme complex like the cellulosome would be invented independently by different bacteria. Heterologous gene transfer of gene clusters followed by slight

TABLE VI CELLULOSOMAL COMPONENTS OF MESOPHILIC CLOSTRIDIA Structure GH48-DD

Possible function Exoglucanase

C. acetobutylicum

C. cellulolyticum

CelF

CelF ManK, CelD, CelA

GH5/DD

Endoglucanase/mannanase

CelA, ManA

(SLH)3-GH5-X-DD

Endoglucanase

CAC3469

GH8-DD

Endoglucanase

GH9-DD

Endoglucanase

CelL

CelM

GH9-CBM3c-DD

Endoglucanase

CelH, CelG

CelG, CelH, CelJ

CBM2-GH9-DD

Endoglucanase

CBM4-Ig-GH9-DD

Endoglucanase

GH11-DD-CE4

Endoxylanase

C. cellulovorans ExgS

C. josui CelD

ManA, EngB EngE

CelC

CelB EngL EngH

CelE

EngY CelE

CelE

EngK, EngM XynA

GH27-DD

-Galactosidase

GH44-DD

Endoglucanase

EngA

GH74-DD

‘‘Sialidase’’

CAC0919

X-CBM2-PL9-DD

Pectate lyase

AgaA

PelA

The components are sorted by structure types. The modules are: GH, glycosyl hydrolase family; DD, dockerin module; SLH, S-layer homologous module; CBM, carbohydrate binding module; Ig, immunoglobulin-like fold; X, hydrophobic module; PL, pectate lyase module; CE, feruloyl esterase module.

EXTRACELLULAR CLOSTRIDIUM ENZYMES

247

rearrangements and adaptation of new components is a probable explanation. This is corroborated by the aforementioned high similarity of the arrangement of the functionally related genes. Moreover, at the 30 end of the C. cellulovorans gene cluster a transposase gene was observed as a possible source of the gene transfer (Tamaru et al., 2000). Although C. acetobutylicum does not degrade cellulose or produce an extracellular cellulosome, some components of the cellulosome were detected in culture supernatants. The presence of cellulosomal gene clusters became obvious during the annotation of the genomic sequence (No¨lling et al., 2001; Sabathe et al., 2002). For a bacterium like C. acetobutylicum, which can grow on starch and different oligosaccharides and monosaccharides, one might expect it to be an unnecessary burden to produce such a large and costly enzyme complex as the cellulosome. It might have acquired the genes in its evolution a long time ago but has managed to shut down its production. Single enzymatic components are helpful for the breakdown of biomass in its natural habitat and are produced at a low rate, but the complete set of enzymes is no longer produced and the genes underwent permanent changes. In the genome we still can see the remnants of this very recent evolutionary process. In an attempt to create a solventogenic bacterium that could use lignocellulosic biomass as substrate for the production of organic solvents on an industrial scale, a number of frame shifts and deletions in the cellulosomal genes were corrected and promoter sequences optimized (Sabathe et al., 2002). However, the bacterium still could not hydrolyze cellulose, and no enzymatically active large protein complex was obtained, although single components were active as recombinant enzymes expressed in other bacteria, and a mini-cellulosome was secreted by the Clostridium (Sabathe and Soucaille, 2003). The search for a cellulose-degrading, solvent-producing bacterium goes on (Montoya et al., 2001). Such a strain might be found among the mesophilic, cellulolytic clostridia that are widespread in natural environments. C. XYLAN AND HEMICELLULOSE DEGRADATION The major components of the noncellulosic polysaccharides in plant biomass are xylans, at least in angiosperms, where they account for 20–30% of the dry weight of woody tissue (Aspinall, 1980). Xylan has a heterogeneous composition, consisting of a homopolymeric, linear backbone of -1,4-linked D-xylopyranosyl residues, which are substituted with -1,3-linked L-arabinofuranosyl, -1,2-linked 4-O-methylglucuronic acid residues, or other sugar residues. Substitutions with acetic, p-coumaric, or ferulic acid are common.

248

SCHWARZ et al.

The enzymatic hydrolysis of arabinoxylan and glucuronoxylan requires the activity of backbone and side-chain cleaving activities. Involved are endo--1,4-xylanases (1,4--D-xylan xylanohydrolase, EC 3.2.1.8), possibly exo--1,4-xylanases (1,4--D-xylan xylohydrolase), and -D-xylosidases (1,4--D-xyloside xylohydrolase, EC 3.2.1.37); the side groups and substituents are removed by -L-arabinofuranosidases (EC 3.2.1.55), -D-glucuronosidases (EC 3.2.1), and esterases such as feruloylesterase (Donaghy et al., 2000). The best-investigated xylan hydrolyzing enzyme system within the clostridia is that of C. stercorarium. Hydrolytic proteins from the culture supernatant of the type strain NCIMB 27405 have been separated and characterized; some selected proteins are also from another strain, the Japanese isolate F-9 (Berenger et al., 1985; Bronnenmeier et al., 1990; Hayashi et al., 1997, 1999). An endo-xylanase and a celloxylanase— both as multiple protein species caused by partial degradation—a xylosidase, an -arabinofuranosidase, and a feruloyl esterase were puriﬁed. They correspond to the proteins Xyn10A, Xyn11C, Bx13B, and Arf51B, respectively (Adelsberger et al., submitted). The gene of the extracellular feruloyl esterase has not yet been identiﬁed (Donaghy et al., 2000). Ali et al. (1999) showed the presence of Xyn11C in the culture supernatant and most probably on the cell surface of strain F-9 by immunological methods. Another xylanase, Xyn11B, and an glucuronidase could not be identiﬁed in culture supernatants but were isolated from cell extracts and seem to be located intracellularly (Bronnenmeier et al., 1995; Sakka et al., 1996). The complete set of enzymes sufﬁcient to hydrolyze arabinoxylan could be cloned from the genome of C. stercorarium. The extracellular xylanases are surprisingly complex enzymes consisting of four and ﬁve modules (Table VII). It is conspicuous that only the intracellular xylanase Xyn11B in C. stercorarium does not have any additional, noncatalytic modules, as extracellular enzymes often possess. The cellulose binding modules CBM6 of Xyn11A and either CBM22 or CBM9 of Xyn10C show strong binding activity to crystalline cellulose (Ali et al., 2001; Bronnenmeier et al., 1996; Sun et al., 1998). Cellulose binding could ﬁx the xylanases ﬁrmly to the surface of the plant cell wall and thus keep them in the vicinity of their substrate xylan. The xylanases that have these binding modules might help the cellulolytic bacteria to polish the cellulose ﬁbers from the attached xylan, which then enables the degradation of the cellulose crystals. The cellodextrins derived from cellulose are of a much higher metabolic value for the bacterium than xylan and probably are the preferred substrate. Xylan is nevertheless a possible fermentation substrate for C. stercorarium.

249

EXTRACELLULAR CLOSTRIDIUM ENZYMES TABLE VII EXTRACELLULAR ENZYMES OF C.

STERCORARIUM INVOLVED IN THE

XYLANOLYTIC PATHWAY

Designation NCIMB

Acc. No. NCIMB

Design. F-9

Synonym

Acc. No. F-9

Module architecture

Xyn11A

AJ508403

Xyn11A

XynA

D13325

GH11-CBM6CBM6-(CBM6)

Xyn10B

AJ508407

Xyn10C

XynB

D12504

GH10

Xyn10C

AJ508408

Xyn10B

XynC

AB024743

CBM22-CBM22GH10-CBM9SLH-SLH

Arf51B

AF00264

-

-

GH51

Bx13B

AJ508405

-

-

GH3

Bg13Z

Z94045

-

-

GH3

Ram78A

AJ238748

-

-

GH78

Feruloylesterase

(not yet cloned)

The accession number of the sequence in the GeneBank is given, as well as the synonyms for the homologous sequences from strain F-9.

The xylanase Xyn10C seems to be able to form an anchor between the bacterial cell and the cellulosic surface. Ali et al. (2001) have shown that Xyn11C is one of the major surface layer proteins. Its high afﬁnity to cellulose brought about co-precipitation of bacterial cells washed of surface layer proteins, but only if the SLH modules were present. This indicates that Xyn10C on one hand binds to the surface layer with virtue of its SLH modules and on the other to cellulose with its CBMs. The binding of the cell to its substrate has a great advantage for the bacterium: a lower amount of exo-enzymes to be produced for saturation of the substrate and a higher concentration of hydrolysis products near the cell surface combined with diminished loss by diffusion. Xyn11A, Xyn10C, Arf51B, and Bx13B are major components in the C. stercorarium culture supernatant. These enzymes, and not the products of other genes also isolated from the genome, are sufﬁcient to hydrolyze arabinoxylan effectively and completely to its monomers (Adelsberger et al., submitted). Arf51B releases -arabinofuranosyl residues from the intact arabinoxylan, as well as from the oligosaccharides produced by the action of the endo-xylanases Xyn11A and/or Xyn10C. The resulting xylo-oligosaccharides are degraded to xylose by Bx13B. The point of attack of the two endo-xylanases seems to be different: the produced oligosaccharides differ as well in size as in

250

SCHWARZ et al.

FIG. 7. Enzymatic hydrolysis of arabinoxylan. The structure of arabinoxylan is drawn schematically (X ¼ xyloside residues, A ¼ arabinoside residues). The hydrolytic sites of attack by the C. stercorarium enzymes (arrows) xylanase XynA, -arabinofuranosidase ArfB, and -xylosidase BxlB are indicated.

the digestibility by -xylosidase Bx13B. The structure of single oligosaccharides could be resolved by sequential digestion with Bx13B and Arf51B. The complete degradation of arabinoxylan is schematically shown in Fig. 7. In analogy to arabinoxylan, it could be supposed that glucuronoxylan was hydrolyzed by extracellular enzymes. However, the only glucuronosidase activity detected in C. stercorarium cultures was located intracellularly. Consequently, it has to be surmised that glucuronoxylan is degraded by the extracellular xylanases to soluble, charged oligosaccharides that are—in contrast to the noncharged arabinoxylo-oligosaccharides—transported through the cell barrier and further degraded there. With in vitro assays using model substrates it has not yet been possible to clarify whether the two major extracellular xylanases Xyn11A and Xyn10C act synergistically on natural xylan. The presence

EXTRACELLULAR CLOSTRIDIUM ENZYMES

251

of different binding modules suggests a role in the digestion of different topologies in naturally occurring plant cell walls. It should be mentioned that the extracellular cellulosome of C. thermocellum contains a number of xylanase components (see Table IV). Five genes have been cloned and biochemically characterized, another two potential genes were identiﬁed in the genomic sequence. The breakdown may be enhanced by the activity of esterases: one esterase connected with Xyn11A (the NodB module), two with Xyn11Y and Xyn11Z (Blum et al., 2000), and two others identiﬁed in the genomic sequence (#38 and 39 in Table IV). The oligosaccharides are broken down by -arbinofuranosidases and -xylosidases, seven genes of which are present in the genome, but none has been characterized yet. In addition, potential genes for two mannanases (manA already characterized), an arabinogalactanase and a xyloglucanase are present, as well as genes for a pectate lyase, a polygalacturonase, and a rhamnogalacturonan lyase for the breakdown of pectin. It is not known if these genes code for active enzymes and if they are expressed. The presence of a chitinase in the cellulosome has already been shown; the gene chiA was identiﬁed (Zverlov et al., 2002a). V. Concluding Remarks Clostridia are very important organisms for modern biotechnology. In the past, they were used for the production of acetone and butanol, ﬂax retting, and indigo dyeing. In this chapter we have highlighted the features of some of the extracellular enzymes produced by these bacteria to degrade/hydrolyze biopolymers such as starch or cellulose. Some of theses enzymes or enzyme systems are unique among microorganisms. The enormous potential of clostridia as producers of industrially important enzymes is obvious. In the last decade, signiﬁcant progress has been achieved in the understanding of the structure-function relationships of the clostridial type of enzymes and their modules. The technique of enzyme modiﬁcation by adding modules or the increase in enzyme activity by complex formation will be a great stimulus for modern enzymology. In addition, genetic tools for clostridia have been developed for custom engineering of new production strains. Thus it seems to be possible now to engineer an enzyme with optimal features for a given purpose or even to create a special Clostridium species that is able to convert cheap, renewable biomass into desired valuable products. Thus it is not utopian to believe that, for example, C. acetobutylicum one day will transform cellulosic wastes directly to solvents.

252

SCHWARZ et al. ACKNOWLEDGMENT

We thank W. L. Staudenbauer for encouragement and many stimulating discussions and Karin Bronnenmeier for providing unpublished results. Experimental work carried out in the authors’ laboratories was supported by the Deutsche Forschungsgemeinschaft, the former Bundesministerium fu¨r Forschung und Technologie, and the Fonds der Chemischen Industrie. This study was also supported by a grant of the AvHumboldt Foundation to V.V.Z.

REFERENCES Adelsberger, H., Hertel, C., Glawischnig, E., Zverlov, V. V., Schwarz, W. H. (2004). Enzyme system of Clostridium stercorarium for hydrolysis of arabinoxylan: Reconstitution of the in vivo system from recombinant enzymes. Microbiology 150, 2257–2266. Ahsan, M. M., Kimura, T., Karita, S., Sakka, K., and Ohmiya, K. (1996). Cloning, DNA sequencing, and expression of the gene encoding Clostridium thermocellum cellulase CelJ, the largest catalytic component of the cellulosome. J. Bacteriol. 178, 5732–5740. Albasheri, K. A., and Mitchell, W. J. (1995). Identiﬁcation of two a-glucosidase activities in Clostridium acetobutylicum NCIB 8052. J. Appl. Bacteriol. 78, 149–156. Ali, M. K., Fukumura, M., Sakano, K., Karita, S., Kimura, T., Sakka, K., and Ohmiya, K. (1999). Cloning, sequencing, and expression of the gene encoding the Clostridium stercorarium xylanase C in Escherichia coli. Biosci. Biotechnol. Biochem. 63, 1596–1604. Ali, M. K., Kimura, T., Sakka, K., and Ohmiya, K. (2001). The multidomain xylanase Xyn10B as a cellulose-binding protein in Clostridium stercorarium. FEMS Microbiol. Let. 198, 79–83. Antranikian, G. (1990). Physiology and enzymology of thermophilic anaerobic bacteria degrading starch. FEMS Microbiol. Rev. 75, 201–218. Arai, T., Ohara, H., Karita, S., Kimura, T., Sakka, K., and Ohmiya, K. (2001). Sequence of celQ and properties of CelQ, a component of the Clostridium thermocellum cellulosome. Appl. Microbiol. Biotechnol. 57, 660–666. Aspinall, G. O. (1980). Chemistry of cell wall polysaccharides. In ‘‘The Biochemistry of Plants’’ (N. E. Tolbert, ed.), pp. 473–500. Academic Press, New York. Aurilia, V., Martin, J. C., McCrae, S. I., Scot, K. P., Rincon, M. T., and Flint, H. J. (2000). Three multidomain esterases from the cellulolytic rumen anaerobe Ruminococcus ﬂavefaciens 17 that carry divergent dockerin sequences. Microbiology 146, 1391–1397. Barr, B. K., Hsieh, Y. L., Ganem, B., and Wilson, D. B. (1996). Identiﬁcation of two functionally different classes of exocellulases. Biochemistry 35, 586–592. Bayer, E. A., and Lamed, R. (1986). Ultrastructure of the cell surface cellulosome of Clostridium thermocellum and its interaction with cellulose. J. Bacteriol. 167, 828–836. Bayer, E. A., Chanzy, H., Lamed, R., and Shoham, Y. (1998). Cellulose, cellulases and cellulosomes. Curr. Opin. Struct. Biol. 8, 548–557. Bayer, E. A., Shoham, Y., and Lamed, R. (2000). Cellulose-decomposing prokaryotes and their enzyme systems. In ‘‘The Prokaryotes: An Evolving Electronic Resource for the

EXTRACELLULAR CLOSTRIDIUM ENZYMES

253

Microbiological Community’’ (M. Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer, and E. Stackebrandt, eds.), 3rd ed. Springer-Verlag, New York. Be´guin, P., Cornet, P., and Aubert, J. P. (1985). Sequence of a cellulase gene of the thermophilic bacterium Clostridium thermocellum. J. Bacteriol. 162, 102–105. Be´laich, J. P., Tardif, C., Be´laich, A., and Gaudin, C. (1997). The cellulolytic system of Clostridium cellulolyticum. J. Biotechnol. 57, 3–14. Be´laich, A., Parsiegla, G., Gal, L., Villard, C., Haser, R., and Belaich, J. P. (2002). Cel9M, a new family 9 cellulase of the Clostridium cellulolyticum cellulosome. J. Bacteriol. 184, 1378–1384. Berenger, J.-F., Frixon, C., Bigliardi, J., and Creuzet, N. (1985). Production, puriﬁcation and properties of thermostable xylanase from Clostridium stercorarium. Can. J. Microbiol. 31, 635–643. Berger, E., Jones, W. A., Jones, D. T., and Woods, D. R. (1990). Sequencing and expression of a cellodextrinase (ced1) gene from Butyrivibrio ﬁbrisolvens H17c cloned in Escherichia coli. Mol. Gen. Genet. 223, 310–318. Bhat, S., Godenough, P. W., Bhat, M. K., and Owen, E. (1994). Isolation of four major subunits from Clostridium thermocellum cellulosome and their synergism in the hydrolysis of crystalline cellulose. Int. J. Biolog. Macromolec. 16, 335–342. Blum, D. L., Kataeva, I. A., Li, X. L., and Ljungdahl, L. (2000). Feruloyl esterase activity of the Clostridium thermocellum cellulosome can be attributed to previously unknown domains of XynY and XynZ. J. Bacteriol. 182, 1346–1351. Bolam, D. N., Ciruela, A., McQueen-Mason, S., Simpson, P., Williamson, M. P., Rixon, J. E., Boraston, A., Hazlewood, G. P., and Gilbert, H. J. (1998). Pseudomonas cellulose-binding domains mediate their effects by increasing enzyme substrate proximity. Biochem. J. 331, 775–781. Brechtel, E., and Bahl, H. (1999). In Thermoanaerobacterium thermosulfurigenes EM1 S-layer homology domains do not attach to peptidoglycan. J. Bacteriol. 181, 5017–5023. Brechtel, E., Matuschek, M., Hellberg, A., Egelseer, E. M., Schmid, R., and Bahl, H. (1999). Cell wall of Thermoanaerobacterium thermosulfurigenes EM1: Isolation of its components and attachment of the xylanase XynA. Arch. Microbiol. 171, 159–165. Bronnenmeier, K., Ebenbichler, C., and Staudenbauer, W. L. (1990). Separation of the cellulolytic and xylanolytic enzymes of Clostridium stercorarium. J. Chromatogr. 521, 301–310. Bronnenmeier, K., Ru¨cknagel, K. P., and Staudenbauer, W. L. (1991). Puriﬁcation and properties of a novel type of exo-1,4--glucanase (Avicelase II) from the cellulolytic thermophile Clostridium stercorarium. Eur. J. Biochem. 200, 379–385. Bronnenmeier, K., Meissner, H., Stocker, S., and Staudenbauer, W. L. (1995). -DGlucuronidases from the xylanolytic thermophiles Clostridium stercorarium and Thermoanaerobacterium saccharolyticum. Microbiol. 141, 2033–2040. Bronnenmeier, K., Adelsberger, H., Lottspeich, F., and Staudenbauer, W. L. (1996). Afﬁnity puriﬁcation of cellulose-binding enzymes of Clostridium stercorarium. Bioseparation 6, 41–45. Bronnenmeier, K., Kundt, K., Riedel, K., Schwarz, W. H., and Staudenbauer, W. L. (1997). Structure of the Clostridium stercorarium gene celY encoding the exo-1,4--glucanase Avicelase II. Microbiol. 143, 891–898. Carrard, G., Koivula, A., So¨derlund, H., and Beguin, P. (2000). Cellulose binding domains promote hydrolysis of different sites on crystalline cellulose. PNAS 97, 10342–10347.

254

SCHWARZ et al.

Ciruela, A., Gilbert, H. J., Ali, B. R. S., and Hazlewood, G. P. (1998). Synergistic interaction of the cellulosome integrating protein (CipA) from Clostridium thermocellum with a cellulosomal endoglucanase. FEBS Lett. 422, 221–224. Coutinho, P. M., and Henrissat, B. (1999a). Carbohydrate-Active Enzymes server at URL:http://afmb.cnrs-mrs.fr/cazy/CAZY/index.html. Coutinho, P. M., and Henrissat, B. (1999b). The modular structure of cellulases and other carbohydrate-active enzymes: An integrated database approach. In ‘‘Genetics, Biochemistry and Ecology of Cellulose Degradation’’ (K. Ohmiya, K. Hayashi, K. Sakka, Y. Kobayashi, S. Karita, and T. Kimura, eds.), pp. 15–23. Uni Publishers Co., Tokyo. Davies, G., and Henrissat, B. (1995). Structures and mechanisms of glycosyl hydrolases. Current Biol. 3, 853–859. Ding, S. Y., Bayer, E. A., Steiner, D., Shoham, Y., and Lamed, R. (1999). A novel cellulosomal scaffoldin from Acetivibrio cellulolyticus that contains a family 9 glycosyl hydrolase. J. Bacteriol. 181, 6720–6729. Doi, R. H., Kosugi, A., Murashima, K., Tamaru, Y., and Han, S. O. (2003). Cellulosomes from mesophilic bacteria. J. Bacteriol. 185, 5907–5914. Donaghy, J. A., Bronnenmeier, K., Soto-Kelly, P. F., and McKay, A. M. (2000). Puriﬁcation and characterization of an extracellular feruloyl esterase from the thermophilic anaerobe Clostridium stercorarium. J. Appl. Microbiol. 88, 458–466. Du¨rre, P. (1998). New insights and novel developments in clostridial acetone/butanol/ isopropanol fermentation. Appl. Microbiol. Biotechnol. 49, 639–648. Engelhardt, H., and Peters, J. (1998). Structural research on surface layers: A focus on stability, surface layer homology domains, and surface layer-cell wall interactions. J. Struct. Biol. 124, 276–302. Fields, M. W., Mallik, S., and Russell, J. B. (2000). Fibrobacter succinogenes S85 ferments ball-milled cellulose as fast as cellobiose until cellulose surface area is limiting. Appl. Microbiol. Biotechnol. 54, 570–574. Fierobe, H. P., Mechaly, A., Tardif, C., Belaich, A., Lamed, R., Shoham, Y., Belaich, J. P., and Bayer, E. A. (2001). Design and production of active cellulosome chimeras. J. Biol. Chem. 276, 21257–21261. Fontes, C. M. G. A., Hazelwood, G. P., Morag, E., Hall, J., Hirst, B. H., and Gilbert, H. J. (1995). Evidence for a general role for non-catalytic thermostabilizing domains in xylanases from thermophilic bacteria. Biochem. J. 307, 151–158. Fuchs, K. P., Zverlov, V. V., Velikodvorskaya, G. A., Lottspeich, F., and Schwarz, W. H. (2003). Lic16A of Clostridium thermocellum, a non-cellulosomal, highly complex endo--1,3-glucanase bound to the outer cell surface. Microbiology 149, 1021–1031. Fujino, T., Be´guin, P., and Aubert, J. P. (1992). Cloning of a Clostridium thermocellum DNA fragment encoding polypeptides that bind the catalytic components of the cellulosome. FEMS Microbiol. Lett. 94, 165–170. Fujino, T., Be´guin, P., and Aubert, J. P. (1993). Organization of a Clostridium thermocellum gene cluster encoding the cellulosomal scaffolding protein CipA and a protein possibly involved in attachment of the cellulosome to the cell surface. J. Bacteriol. 175, 1891–1899. Ganghofner, D., Kellermann, J., Staudenbauer, W. L., and Bronnenmeier, K. (1998). Puriﬁcation and properties of an amylopullulanase, a glucoamylase, and an glucosidase in the amylolytic enzyme system of Thermoanaerobacterium thermosaccharolyticum. Biosci. Biotechnol. Biochem. 62, 302–308. Gapes, J. R. (2000). The economics of acetone-butanol fermentation: Theoretical and market considerations. J. Mol. Microbiol. Biotechnol. 2, 27–32.

EXTRACELLULAR CLOSTRIDIUM ENZYMES

255

Gerischer, U., and Du¨rre, P. (1988). Cloning, sequencing, and molecular analysis of the acetoacetate decarboylase gene region from Clostridium acetobutylicum. J. Bacteriol. 172, 6907–6918. Gerngross, U. T., Romaniec, M. P. M., Kobayashi, T., Huskisson, N. S., and Demain, A. L. (1993). Sequencing of a Clostridium thermocellum gene (cipA) encoding the cellulosomal SL-protein reveals an unusual degree of internal homology. Molec. Microbiol. 8, 325–334. Grepinet, O., and Beguin, P. (1986). Sequence of the cellulase gene of Clostridium thermocellum coding for endoglucanase B. Nucleic Acids Res. 14, 1791–1799. Grepinet, O., Chebrou, M.-C., and Beguin, P. (1988). Nucleotide sequence and deletion analysis of the xylanase gene (xynZ) of Clostridium thermocellum. J. Bacteriol. 170, 4582–4588. Guglielmi, G., and Be´guin, P. (1998). Cellulase and hemicellulase genes of Clostridium thermocellum from ﬁve independent collections contain few overlaps and are widely scattered across the chromosome. FEMS Microbiol. Lett. 161, 209–215. Halstead, J. R., Vercoe, P., Gilbert, H. J., Hazlewood, G. P., and Davidson, K. (1999). A family 26 mannanase produced by Clostridium thermocellum as a component of the cellulosome contains a domain which is conserved in mannanases from anaerobic fungi. Microbiology 145, 3101–3108. Hayashi, H., Takagi, K., Fukumura, M., Kimura, T., Karita, S., Sakka, K., and Ohmiya, K. (1997). Sequence of xynC and properties of XynC, a major component of the Clostridium thermocellum cellulosome. J. Bacteriol. 179, 4246–4253. Hayashi, H., Takehara, M., Hattori, T., Kimura, T., Karita, S., Sakka, K., and Ohmiya, K. (1999). Nucleotide sequences of two contiguous and highly homologous xylanase genes xynA and xynB and characterization of XynA from Clostridium thermocellum. Appl. Microbiol. Biotechnol. 51, 348–357. Hazlewood, G. P., Davidson, K., Clarke, J. H., Durrant, A. J., Hall, J., and Gilbert, H. J. (1990). Endoglucanase-E, produced at high-level in Escherichia coli as a lacZ0 fusion protein, is part of the Clostridium thermocellum cellulosome. Enz. Microb. Technol. 12, 656–662. Hazlewood, G. P., Davidson, K., Laurie, J. I., Huskisson, N. S., and Gilbert, H. J. (1993). Gene sequence and properties of Cell, a family E endoglucanase from Clostridium thermocellum. J. Gen. Microbiol. 139, 307–316. He, S. Y., and Collmer, A. (1990). Molecular cloning, nucleotide sequence, and marker exchange mutagenesis of the exo-poly--D-galacturonidase-encoding pehX gene of Erwinia chrysanthemi EC16. J. Bacteriol. 172, 4988–4995. Henrissat, B. (1991). A classiﬁcation of glycosyl hydrolases based on amino-acid sequence similarities. Biochem. J. 280, 309–316. Ilk, N., Kosma, P., Puchberger, M., Egelseer, E. M., Mayer, H. F., Sleytr, U. B., and Sa´ra, M. (1999). Structural and functional analyses of the secondary cell wall polymer of Bacillus sphaericus CCM 2177 that serves as an S-layer anchor. J. Bbacteriol. 181, 7643–7646. Jauris, S., Ru¨cknagel, K. P., Schwarz, W. H., Kratzsch, P., Bronnenmeier, K., and Staudenbauer, W. L. (1990). Sequence analysis of the Clostridium stercorarium celZ gene encoding a thermoactive cellulase (Avicelase I): Identiﬁcation of catalytic and cellulose-binding domains. Mol. Gen. Genet. 223, 258–267. Jervis, E. J., Haynes, C. A., and Kilburn, D. G. (1997). Surface diffusion of cellulases and their isolated binding domains on cellulose. J. Biol. Chem. 272, 24016–24023. Joliff, G., Beguin, P., and Aubert, J. P. (1986). Nucleotide sequence of the cellulase gene celD encoding endoglucanase D of Clostridium thermocellum. Nucleic Acids Res. 14, 8605–8613.

256

SCHWARZ et al.

Kakiuchi, M., Isui, A., Suzuki, K., Fujino, T., Fujino, E., Kimura, T., Karita, S., Sakka, K., and Ohmiya, K. (1998). Cloning and DNA sequencing of the genes encoding Clostridium josui scaffolding protein CipA and cellulase CelD and identiﬁcation of their gene products as major components of the cellulosome. J. Bacteriol. 180, 4303–4308. Kataeva, I. A., Seidel III, R. D., Shah, A., West, L. T., Li, X.-L., and Ljungdahl, L. G. (2002). The Fn3-like repeat from CbhA of Clostridium thermocellum cellulosome promotes hydrolysis of cellulose by modifying its surface. Appl. Environ. Microbiol. 68, 4292–4300. Kim, H., Jung, K. H., and Pack, M. Y. (2000). Molecular characterization of xynX, a gene encoding a multidomain xylanase with a thermostabilizing domain from Clostridium thermocellum. Appl. Microbiol. Biotechnol. 54, 521–527. Kitamoto, N., Yamagata, H., Kato, T., Tsukagoshi, N., and Udaka, S. (1988). Cloning and sequencing of the gene encoding thermophilic beta-amylase of Clostridium thermosulfurogenes. J. Bacteriol. 170, 5848–5854. Kobayashi, T., Romaniec, M. P. M., Barker, P. J., Gerngross, U. T., and Demain, A. L. (1993). Nucleotide sequence of gene celM encoding a new endoglucanase (CelM) of Clostridium thermocellum and puriﬁcation of the enzyme. J. Ferment. Bioeng. 76, 251–256. Kurokawa, J., Hemjinda, E., Arai, T., Karita, S., Kimura, T., Sakka, K., and Ohmiya, K. (2001). Sequence of the Clostridium thermocellum mannanase gene man26B and characterization of the translated product. Biosci. Biotechnol. Biochem. 65, 548–554. Kurokawa, J., Hemjinda, E., Arai, T., Kimura, T., Sakka, K., and Ohmiya, K. (2002). Clostridium thermocellum cellulase CelT, a family 9 endoglucanase without an Iglike domain or family 3c carbohydrate-binding module. Appl. Microbiol. Biotechnol. 59, 455–461. Lamed, R., Setter, E., and Bayer, E. A. (1983). Characterization of a cellulose-binding cellulase-containing complex in Clostridium thermocellum. J. Bacteriol. 156, 828–836. Lamed, R., Naimark, J., Morgenstern, E., and Bayer, E. A. (1987). Specialized surface structure in cellulolytic bacteria. J. Bacteriol. 169, 3792–3800. Lamed, R., Morag, E., Moryosef, O., and Bayer, E. A. (1991). Cellulosome-like entities in Bacteroides cellulosolvens. Curr. Microbiol. 22, 27–34. Lee, I., Evans, B. R., and Woodward, J. (2000). The mechanism of cellulase action on cotton ﬁbers: Evidence from atomic force microscopy. Ultramicroscopy 82, 213–221. Lee, Y. E., Jain, M. K., Lee, C., Lowe, S. E., and Zeikus, J. G. (1993). Taxonomic distinction of saccharolytic thermophilic anaerobes: Description of Thermoanaerobacterium xylanolyticum gen. nov., sp. nov., and Thermoanaerobacterium saccharolyticum gen. nov., sp. nov.; reclassiﬁcation of Thermoanaerobium brockii, Clostridium thermosulfurogenes, and Clostridium thermohydrosulfuricum E100-69 as Thermoanaerobacter brockii comb. nov., Thermoanaerobacterium thermosulfurigenes comb. nov., and Thermoanaerobacter thermohydrosulfuricus comb. nov., respectively; and transfer of Clostridium thermohydrosulfuricum 39E to Thermoanaerobacter ethanolicus. Int. J. Syst. Bacteriol. 43, 41–51. Leibovitz, E., and Be´guin, P. (1996). A new type of cohesin domain that speciﬁcally binds the dockerin domain of the Clostridium thermocellum cellulosome-integrating protein CipA. J. Bacteriol. 178, 3077–3084. Lemaire, M., and Beguin, P. (1993). Nucleotide sequence of the celG gene of Clostridium thermocellum and characterization of its product, endoglucanase CelG. J. Bacteriol. 175, 3353–3360.

EXTRACELLULAR CLOSTRIDIUM ENZYMES

257

Lemaire, M., Ohayon, H., Gounon, P., Fujino, T., and Be´guin, P. (1995). OlpB, a new outer layer protein of Clostridium thermocellum, and binding of its S-layer-like domains to components of the cell envelope. J. Bacteriol. 177, 2451–2459. Linder, M., and Teeri, T. T. (1997). The roles and function of cellulose-binding domains. J. Biotechnol. 57, 15–28. Little, E., Bork, P., and Doolittle, R. F. (1994). Tracing the spread of ﬁbronectin type III domains in bacterial glycohydrolases. J. Mol. Evol. 39, 631–643. Ljungdahl, L. G., and Eriksson, K. E. (1985). Ecology of microbial cellulose degradation. Adv. Microb. Ecol. 8, 237–299. Lupas, A., Engelhardt, H., Peters, J., Santarius, U., Volker, S., and Baumeister, W. (1994). Domain structure of the Acetogenium kivui surface layer revealed by electron crystallography and sequence analysis. J. Bacteriol. 176, 1224–1233. Lynd, L., Weimer, P. J., van Zyl, W. H., and Pretorius, I. S. (2002). Microbial cellulose utilization: Fundamentals and biotechnology. Microbiol. Molec. Biol. Rev. 66, 506–577. Lytle, B. L., Volkman, B. F., Westler, W. M., and Wu, J. H. D. (2000). Secondary structure and calcium-induced folding of the Clostridium thermocellum dockerin domain determined by NMR spectroscopy. Arch. Biochem. Biophys. 379, 237–244. Madkour, M., and Mayer, F. (2003). The structural organization of the intact bacterial cellulosome as revealed by electron microscopy. Cell Biol. Int. 27, 831–836. Mathupala, S. P., Saha, B. C., and Zeikus, J. G. (1990). Substrate competition and speciﬁcity at the active site of amylopullulanase from Clostridium thermohydrosulfuricum. Biochem. Biophys. Res. Commun. 166, 126–132. Matuschek, M., Burchhardt, G., Sahm, K., and Bahl, H. (1994). Pullulanase of Thermoanaerobacterium thermosulfurigenes EM1 (Clostridium thermosulfurogenes): Molecular analysis of the gene, composite structure of the enzyme and a common model for its attachment to the cell surface. J. Bacteriol. 176, 3295–3302. Matuschek, M., Sahm, K., Zibat, A., and Bahl, H. (1996). Characterization of genes from Thermoanaerobacterium thermosulfurigenes EM1 that encode two glycosyl hydrolases with conserved S-layer-like domains. Mol. Gen. Genet. 252, 493–496. Mayer, F., Coughlan, M. P., Mori, Y., and Ljungdahl, L. G. (1987). Macromolecular organization of the cellulolytic enzyme complex of Clostridium thermocellum as revealed by electron microscopy. Appl. Environ. Microbiol. 53, 2785–2792. Mechaly, A., Yaron, S., Lamed, R., Fierobe, H. P., Belaich, A., Belaich, J. P., Shoham, Y., and Bayer, E. A. (2000). Cohesin-dockerin recognition in cellulosome assembly: Experiment versus hypothesis. Proteins Struct. Func. Gen. 39, 170–177. Melasniemi, H., Paloheimo, M., and Hemio¨, L. (1990). Nucleotide sequence of the -amylase-pullulanase gene from Clostridium thermohydrosulfuricum. J. Gen. Microbiol. 136, 447–454. Meissner, K., Wassenberg, D., and Liebl, W. (2000). The thermostabilizing domain of the modular xylanase XynA of Thermotoga maritima represents a novel type of binding domain with afﬁnity for soluble xylan and mixed-linkage -1,3/-1,4-glucan. Molec. Microbiol. 36, 898–912. Mesnage, S., Tosi-Couture, E., and Fourt, A. (1999). Production and cell surface anchoring of functional fusions between the SLH motifs of the Bacillus anthracis S-layer proteins and the Bacillus subtilis levansucrase. Mol. Microbiol. 31, 927–936. Mesnage, S., Fontaine, T., Mignot, T., Delepierre, M., Mock, M., and Fouet, A. (2000). Bacterial SLH domains proteins are non-covalently anchored to the cell surface via a conserved mechanism involving wall polysaccharide pyruvylation. EMBO J. 17, 4473–4484.

258

SCHWARZ et al.

Mitchell, W. J. (2001). Biology and Physiology. In ‘‘Clostridia: Biotechnology and Medical Applications’’ (H. Bahl and P. Du¨rre, eds.). Wiley-VCH, Weinheim. Montoya, D., Are´valo, C., Gonzales, S., Aristizabal, F., and Schwarz, W. H. (2001). New solvent-producing Clostridium sp. strains, hydrolyzing a wide range of polysaccharides, are closely related to Clostridium butyricum. J. Ind. Microbiol. Biotechnol. 27, 329–335. Morag, E., Bayer, E. A., and Lamed, R. (1992). Afﬁnity digestion for the near-total recovery of puriﬁed cellulosome from Clostridium thermocellum. Enzyme Microb. Technol. 14, 289–292. Morag, E., Yaron, S., Lamed, R., Kenig, R., Shoham, Y., and Bayer, E. A. (1996). Dissociation of the cellulosome of Clostridium thermocellum under nondenaturing conditions. J. Biotechnol. 51, 235–242. Mosbah, A., Belaich, A., Bornet, O., Belaich, J. P., Henrissat, B., and Darbon, H. (2000). Solution structure of the module X2 1 of unknown function of the cellulosomal scaffolding protein CipC of Clostridium cellulolyticum. J. Mol. Biol. 304, 201–207. Navarro, A., Chebrou, M. C., Beguin, P., and Aubert, J. P. (1991). Nucleotide sequence of the cellulase gene celF of Clostridium thermocellum. Res. Microbiol. 142, 927–936. No¨lling, J., Breton, G., Omelchenko, M. V., Makarova, K. S., Zeng, Q., Gibson, R., Lee, H. M., Dubois, J., Qiu, D., Hitti, J., Wolf, Y. I., Tatusov, R. L., Sabathe, F., DoucetteStamm, L., Soucaille, P., Daly, M. J., Bennett, G. N., Koonin, E. V., and Smith, D. R. (2001). Genome sequence and comparative analysis of the solvent-producing bacterium Clostridium acetobutylicum. J. Bacteriol. 183, 4823–4838. Ohara, H., Karita, S., Kimura, T., Sakka, K., and Ohmiya, K. (2000). Characterization of the cellulolytic complex (cellulosome) from Ruminococcus albus. Biosci. Biotechnol. Biochem. 64, 254–260. Olabarrı´a, G., Carrascosa, L., de Pedro, and Berenguer, J. (1996). A conserved motif in S-layer proteins is involved in peptidoglycan binding in Thermus thermophilus. J. Bacteriol. 178, 4765–4772. Paquet, V., Croux, C., Goma, G., and Soucaille, P. (1991). Puriﬁcation and characterization of the extracellular alpha-amylase from Clostridium acetobutylicum. Appl. Environ. Microbiol. 57, 212–218. Parsiegla, G., Reverbel-Leroy, C., Tardif, C., Belaich, J. P., Driguez, H., and Haser, R. (2000). Crystal structures of the cellulase Cel48F in complex with inhibitors and substrates give insights into its processive action. Biochemistry 39, 11238–11246. Pohlschro¨der, M., Canale-Parola, E., and Leschine, S. B. (1995). Ultrastructural diversity of the cellulase complexes of Clostridium papyrosolvens C7. J. Bacteriol. 177, 6625–6629. Ramesh, M. V., Podkovyrov, S. M., Lowe, S. E., and Zeikus, J. G. (1994). Cloning and sequencing of the Thermoanaerobacterium saccharolyticum B6A-RI apu gene and puriﬁcation and characterization of the amylopullulanase from Escherichia coli. Appl. Environ. Microbiol. 60, 94–101. Reichenbecher, M., Lottspeich, F., and Bronnenmeier, K. (1997). Puriﬁcation and properties of a cellobiose phosphorylase (CepA) and a cellodextrin phosphorylase (CepB) from the cellulolytic thermophile Clostridium stercorarium. Eur. J. Biochem. 247, 262–267. Reverbel-Leroy, C., Pages, S., Belaich, A., Belaich, J. P., and Tardif, C. (1997). The processive endocellulase CelF, a major component of the Clostridium cellulolyticum cellulosome: Puriﬁcation and characterization of the recombinant form. J. Bacteriol. 179, 46–52.

EXTRACELLULAR CLOSTRIDIUM ENZYMES

259

Riedel, K., Ritter, J., and Bronnenmeier, K. (1997). Synergistic interaction of the Clostridium stercorarium cellulases Avicelase I (CelZ) and Avicelase II (CelY) in the degradation of microcrystalline cellulose. FEMS Microbiol. Lett. 147, 239–243. Riedel, K., Ritter, J., Bauer, S., and Bronnenmeier, K. (1998a). The modular cellulase CelZ of the thermophilic bacterium Clostridium stercorarium contains a thermostabilizing domain. FEMS Microbiol. Lett. 164, 261–267. Riedel, K., and Bronnenmeier, K. (1998b). Intramolecular synergism in an engineered exoendo-1,4--glucanase fusion protein. Molec. Microbiol. 28, 767–775. Ries, W., Hotzy, C., Schocher, I., Sleytr, U. B., and Sara´, M. (1997). Evidence that the N-terminal part of the S-layer protein from Bacillus stearothermophilus PV72/p2 recognizes a secondary cell wall polymer. J. Bacteriol. 179, 3892–3898. Sabathe, F., Belaich, A., and Soucaille, P. (2002). Characterization of the cellulolytic complex (cellulosome) of Clostridium acetobutylicum. FEMS Microbiol. Lett. 19, 15–22. Sabathe, F., and Soucaille, P. (2003). Characterization of the CipA scaffolding protein and in vivo production of a minicellulosome in Clostridium acetobutylicum. J. Bacteriol. 185, 1092–1096. Sakka, K., Fukumura, M., Tanaka, A., and Ohmiya, K. (1996). Characterization of thermostability of Clostridium stercorarium xylanase. In ‘‘Enzyme Engineering XIII’’ vol. 799, pp. 341–345. Annals of the New York Academy of Sciences, New York. Sakon, J., Irwin, D., Wilson, D. B., and Karplus, P. A. (1997). Structure and mechanism of endo/exocellulase E4 from Thermomonopsora fusca. Nat. Struct. Biol. 4, 810–818. Sa´ra, M., Kuen, B., Mayer, H. F., Mandl, F., Schuster, K. C., and Sleytr, U. B. (1996). Dynamics in oxygen-induced changes in S-layer protein synthesis from Bacillus stearothermophilus PV72 and the S-layer-deﬁcient variant T5 in continuous culture and studies of the cell wall composition. J. Bacteriol. 178, 2108–2117. Schwarz, W. H. (2001). The cellulosome and cellulose degradation by anaerobic bacteria. Appl. Microbiol. Biotechnol. 56, 634–649. Schwarz, W. H. (2003). http://www.wzw.tum.de/mbiotec/cellmo.htm Schwarz, W. H., Schimming, S., Ru¨cknagel, K. P., Burgschwaiger, S., Kreil, G., and Staudenbauer, W. L. (1988). Nucleotide sequence of the celC gene encoding endoglucanase C of Costridium thermocellum. Gene 63, 23–30. Schwarz, W. H., Jauris, S., Kouba, M., Bronnenmeier, K., and Staudenbauer, W. L. (1989). Cloning and expression of Clostridium stercorarium cellulase genes in Escherichia coli. Biotechnol. Lett. 11, 461–466. Shimon, L. J. W., Bayer, E. A., Morag, E., Lamed, R., Yaron, S., Shoham, Y., and Frolow, F. (1997). A cohesin domain from Clostridium thermocellum: The crystal structure provides new insights into cellulosome assembly. Structure 15, 381–390. Specka, U., Mayer, F., and Antranikian, G. (1991). Puriﬁcation and properties of a thermoactive glucoamylase from Clostridium thermosaccharolyticum. Appl. Environ. Microbiology 57, 2317–2323. Spinnler, H. E., Lavigne, B., and Blachere, H. (1986). Pectinolytic activity of Clostridium thermocellum: Its use for anaerobic fermentation of sugar beet pulp. Appl. Microbiol. Biotechnology 23, 434–437. Spreinat, A., and Antranikian, G. (1990). Puriﬁcation and properties of a thermostable pullulanase from Clostridium thermosulfurogenes EM1 which hydrolyzes both -1,6 and -1,4-glycosidic linkages. Appl. Microbiol. Biotechnology 33, 511–518. Spreinat, A., and Antranikian, G. (1992). Analysis of the amylolytic enzyme system of Clostridium thermosulfurigenes EM1: Puriﬁcation and synergistic action of pullulanases and maltohexaose forming -amylase. Starch 44, 305–312.

260

SCHWARZ et al.

Sun, J. L., Sakka, K., Karita, S., Kimura, T., and Ohmiya, K. (1998). Adsorption of Clostridium stercorarium xylanase A to insoluble xylan and the importance of the CBDs to xylan hydrolysis. J. Ferment. Bioeng. 85, 63–68. Tamaru, Y., Karita, S., Ibrahim, A., Chan, H., and Doi, R. H. (2000). A large gene cluster for the Clostridium cellulovorans cellulosome. J. Bacteriol. 182, 5906–5910. Tokatlidis, K., Salamitou, S., Be´guin, P., Dhurjati, P., and Aubert, J. P. (1991). Interaction of the duplicated segment carried by Clostridium thermocellum cellulases with cellulosome components. FEBS Lett. 291, 185–188. Tormo, J., Lamed, R., Chirino, A. J., Morag, E., Bayer, E. A., Shoham, Y., and Steitz, T. A. (1996). Crystal structure of a bacterial family-III cellulose-binding domain: A general mechanism for attachment to cellulose. EMBO J. 15, 5739–5751. Tomme, P., Creagh, A. L., Kilburn, D. G., and Haynes, C. A. (1996). Interaction of polysaccharides with the N-terminal cellulose-binding domain of Cellulomonas ﬁmi CenC. 1. Binding speciﬁcity and calorimetric analysis. Biochemistry 35, 13885–13894. Wang, W. K., Kruus, K., and Wu, J. H. D. (1993). Cloning and the DNA sequence of the gene coding for the Clostridium thermocellum cellulase Ss (CelS), a major cellulosome component. J. Bacteriol. 175, 1293–1302. Watanabe, T., Ito, Y., Yamada, T., Hashimoto, M., Sekine, S., and Tanaka, H. (1994). The roles of the C-terminal domain and type III omains of chitinase A1 from Bacillus circulans WL-12 in chitin degradation. J. Bacteriol. 176, 4465–4472. Wind, R. D., Liebl, W., Buitelaar, R. M., Penninga, D., Spreinat, A., Dijkhuizen, L., and Bahl, H. (1995). Cyclodextrin formation of the thermostable -amylase of Thermoanaerobacterium thermosulfurigenes EM1 and reclassiﬁcation of the enzyme as a cyclodextrin glycosyltransferase. Appl. Environ. Microbiol. 61, 1257–1265. Wu, J. H. D., Orme-Johnson, W. H., and Demain, A. L. (1987). Two components of an extracellular protein aggregate of Clostridium thermocellum together degrade crystalline cellulose. Biochem. 27, 1703–1709. Yaguee, E., Beguin, P., and Aubert, J.-P. P. (1990). Nucleotide sequence and deletion analysis of the cellulase-encoding gene celH of Clostridium thermocellum. Gene 89, 61–67. Zverlov, V. V., Fuchs, K. P., Schwarz, W. H., and Velikodvorskaya, G. A. (1994). Puriﬁcation and cellulosomal localization of Clostridium thermocellum mixed linkage glucanase LicB (1,3-1,4--D-glucanase). Biotechnol. Lett. 16, 29–34. Zverlov, V., Mahr, S., Riedel, K., and Bronnenmeier, K. (1998a). Properties and gene structure of a bifunctional cellulolytic enzyme (CelA) from the extreme thermophile ‘‘Anaerocellum thermophilum’’ with separate glycosyl hydrolase family 9 and 48 catalytic domains. Microbiology 144, 457–465. Zverlov, V. V., Velikodvorskaya, G. A., Schwarz, W. H., Bronnenmeier, K., Kellermann, J., and Staudenbauer, W. L. (1998b). Multidomain structure and cellulosomal localization of the Clostridium thermocellum cellobiohydrolase CbhA. J. Bacteriol. 180, 3091–3099. Zverlov, V. V., Velikodvorskaya, G. A., and Schwarz, W. H. (1999). Duplicated Clostridium thermocellum cellobiohydrolase gene encoding cellulosomal subunits S3 and S5. Appl. Microbiol. Biotechnol. 51, 852–859. Zverlov, V. V., Fuchs, K. P., and Schwarz, W. H. (2002a). Chi18A, the endochitinase in the cellulosome of the thermophilic, cellulolytic bacterium Clostridium thermocellum. Appl. Environ. Microbiol. 68, 3176–3179. Zverlov, V. V., Velikodvorskaya, G. A., and Schwarz, W. H. (2002b). A newly described cellulosomal cellobiohydrolase, CelO, from Clostridium thermocellum: Investigation

EXTRACELLULAR CLOSTRIDIUM ENZYMES

261

of the exo-mode of hydrolysis, and binding capacity to crystalline cellulose. Microbiology 148, 247–255. Zverlov, V. V., Velikodvorskaya, G. A., and Schwarz, W. H. (2003). Two new cellulosome components encoded downstream of cell in the genome of Clostridium thermocellum: The non-processive endoglucanase CelN and the possibly structural protein CseP. Microbiology 149, 515–524. Zverlov et al., in preparation. Genome sequence available at URL http://genome.ornl.gov/ microbial/cthe/

This Page Intentionally Left Blank

Kernel Knowledge: Smut of Corn MARI´A D. GARCI´A-PEDRAJAS

AND

SCOTT E. GOLD*

Department of Plant Pathology University of Georgia Athens, Georgia 30602 *Author for correspondence. E-mail: [email protected]

I. Introduction II. The Fungal Saprophyte III. The Fungal Pathogen A. The ABCs of Fungal Sex (No C Involved) B. Host–Pathogen Interaction C. In Planta Gene Expression IV. The Host Reaction V. Conclusions References

263 263 267 267 271 281 282 284 285

I. Introduction The earliest recordings of Ustilago maydis, the causal agent of corn smut, date back to the 1750s and 1760s (Christensen, 1963). Yield loss caused by corn smut is generally kept below 2% because of available partially resistant varieties (Shutleff, 1980). However, if one considers that maize is the most important economic crop in the United States, generating $18 billion annually, with approximately 80 million acres planted (2001 World of Corn, National Corn Growers Association (http://www.ncga.com/03 world/main/index.html), even small yield decreases take on major signiﬁcance. Even a 1% loss (an underestimate of the damage caused by this fungus) represents nearly $200 million annually. In addition to domestic consumption, corn is a major proﬁtable U.S. export. U. maydis is a commonly studied model fungus for mating type, mophogenesis, and pathogenicity. In this chapter we attempt to paint a picture of the biology of this fungus that is now fairly well documented and to provide a brief outline of the little information available on the host response to infection. II. The Fungal Saprophyte U. maydis is a heterothallic fungus with a tetrapolar mating system and a dimorphic life cycle consisting of a saprophytic asexual phase and a parasitic phase that generates sexual spores that undergo meiosis upon germination (Fig. 1). Phytopathogenic fungi vary in the strategies 263 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

264

GARCI´A-PEDRAJAS AND GOLD

FIG. 1. Disease cycle of corn smut, caused by Ustilago maydis. Reproduced from (Agrios, 1997) with the permission of Academic Press.

to survive between rounds of plant infection. U. maydis relies on the production of teliospores, resistant diploid cells with thick cell walls produced in mature galls. Teliospores of U. maydis are believed to overwinter in the soil and plant residues (Christensen, 1963). Fischer (1936) showed that teliospores can be stored for 2 years without loss of viability. However, little is known about the survival of teliospores in soil under varying environmental conditions. In U. maydis, germination of teliospores is considered the ﬁrst step of infection. Meiosis occurs in germinating teliospores (Christensen, 1963; ODonnell and McLaughlin, 1984). As the teliospore germinates, the nucleus migrates to the center of the developing promycelium and the ﬁrst meiotic division occurs in the midregion of the promycelium (Christensen, 1963; ODonnell and McLaughlin, 1984). The resultant daughter nuclei divide again to produce four meiotic products. One of these nuclei generally remains in the teliospore, and the other three are typically observed in

KERNEL KNOWLEDGE: SMUT OF CORN

265

the promycelium (Christensen, 1963; ODonnell and McLaughlin, 1984). Septa are formed between these nuclei. In general, the behavior of the nucleus and spindle pole body during meiosis is comparable to that of other basidiomycetes (ODonnell and McLaughlin, 1984). Typically, although not always, in the four-celled promycelium, each cell gives rise to a single primary sporidium. While the description of teliospore germination and sporidial formation is classical, Christensen (1963) stated that it is not the most frequent. A single promycelium may give rise to ﬁve to seven primary sporidia and some of these may be diploid (Christensen, 1963). Fibrillar outgrowths on sporidia have been termed ﬁmbriae and may aid in the attachment to host tissues (Snetselaar and Mims, 1993). It should be noted that, unlike many other basidiomycetes, the sporidia (basidiospores) of U. maydis do not develop on sterigmata and are not forcibly discharged. Rather, the primary sporidia develop on the surface of promycelia (Alexopoulus et al., 1996). Meiosis in the promycelium leads to the production of haploid sporidia that are normally not capable of causing infection singly; they are saprophytic and grow by budding. Haploid budding cells grow readily in laboratory conditions. The role of this growth form in the life and disease cycle of U. maydis in nature, prior to ﬁnding a mating partner and producing the infectious dikaryon, converted to ﬁlamentous growth and obligate biotrophy (Fig. 2), is not well understood. Budding plays important roles in allowing for logarithmic growth rates of the haploid fungus and possibly in dispersal. Indeed, secondary sporidia produced by budding have

FIG. 2. 2b or not 2b, that is the question. The upper portion of the photo shows an a1b1 strain overlaid with an a2b2 strain yielding the white ﬁlamentous dikaryotic growth of the compatible mating reaction. The bottom portion shows an a1b1 strain overlaid with an a2b1 strain yielding yeast growth. These results are indicative of the feature that two different b alleles are essential for the formation of the ﬁlamentous (and pathogenic) dikaryon.

266

GARCI´A-PEDRAJAS AND GOLD

been proposed as the primary infectious agents in nature (Alexopoulus et al., 1996). Like teliospores, sporidia can be transported by wind over hundreds of miles, thus increasing the range over which U. maydis can survive. Sporidia of U. maydis can also be desiccated for 1–5 months without loss in viability, although alternate periods of freezing and thawing are harmful to sporidia (Christensen, 1963). Further, genes speciﬁcally expressed in the budding form may be involved in growth on plant surfaces before mating. Processes such as adhesion of budding cells to plant surfaces could also play important roles in pathogenicity. DNA synthesis during budding growth in U. maydis has been investigated. Using microdensitometry, a technique to measure DNA content in nuclei, Snetselaar and McCann (1997) correlated nuclear density with sporidial cell morphology. DNA synthesis occured in sporidia before bud formation (Snetselaar and McCann, 1997) and production of secondary sporidia. Older cultures of uniformly budding sporidia sometimes produced large numbers of binucleate cells when transferred to fresh culture medium. Septa were formed between these nuclei and the cells bud at both ends. Thus it is important to observe and maintain U. maydis cultures appropriately for morphological observations. Little is known about the genetic regulation of teliospore germination. Caltrider and Gottlieb (1966) showed that the presence of a carbon source in the environment stimulated teliospores to germinate. In recent work, Sacadura and Saville (2003) have started to identify gene expression associated with teliospore germination on a large scale. They sequenced 2,871 ESTs (expressed sequence tags) from a cDNA library prepared from germinating teliospores and performed northern blots on selected cDNA clones. In our laboratory, to gain insight of the role and signiﬁcance of the budding phase of growth in the life and disease cycle of U. maydis in nature, we have used a subtractive technique, suppression subtractive hybridization (SSH), to identify genes up-regulated in this growth form. For this subtractive approach we took advantage of a uac1 mutant strain, which is constitutively ﬁlamentous, to remove sequences common to both growth morphologies. We identiﬁed 37 genes speciﬁcally up-regulated in budding cells. For nine genes, expression could only be detected in budding cells. For genes expressed in both growth forms, levels of differential expression varied from as much as 65-fold to only twofold higher levels in budding cells. Twenty-seven of these genes showed similarity to database sequences and fell into several putative functional categories (Garcı´a-Pedrajas and Gold, 2004).

KERNEL KNOWLEDGE: SMUT OF CORN

267

III. The Fungal Pathogen A. THE ABCS OF FUNGAL SEX (NO C INVOLVED) Genetic analysis of the sexual system has been a major focus of work on Ustilago maydis, and this species has become a paradigm for the higher basidiomycetes (Casselton, 2002; Casselton and Olesnicky, 1998). The heterothallism of U. maydis was recognized as early as 1927 (Stakman and Christensen, 1927). Employing tetrad analysis, Hana (1929) showed conclusively that there were at least two pairs of segregating sex factors because in some cases four distinct mating types were derived in a single tetrad. He designated the diploid nucleus as possessing an AaBb genotype and the progeny as AB, Ab, aB, and ab. Similar notations for the mating type genes are still used. No distinction was made for the function of A or B in Hana’s work. Rowell and DeVay (1954) designated these two mating factors as ‘‘a’’ and ‘‘b’’ and determined that two speciﬁcities of a and multiple speciﬁcities of b existed. Different speciﬁcities were necessary at both the a locus and the b locus to generate productive maize infection in which teliospores (chlamydospores) were generated, and the currently used designations for these mating type loci were established. In this same study it was determined that amphisexual progeny were occasionally generated that had both a speciﬁcities such that they could productively be paired with any strain possessing a different b allele. However, these amphisexual strains were not solopathogenic, whereas diploids heterozygous at both a and b were solopathogens. This indicated to the authors that the a compatibility factor was clearly not a primary pathogenicity characteristic, a fact further corroborated by Banuett and Herskowitz (1989). Rowell clearly demonstrated that the roles of heterozygosity at a and b were in fusion and dikaryon vigor/stability, respectively (Rowell, 1955). He also noted that alleles of both factors had to be different in the mating partners to generate the virulent pathogen. Trueheart and Herskowitz (1992) generated a cytoduction assay in which cell fusion was strictly controlled by possession of unequal alleles at the a locus, while the b locus played no role. Later, Laity et al. (1995) complemented this work, showing that heterozygosity at the b locus within a strain inhibits further mating (Laity et al., 1995). In summary then, a1 strains will fuse with a2 strains regardless of the condition at the b locus except that once b becomes heterozygous the cell will fuse no further with any other strain. Additionally, a1a2bn stains fuse promiscuously with any normal haploid mating type. By addition of charcoal to

268

GARCI´A-PEDRAJAS AND GOLD

solid medium the development of the functional dikaryon can easily be monitored in culture (Day and Anagnosakis, 1971; Holliday, 1974). White ﬁlamentous growth is observed on this medium only when both a and b differ in cospotted compatible strain pairs (Fig. 2). The b locus is multiallelic (Rowell and DeVay, 1954). Puhalla found only 2 a but 18 different b mating type alleles in 62 lines among 33 different isolations from the United States and Canada. He predicted that there should be no more than 25 distinct alleles of b in the population (Puhalla, 1970). The master control genes of mating and pathogenicity, the a and b mating type genes, have now been cloned and characterized (Bolker et al., 1992; Froeliger and Leong, 1991; Kronstad and Leong, 1989, 1990; Schulz et al., 1990). Holliday had noted that the pan1 gene was tightly linked at about 2.5 map units from the a mating type locus (Holliday, 1961a, 1974). Using this information, Froeliger and Leong (1991) cloned the a2 mating type determinant by complementation of a pan1 mutant a1 strain with a cosmid from a prototrophic a2 strain. By mating a1b1/a2 transformant strains with an a1b2 strain to generate a ﬁlamentous and pathogenic dikaryon, the presence of the a2 allele was conﬁrmed. The intial cloning of the a locus indicated that the a1 and a2 allelic sequences were idiomorphs, (i.e., they lacked sequence homology). Homologous ﬂanks were then employed to isolate the a2 mating type idiomorph and similar methods used to conﬁrm function (Froeliger and Leong, 1991). The a mating type genes were then sequenced and Bolker et al. (1992) demonstrated that the mating type speciﬁcity in each idiomorph was determined by two genes. One gene encodes a lipopeptide mating factor and the other a pheromone receptor. Thus the a locus encompasses mfa and pra, two tightly linked genes that encode secreted pheromone and membrane spanning pheromone receptors, respectively (Bolker et al., 1992). The pheromone encoded by the mfa gene is thought to interact directly with the pheromone receptor product encoded by the pra gene of the opposite a mating speciﬁcity (Spellig et al., 1994). Synthetic pheromone causes cell cycle arrest in the G2 phase (Garcia-Muse et al., 2003). This is in contrast with the situation described in ascomycete yeasts such as S. cerevisiae and Schizosaccharomyces pombe, in which pheromone induces cell cycle arrest at G1. The function of the genes at the a locus helps explain the fact that a6¼ (possession of two different allelic speciﬁcities of a) is required for a diploid heterozygous at b to become ﬁlamentous on charcoal mating media (Banuett and Herskowitz, 1989). In addition to its function as a mating attraction system, dikaryon heterozygosity at the a locus (in addition to heterozygosity at b) also

KERNEL KNOWLEDGE: SMUT OF CORN

269

contributes to the in vitro production of the postmating dikaryotic ﬁlamentous form through an autocrine response in which the pheromones and receptors of opposite allelic speciﬁcity are present within the same cell and therefore may continually interact (Banuett and Herskowitz, 1989; Spellig et al., 1994). Downstream events generating the ﬁnal response to pheromone (Fig. 3) appear to involve components similar to those encountered in S. cerevisiae. In this budding yeast, signal transduction from the pheromone-receptor interaction to the ﬁnal cellular responses involves a trimeric G protein and a mitogen-activated protein (MAP) kinase cascade with the ﬁnal phosphorylation and activation of two critical proteins. These proteins are the Ste12p transcription factor, which when activated regulates transcription of target genes, and Far1p, which causes cell cycle arrest by inhibition of the kinase activity of the G1 cyclin complex Cdc28-Cln (Banuett, 1998; Valdivieso et al., 1993). In U. maydis, none of the four cloned G subunits of the trimeric G proteins

FIG. 3. Interaction of the PKA and MAP kinase pathways. Current information indicates that the cAMP and MAP kinase pathways impinge on posttranslational modiﬁcation of the transcription factor Prf1. This is likely not the only target of these kinases, however. Smu1 is a Ste20-like kinase (Smith et al., 2004).

270

GARCI´A-PEDRAJAS AND GOLD

appear to be directly involved in transmission of the pheromone signal (Kruger et al., 1998; Regenfelder et al., 1997). As is the case in Schizosaccharomyces pombe (Sipiczki, 1988), a ras gene (ras2) functions to stimulate ﬁlamentous growth through the pheromone responsive MAP kinase cascade (Lee and Kronstad, 2002). Additional work suggested that the cdc25p homolog Sq11 may function as an activator of Ras2 (Muller et al., 2003). An additional ﬁnding in this work was that activated Ras1, the product of a second ras gene, increased pheromone gene expression. The three members of the pheromone responsive MAP kinase cascade have been identiﬁed. These are ubc4/kpp4 encoding the ste11p MAPKK kinase homolog (Andrews et al., 2000; Mayorga and Gold, 1998; Muller et al., 2003), fuz7/ubc5 encoding the ste7p MAPK kinase homolog (Andrews et al., 2000; Banuett and Herskowitz, 1994; Mayorga and Gold, 1998), and the fus3p and kss1p MAP kinase homolog ubc3/kpp2 (Mayorga and Gold, 1998, 1999; Muller et al., 1999). A putative adaptor protein Ubc2 may link the MAP kinase cascade with the upstream components of signaling through Ras proteins (Mayorga and Gold, 2001). A strong interaction of Ubc2 and the Ubc4 MAPKKK was shown in two-hybrid tests (Klosterman et al., unpublished), conﬁrming previous observed genetic interaction (Mayorga and Gold, 1998). A gene designated prf1 encodes an HMG family transcription factor that links the pheromone response pathway to the expression of the b locus and thus to pathogenicity (Hartmann et al., 1996). The prf1 protein has potential phosphorylation sites for both a MAP kinase (presumably ubc3/kpp2, see below) and for the cyclic AMP dependent protein kinase (Kahmann et al., 1999; Muller et al., 1999). The putative MAP kinase phosphorylation sites appear important for the biological function of the protein (Muller et al., 1999). The prf1 gene is required for pathogenicity because of its essential function in the regulation of the b mating type genes. Constitutive expression of the b genes restores pathogenicity in prf1 mutants (Hartmann et al., 1996). Additional transcription factor(s) are likely involved in transmitting the pheromone responsive MAP kinase and/or cAMP pathway signals besides Prf1. As noted by Lee and Kronstad (2002), epistasis experiments indicated that Ras2 may regulate ﬁlamentation via the pheromone responsive MAP kinase cascade including Ubc3, but not through the activation of Prf1. Additionally, work from our laboratory indicates that the MAP kinase cascade is required for acid-induced ﬁlamentation while prf1 is not (Martinez-Espinoza et al., 2004). The b locus controls events after cell fusion necessary for establishment of the infectious ﬁlamentous dikaryon. lga2, a gene of unknown function (Urban et al., 1996) located within the a2 idiomorph,

KERNEL KNOWLEDGE: SMUT OF CORN

271

is directly and positively regulated by the b-heterodimer (Romeis et al., 2000). Employing inducible promoters to replace those native to b, both positively and negatively b transcriptionally regulated genes are being identiﬁed (Brachmann et al., 2001). However, deletion of a number of these genes did not produce any discernible effect on morphology or pathogenicity, indicating that the ones characterized so far do not individually play a major role in pathogenesis and development. Additionally, genes that, when mutated, induced expression of the b genes in haploid cells as well as other dikaryon speciﬁc genes, have been identiﬁed using another reporter system (Quadbeck-Seeger et al., 2000; Reichmann et al., 2002). Deletion of these genes affect teliosporogenesis and will be discussed in the corresponding section. B. HOST–PATHOGEN INTERACTION Although mating and dikaryon formation can be induced in vitro, Snetselaar and Mims (1992) observed that this process occurs more consistently and rapidly on maize leaves. In the resulting hyphae, infection structures are produced that penetrate plant cells directly through the cell walls. All meristematic plant tissues above ground are susceptible to infection by U. maydis. In early work concerning the mode of entrance of U. maydis into maize, Walter (1934) observed that sporidial germ tubes displayed characteristic curling and swelling prior to penetrating leaf tissue directly through the cell walls. Depressions in the leaf surface were the most common points of penetration. In such areas, hyphae bulged, ﬂattened, and pressed close to the epidermis. Unlike typical appressoria, however, these structures constricted again, grew farther, coiled, and only then produced a penetration peg. Snetselaar and Mims (1992) reported that hyphae, formed after mating of compatible haploid strains on plant surfaces, produced swollen appressorium-like structures from which penetrating hyphae emerged (Fig. 4). These penetrated the cuticle and entered the underlying epidermal cell directly but remained separated from the host cytoplasm by the invaginated host plasma membrane. Although all above ground organs of maize can be infected by U. maydis, in the ﬁeld the most prominent galls are commonly found on maize ears. In a given ear, adjacent smutted and healthy kernels are found, so it was assumed that kernels were infected through stigmas (silks) and that this could be a frequent infection pathway under natural conditions. To assess this hypothesis, Snetselaar and Mims (1993) inoculated stigmas with mixtures of compatible haploid strains or a solopathogenic diploid strain. They conﬁrmed fungal penetration

272

GARCI´A-PEDRAJAS AND GOLD

FIG. 4. Scanning electron micrograph of ﬁxed Ustilago maydis appressorium, 12–14 h after inoculation. Bar ¼ 10 m. Reproduced from Snetselaar and Mims, 1993, with permission.

of stigmas. In most penetration sites on the silks, the same appresorium-like structures previously described on leaf surfaces were encountered. Beneath these structures, penetrating hyphae often entered the stigma between epidermal cells and then turned sharply to enter one of the cells. Cross sections showed that the depressions between the two lobes of maize stigmas were the most frequent points of penetration. Several general conclusions can be drawn regarding early infection events in the U. maydis-maize pathosystem. On both leaves and silks, infection structures are formed almost exclusively on immature epidermal cells that therefore present little mechanical resistance (Snetselaar and Mims, 1993). Maize response to U. maydis penetration is not obvious; necrosis or other dramatic symptoms of plant response are not observed (Snetselaar and Mims, 1992). However, anthocyanin production is a characteristic symptom of inoculated maize seedlings (Banuett and Herskowitz, 1996; Hana, 1929). Formation of swollen appresorium-like structures and their subsequent production of invading hyphae that penetrate epidermal cells appear to be distinct steps in the infection process. A mutant strain defective in Kpp6 activity, a b-regulated MAP kinase, has recently been constructed. This mutant is able to produce appresoria but unable to penetrate plant cells (Brachmann et al., 2003). Microscopic observations of plant surfaces after inoculation with compatible strains,

KERNEL KNOWLEDGE: SMUT OF CORN

273

both carrying an inactivated mutant allele, kpp6T355A,Y357F, showed appressorium formation. However, the majority of those appressoria yielded only short ﬁlaments that failed to penetrate plant cells. Appressorium-like structures are only observed on plant surfaces, suggesting that either contact with the plant surface or signals coming from the host induce their formation. To date, however, little is known about the genetic regulation of the formation of infection structures in U. maydis. Tight regulation of the cAMP pathway is needed for normal disease development. Compatible strains lacking adenylate cyclase activity (uac1), the subunit of the G protein Gpa3 (gpa3), or the catalytic subunit of PKA (adr1) do not produce any symptoms in inoculated plants (Barrett et al., 1993; Durrenberger, 2001; Gold et al., 1994; Regenfelder et al., 1997). These data suggest that PKA activity is required for the initial steps of infection. Interestingly, inactivation of any of these genes in a wild-type haploid background leads to ﬁlamentous growth, indicating that ﬁlamentous morphology is associated with, but not sufﬁcient for, pathogenic behavior. On the other hand, levels of PKA activity above normal do not appear to affect the penetration process. Inoculation with mutant strains with different degrees of PKA activation affect the progression of disease at later stages, most markedly gall formation and teliosporogenesis, and will be discussed later. A thorough review of the role of cAMP in U. maydis and several other plant pathogenic fungi has recently been published (Lee et al., 2003). The deformation of the plant cell wall around hyphae growing from cell to cell indicates mechanical penetration (Snetselaar and Mims, 1993). However, appressorium-like structures visualized in U. maydis are morphologically undifferentiated relative to appressoria formed by other pathogenic fungi, and they are not melanized. It is unclear whether these structures can penetrate the plant cuticle by mechanical forces alone, and thus, other mechanisms, such as production of lytic enzymes, may be involved. Little is known regarding the expression of lytic enzymes in U. maydis nor about their potential role in the infection process. Schauwecker et al. (1995) identiﬁed a gene, egl1, coding for an endo-glucanase, which is not expressed in haploid cells but highly induced in the b-dependent ﬁlamentous form. However, mutants deleted for this gene were not affected in disease development. Cano-Canchola and co-workers (2000) measured pectate lyase, polygalacturonase, cellulase, and xylanase activities in both haploid and solopathogenic strains by using different carbon sources, including plant tissues. They also investigated the induction of these lytic enzymes over time in inoculated plants. Pectate lyase activity was

274

GARCI´A-PEDRAJAS AND GOLD

rapidly induced, reaching a peak of activity approximately 2 days after inoculation, suggesting a potential role in the early steps of infection. Activity for all other enzymes was induced only at later stages of disease development. Haploid strains are able to penetrate host cells, although their ability to progress further in the disease cycle is very limited. Interestingly, pectate lyase activity was induced in inoculations with both haploid and solopathogenic strains (but not in mock infections with water), while all other enzyme activities were induced only during infection with solopathogenic strains. Further studies to establish the role of lytic enzymes in U. maydis pathogenicity are necessary. A gene with high similarity to endo-1,4 -D-xylanases was up-regulated 23-fold in budding haploid cells in comparison with a constitutively ﬁlamentous uac1 mutant. Interestingly, preliminary results showed that this gene was also highly expressed in galls (Garcı´a-Pedrajas and Gold, unpublished). We are further investigating the role of this putative xylanase in pathogenicity in U. maydis. The developmental process from plant inoculation to teliospore maturation combined requires approximately 21 days for completion and can be broken into several phases. Chlorotic spots can be observed as soon as 1 day after inoculation (Banuett and Herskowitz, 1996). Chlorosis becomes extensive within 2 to 3 days (Banuett and Herskowitz, 1996; Snetselaar and Mims, 1992, 1994). In addition, anthocyanin production may occur away from the infection site (Banuett and Herskowitz, 1996). There seems to be no preference for hyphal invasion of particular cell types because hyphae grow within epidermal cells, parenchyma cells, and vascular bundle cells within the ﬁrst 4 days following inoculation (Snetselaar and Mims, 1994). The deformed rupture site between host cell walls suggests that U. maydis may rely on, at least partially, a mechanical means to break cell walls and proliferate from cell to cell (Snetselaar and Mims, 1994). Infected plant cells appear normal except that chloroplasts are sometimes enlarged and contain many starch granules (Snetselaar and Mims, 1994). Branch primordia that resemble clamp connections are observed in U. maydis hyphae during colonization (Banuett and Herskowitz, 1996; Christensen, 1963; Snetselaar and Mims, 1994). However, these structures do not appear to serve the same function as true clamp connections; they do not fuse with the adjacent hyphal cells (Banuett and Herskowitz, 1996) and nuclear migration into them has not been observed (Snetselaar and Mims, 1994). Additionally, there are some conﬂicting reports regarding the moment of appearance of these structures. While Banuett and Herskowitz (1996) observed them as soon as

KERNEL KNOWLEDGE: SMUT OF CORN

275

one day after inoculation, Snetselaar and Mims (1994) stated that clamp-like structures appear only in binucleate hyphae at later stages of colonization. During colonization, hyphae appear vacuolated and collapsed in some areas (Banuett and Herskowitz, 1996; Snetselaar and Mims, 1993). In contrast with branching that is only observed within plant tissue, migration of cytoplasm to the tip during hyphal extension, leaving areas of empty cells, is also a characteristic of the dikaryon formed in vitro (Day and Anagnostakis, 1971). Two studies in particular have contributed toward the understanding of this process. kin2, encoding a heavy chain of conventional kinesin, was identiﬁed in a candidate gene approach (Lehmler et al., 1997). Dikaryons formed by fusion of kin2 mutants produced hyphae that remained short and ﬁlled with cytoplasm (Lehmler et al., 1997). Further investigation of dikaryotic hyphae formed by kin2 mutants (Steinberg et al., 1998) showed that they lacked the large basal vacuole present in wild-type dikaryons. kin2 strains showed a severe reduction in pathogenicity (Lehmler et al., 1997), suggesting that both vacuolization and cytoplasmic migration were important during normal development of the dikaryon in planta. The extent of hyphal ramiﬁcation may depend on the location of the plant infection. In stigma infections, hyphae grew rapidly without branching; in contrast, more hyphal branching was observed in ovaries and leaves (Snetselaar and Mims, 1993). Hyphae were generally oriented longitudinally, parallel to the vascular bundles in stigma infections but not preferentially in pollen tracts (Snetselaar et al., 2001). Interestingly, the number of smutted kernels was signiﬁcantly reduced if ﬂowers were pollinated prior to infecting the silks with U. maydis (Snetselaar et al., 2001). In silks attached to ovaries that were pollinated before inoculation an abscission zone formed at the base of the silk and the fungus could still be seen growing in the silk towards the ovary. However, very few hyphae were found growing beyond the abscission zone and into the ovary. The inability of U. maydis to grow through a style consisting of collapsed cells and disorganized tissue was postulated to cause the reduced kernel infection. Fungal in planta growth and development is distinct from that in culture. Although mating can be induced in vitro, the resulting dikaryon cannot be maintained in culture. Additionally, structures characteristic of the development of the dikaryon during colonization, such as branching and clamp-like structures, are never observed in vitro. This strongly suggests that signals from the plant are important for maintenance of the dikaryon and hence successful colonization. A further piece of evidence that signals from the plant induce developmental

276

GARCI´A-PEDRAJAS AND GOLD

processes in U. maydis is the fact that diploid solopathogenic strains homozygous at the a locus are unable to ﬁlament in vitro but readily ﬁlament in the infected plant (Banuett and Herskowitz, 1996). Furthermore, mutant strains lacking ubc1 (regulatory subunit of PKA) are nonﬁlamentous in vitro but can colonize the plant and grow as ﬁlaments in infected tissue (Gold et al., 1997). These results are consistent with the observation that pathogenicity is always associated with ﬁlamentous growth in the plant. Developmental events in the proliferating hyphae may be triggered on cue in response to multiple plant signals. However, little is known about such signals and the genes involved in inducing developmental responses in the fungus. How U. maydis acquires nutrients during development inside the host is another area that is currently not well understood. Analyses of electron micrographs of U. maydis invading host tissues has revealed the presence of an intracellular structure that somewhat resembles haustoria described in other fungi. However, there is no clear demarcation of an interface analogous to the haustoria of rusts and other biotrophic fungi (Hahn and Mendgen, 2001). Furthermore, these structures are not consistently observed, and when present they are very irregularly branched (Luttrell, 1987; Snetselaar and Mims, 1994). Conceivably, this irregularly branched structure may correspond to the multi-lobed, sporogenous hyphae that appear intracellularly immediately prior to spore formation (Banuett and Herskowitz, 1996) rather than structures formed to obtain nutrients. The plant may exhibit disease symptoms, such as chlorosis, well in advance of the hyphae of U. maydis (Callow and Ling, 1973), suggesting release of toxins and/ or degradative enzymes by the fungus. It has also been established that when infected maize leaves are provided with the radioactive [14CO2], 14 C assimilates are increasingly imported into infected tissue, even prior to gall formation (Billett and Burnett, 1978). Nutrient acquisition by the pathogen in planta is an area that requires further research attention. Once fungal colonization is underway the most remarkable symptom induced by U. maydis is hypertrophy of plant cells, which is observed macroscopically in the form of tumors. Within those tumors a massive proliferation of fungal sporogenous hyphae takes place that differentiate into diploid teliospores. Fungal development in galls leading toward the formation of mature diploid teliospores is an ordered process with distinctive stages that have been studied in detail, especially in galls formed in leaves and stems. As the fungus proliferates within the plant tumors, an increase in hyphal branching is observed. Increasingly shorter branches are formed, especially at the hyphal tips, and

KERNEL KNOWLEDGE: SMUT OF CORN

277

appear to signal the switch from vegetative to sporogenous hyphae (Banuett and Herskowitz, 1996). At this stage, the fungus is embedded in a mucilaginous material presumably derived from hyphal walls (Christensen, 1932; Snetselaar and Mims, 1994). Banuett and Herskowitz (1996) observed that hyphae embedded in this matrix tend to stick together even when squeezed out of the plant cell. The tips of the hyphae become lobed, followed by hyphal fragmentation that produces segments of one to several cells. There are contradicting reports as to whether mycelial fragmentation occurs intracellularly or intercellularly. Several authors have reported that this process takes place mostly intercellularly (Mills and Kotze, 1981; Snetselaar and Mims, 1994). However, Banuett and Herskowitz (1996) showed in nonﬁxed samples that proliferation and fragmentation of hyphae occurred within tumor cells. An explanation for these different observations could be artifacts in ﬁxation techniques, but it is also possible that plant cells rupture due to the pressure exerted by fungal hyphae and mucilaginous material. At this stage karyogamy has presumably taken place, because DAPI staining shows a single nucleus per cell (Banuett and Herskowitz, 1996; Snetselaar and Mims, 1994). It has been assumed that for most smut fungi meiosis directly follows nuclear fusion. However, in U. maydis, karyogamy appears to take place early in the development of sporogenous hyphae, and it has been suggested that after nuclear fusion, diploid nuclei divide mitotically, giving rise to masses of diploid uninucleate cells (Snetselaar and Mims, 1994). Diploid nuclei are clearly capable of mitotic division because diploid budding sporidia are easily isolated from immature galls (Holliday, 1961b; Snetselaar and Mims, 1992, 1994). Following fragmentation, hyphal cells become rounded, and deposition of the secondary cell wall takes place. Cell walls are yellow-brown at ﬁrst and become dark brown and show a characteristic echinulation in mature teliospores. As the number of mature teliospores increases, tumors become dark. These processes are not synchronized in different tumors, and even in different parts of the same tumor, teliospores at different stages of development are encountered (Banuett and Herskowitz, 1996). When infection occurs via silks, tumors develop in the form of enlarged hollow ovaries with galled ovary walls (Snetselaar et al., 2001). The fungal genetic determinants of gall formation and teliosporogenesis have been a subject of research focus. Gall induction, teliosporogenesis, and cAMP signaling are intertwined processes. The cAMP signaling pathway appears to play a critical role in gall formation, especially during the induction of tumors, hyphal fragmentation, and the formation of prespores (Lee et al., 2003). Mutant strains with

278

GARCI´A-PEDRAJAS AND GOLD

different levels of perturbation of cAMP signaling pathway arrest at different developmental stages, indicating that changes in PKA activity play an important role in regulating these transitions. For example, mutant strains with constitutive PKA activity achieved by disruption of ubc1, the regulatory subunit of PKA, are able to infect and colonize plant tissue but unable to induce gall formation (Gold et al., 1997). Kruger et al. (2000) showed that even moderate activation of the cAMP signaling pathway has an effect on tumor morphology and the amount of fungal material in the tumors formed. They identiﬁed mutant alleles in gpa3 (gpa3Q206L) and ubc1 (ubc1R321Q), both of which presumably result in activation of the cAMP pathway. However, in contrast with ubc1 mutant, infection with two compatible gpa3Q206L or ubc1R321Q strains led to 50% and 12% of infected plants developing tumors, respectively. In addition to this decrease in disease severity in comparison with wild-type infections, tumors induced by these mutant strains were altered in morphology, and no teliospores were produced. Detailed microscopic observations showed arrest of fungal development after hyphae had formed lobed tips. Kruger and coworkers concluded that these mutant strains represent different levels of cAMP pathway activation, with the activation caused by the ubc1R321Q allele intermediate between the gpa3Q206 and ubc1 strains. These differences in cAMP pathway deregulation would account for the differences observed in disease severity among the mutants, with the lowest severity found in the mutant with a highest level of cAMP pathway activation. In contrast with the pathogenic behavior of mutants with an activated cAMP pathway, strains with low PKA activity caused by inactivating mutations in genes of the cAMP pathway, such as adenylate cyclase (uac1) or the catalytic subnunit of PKA (adr1), do not induce any symptoms in the host plant. Taken together, these results suggest that the levels of cAMP and PKA activity are high during penetration and initial colonization of plant tissue, while a drop in PKA activity signals the induction of gall formation. Mutations in genes encoding proteins that are putative targets for PKA also affect gall and teliospore formation. A search for suppressors of the ﬁlamentous phenotype of the adr1 mutant identiﬁed a gene, hgl1, whose product is a potential target for phosphorylation by PKA (Durrenberger, 2001). Inoculation of plants with compatible combinations of hgl1 mutant strains produced galls, generally larger than those produced by wild-type strains but lacking mature, darkly pigmented teliospores (Fig. 5). Close examination of these tumors showed that the fungus was unable to progress from the stage of hyphal fragmentation to teliospore development. In addition to PKA, two other protein

KERNEL KNOWLEDGE: SMUT OF CORN

279

FIG. 5. Teliospore formation on ﬂoral tissue by wild-type strains and hgl1 mutants. (A) Infected ears of maize were photographed 20 days after inoculation with hgl1 mutants (left) or wild-type strains (right). (B) Section of tumor tissue from the ear inoculated with the hgl1 mutants. (C) Section of tumor tissue from the ear inoculated with the wild-type strains showing the presence of round, melanized teliospores. The bar in (B) and (C) represents 20 m. Reproduced from Durrenberger et al., 2001, with permission.

280

GARCI´A-PEDRAJAS AND GOLD

kinases have been shown to be important for disease development and speciﬁcally for gall formation. Mutants in ukb1, a gene that encodes a putative serine/threonine protein kinase, are able to colonize plant tissue but incapable of inducing gall formation (Abramovitch et al., 2002). Plants inoculated with compatible haploid strains both deleted for ukc1, another protein kinase, or diploid strains harboring the same mutation, did not produce galls (Durrenberger and Kronstad, 1999). A number of additional genes involved in gall formation and teliosporogenesis have been identiﬁed. Quadbeck-Seeger et al. (2000) searched for components of the b-dependant regulatory cascade using a gene encoding an endo-glucanase, egl1, speciﬁcally expressed in the dikaryon (Schauwecker et al., 1995) as a reporter gene. Identiﬁcation of mutations that lead to expression of egl1 in haploid cells and complementation of one such mutant identiﬁed a gene coding for a protein, Rum1, which exhibited similarity to the human retinoblastoma binding protein 2. In a similar screen, Reichmann et al. (2002) identiﬁed a gene encoding Hda1, a protein with histone deacetylase activity. Deletion of either gene led to expression in haploid cells of several genes known to be b-regulated, as well as induction of the bE and bW genes themselves. These mutations had a similar effect on disease development: both blocked teliospore formation after karyogamy at the stage of hyphal fragmentation. However, detailed microscopic observation revealed that tumors induced by hda1 or rum1 strains have small and large areas of fragmented hyphae, respectively. Thus the block in teliospore development appears to occur earlier in hda1 cells than in rum1 mutants. As already discussed, it has been suggested that after karyogamy U. maydis proliferates within galled tissue by mitotic division of the diploid cells before mature teliospores are formed. Reichmann and coworkers (2002) postulated that for hda1 mutants, teliospore development appears to be blocked right after karyogamy and before mitotic division of the diploid nuclei. On the other hand, rum1 mutants can divide mitotically after nuclear fusion but are unable, at this stage, to reprogram their development in the direction of spore maturation. For both mutants, the lack of mature teliospores was postulated to be the result of the deregulation of a set of genes whose temporal or spatial misexpression prevents the completion of the disease cycle. Further, it was hypothesized that Hda1 functions in a complex with Rum1 and also independently from it (Reichmann et al., 2002). Infection of maize by U. maydis leads to obvious morphological alterations in the host plant hypothesized to be due to the action of phytohormones, either produced by the fungus or induced in the plant.

KERNEL KNOWLEDGE: SMUT OF CORN

281

Over 60 years ago, Moulton (1942) found that galled tissue contained an auxin in much higher amounts than uninfected tissue. This auxin was later identiﬁed as indole-3-acetic acid (IAA) (Turian and Hamilton, 1960; Wolf, 1952). In more recent years, as a ﬁrst step towards elucidating the putative role of this auxin in pathogenicity, attempts have been made to identify mutant U. maydis strains impaired in their ability to produce IAA in culture. Guevara-Lara et al. (2000) have characterized a number of mutants with reduced IAA production in vitro. Unfortunately, the only strain null for IAA production in vitro available to them, named udi-1 (Sosa-Morales et al., 1997), was also auxotrophic for L-methionine. Because L-methionine auxotrophic strains are compromised for pathogenicity (Fischer et al., 2001; Holliday, 1961a) and show reduced in vitro mating reactions (Fischer et al., 2001), the effect of the IAA (udi-1) mutation on disease severity was confounded. Attempts to obtain IAA/methþ segregants generated only strains with partial loss of IAA production in vitro. Thus, although inoculation with these strains did suggest a role for IAA production in gall formation, these results were not conclusive. Even wild-type strains exhibit variability in IAA production in vitro. Another approach to study the role of IAA in gall formation has been to identify genes predicted to be involved in IAA production, and then generate mutant strains deleted for such genes. Because U. maydis is able to produce IAA in vitro from tryptophan, the role of an indole3-acetaldehyde dehydrogenase, Iad1, was thought to be involved (Basse et al., 1996). However, not only were iad1 mutant strains fully pathogenic, but they still produced IAA in vitro from tryptophan, suggesting a different pathway for IAA synthesis. C. IN PLANTA GENE EXPRESSION To understand the disease interaction more completely, in addition to analysis of mutants, the identiﬁcation of fungal genes speciﬁcally expressed in planta is being pursued. Using differential display, a maize-induced gene (mig1), which encodes a secreted protein, was identiﬁed. In planta observations of mig1 expression with GFP as a reporter system revealed its strong up-regulation following penetration until the development of sporogenic hyphae (Basse et al., 2000). Fluorescence was no longer detectable in pigmented teliospores. However, mig1 does not play an essential role during colonization, as its deletion did not compromise pathogenicity. The authors suggested that Mig1 could be an enzyme involved in nutrient uptake or acquisition, or, alternatively, it may function as an elicitor (Basse et al., 2000). Five

282

GARCI´A-PEDRAJAS AND GOLD

additional mig genes similar to mig1 have now been identiﬁed (Basse et al., 2002). These genes (mig2–1, mig2–2, mig2–3, mig2–4, and mig2–5) are arranged as direct repeats in a 7.1 kb cluster. Deletion of the entire cluster did not have an effect on pathogenicity. Features of the mig genes such as secretion, plant-inducible expression, and an even number of cysteines are reminiscent of fungal avr genes (reviewed in Basse et al., 2002). Thus far, a gene-for-gene interaction has not been observed in the U. maydis-maize interaction. In a recent work, Aichinger et al. (2003) have reported further identiﬁcation of in planta induced genes. Using an approach that combines REMI (restriction enzyme mediated integration) mutagenesis with enhancer trapping by using the gene for green ﬂuorescent protein (GFP) as a reporter for in the plant detection, these authors found three plant induced genes (pig) among a collection of 2,350 insertion strains. pig1 was found to be mfa1, and pig2 showed similarity to a disulﬁde isomerase. The third integration event occurred in a locus that was designated the p-locus. It contains 11 genes and spans 24 kb. Five of the genes in the p-locus showed a plant regulated expression pattern. As previously found with the mig genes, deletion of this new set of plant induced genes did not produce any effect on virulence. Nevertheless, it is hoped that further characterization of plant induced fungal genes will provide insight into the developmental program of U. maydis during pathogenesis. IV. The Host Reaction A thus-far understudied area is the plant side of the interaction. With over $130 million invested by the U.S. National Science Foundation’s Plant Genome Research Grants on maize genome–related projects since 1998, this situation is likely to change soon. The promise of tools such as large microarray chips for analysis of plant gene expression concurrently with that of the infecting fungus offers an obvious avenue of exploration. Under the Maize Gene Discovery Project, an NSF-funded plant genome initiative that started in 1998, several laboratories collaborate with the common goal of discovering new maize genes and developing tools for the phenotypic characterization of maize mutants (Lunde et al., 2003). This initiative includes EST sequencing, cDNA microarray production, and tools to study gene function by production of insertional mutants in maize using a recombinant Mu1 transposon (RescueMu). These resources are for public use and their description is available online at ZmDB (http://www.zmdb.iastate.edu/). The very specialized interaction between U. maydis and its host obviously suggests cross-species signaling; the fungus induces

KERNEL KNOWLEDGE: SMUT OF CORN

283

signiﬁcant developmental alterations in the plant, leading to the formation of galls. An interesting aspect of this process is that U. maydis appears to be able to send signals that alter normal maize morphology even in the absence of extensive proliferation of fungal hyphae in the galls. Thus, Kruger and coworkers (2000) observed that inoculations with compatible gpa3Q206L mutant strains, in which cAMP signaling is enhanced through an activated allele of gpa3, induced tumors similar in size to those induced by wild-type inoculation. However, those tumors contained a much smaller amount of fungal material than normal galls induced by wild-type inoculation. Because of the formation of these highly modiﬁed structures, it is speculated that some of the genes that are either up- or down-regulated in the plant on gall formation are involved in developmental processes, primarily representing genes ectopically expressed in tissue or developmental stages in which they are not usually expressed. To date, however, no genes involved in gall formation from maize have been identiﬁed. Nothing is therefore known about the plant side of the interaction at the genetic level—that is, which maize genes are either up- or down-regulated in response to fungal signals and involved in the morphological alterations observed in the plant. Phytohormones, either produced by the fungus or induced in the plant, likely play a role in tumor formation. As discussed in Section III.B, galled tissue contains a much higher level of IAA than does normal healthy tissue. The contribution of the plant to this or other auxins during gall formation has not been investigated. Our laboratory has used a subtractive technique, SSH-PCR, to identify genes differentially expressed in the disease interaction. For this, we made use of the wild-type gall producing U. maydis dikaryon and the contrasting disease symptoms caused by a ubc1 mutant dikaryon (Gold et al., 1997), able to colonize maize but unable to induce gall formation. Use of the mutant allows the removal of sequences from genes expressed in phases of plant infections other than gall formation (Garcı´a-Pedrajas and Gold, unpublished). The sequencing of clones from our SSHarrayed library, shown to be differentially expressed in galls by reserve northern blots, and comparison of these sequences with ESTs at ZmDB, are allowing us to identify genes of maize origin whose expression patterns change on gall formation. As an example, among the genes speciﬁcally up-regulated in galls that we have thus far identiﬁed is an EST with similarity to a gene encoding an ETA subunit of a chaperonin (Fig. 6) essential for proper folding of cyclin E (Won et al., 1998), which in turn is required for cell cycle progression. We postulate that this chaperonin could be involved in cell-cycle deregulation leading to

284

GARCI´A-PEDRAJAS AND GOLD

FIG. 6. Expression of a maize cDNA clone 2F1 (encoding an ETA subunit of CCTchaperonin) in galls versus nongalled symptomatic tissue. Lanes 1, 3 and 5, RNA from tissue infected with the non–gall forming U. maydis ubc1 mutant; 2, 4 and 6, RNA from galls obtained by wild-type infection. Lanes 1 and 2, total RNA; lanes 3, 4, 5 and 6, PolyAþRNA. Lanes 3 and 4 represent RNA from experiment 1 and lanes 5 and 6 from experiment 2. Samples from two separate experiments were tested to conﬁrm induction of expression of this gene in galls.

hyperplasia (Nadal et al., unpublished). After characterization of temporal differential expression by quantitative methods such as real-time PCR, we plan to search for insertional mutants in this and/or other genes differentially expressed in galls, and phenotypes such as plant development and susceptibility to U. maydis will be investigated. As transformation of maize becomes more technically feasible overexpression of some of these genes in maize is also likely to generate important information regarding their role in pathogenicity. In this section we focus our attention on gall formation, since it is the most obvious symptom produced by U. maydis, and the phase of infection most likely to involve alterations in gene expression in the host. Maize response to the initial steps of U. maydis colonization is not obvious; necrosis or other dramatic symptoms of plant response are not observed (Snetselaar and Mims, 1992). Therefore it has been assumed that this fungal species does not a elicit defense response in the plant. However, anthocyanin production, associated with a stress response to fungal penetration, is a characteristic symptom of inoculated maize seedlings (Banuett and Herskowitz, 1996; Hana, 1929). Response of maize to penetration by U. maydis and possible induction of a defense response remains unexplored in this pathosystem. V. Conclusions Ustilago maydis provides an excellent model for the study of fungal development and host-pathogen interactions. There are many powerful tools that have been generated to assist researchers in their studies on Ustilago species. Several research groups have contributed

KERNEL KNOWLEDGE: SMUT OF CORN

285

signiﬁcantly to our understanding of various aspects of the biology of this fungus. As described in this chapter, mating, morphogenesis, and pathogenicity have been a major focus of research attention to date. Much of the effort has indicated the intertwined nature of the signaling pathways that cross-talk to inﬂuence all of these processes. The era of genomics promises more rapid and comprehensive analysis of the biology of U. maydis in all areas of interest. Recently a cohesive international effort facilitated and managed by the Whitehead Institute for Genome Research has carried out the publicly funded genomic sequencing of U. maydis. Both Lion Bioscience AG/Bayer CropScience and Exelixis Inc. had generated genomic sequences of this fungus, and they donated this information to the Whitehead effort. With wide access to the U. maydis genomic sequence (http://www-genome.wi.mit. edu/annotation/fungi/ustilago_maydis/index.html), methods for understanding the biology of this fascinating plant pathogenic fungus will become even more powerful. Carrying forth this momentum further on an international level is evident by the ﬁrst international Ustilago conference (Kronstad, 2003) organized by Regine Kahmann and Flora Banuett. These same organizers are actively planning a second meeting in 2004. Discussions at the 2003 22nd Fungal Genetics Conference, Asilomar, were very encouraging regarding initiation of international cooperation on use of the genomic sequence. This cooperation is expected to initially involve genome annotation, production of full-length cDNAs, and functional genomics including expression analysis and a genome-wide gene deletion set. REFERENCES Abramovitch, R., Yang, G., and Kronstad, J. (2002). The ukb1 gene encodes a putative protein kinase required for bud site selection and pathogenicity in Ustilago maydis. Fungal Genet. Biol. 37, 98–108. Agrios, G. N. (1997). ‘‘Plant Pathology.’’ Academic Press, San Diego. Aichinger, C., Hansson, K., Eichhorn, H., Lessing, F., Mannhaupt, G., Mewes, W., and Kahmann, R. (2003). Identiﬁcation of plant-regulated genes in Ustilago maydis by enhancer-trapping mutagenesis. Mol. Genet. Genomics 270, 303–314. Alexopoulus, C. J., Mims, C. W., and Blackwell, M. (1996). ‘‘Introductory Mycology.’’ Wiley, New York. Andrews, D. L., Egan, J. D., Mayorga, M. E., and Gold, S. E. (2000). The Ustilago maydis ubc4 and ubc5 genes encode members of a MAP kinase cascade required for ﬁlamentous growth. Mol. Plant-Microbe Interac. 13, 781–786. Banuett, F. (1998). Signalling in the yeasts: An informational cascade with links to the ﬁlamentous fungi. Microbiol. Mol. Biol. Rev. 62, 249–274. Banuett, F., and Herskowitz, I. (1989). Different a-alleles of Ustilago maydis are necessary for maintenance of ﬁlamentous growth but not for meiosis. Proc. Natl. Acad. Sci. USA 86, 5878–5882.

286

GARCI´A-PEDRAJAS AND GOLD

Banuett, F., and Herskowitz, I. (1994). Identiﬁcation of fuz7, a Ustilago maydis MEK/ MAPKK homolog required for a-locus-dependent and -independent steps in the fungal life cycle. Genes Dev. 8, 1367–1378. Banuett, F., and Herskowitz, I. (1996). Discrete developmental stages during teliospore formation in the corn smut fungus, Ustilago maydis. Development 122, 2965–2976. Barrett, K. J., Gold, S. E., and Kronstad, J. W. (1993). Identiﬁcation and complementation of a mutation to constitutive ﬁlamentous growth in Ustilago maydis. Mol. PlantMicrobe Interac. 6, 274–283. Basse, C. W., Kolb, S., and Kahmann, R. (2002). A maize-speciﬁcally expressed gene cluster in Ustilago maydis. Mol. Microbiol. 43, 75–93. Basse, C. W., Lottspeich, F., Steglich, W., and Kahmann, R. (1996). Two potential indole3-acetaldehyde dehydrogenases in the phytopathogenic fungus Ustilago maydis. Eur. J. Biochem. 242, 648–656. Basse, C. W., Stumpferl, S., and Kahmann, R. (2000). Characterization of a Ustilago maydis gene speciﬁcally induced during the biotrophic phase: Evidence for negative as well as positive regulation. Mol. Cell. Biol. 20, 329–339. Billett, E. E., and Burnett, J. H. (1978). The host-parasite physiology of the maize smut fungus, Ustilago maydis II. Translocation of 14C-labelled assimilates in smutted maize plants. Physiol. Plant Pathol. 12, 103–112. Bolker, M., Urban, M., and Kahmann, R. (1992). The a mating type locus of U. maydis speciﬁes cell signaling components. Cell 68, 441–450. Brachmann, A., Weinzierl, G., Kamper, J., and Kahmann, R. (2001). Identiﬁcation of genes in the bW/bE regulatory cascade in Ustilago maydis. Mol. Microbiol. 42, 1047–1063. Brachmann, A., Schirawski, J., Muller, P., and Kahmann, R. (2003). An unusual MAP kinase is required for efﬁcient penetration of the plant surface by Ustilago maydis. EMBO J. 22, 2199–2210. Callow, J. A., and Ling, I. T. (1973). Histology of neoplasms and chlorotic lesions in maize seedlings following injection of sporidia of Ustilago maydis (Dc) Corda. Physiol. Plant Pathol. 3, 489. Caltrider, P. G., and Gottlieb, D. (1966). Effect of sugars on germination and metabolism of teliospores of Ustilago maydis. Phytopathology 56, 479–484. Cano-Canchola, C., Acevedo, L., Ponce-Noyola, P., Flores-Martinez, A., Flores-Carreon, A., and Leal-Morales, C. A. (2000). Induction of lytic enzymes by the interaction of Ustilago maydis with Zea mays tissues. Fungal Genet. Biol. 29, 145–151. Casselton, L. A. (2002). Mate recognition in fungi. Heredity 88, 142–147. Casselton, L. A., and Olesnicky, N. S. (1998). Molecular genetics of mating recognition in basidiomycete fungi. Microbiol. Mol. Biol. Rev. 62, 55–70. Christensen, J. J. (1932). Studies on the genetics of Ustilago zeae. Phytopathol. Zeit. 4, 129–188. Christensen, J. J. (1963). Corn Smut caused by Ustilago maydis. Monograph No. 2. American Phytopathological Society, Saint Paul. Day, P. R., and Anagnostakis, S. L. (1971). Corn smut dikaryon in culture. Nature-New Biol. 231, 19–20. Durrenberger, F., and Kronstad, J. (1999). The ukc1 gene encodes a protein kinase involved in morphogenesis, pathogenicity and pigment formation in Ustilago maydis. Mol. Gen. Genet. 261, 281–289. Durrenberger, F., Laidlaw, R. D., and Kronstad, J. W. (2001). The hgl1 gene is required for dimorphism and teliospore formation in the fungal pathogen Ustilago maydis. Mol. Microbiol. 41, 337–348.

KERNEL KNOWLEDGE: SMUT OF CORN

287

Fischer, G. W. (1936). The longevity of smut spores in herbarium specimens. Phytopathology 26, 1118–1127. Fischer, J. A., McCann, M. P., and Snetselaar, K. M. (2001). Methylation is involved in the Ustilago maydis mating response. Fungal Genet. Biol. 34, 21–35. Froeliger, E. H., and Leong, S. A. (1991). The a mating-type alleles of Ustilago maydis are idiomorphs. Gene 100, 113–122. Garcia-Muse, T., Steinberg, G., and Perez-Martin, J. (2003). Pheromone-induced G(2) arrest in the phytopathogenic fungus Ustilago maydis. Eukaryt. Cell. 2, 494–500. Garcı´a-Pedrajas, M. D., and Gold, S. E. (2004). Fungal dimorphism regulated gene expression in Ustilago maydis: II. Filament downregulated genes. Mol. Plant Path. 5, 295–308. Gold, S., Duncan, G., Barrett, K., and Kronstad, J. (1994). cAMP regulates morphogenesis in the fungal pathogen Ustilago maydis. Genes Dev. 8, 2805–2816. Gold, S. E., Brogdon, S. M., Mayorga, M. E., and Kronstad, J. W. (1997). The Ustilago maydis regulatory subunit of a cAMP-dependent protein kinase is required for gall formation in maize. Plant Cell. 9, 1585–1594. Guevara-Lara, F., Valverde, M. E., and Paredes-Lopez, O. (2000). Is pathogenicity of Ustilago maydis (huitlacoche) strains on maize related to in vitro production of indole-3-acetic acid? World J. Microbiol. Biotechnol. 16, 481–490. Hahn, M., and Mendgen, K. (2001). Signal and nutrient exchange at biotrophic plantfungus interfaces. Curr. Opin. Plant Biol. 4, 322–327. Hana, W. F. (1929). Studies in the physiology and cytology of Ustilago zeae and Sorosporium reilianum. Phytopathology 19, 415–442. Hartmann, H. A., Kahmann, R., and Bolker, M. (1996). The pheromone response factor coordinates ﬁlamentous growth and pathogenicity in Ustilago maydis. EMBO J. 15, 1632–1641. Holliday, R. (1961a). Induced mitotic crossing-over in Ustilago maydis. Genet. Res. 2, 231–248. Holliday, R. (1961b). The genetics of Ustilago maydis. Genet. Res. 2, 204–230. Holliday, R. (1974). Ustilago maydis. In ‘‘Handbook of Genetics’’ (R. C. King, ed.), pp. 575–595. Plenum, New York. Kahmann, R., Basse, C., and Feldbrugge, M. (1999). Fungal-plant signalling in the Ustilago maydis-maize pathosystem. Curr. Opin. Microbiol. 2, 647–650. Kronstad, J. W. (2003). Castles and cuitlacoche: The ﬁrst international Ustilago conference. Fungal Genet. Biol. 38, 265–271. Kronstad, J. W., and Leong, S. A. (1989). Isolation of two alleles of the b locus of Ustilago maydis. Proc. Natl. Acad. Sci. USA 86, 978–982. Kronstad, J. W., and Leong, S. A. (1990). The b mating-type locus of Ustilago maydis contains variable and constant regions. Genes Dev. 4, 1384–1395. Kruger, J., Loubradou, G., Regenfelder, E., Hartmann, A., and Kahmann, R. (1998). Crosstalk between cAMP and pheromone signalling pathways in Ustilago maydis. Mol. Gen. Genet. 260, 193–198. Kruger, J., Loubradou, G., Wanner, G., Regenfelder, E., Feldbrugge, M., and Kahmann, R. (2000). Activation of the cAMP pathway in Ustilago maydis reduces fungal proliferation and teliospore formation in plant tumors. Mol. Plant-Microbe Interac. 13, 1034–1040. Laity, C., Giasson, L., Campbell, R., and Kronstad, J. (1995). Heterozygosity at the b mating-type locus attenuates fusion in Ustilago maydis. Curr. Genet. 27, 451–459.

288

GARCI´A-PEDRAJAS AND GOLD

Lee, N., and Kronstad, J. W. (2002). ras2 controls morphogenesis, pheromone response, and pathogenicity in the fungal pathogen Ustilago maydis. Eukaryot. Cell. 1, 954–966. Lee, N., De’Souza, C. A., and Kronstad, J. W. (2003). Of smuts, blasts, mildews, and blights: cAMP signaling in phytopathogenic fungi. Ann. Rev. Phytopathol. 41, 399–427. Lehmler, C., Steinberg, G., Snetselaar, K. M., Schliwa, M., Kahmann, R., and Bolker, M. (1997). Identiﬁcation of a motor protein required for ﬁlamentous growth in Ustilago maydis. EMBO J. 16, 3464–3473. Lunde, C. F., Morrow, D. J., Roy, L. M., and Walbot, V. (2003). Progress in maize gene discovery: A project update. Funct. Integr. Genomics 3, 25–32. Luttrell, E. S. (1987). Relations of hyphae to host-cells in smut galls caused by species of tilletia, tolyposporium, and Ustilago. Can. J. Bot. 65, 2581–2591. Martinez-Espinoza, A. D., Ruiz-Herrera, J., Leon-Ramirez, C. G., and Gold, S. E. (2004). MAP kinase and cAMP signaling pathways modulate the pH-induced yeast-to-mycelium dimorphic transition in the corn smut fungus Ustilago maydis. Current Microbiol. 49, 274–281. Mayorga, M. E., and Gold, S. E. (1998). Characterization and molecular genetic complementation of mutants affecting dimorphism in the fungus Ustilago maydis. Fungal Genet. Biol. 24, 364–376. Mayorga, M. E., and Gold, S. E. (1999). A MAP kinase encoded by the ubc3 gene of Ustilago maydis is required for ﬁlamentous growth and full virulence. Mol. Microbiol. 34, 485–497. Mayorga, M. E., and Gold, S. E. (2001). The ubc2 gene of Ustilago maydis encodes a putative novel adaptor protein required for ﬁlamentous growth, pheromone response and virulence. Mol. Microbiol. 41, 1365–1379. Mills, L. J., and Kotze, J. M. (1981). Scanning electron-microscopy of the germination, growth and infection of Ustilago maydis on maize. Phytopathol. Zeit.-J. Phytopathol. 102, 21–27. Moulten, J. E. (1942). Extraction of auxin from maize, from smut tumor of maize and from Ustilago zeae. Bot. Gazete 103, 725–729. Muller, P., Aichinger, C., Feldbrugge, M., and Kahmann, R. (1999). The MAP kinase kpp2 regulates mating and pathogenic development in Ustilago maydis. Mol. Microbiol. 34, 1007–1017. Muller, P., Katzenberger, J. D., Loubradou, G., and Kahmann, R. (2003). Guanyl nucleotide exchange factor Sql2 and Ras2 regulate ﬁlamentous growth in Ustilago maydis. Eukaryot. Cell. 2, 609–617. Muller, P., Weinzierl, G., Brachmann, A., Feldbrugge, M., and Kahmann, R. (2003). Mating and pathogenic development of the Smut fungus Ustilago maydis are regulated by one mitogen-activated protein kinase cascade. Eukaryot. Cell 2, 1187–1199. ODonnell, K. L., and McLaughlin, D. J. (1984). Ultrastructure of meiosis in Ustilago maydis. Mycologia 76, 468–485. Puhalla, J. E. (1970). Genetic studies on the incompatibility locus of Ustilago maydis. Genet. Res. Camb. 16, 229–232. Quadbeck-Seeger, C., Wanner, G., Huber, S., Kahmann, R., and Kamper, J. (2000). A protein with similarity to the human retinoblastoma binding protein 2 acts speciﬁcally as a repressor for genes regulated by the b mating type locus in Ustilago maydis. Mol. Microbiol. 38, 154–166.

KERNEL KNOWLEDGE: SMUT OF CORN

289

Regenfelder, E., Spellig, T., Hartmann, A., Lauenstein, S., Bolker, M., and Kahmann, R. (1997). G proteins in Ustilago maydis: Transmission of multiple signals? EMBO J. 16, 1934–1942. Reichmann, M., Jamnischek, A., Weinzierl, G., Ladendorf, O., Huber, S., Kahmann, R., and Kamper, J. (2002). The histone deacetylase Hda1 from Ustilago maydis is essential for teliospore development. Mol. Microbiol. 46, 1169–1182. Romeis, T., Brachmann, A., Kahmann, R., and Kamper, J. (2000). Identiﬁcation of a target gene for the bE-bW homeodomain protein complex in Ustilago maydis. Mol. Microbiol. 37, 54–66. Rowell, J. B. (1955). Functional role of compatibility factors and an in vitro test for sexual compatibility of haploid lines of Ustilago zea. Phytopathology 45, 370–374. Rowell, J. B., and DeVay, J. E. (1954). Genetics of Ustilago zeae in relation to basic problems of its pathogenicity. Phytopathology 44, 356–362. Sacadura, N. T., and Saville, B. J. (2003). Gene expression and EST analyses of Ustilago maydis germinating teliospores. Fungal Genet. Biol. 40, 47–64. Schauwecker, F., Wanner, G., and Kahmann, R. (1995). Filament-speciﬁc expression of a cellulase gene in the dimorphic fungus Ustilago maydis. Biol. Chem. Hoppe Seyler. 376, 617–625. Schulz, B., Banuett, F., Dahl, M., Schlesinger, R., Schafer, W., Martin, T., Herskowitz, I., and Kahmann, R. (1990). The b alleles of U. maydis, whose combinations program pathogenic development, code for polypeptides containing a homeodomain-related motif. Cell 60, 295–306. Shutleff, M. C. (1980). ‘‘Compendium of Corn Diseases.’’ APS Press, St. Paul, MN. Sipiczki, M. (1988). The role of sterility genes (ste and aff ) in the initiation of sexual development in Schizosaccharomyces pombe. Mol. Gen. Genet. 213, 529–534. Smith, D. G., Garcı´a-Pedrajas, M. D., Hong, W., Gold, S. E., and Perlin, M. H. (2004). A ste20 homologue in Ustilago maydis plays a role in mating, dimorphism, and pathogenicity. Eukaryot. Cell 3, 180–189. Snetselaar, K. M., and Mims, C. W. (1992). Sporidial fusion and infection of maize seedlings by the smut fungus Ustilago maydis. Mycologia 84, 193–203. Snetselaar, K. M., and Mims, C. W. (1993). Infection of maize stigmas by Ustilago maydis: light and electron-microscopy. Phytopathology 83, 843–850. Snetselaar, K. M., and Mims, C. W. (1994). Light and electron-microscopy of Ustilago maydis hyphae in maize. Mycol. Res. 98, 347–355. Snetselaar, K. M., and McCann, M. P. (1997). Using microdensitometry to correlate cell morphology with the nuclear cycle in Ustilago maydis. Mycologia 89, 689–697. Snetselaar, K. M., Carﬁoli, M. A., and Cordisco, K. M. (2001). Pollination can protect maize ovaries from infection by Ustilago maydis, the corn smut fungus. Can. J. Bot. 79, 1390–1399. Sosa-Morales, M. E., Guevara-Lara, F., Martinez-Juarez, V. M., and Paredes-Lopez, O. (1997). Production of indole-3-acetic acid by mutant strains of Ustilago maydis (maize smut huitlacoche). Appl. Microbiol. Biotechnol. 48, 726–729. Spellig, T., Bolker, M., Lottspeich, F., Frank, R. W., and Kahmann, R. (1994). Pheromones trigger ﬁlamentous growth in Ustilago maydis. EMBO J. 13, 1620–1627. Stakman, E. C., and Christensen, J. J. (1927). Heterothallism in Ustilago zeae. Phytopathology 17, 827–834. Steinberg, G., Schliwa, M., Lehmler, C., Bolker, M., Kahmann, R., and McIntosh, J. R. (1998). Kinesin from the plant pathogenic fungus Ustilago maydis is involved in vacuole formation and cytoplasmic migration. J. Cell. Sci. 111, 2235–2246.

290

GARCI´A-PEDRAJAS AND GOLD

Trueheart, J., and Herskowitz, I. (1992). The a locus governs cytoduction in Ustilago maydis. J. Bacteriol. 174, 7831–7833. Turian, G., and Hamilton, R. H. (1960). Chemical detection of 3-indolylacetic acid in Ustilago zeae tumors. Biochim. Et Biophys. Acta 41, 148–150. Urban, M., Kahmann, R., and Bolker, M. (1996). The biallelic a mating type locus of Ustilago maydis: Remnants of an additional pheromone gene indicate evolution from a multiallelic ancestor. Mol. Gen. Genet. 250, 414–420. Valdivieso, M. H., Sugimoto, K., Jahng, K. Y., Fernandes, P. M., and Wittenberg, C. (1993). FAR1 is required for posttranscriptional regulation of CLN2 gene expression in response to mating pheromone. Mol. Cell. Biol. 13, 1013–1022. Walter, J. M. (1934). The mode of entrance of Ustilago zeae into corn. Phytopathology 24, 1012–1020. Wolf, F. T. (1952). The production of indole acetic acid by Ustilago zeae, and its possible signiﬁcance in tumor formation. Proc. Natl. Acad. Sci. USA 38, 106–111. Won, K. A., Schumacher, R. J., Farr, G. W., Horwich, A. L., and Reed, S. I. (1998). Maturation of human cyclin E requires the function of eukaryotic chaperonin CCT. Mol. Cell. Biol. 18, 7584–7589.

Bacterial ACC Deaminase and the Alleviation of Plant Stress BERNARD R. GLICK Department of Biology, University of Waterloo Waterloo, Ontario, Canada N2L 3G1 E-mail: [email protected]

I. ACC Deaminase–Containing Bacteria II. Ethylene and Plant Stress III. Decreasing Plant Stress with ACC Deaminase–Containing Bacteria A. Flooding B. Organic Toxicants C. Metals D. Biocontrol of Pathogens E. Flower Wilting F. Drought G. High Salt IV. Modulating Nodulation of Legumes V. Decreasing Stress in ACC Deaminase Transgenic Plants VI. Conclusions References

291 293 295 295 297 298 300 301 302 303 304 306 307 308

I. ACC Deaminase–Containing Bacteria One of the mechanisms that a number of plant growth-promoting bacteria use to facilitate plant growth and development is the lowering of a plant’s ethylene concentration through the action of the enzyme 1-aminocyclopropane-1-carboxylate (ACC) deaminase (Glick, 1995; Glick et al., 1999). In a model that was developed to explain the role of ACC deaminase in plant growth promotion, the following scenario was envisioned (Glick et al., 1998). The ACC deaminase–containing plant growth-promoting bacteria bind to the surface of either the seed or root of a developing plant, and in response to tryptophan and other small molecules that are found in seed and/or root exudates (Bayliss et al., 1997; Penrose and Glick, 2001; Whipps, 1990), the bacteria synthesize and secrete indole acetic acid (IAA) (Fallik et al., 1994; Frankenberger and Arshad, 1995; Patten and Glick, 1996; Xie et al., 1996), some of which is taken up by the plant. This IAA, together with endogenous plant IAA, can stimulate plant cell proliferation or plant cell elongation and/or can induce the synthesis of the plant enzyme ACC synthase that converts S-adenosylmethionine to ACC (Kende, 1993). A portion of the ACC produced by this latter reaction may be exuded from seeds or plant roots along with other small molecules that 291 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

292

BERNARD R. GLICK

are normally present in seed or root exudates (Penrose and Glick, 2001). The ACC in the exudates (as well as some of the other small molecules) is then taken up by the bacteria and subsequently converted by the enzyme, ACC deaminase, to ammonia and -ketobutyrate, both of which are readily metabolized by most soil bacteria. The presence of a bacterium on a seed or root both induces a plant to synthesize more ACC than it would otherwise need and stimulates the exudation of ACC and other small molecules from the plant. As a result of lowering the ACC level within a plant, either the endogenous level or the IAAstimulated level, the amount of ethylene that can subsequently form in the plant is also reduced. Hence, when ACC deaminase–containing bacteria are bound to a plant’s tissues, the bacteria act as a sink for ACC. Unpublished experiments from our lab indicate that ACC deaminase-containing bacteria may be found on leaves and ﬂowers as well as on seeds and roots, so that the model that has been developed based on the interaction of bacteria with seeds and roots is likely applicable to the entire plant. Thus, ACC deaminase-containing plant growthpromoting bacteria can take up and cleave plant ACC and thereby lower plant ethylene levels, regardless of whether the ACC was produced as a consequence of normal development, interaction with bacteria, or some sort of environmental stress. The enzyme ACC deaminase has been found to be associated with a large number of different soil microorganisms, both bacteria and fungi (Belimov et al., 2001, 2004; Burd et al., 1998; Campbell and Thompson, 1996; Ghosh et al., 2003; Glick, 1995; Glick et al., 1995; Honma, 1993; Jacobson et al., 1994; Jia et al., 1999; Klee et al., 1991; Ma et al., 2003; Minami et al., 1998; Shah et al., 1998; Sheehy et al., 1991; Wang et al., 2001). However, a certain amount of caution should be exercised when deciding whether a particular gene does or does not encode ACC deaminase. For example, DNA sequence analysis suggests that Escherichia coli K-12 contains ACC deaminase although, despite repeated attempts with a number of different strains, ACC deaminase activity has never been detected in E. coli (Glick et al., unpublished observations). In addition to E. coli, DNA sequence analysis has been used to putatively identify ACC deaminases from Pyrococcus horikoshii, Pyrococcus abyssi, Thermotoga maritima, Caulobacter crescentus, and Arabidopsis thaliana. A comparison of the DNA sequences of the putative ACC deaminases with DNA sequences that encode veriﬁable enzyme indicates that the putative enzymes show increasing divergence from Enterobacter cloacae UW4, especially the archaebacteria Thermotoga maritima and the plant Arabidopsis thaliana so that at this time, it is unclear whether or not these putative ACC deaminases are

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

293

bona ﬁde ACC deaminases. This caveat notwithstanding, there is also some evidence from two different laboratories (that is as yet unpublished and largely based on DNA sequence similarity) that both poplar and birch trees may contain the gene for ACC deaminase. If this is in fact the case, then it is possible that these plants acquired ACC deaminase genes by DNA transfer from soil microorganisms that normally interact with these plants. II. Ethylene and Plant Stress The gaseous plant hormone ethylene is important for normal development in plants as well as for their response to stress (Abeles et al., 1992; Arshad and Frankenberger, 2002; Matoo and Suttle, 1991). Ethylene mediates a wide range of plant responses and developmental steps; it is involved in seed germination, root elongation, tissue differentiation, formation of root and shoot primordia, lateral bud development, ﬂowering initiation, anthocyanin synthesis, ﬂower opening and senescence, fruit ripening and degreening, production of volatile organic compounds responsible for aroma formation in fruits, storage product hydrolysis, leaf and fruit abscission, and the response of plants to biotic and abiotic stress. For some of these biological processes ethylene is stimulatory while for others it is inhibitory. The term ‘‘stress ethylene,’’ originally coined by Abeles, describes the peak of ethylene biosynthesis that is associated with biological and environmental stresses, and pathogen attack (Hyodo, 1991). The increased level of ethylene formed in response to trauma inﬂicted by chemicals, temperature extremes, water stress, ultraviolet light, insect damage, disease, and mechanical wounding can be both the cause of some of the symptoms of stress (e.g., onset of epinastic curvature and formation of arenchyma) and the inducer of responses that will enhance survival of the plant under adverse conditions (Stearns and Glick, 2003; van Loon and Glick, 2004; van Loon et al., 1997). Ethylene that is synthesized in response to a variety of environmental stresses appears to be produced in two peaks (Abeles et al., 1992). The best studied stress in this regard is infection by fungal pathogens. The ﬁrst peak of ethylene is small, generally occurs within a few hours after the stress, and is often difﬁcult to detect by gas chromatography— the standard means of measuring plant ethylene production. The difﬁculty of measuring this peak of ethylene synthesis notwithstanding, its presence may be inferred from genetic experiments with the plant Arabidopsis thaliana in which reduced sensitivity to ethylene cosegregates with the inability to express the defense response that has

294

BERNARD R. GLICK

been termed induced systemic resistance (Ton et al., 2001). This ﬁrst peak of ethylene is thought to act as a signal to turn on transcription of genes that encode proteins that help to protect the plant against the pathogen—for example, as part of the mechanism of induced systemic resistance (van Loon and Glick, 2004; van Loon et al., 1997). The second ethylene peak is much larger than the ﬁrst peak, is readily detected by using standard gas chromatography techniques, and generally occurs several days after the stress, often concomitant with the appearance of visible damage to the plant. The second ethylene peak can exacerbate the impact of the environmental stress—for example, by triggering a senescence response (localized or otherwise) in the plant. Indeed, a number of studies indicate that high ethylene levels are associated with increased damage to the plant by the pathogen/stressor (Abeles et al., 1992; Bashan, 1994; Biles et al., 1990; Cohen et al., 1986; Cronshaw and Pegg, 1976; Elad, 1988, 1990; van Loon, 1984). Ideally, it would appear to be advantageous for plants to synthesize the ﬁrst but not the second peak of ethylene in response to an environmental stress. This would have the effect of turning on some of the plant’s defense responses and at the same time avoiding any senescence response. While it is not simple to engineer a plant to respond to environmental stresses in this manner, a selective decrease in the second but not the ﬁrst ethylene peak may be achieved by using ACC deaminase–containing plant growth–promoting bacteria. It was previously observed, presumably as a consequence of the complex mode of transcriptional regulation of ACC deaminase expression in many bacterial strains (Grichko and Glick, 2000; Li and Glick, 2001), that the expression of ACC deaminase is relatively low in the absence of ACC and requires several hours following the addition of ACC before enzyme activity is fully induced (Jacobson et al., 1994; Ma et al., 2003). Thus it is envisioned that following an environmental stress, a portion of a plant’s existing pool of ACC will be converted to ethylene by the low level of endogenous ACC oxidase resulting in the ﬁrst ethylene peak. Subsequently, the stress may induce the transcription of the enzyme ACC synthase, resulting in a relatively large increase in ACC and then ethylene levels, unless an ACC deaminasecontaining bacterium is available to lower the level of ACC before the expression of the enzyme ACC oxidase is induced to any signiﬁcant extent. While ethylene signaling is required for the induction of systemic resistance elicited by rhizobacteria, a signiﬁcant increase in the level of ethylene is not. Hence, lowering of ethylene levels by bacterial ACC deaminase does not appear to be incompatible with the induction of systemic resistance. Indeed, some bacterial strains possessing ACC deaminase also induce systemic resistance. What plant genes become

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

295

expressed during these processes, what regulatory factors are involved, and how these processes interact are questions that need to be further addressed to obtain a fuller understanding of how plants can cope with various stresses by beneﬁting from their rhizosphere microﬂora. III. Decreasing Plant Stress with ACC Deaminase–Containing Bacteria Several different chemicals have been successfully used to lower ethylene levels in plants (e.g., L--(aminoethoxyvinyl)-glycine, AVG; and aminooxyacetic acid, AOA) or to alter a plant’s sensitivity to ethylene (e.g., 1-methylcyclopropene, 1-MCP), especially during fruit ripening and ﬂower wilting (Abeles et al., 1992; Sisler and Serek, 1997). Many of these chemicals are either expensive or potentially harmful to the environment, and, as far as we are aware, none of them have been utilized to decrease the effects of various environmental stresses on plants. Nevertheless, 1-MCP has been approved for commercial use and has been tested with a wide range of fruits and ﬂowers to limit post-harvest fruit spoilage and ﬂower wilting that occurs as a consequence of ethylene synthesis (Watkins and Miller, 2003). On the other hand, if the model of how ACC deaminase–containing bacteria can act as a sink for ACC and thereby lower the level (and thus decrease the impact) of stress ethylene has any validity, then one would expect that these bacteria should decrease the deleterious effects of a large number of different ethylene-inducing stresses (Fig. 1). As indicated below, ACC deaminase-containing plant growth-promoting bacteria have been tested to see whether they could be used as an environmentally friendly means of lowering ethylene levels in plants subjected to different stresses. It should be emphasized here that the strategies and approaches discussed below are not equally applicable to all plants, as not all plants are equally sensitive to ethylene. For example, in one study it was found that dicots were much more ethylene sensitive than were monocots (Hall et al., 1996); the ﬂowers of different plants were found to have widely varying sensitivities to ethylene (Woltering and van Doorn, 1988); and rice plants are ethylene insensitive when they are grown in the ﬁeld under ﬂooded conditions and ethylene sensitive at an early seedling stage. A. FLOODING Flooding is a common biotic stress that affects many plants, often several times during the same growing season. Plant roots are exposed to decreased oxygen concentrations as a consequence of ﬂooding; then

296

BERNARD R. GLICK

FIG. 1. Schematic representation of the effect of the absence (A) or presence (B) of ACC deaminase–containing plant growth–promoting bacteria on plant proliferation following any one of a variety of different ethylene-producing stresses (e.g., high and low temperature, high salt, phytopathogens, high salt, metals, organic xenobiotics, ﬂooding, and drought). The size of the arrow indicates the relative degree of ﬂux through a particular pathway, that is, large arrows represent a signiﬁcant ﬂux and small arrows a more limited ﬂux.

some of the plant’s ACC synthase genes, encoding the enzyme that converts the compound S-adenosylmethionine into ACC, are rapidly induced in the roots (Olson et al., 1995). Since ACC oxidase-catalyzed ethylene synthesis cannot occur in the anaerobic environment of ﬂooded roots, the ACC that has accumulated within the roots is transported to the shoots where there is an aerobic environment and the ACC can be converted to ethylene (Bradford and Yang, 1980; Else and Jackson, 1988). This synthesis of ethylene by ﬂooded plants in turn causes deleterious effects for the plant, such as epinasty, leaf chlorosis, necrosis, and reduced fruit yield. However, treatment of plants with ACC deaminase–containing, plant growth–promoting bacteria

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

297

can signiﬁcantly decrease the damage that would otherwise be suffered by these plants as a result of ﬂooding (Grichko and Glick, 2001a). These bacteria lower the level of ethylene that can be formed in the shoots and thus ‘‘protect’’ plants against some of the damage caused by ﬂooding. B. ORGANIC TOXICANTS Considerate effort has been made to try to develop some new and improved methods to deal with a wide range of environmental contaminants. However, most of the procedures that are used at the present time are either very expensive or are not especially effective. One recently developed method of environmental clean-up called phytoremediation is deﬁned as the use of plants to remove, destroy, or sequester hazardous substances from the environment (Tsao, 2003). Phytoremediation of organic compounds often involves either the stimulation of microbial biodegradation in the rhizosphere, the area around the roots of plants, or the absorption and degradation of contaminants by the plant. In practice, organic environmental contaminants, such as polyaromatic hydrocarbons (PAHs) and polycyclic biphenyls (PCBs), are generally mixtures of a large number of different compounds of varying size and solubility. The various compounds in one of these mixtures have different chemical and physical properties and are degraded at different rates in the environment. Since smaller organic compounds are more likely to be at least somewhat soluble in water, they are more likely than the larger more complex organic compounds to be transported into roots. Several varieties of plants and trees can take up and degrade some organic contaminants. For example, plants with phytotransformation activity may contain nitroreductases, which are useful for degrading the explosive trinitrotoluene (TNT) and other nitroaromatics, dehalogenases for the degradation of chlorinated solvents and pesticides, and laccases that can degrade anilines such as triaminotoluene. While plants may degrade the smaller, more-water-soluble organic molecules, the larger, less-water-soluble molecules are likely to be more recalcitrant to breakdown by plants alone and to require bacteria for their remediation. The bulk soil contains around 107 to 108 bacterial cells per gram of dry weight of soil, while the bacterial population in the rhizosphere is often 100–1,000 times greater, primarily as a consequence of the exudates released by plant roots. Because rhizosphere bacteria are present in large numbers, they are better able to rapidly degrade organic compounds in their vicinity.

298

BERNARD R. GLICK

A number of different types of plants are effective at stimulating the degradation of organic molecules in the rhizosphere. Typically, these plants all have extensive and ﬁbrous roots that form an extended rhizosphere; these plants include many common grasses as well as corn, wheat, soybean, peas, and beans. At present, the largest number of sites being phytoremediated contain organic contaminants rather than metals, probably because organics are easier and less expensive than metals to remediate. However, because of their size and complexity, some organic contaminants are difﬁcult to remove from the environment. While plants grown on mixtures of organic contaminants are able to degrade some of these compounds, high concentrations of organic contaminants are often inhibitory to a plant. This is both the result of the direct inhibition of plant enzymatic processes and the synthesis by the plant of stress ethylene. Preliminary experiments that have attempted to deal with the inhibition of plant growth by high concentrations of organic contaminants indicate that phytoremediation of recalcitrant organic molecules proceeds to a much greater extent when ACC deaminase–containing plant growth-promoting bacteria are added to the soil (Huang et al., 2004a and b). C. METALS Phytoremediation of metals and other inorganic compounds generally involves the absorption and concentration of metals from the soil into the roots and shoots of the plant. A considerable amount of the focus of many of the metal phytoremediation studies undertaken in the last 10– 15 years has been the testing of a large number of different plants in an effort to identify those plants that can naturally accumulate large amounts of metals (Cunningham, 1995; Salt et al., 1995). These plants, which have been termed hyperaccumulators, are often found growing in areas with elevated metal concentrations in the soil. Unfortunately, in the presence of very high concentrations of these contaminants, hyperaccumulating plants tend to grow slowly and usually attain only a small size. That is, high concentrations of contaminants are generally inhibitory to the growth of plants, even metal-hyperaccumulating plants. Thus to completely remediate a site, even using hyperaccumulating plants, the process is usually considered to be too slow for practical application. While plants grown on metal-contaminated soils might be able to withstand some of the inhibitory effects of high concentrations of metals within a plant, at least two features of most plants could result in a decrease in plant growth and viability. In the presence of high

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

299

levels of metals most plants (1) synthesize stress ethylene and (2) become severely depleted in the amount of iron that they contain. ACC deaminase–containing plant growth-promoting bacteria might be used to relieve the some of the toxicity of metals to plants. First, the use of these organisms would be expected to decrease the level of stress ethylene in a plant growing in soil with high levels of metal. In addition, plants are able to take up and utilize complexes between bacterial siderophores—low-molecular-weight iron-chelating compounds—and iron. Bacterial siderophores bind to iron with a much higher afﬁnity than plant siderophores, so that in metal contaminated soils, bacteria can help plants to accumulate a sufﬁcient amount of iron so that their metabolism is no longer impaired. In one series of experiments designed to improve the performance of a naturally occurring metal-resistant bacterial strain that could promote plant growth, a spontaneous siderophore overproducing mutant was selected (Burd et al., 2000). The bacterium was grown on a minimal solid growth medium that did not contain any measurable amounts of iron and, in fact, contained a chemical iron chelating agent. Starting with a naturally occurring nickel-resistant plant growthpromoting bacterial strain, of the many bacteria that were added to this medium, only a few were able to grow. The bacteria that grew under these extremely iron-limited conditions contained a spontaneous mutation that caused the overproduction of the bacterial siderophore. When this mutant bacterial strain was added to the roots of plants grown in the presence of high levels of nickel, bacterial siderophore overproduction enabled the bacterium to sequester a sufﬁcient amount of iron, even though the iron was present at extremely low levels, to permit both the bacterium and the plant to grow. When the wild-type bacterium and the siderophore overproducing mutant were tested in the laboratory, both of them promoted the growth of (tomato, canola, and Indian mustard) plants in the presence of inhibitory levels of nickel, lead, or zinc. However, the siderophore overproducing mutant decreased the inhibitory effect of the added metal on plant growth signiﬁcantly more than the wild-type bacterium that produced a lower level of siderophores. When the siderophore overproducing mutant was tested in a ﬁeld that had been contaminated with nickel over a period of many years, both the number of (Indian mustard, Brassica juncea) seeds that germinated in the nickel-contaminated soil and the size that the plants were able to attain was increased by 50–100%—that is, overall there was a twofold to fourfold increase in the amount of nickel removed from the soil by the addition of the mutant compared to the wild-type bacterium.

300

BERNARD R. GLICK

D. BIOCONTROL OF PATHOGENS Phytopathogen damage that can reduce crop yields from 25% to 100% is generally dealt with by the use of chemical agents, although other treatments have also been employed. Many of the chemicals that are used to control phytopathogens are hazardous to animals and humans, and both persist and accumulate in natural ecosystems. It is therefore desirable, whenever possible, to replace these chemicals with biological control agents that are more ‘‘friendly’’ to the environment. One biological approach for the control of phytopathogenic agents is the development of transgenic plants that are resistant to one or more pathogenic agents (Greenberg and Glick, 1993). Alternatively, scientists have begun to use plant growth-promoting bacteria as biocontrol agents to suppress or prevent phytopathogen damage (Glick and Bashan, 1997; Glick et al., 1999). Plant growth-promoting bacteria may produce a variety of substances that can be used to limit the damage to plants by phytopathogens. These substances include siderophores, antibiotics, other small molecules, and a variety of enzymes that can, for example, lyse fungal cell walls. Bacterial and fungal pathogens not only directly inhibit plant growth, but they also cause the plant to synthesize stress ethylene, and, as is the case with other environmental stresses, much of the damage sustained by plants infected with phytopathogens occurs as a result of the response of the plant to the increased levels of ethylene. For example, it is well known that exogenous ethylene often increases the severity of a fungal infection, while some ethylene synthesis inhibitors signiﬁcantly decrease the severity of a fungal infection (Bashan, 1994; Biles et al., 1990; Cohen et al., 1986; Cronshaw and Pegg, 1976; Elad, 1988, 1990; Robison et al., 2001). In one series of experiments, two biocontrol bacterial strains were transformed with the Enterobacter cloacae UW4 ACC deaminase gene, and the effect of the transformed and nontransformed bacteria on the damage to cucumbers caused by Pythium ultimum was assessed (Wang et al., 2000). In all instances, the ACC deaminase–containing biocontrol bacterial strains were more effective than biocontrol strains that did not possess this enzyme. With the ACC deaminase–transformed strains, not only was the root and shoot fresh weight greater than with the wild-type biocontrol strains, but the number of seeds that germinated in pathogen-containing soil was also higher. In addition, one ACC deaminase-transformed biocontrol strain signiﬁcantly reduced the extent of soft rot of potato slices caused by the bacterial pathogen Erwinia carotovora subsp. carotovora in sealed plastic bags. This is in stark

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

301

contrast to the nontransformed biocontrol strain, which did not prevent damage to the potato slices. In these experiments, by lowering the level of stress ethylene, ACC deaminase acts synergistically with other mechanisms of biocontrol, such as the production of antibiotics or pathogenesis-related proteins, to prevent phytopathogens from damaging plants. E. FLOWER WILTING Ethylene is a key signal in initiation of the senescence of ﬂowers of most plants. For example, carnation ﬂowers produce minute amounts of ethylene until an endogenous rise (a climacteric burst) of this hormone occurs. However, ethylene does not cause senescence in all ﬂower families, and even the senescence symptoms that are caused by ethylene differ from plant to plant (Woltering and van Doorn, 1988). In this regard, Caryophyllaceae (e.g., carnations) show dramatic ethylenemediated wilting of their petals; in Rosaceae (e.g., roses), ethylene causes petal abscission; in Compositae (e.g., sunﬂowers), ethylene does not cause any detectable senescence of petals. The compound L--(aminoethoxyvinyl)-glycine (AVG), which prevents the functioning of pyroxidal phosphate-linked enzymes and is therefore an inhibitor of the enzyme ACC synthase, is a highly effective ethylene inhibitor that can delay senescence in some ﬂowers. Unfortunately, AVG is too expensive for its routine use to delay ﬂower senescence to be a practical consideration. In an effort to delay the senescence of ﬂower petals, many cut ﬂowers (e.g., carnations and lilies) are routinely treated with the ethylene inhibitor silver thiosulfate prior to their sale. However, a high silver thiosulfate concentration is potentially phytotoxic and environmentally hazardous. In our laboratory, treatment of carnation petals (carefully dissected from carnation ﬂowers) with the ethylene precursor ACC increased their rate of senescence, while treatment of carnation petals with the ethylene inhibitor AVG decreased their senescence rate. Fifty percent of untreated petals senesced in approximately 4.5 days, 50% of petals treated with ACC senesced in approximately 2 days, and 50% of petals treated with AVG senesced in approximately 9 days. Thus in this system the rate of petal senescence was directly related to the amount of ACC. On the other hand, zinnia petals even after 10 days did not senesce to any detectable extent regardless of the presence of ACC or AVG. These observations are consistent with the fact that zinnia is relatively ethylene insensitive while carnation is highly sensitive to ethylene.

302

BERNARD R. GLICK

To lower the concentration of ACC, and hence ethylene, within a ﬂower, it may be possible, as an alternative to the use of chemical ethylene inhibitors, to use ACC deaminase–containing plant growthpromoting bacteria. When carnation petals were incubated in an ACC deaminase–containing plant growth-promoting bacterial suspension, the lifetime of the petals increased by approximately 5–6 days compared to the petals of untreated ﬂowers (Nayani et al., 1998). In this experiment, 50% of the untreated petals senesced in approximately 6 days, while 50% of the bacterially treated petals senesced after 11–12 days. This is roughly equal to or greater than the level of protection that AVG afforded to carnation petals. Unfortunately, it is not at all straightforward to apply the aforementioned laboratory results with plant growth-promoting bacteria in a way that could be used to increase the life of cut ﬂowers on a large scale. In the ﬁrst instance, the bacteria that have been tested so far are all soil bacteria that bind efﬁciently to plant roots and not phyllosphere bacteria that bind more efﬁciently to the surface of leaves or ﬂowers. However, it is possible to isolate phyllosphere bacteria with appreciable levels of ACC deaminase activity (Glick et al., unpublished results), and these organisms may bind more readily to the ﬂower surface than ACC deaminase–containing rhizobacteria. Additionally, at least with carnations, these bacteria are not taken up from solution by intact ﬂowers (as opposed to isolated petals), so that merely dipping ﬂowers in a bacterial solution is not an effective way to proceed. On the other hand, there is some evidence that some plants, other than carnations, may be able to take up bacteria from solution. Thus there is reason to be optimistic that, at least in some instances, ACC deaminase–containing plant growth-promoting bacteria may eventually be used instead of chemicals to prolong the life of cut ﬂowers. F. DROUGHT It was hypothesized that plant growth-promoting bacteria endemic to locales where water is limited are likely to be able to better facilitate plant growth compared with bacteria isolated from sites where water is abundant (Mayak et al., 2004a). Based on this hypothesis, an ACC deaminase–containing bacterium Achromobacter piechaudii ARV8 was isolated from a rhizosphere soil sample from a Lycium shawii plant growing in a dry riverbed in the Arava region of Israel (i.e., the southern portion of the Negev desert). Tomato plant seedlings treated with A. piechaudii ARV8 and then subjected to a drought regimen were signiﬁcantly larger than both

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

303

untreated seedlings and seedlings treated with the ACC deaminasecontaining plant growth-promoting bacterium Pseudomonas putida GR12-2 that was originally isolated from the rhizosphere of grasses in the High Canadian Arctic (Lifshitz et al., 1986), where water is abundant. At 3 weeks of age, watering was stopped for 1 week and then resumed every other day for 1 week. At the end of the experiment (5 weeks) the mean dry weight of untreated plants was 15.6 mg; P. putida GR12-2 treated plants had a mean of 41.0 mg; A. piechaudii ARV8-treated plants were 59.7 mg. Each of these treatments yields plants of a size that was signiﬁcantly different from the other treatments. Consistent with this data and with the assumed role of bacterial ACC deaminase, the untreated tomato plant seedlings produced approximately four and a half times as much ethylene as the plants treated with A. piechaudii ARV8. The detailed mechanism(s) by which ACC deaminase–containing plant growth-promoting bacteria decrease the damage to plants that occurs under drought conditions, is not especially well understood at this time, despite the fact that the lowering of stress ethylene levels in the plant appears to be an important component of this effect. Nevertheless, the use of organisms such as A. piechaudii ARV8 to lower plant stress may be a potentially important adjunct to agricultural practice in locales where drought is endemic. G. HIGH SALT Increasing saline concentrations are known to progressively suppress the growth of plants (Cuartero and Fernandez-Munoz, 1999; Feng and Barker, 1992). However, when plants were treated with a suspension of the bacterium A. piechaudii ARV8, mentioned previously in the context of promoting the growth of drought stressed plants, the suppression of plant growth was lessened and the bacterially treated plants accumulated signiﬁcantly more fresh and dry weight than untreated plants (Mayak et al., 2004b). In fact, tomato plants irrigated with salt solutions ranging from 2.5 to 10.0 g/L produced approximately 40-50% more biomass in the presence of this added ACC deaminasecontaining plant growth-promoting bacterium. Thus this bacterium can alleviate some of the debilitating effects of salt stress. Consistent with the notion that the decrease in salt inhibition of plant growth resulted from a lowering of stress ethylene production by the plant growth-promoting bacterium A. piechaudii ARV8 was the observation that concurrent with salt stress, a rise in ethylene production occurred in tomato seedlings, and this increase in

304

BERNARD R. GLICK

ethylene production was limited by the presence of the bacterium. Increased production of ethylene in response to salt stress has been previously demonstrated in tomato and cucumber plants (Feng and Barker, 1992). Furthermore, ethylene is involved in transduction of a signal between the recognition of salt stress and a plant’s response in terms of physiological processes (Wilkinson et al., 1995). Increased salt concentration decreases the osmotic potential of a growth medium, thus reducing the water availability so that the response of a plant to salt stress is thought to be in part similar to its response to drought stress (Blumwald, 2000). In fact, the ‘‘relative water content’’ (a good indicator of water stress) of tomato plants subjected to increasing amounts of salt decreased while the ‘‘osmotic concentration’’ (another indicator of water stress) increased under the same conditions. However, the presence of the plant growth-promoting bacterium A. piechaudii ARV8 did not affect either the relative water content or the osmotic concentration. This suggests that the bacterial treatment did not inﬂuence the extent of water stress. The limited data on the effect of the plant growth-promoting bacterium A. piechaudii ARV8 on plants subjected to salt stress suggest that, without any genetic manipulation of the plant, it may be possible to productively cultivate a variety of crop plants under saline conditions, provided that the plants are grown in the presence of a suitable plant growth–promoting bacterium. IV. Modulating Nodulation of Legumes It has been known for some time that ethylene is an inhibitor of the nodulation of legumes by Rhizobia spp. (Drennan and Norton, 1972; Grobbelaar et al., 1971). Moreover, the addition of the ethylene precursor ACC to plant roots blocks nodulation (Penmesta and Cook, 1997), while the addition of the ethylene inhibitor AVG increases the number of nodules (Nukui et al., 2000). In nature, the compound rhizobitoxine (2- amino-4-(2-amino-3-hydropropoxy)-trans-but-3-enoic acid), which is synthesized by the legume symbiont Bradyrhizobium elkanii and the plant pathogen Burkholderia andropogonis (Yasuta et al., 2001), has been found to enhance nodulation and competitiveness of the legumes Macroptilium atropurpureum (siratro) and Vigna radiata (mung bean) by inhibiting endogenous ethylene synthesis in the host plant (Duodu, 1999; Yasuta et al., 1999). This compound—AVG is a structural analog of rhizobitoxine—acts by inhibiting cystathionine synthase in the methionine biosynthesis pathway and ACC synthase in the ethylene synthesis pathway.

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

305

To determine whether other microbial systems that modulate ethylene synthesis and thereby enhance nodulation might exist, a number of rhizobial strains were assayed for ACC deaminase activity. Upon surveying 13 different known rhizobial strains, it was observed that ﬁve strains had ACC deaminase activity while the others were devoid of this enzyme (Ma et al., 2003b). Interestingly, four of the ﬁve ACC deaminase positive strains were commercial strains that had previously undergone extensive biological screening/testing for nodulation and nitrogen ﬁxing activity. Only one of the ﬁve commercial strains that was tested did not have ACC deaminase activity. On the other hand, only one of the remaining eight (noncommercial) strains had ACC deaminase activity. From these limited data, one is tempted to speculate that (1) only a small fraction of rhizobial strains found in the soil contain ACC deaminase and (2) conversely, a large number of commercial rhizobial strains possess this enzyme. The tacit assumption behind all studies of ACC deaminase in Rhizobia spp. is that strains that contain this enzyme should be more efﬁcient at forming nitrogen-ﬁxing nodules than strains lacking this enzyme. To test this hypothesis, an ACC deaminase gene from a strain of Rhizobium leguminosarum that nodulates pea plants was isolated and characterized (Ma et al., 2003a). This gene was used to construct a mutant of the original strain that no longer produced ACC deaminase as well as a mutant that overproduced ACC deaminase. When these engineered mutant strains were compared with the wild-type strain for the ability to nodulate pea plants and promote plant growth, it was found that the ACC deaminase minus strain of R. leguminosarum formed approximately 30% fewer nodules than the wild-type strain and also produced approximately 30% less plant biomass. On the other hand, pea plants treated with the ACC deaminase overproducing strain behaved identically to those treated with the wild-type strain. In another series of experiments, a strain of Sinorhizobium meliloti that was devoid of ACC deaminase activity was transformed to express a R. leguminosarum ACC deaminase gene and its regulatory region (Ma et al., 2004). The transformed bacterium expressed ACC deaminase activity and produced nearly 40% more nodules and biomass than the nontransformed strain when it was used to treat alfalfa seedlings. This result is potentially important for agricultural practice and suggests that rhizobial strains that are intended for use as inoculants of host legumes should ﬁrst be selected/tested for the presence of a functional ACC deaminase (Penrose and Glick, 2003). It should be noted that to date all of the rhizobial strains that have been found to produce active ACC deaminase contain a relatively low

306

BERNARD R. GLICK

level of this activity (approximately 2–8%) compared with the amount of enzyme activity generally found in free-living soil bacteria (Ma et al., 2003). However, upon infection of root hairs by rhizobial strains, only low levels of ethylene are produced, and this tends to be localized to the site of the rhizobial infection. In comparison, free-living bacteria that contain ACC deaminase often act as a sink for relatively high amounts of ACC produced as a consequence of an environmental stress that affects all or large parts of the plant. Thus, despite having an apparently low level of ACC deaminase activity, rhizobial strains have a sufﬁcient amount of this enzyme to prevent ethylene inhibition of nodulation that might otherwise occur as a consequence of the infection process. Although it has not yet been tested, an intriguing possibility is that ACC deaminase–containing plant growth-promoting bacteria might act to promote the formation of arbuscular mycorrhizal symbioses with plants (Guinel and Geil, 2002). This could happen as a result of the added bacterium decreasing the amount of ethylene that forms following infection of a plant by mycorrhizae. Thus, ACC deaminasecontaining plant growth-promoting bacteria and mycorrhizae might act synergistically to promote plant growth; however, this conjecture remains to be tested. V. Decreasing Stress in ACC Deaminase Transgenic Plants Several different plants have been genetically engineered to express bacterial ACC deaminase genes under the control of plant promoters. In the earliest examples of this work the bacterial gene was expressed in tomato plants under the control of the constitutive 35S promoter from cauliﬂower mosaic virus (Klee et al., 1991; Sheehy et al., 1991) with the objective of delaying the ethylene-catalyzed fruit ripening process. More recently, transgenic plants expressing ACC deaminase have been shown to increase their resistance to a variety of fungal pathogens (Lund et al., 1998; Robison et al., 2001) as well as to a number of different stresses, including metals (Grichko et al., 2000; Nie et al., 2002; Stearns et al., unpublished observations), ﬂooding (Grichko and Glick, 2001b), and high salt (Sergeeva et al., unpublished observations). In these experiments, transgenic plants that are inhibited to the least extent by the stressor are those in which the ACC deaminase gene is under the transcriptional control of the rolD (root speciﬁc) promoter. Interestingly, for all of the environmental stresses that have been tested so far, nontransgenic plants that were treated with ACC deaminase–containing plant growth–promoting bacteria were as

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

307

resistant to the inhibitory effects of the stress as were transgenic plants expressing the same enzyme. Thus, from this limited number of experiments, there does not seem to be any intrinsic advantage of expressing ACC deaminase within a transgenic plant, as opposed to within a soil bacterium that is able to bind to plant roots. VI. Conclusions The successful demonstrations of the efﬁcacy of using ACC deaminase-containing plant growth-promoting bacteria to obviate the deleterious effects of a number of environmental stresses in laboratory and greenhouse settings notwithstanding, there has been a reluctance to adopt the use of these bacteria on a large scale in the ﬁeld. There are several reasons why this is the case. In the more developed countries of the world, agricultural chemicals are relatively inexpensive, effective, and easy to apply, while the use of microbial inoculants is more labor intensive and the technology is considered to be unproven. On the other hand, the use of microbial inoculants is viewed as a more appropriate technology in many of the poorer and less-developed countries of the world where agricultural chemicals are relatively expensive and labor is inexpensive. A few plant growth-promoting bacterial strains with different properties as well as several types of biocontrol bacteria, are currently commercially available and are being used to increase crop yields in, for example, Russia, China and Latin America. Given the reluctance on the part of many consumers worldwide to use as foods plants that have been genetically modiﬁed, for the foreseeable future it may be advantageous to use either natural or genetically engineered plant growth-promoting bacteria as a means to promote growth by lowering plant ethylene levels or reduce disease through induction of resistance rather than genetically modifying the plant itself to the same end. Moreover, given the large number of different plants, the various cultivars of those plants and the multiplicity of genes that would need to be engineered into plants, at this time it is not feasible to genetically engineer all plants to be resistant to all pathogens and environmental stresses. Rather, it makes a lot more sense to either select or engineer plant growth-promoting bacteria to do this job.

ACKNOWLEDGMENTS The work from the author’s laboratory that is described in this review was funded by grants from the Natural Sciences and Engineering Research Council of Canada. Thanks are also due to the following collaborators and students including (in alphabetical order):

308

BERNARD R. GLICK

Yoav Bashan, Andrei Belimov, Genrich Burd, Genevieve De´fago, D. George Dixon, Sibdas Ghosh, Bruce Greenberg, Varvara Grichko, Fre`de`rique Guinel, Jeremy Hall, Gina Holguin, Nikos Hontzeas, XiaoDong Huang, Christian Jacobson, Jiping Li, Wenbo Ma, Shimon Mayak, Barbara Moffat, Jack Pasternak, Cheryl Patten, Donna Penrose, Saleh Shah, Jennifer Stearns, Tsipi Tirosh, and Chunxia Wang.

REFERENCES Abeles, F. B., Morgan, P. W., and Saltveit, M. E., Jr. (1992). ‘‘Ethylene in Plant Biology,’’ 2nd ed. Academic Press, New York. Arshad, M., and Frankenberger, W. T. (2002). ‘‘Ethylene: Agricultural Sources and Applications.’’ Kluwer Academic/Plenum Publishers, New York. Bashan, Y. (1994). Symptom expression and ethylene production in leaf blight of cotton caused by Alternaria macrospora and Alternaria alternata alone and combined. Can. J. Bot. 72, 1574–1579. Bayliss, C., Bent, E., Culham, D. E., MacLellan, S., Clarke, A. J., Brown, G. L., and Wood, J. M. (1997). Bacterial genetic loci implicated in the Pseudomonas putida GR122R3–canola mutualism: Identiﬁcation of an exudate-inducible sugar transporter. Can. J. Microbiol. 43, 809–818. Belimov, A. A., Safranova, V. I., Sergeyeva, T. A., Egorova, T. N., Matveyeva, V. A., Tsyganov, V. E., Borisov, A. Y., Tikhonovich, I. A., Kluge, C., Preisfeld, A., Dietz, K.-J., and Stepanok, V. V. (2001). Characterization of plant growth promoting rhizobacteria isolated from polluted soils and containing 1- aminocyclopropane-1-carboxylate deaminase. Can. J. Microbiol. 47, 642–652. Belimov, A. A., Hontzeas, N., Safronova, V. I., Dodd, I. C., Demchinskaya, S. V., Piluzza, G., Bullitta, S., Davies, W. J., and Glick, B. R. (2004). Cadmium tolerant root growthpromoting bacteria of the rhizoplane of Indian mustard (Brassica juncea L. Czern.). Submitted for publication. Biles, C. L., Abeles, F. B., and Wilson, C. L. (1990). The role of ethylene in anthracnose of cucumber, Cucumis sativus, caused by Colletotrichum lagenarium. Phytopathology 80, 732–736. Blumwald, E. (2000). Sodium transport and salt tolerance in plants. Curr. Opin. Cell Biol. 12, 431–434. Bradford, K. J., and Yang, S. F. (1980). Xylem transport of 1-aminocyclopropane-1-carboxylic acid, an ethylene precursor, in waterlogged tomato plants. Plant Physiol. 65, 322–326. Burd, G. I., Dixon, D. G., and Glick, B. R. (1998). A plant growth promoting bacterium that decreases nickel toxicity in plant seedlings. Appl. Environ. Microbiol. 64, 3663–3668. Burd, G. I., Dixon, D. G., and Glick, B. R. (2000). Plant growth-promoting bacteria that decrease heavy metal toxicity in plants. Can. J. Microbiol. 46, 237–245. Campbell, B. G., and Thompson, J. A. (1996). 1-Aminocyclopropane-1-carboxylate deaminase genes from Pseudomonas strains. FEMS Microbiol. Lett. 138, 207–210. Cohen, R., Riov, J., Lisker, N., and Katan, J. (1986). Involvement of ethylene in herbicide-induced resistance to Fusarium oxysporum f. sp. melonis. Phytopathol. 76, 1281–1285. Cronshaw, D. K., and Pegg, G. F. (1976). Ethylene as a toxin synergist in Verticillium wilt of tomato. Physiol. Plant Pathol. 9, 33–38. Cuartero, J., and Fernandez-Munoz, R. (1999). Tomato and salinity. Scientia Hortic. 78, 83–125.

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

309

Cunningham, S. D., Berti, W. R., and Huang, J. W. (1995). Phytoremediation of contaminated soils. Trends Biotechnol. 13, 393–397. Drennan, D. S., and Norton, C. (1972). The effect of ethrel on nodulation in Pisum sativum. Plant Soil 36, 53–57. Duodu, S., Bhuvaneswari, T. V., Stokkermans, T. J., and Peters, N. K. (1999). A positive role for rhizobitoxine in Rhizobium-legume symbiosis. Mol. Plant-Microbe Interact. 12, 1082–1089. Elad, Y. (1988). Involvement of ethylene in the disease caused by Botrytis cinerea on rose and carnation ﬂowers and the possibility of control. Ann. Appl. Biol. 113, 589–598. Elad, Y. (1990). Production of ethylene in tissues of tomato, pepper, French-bean and cucumber in response to infection by Botrytis cinerea. Physiol. Mol. Plant Pathol. 36, 277–287. Else, M. A., and Jackson, M. B. (1988). Transport of 1-aminocyclopropane-1-carboxylic acid (ACC) in the transpiration stream of tomato (Lycopersicon esculentum) in relation to foliar ethylene production and petiole epinasty. Aust. J. Plant Physiol. 25, 453–458. Fallik, E., Sarig, S., and Okon, Y. (1994). Morphology and physiology of plant roots associated with Azospirillum. In ‘‘Azospirillum/Plant Associations’’ (Y. Okon, ed.), pp. 77–85. CRC Press, Boca Raton, FL. Feng, J., and Barker, A. V. (1992). Ethylene evolution and ammonium accumulation by tomato plants under water and salinity stresses. II. J. Plant Nutr. 15, 2471–2490. Frankenberger, W. T., Jr., and Arshad, M. (1995). ‘‘Phytohormones in Soil.’’ Marcel Dekker, New York. Ghosh, S., Penterman, J. N., Little, R. D., Chavez, R., and Glick, B. R. (2003). Three newly isolated plant growth-promoting bacilli facilitate the growth of canola seedlings. Plant Physiol. Biochem. 41, 277–281. Glick, B. R. (1995). The enhancement of plant growth by free-living bacteria. Can. J. Microbiol. 41, 109–117. Glick, B. R., and Bashan, Y. (1997). Genetic manipulation of plant growth-promoting bacteria to enhance biocontrol of fungal phytopathogens. Biotechnol. Adv. 15, 353–378. Glick, B. R., Karaturovı´c, D., and Newell, P. (1995). A novel procedure for rapid isolation of plant growth-promoting rhizobacteria. Can. J. Microbiol. 41, 533–536. Glick, B. R., Patten, C. L., Holguin, G., and Penrose, D. M. (1999). ‘‘Biochemical and Genetic Mechanisms Used by Plant Growth Promoting Bacteria.’’ Imperial College Press, London. Glick, B. R., Penrose, D. M., and Li, J. (1998). A model for the lowering of plant ethylene concentrations by plant growth promoting bacteria. J. Theor. Biol. 190, 63–68. Greenberg, B. M., and Glick, B. R. (1993). The use of recombinant DNA technology to produce genetically modiﬁed plants. In ‘‘Methods in Plant Molecular Biology and Biotechnology.’’ (B. R. Glick and J. E. Thompson, eds.), pp. 1–10. CRC Press, Boca Raton, FL. Grichko, V. P., and Glick, B. R. (2000). Identiﬁcation of DNA sequences that regulate the expression of the Enterobacter cloacae UW4 1-aminocyclopropane-1-carboxylate deaminase gene. Can. J. Microbiol. 46, 1159–1165. Grichko, V. P., and Glick, B. R. (2001a). Amelioration of ﬂooding stress by ACC deaminase-containing plant growth-promoting bacteria. Plant Physiol. Biochem. 39, 11–17. Grichko, V. P., and Glick, B. R. (2001b). Flooding tolerance of transgenic tomato plants expressing the bacterial enzyme ACC deaminase controlled by the 35S, rolD or PRB-1b promoter. Plant Physiol. Biochem. 39, 19–25.

310

BERNARD R. GLICK

Grichko, V. P., Filby, B., and Glick, B. R. (2000). Increased ability of transgenic plants expressing the bacterial enzyme ACC deaminase to accumulate Cd, Co, Cu, Ni, Pb and Zn. J. Biotechnol. 81, 45–53. Grobbelaar, N., Clarke, B., and Hough, M. C. (1971). The nodulation and nitrogen ﬁxation of isolated roots of Phaseolus vulgaris L. III. The effect of carbon dioxide and ethylene. Plant Soil, (spec. vol.) 214–223. Guinel, F. C., and Geil, R. D. (2002). A model for the development of the rhizobial and arbuscular mycorrhizal symbioses in legumes and its use to understand the roles of ethylene in the establishment of these two symbioses. Can. J. Bot. 80, 695–720. Hall, J. A., Peirson, D., Ghosh, S., and Glick, B. R. (1996). Root elongation in various agronomic crops by the plant growth promoting rhizobacterium Pseudomonas putida GR12-2. Isr. J. Plant Sci. 44, 37–42. Honma, M., Kawai, J., and Yamada, M. (1993). Identiﬁcation of the reactive sulfhydryl group of 1-aminocyclopropane-1-carboxylate deaminase. Biosci. Biotechnol. Biochem. 57, 2090–2093. Huang, X.-D., El-Alawi, Y., Penrose, D. M., Glick, B. R., and Greenberg, B. M. (2004a). Responses of three grass species to creosote during phytoremdiation. Environ. Pollut. 130, 453–463. Huang, X.-D., El-Alawi, Y., Penrose, D. M., Glick, B. R., and Greenberg, B. M. (2004b). A multi-process phytoremediation system for removal of polycyclic aromatic hydrocarbons from contaminated soils. Environ. Pollut., 130, 465–476. Hyodo, H. (1991). Stress/wound ethylene. In ‘‘The Plant Hormone Ethylene’’ (A. K. Mattoo and J. C. Suttle, eds.), pp. 65–80. CRC Press, Boca Raton, FL. Jacobson, C. B., Pasternak, J. J., and Glick, B. R. (1994). Partial puriﬁcation and characterization of ACC deaminase from the plant growth-promoting rhizobacterium Pseudomonas putida GR12-2. Can. J. Microbiol. 40, 1019–1025. Jia, Y.-J., Kakuta, Y., Sugawara, M., Igarashi, T., Oki, N., Kisaki, M., Shoji, T., Kanetuna, Y., Horita, T., Matsui, H., and Honma, M. (1999). Synthesis and degradation of 1aminocyclopropane-1-carboxylic acid by Penicillium citrinum. Biosci. Biotechnol. Biochem. 63, 542–549. Kende, H. (1993). Ethylene biosynthesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 44, 283–307. Klee, H. J., Hayford, M. B., Kretzmer, K. A., Barry, G. F., and Kishore, G. M. (1991). Control of ethylene synthesis by expression of a bacterial enzyme in transgenic tomato plants. Plant Cell 3, 1187–1193. Li, J., and Glick, B. R. (2001). Transcriptional regulation of the Enterobacter cloacae UW4 1-aminocyclopropane-1-carboxylate (ACC) deaminase gene (acdS). Can. J. Microbiol. 47, 259–267. Lund, S. T., Stall, R. E., and Klee, H. J. (1998). Ethylene regulates the susceptible response to pathogen infection in tomato. Plant Cell 10, 371–382. Lifshitz, R., Kloepper, J. W., Scher, F. M., Tipping, E. M., and Laliberte, M. (1986). Nitrogen-ﬁxing pseudomonads isolated from roots of plants grown in the Canadian High Arctic. Appl. Environ. Microbiolol. 51, 251–255. Ma, W., Charles, T. C., and Glick, B. R. (2004). Expression of an exogenous 1-aminocyclopropane-1-carboxylate deaminase gene in Sinorhizobium meliloti increases its ability to nodulate alfalfa. Appl. Environ. Microbiol., in press. Ma, W., Guinel, F. C., and Glick, B. R. (2003a). Rhizobium leguminosarum biovar viciae 1-aminocyclopropane-1-carboxylate deaminase promotes nodulation of pea plants. Appl. Environ. Microbiol. 69, 4396–4402.

ACC DEAMINASE AND THE ALLEVIATION OF PLANT STRESS

311

Ma, W., Sebestianova, S., Sebestian, J., Burd, G. I., Guinel, F., and Glick, B. R. (2003b). Prevalence of 1-aminocyclopropaqne-1-carboxylate in deaminase in Rhizobia spp. Anton. Van Leeuwenhoek 83, 285–291. Mattoo, A. K., and Suttle, C. S. (eds.) (1991). ‘‘The Plant Hormone Ethylene.’’ CRC Press, Boca Raton, FL. Mayak, S., Tirosh, T., and Glick, B. R. (2004a). Plant growth-promoting bacteria that confer resistance to water stress in tomatoes and peppers. Plant Sci., 166, 525–530. Mayak, S., Tirosh, T., and Glick, B. R. (2004b). Bacteria that confer resistance in tomato to salt stress. Plant Physiol. Biochem. 42, 565–572. Minami, R., Uchiyama, K., Murakami, T., Kawai, J., Mikami, K., Yamada, T., Yokoi, D., Ito, H., Matsui, H., and Honma, M. (1998). Properties, sequence, and synthesis in Escherichia coli of 1-aminocyclopropane-1-carboxylate deaminase from Hansenula saturnus. J. Biochem. 123, 1112–1118. Nayani, S., Mayak, S., and Glick, B. R. (1998). The effect of plant growth promoting rhizobacteria on the senescence of ﬂower petals. Ind. J. Exptl. Biol. 36, 836–839. Nie, L., Shah, S., Burd, G. I., Dixon, D. G., and Glick, B. R. (2002). Phytoremediation of arsenate contaminated soil by transgenic canola and the plant growth-promoting bacterium Enterobacter cloacae CAL2. Plant Physiol. Biochem. 40, 355–361. Nukui, N., Ezura, H., Yuhashi, K. I., Yasuta, T., and Minamisawa, K. (2000). Effects of ethylene precursor and inhibitors for ethylene biosynthesis and perception and nodulation in Lotus japonicus and Macroptilium atropurpureum. Plant Cell Physiol. 41, 893–897. Olson, D. C., Oetiker, J. H., and Yang, S. F. (1995). Analysis of LE-ACS3, a 1-aminocyclopropane-1-carboxylic acid synthase gene expressed during ﬂooding in the roots of tomato plants. J. Biol. Chem. 270, 14056–14061. Patten, C., and Glick, B. R. (1996). Bacterial biosynthesis of indole-3-acetic acid. Can. J. Microbiol. 42, 207–220. Penmesta, R. V., and Cook, D. R. (1997). A legume ethylene-insensitive mutant hyperinfected by its rhizobial symbiont. Science 275, 527–530. Penrose, D. M., and Glick, B. R. (2001). Levels of 1-aminocyclopropane-1-carboxylic acid (ACC) in exudates and extracts of canola seeds treated with plant growth-promoting bacteria. Can. J. Microbiol. 47, 368–372. Penrose, D. M., and Glick, B. R. (2003). Methods for isolating and characterizing ACC deaminase–containing plant growth–promoting rhizobacteria. Physiol. Plant. 118, 10–15. Robison, M. M., Shah, S., Tamot, B., Pauls, K. P., Moffatt, B. A., and Glick, B. R. (2001). Reduced symptoms of Verticillium wilt in transgenic tomato expressing a bacterial ACC deaminase. Molec. Plant Pathol. 2, 135–145. Salt, D. E., Blaylock, M., Kumar, N. P. B. A., Dushenkov, V., Ensley, B. D., Chet, I., and Raskin, I. (1995). Phytoremediation: a novel strategy for the removal of toxic metals from the environment using plants. Bio/Technology 13, 468–474. Sisler, E. C., and Serek, M. (1997). Inhibitors of ethylene responses in plants at the receptor level: Recent developments. Physiol. Plant. 100, 577–582. Shah, S., Li, J., Moffatt, B. A., and Glick, B. R. (1998). Isolation and characterization of ACC deaminase genes from two different plant growth promoting rhizobacteria. Can. J. Microbiol. 44, 833–843. Sheehy, R. E., Honma, M., Yamada, M., Sasaki, T., Martineau, B., and Hiatt, W. R. (1991). Isolation, sequence, and expression in Escherichia coli of the Pseudomonas sp. strain ACP gene encoding 1-aminocyclopropane-1-carboxylate deaminase. J. Bacteriol. 173, 5260–5265.

312

BERNARD R. GLICK

Stearns, J., and Glick, B. R. (2003). Transgenic plants with altered ethylene biosynthesis or perception. Biotechnol. Adv. 21, 193–210. Tsao, D. T. (ed.) (2003). ‘‘Phytoremediation.’’ Series: Advances in Biochemical Engineering/Biotechnology, vol. 78, Springer-Verlag, New York. Ton, J., Davison, S., van Wees, S. C. M., van Loon, L. C., and Pieterse, C. M. J. (2001). The Arabidopsis ISR1 locus controlling rhizobacteria-mediated induced systemic resistance is involved in ethylene signaling. Plant Physiol. 125, 652–661. van Loon, L. C., Bakker, P. A. H. M., and Pieterse, C. M. J. (1997). Mechanisms of PGPRinduced resistance against pathogens. In ‘‘Plant Growth-Promoting Rhizobacteria: Present Status and Future Prospects’’ (A. Ogoshi, K. Kobayashi, Y. Homma, F. Kodama, N. Kondo, and S. Akino, eds.), pp. 50–57. OECD, Paris. van Loon, L. C., and Glick, B. R. (2004). Increased plant ﬁtness by rhizobacteria. In ‘‘Molecular Ecotoxicology of Plants’’ (H. Sandermann, ed.), pp. 177–205. SpringerVerlag, Berlin. van Loon, L. C. (1984). Regulation of pathogenesis and symptom expression in diseased plants by ethylene. In ‘‘Ethylene: Biochemical, Physiological and Applied Aspects’’ (Y. Fuchs and E. Chalutz, eds.), pp. 171–180. Martinus Nijhoff/Dr W. Junk, The Hague. Wang, C., Knill, E., Glick, B. R., and De´Fago, G. (2000). Effect of transferring 1-aminocyclopropane-1-carboxylic acid (ACC) deaminase genes into Pseudomonas ﬂuorescens strain CHAO and its gacA derivative CHA96 on their growth-promoting and diseasesuppressive capacities. Can. J. Microbiol. 46, 898–907. Wang, C., Ramette, A., Punjasamarnwong, P., Zala, M., Natsch, A., Moe¨nne-Loccoz, Y., and De´fago, G. (2001). Cosmopolitan distribution of phlD-containing dicotyledonous crop associated biological control pseudomonads of worldwide origin. FEMS Microbiol. Ecol. 37, 105–116. Watkins, C. B., and Miller, W. B. (2003). Implications of 1-methylcyclopropene registration for use on horticultural crops. In ‘‘Biology and Biotechnology of the Plant Hormone Ethylene III’’ (M. Vendrell, H. Klee, J. C. Pech, and F. Romojaro, eds.), pp. 149–150. IOS Press, Amsterdam. Whipps, J. M. (1990). Carbon utilization. In ‘‘The Rhizosphere’’ (J. M. Lynch, ed.), pp. 59–97. Wiley Interscience, Chichester, UK. Wilkinson, J. Q., Lanahan, M. B., Yen, H. C., Giovannoni, J. J., and Klee, H. J. (1995). An ethylene-inducible component of signal transduction encoded by never-ripe. Science 270, 1807–1809. Woltering, E. J., and van Doorn, W. G. (1988). Role of ethylene in senescence of petals— morphological and taxonomical relationships. J. Exp. Bot. 39, 1605–1616. Xie, H., Pasternak, J. J., and Glick, B. R. (1996). Isolation and characterization of mutants of the plant growth-promoting rhizobacterium Pseudomonas putida GR12-2 that overproduce indoleacetic acid. Curr. Microbiol. 32, 67–71. Yasuta, T., Okazaki, S., Mitsui, H., Yuhashi, K.-I., Ezura, H., and Minamisawa, K. (2001). DNA sequence and mutational analysis of rhizobitoxine production in Bradyrhizobium elkanii. Appl. Environ. Microbiol. 67, 4999–5009. Yasuta, T., Satoh, S., and Minamisawa, K. (1999). New assay for rhizobitoxine based on inhibition of 1-aminocyclopropane-1-carboxylate synthase. Appl. Environ. Microbiol. 65, 849–852.

Uses of Trichoderma spp. to Alleviate or Remediate Soil and Water Pollution G. E. HARMAN,*,} M. LORITO,{

AND

J. M. LYNCH{

*Departments of Horticultural Sciences and Plant Pathology Cornell University, Geneva, New York 14456 {

Dipartimento di Arboricoltura, Botanica e Patologia Vegetale Sezione di Patologia Vegetale, Universita degli Studi di Napoli Federico II e Istituto CNR-IPP sez. Portici, 100-80055 Portici (Napoli), Italy {

Forest Research, Alice Holt Lodge, Farnham, Surrey GU10 4LH United Kingdom }

Author for correspondence. E-mail: [email protected]

I. II. III. IV. V.

Introduction Trichoderma spp. Are Opportunistic Plant Symbionts Rhizosphere Competence and Co-Metabolism Root Enhancement by Trichoderma spp. Enhanced Extraction and Biodegradation of Toxicants A. Enhanced Plant Removal of Toxicants B. Degradation of Pollutants VI. Conclusions and Future Prospects References

313 314 315 316 316 318 321 326 327

I. Introduction Trichoderma spp. are very common in soil and on plant roots; they are among the most abundant culturable fungi in many soils. They also colonize plant debris, and in some cases the sexual stage (genus Hypocrea) is found on twigs or other similar materials (Harman et al., 2004a; Klein and Eveleigh, 1998). The fungi in this genus are genetically quite diverse, with a number of different capabilities between different strains. It is the purpose of this chapter to consider how these proliﬁc fungi may be used in remediation or amelioration of pollutants in the environment, with foci on mechanisms and capabilities of the fungi and their emergence as novel and useful tools to improve agriculture and environmental quality. Other reviews (Harman, 2000; Harman et al., 2004a; Lynch, 2003) examined the agricultural and biocontrol applications of these fungi including their abilities to control plant diseases and to enhance plant growth and the uses and nature of their economically important enzymes (Harman and Kubicek, 1998), and so these uses will be considered only in passing here. 313 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

314

HARMAN et al.

II. Trichoderma spp. Are Opportunistic Plant Symbionts Recently these fungi have been demonstrated to be opportunistic avirulent plant symbionts (Harman et al., 2004a). They clearly are opportunistic, since they can proliferate, compete, and survive in soil and other complex ecosystems. They are capable of invading roots but are typically restricted to the outer layers of the cortex (Yedidia et al., 1999), probably because of production by the fungi of several classes of compounds that act as signals for the plant to activate resistance responses based on chemical and structural mechanisms (Harman et al., 2004a). This root infection, followed by limitation of fungal proliferation within the root, allows the fungi to grow and develop by using the energy sources of the plant. Not only do the fungi grow based on resources provided by the plant, but they also are carried through soil and occupy new soil niches as a consequence of root colonization. Thus root-associated Trichoderma spp. derive numerous beneﬁts from plants. Plants also derive numerous advantages from root colonization by these opportunistic root symbionts. These include the following: . Protection of plants against diseases by direct action of the Trichoderma strains on pathogenic microbes (Chet, 1987) or other deleterious soil microﬂora (Bakker and Schippers, 1987). . Protection against plant pathogens because of systemic induction of resistance; this permits plants to be protected widely separated temporally or spatially from application of Trichoderma (Bigirimana et al., 1997; Harman et al., 2004a; Yedidia et al., 1999, 2000, 2003). For example, through induced resistance, Trichoderma spp. can control foliar pathogens even when it is present only on the roots. . Enhancement of plant growth and development, especially of roots. The activity of Trichoderma spp. added to soil increases plant growth and development. This fact seems counterintuitive, since no doubt the root colonization and induction of resistance is energetically expensive to the plants, but it is a phenomenon that is commonly observed on a variety of plants (Chang et al., 1986; Harman, 2000; Lindsey and Baker, 1967). Some of this improved plant growth no doubt occurs as a consequence of control of pathogenic or other deleterious microbes, but it also has been demonstrated in axenic systems (Lindsey and Baker, 1967; Yedidia et al., 2001), so it is no doubt a consequence of direct effects on plants as well as a biological control phenomenon (Harman et al., 2004b).

POLLUTION REMEDIATION USING TRICHODERMA SPP.

315

These facts directly demonstrate that Trichoderma spp. have a strong beneﬁcial effect upon plants. Thus, at least some strains function as plant symbionts. This is a strain-speciﬁc ability, however, since some strains in some conditions may produce toxic metabolites, and so the balance between toxicants and growth-promoting effects determines their net effect (Ousley et al., 1993). However, with other strains, negative effects usually are not seen regardless of inoculum level or environmental conditions (Harman, 2000). It should be noted that root-colonizing Trichoderma strains are not the only organisms that provide similar beneﬁts. For example, the PGPR (plant growth promoting rhizobacteria), including strains of Pseudomonas and Bacillus spp. (Kloepper et al., 1993; Ryu et al., 2003), both induce systemic resistance and enhance plant growth. Several other fungi—including nonpathogenic strains of Fusarium and Rhizoctonia spp., mycorrhizal fungi, and Penicillium spp.—may colonize superﬁcial layers of roots and induce systemic resistance (Fravel et al., 2003; Hwang and Benson, 2003; Pozo et al., 2002). This suggests that the ability to (1) infect plant roots, (2) induce the plants to limit the level of infection and induce generalized resistance mechanisms in the plant, and (3) enhance plant growth and development evolved independently numerous times within different fungal genera and is a useful survival strategy (Harman et al., 2004a).

III. Rhizosphere Competence and Co-Metabolism A few strains of Trichoderma are highly rhizosphere competent, but most strains are not (Chao et al., 1986). The most effective strains can colonize roots of virtually all plant species, as exempliﬁed by T. harzianum strain T22, and this has been tested thousands of times in academic and commercial trials (Harman, 2000). The spores of the fungus, or other forms of biomass, can be added in any way that the fungi come into contact with roots. Once this occurs, T22 proliferates over the entire root system and grows with roots as they penetrate into soil (Harman, 2000). Thus the fungus is carried deeply within the soil proﬁle even when applied only to the upper soil layer. Moreover, the addition of T22 or other strains results in denser root growth and may increase deep rooting by as much as twofold. In this way, the fungus enhances root exploitation of the soil volume both laterally and vertically, which also increases the volume of soil exposed to the fungus (Harman, 2000). Finally, root colonization occurs across a wide range of soil types (Harman, 2000; Harman and Bjo¨rkman, 1998), perhaps

316

HARMAN et al.

because the fungus exists within the outer layers of roots and therefore is less exposed to adverse soil conditions. An important part of the ecological success of Trichoderma spp. relies upon their biocontrol abilities. As the levels of T22 in soil are increased, the fungus becomes more and more dominant on roots, thus excluding other microbes (Harman, 2000); the reliability of this ability of T22 has been proven in more than a decade of use in commercial horticulture and agronomy (Harman, 2000). Thus T22 and other strains of Trichoderma form robust and stable, self-organizing symbiotic root-microbial systems. While many other microbes are known to colonize roots, we are unaware of any that are as reliable in their performance. Trichoderma spp. can grow on and in plant roots and derive nutrients from plants. At the same time, the fungus provides beneﬁts to the plant (Harman, 2004a). This creates a co-metabolic system that probably has substantial potential for alleviation of soil pollution problems (Fig. 1). IV. Root Enhancement by Trichoderma spp. T. harzianum strain T22 now is widely used in plant agriculture, both for its abilities to control plant diseases and to increase plant root growth (Harman, 2000). The organism applied even as a seed treatment can provide long-term beneﬁts to plants. An example of increased root growth caused by a seed treatment on maize and on soybeans from grower ﬁelds is shown in Fig. 2. Root colonization from seed treatments typically results in about one order of magnitude greater colonization (colony forming units per g dry weight of roots) than occurs natually from the native strains in soils (Harman and Bjo¨rkman, 1998). This level of colonization can be increased by additions of conidia or other inoculum to the developing root system. V. Enhanced Extraction and Biodegradation of Toxicants Microbes with capabilities to degrade or remove toxicants are known in many genera. However, a common failing in the use of some of these is that the microbes used frequently are overtaken or outgrown by other microbes, especially when glucose or other nutrients are added. This may necessitate their use in highly controlled conditions, such as a bioreactor. This may increase costs associated with removal of polluted materials, sterilization, and media components. The Trichoderma root

POLLUTION REMEDIATION USING TRICHODERMA SPP.

317

FIG. 1. Detoxiﬁcation of cyanide by Trichoderma spp. growing on plant seeds and roots; an example of the utility of rhizosphere competence and co-metabolism. Soils were amended or not amended with various levels of KCN, as noted, and planted to wheat seeds. Even at 50 g/g of cyanide, wheat seeds were killed in the absence of Trichoderma, but in its presence, the seedling survived even at 100 g/g. Trichoderma in this case grew on the surfaces of germinating seeds and roots and degraded cyanide as indicated by its disappearance from soil. The fungus produces at least two separate cyanide-degrading enzymes, rhodanese and formamide hydrolase (Ezzi, 2002, 2003; see text also). The image is from M. Ezzi and J. M. Lynch, unpublished.

system, especially in the case of strains such as T22, is sufﬁciently stable that such issues can largely be overcome. Moreover, since the fungus-root association is stable and the fungi actually colonize root surfaces, there is an exchange of bioactive molecules between the fungus and the plant. The fungi gain their nutrients from the plant and produce a series of disease resistance signaling molecules that have a strong effect upon plant metabolism (Harman et al., 2004a) and provide the basis for this symbiotic co-metabolic plant-fungus association. Other molecules—for example, enzymes that degrade toxicants in soils or water—also can be produced from this synergistic platform (Fig. 1). This co-metabolic system, composed of both plant and fungal metabolites, including perhaps some factors uniquely synthesized only during the interaction, may represent an

318

HARMAN et al.

FIG. 2. Enhancement of root growth of maize and soybeans by a seed treatment with T. harzianum strain T22 in grower ﬁelds. Enhancement of root growth occurs over at least the life of annual crops and is a major component that leads to yield increases and its commercial use. Figure from Harman (2000), with permission.

advantage for applications in bioremediation systems for cleanup for strategies either on accumulation of toxicants in plant tissues or for breakdown of toxic compounds.

A. ENHANCED PLANT REMOVAL OF TOXICANTS 1. Arsenic Arsenic was applied to orchards and other farm lands as a pesticide and herbicide between approximately 1930 and 1960 in various forms, most commonly lead arsenate (Woolson, 1975). Lead concentrations in at least some sites do not rise to actionable levels, but arsenic more frequently is of concern. Typical values in old orchard soils range between 30 and 150 g/g of soil, while eastern U.S. background levels are 7.5 to 12 g/g. Assessments of risks to human health posed by arsenic-polluted soils suggest that levels above about 20 ppm of bioavailable arsenic may be a health hazard. This is an emerging problem; many hundreds of thousands of hectares are affected, and so this is likely to become a serious problem for farmers and developers. Other sites also are polluted with this metalloid as a result of smelting operations and wood-treating sites and leaching from treated lumber used for landscaping, playgrounds, and other purposes. There are few low-cost and effective solutions for the cleanup of relatively low but toxic levels (e.g., 20–150 g/g) of arsenic-polluted soils. However, ferns have been identiﬁed that hyperaccumulate arsenic in their fronds from soils at levels between 2 and 200 times the level in soils (Ma et al.,

POLLUTION REMEDIATION USING TRICHODERMA SPP.

319

2001a,b). Commercial fern growers routinely treat very small fern plantlets with T22 because it enhances growth and survival. Such ferns are being commercialized by Edenspace Systems Corporation TM (Edenferns ). In the summer of 2003, we purchased ferns for ﬁeld trials in an arsenic-polluted site that were, of course, already treated with T22. Soon after we transplanted the ferns to the ﬁeld, we applied T22 as a drench at the rate of 9 kg/ha (the product contained about 109 cfu/g of T22). About 6 weeks later we sampled the roots. The level of Trichoderma spp. on roots of ferns that did not receive the drench treatment but that were treated at the nursery had a mean value of log 1.9 cfu/g roots, while ones that received the drench had 3.3 cfu/g roots, which is an increase of more than one order of magnitude. The level of arsenic taken up into ferns was about tenfold higher than the level in soil regardless of treatment. Unfortunately, this level of uptake times the total plant biomass still resulted in very small reductions in the course of a single growing season. Use of these tropical-adapted ferns may be more practical in southern regions, where they can grow as a perennial than in upstate New York, where they are poorly adapted. However, there are other plants that are better adapted to northern climates and that under proper conditions can hyperaccumulate arsenic. Testing of these plant-microbe associations is currently underway. In addition, the abilities of T22 and other root-associated microbes to enhance uptake of heavy metals in hyperaccumulating plants is being evaluated concurrently. 2. Nitrates Agricultural systems and industrial processes place large quantities of nitrates and phosphorus into waterways. This pollution problem contributes to the zone of hypoxia along the coast of the United States in the Gulf of Mexico and other regions worldwide and may also encourage growth of toxic estuarine microbes such as Pﬁesteria. These environmental costs are high—the EPA estimates that harmful algal blooms may have been responsible for an estimated $1 billion in economic losses during the past decade. In the mid-1990s, G. E. Harman and his coworkers noticed that maize plants grown from T22-treated seeds were greener and larger than ones grown without the seed treatment, which suggested a capability of these fungi to enhance nitrogen uptake. A series of ﬁeld experiments were undertaken in which we grew maize under different nitrogen fertilizer levels with or without a seed treatment with the beneﬁcial fungus. Maize usually responds to nitrogen fertilizer by increases in yields up to a level, deﬁned as the yield plateau, beyond which yields

320

HARMAN et al.

usually do not increase further. This yield plateau is considered the maximum yield for the particular cultivar and environment. In the presence of T22, (1) plants responded (increased growth) more rapidly to N fertilizer than in its absence (Harman, 2000), and (2) at tasselling, plants were greener and had a greater stalk diameter (Harman, 2000). The nitrogen response curves consistently showed an improvement in nitrogen utilization with T22 in our trials (Harman, 2000) (Fig. 3).

FIG. 3. (A) Nitrogen fertilizer yield relationships in a trial at Cornell University in 2000 in the presence or absence of T22 and (B) yields at two different nitrogen levels in a large (23 replications) trial at Cornell the same year. The lines in (A) are signiﬁcantly different at P = 0.05, and in (B), bars with different letters are signiﬁcantly different at the same probability level. (A) is from Harman (2001); (B) is from Harman and Donzelli (2001). Both are used with permission.

POLLUTION REMEDIATION USING TRICHODERMA SPP.

321

These experiments suggested that nitrogen fertility levels could be reduced by 40–50% (Harman, 2000, 2001; Harman and Donzelli, 2001, see also Fig. 3), with no reduction in yield. It is not clear whether this response was due simply to more efﬁcient use of fertilizer or to mining of organic nitrogen pools in the soil. Such responses were associated with more abundant deep roots (Harman, 2000), which suggests that one key factor is greater density and depth of rooting, since deeper and denser root systems would be expected to both better intercept applied nitrogen and also utilize the nutrient at deeper depths of roots. However, there is a complication to this positive story. Persons who conducted commercial trials (more than 500 have been done so far) reported that occasionally there were negative growth responses to T22 in maize. We now have veriﬁed that this is true by examination of inbred lines. Maize line Mo17 responds very positively (Harman et al., 2004b), other lines respond less strongly, and a few—such as inbred line A661—are negatively affected by T22 (Harman et al., 2004b, unpublished). Fortunately such responses can be quantiﬁed when plants are very young; 2-week-old plants already show changes in growth that are predictive of responses and plant performance throughout the season. Thus the increased growth capabilities and associated nitrogen responses are rapidly predictable. It is likely that reactions of other plants may also be measurable at an early stage, thereby permitting rapid assessment of the genetic responses of different varieties to T22. B. DEGRADATION OF POLLUTANTS The largest commercial use of Trichoderma spp. is in the production of enzymes. These fungi are proliﬁc producers and secreters of enzymes and are widely used, as both native (mutated) and transgenic strains for production of cellulases and other carbohydrate lyases of plant cell wall components; they also are efﬁcient producers of enzymes such as chitinases. Trichoderma chitinases probably will be used for commercial applications soon (Donzelli et al., 2003). In all cases, multiple forms of the enzymes exist with different modes of action; these enzymes and their genes were thoroughly reviewed in 1998 (Harman and Kubicek, 1998). 1. Cyanide Cyanide and metallocyanides have been released into the environment from the metal plating and mining industries, and large quantities are also produced as wastes and emissions from the manufacture of

322

HARMAN et al.

paints, synthetic ﬁbers, pesticides, and formerly, in the production of fuel gas from carbonization of oil and coal (Dubey and Holmes, 1995; Shiriﬁn et al., 1966). There are an estimated 50,000 former coal gasiﬁcation sites in the United States left from the coal gas era of 1840–1950 (http://www.heritageresearch.com/manufactured_gas_E.htm). In the water gas process, retorted coal was blasted with super-heated steam to produce a line fuel composed of hydrogen, methane, and carbon monoxide plus HCN, NH4, and H2S impurities. The ammonia was captured with sulfuric acid and sold as fertilizer; the H2S was scrubbed with ferric oxide and went to the manufacture of sulfuric acid; and HCN also reacted with ferric oxide to form ferric ferrocyanide, a.k.a. Prussian blue. This valuable pigment was used in paints, textiles, and inks. None of these operations were quantitative; thus, former gas sites may contain high levels of cyanide, metallocyanides, and metal sulﬁdes and organic components of the coal tar residues that constituted approximately 7–8% of the coal gasiﬁed. A second, more recent source of environmental concern is the practice of leaching trace amounts of gold and silver from abandoned mine tailings with CN solutions. Thousands of tons of cyanide are applied to open piles of rubble; after inﬁltrating the pulverized rock, the solutions run into impoundments where the metals are extracted and the cyanide is recycled to the tailings. Large volumes of dilute cyanide solutions may be lost to ground and surface waters through leaks and spills, plus catastrophic spills have poisoned whole river systems (http://www. grida.no/inf/news/news00/news17.htm). Several states and countries (e.g., Wisconsin, Montana, Turkey) are enacting laws to forbid further open cyanide leaching operations, but millions of acres of mine lands and water sheds are already contaminated from some of these ventures. Trichoderma spp. have recently been reported to produce two separate enzymes that degrade cyanide (Ezzi and Lynch, 2002; Ezzi et al., 2003). The enzymes and the reactions they catalyze are provided below: Formamide hydrolase HCN þ H2 O ! HCONH2 !! CO2 and NH3 Rhodanese 2 S2 O 3 þ CN ! SO3 þ CNS These enzymes are constitutively produced and secreted into the medium by all strains that have been tested (Ezzi and Lynch, 2002; Ezzi et al., 2003), and the rhodanese has been characterized (Ezzi et al., 2003). This discovery provides a useful biological method for

POLLUTION REMEDIATION USING TRICHODERMA SPP.

323

remediation of cyanide. In addition to cyanide, most polluted sites contain metallocyanides (e.g., Prussian blue). These compounds are insoluble and are not particularly toxic themselves, but they release soluble cyanide over time. However, Trichoderma spp. take up Prussian blue—Fe4[Fe(CN)6]—into the thallus of the fungi, where it is then degraded (J. M. Lynch, D. Redman, and N. Isla, unpublished). Importantly, Trichoderma spp. are resistant to a wide range of toxic compounds (Harman et al., 2004a), and some strains will grow even at levels of up to 2,000 g of cyanide per g of soil, a level that is 10,000 times the EPA allowable limit. This rather remarkable tolerance for cyanide and other toxicants is associated with a very active cellular detoxiﬁcation system based on permeases such as ABC transporters recently discovered in these fungi (Harman et al., 2004a). Thus Trichoderma spp. (1) degrade cyanide by activity of characterized enzymes, (2) take up and degrade metallocyanides, (3) are resistant to high levels of a variety of toxicants, and (4) some strains are highly rhizosphere competent and form stable, long-lasting plant microbe communities. This self-organization provides a highly robust platform for degradation of cyanide and other pollutants. This system does, in fact, work rather well in remediation of cyanide in microcosm studies, as shown in Fig. 1, and this has also been reproduced by using other plants, including dicots. The wheat plants shown in the ﬁgure are sensitive to cyanide, but they are protected from its toxicity by release of enzymes from the plant-colonizing Trichoderma strains. Thus the Trichoderma-plant associations provide a useful point from which to develop novel remediation strategies for this serious pollutant. However, this system would be even more effective if the plants used were themselves resistant to cyanide. Recently such plants have been identiﬁed. A variety of willow (Salix eriocephala var. Michaux) was found to take up both potassium cyanide and potassium ferrocyanide and to translocate them throughout the plant. Most importantly, the cyanide and ferrocyanide are degraded within the plant so that they do not accumulate within the aerial portions of the plant (Ebbs et al., 2003). The combination of willows plus rhizosphere competent Trichoderma strains are expected to provide very effective methods of removal of cyanide and metallocyanides from polluted sites; research and development on this combination is underway. 2. Polyphenols When olives are pressed, about 11 L of water are released for every liter of oil. This ‘‘black water’’ contains very high levels of phenolic and polyphenolic compounds; the biological oxygen demand may be

324

HARMAN et al.

400-fold higher than that of usual municipal waste waters. The polyphenolics are toxic, so the water is essentially sterile. Moreover, at least a portion of the polyphenolics are resistant to anaerobic microbial degradation, making processing of such waste waters difﬁcult. However, Trichoderma spp. are highly tolerant to the toxicity of olive oil waste water, just as they are to many other compounds (M. Lorito, unpublished). Thus, if the waste water is diluted 1:3 and aerated, the fungi can grow and proliferate. While so doing, they produce polyphenol-degrading enzymes that destroy the polyphenolics (Fig. 4A, see color insert) and so remove them from solution (Fig. 4B). Similar procedures have been used to treat waste water with Phanerochaete chrysosporium or Pleurotus ostreatus, which are basidomycetes known to produce copious levels of polyphenol oxidases, laccases, and related enzymes expected to degrade polyphenolics (Kissi et al., 2001). P. chrysosporium after 9 days of incubation removed about 50% of the phenolics and color, while P. ostreatus required 12 days to reach this same level (Kissi et al., 2001). By contrast, Trichoderma strain TC3 removed about 90% of the materials, which resulted in a high level of color removal, after only 6 days of incubation. Thus these fungi appear to provide useful methods to cleanse olive oil waste waters. 3. Polycyclic Aromatic Hydrocarbons Along with cyanides, gas works sites—as well as other polluted soils, including those polluted with petroleum, coal tar, and shale oil— contain complex polycyclic aromatic hydrocarbons (PAHs). These compounds are frequently toxic, mutagenic, and/or carcinogenic (Keith and Telliard, 1979). Numerous fungi, especially wood-rotting fungi (Andersson et al., 2003; Pointing, 2001) and bacteria, sometimes in combination (Boonchan et al., 2000), have been demonstrated to degrade these compounds to one degree or another. More recently, other fungi including Mucor spp., Penicillium spp., T. harzianum, and T. virens have been shown to effectively degrade pristane when the compound was used as the sole carbon source (Ravelet et al., 2000; Saraswathy and Hallberg, 2002). In our research, T. harzianum TH1 has been shown to degrade hexadecane and pristane in ﬂask culture (J. M. L. and P. J. Phillips, unpublished). The addition of glucose as a co-metabolic substrate tripled the degradation rate compared with ﬂasks containing only the hydrocarbons as substrates (J. M. L. and P. J. Phillips, unpublished). These capacities are usually strain dependent. Moreover, there exists great variation in the resistance of different strains to the toxic effects of PAHs. For example, strain T22 was only slightly inhibited in the application of 20 g/L to the

POLLUTION REMEDIATION USING TRICHODERMA SPP.

325

FIG. 4. Degradation of polyphenolics from olive oil waste waters by the activities of Trichoderma spp. The fungi were added to diluted (1:3 waste:water dilution) waste water and the content of polyphenolics in the mixture was followed over time. Phenolic compound content was determined by using the method of Folin-Denis based on the commercial substrate (Folin-Ciocalteu reagent; Sigma Scientiﬁc). A 5 ml total reaction mixture consisted of 250 l of the reagent, 500 l of sodium carbonate 20%, 100 l of the sample to be tested, and distilled water. The phenolic concentration was determined based on a standard curve with differing concentrations of tannic acid. (A) The percentage degradation of the polyphenols; (B) the change in water color by the process. Analysis of variance of arcsine transformed percent values of all Trichoderma treatments were signiﬁcantly different (P < 0.05) from controls without the strains.

326

HARMAN et al.

surface of nutrient medium, but strain T12 was completely inhibited at this same concentration. This is particularly intriguing, since T22 was prepared by protoplast fusion between T. harzianum strain T12 and T. atroviride strain T95, and the genome of the resulting progeny is that of T12 (Stasz, 1990; Stasz et al., 1988). Thus there is substantial potential for selection of Trichoderma strains that are both resistant to and degrade PAHs, perhaps as root colonizing symbionts. VI. Conclusions and Future Prospects Trichoderma spp. are used substantially for enzyme production and in agriculture for both plant protection against diseases and for plant growth promotion. Products and services based on these fungi are poised to make signiﬁcant contributions to the remediation or alleviation of soil or water. The fungal genus is common, and there is great diversity between different strains. They also are very proliﬁc producers of a wide range of enzymes. They also have been shown to be plant symbionts and are capable of increasing plant biomass and root growth and simultaneously protecting plants from disease by a number of different mechanisms. Some strains are strongly rhizosphere competent, which permits them to colonize roots, grow, and persist on roots and to provide long-term beneﬁts in terms of plant health and productivity. This capability permits the fungi to form durable and robust plant associations in a wide variety of soil conditions. The symbiotic nature of the interaction permits co-metabolic processes in which the fungus gains nutrients from the plants and produces bioactive molecules that stimulate plant growth and resistance to stresses and stimulate the production of metabolites that are of value to the plants. Some of these compounds, including enzymes, may be highly useful in degradation of toxic soil pollutants; this capability is enhanced by the fact that Trichoderma spp. possess high intrinsic resistance to toxic compounds, probably by virtue of a strong detoxiﬁcation system. The abilities of rhizosphere-competent Trichoderma spp. to enhance root growth is expected to enhance the capability of hyperaccumulating plants to remove toxic metals and metalloids. Research just conducted suggests that the presence of the fungi increased removal of arsenic from soils by hyperaccumulating ferns in the genus Pteris. Other data demonstrate that root colonization by the fungi increases nitrogen fertilizer use efﬁciency in maize but that there is a strong maize genotype response. Positive responding lines can be identiﬁed by growth after only 2 weeks. The use of positive responding maize lines and reduced levels of nitrogen fertilizer are expected to reduce nitrate

POLLUTION REMEDIATION USING TRICHODERMA SPP.

327

pollution of waterways and, incidentally, to increase farm proﬁts. Further, the same fungi produce enzymes that degrade cyanide, and they accumulate, and then degrade, metallocyanides such as Prussian blue. They do so in co-metabolic combinations with plants. We anticipate that combinations of rhizosphere competent Trichoderma strains with willows that take up and themselves degrade ferrocyanides and cyanide will provide novel and effective solutions to problems of soil pollution with cyanide. Further, these fungi also produce enzymes that, in aerated reactors, degrade polyphenolics. Finally, they possess capabilities to degrade at least some PAHs. No doubt additional uses of the fungi, and their associations with plants, will be found in the pollution remediation ﬁeld. They appear to provide highly promising new tools for the treatment or amelioration of soil and water pollution. ACKNOWLEDGMENTS The authors thank Dan Berler, Phytobials, LLC, and Terry Spittler and Kristen Ondik for editorial assistance. G. E. H. thanks Rixana Petzoldt and Kristen Ondik for technical support; his research was supported in part by grants from the U.S.-Israel Binational Agricultural Research and Development Fund, the Cornell Center for Advanced Technology, Phytobials, LLC, Advanced Biological Marketing and BioWorks, Inc. Research by M. L. was supported by the following projects: FIRB-MIUR 2002 Genomica Funzionale, PON-MIUR 2002 NTSPAE, EU ‘‘TRICHOEST’’, EU FAIR98PL-4140; MIUR-MIPAF 2002 ‘‘Costituzione di un Germoplasma Microbico’’, MIUR PRIN 2002 Pseudomonas syringae, CNR-IPP SCAFCF, PRIN 2003.

REFERENCES Andersson, B. E., Lundstedt, S., Tornberg, K., Schnurer, Y., Oberg, L. G., and Mattiasson, B. (2003). Incomplete degradation of polycyclic aromatic hydrocarbons in soil inoculated with wood-rotting fungi and their effect on indigenous soil bacteria. Environ. Toxic. Chem. 22, 1238–1243. Bakker, A. W., and Schippers, B. (1987). Microbial cyanide production in the rhizosphere in relation to potato yield reduction and Pseudomonas spp-mediated plant growthstimulation. Soil Biol. Biochem. 19, 451–457. Bigirimana, J., De Meyer, G., Poppe, J., Elad, Y., and Hofte, M. (1997). Induction of systemic resistance on bean (Phaseolus vulgaris) by Trichoderma harzianum. Med. Fac. Landbouww. Univ. Gent. 62, 1001–1007. Boonchan, S., Britz, M. L., and Stanley, G. A. (2000). Degradation and mineralization of high-molecular-weight polycyclic aromatic hydrocarbons by deﬁned fungal-bacterial cocultures. Appl. Environ. Microbiol. 66, 1007–1019. Chang, Y.-C., Chang, Y.-C., Baker, R., Kleifeld, O., and Chet, I. (1986). Increased growth of plants in the presence of the biological control agent Trichoderma harzianum. Plant Dis. 70, 145–148.

328

HARMAN et al.

Chao, W. L., Nelson, E. B., Harman, G. E., and Hoch, H. C. (1986). Colonization of the rhizosphere by biological control agents applied to seeds. Phytopathology 76, 60–65. Chet, I. (1987). Trichoderma—application, mode of action, and potential as a biocontrol agent of soilborne plant pathogenic fungi. In ‘‘Innovative Approaches to Plant Disease Control’’ (I. Chet, ed.), pp. 137–160. J. Wiley and Sons, New York. Donzelli, B. G. G., Ostroff, G., and Harman, G. E. (2003). Enhanced enzymatic hydrolysis of langostino shell chitin with mixtures of enzymes from bacterial and fungal sources. Carboh. Res. 338, 1823–1833. Dubey, S. K., and Holmes, D. S. (1995). Biological cyanide destruction mediated by microorganisms. World J. Microbiol. Biotechnol. 11, 257–265. Ebbs, S., Bushey, J., Poston, S., Kosma, D., Samiotakis, M., and Dzombak, D. (2003). Transport and metabolism of free cyanide and iron cyanide complexes by willow. Plant Cell Environ. 26, 1467–1478. Ezzi, M. I., and Lynch, J. M. (2002). Cyanide catabolizing enzymes in Trichoderma spp. Enz. Microb. Technol. 31, 1042–1047. Ezzi, M. I., Pascual, J. A., Gould, B. J., and Lynch, J. M. (2003). Characterisation of the rhodanese enzyme in Trichoderma spp. Enz. Microb. Technol. 32, 629–634. Fravel, D., Olivain, C., and Alabouvette, C. (2003). Fusarium oxysporum and its biocontrol. New Phytol. 157, 493–502. Harman, G. E. (2000). Myths and dogmas of biocontrol. Changes in perceptions derived from research on Trichoderma harzianum T-22. Plant Dis. 84, 377–393. Harman, G. E. (2001). Microbial tools to improve crop performance and proﬁtability and to control plant diseases. In ‘‘Proceedings of International Symposium on Biological Control of Plant Diseases for the New Century-Mode of Action and Application Technology’’ (D. D.-S. Tzeng and J. W. Huang, eds.), pp. 71–84. National Chung Hsing University, Taichung, Taiwan. Harman, G. E., and Bjo¨rkman, T. (1998). Potential and existing uses of Trichoderma and Gliocladium for plant disease control and plant growth enhancement. In ‘‘Trichoderma and Gliocladium’’ Vol. 2 (G. E. Harman and C. P. Kubicek, eds.), pp. 229–265. Taylor and Francis, London. Harman, G. E., and Kubicek, C. P. (1998). ‘‘Trichoderma and Gliocladium, Vol. 2. Enzymes, Biological Control and Commercial Applications.’’ Taylor and Francis, London. Harman, G. E., and Donzelli, B. G. G. (2001). Enhancing crop performance and pest resistance with genes from biocontrol agents. In ‘‘Enhancing Biocontrol Agents and Handling Risks’’ (M. Vurro, J. Gressel, T. Butt, G. E. Harman, A. Pilgeram, R. J. St. Ledger, and D. L. Nuss, eds.), pp. 114–125. IOS Press, Amsterdam. Harman, G. E., Howell, C. R., Viterbo, A., Chet, I., and Lorito, M. (2004a). Trichoderma species-opportunistic, avirulent plant symbionts. Nat. Rev. Microbiol. 2, 43–56. Harman, G. E., Petzoldt, R., Comis, A., and Chen, J. (2004b). Interactions between Trichoderma harzianum strain T22 and maize inbred line Mo17 and effects of these interactions on diseases caused by Pythium ultimum and Colletotrichum graminicola. Phytopathology 94, 147–153. Hwang, J., and Benson, D. M. (2003). Expression of induced systemic resistance in poinsettia cuttings against Rhizoctonia stem rot by treatment of stock plants with binucleate Rhizoctonia. Biol. Contr. 27, 73–80. Keith, L. H., and Telliard, W. A. (1979). Priority pollutants I-a perspective view. Environ. Sci. Technol. 13, 416–423. Kissi, M., Mountadar, M., Assobhei, O., Gargiulo, E., Palmieri, G., Giardina, P., and Sannia, G. (2001). Roles of two white-rot basidiomycete fungi in decolorisation

POLLUTION REMEDIATION USING TRICHODERMA SPP.

329

and detoxiﬁcation of olive mill waste water. Appl. Microbiol. Biotechnol. 57, 221–226. Klein, D., and Eveleigh, D. E. (1998). Ecology of Trichoderma. In ‘‘Trichoderma and Gliocladium’’ (C. P. Kubicek and G. E. Harman, eds.), Vol. 1, pp. 57–74. Taylor and Francis, London. Kloepper, J. W., Tuzun, S., Liu, L., and Wei, G. (1993). Plant growth-promoting rhizobacteria as inducers of systemic disease resistance. In ‘‘Pest Management: Biologically Based Technologies’’ (R. D. Lumsden and J. L. Vaughn, eds.), pp. 156–165. Am. Chem. Soc. Washington, D. C. Lindsey, D. L., and Baker, R. (1967). Effect of certain fungi on dwarf tomatoes grown under gnotobiotic conditions. Phytopathology 57, 1262–1263. Lynch, J. M. (2003). Plant growth promoting agents. In ‘‘Microbial Diversity and Biprospecting’’ (T. A. Bull, ed.) Washington, D. C. Ma, L. Q., Komar, K. M., and Kennelley, E. D. (2001a). Methods for removing pollutants from contaminated soil materials with a fern plant. U. S. Patent 6,302,942. Ma, L. Q., Komar, K. M., Tu, C., Zhang, W., Cai, Y., and Kennelley, E. D. (2001b). A fern that hyperaccumulates arsenic. Nature 409, 579. Ousley, M. A., Lynch, J. M., and Whipps, J. M. (1993). Effect of Trichoderma on plant growth: A balance between inhibition and growth promotion. Microbial Ecol. 26, 277–285. Pointing, S. B. (2001). Feasibility of bioremediation by white-rot fungi. Appl. Microbiol. Biotechnol. 57, 20–33. Pozo, M. J., Cordier, C., Dumas-Gaudot, E., Gianinazzi, S., Barea, J. M., and AzconAguilar, C. (2002). Localized versus systemic effect of arbuscular mycorrhizal fungi on defence responses to Phytophthora infection in tomato plants. J. Exp. Bot. 53, 525–534. Ravelet, C., Krivobok, S., Sage, L., and Steiman, R. (2000). Biodegradation of pyrene by sediment fungi. Chemosphere 40, 557–563. Ryu, C.-M., Farag, M. A., Hu, C.-H., Reddy, M. S., Wei, H.-X., Pare, P. W., and Kloepper, J. W. (2003). Bacterial volatiles promote growth in Arabidopsis. Proc. Nat. Acad. Sci. USA 100, 4927–4932. Saraswathy, A., and Hallberg, R. (2002). Degradation of pyrene by indigenous fungi from a former gasworks site. FEMS Microbiol. Lett. 210, 227–232. Shiriﬁn, N. S., Beck, B. D., Gauthier, T. D., Chapnick, S. D., and Goodman, D. (1966). Chemistry, toxicology and human health risk of cyanide compounds at former manufactured gas sites. Regulat. Toxicol. Pharmacol. 23, 106–116. Stasz, T. E. (1990). Genetic improvement of fungi by protoplast fusion for biological control of plant pathogens. Can. J. Plant Pathol. 12, 322–327. Stasz, T. E., Harman, G. E., and Weeden, N. F. (1988). Protoplast preparation and fusion in two biocontrol strains of Trichoderma harzianum. Mycologia 80, 141–150. Woolson, E. A. (1975). ‘‘Arsenical Pesticides.’’ American Chemical Society, Washington, D. C. Yedidia, I., Benhamou, N., and Chet, I. (1999). Induction of defense responses in cucumber plants (Cucumis sativus L.) by the biocontrol agent Trichoderma harzianum. Appl. Environ. Microbiol. 65, 1061–1070. Yedidia, I., Benhamou, N., Kapulnik, Y., and Chet, I. (2000). Induction and accumulation of PR proteins activity during early stages of root colonization by the

330

HARMAN et al.

mycoparasite Trichoderma harzianum strain T-203. Plant Physiol. Biochem. 38, 863–873. Yedidia, I., Shoresh, M., Kerem, Z., Benhamou, N., Kapulnik, Y., and Chet, I. (2003). Concomitant induction of systemic resistance to Pseudomonas syringae pv lachrymans in cucumber by Trichoderma asperellum (T-203) and accumulation of phytoalexins. Appl. Environ. Microbiol. 69, 7343–7353. Yedidia, I., Srivastva, A. K., Kapulnik, Y., and Chet, I. (2001). Effect of Trichoderma harzianum on microelement concentrations and increased growth of cucumber plants. Plant Soil 235, 235–242.

Bacteriophage Defense Systems and Strategies for Lactic Acid Bacteria JOSEPH M. STURINO*

AND

TODD R. KLAENHAMMER{,{,},}

*Chr. Hansens, Inc. Milwaukee, Wisconsin 53214 { { }

Genomic Sciences Program, North Carolina State University Raleigh, North Carolina, 27695

Department of Food Science, North Carolina State University Raleigh, North Carolina, 27695

Southeast Dairy Foods Research Center, North Carolina State University Raleigh, North Carolina, 27695 }

Author for correspondence. E-mail: [email protected]

I. Introduction A. Streptococcus thermophilus B. Discovery of Bacteriophages C. Life Cycles D. Bacteriophage Defense Strategies and Systems Defined II. Traditional Strategies A. Improved Sanitation and Manufacturing Processes B. Phage Inhibitory Media C. Phage-Unrelated Strains: Rotation III. Molecular Strategies A. Elimination or Alteration of Host-Encoded Factors IV. Native Defense Systems A. Adsorption Blocking B. Restriction and Modification C. Abortive Infection V. Recent Advancements in Genomics A. Comparative Genomics B. Anti-Receptor Identification VI. Engineered Defense Systems A. Superinfection Exclusion and Immunity B. Antisense RNA C. Origin-Derived Phage-Encoded Resistance D. Subunit Poisoning E. Phage-Triggered Suicide Systems VII. Concluding Remarks References

332 334 336 337 339 339 339 340 340 342 342 343 344 345 349 351 351 352 353 354 355 362 366 368 368 369

The next morning, on opening the incubator, I experienced one of those moments of intense emotion which reward the research worker for all his pains: at the first glance I saw that the culture which the night before had been very turbid, was perfectly clear: all the bacteria had vanished, they 331 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

332

STURINO AND KLAENHAMMER

had dissolved away like sugar in water. As for the agar spread, it was devoid of all growth and what caused my emotion was that in a flash I had understood: what caused my clear spots was in fact an invisible microbe, a filterable virus, but a virus which is parasitic on bacteria. Felix d’Herelle, French-Canadian Bacteriologist, on his discovery of bacteriophages in 1916 (Science News 14: 44–59, 1949).

I. Introduction Food preservation has been critical for the survival of humankind. Historically, populations depended on a variety of techniques to preserve raw foodstuffs, including fruits, grains, meats, milk, vegetables, and water. Although drying, salting, and smoking are perhaps the most rudimentary forms of food preservation, fermentation is one of the oldest and, arguably, most elegant preservation techniques. Together, these preservation processes enabled humankind to explore the planet in ways not possible before their discovery and allowed commerce to develop between distant populations. These two factors formed an essential platform that allowed increasingly complex civilizations to develop and ﬂourish. Today, with an exploding population and the demand for food growing worldwide, the dairy industry is under considerable pressure to produce large volumes of consistent and highquality fermented products. The industry has adapted to meet these demands. Small-scale production plants have been consolidated into large-scale production facilities that process millions of gallons of milk per day. Modern production processes continue to become increasingly automated, and signiﬁcant efforts have been made to shorten production times while extending production schedules. In fact, many food-processing plants are run at capacity 24 hours a day. In addition, industrial-scale fermentations no longer rely on spontaneous fermentation, and the industry has turned to more consistent inoculation strategies. Today the industry has turned to the use of concentrated, well-deﬁned, and optimized starter cultures to catalyze their fermentations. The dairy industry utilizes extensively strains of Streptococcus thermophilus and species from the genera Lactococcus, Lactobacillus, Leuconostoc, and Pediococcus as starter cultures or culture adjuncts for use in the manufacture of a variety of fermented dairy products. These starter cultures are all members of the lactic acid bacteria (LAB), a heterogeneous family of Gram-positive eubacteria that derive metabolic energy from the fermentation of carbohydrates to lactate via substrate level phosphorylation. These bacteria impart many of the

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

333

textural, organoleptic, and preservative characteristics that can exclude the growth of many pathogens (for a review, see Ross et al., 2002). Most of these desirable properties are largely a byproduct of cellular metabolism reﬂecting the activity of enzymes or the production of organic acids, especially lactic acid. From a macroscopic perspective, however, much more than the carbohydrate content changes over the course of food fermentation: the basic characteristics of the foodstuffs, including texture, taste, aroma, and nutritional value, are transformed. These changes are primarily due to the gradual hydrolysis of proteins, lipids, polysaccharides, and other compounds associated with the fermentation substrate. The signiﬁcance of bacteriophage infections during product manufacture was unknown until Whitehead and Cox (1935) identiﬁed the ﬁrst phages speciﬁc for strains of Lactococcus lactis (then called Streptococcus lactis). Since their discovery, bacteriophages speciﬁc for dairy starter cultures, notably lactococci and recently S. thermophilus, have been recognized as a signiﬁcant and persistent problem for the dairy industry. Loss of fermentative capacity associated with starter culture lysis can signiﬁcantly retard or halt batch fermentations, thereby causing signiﬁcant losses of time and production capital. These losses are particularly severe when highly specialized strains, which are themselves a valuable product of scientiﬁc discovery and product development, become susceptible to phage attack. In this case, costs committed for strain development will not be recovered if the expected lifetime of a new, highly specialized strain is diminished by the appearance of lytic phages capable of infecting it. Despite the development of a variety of countermeasures, including culture rotation, improved sanitation strategies, and the use of bacteriophage-resistant starter strains, phage contamination during product manufacture continues to be the leading cause of failed or retarded batch fermentations. The problem endures because the dairy environment is a consistent reservoir for phage contamination, especially the nonsterile fermentation substrate and lysogenic starter cultures (Bruttin et al., 1997a; Moineau et al., 1996). In addition, existing phage populations can evolve resistance to phage defense systems by mutation and recombination (Bouchard and Moineau, 2000; Durmaz and Klaenhammer, 2000). As a result, the phage population as a whole is in a constant state of ﬂux. Together, these selective pressures necessitate the isolation or construction of starter cultures with enhanced phage resistance properties. This chapter describes the complex relationship that exists between strains of S. thermophilus and their bacteriophages. Examples from other genera and species of LAB are cited only when necessary

334

STURINO AND KLAENHAMMER

to ﬁll subtle gaps in current knowledge. In particular, this chapter highlights the various defense strategies and systems that have been developed to curb the propagation and evolution of lytic phages. A. STREPTOCOCCUS THERMOPHILUS The genome of S. thermophilus is 1.8 Mb, placing it among the smallest genomes of all the dairy LAB. The molar G þ C ratio is 40%. Morphologically, S. thermophilus cells are spherical to ovoid, 0.7–0.9 m in diameter, and grow in pairs to long chains. S. thermophilus strains are moderately thermophilic and can grow in the temperature range of 15–45 C (Hardie, 1994). They do not grow at pH 9.6; growth in 2.0% NaCl is strain dependent. S. thermophilus strains are fastidious and require nutrient-rich environments, such as milk, to support growth. In addition, S. thermophilus strains have a somewhat limited carbohydrate fermentation proﬁle in comparison with other dairy LAB, making them readily identiﬁable with API 50 CH fermentation strips (bioMe´rieux, Marcy l’Etoile, France). Typically, S. thermophilus produces L(þ)-lactic acid as the principal byproduct of the fermentation of fructose, glucose, lactose, mannose, and sucrose but not from arabinose, dextrin, glycerol, inulin, mannitol, rhamose, salicin, sorbitol, starch, and xylose (Hardie, 1994). Like most other LAB, S. thermophilus strains are chemoorganotrophic, nonsporulating, catalase negative, devoid of cytochromes, facultatively aerobic, and acid-tolerant. They are found naturally in milk and decaying plant material. Strains of S. thermophilus are among the most economically important of the lactic acid bacteria. S. thermophilus strains are used during the manufacture of Italian-style cheese varieties including Asiago, Mozzarella, Parmesan, Provolone, and Romano and surface-ripened cheeses such as Bel Paese, Limburger, Port du Salut, Tilsit, and Trappist (Olson, 1969; Reinbold, 1963). S. thermophilus is also used in combination with the mesophilic Lactococcus lactis during the production of Cheddar cheese. In this case, the thermophilic and mesophilic components are phage unrelated, and one of the two components will continue to produce acid if the other is lysed by bacteriophages. This microorganism is perhaps best known for its use in conjunction with Lactobacillus species, especially Lactobacillus delbrueckii spp. bulgaricus, during the manufacture of yogurt. These two microorganisms share a remarkable synergistic relationship when mixed together in a 1:1 ratio: they grow faster and produce more lactic acid and acetaldehyde, the principle volatile ﬂavor compound

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

335

associated with yogurt, when grown in co-culture (for reviews, see Matalon and Sandine, 1986; Zourari et al., 1992). Efforts to better understand this relationship are now underway. Milk is very rich in proteins, especially caseins, but contains few free amino acids (Jensen, 1995). Both strains harbor multiple amino acid auxotrophies; however S. thermophilus has fewer nutritional requirements in chemically deﬁned media (Grobben et al., 1998; Letort and Julillard, 2001) and in milk (Desmazeaud, 1983). L. delbrueckii subsp. bulgaricus typically possesses a more robust proteolytic system when compared with S. thermophilus (Courtin et al., 2002; Rajagopal and Sandine, 1990). The analysis of S. thermophilus and L. delbrueckii subsp. bulgaricus proteinase (Prt) null strains was recently used to investigate the importance of these two proteinases when the strains are grown in milk in co-culture. These studies revealed that the streptococcal proteinase (PrtS) was essential for S. thermophilus grown in milk in pure culture, but its absence had no effect on the ﬁnal pH when grown in milk in the presence of PrtBþ strains of L. delbrueckii subsp. bulgaricus. The proteinase PrtB, on the other hand, is required for optimal growth of S. thermophilus in milk, and its absence resulted in a higher ﬁnal pH and lower streptococcal cell counts when grown in mixed culture (Courtin et al., 2002). These cell wall–associated proteinases cleave caseins into short peptides that are then transported into both organisms via their respective oligopeptide transport permeases and are further degraded intracellularly into free amino acids by a variety of peptidases (for a review, see Kunji et al., 1996). As a result, increased levels of proteolysis stimulate the growth of S. thermophilus, which in turn stimulates the growth of L. delbrueckii subsp. bulgaricus through the production of a variety of compounds, including CO2 and formate, and by reducing the redox potential of the growth substrate (for reviews, see Matalon and Sandine, 1986; Zourari et al., 1992). Thus this synergistic relationship results in a dynamic fermentation where S. thermophilus is responsible for driving much of the acid production during the early stages of the fermentation, while acidiﬁcation is driven later by L. delbrueckii subsp. bulgaricus. S. thermophilus also plays an important role as a probiotic, alleviating symptoms of lactose intolerance and other gastrointestinal disorders (Kolars et al., 1984; Martini et al., 1991). This microorganism has shown promise in maintaining remission of ulcerative colitis and has also been shown to help prevent the postoperative recurrence of Crohn’s disease (Venturi et al., 1999). S. thermophilus also produces high levels of the vitamin folate (vitamin B-9), which plays a variety of important roles in human health (Crittenden et al., 2003). Folate is

336

STURINO AND KLAENHAMMER

important for iron metabolism and maintaining cardiovascular function but is perhaps best known for its role in modulating fetal development in utero (Krishnaswamy et al., 2001). Folate deﬁciency during pregnancy has been linked to the increased incidence of fetal neural tube defects and other congenital anomalies (for a review, see Wilson et al., 2003). The mechanism of action of probiotics may include receptor competition, pathogen antagonism, effects on mucin secretion, vitamin or co-factor production, or probiotic immunomodulation of gut-associated lymphoid tissue (Reid et al. 2003).

B. DISCOVERY OF BACTERIOPHAGES Phages were the last of the three major classes of viruses to be discovered; their discovery was preceded by the discovery of the plant viruses (tobacco mosaic virus) (Ivanovsky, 1892) and the animal viruses (foot-and-mouth disease virus) (Loefﬂer and Frosch, 1898). As with many of the great discoveries of our age, phages were discovered by accident. Phages were independently identiﬁed early in the twentieth century by British bacteriologist Frederick William Twort (1915) working at the Brown Institution (London, England) and then 2 years later by the French Canadian bacteriologist Felix D’Herelle (1917) at the Pasteur Institute (Paris, France). Both researchers observed small, symmetrical zones of clearing, called plaques, on agar plates containing an otherwise conﬂuent lawn of bacteria. When a soft agar overlay containing an abundance of sensitive bacteria and a ﬁnite number of phage particles is poured over the surface of a nutrient agar plate and incubated under the appropriate conditions, two distinct topographies will be evident over the surface of the plate. The bacteria will grow to saturation in regions of the plates devoid of phage particles, forming an opaque layer (or lawn) in the nutrient agar overlay. In regions of the plate where phage particles are present, however, the phage particles will come into contact with and lyse the bacteria. Following lysis, the progeny phages are released and diffuse through the semisolid substrate to come into contact with neighboring bacteria, beginning the infection cycle anew. The eventual result of these and subsequent rounds of local infection and lysis is a growing zone of lysis, or plaque, teeming with phage particles and cellular debris. This zone continues to expand while there are actively metabolizing bacteria available. Eventually, the infection becomes visible to the naked eye against the backdrop of an otherwise conﬂuent lawn.

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

337

C. LIFE CYCLES Bacteriophages exist in two alternative states: (1) the extracellular and metabolically inert infectious particle and (2) the intracellular genetic parasite. In the extracellular state, the phage particle is deﬁned as a nucleic acid genome encased within a highly symmetric capsid that is normally composed solely of proteins. The capsid acts as a protective vehicle for the delivery of the phage genome to another suitable host. To initiate phage replication, a phage particle must adsorb to an actively growing cell by binding speciﬁc receptor(s) associated with the extracellular envelope. Following adsorption, the phage injects its genome into the bacterial cytosol. Once introduced into the cytosol, the genome will respond to its new environment in a manner that is consistent with its genetic makeup. There are three types of bacteriophage life cycles: lytic, lysogenic, and chronic. Obligate lytic phages must ultimately terminate their infection and lyse their host to release progeny phage particles; most of the phages that are problematic to the fermentation industries are obligately lytic in nature. Lysogenic infections do not result in the production of phage particles and do not release progeny into the extracellular environment; this type of life cycle is described later in this review. A chronically infecting phage can release progeny phages into the extracellular environment without killing its host (Maniloff et al., 1981). In this case, the phage and the host co-exist, and bacteriophages are constantly shed over the course of the infection. Bacteriophages capable of chronically infecting LAB have not been described to date. Intracellularly, bacteriophages exist in a state of ﬂux, and commitment to the lytic life cycle initiates a ﬁnely tuned and temporally coordinated cascade of biosynthetic reactions. These reactions rely heavily on the preexisting machinery and energetic capacity of the host system at the time of infection. The degree to which a virus relies on host-encoded factors differs from bacteriophage to bacteriophage, however. During the lytic life cycle, phage-encoded genes are transcribed and translated. This results in the catalysis of phage genome replication and the synthesis of phage polymers. These polymers are then assembled into mature phage particles. These particles are ﬁnally released into the environment when the cellular envelope is breached due to the activity of two phage-encoded proteins, holin and lysin (for reviews, see Sable and Lortal, 1995; Wang et al., 2000). In addition to the lytic cycle, temperate phages are capable of undergoing a lysogenic life cycle (Fig. 1, see color insert). In this case, the

338

STURINO AND KLAENHAMMER

FIG. 1. Simpliﬁcation of the life cycles of a temperate bacteriophage. The lytic life cycle is depicted by blue arrows. Entry into and exit from the lysogenic life cycle is depicted by red arrows.

phage does not multiply lytically. Instead, the phage integrates its genome into the host chromosome. The bacterium, which is now referred to as a lysogen, possesses the ability to produce bacteriophages and transmits this ability to its daughter progeny. Hence, the phage is effectively replicated once with every division of the bacterium. The lysogenic state will be maintained until the prophage is induced because of the exposure of the lysogen to exogenous stress(es) (e.g., SOS response triggered by DNA damage). Once induced, the integrated phage, called a prophage, will excise from the host genome and enter the lytic cycle. The progeny phages that are subsequently released can go on to lysogenize neighboring cells. As a result, lysogenic starter cultures could be problematic if used industrially, because they represent a renewable source of phage within the dairy facility. Lysogenic starter strains appear to be less of a problem for S. thermophilus strains than they are for Lactococcus lactis, however. Fortunately, very few strains of S. thermophilus have been shown to be lysogenic. Le Marrec et al. (1997) found one lysogen out of the 51 strains (2%) tested; Bru¨ssow et al. (1994a) found two lysogens of the 100 strains (2%) tested; Fayard et al. (1993) found 12 lysogens of the 120 strains (10%) tested; and Carminati and Giraffa (1992) found one lysogen of the 45 strains tested (2%). In addition, candidate starter strains are now routinely testing for the presence of prophages. If the strain

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

339

exhibits a great deal of economic potential, a prophage-cured derivative of the strain may be constructed. Interestingly, some prophages appear to be hypersensitive to physiological stresses associated with growth, such that even mild stresses can trigger premature lysis (for a review, see Gasson, 1996). It has been proposed that the prophage-encoded genes (speciﬁcally the holin and lysin) are constitutively expressed at a low and nonlethal level within the population and that this expression gives rise to the autolytic phenotype without massive prophage induction (Husson-Kao et al., 2000a,b). While this phenotype is clearly not beneﬁcial for the host, it is of great value to cheese makers. Autolytic strains of LAB are often used as culture adjuncts to deliver intracellular enzymes into the cheese matrix, which can accelerate cheese ripening (for a review, see Gasson, 1996). D. BACTERIOPHAGE DEFENSE STRATEGIES AND SYSTEMS DEFINED In this chapter, countermeasures designed to minimize the impact of phages are broken down into strategies and systems. Phage defense strategies, which will be discussed only brieﬂy, are speciﬁc actions that, when properly administered, reduce the number or kind of phages in the dairy environment. Phage defense systems, on the other hand, are deﬁned herein as genetic constructs that, when expressed in a host bacterium, lead to a bacteriophage-resistant phenotype. Phage defense systems are the primary focus for the remainder of this chapter. II. Traditional Strategies A. IMPROVED SANITATION AND MANUFACTURING PROCESSES Bacteriophages are found ubiquitously in milk (Bruttin et al., 1997a; Moineau et al., 1996; for a review, see Bru¨ssow et al., 1998). As a result, the use of higher-quality milk substrate(s) and proper pasteurization and sanitation regimes are critical for controlling phage levels within dairy facilities (for a review, see Daly, 1983). In addition, most phages are not completely inactivated by standard pasteurization treatments, and those that do survive are able to infect and lyse starter cultures (Binetti and Reinheimer, 2000). Sodium hypochlorite (100 ppm) and 0.15% (w/v) peracetic acid are very effective at inactivating phages during routine sanitation. Other biocides, such as 75–100% ethanol and isopropanol, exhibit suboptimal biocidal activity and are generally useful only in laboratory settings (Binetti and Reinheimer, 2000).

340

STURINO AND KLAENHAMMER

Inside a dairy facility, phages are spread by improper whey handling and aerosols and become ﬁxed on fermentation equipment. Special care must be taken to minimize the risk of phage contamination in the starter room. If the bulk starter is infected, then it is possible that the culture will be lysed and yield a cheese with high pH and high lactose content. Other methods, including the use of closed cheese vats and concentrated direct vat inoculation (DVI), which eliminates the need for bulk starter systems, have also helped to minimize the impacts of phage contamination. The use of frozen concentrates has largely eliminated the need for intermediate transfers, which helps to minimize the problems associated with scale-up propagations. Prior to the development of DVI cultures, failures in starter systems usually resulted from phage contamination of the bulk starter. B. PHAGE INHIBITORY MEDIA The use of enriched, phage-inhibitory media (PIM) has been widely adopted since its introduction (Bester and Lombard, 1975; Gulstrom et al., 1979; for a review, see Daly, 1983). In general, calcium ions are required for the efﬁcient adsorption of phages, including those speciﬁc for L. lactis and S. thermophilus. As an example, calcium ions are essential for the adsorption of phage c2 to the extracellular envelope of L. lactis (Lowrie and Pearce, 1971; Lubbers et al., 1995). Antibodies speciﬁc for the minor capsid protein gp110 (encoded by l10) speciﬁcally bound to the tip of the tail, suggesting that gp110 could be responsible for mediating adsorption of phage c2 to the host receptor (Lubbers et al., 1995). Interestingly, an EF-hand signature (pfam00036) was detected within the deduced gp110 protein sequence, which was not found in any other protein encoded by phage c2. This signature typically facilitates the binding of calcium ions and is composed of a 12-amino acid residue loop followed by a hydrophobic residue (Nakayama et al., 1992). Based on these requirements, PIM generally contain one or more cation-scavenging compounds (e.g., phosphates and citrates) to bind calcium, in addition to a variety of standard nutrients, including milk solids, yeast extract, and peptones, which support robust growth of the bacteria (for a review, see Whitehead et al., 1993). C. PHAGE-UNRELATED STRAINS: ROTATION Starter cultures may be of deﬁned or undeﬁned strain composition and may be used with or without culture or strain rotation (Cogan et al., 1991). The use of deﬁned strain cultures has provided a signiﬁcant

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

341

degree of control over fermentation systems and has been widely adopted in large-scale production facilities. These starter cultures usually contain two to ﬁve well-characterized strains that are phageunrelated and possess deﬁned fermentation properties (Thunell and Sandine, 1985). In deﬁned systems, rotation is a process whereby sensitive strains are withdrawn as needed upon the emergence of lytic phages and replaced with one or more nonisogenic and phageunrelated strain(s) with similar fermentative properties. Plasmid intracellular rotation is a novel alternative to traditional strain rotation developed in lactococci but should be equally useful in S. thermophilus (Durmaz and Klaenhammer, 1995; O’Sullivan et al., 1998; Sing and Klaenhammer, 1993). During this process, a number of phage-resistant derivatives of a single strain are constructed by introducing a variety of phage defense plasmids of different natures and speciﬁcity (e.g., abortive infection and restriction and modiﬁcation systems). These strains are then rotated as necessary. If the starter strains are to be used for the manufacture of consumer products, the most widely accepted approach for the introduction of heterologous DNA is the use of conjugation (Sanders et al., 1986; for a review, see Klaenhammer and Fitzgerald, 1994). When used properly, these rotation strategies can signiﬁcantly extend the longevity of strains in the dairy environment—especially when used in conjunction with effective methods to select for bacteriophage-insensitive mutants (Heap and Lawrence, 1976; Huggins, 1984; Klaenhammer, 1984). Although phage rotation is a powerful tool, care must be taken to ensure that not too many phage-unrelated strains are used at one time. Hull (1985) warned that the concurrent usage of large numbers of phage-unrelated strains at one time would increase the size of the available gene pool and might stimulate the emergence of new virulent phages by mutation or recombination. Unfortunately, the degree of phage relatedness must be determined empirically. Since few S. thermophilus strains contain plasmids, most strains cannot be differentiated by standard plasmid proﬁling, which is routinely used for lactococci. Although most S. thermophilus strains do not harbor plasmids, they do encode other mobile elements, including insertion sequences (IS), that readily allow for genetic typing (Guedon et al., 1995). Pulsed-ﬁeld gel electrophoresis (PFGE) in conjunction with ribotyping has proven to be reliable for the differentiation of S. thermophilus strains (Roussel et al., 1997). DNA probes have been used to differentiate even very closely related strains by targeting (1) 10 single copy genes, (2) the genes associated with the ribosomal RNA operon (rrn), and (3) three different insertion sequences—IS1191,

342

STURINO AND KLAENHAMMER

IS981, and ISS1 (Roussel et al., 1997). Prior to this, Pebay et al. (1992) demonstrated that signiﬁcant variability in hybridization patterns were observed among strains of S. salivarius subsp. thermophilus based on (1) restriction fragment length polymorphism (RFLP) and (2) variability in the number of rrn operons present.

III. Molecular Strategies A. ELIMINATION OR ALTERATION OF HOST-ENCODED FACTORS Following phage adsorption, the phage genome is injected into the bacterial cytoplasm to establish a lytic infection. The receptors and other factors required for phage replication, such as membraneassociated genome carrier proteins, are encoded by the host but have been poorly characterized in S. thermophilus. Quiberoni et al. (2000) have recently made some progress in identifying the character of these receptors in S. thermophilus. In this study, puriﬁed cell walls from two strains of S. thermophilus, YSD10 and BJ15, were treated with various macromolecule antagonists. Treatment of these cell walls with sodium dodecyl sulphate (SDS) and proteinase K failed to reduce phage adsorption, whereas mutanolysin and trichloroacetic acid reduced phage adsorption. These results suggested that a component of the phage receptor is either the peptidoglycan itself or moieties associated with the peptidoglycan. In separate experiments, the authors tested several carbohydrates for their ability to inhibit phage adsorption (Quiberoni et al., 2000). These experiments showed that phage CYM adsorbed to the glucosamine and rhamnose moieties associated with the YSD10 cell wall, whereas phage OBJ adsorbed to glucosamine and ribose moieties associated with the BJ15 cell wall. In L. lactis, the adsorption of c2-type phages is a two-step process (Monteville et al., 1994). In the ﬁrst step, the phage tail reversibly adsorbs to a carbohydrate component (rhamnose) of the cell wall. In the second step, the phage particle becomes irreversibly anchored to a membrane-associated phage infection protein (Pip) (Geller et al., 1993; Monteville et al., 1994). The involvement of a previously identiﬁed unnamed 32-kDa protein in phage genome infection is as yet unknown (Valyasevi et al., 1991). The generation of bacteriophage insensitive mutants (BIM) by spontaneous mutation or chemical mutagenesis has a rich history in starter culture development (for reviews, see Coffey and Ross, 2002; Forde and Fitzgerald, 1999). The random introduction of one or more speciﬁc mutation(s) may confer partial or complete insensitivity to phages,

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

343

which enables the straightforward selection of BIMs. Although BIMs are generally easy to isolate, they often exhibit a variety of negative qualities that may exclude them from use during product manufacture. Among the least desirable qualities that often accompany phage insensitivity are slow growth, diminished capacity to produce lactic acid and/or ﬂavor compounds, and altered agglutination properties. Other problems commonly associated with the use of with BIMs are frequent reversion to the phage-sensitive phenotype and insensitivity to only closely related phages. The study of BIMs has facilitated the study of the phage receptors involved in adsorption; however, it is often difﬁcult to localize the gene(s) that have been mutated, since they may be located anywhere in the bacterial genome. To address these difﬁculties, Lucchini et al. (2000) described the use of pGþhost9::ISS1-based insertional mutagenesis to identify genes involved in bacteriophage sensitivity. The plasmid pGþhost9::ISS1 encodes an antibiotic resistance marker, a temperature-sensitive replicon, and a single copy of ISS1, which has previously been shown to integrate randomly in S. thermophilus (Maguin et al., 1996). One of the principal advantages of plasmid-based mutagenesis systems over the use of spontaneous or chemically induced mutagenesis is that the genes interrupted by the integrated plasmid are readily cloned. Further, the vector sequences can be removed from the chromosome by recombination while leaving a single integrated copy of ISS1 in the chromosome. Using this approach, four distinct host-encoded loci involved in bacteriophage sensitivity were identiﬁed (Lucchini et al., 2000). Among the most effective loci identiﬁed was an open reading frame (orf394) that encoded a putative transmembrane protein, gene product gp394. When mutated, gp394 conferred complete resistance to all S. thermophilus phages tested. As a result, the authors proposed that gp394 is functionally analogous to the lactococcal Pip, which is essential for infection of L. lactis by c2-type bacteriophages (Garbutt et al., 1997). IV. Native Defense Systems The dairy has proven to be a dynamic venue for the study of the interactions between LAB and their bacteriophages. Through prolonged exposure over time, the persistence of phages within this environment has enriched for robust bacterial strains that have acquired a variety of bacteriophage defense systems. The characterization of these defenses allows researchers to (1) better understand the dynamics that exist between phage and host and (2) exploit these ﬁndings through the

344

STURINO AND KLAENHAMMER

construction of strains with greater phage resistance that exhibit enhanced ﬁtness in the industrial setting. In lactococci, these defenses are typically, but not exclusively, plasmid-encoded. Industrial strains of lactococci typically carry multiple plasmids in a wide range of sizes, and multiple phage defense systems are often stacked within the same strain (Klaenhammer, 1989). Strikingly, this contrasts with the situation found in S. thermophilus, where strains encode fewer plasmids. Herman and McKay (1985) examined 23 strains of S. thermophilus and found that ﬁve (22%) contained a single small cryptic plasmid. In a more recent study, Turgeon and Moineau (2001) assayed the plasmid content of 22 strains of S. thermophilus. Thirteen of them (59%) were found to contain one or two plasmids. Fifteen S. thermophilus plasmids were divided into four DNA homology groups (groups A–D). Two thirds of these plasmids belonged to group A. Three of the plasmids belonged to group B, while groups C and D each contained a single plasmid. The authors went on to determine that groups A, C, and D replicated via a rollingcircle mechanism, but they were unable to determine the replication mechanism for plasmids in group B. The scarcity of plasmid-associated defense mechanisms has made the study of native bacteriophage defense systems more challenging in S. thermophilus. Although studies on native bacteriophage defense systems in S. thermophilus remain limited, three distinct classiﬁcations of native bacteriophage defense strategies are discussed next. They are listed in order of their interference with the lytic life cycle. Wherever possible, each of these functional classes are discussed, and representative examples of each are highlighted as appropriate. A. ADSORPTION BLOCKING The ﬁrst line of extracellular defense against bacteriophage infection is to prevent the adsorption of the phage particle to the extracellular envelope (for reviews, see Coffey and Ross, 2002; Forde and Fitzgerald, 1999; Klaenhammer and Fitzgerald, 1994). As seen in other bacterial systems, electron microscopic examinations of lactococcal phage-host complexes indicated that phage attachment to the host is mediated by the tail structure (Budde-Niekiel and Teuber, 1987). Further, phage attachment to host-encoded receptors may be uniformly distributed over the surface of a bacterium or may occur in localized hot spots randomly distributed over the surface of the extracellular envelope (Budde-Niekiel and Teuber, 1987). Adsorption inhibition (Ads) has been proposed to occur by two distinct mechanisms (for a

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

345

review, see Klaenhammer and Fitzgerald, 1994). The ﬁrst mechanism involves the expression of extracellular factors, including exopolysaccharides (EPSs), which either bind to or sterically mask the phage receptor(s). The second mechanism results in the reduction or complete elimination of the phage receptors expressed on the extracellular surface. These mechanisms have been characterized, extensively in lactococci; however, they have not yet been identiﬁed in strains of S. thermophilus, although it is very likely to occur since strains of S. thermophilus produce a variety of chemically distinct EPSs (for a review, see Broadbent et al., 2003). B. RESTRICTION AND MODIFICATION Following particle adsorption and genome injection, replication of the phage genome can be terminated through the activity of restriction and modiﬁcation (R-M) systems, which act as the ﬁrst line of intracellular defense against bacteriophages (Klaenhammer, 1989; for reviews, see Allison and Klaenhammer, 1998; Coffey and Ross, 2002; Forde and Fitzgerald, 1999; Klaenhammer et al., 1991). One of the greatest beneﬁts of R-M systems is that they terminate the infection prior to the initiation of phage-directed cell death. As a result, many of the R-Mþ-infected cells will continue to be viable following the degradation of the phage genome. R-M systems are composed of two complementary enzymatic functions that work in concert to differentiate endogenous from exogenous DNA. The ﬁrst functional component is a restriction endonuclease (REase), which cleaves double stranded DNA (dsDNA). The second component is a modiﬁcation enzyme, typically a methyltransferase (MTase) that covalently modiﬁes DNA at sequence speciﬁc loci located throughout the genome. Currently, four functional classiﬁcations of R-M systems have been described based on the nature and complexity of their (1) target recognition sequences, (2) cleavage site(s), and (3) enzyme structure (for a review, see Roberts et al., 2003a). In Type I, II, and III R-M systems, the MTase is responsible for protecting endogenous DNA from REase-mediated cleavage. The converse is true for Type IV R-M systems, where the MTase targets modiﬁed DNA for REase-mediated cleavage. To date, only members of the Type I, II, and III R-M systems have been identiﬁed in the dairy LAB. In S. thermophilus, two Type I and eight Type II R-M systems have been described and partially characterized in strains of S. thermophilus (Table I). The vast majority of these R-M systems have never been cloned, sequenced, or used to augment the levels of phage resistance, however. Recently, Burrus et al. (2001)

TABLE Ia CHARACTERIZED R-M SYSTEMS IN STREPTOCOCCUS THERMOPHILUS Nameb

Type

Specificityc,d

S.SthCI65IP

I.hsdS

-f

ST NDI-6

pCI65st

AF02167

O’Sullivan et al. (1999)

SthSFiI

I

-

ST Sﬁ1

Chromosome

-

Lucchini et al. (2000)

SthER35IP

I

-

ST 135

pER35

AF177167

Solow and Somkuti (2001)

0

Hoste

0

Location

Accession number

Reference

Sth134I

II

5 -C#CGG-3

ST 134

Chromosome

-

Solaiman and Somkuti (1990)

Sth117I

II

50 -CC#WGG-30

ST 117

Chromosome

-

Solaiman and Somkuti (1991)

SslI

II

50 -CCWGG-30

ST T

Chromosome

-

Benbadis et al. (1991)

Sth455I

II

50 -CCWGG-30

ST CNRZ455

Chromosome

-

Guimont et al. (1993)

0

0

Sth132I

II

5 -CCCGNNNN#NNNN-3 30 -GGGCNNNN NNNN"-50

ST 132

Chromosome

-

Poch et al. (1997)

Sth368I

II

50 -GATC-30

ST CNRZ368

Chromosome

AJ271594

Burrus et al. (2001)

Sth0I

II

-

ST 0

pSt0

AJ242480

Geis et al. (2003)

Sth8I

II

-

ST 8

pSt08

AJ239049

Geis et al. (2003)

a

Adapted from table provided by Aidan Coffey (personal communication). Nomenclature guidelines according to REBASE (http://rebase.neb.com; Roberts et al., 2003b). c When known, the cleavage point is indicated (#). d W ¼ A or T; R ¼ A or G; Y ¼ C or T; N ¼ ACG or T. e ST ¼ Streptococcus thermophilus. f ‘‘-’’ indicates an unknown specificity or no accession number. b

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

347

cloned, sequenced, and functionally characterized Sth368I, a Type II R-M system associated with the 34,734-bp integrative and conjugative element ICESt1, which was isolated from S. thermophilus CNRZ368 (Burrus et al., 2002). Sth368I is comprosed of two different genes, sth368IR and sth368IM. The sth368IR-encoded protein exhibited signiﬁcant sequence similarity to a variety of Type II endonucleases, including R.L1aKR2I and R.Sau3AI, which recognize and cleave (#) the sequence 50 -#GATC-30 . In addition, the sth368IM-encoded protein exhibited similarity to a variety of Type II 5-methylcytosine methyltransferases, including M.LlaKR2I and M.Sau3AI. Cloning and integration of the sth368IRM genes into the chromosome of S. thermophilus A054 resulted in signiﬁcant resistance to phage ST84 and reduced the efﬁciency of plaquing (EOP) to approximately 104. This was the ﬁrst report of using a native R-M system to enhance the phage resistance of a closely related strain of S. thermophilus. Prior to that, the only other report that described the use of an R-M system to enhance the level of phage resistance in S. thermophilus described the expression of a heterologous R-M system. Moineau et al. (1995) found that the plasmidborne expression of the lactococcal LlaDCHI (formerly LlaII) conferred broad-range bacteriophage resistance in various strains of S. thermophilus. When expressed from pNZ123, a high-copy-number vector, the expression of LlaDCHI reduced the EOP to between 105 and 108, depending on the phage and host background tested (Moineau et al., 1995). As seen in Table I, two complete Type I R-M systems have been identiﬁed in S. thermophilus. Type I systems differ from Type II, III, and IV R-M systems in that sequence speciﬁcity is not determined by their respective REase or MTase subunits (Roberts et al., 2003a). Rather, speciﬁcity is directed by a third subunit called a host speciﬁcity determinant (HsdS). Schouler et al. (1998) ﬁrst described the existence of plasmid-borne hsdS genes in lactococci and demonstrated that the encoded speciﬁcity subunits interacted with chromosomally encoded HsdR and HsdM proteins. Since then, speciﬁcity domain stacking has been suggested to be a means of augmenting the phage resistance of S. thermophilus strains (O’Sullivan et al., 1999). For example, the 6.5 kb plasmid pCI65st isolated from the S. thermophilus strain NDI-6 encodes a putative HsdS protein (O’Sullivan et al., 1999). Interestingly, the plasmid-bearing parent was resistant to phage bas19, whereas a pCI65st-cured derivative was found to be sensitive to phage bas19, suggesting that the HsdS played a role in modulating phage resistance. It had been reported previously that another S. thermophilus plasmid, pCRB33, was similarly able to confer bacteriophage

348

STURINO AND KLAENHAMMER

resistance to a variety of S. thermophilus strains (Cocconcelli et al., 1995). This is believed to be a natural process in both organisms, since several HsdS domains (i.e., without the cognate REase and/or MTase) have been found on the same or different native plasmids (Forde et al., 1999; O’Sullivan et al., 1999; Schouler et al., 1998; Seegers et al., 2000), but are also found in the host chromosome (Schouler et al., 1998; Seegers et al., 2000). In addition, variable speciﬁcity domains can be shufﬂed between different hsdS genes via recombination and can result in novel HsdS subunits with altered speciﬁcities (Fuller-Pace and Murray, 1986). R-M systems confer an added beneﬁt at the population level in that they actually remove lytic phages from the environment that might have otherwise gone on to infect other R-M hosts that might be grown in co-culture. R-M systems are powerful defenses; however, failure of either component can have dire consequences not only for the infected bacterium but also for the population as a whole (for reviews, see Bickle and Kruger, 1993; Kruger and Bickle, 1983). At some frequency, a ﬁnite number of phage genomes will escape restriction and become modiﬁed by the bacterial MTase during replication. Once modiﬁed, the phage genome becomes impervious to the cognate REase and can then initiate its developmental program unimpeded. As a result, the progeny phages that result from this infection will be modiﬁed by the host MTase by default. Once released, the phages will be able to circumvent the R-M systems of neighboring bacteria in subsequent infections, thereby rendering this particular R-M system ineffective at the population level. The initial modiﬁcation event is a random epigenetic event that leads to heritable resistance to the cognate REases and REases that share similar core recognition and/or cleavage sequences. The genetic capacity of the bacteriophage is unchanged, however. As such, the conferred resistance is host dependent. Invasion of a homologous, R-Mþ host by the modiﬁed phage will lead to the generation of R-Mþ resistant phages, whereas invasion of an R-M host will lead to the generation of R-M sensitive phages. The chances of evading restriction via modiﬁcation of the genome are dependent on the R-M system and the number of restriction sites encoded on the phage genome. Since it has been found that the EOP of the phage decreases logarithmically as the number of sites in the phage genome increases, the elimination of even a single REase recognition site by deletion or point mutation can dramatically increase a phage’s chances of evading the REase via this method (Moineau et al., 1993; Powell and Davidson, 1986; for a review, see Wilson and Murray, 1991).

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

349

Industrial strains of S. thermophilus typically encode one or more R-M systems. As a result of these extreme selective pressures, phages have evolved a variety of genetic countermeasures that allow mutant derivatives to circumvent these defenses (for reviews, see Bickle and Kruger, 1993; Kruger and Bickle, 1983). Among these countermeasures are (1) the acquisition of methyltransferase genes from the host genome via recombination (Hill et al., 1991b), (2) the incorporation of modiﬁed nucleotides into bacteriophage DNA, (3) the production of antagonistic proteins that inhibit the activity of bacterially encoded restriction endonucleases, and (4) the elimination of restriction endonucleases recognition sites throughout the genome (Moineau et al., 1993). To date, none of these escape countermeasures have been observed in S. thermophilus. In lactococci, the identiﬁcation, characterization, and—importantly— the functional exploitation of these and other defense systems has extended the utility of many industrially important strains, especially when used in rotation (Durmaz and Klaenhammer, 1995; Sing and Klaenhammer, 1993). This has been greatly facilitated by the fact that R-M systems are typically, although not exclusively, plasmid-encoded in lactococci. Of the 10 R-M systems that have been identiﬁed in lactococci, eight have been plasmid-encoded. In addition, several of these plasmids, such as pTR2030, are self-transmissible, which greatly facilitates the introduction of the phage resistance plasmids into lactococcal starter cultures (Klaenhammer and Sanozky, 1985). Increased use of S. thermophilus starter cultures worldwide has resulted in an increased incidence of phage attacks, which has prompted the study of S. thermophilus R-M systems. The vast majority of R-M systems identiﬁed in S. thermophilus have been chromosomally encoded (70%), which has made them more difﬁcult to clone and characterize (Table I). C. ABORTIVE INFECTION If a phage circumvents restriction, abortive infection (Abi) defense systems can act as the second line of intracellular defense against bacteriophages (for reviews, see Klaenhammer, 1987; Klaenhammer et al., 1991). In lactococci, abortive defense systems that interfere with many steps in the lytic life cycle have been identiﬁed, including genome replication, transcription, translation, encapsidation, and particle morphogenesis. Twenty-four genes have been shown to abort the replication of lactococcal bacteriophages (for reviews, see Allison and Klaenhammer, 1998; Coffey and Ross, 2002; Forde and Fitzgerald, 1999). The vast majority of these genes show no homology to one

350

STURINO AND KLAENHAMMER

another or to any other entry in the GenBank database. Of the 12 phage species that have been identiﬁed, only three are responsible for the majority of industrial phage attacks (i.e., P335, c2, and 936) (Jarvis et al., 1991). Based on this, the spectrum of phage protection conferred by any one of the Abi defenses may be rated as narrow or broad. Defenses with narrow speciﬁcity are conﬁned to affect a single phage species (e.g., AbiB, AbiE, AbiH, and AbiJ), whereas defenses with broad speciﬁcity are able to affect constituents of two (e.g., AbiC, AbiD, AbiD1, AbiF, AbiG, AbiI, and AbiL) or three (AbiA, AbiK and AbiU) phage types. Not all of the defenses have been tested against all three phage species, however. It is important to note that these defenses may not be effective against all phages within a given species but may be effective only against a handful of phages within the species in question. Abi systems are typically characterized as acting either prior to or at the level of genome replication (i.e., early) or after genome replication (i.e., late) (Garvey et al., 1995). Native abortive defense mechanisms typically result in the death of the host cell; however, it is not known if these systems kill the host by design (i.e., a form of altruism) or if this is simply a function of arresting phage development. A single reference to native abortive infection in S. salivarius subsp. thermophilus NST5 has been described in the literature; however, the putative defense system(s) have not been cloned or characterized (Larbi et al., 1992). The activity was found to be temperature dependent, active at 42 C but not at 30 C, as measured by a signiﬁcant increase in both plaque size and EOP at the non permissive temperature. This temperature-dependent activity of Abi systems had previously been shown in lactococci (Klaenhammer and Sanozky, 1985). Recently the AbiA and AbiG abortive defense systems isolated from L. lactis subsp. cremoris AC8147 and LOC735, respectively, were tested for their ability to confer resistance to phages in S. thermophilus 4035 (Tangney and Fitzgerald, 2002). In this study, the AbiA system was shown to be effective at 30 C, but not at 37 or 42 C. It is important to note that 30 C is optimal for the growth of many lactococci. Unfortunately, the permissive temperature for AbiA function in S. thermophilus is well below its optimum for propagation and, importantly, the temperatures at which thermophilic processes are generally conducted during product manufacture (i.e., between 40 and 45 C). As seen against lactococcal phages, AbiA interfered with phage genome replication (Hill et al., 1991a). BlastP analysis detected a putative reverse transcriptase domain (pfam00078) within the AbiA primary amino acid structure; however, the function of AbiA is not yet known. In contrast, the AbiG system failed to interfere with S. thermophilus

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

351

phage replication at any temperature tested. These two systems were speciﬁcally chosen because they are effective against lactococcal P335 phages, which share up to 60% sequence similarity at the nucleotide level with regions of S. thermophilus phages Sﬁ19 and Sﬁ21 genomic DNA (Chopin et al., 2001). V. Recent Advancements in Genomics A. COMPARATIVE GENOMICS Our understanding of S. thermophilus bacteriophages has progressed rapidly since the release of six bacteriophage whole-genome sequences: DT1 (Tremblay and Moineau, 1999), O1205 (Stanley et al., 1997), Sﬁ11 (Lucchini et al., 1998), Sﬁ19 (Lucchini et al., 1999a), Sﬁ21 (Lucchini et al., 1999a), and 7201 (Stanley et al., 2000). With regard to the genetic content and genomic organization of these phages, they exhibit signiﬁcant similarities to other members of the super family of phages (Bru¨ssow and Desiere, 2001). Bioinformatic analyses have provided signiﬁcant insight into the evolution of S. thermophilus phages, revealing that their genomes are molecular mosaics assembled on a relatively simple scaffold consisting of independently evolving modules such that each module directs distinct developmental processes (e.g., DNA replication) (Lucchini et al., 1999b). In addition, functionally coupled genes were found to be arranged into conserved clusters within these modules, such that gene order was predicated by developmental order, or vice versa. These clusters were generally, but not exclusively, found to be arranged into putative operons such that they are co-transcribed and, likely, co-regulated (Ventura et al., 2002b). Interestingly, one or more distinct pathways encoded on alternative and interchangeable modules can carry out each developmental process (Lucchini et al., 1999b). For example, two DNA replication modules have been identiﬁed among the various S. thermophilus phages: Sﬁ21-type and 7201-type. The Sﬁ21-type replication module is composed of a single origin of DNA replication and several open reading frames that encode a putative helicase, a putative primase, and a number of other proteins of undetermined function (Lucchini et al., 1999b). The phage 7201-type replication module, on the other hand, includes two distinct oris and encodes a probable single-stranded DNA binding protein, a putative replication protein, a putative DnaC homolog, and a number of other proteins of undetermined function (Stanley et al., 2000). The variants of the Sﬁ21-type genome replication module are found among the majority of industrial phage isolates (Bru¨ssow et al., 1994b).

352

STURINO AND KLAENHAMMER

These modules are subject to both horizontal and vertical evolution. The horizontal evolutionary component is primarily the result of recombination between interbreeding phage populations, including resident prophages and remnants thereof, but may also occur between an invading phage and the host genome (Desiere et al., 1998). As a result of horizontal evolution, on the other hand, two parent phages produce a chimeric third. The vertical evolutionary component is the gradual accumulation of spontaneous mutations, predominantly point mutations, but also short insertions and deletions, passed from parent to progeny (Desiere et al., 1998). The availability of a growing number of bacteriophage and bacterial genomes has become an invaluable resource for the understanding these interactions. Bacteriophage genomes may be analyzed to identify conserved gene targets. Developmental pathways can be scrutinized to anticipate evolutionary escape routes. Analysis of bacterial sequences may be used to identify native defense systems or to facilitate the removal of host-encoded factors involved in phage replication or evolution. Given this, special focus is given to the impact of genomic data on the development of novel, genetically engineered solutions to the phage problem. These information-based strategies include the use of origin-conferred phage-encoded resistance, phage-derived antisense RNAs, and the overexpression of phage-encoded proteins, or mutant derivatives thereof. B. ANTI-RECEPTOR IDENTIFICATION Until recently, phage-host interactions remained poorly understood in most Gram-positive bacteria, including the LAB. Signiﬁcant advances have been made in this area, largely because of insights garnered from comparative genomic analyses. Based on these analyses, several clues suggested that the S. thermophilus phage DT1-encoded orf18 gene the phage anti-receptor (Tremblay and Moineau, 1999). First, orf18 is located within the same region as the coliphage antireceptor gene, J (Werts et al., 1994). Second, the deduced proteins, gp18 from DT1 and gpJ from coliphage , have similar molecular weights and isoelectric points (pI). Finally, gp18 exhibits a modular organization of conserved and hyper-variable regions shared with the anti-receptor genes of T-even phages (Te´tart et al., 1996, 1998). This last point was explored further in a later study by Duplessis and Moineau (2001). The authors aligned the orf18 gene from S. thermophilus phage DT1 with alleles from six other phages. The deduced gp18 proteins were divided into three distinct domains. The ﬁrst, the

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

353

N-terminal domain (491 residues), was highly conserved among the seven phages, exhibiting 83–100% identity at the amino acid level between variants. The second domain (approximately 400 residues) was found in only two of the seven phages (MD2 and DT2). Domain 2 also contained a short (133 residue) internal variable region (VR) called VR1. The second (central) domain was demarcated by two collagen-like repeats. The third, the C-terminal domain, was present in all seven phages and contained another variable region (VR2) (145 residues). VR2 was also found in phages Sﬁ11, Sﬁ19, Sﬁ21, O1205, and 7201, the ﬁve other phages that have been completely sequenced. The authors went on to use recombination-mediated domain swapping to generate chimeric variants of phage DT1 that had acquired the host range of MD4 to provide convincing biological evidence in support of their claim that orf18 encoded the phage host speciﬁcity determinant (Duplessis and Moineau, 2001). Prior to this, a spontaneous deletion mutant of phage Sﬁ21, designated D3, was isolated through routine serial propagation. In this case, the second domain between the collagen-like repeats III and I was deleted in the mutant, which indicated that these recombination hotspots contribute to the allelic diversity within the population (Bruttin and Bru¨ssow, 1996). The nucleotide sequences encoding the collagen-like repeats had previously been suggested to be hotspots for recombination-mediated gene shufﬂing (Desiere et al., 1998). These motifs consist of repeated amino acid triplets where glycine is the ﬁrst residue in each triplet (GX2)n (Beck and Brodsky, 1998). VI. Engineered Defense Systems In addition to those native bacteriophage defenses described previously, a wide variety of engineered phage resistance strategies have been constructed. With regard to the practical efﬁcacy of antisense cassettes in the dairy environment, identiﬁcation of target genes that are effective against a variety of industrially relevant bacteriophages is of utmost importance. When engineering phage-encoded resistance systems, there are potential beneﬁts of implementing comparative genomics as an initial screen for choosing potential targets. These analyses expedite the identiﬁcation of well-conserved genes and cis regulatory elements shared between the various genomes, while conversely enabling the elimination of poorly conserved targets in silico. These advantages are of critical importance when engineering defense strategies intended for use in large-scale industrial settings, where protection is required against both the residing and potentially

354

STURINO AND KLAENHAMMER

emerging phage populations. In many cases, however, these defense systems have been tested in the laboratory but have not been utilized during product manufacture because of consumer reluctance and/or government mandate (Scannerini, 2003). A. SUPERINFECTION EXCLUSION AND IMMUNITY Approximately half of the bacterial genomes sequenced to date contain prophage-associated sequences (Lawrence et al., 2001). Examination of these lysogens reveals that prophage or prophage-remnant sequences comprise between 3–10% of the total genomic content of lysogens (Bru¨ssow and Hendrix, 2002). Conceptually, this extra genomic content signiﬁcantly increases the metabolic burden of the host, which should signiﬁcantly decrease the ﬁtness of the lysogen relative to prophage-free strains. Interestingly, prophage-containing strains abound in nature. This is apparently due to the fact that prophages often provide beneﬁts to the lysogen by encoding factors that may increase the ﬁtness of the bacterium, such as the lysogenic conversion genes (Desiere et al., 2002). The genes encoding lysogenic conversion functions are located between the lysin gene and the phage attachment site (Desiere et al., 2002). In the pathogenic streptococci, prophages often encode virulence factors. For instance, the streptococcal erythrogenic toxin A gene is encoded by a variety of S. pyogenes prophages, including phage T12 (Yu and Ferretti, 1991). While prophages associated with LAB certainly do not encode virulence factors, they do encode genes that may provide some beneﬁt to the lysogen. In S. thermophilus, the lysogenic conversion genes are among the small handful of phageencoded genes that are actively and consistently transcribed by the prophage (Ventura et al., 2002a). From a phage resistance point of view, superinfection exclusion and immunity genes are well-characterized examples of beneﬁcial genes associated with the prophages of Gram-positive bacteria, including lactococci (Bruttin et al., 1997b; McGrath et al., 2002). These genes protect lysogens from becoming infected with additional phages. In the S. thermophilus temperate phage Sﬁ21, RNAs speciﬁc for orf203 were found to be the predominant phage-speciﬁc transcripts detected in Sﬁ21 lysogens (Ventura et al., 2002a). This gene, which is located upstream of the phage integrase, was shown to mediate superinfection exclusion when it was expressed from a high-copy-number plasmid, and it resulted in signiﬁcant protection from a diverse collection of lytic bacteriophages (Bruttin et al., 1997b). In a related form of resistance, Durmaz et al. (2002) recently cloned two derivates of the cI

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

355

repressor from the lactococcal P335-type phage 31 into the high-copynumber plasmid pTRKH2. The ﬁrst construct (pTRKH2::CI-per1) harbored an ochre mutation in the cI gene after the ﬁrst 128 amino acids of the predicted 180-amino-acid protein, whereas the second construct (pTRKH2::CI-per2) was completely devoid of the sequences downstream of the ochre mutation. The EOP of 31 was reduced to 106 in the presence of pTRKH2::CI-per1 construct, whereas EOP was further reduced to less than 107 in the presence of pTRKH2::CI-per2 construct. The authors went on to show that 12 of 16 heterologous lytic P335-type phages were completely inhibited by pTRKH2::CI-per2, while four phages were completely resistant to the defense system. B. ANTISENSE RNA Antisense RNAs may be used to interfere with phage development by inhibiting the translation of phage-encoded genes necessary for normal development. Mechanistically, antisense RNA hybridizes to the sense RNA strand and creates a translationally inactive double stranded RNA (dsRNA) molecule (for a review, see Inouye, 1988) (Fig. 2). Formation of the dsRNA molecule silences gene expression through the cooperative action of one or more intermolecular mechanisms. If the antisense RNA includes sequences complementary to the ribosome-binding site (RBS), then the formation of dsRNA may mask the RBS, thereby preventing efﬁcient ribosome loading and reducing translation of the gene of interest. Formation of dsRNA downstream of the RBS may also interfere with translation by sterically impeding, to some degree, the procession of the mRNA through the ribosome. In addition, the formation of dsRNA may destabilize the sense mRNA by promoting the action of dsRNA-speciﬁc ribonucleases. Last, if the gene of interest is transcribed on a polycistronic mRNA, then antisense targeting may also negatively impact the expression of translationally coupled genes located downstream, causing pleotropic effects that might further inhibit bacteriophage proliferation. Watson-Crick base pairings between the antisense RNA and its complementary, phage-encoded target mRNA(s) are initiated through a limited number of intermolecular nucleation events occurring within complementary unstructured regions, including 50 or 30 single-stranded tails and internal loops and/or bulges (Hjalt and Wagner, 1995; Kolb et al., 2001). These associations are subsequently stabilized through progressive Watson-Crick base-pairings proximal to the initial nucleation site(s). During this process, intermolecular base pairing occurs at the expense of intramolecular interactions individually associated

356

STURINO AND KLAENHAMMER

FIG. 2. Proposed mechanism of antisense RNA-mediated gene silencing. Antisense RNAs interfere with phage development by inhibiting the translation of phage-encoded genes necessary for normal development. Mechanistically, antisense RNA hybridizes to the sense RNA strand and creates a translationally inactive dsRNA molecule, which is subject to degradation. Abbreviations: RNAP, RNA polymerase; orfA, open reading frame targeted by antisense RNA; orfD, downstream orf; gpD, gene product D; RNase, dsRNA-speciﬁc ribonuclease.

with both the antisense and the target RNA molecules. Previous studies in vitro demonstrated that two strands of even short (less than 200-nt) complementary RNAs become ﬁxed in thermodynamic troughs and may not form double-stranded RNA over the entire length of both molecules (Kolb et al., 2001). The remaining single-stranded regions were found to either remain unpaired or to undergo extensive intramolecular associations. If this is also true in vivo, it is possible that a single antisense RNA may interact with and negatively impact more than one target RNA at a time, especially when the antisense RNA is exceptionally long, as is the case of the engineered varieties that have been expressed routinely in the dairy LAB for phage defense. The genomes of six S. thermophilus phages were compared to identify genes that could be targeted by antisense RNA defense systems with potentially widespread efﬁcacy. The genes associated with the

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

357

Sﬁ21-type genome replication module, including a putative primase and helicase, were found to be among the best candidates because of their frequency of distribution in industrial phage isolates, striking sequence conservation between independent isolates, and intrinsic strategic importance in early phage development (Sturino and Klaenhammer, 2002). The phage 3-derived putative helicase (hel3) and putative primase genes (pri3) were cloned in the antisense orientation behind a strong promoter and expressed from a high-copy number vector (Sturino and Klaenhammer, 2002, 2004). These antisense RNAs consistently reduced the EOP of phage 3 to between 5 101 and 2 103, depending on the (1) gene targeted and (2) region of the gene that was targeted by antisense RNA; see Table II for a list of all of the antisense RNA constructs that have been evaluated in S. thermophilus to date. A 1.5-kb antisense RNA complementary to the complete S. thermophilus phage 3-derived putative helicase gene (hel3.1-AS) inhibited the proliferation of independently isolated Sﬁ21-type phages, mediating a phage-speciﬁc 40–70% reduction in the EOP with a concomitant reduction in plaque size (Sturino and Klaenhammer, 2002). In a second study, the expression of pri3.10-AS, a 1.5-kb antisense RNA complementary to the S. thermophilus phage 3-derived putative primase gene resulted in a 2.7-log cycle reduction in EOP and a 50% reduction in ECO1 formation, meaning that only one of every two phage-infected cells released progeny, even when only the ﬁrst lytic cycle was impeded by antisense RNA (Sturino and Klaenhammer, 2004a). Given the effectiveness of pri3.10-AS expression on phage 3 development, efforts were made to identify regions that were more or less important for optimal efﬁcacy in vivo. To address this, the pri3.10-AS region was systematically reduced through the construction of 12 additional subclones, which spanned various structural or putative regulatory regions of the primase gene. The expression of 13 antisense constructs resulted in statistically signiﬁcant reductions in EOP that ranged from 0.2- to 2.7-log cycles. The largest antisense RNAs were generally found to confer the largest reductions in EOP; however, shorter antisense RNAs designed to the 50 region of the gene retained much of the inhibitory function. Larger antisense RNAs may (1) have more opportunities over their length to maximize intermolecular base pairing and thus exert their inhibitory effects, or (2) exhibit decreased stability when bound to the target RNA (Sturino and Klaenhammer, 2004a). No mutant phages were recovered that were insensitive to either primase- or helicase-targeted antisense RNA after numerous attempts to select or enrich for derivatives (Sturino and Klaenhammer, 2002, 2004a). Because of the length of

TABLE II ANTISENSE RNA-BASED PHAGE DEFENSE STRATEGIES EFFECTIVE (EOP < 1) Antisense construct

Base Vector

Promoter

Phage

Predicted target

Function

pTRK787:: pri3.4-AS

pNZ123

P6

3

Primase

Early

pTRK788:: pri3.7-AS

pNZ123

P6

3

Primase

pTRK789:: pri3.8-AS

pNZ123

P6

3

pTRK790:: pri3.10-AS

pNZ123

P6

pTRK791:: pri3.11-AS

pNZ123

pTRK792:: pri3.12-AS

Lengtha

IN

STREPTOCOCCUS THERMOPHILUS

ORFb

anti-RBSc

EOP

Reference

587

P

Yes

1.4 101

Sturino and Klaenhammer (2004a)

Early

1,023

P

Yes

3.2 102

Sturino and Klaenhammer (2004a)

Primase

Early

519

P

Yes

3.8 102

Sturino and Klaenhammer (2004a)

3

Primase

Early

1,512

C

No

2.0 103

Sturino and Klaenhammer (2004a)

P6

3

Primase

Early

1,008

P

No

1.1 101

Sturino and Klaenhammer (2004a)

pNZ123

P6

3

Primase

Early

504

P

No

9.3 102

Sturino and Klaenhammer (2004a)

pTRK793:: pri3.13-AS

pNZ123

P6

3

Primase

Early

1,506

P

No

3.2 103

Sturino and Klaenhammer (2004a)

pTRK794:: pri3.14-AS

pNZ123

P6

3

Primase

Early

1,486

P

No

4.8 103

Sturino and Klaenhammer (2004a)

pTRK795:: pri3.15-AS

pNZ123

P6

3

Primase

Early

982

P

No

7.1 102

Sturino and Klaenhammer (2004a)

pTRK796:: pri3.16-AS

pNZ123

P6

3

Primase

Early

487

P

No

2.1 102

Sturino and Klaenhammer (2004a)

pTRK797:: pri3.18-AS

pNZ123

P6

3

Primase

Early

1,008

P

No

1.5 101

Sturino and Klaenhammer (2004a)

pTRK798:: pri3.19-AS

pNZ123

P6

3

Primase

Early

504

P

No

6.1 101

Sturino and Klaenhammer (2004a)

pTRK799:: pri3.21-AS

pNZ123

P6

3

Primase

Early

504

P

No

4.0 101

Sturino and Klaenhammer (2004a)

pTRK689:: hel3.1-AS

pNZ123

P6

3

Primase

Early

1,431

?

No

5 101

Sturino and Klaenhammer (2002)

a b c

Length in nucleotides unless otherwise indicated. C, complete open reading frame (start to stop codon); P, partial open reading frame. Presence or absence of sequences complementary for the predicted ribosome binding site (RBS).

360

STURINO AND KLAENHAMMER

the antisense RNAs, it was proposed that the potential for multiple associations could limit the ease at which phages might overcome antisense inhibition via point mutation(s) (Bull et al., 1998; Sturino and Klaenhammer, 2004a). Antisense constructs that contained sequences complementary to the putative RBS generally reduced the EOP below the level of constructs that lacked them. This may be due to the formation of dsRNA over the length of the RBS, thus preventing efﬁcient ribosome loading and reducing translation of the target gene (Sturino and Klaenhammer, 2002, 2004a). The expression of these antisense RNAs also retarded phage genome replication and severely limited the number of progeny phages released from an infected cell, which indicated that the antisense RNAs acted to abort phage infection (Sturino and Klaenhammer, 2004a). Antisense RNAs generally reduced the abundance of the target transcript in a manner consistent with the observed EOP reductions. Further, antisense RNAs that strongly reduced EOP also resulted in the synthesis of fewer phage genomes, indicating a correlation between the lowered abundance of primase transcripts, lowered levels of genome replication, and interference with progeny development. Reduced primase transcript abundance was accompanied by a concomitant decrease in antisense RNA abundance, suggesting that both RNA species were degraded by ribonuclease(s). Table III lists antisense RNA targets that were effective against lactococcal phages. In lactococci, six genes putatively involved in lactococcal P335-type phage genome replication were targeted with antisense RNA (McGrath et al., 2001). The targeted genes were orf14 (encoding a putative topoisomerase), orf15 (putative single-stranded DNA binding protein), orf16 (putative replisome organizer), orf18 (putative methylase), and two open reading frames encoding proteins of undetermined function (i.e., orf17 and orf19). For each gene, the expressed antisense RNAs were complementary to the complete open reading frame, including its upstream putative RBS. When challenged with four different P335-type phages, the authors found that the expression of antisense RNAs speciﬁc for orf14, orf15, and orf18 each reduced the EOP of phage Tuc2009 tenfold but did not have any effect on phages Q30, Q33, or ul36. In contrast, the expression of orf16 and orf17 conferred signiﬁcant and highly variable resistance to all four phages, as measured by 0.5 to 106 log reductions in EOP. Antisense RNA speciﬁc for orf19 failed to inhibit any of the four phages. Kim et al. (1992) found that antisense expression of two polycistronic open reading frames, designated gp18C and gp24C, inhibited the P335-type phage 7–9, as measured by a 55% reduction in EOP. The reduction in EOP dropped to

TABLE III ANTISENSE RNA-BASED PHAGE DEFENSE STRATEGIES EFFECTIVE (EOP < 1.0) AGAINST P335-TYPE PHAGES IN LACTOCOCCUS LACTIS Antisense construct

Base Vector

pNZ44::topo-rev pNZ123

Promoter P44

Phage

Predicted target

Tuc2009 Topoisomerase

Function Lengtha ORFb anti-RBSc Early

640

C

Yes

EOP 1

1 10

1

Reference McGrath et al. (2001)

pNZ44::ssb-rev

pNZ123

P44

Tuc2009 ssDNA binding protein

Early

478

C

Yes

1 10

McGrath et al. (2001)

pNZ44:: rep2009-rev

pNZ123

P44

Tuc2009 Replisome organizer

Early

802

C

Yes

1 106

McGrath et al. (2001)

pNZ44:: orf17-rev

pNZ123

P44

Tuc2009 Topoisomerase

Early

747

C

Yes

1 106

McGrath et al. (2001)

pNZ44:: meth-rev

pNZ123

P44

Tuc2009 Methylase

Early

773

C

Yes

5 101

McGrath et al. (2001)

pSGK1.0R:: pGKV210 gp18C::gp24C

P59

mi7-9

Glycoprotein:: (unknown)

Early

1.0 kb

P::C

No:: yes

6.9 101 Kim et al. (1992)

pSGK1.5R:: pGKV210 gp18C::gp24C

P59

mi7-9

Glycoprotein:: (unknown)

Early

1.5 kb

C::C

Yes:: yes

4.5 101 Kim et al. (1992)

pDC100::mcp

pGKV210

P59

F4-1

Major capsid protein

Late

926

C

Yes

5.8 101 Chung et al. (1992)

pSC1::mcp-222

pGKV210

P59

F4-1

Major capsid protein

Late

301

P

Yes

5.0 101 Chung et al. (1992)

pDC101:: mcp-246

pGKV210

P59

F4-1

Major capsid protein

Late

227

P

Yes

7.7 101 Chung et al. (1992)

pSGK1.6R:: gp51C

pGKV210

P59

mi7-9

Translation factor?

1,654

C

Yes

4 103

a b c

?

Length in nucleotides unless otherwise indicated. C, complete open reading frame (ORF) from start to stop codon; P, partial open reading frame. Presence or absence of sequences complementary for the predicted ribosome binding site (RBS).

Kim and Batt (1991)

362

STURINO AND KLAENHAMMER

30% if the RBS and coding region for the ﬁrst 15 amino-terminal residues of gp18C were omitted from the antisense construct. In both cases, the plaque size was also reduced by approximately tenfold. Chung et al. (1992) obtained variable reductions in EOP, which ranged between 0.5 and 0.8, as they expressed different lengths of the F4–1 major coat protein (mcp) gene. Kim and Batt (1991) found that antisense expression of the full-length, phage 7–9-derived gp15C mediated a 100-fold reduction in the EOP of 7–9 and other gp15C-containing phages. In these and other studies, the effectiveness of antisense RNA-based phage defense strategies has been highly variable, exhibiting both target- and phage-speciﬁc differences. The collective observations thus far do suggest some key characteristics of an ideal antisense RNA target. Genes that are transiently expressed, expressed at a very low level, and/or coded for by unstable, inefﬁciently translated mRNAs should make excellent candidates for antisense RNA targeting. From a practical standpoint, the target RNA must be essential for phage development, or at least critical to the synthesis or maturation of virulent progeny phages. Table IV lists antisense RNA targets that were ineffective against lactococcal phages. In general, antisense RNAs that target early-expressed genes involved in genome replication (McGrath et al., 2001; Sturino and Klaenhammer, 2002, 2004a) have been more effective targets than genes expressed later in the lytic cycle (McGrath et al., 2001; Walker and Klaenhammer, 2000). It is important to note, however, that not all genes involved in genome replication are effective targets. Polzin et al. (1996) found that the antisense expression of four early open reading frames, including e5 (encoding a putative subunit of DNA polymerase), e12 (putative transcription regulator), and e15 (putative recombinase), were all ineffective in their ability to inhibit the replication of the lactococcal prolate-headed phage c2, regardless of the gene dosage tested. In this context, the failure of these strategies to inhibit phage replication may have been due to functional complementation by host-encoded factors or a more general recalcitrance to antisense RNA, which might be mediated by differences in phage c2-directed RNA metabolism. C. ORIGIN-DERIVED PHAGE-ENCODED RESISTANCE Origin-derived phage-encoded resistance (PER) was ﬁrst reported to be effective in L. lactis (Hill et al., 1990). When a bacteriophage origin of genome replication (ori) is provided in trans on a recombinant plasmid, the origin acts as a molecular decoy that competes for and

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

363

titrates away both bacteriophage- and host-derived replication factors that catalyze phage genome replication. As a result, the number of bacteriophage genomes replicated over the course of the lytic infection is reduced. In addition, the plasmid-associated replication factors catalyze a dramatic increase in plasmid copy-number. Origin-derived PER has been found to be highly dependent on gene dosage, and can be dependent on plasmid copy-number (O’Sullivan et al., 1993) or the number of copies of the ori that are cloned within the same plasmid (McGrath et al., 2001). Origin-derived PER has recently been applied to S. thermophilus strains. Foley et al. (1998) were the ﬁrst to use origin-derived PER as a means of increasing phage resistance in S. thermophilus. They found that the phage Sﬁ21-derived ori conferred strong resistance to related phages, as measured by at least seven log cycle reductions in the number of PFU/ml obtained when grown in broth. The authors pointed out that the phage Sﬁ21-derived ori shows 80% sequence similarity to the putative single-strand origin of the cryptic S. thermophilus plasmid, pST1. These results suggested that the phage- and plasmidencoded oris might share a common ancestor. More recently, Stanley et al. (2000) identiﬁed four loci from two different S. thermophilus phage genomes that were able provide origin-derived PER. These authors divided 11 phages into two replication groups: group I and group II. At least one of the phages in replication group I (i.e., phage O1205) encoded an Sﬁ21-type DNA replication module, whereas at least one member of the replication group II phages (i.e., phage 7201) encoded a different (non-Sﬁ21-type) genome replication module. It is not known, however, if all of the group I or group II phages encoded variants of the Sﬁ21-type or 7201-type replication modules, respectively. As seen in lactococci (O’Sullivan et al., 1993), the presentation of the phage O1205-derived origin (ori1205) on a low-copy-number vector failed to provide protection from the homologous phage but provided protection from group I type phages when cloned onto the high-copynumber plasmid pNZ8048 (Stanley et al., 2000). The range of protection provided by this plasmid was phage-speciﬁc, and the EOP for these phages ranged between 103 to less than 107. Interestingly, the pORI1205 construct was also able to reduce the plaque diameter of three out of the ﬁve group II phages that were tested but did not have any impact on EOP of any of these phages. These results suggested that a common host-encoded protein is required for the replication of the two phage groups, albeit to varying degrees. In addition, one of the phage O1205-derived PER constructs (p1205-orf9) provided signiﬁcant

TABLE IV ANTISENSE RNA-BASED PHAGE DEFENSE STRATEGIES INEFFECTIVE (EOP ¼ 1) AGAINST VARIOUS PHAGES IN LACTOCOCCUS LACTIS Antisense construct

Base Vector

pNZ44:: orf19-rev

pNZ123

Promoter

Predicted target

Phage

Function Lengtha ORFb anti-RBSc

Reference

P44

Tuc2009 Resolvase

Early

410

C

Yes

McGrath et al. (2001)

pGKV259::e5 pGKV210

P59

c2

DNA pol. subunit

Early

?

?

?

Polzin et al. (personal communication, 1996)

pGKV259:: e12

pGKV210

P59

c2

Transcription factor

Early

?

?

?

Polzin et al. (personal communication, 1996)

pGKV259:: e15

pGKV210

P59

c2

Recombinase

Early

?

?

?

Polzin et al. (personal communication, 1996)

pTRK594:: anti-orf1

pTRKH2

P6

31

?

Early or middle

?

C

?

Walker and Klaenhammer (2000)

pTRKH2 pTRK595:: anti-(orf1:: tac31A)

P6

31

?/ Transcriptional activ.

Early or middle

?

C::C

?

Walker and Klaenhammer (2000)

pNZ44:: msp1-rev

pNZ123

P44

Tuc2009 Major structural prot.

Late

551

C

Yes

McGrath et al. (2001)

pNZ44:: msp2-rev

pNZ123

P44

Tuc2009 Major structural prot.

Late

546

C

Yes

McGrath et al. (2001)

pTRK596:: anti-orf3

pTRKH2

P6

31

?

Late

358

C

?

Walker and Klaenhammer (2000)

pTRK597:: anti-orf4H

pTRKH2

P6

31

?

Late

361

P

Yes

Walker and Klaenhammer (2000)

pTRK598:: anti-orf5H

pTRKH2

P6

31

?

Late

493

P

Yes

Walker and Klaenhammer (2000)

pTRK599:: anti-orf6H

pTRKH2

P6

31

?

Late

467

P

Yes

Walker and Klaenhammer (2000)

pTRK600:: anti-orf6

pTRKH2

P6

31

?

Late

?

C

Yes

Walker and Klaenhammer (2000)

pGKV259::17 pGKV210

P59

c2

Major tail protein

Late

?

?

?

Polzin et al. (1996)

pGKV259:: 112

pGKV210

P59

c2

Terminase

Late

?

?

?

Polzin et al. (1996)

pSGK1.0R:: gp51C

pGKV210

P59

mi7-9

Translation factor

?

695

P

No

Kim and Batt (1991)

pSGK0.8R:: gp51C

pGKV210

P59

mi7-9

Translation factor

?

422

P

No

Kim and Batt (1991)

a b c

Length in nucleotides unless otherwise indicated. C, complete open reading frame (start to stop codon); P, partial open reading frame. Presence or absence of sequences complementary for the predicted ribosome binding site (RBS).

366

STURINO AND KLAENHAMMER

protection to bacteriophages from both replication groups in one host (i.e., strain CNRZ1205-3); however, it failed to provide detectable levels of phage protection in another background (i.e., strain 4035). These results clearly implicated the importance of certain host-encoded factors in phage replication and suggested that the available levels of these factors were also critical to confer phage resistance (Stanley et al., 2000). D. SUBUNIT POISONING The expression in trans of mutant protein subunits can suppress, or poison, the function of native, multimeric proteins in a dominant negative fashion (Herskowitz, 1987). Subunit poisoning has recently been used as a novel phage defense system in S. thermophilus (Sturino and Klaenhammer, 2004b). In this study, the authors describe the use of multiple alignments of related target protein sequences in order to identify critical amino acid residues involved in enzyme catalysis and/ or protein subunit oligomerization. In this case, the putative primase, which is a component of the phage Sﬁ21-type genome replication module encoded by the S. thermophilus phage 3, was used as a model system. Directed by this approach, invariant and highly conserved amino acids within a phage primase consensus ATPase/helicase domain (pfam01057) were targeted by site-speciﬁc mutations. When expressed in trans from pTRK687, a high-copy-number vector, the K238(A/T) and RR340-1AA mutant proteins appeared to completely inhibit phage genome replication and reduced the EOP of phage 3 by greater than nine log cycles. Given the magnitude of the resistance conferred, it was concluded that the putative primase protein is an essential enzyme required for genome replication in S. thermophilus Sﬁ21-type phages. Further, host-encoded factors were unable to complement the deﬁciency caused by transdominant primase expression, indicating that the phage-encoded primase must have unique activities and/or associations that are essential for phage genome replication. The dominant negative phenotype suggests that the plasmidencoded mutant primase proteins must be structurally intact and are able to form stable interactions with the native, phage-encoded primase proteins, thus inhibiting their activity (Fig. 3, see color insert). Alternatively, the mutant subunits might form other nonproductive associations, such as substrate binding in the absence of catalysis and/or titrating away other phage- or host-encoded genome replication factors. Amber mutations (N151am) introduced upstream of the transdominant RR340-341AA and K238(A/T) mutations restored phage genome

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

367

FIG. 3. Proposed mechanism of subunit poisoning in phage defense. Mutant subunits (pink) of a multimeric protein (e.g., primase) associate with the wild-type subunits (blue) to form functionally inactive heteromeric proteins with either reduced or null activity.

replication and parental-type EOP and completely suppressed phage resistance. This indicated that translation of the transdominant mutant primase proteins was required to confer phage resistance. In a separate example, when an E437A mutation was introduced downstream of the transdominant K238T mutation, the E437A mutation completely suppressed phage resistance as well. These results indicated that the E437A mutation precluded the association of the mutant primase protein from the native, phage-encoded primase. Hence, E437 was postulated to be a component of a carboxy-terminal protein oligomerization domain. The oligomerization domain of the gp4 primase from coliphage T7 is also located near the carboxy-terminal region of the protein (Notarnicola et al., 1995).

368

STURINO AND KLAENHAMMER

Signiﬁcantly, no bacteriophages resistant to transdominant primase proteins have been isolated to date. This may be due to the fact that protein structure is tightly associated with protein function and strong conservation is found at the tertiary and quaternary structural levels of related proteins. E. PHAGE-TRIGGERED SUICIDE SYSTEMS Phage-triggered suicide systems put the expression of a toxic gene product under the control of a phage-inducible promoter (Djordjevic and Klaenhammer, 1997; Djordjevic et al., 1997). For example, Djordjevic et al. (1997) placed the expression of a restriction endonuclease gene (llaIR) (O’Sullivan et al., 1995) under the control of 31p, a late-expressed inducible promoter isolated from the lytic lactococcal phage 31. The expression of the suicide cassette, composed of the restriction endonuclease without its cognate methyltransferase gene (llaIM), from a high-copy-number vector reduced the EOP and ECOI of phage 31 to approximately 104 and 0.15, respectively (Djordjevic et al., 1997). Once expressed, the LlaIR restriction endonuclease killed the host and aborted phage infection by restricting both the unmethylated host and the phage genome. In the absence of phage infection, even a very low level of LlaIR expression can be toxic to the cell. As a result, these systems must be stringently regulated. Mutant phages that exhibited reduced sensitivity to the original suicide system (EOP of approximately 0.4) were isolated and found to harbor mutations that mapped to the regulator, Tac31. These phages exhibited lower levels of transcription from the 31p promoter, thus decreasing effectiveness of the phage-trigged suicide cassette expressed in trans (Djordjevic and Klaenhammer, 1997). VII. Concluding Remarks Point mutation and recombination are the two great engines of phage evolution. Their short generation time and large burst sizes can act to accelerate the rate at which mutant phages may overcome a given defense. The plasticity of phage genomes is critical to their rapid evolution. Unfortunately, phages have been able to evolve resistance to a great many of the defenses that have been implemented in the industrial setting over the years (Klaenhammer and Fitzgerald, 1994). In fact, nearly 70 years after the discovery of bacteriophages that infect L. lactis, phages continue to be a leading cause of failure in industrial fermentations. Further, as the demand for fermented food products

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

369

made with strains of S. thermophilus has increased, so has the incidence and severity of phage attacks against these thermophilic starter strains. This trend has been seen previously with other LAB, including L. lactis. With the expansion of fermentation and bioprocessing systems reliant on LAB, disruption by bacteriophages remains a serious concern. For example, strains of LAB are being further exploited for the manufacture of industrial chemicals (e.g., lactate) and employed as vehicles for the delivery of biologics (e.g., vaccines and enzymes). Together, these persistent pressures necessitate the continued development of starter cultures with enhanced phage resistance properties. Over the years, the use of phage defense systems, both endogenous and engineered, have proven to be invaluable for the protection of LAB that are expected to perform consistently and over extended time frames within industrial applications. The number of completely sequenced phage and bacterial genomes continues to grow rapidly, and this has aided in the development of these systems. These emerging tools are invaluable for the understanding of the fundamental relationships that exist between the dairy LAB and their phages. When leveraged against a greater understanding of gene regulation, these sequences reveal the genetic content and comparative organization of phage genomes and facilitate the development of more efﬁcient phage defenses. ACKNOWLEDGMENTS Research on lactic acid bacteria and their bacteriophages at North Carolina State University is conducted with the support of the North Carolina Agricultural Research Service, the USDA National Research Initiative Competitive Grants Program, and Danisco, Inc., of Madison, Wisconsin. Joseph Sturino was supported, in part, by an NIH-Biotechnology Training Fellowship.

References Allison, G. E., and Klaenhammer, T. R. (1998). Phage resistance mechanisms in lactic acid bacteria. Int. Dairy J. 8, 207–226. Beck, K., and Brodsky, B. (1998). Supercoiled protein motifs: The collagen triple-helix and the -helical coiled coil. J. Struct. Biol. 122, 17–29. Bester, B. H., and Lombard, S. H. (1975). Protection of starter cultures against bacteriophages by propagation in a phage-resistant medium. South African J. Dairy Technol. 7, 235–450. Benbadis, L., Garel, J. R., and Hartley, D. L. (1991). Puriﬁcation, properties, and sequence speciﬁcity of SslI, a new type II restriction endonuclease from Streptococcus salivarius subsp. thermophilus. Appl. Environ. Microbiol. 57, 3677–3678. Bickle, T. A., and Kruger, D. H. (1993). Biology of DNA restriction. Microbiol. Rev. 57, 434–450.

370

STURINO AND KLAENHAMMER

Binetti, A. G., and Reinheimer, J. A. (2000). Thermal and chemical inactivation of indigenous Streptococcus thermophilus bacteriophages isolated from Argentinean dairy plants. J. Food Prot. 63, 509–515. Bouchard, J. D., and Moineau, S. (2000). Homologous recombination between a lactococcal bacteriophage and the chromosome of its host strain. Virology 270, 65–75. Broadbent, J. R., McMahon, D. J., Welker, D. L., Oberg, C. J., and Moineau, S. (2003). Biochemistry, genetics, and applications of exopolysaccharide production in Streptococcus thermophilus: A review. J. Dairy Sci. 86, 407–423. Burrus, V., Bontemps, C., Decaris, B., and Gue´don, G. (2001). Characterization of a novel type II restriction-modiﬁcation system, Sth368I, encoded by the integrative element ICESt1 of Streptococcus thermophilus CNRZ368. Appl. Environ. Microbiol. 67, 1522–1528. Burrus, V., Pavlovic, G., Decaris, B., and Guedon, G. (2002). The ICESt1 element of Streptococcus thermophilus belongs to a large family of integrative and conjugative elements that exchange modules and change their speciﬁcity of integration. Plasmid 48, 77–97. Bru¨ssow, H., Fremont, M., Bruttin, A., Sidoti, J., Constable, A., and Fryder, V. (1994a). Detection and classiﬁcation of Streptococcus thermophilus bacteriophages isolated from industrial milk fermentation. Appl. Environ. Microbiol. 60, 4537–4543. Bru¨ssow, H., Probst, A., Fremont, M., and Sidoti, J. (1994b). Distinct Streptococcus thermophilus bacteriophages share an extremely conserved DNA fragment. Virology 200, 854–857. Bru¨ssow, H., Bruttin, A., Desiere, F., Lucchini, S., and Foley, S. (1998). Molecular ecology and evolution of Streptococcus thermophilus bacteriophages—a review. Virus Genes 16, 95–109. Bru¨ssow, H., and Desiere, F. (2001). Comparative phage genomics and the evolution of Siphoviridae: Insights from dairy phages. Mol. Microbiol. 39, 213–222. Bru¨ssow, H., and Hendrix, R. W. (2002). Phage genomics: Small is beautiful. Cell 108, 13–16. Bruttin, A., and Bru¨ssow, H. (1996). Site-speciﬁc spontaneous deletions in three genome regions of a temperate Streptococcus thermophilus phage. Virology 219, 96–104. Bruttin, A., Desiere, F., d’Amico, N., Guerin, J. P., Sidoti, J., Huni, B., Lucchini, S., and Bru¨ssow, H. (1997a). Molecular ecology of Streptococcus thermophilus bacteriophage infections in a cheese factory. Appl. Environ. Microbiol. 63, 3144–3150. Bruttin, A., Desiere, F., Lucchini, S., Foley, S., and Bru¨ssow, H. (1997b). Characterization of the lysogeny DNA module from the temperate Streptococcus thermophilus bacteriophage Sﬁ21. Virology 233, 136–148. Budde-Niekiel, A., and Teuber, M. (1987). Electron microscopy of the adsorption of bacteriophages to lactic acid streptococci. Milchwissenschaft 42, 551–553. Bull, J. J., Jacobson, A., Badgett, M. R., and Molineux, J. (1998). Viral escape from antisense RNA. Mol. Microbiol. 28, 853–846. Carminati, D., and Giraffa, G. (1992). Evidence and characterization of temperate bacteriophages in Streptococcus salivarius subsp. thermophilus St18. J. Dairy Res. 59, 71–79. Chopin, A., Bolotin, A., Sorokin, A., Ehrlich, S. D., and Chopin, M. (2001). Analysis of six prophages in Lactococcus lactis IL1403: Different genetic structure of temperate and virulent phage populations. Nucleic Acids Res. 29, 644–651. Chung, D. K., Chung, S. K., and Batt, C. A. (1992). Antisense RNA directed against the major capsid protein of Lactococcus lactis subsp. cremoris bacteriophage F4-1 confers partial resistance to the host. Appl. Microbiol. Biotechnol. 37, 79–83.

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

371

Cocconcelli, P. S., Callegari, M. L., Porro, D., and Morelli, L. (1995). Plasmid recombination mediated by IS-like structures in Streptococcus thermophilus T003. In ‘‘Abstracts of Posters, European Commission Biotechnology and FAIR Programmes Conference on Lactic Acid Bacteria: From Fundamental Research to Innovative Applications’’, Abstract F-8. Cork, Ireland, 22–26 October. Coffey, A., and Ross, R. P. (2002). Bacteriophage-resistance systems in dairy starter strains: molecular analysis to application. Antonie van Leeuwenhoek 82, 303–321. Cogan, T. M., Peitersen, N., and Sellars, R. L. (1991). Starter systems. In ‘‘Bulletin of the International Dairy Federation, no. 263/1991. Practical Phage Control’’ pp. 16–23. International Dairy Federation, Brussels. Courtin, P., Monnet, V., and Rul, F. (2002). Cell-wall proteinases PrtS and PrtB have a different role in Streptococcus thermophilus/Lactobacillus bulgaricus mixed cultures in milk. Microbiology 148, 3413–3421. Crittenden, R. G., Martinez, N. R., and Playne, M. J. (2003). Synthesis and utilization of folate by yogurt starter cultures and probiotic bacteria. Int. J. Food Microbiol. 80, 217–222. Daly, C. (1983). The use of mesophilic cultures in the dairy industry. Antonie Van Leeuwenhoek 49, 297–312. D’Herelle, F. (1917). Sur un microbe invisible antagoniste des bacilles dysenteriques. In ‘‘Milestones in Microbiology: 1556 to 1940’’ (T. D. Brock, ed.), p. 157. American Society for Microbiology Press, Washington, DC. Comp. Rend. Acad. Sci. 165:373–375. Desiere, F., Lucchini, S., and Bru¨ssow, H. (1998). Evolution of Streptococcus thermophilus bacteriophage genomes by modular exchanges followed by point mutations and small deletions and insertions. Virology 241, 345–356. Desiere, F., Lucchini, S., Canchaya, C., Ventura, M., and Bru¨ssow, H. (2002). Comparative genomics of phages and prophages in lactic acid bacteria. Antonie Van Leeuwenhoek 82, 73–91. Desmazeaud, M. (1983). L’e´tat des connaissances en matie`re de nutrition des bacte´ries lactiques. Lait 63, 267–316. Djordjevic, G. M., O’Sullivan, D. J., Walker, S. A., Conkling, M. A., and Klaenhammer, T. R. (1997). A triggered-suicide system designed as a defense against bacteriophages. J. Bacteriol. 179, 6741–6748. Djordjevic, G. M., and Klaenhammer, T. R. (1997). Bacteriophage-triggered defense systems: Phage adaptation and design improvements. Appl. Environ. Microbiol. 63, 4370–4376. Duplessis, M., and Moineau, S. (2001). Identiﬁcation of a genetic determinant responsible for host speciﬁcity in Streptococcus thermophilus bacteriophages. Mol. Microbiol. 41, 325–336. Durmaz, E., and Klaenhammer, T. R. (1995). A starter culture rotation strategy incorporating paired restriction/modiﬁcation and abortive infection bacteriphage defenses in a single Lactococcus lactis strain. Appl. Environ. Microbiol. 61, 1266–1273. Durmaz, E., and Klaenhammer, T. R. (2000). Genetic analysis of chromosomal regions of Lactococcus lactis acquired by recombinant lytic phages. Appl. Environ. Microbiol. 66, 895–903. Durmaz, E., Madsen, S. M., Israelsen, H., and Klaenhammer, T. R. (2002). Lactococcus lactis lytic bacteriophages of the P335 group are inhibited by overexpression of a truncated CI repressor. J. Bacteriol. 184, 6532–6544. Fayard, B., Haeﬂiger, M., and Accolas, J. P. (1993). Interactions of temperate bacteriophages of Streptococcus salivarius subsp. thermophilus with lysogenic indicators affect phage DNA restriction patterns and host ranges. J. Dairy Res. 60, 385–399.

372

STURINO AND KLAENHAMMER

Foley, S., Lucchini, S., Zwahlen, M. C., and Bru¨ssow, H. (1998). A short noncoding viral DNA element showing characteristics of a replication origin confers bacteriophage resistance to Streptococcus thermophilus. Virology 250, 377–387. Forde, A., and Fitzgerald, G. F. (1999). Bacteriophage defense systems in lactic acid bacteria. Antonie van Leeuwenhoek 76, 89–113. Forde, A., Daly, C., and Fitzgerald, G. F. (1999). Identiﬁcation of four phage resistance plasmids from Lactococcus lactis subsp. cremoris HO2. Appl. Environ. Microbiol. 65, 1540–1547. Fuller-Pace, F. V., and Murray, N. E. (1986). Two DNA recognition domains of the speciﬁcity polypeptides of a family of type I restriction enzymes. Proc. Natl. Acad. Sci. USA 83, 9368–9372. Garbutt, K. C., Kraus, J., and Geller, B. L. (1997). Bacteriophage resistance in Lactococcus lactis engineered by replacement of a gene for a bacteriophage receptor. J. Dairy Sci. 80, 1512–1519. Garvey, P., van Sinderen, D., Twomey, D. P., Hill, C., and Fitzgerald, G. F. (1995). Molecular genetics of bacteriophage and natural phage defense systems in the genus Lactococcus. Int. Dairy J. 5, 905–947. Gasson, M. J. (1996). Lytic systems in lactic acid bacteria and their bacteriophages. Antonie Van Leeuwenhoek 70, 147–159. Geis, A., Demerdash, H. A., and Heller, K. J. (2003). Sequence analysis and characterization of plasmids from Streptococcus thermophilus. Plasmid 50, 53–69. Geller, B. L., Ivey, R. G., Trempy, J. E., and Hettinger-Smith, B. (1993). Cloning of a chromosomal gene required for phage infection in Lactococcus lactis subsp. lactis C2. J. Bacteriol. 175, 5510–5519. Grobben, G. J., Chin-Joe, I., Kitzen, V. A., Boels, I. C., Boer, F., Sikkema, J., Smith, M. R., and Bont, J. A. M. (1998). Enhancement of exopolysaccharide production by Lactobacillus delbrueckii subsp. bulgaricus NCFB 2772 with a simpliﬁed deﬁned medium. Appl. Environ. Microbiol. 64, 1333–1337. Guedon, G., Bourgoin, F., Pebay, M., Roussel, Y., Colmin, C., Simonet, J. M., and Decaris, B. (1995). Characterization and distribution of two insertion sequences, IS1191 and iso-IS981, in Streptococcus thermophilus: Does intergeneric transfer of insertion sequences occur in lactic acid bacteria co-cultures? Mol. Microbiol. 16, 69–78. Guimont, C., Henry, P., and Linden, G. (1993). Restriction/modiﬁcation in Streptococcus thermophilus: Isolation and characterization of a type II restriction endonuclease Sth455I. Appl. Microbiol. Biotechnol. 39, 216–220. Gulstrom, T. J., Pearce, L. E., Sandine, W. E., and Elliker, P. R. (1979). Evaluation of commercial phage inhibitory media. J. Dairy Sci. 62, 208–221. Hardie, J. M. (1994). Streptococcus. In ‘‘Bergy’s Manual of Determinative Bacteriology’’ (J. G. Holt, ed.), 9th ed. Williams and Wilkins, Baltimore, MD. Heap, H. A., and Lawrence, R. C. (1976). The selection of starter strains for cheese making. N. Z. J. Dairy Sci. Technol. 11, 16–53. Herman, R. E., and McKay, L. L. (1985). Isolation and partial characterization of plasmid DNA from Streptococcus thermophilus. Appl. Environ. Microbiol. 50, 1103–1106. Herskowitz, I. (1987). Functional inactivation of genes by dominant negative mutation. Nature 329, 219–222. Hill, C., Miller, L. A., and Klaenhammer, T. R. (1990). Cloning, expression, and sequence determination of a bacteriophage fragment encoding bacteriophage resistance in Lactococcus lactis. J. Bacteriol. 172, 6419–6426. Hill, C., Massey, I. J., and Klaenhammer, T. R. (1991a). A rapid method to characterize lactococcal bacteriophage genomes. Appl. Environ. Microbiol. 57, 283–288.

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

373

Hill, C., Miller, L. A., and Klaenhammer, T. R. (1991b). In vivo genetic exchange of a functional domain from a type IIA methylase between lactococcal plasmid pTR2030 and a virulent bacteriophage. J. Bacteriol. 173, 4363–4370. Hjalt, T. A., and Wagner, E. G. (1995). Bulged-out nucleotides in an antisense RNA are required for rapid target RNA binding in vitro and inhibition in vivo. Nucleic Acids Res. 23, 580–587. Huggins, A. R. (1984). Progress in dairy starter culture technology. Food Technol. 38, 41. Hull, R. R. (1985). Recent developments in the genetics of lactic acid bacteria. CSIRO Food Res. Q. 45, 40–46. Husson-Kao, C., Mengaud, J., Gripon, J. C., Benbadis, L., and Chapot-Chartier, M. P. (2000a). Characterization of Streptococcus thermophilus strains that undergo lysis under unfavorable environmental conditions. Int. J. Food Microbiol. 55, 209–213. Husson-Kao, C., Mengaud, J., Cesselin, B., van Sinderen, D., Benbadis, L., and ChapotChartier, M. P. (2000b). The Streptococcus thermophilus autolytic phenotype results from a leaky prophage. Appl. Environ. Microbiol. 66, 558–565. Inouye, M. (1988). Antisense RNA: Its functions and applications in gene regulation—a review. Gene 72, 25–34. Ivanowsky, D. I. (1892). On two diseases of tobacco. Sel’ Khoz. Lesov. 169:108–121; an English translation appears as an appendix to S. S. Hughes, 1972. In ‘‘The origins and development of the concept of the virus in the late nineteenth century’’ Ph.D. thesis, London University. Jarvis, A. W., Fitzgerald, G. F., Mata, M., Mercenier, A., Neve, H., Powell, I. B., Ronda, C., Saxelin, M., and Teuber, M. (1991). Species and type phages of lactococcal bacteriophages. Intervirology 32, 2–9. Jensen, R. G. (ed.) (1995). ‘‘Handbook of Milk Composition.’’ Academic Press, New York. Kim, J. H., and Batt, C. A. (1991). Antisense RNA mediated bacteriophage resistance in Lactococcus lactis. Appl. Environ. Microbiol. 57, 1039–1045. Kim, S. G., Bor, Y. C., and Batt, C. A. (1992). Bacteriophage resistance in Lactococcus lactis spp. lactis using antisense ribonucleic acid. J. Dairy Sci. 75, 1761–1767. Klaenhammer, T. R. (1984). Interactions of bacteriophages with lactic streptococci. In ‘‘Advances in Applied Microbiology’’ (A. I. Laskin, ed.), Vol. 30, pp. 1–29. Academic Press, New York. Klaenhammer, T. R., and Sanozky, R. (1985). Conjugal transfer from Streptococcus lactis ME2 of plasmids encoding phage resistance, nisin resistance and lactosefermenting ability: Evidence for a high-frequency conjugative plasmid responsible for abortive infection of virulent bacteriophage. J. Gen. Microbiol. 131, 1531–1541. Klaenhammer, T. R. (1987). Plasmid-directed mechanisms for bacteriophage defense in lactic streptococci. FEMS Microbiol. Rev. 46, 313–325. Klaenhammer, T. R. (1989). Genetic characterization of multiple mechanisms of phage defense from a prototype phage-insensitive strain Lactococcus lactis ME2. J. Dairy Sci. 72, 3429–3443. Klaenhammer, T. R., Romero, D., Sing, W., and Hill, C. (1991). Molecular analysis of pTR2030 gene systems that confer phage resistance to lactococci. In ‘‘Genetics and Molecular Biology of Streptococci, Lactococci, and Enterococci’’ (G. Dunny, P. Cleary, and L. McKay, eds.), pp. 124–130. American Society for Microbiology Press, Washington, DC. Klaenhammer, T. R., and Fitzgerald, G. F. (1994). Bacteriophage and bacteriophage resistance. In ‘‘Genetics and Biotechnology of Lactic Acid Bacteria’’ (M. J. Gasson and W. M. de Vos, eds.), pp. 106–168. Chapman and Hall, London.

374

STURINO AND KLAENHAMMER

Kolars, J. C., Levitt, M. D., Aouji, M., and Savaiano, D. A. (1984). Yogurt—an autodigesting source of lactose. N. Engl. J. Med. 310, 1–3. Kolb, F. A., Westhof, E., Ehresmann, C., Ehresmann, B., Gerhart, E., Wagner, H., and Romby, P. (2001). Bulged residues promote the progression of a loop–loop interaction to a stable and inhibitory antisense–target RNA complex. Nucleic Acids Res. 29, 3145–3153. Krishnaswamy, K., and Madhavan, N. K., (2001). Importance of folate in human nutrition. Br. J. Nutr. (Suppl. 2), S115–S124. Kruger, D. H., and Bickle, T. A. (1983). Bacteriophage survival: Multiple mechanisms for avoiding the deoxyribonucleic acid restriction systems of their hosts. Microbiol. Rev. 47, 345–360. Kunji, E. R., Mierau, I., Hagting, A., Poolman, B., and Konings, W. N. (1996). The proteolytic systems of lactic acid bacteria. Antonie Van Leeuwenhoek 70, 187–221. Larbi, D., Decaris, B., and Simonet, J. M. (1992). Different bacteriophage resistance mechanisms in Streptococcus salivarius subsp. thermophilus. J. Dairy Res. 59, 349–357. Lawrence, J. G., Hendrix, R., and Casjens, S. (2001). Where are the pseudogenes in bacterial genomes? Trends Microbiol. 9, 535–540. Le Marrec, C., van Sinderen, D., Walsh, L., Stanley, E., Vlegels, E., Moineau, S., Heinze, P., Fitzgerald, G., and Fayard, B. (1997). Two groups of bacteriophages infecting Streptococcus thermophilus can be distinguished on the basis of mode of packaging and genetic determinants for major structural proteins. Appl. Environ. Microbiol. 63, 3246–3253. Letort, C., and Juillard, V. (2001). Development of a minimal chemically deﬁned medium for the exponential growth of Streptococcus thermophilus. J. Appl. Microbiol. 91, 1023–1029. Loefﬂer, F., and Frosch, P. (1898). Berichte der Kommission zur Erforschung der Maulund Klauenseuche bei dem Institut fur Infektionskrankheiten. 23,371–391. In ‘‘Milestones in Microbiology: 1556 to 1940’’ (T. D. Brock, ed.), p. 149. American Society for Microbiology Press, Washington, DC. Lowrie, R. J., and Pearce, L. E. (1971). The plating efﬁciency of bacteriophages of lactic streptococci. N. Z. J. Dairy Sci. Technol. 6, 166–171. Lubbers, M. W., Waterﬁeld, N. R., Beresford, T. P., Le Page, R. W., and Jarvis, A. W. (1995). Sequencing and analysis of the prolate-headed lactococcal bacteriophage c2 genome and identiﬁcation of the structural genes. Appl. Environ. Microbiol. 61, 4348–4356. Lucchini, S., Desiere, F., and Bru¨ssow, H. (1998). The structural gene module in Streptococcus thermophilus bacteriophage Sﬁ11 shows a hierarchy of relatedness to Siphoviridae from a wide range of bacterial hosts. Virology 246, 63–73. Lucchini, S., Desiere, F., and Bru¨ssow, H. (1999a). The genetic relationship between virulent and temperate Streptococcus thermophilus bacteriophages: Whole genome comparison of cos-site phages Sﬁ19 and Sﬁ21. Virology 260, 232–243. Lucchini, S., Desiere, F., and Bru¨ssow, H. (1999b). Comparative genomics of Streptococcus thermophilus phage species supports a modular evolution theory. J. Virology 73, 8647–8656. Lucchini, S., Sidoti, J., and Bru¨ssow, H. (2000). Broad-range bacteriophage resistance in Streptococcus thermophilus by insertional mutagenesis. Virology 275, 267–277. Maguin, E., Prevost, H., Ehrlich, S. D., and Gruss, A. (1996). Efﬁcient insertional mutagenesis in lactococci and other gram-positive bacteria. J. Bacteriol. 178, 931–935.

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

375

Maniloff, J., Cadden, S. P., and Putzrath, R. M. (1981). Maturation of an enveloped budding phage: Mycoplasmavirus L2. Prog. Clin. Biol. Res. 64, 503–513. Martini, M. C., Lerebours, E. C., Lin, W. J., Harlander, S. K., Barrada, N. M., Antoine, J. M., and Savaiano, D. A. (1991). Strains and species of lactic acid bacteria in fermented milks (yogurts): Effect on in vivo lactose digestion. Am. J. Clin. Nutr. 54, 1041–1046. Matalon, M. E., and Sandine, W. E. (1986). Lactobacillus bulgaricus, Streptococcus thermophilus and yogurt: A review. Cult. Dairy Prod. J. 21, 6–23. McGrath, S., Fitzgerald, G. F., and van Sinderen, D. (2001). Improvement and optimization of two engineered phage resistance mechanisms in Lactococcus lactis. Appl. Environ. Microbiol. 67, 608–616. McGrath, S., Fitzgerald, G. F., and van Sinderen, D. (2002). Identiﬁcation and characterization of phage-resistance genes in temperate lactococcal bacteriophages. Mol. Microbiol. 43, 509–520. Moineau, S., Pandian, S., and Klaenhammer, T. R. (1993). Restriction/modiﬁcation systems and restriction endonucleases are more effective on lactococcal bacteriophages that have emerged recently in the dairy industry. Appl. Environ. Microbiol. 59, 197–202. Moineau, S., Walker, S. A., Holler, B. J., Vedamuthu, E. R., and Vandenbergh, P. A. (1995). Expression of a Lactococcus lactis phage resistance mechanism by Streptococcus thermophilus. Appl. Environ. Microbiol. 61, 2461–2466. Moineau, S., Borkaev, M., Holler, B. J., Walker, S. A., Kondo, J. K., Vedamuthu, E. R., and Vandenbergh, P. A. (1996). Isolation and characterization of lactococcal bacteriophages from cultured buttermilk plants in the United States. J. Dairy Sci. 79, 2104–2111. Monteville, M. R., Ardestani, B., and Geller, B. L. (1994). Lactococcal bacteriophages require a host cell wall carbohydrate and a plasma membrane protein for adsorption and ejection of DNA. Appl. Environ. Microbiol. 60, 3204–3211. Nakayama, S., Moncrief, N. D., and Kretsinger, R. H. (1992). Evolution of EF-hand calcium-modulated proteins. II. Domains of several subfamilies have diverse evolutionary histories. J. Mol. Evol. 34, 416–448. Notarnicola, S. M., Park, K., Grifﬁth, J. D., and Richardson, C. C. (1995). A domain of the gene 4 helicase/primase of bacteriophage T7 required for the formation of an active hexamer. J. Biol. Chem. 270, 20215–20224. Olson, N. F. (1969). Ripened semisoft cheeses. ‘‘Pﬁzer Cheese Monographs,’’ Vol. 4, Pﬁzer, Inc., New York. O’Sullivan, D. J., and Klaenhammer, T. R. (1998). Control of expression of LlaI restriction in Lactococcus lactis. Mol. Microbiol. 27, 1009–1020. O’Sullivan, D. J., Hill, C., and Klaenhammer, T. R. (1993). Effect of increasing the copy number of bacteriophage origins of replication in trans, on incoming-phage proliferation. Appl. Environ. Microbiol. 59, 2449–2456. O’Sullivan, D. J., Zagula, K., and Klaenhammer, T. R. (1995). In vivo restriction by LlaI is encoded by three genes, arranged in an operon with llaIM, on the conjugative Lactococcus plasmid pTR2030. J. Bacteriol. 177, 134–143. O’Sullivan, D., Coffey, A., Fitzgerald, G. F., Hill, C., and Ross, R. P. (1998). Design of a phage-insensitive lactococcal dairy starter via sequential transfer of naturally occurring conjugative plasmids. Appl. Environ. Microbiol. 64, 4618–4622. O’Sullivan, T., van Sinderen, D., and Fitzgerald, G. (1999). Structural and functional analysis of pCI65st, a 6.5 kb plasmid from Streptococcus thermophilus NDI-6. Microbiology 145, 127–134.

376

STURINO AND KLAENHAMMER

Pebay, M., Colmin, C., Guedon, G., DeGasperi, C., Decaris, B., and Simonet, J. M. (1992). Detection of intraspeciﬁc DNA polymorphism in Streptococcus salivarius subsp. thermophilus by a homologous rDNA probe. Res. Microbiol. 143, 37–46. Poch, M. T., Somkuti, G. A., and Solaiman, D. K. Y. (1997). Sth132I, a novel class-IIS restriction endonuclease of Streptococcus thermophilus ST132. Gene 195, 201–206. Polzin, K. M., Collins, L. J., Lubbers, M. W., and Jarvis, A. W. (1996). Effect of various antisense mRNAs on bacteriophage c2 replication. In Abstracts of Posters at the Fifth Symposium on Lactic Acid Bacteria: Genetics, Metabolism, and Applications, Abstract F2. Veldhoven, The Netherlands, 8–12 October, 1996. Powell, I. A., and Davidson, B. E. (1986). Resistance to in vitro restriction of DNA from lactic streptococcal bacteriophage c6A. Appl. Environ. Microbiol. 51, 1358–1360. Quiberoni, A., Stiefel, J. I., and Reinheimer, J. A. (2000). Characterization of phage receptors in Streptococcus thermophilus using puriﬁed cell walls obtained by a simple protocol. J. Appl. Microbiol. 86, 1059–1065. Rajagopal, S. N., and Sandine, W. E. (1990). Associative growth and proteolysis of Streptococcus thermophilus and Lactobacillus bulgaricus in skim milk. J. Dairy Sci. 73, 894–899. Reid, G., Sanders, M. E., Gaskins, H. R., Gibson, G. R., Mercenier, A., Rastall, R., Roberfroid, M., Rowland, I., Cherbut, C., and Klaenhammer, T. R. (2003). New scientiﬁc paradigms for probiotics and prebiotics. J. Clin. Gastroenterol. 37, 105–108. Reinbold, G. W. (1963). Italian cheese varieties. ‘‘Pﬁzer Cheese Monographs,’’ Vol. 1, Pﬁzer, Inc., New York. Roberts, R. J., Belfort, M., Bestor, T., Bhagwat, A. S., Bickle, T. A., Bitinaite, J., Blumenthal, R. M., Degtyarev, S. K., Dryden, D. T., Dybvig, K., Firman, K., Gromova, E. S., Gumport, R. I., Halford, S. E., Hattman, S., Heitman, J., Hornby, D. P., Janulaitis, A., Jeltsch, A., Josephsen, J., Kiss, A., Klaenhammer, T. R., Kobayashi, I., Kong, H., Kruger, D. H., Lacks, S., Marinus, M. G., Miyahara, M., Morgan, R. D., Murray, N. E., Nagaraja, V., Piekarowicz, A., Pingoud, A., Raleigh, E., Rao, D. N., Reich, N., Repin, V. E., Selker, E. U., Shaw, P. C., Stein, D. C., Stoddard, B. L., Szybalski, W., Trautner, T. A., Van Etten, J. L., Vitor, J. M., Wilson, G. G., and Xu, S. Y. (2003a). A nomenclature for restriction enzymes, DNA methyltransferases, homing endonucleases and their genes. Nucleic Acids Res. 31, 1805–1812. Roberts, R. J., Vincze, T., Posfai, J., and Macelis, D. (2003b). REBASE: Restriction enzymes and methyltransferases. Nucleic Acids Res. 31, 418–420. Ross, R. P., Morgan, S., and Hill, C. (2002). Preservation and fermentation: Past, present and future. Int. J. Food Microbiol. 79, 3–16. Roussel, Y., Bourgoin, F., Guedon, G., Pebay, M., and Decaris, B. (1997). Analysis of the genetic polymorphism between three Streptococcus thermophilus strains by comparing their physical and genetic organization. Microbiology 143, 1335–1343. Sable, S., and Lortal, S. (1995). The lysins of bacteriophages infecting lactic acid bacteria. Appl. Microbiol. Biotechnol. 43, 1–6. Sanders, M. E., Leonhard, P. J., Sing, W. D., and Klaenhammer, T. R. (1986). Conjugal strategy for construction of fast-acid producing, bacteriophage resistant lactic streptococci for use in dairy fermentations. Appl. Environ. Microbiol. 52, 1001–1007. Scannerini, S. (2003). The GMO war. Riv. Biol. 96, 181–187. Schouler, C., Gautier, M., Ehrlich, D., and Chopin, M. C. (1998). Combinational variation of restriction modiﬁcation in Lactococcus lactis. J. Mol. Microbiol. 28, 169–178. Seegers, J. F., van Sinderen, D., and Fitzgerald, G. F. (2000). Molecular characterization of the lactococcal plasmid pCIS3: Natural stacking of speciﬁcity subunits of a type I

BACTERIOPHAGE DEFENSE SYSTEMS AND STRATEGIES

377

restriction/modiﬁcation system in a single lactococcal strain. Microbiology 146, 435–443. Sing, W. D., and Klaenhammer, T. R. (1993). A strategy for rotation of different bacteriophage defenses in a lactococcal single-strain starter culture system. Appl. Environ. Microbiol. 59, 365–372. Solow, B. T., and Somkuti, G. A. (2001). Molecular properties of Streptococcus thermophilus plasmid pER35 encoding a restriction modiﬁcation system. Curr. Microbiol. 42, 122–128. Solaiman, D. K. Y., and Somkuti, G. A. (1990). Isolation and characterization of a type II restriction endonuclease from Streptococcus thermophilus. FEMS Microbiol. Lett. 67, 261–266. Solaiman, D. K. Y., and Somkuti, G. A. (1991). A type II restriction endonuclease of Streptococcus thermophilus ST117. FEMS Microbiol. Lett. 80, 75–80. Stanley, E., Fitzgerald, G. F., Le Marrec, M. C., Fayard, B., and van Sinderen, D. (1997). Sequence analysis and characterization of O1205, a temperate bacteriophage infecting Streptococcus thermophilus CNRZ1205. Microbiology 143, 3417–3429. Stanley, E., Walsh, L., van der Zwet, A., Fizgerald, G. F., and van Sinderen, D. (2000). Identiﬁcation of four loci isolated from two Streptococcus thermophilus phage genomes responsible for mediating bacteriophage resistance. FEMS Microbiol. Lett. 182, 271–277. Sturino, J. M., and Klaenhammer, T. R. (2002). Expression of antisense RNA targeted against Streptococcus thermophilus bacteriophages. Appl. Environ. Microbiol. 68, 588–596. Sturino, J. M., and Klaenhammer, T. R. (2004a). Antisense RNA targeting primase interferes with bacteriophage replication in Streptococcus thermophilus. Appl. Environ. Microbiol. 70, 1735–1743. Sturino, J. M., and Klaenhammer, T. R. (2004b). Inhibition of bacteriophage genome replication through the expression of transdominant mutant primase proteins in Streptococcus thermophilus. In review. Tangney, M., and Fitzgerald, G. F. (2002). AbiA, a lactococcal abortive infection mechanism functioning in Streptococcus thermophilus. Appl. Environ. Microbiol. 68, 6388–6391. Te´tart, F., Repolla, R., Monod, C., and Krisch, H. M. (1996). Bacteriophage T4 host range is expanded by duplication of a small domain of the tail ﬁber adhesin. J. Mol. Biol. 258, 726–731. Te´tart, F., Desplats, C., and Krisch, H. M. (1998). Genome plasticity in the distal tail ﬁber locus of the T-even bacteriophage recombination between conserved motifs swaps adhesin speciﬁcity. J. Mol. Biol. 282, 543–556. Thunell, R. K., and Sandine, W. E. (1985). Types of starter cultures. In ‘‘Bacterial Starter Cultures for Foods’’ (S. E. Gilland, ed.), pp. 127–144. CRC Press, Boca Raton, FL. Tremblay, D. M., and Moineau, S. (1999). Complete genomic sequence of the lytic bacteriophage DT1 of Streptococcus thermophilus. Virology 255, 63–76. Turgeon, N., and Moineau, S. (2001). Isolation and characterization of a Streptococcus thermophilus plasmid closely related to the pMV158 family. Plasmid 45, 171–183. Twort, F. (1915). An investigation on the nature of ultra-microscopic viruses. Lancet 2, 1241–1243. Valyasevi, R., Sandine, W. E., and Geller, B. L. (1991). A membrane protein is required for bacteriophage c2 infection of Lactococcus lactis subsp. lactis C2. J. Bacteriol. 173, 6095–6100.

378

STURINO AND KLAENHAMMER

Ventura, M., Bruttin, A., Canchaya, C., and Bru¨ssow, H. (2002a). Transcription analysis of Streptococcus thermophilus phages in the lysogenic state. Virology 302, 21–32. Ventura, M., Foley, S., Bruttin, A., Chennouﬁ, S., Canchaya, C., and Bru¨ssow, H. (2002b). Transcription mapping as a tool in phage genomics: The case of the temperate Streptococcus thermophilus phage Sﬁ21. Virology 296, 62–76. Venturi, A., Gionchetti, P., Rizzello, F., Johansson, R., Zucconi, E., Brigidi, P., Matteuzzi, D., and Campieri, M. (1999). Impact on the composition of the fecal ﬂora by a new probiotic preparation: preliminary data on maintenance treatment of patients with ulcerative colitis. Aliment. Pharmacol. Ther. 13, 1103–1108. Wang, I. N., Smith, D. L., and Young, R. (2000). Holins: The protein clocks of bacteriophage infections. Annu. Rev. Microbiol. 54, 799–825. Walker, S. A., and Klaenhammer, T. R. (2000). An explosive antisense RNA strategy for inhibition of a lactococcal bacteriophage. Appl. Environ. Microbiol. 66, 310–319. Werts, C., Michel, V., Hofnung, M., and Charbit, C. (1994). Adsorption of bacteriophage lambda on the LamB protein of Escherichia coli K-12: Point mutations in gene J of lambda responsible for extended host range. J. Bacteriol. 176, 941–947. Whitehead, H. R., and Cox, G. A. (1935). The occurrence of bacteriophage in cultures of lactic streptococci, a preliminary note. N. Z. J. Sci. Technol. 16, 319. Whitehead, W. E., Ayres, J. W., and Sandine, W. E. (1993). A review of starter media for cheese making. J. Dairy Sci. 76, 2344–2353. Wilson, G. G., and Murray, N. E. (1991). Restriction and modiﬁcation systems. Annu. Rev. Genet. 25, 585–627. Wilson, R. D., Davies, G., Desilets, V., Reid, G. J., Summers, A., Wyatt, P., and Young, D. (2003). The use of folic acid for the prevention of neural tube defects and other congenital anomalies. J. Obstet. Gynaecol. Can. 25, 959–973. Yu, C. E., and Ferretti, J. J. (1991). Molecular characterization of new group A streptococcal bacteriophages containing the gene for streptococcal erythrogenic toxin A (speA). Mol. Gen. Genet. 231, 161–168. Zourari, A., Accolas, J. P., and Desmazeaud, M. J. (1992). Metabolism and biochemical characteristics of yogurt bacteria: A review. Lait 72, 1–34.

Current Issues in Genetic Toxicology Testing for Microbiologists KRISTIEN MORTELMANS*

AND

DOPPALAPUDI S. RUPA

Toxicology and Pharmacology Laboratory Biosciences Division, SRI International Menlo Park, California 94025 *Author for correspondence. E-mail: [email protected]

I. Introduction II. Genesis of Genetic Toxicology III. Regulatory Genetic Toxicology Tests A. The Salmonella/Mammalian-Microsome Mutagenicity Assay B. In Vitro Mouse Lymphoma TKþ/ Gene Mutation Assay C. In Vitro Mammalian Chromosomal Aberration Assay D. In Vivo Mammalian Erythrocyte Micronucleus Test IV. Regulatory Genetic Toxicology Guidelines A. International Conference on Harmonization Guidelines B. OECD Guidelines V. Concluding Remarks and Outlook References

379 381 382 382 387 389 390 391 392 394 396 397

I. Introduction Genetic toxicology tests are designed to detect chemical agents capable of damaging DNA and chromosomes of cells. Such damage is generally detected as mutations, which are changes in the DNA base sequence, chromosomal aberrations, and DNA strand breaks, all of which may lead to inheritable alterations of gene function. Chemicals that induce mutations are considered ‘‘toxic’’ to the genome and are thus referred to as ‘‘genotoxic.’’ There are two major concerns when chemical or physical agents induce genotoxicity. First, gene alterations in germ cells may result in birth defects and impose risks to future generations. Second, when the genetic material in somatic cells is altered, it may become the ﬁrst step in the initiation and progression of cancer according to the somatic mutation theory of cancer. It is to be argued that genetic damage in germ cells potentially poses a far greater threat to immediate offspring, in addition to adverse health effects in future generations as compared with cancer. However, the high incidence of cancer, as well as the high cancer-related mortality are of greater immediate concern to researchers and regulatory agencies. 379 ADVANCES IN APPLIED MICROBIOLOGY, VOLUME 56 Copyright 2004, Elsevier Inc. All rights reserved. 0065-2164/04 $35.00

380

MORTELMANS AND RUPA

Almost all known human carcinogens evaluated by the International Agency for Research on Cancer (IARC) are mutagens, except for ﬁber, hormones, and lifestyle (IARC, 1997). There is convincing evidence that mutations are mediating factors in at least two well-characterized mechanisms of human carcinogenesis: activation of oncogenes and inactivation of tumor suppressor genes (Knudson, 1993). This evidence provides support for the somatic mutation theory of carcinogenesis. It has also been shown in cytogenetic studies of human cancers that all cancer cells are heteroploid and that many of the chromosomal changes are speciﬁc for certain types of cancer (Mitelman, 1988). Thus, genetic toxicology tests that are highly sensitive and reliable in detecting mutagens provide pivotal information to regulatory agencies. There are a number of in vitro short-term genetic toxicology tests available to evaluate the mutagenic potential of chemical and physical agents. Typically, mutations are the endpoint in bacterial systems, whereas gene and chromosome damage are determined in cultured mammalian cells. In vivo short-term tests are also available and usually involve the detection of chromosome damage in bone marrow cells of mice or rats. Genetic toxicology tests are used by the industry for the prediction of carcinogenicity of new chemicals, or of chemicals under development, because there is a high association between positive genotoxicity results and rodent carcinogenicity (McCann and Ames, 1976; Zeiger, 1998). It should be noted that genetic toxicology tests should not be considered substitutes for cancer studies in animals. They should be used as an initial screen to prioritize chemicals for additional toxicological testing, to determine the spectrum of genetic effects that can be produced by speciﬁc chemicals, and to aid in the interpretation of the genotoxicity results (Zeiger, 2001). Genetic toxicology testing by industry for regulatory approval or for registration of a product is generally conducted according to guidelines established by national or international authorities. In addition, the testing is done in adherence to Good Laboratory Practices (GLP). Usually a battery of genetic toxicology tests is required. A battery usually includes (1) a test for gene mutation in bacteria, (2) an in vitro cytogenetic test (e.g., chromosomal aberrations) in cultured mammalian cells, and (3) an in vivo test for chromosomal damage in bone marrow cells of rodents. It may also include a test for gene mutation in cultured mammalian cells. This chapter provides a short history of genetic toxicology, a brief description of the most commonly used regulatory tests, and an overview of international regulatory guidelines, most of which are accepted by regulatory agencies in the United States.

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

381

II. Genesis of Genetic Toxicology The work published in 1975 by Dr. Ames and colleagues at the University of California at Berkeley played a crucial role in the establishment of genetic toxicology, which has become an integral part of the toxicological evaluation of new chemicals and drugs (Ames et al., 1975). The publication described the methods for performing the Ames Salmonella/mammalian-microsome mutagenicity assay, which is a reverse mutation test. The assay employs several specially designed histidine auxotrophic Salmonella typhimurium strains that detect mutagens that revert base pair substitution or frameshift mutations present in the tester strains. The reversion restores the functional capability of the bacteria to synthesize histidine, an essential amino acid. The test is simple; about 108 histidine-dependent bacteria are exposed to several concentrations of the test article on minimal glucose agar plates supplemented with a small amount of histidine. The small amount of histidine allows all the plated bacteria to undergo a few cell divisions, which is essential for the ﬁxation of mutations. After 2 days of incubation at 37 C, the plates are observed for the presence of histidine-independent colonies. Only the bacteria that revert back to histidine independence are able to form visible colonies on the agar plates. An exogenous mammalian metabolic activation system, usually a rodent liver homogenate, is included in the test to mimic mammalian metabolism. A chemical is considered a mutagen when the number of revertant colonies on the treated plates is higher than the number of revertant colonies on the negative control plate. Ideally, the increase in the number of revertant colonies should be dose related. Dr. Ames and colleagues published two additional papers in which they reported on the mutagenicity of 300 chemicals (McCann and Ames, 1976; McCann et al., 1975a). These two publications revealed that approximately 90% of the carcinogens tested yielded a positive mutagenic response and convincingly proved that the Salmonella histidine reverse mutation assay (now referred to as the Ames test) is a quick and inexpensive method for screening large numbers of chemicals as a predictor for rodent carcinogens. Commercial companies eagerly embraced the Ames Salmonella assay for screening new chemicals and drugs. In addition, many national and international government agencies, interested in using data from short-term tests for regulatory purposes, welcomed the Ames test. Some of these agencies provided support to validate the Ames assay, sometimes in multiple laboratories. The initial reports on the effectiveness of the Ames test as a predictor of carcinogenesis prompted the development of numerous bacterial

382

MORTELMANS AND RUPA

and mammalian test systems to detect chemical-induced mutations, as well as chromosomal aberrations. Hollstein et al. (1979) and Lohman et al. (1992) reported the existence of 119 and 85 test systems or endpoints, respectively. However, over the years only a few test systems were sufﬁciently validated with the generation of large databases. This warranted the inclusion of these test systems in the battery of tests required by regulatory agencies. The other test systems that ‘‘didn’t make it’’ may very well be as effective, if not more effective, as predictors of rodent carcinogenesis. However, the limited information available on them does not at the present time justify their routine use in the initial screening of chemicals for their mutagenic potential. III. Regulatory Genetic Toxicology Tests The procedures of the most commonly used regulatory genetic toxicology tests are presented below. In addition, a short historical note is included on how the Ames Salmonella/mammalian-microsome test was developed. It is a wonderful illustration of how classical bacterial genetics, before the advent of recombinant DNA technology, was used to genetically alter the Salmonella tester strains to improve the sensitivity of the assay. A. THE SALMONELLA /MAMMALIAN-MICROSOME MUTAGENICITY ASSAY 1. Historical Aspects The papers published by Dr. Ames and colleagues in 1975 and 1976 (Ames et al., 1975; McCann et al., 1975a; McCann and Ames, 1976) represent a culmination of many years of research in ‘‘perfecting’’ the Salmonella bacterial mutagenicity assay. In the 1960s, Dr. Ames was one of many researchers who used bacterial systems to study mutagenesis. Up to that time the most commonly used mutagenesis system employed E. coli WP2 strains, auxotrophic for the essential amino acid tryptophan, and defective in different DNA repair mechanisms. These strains contained the marker trpE, which is an ochre nonsense mutation, that was used by geneticists interested in radiation studies and in the analysis of DNA repair pathways and mutagenesis. Reversion from tryptophan dependence to tryptophan independence was used for measuring mutagenesis (Bridges et al., 1967; Hill, 1965; Witkin, 1956). In the early 1970s, researchers started using the E. coli tryptophan reverse mutation assay to evaluate chemicals for mutagenicity (Bridges et al., 1973; McCalla and Voutsinos, 1974).

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

383

Instead of working with the popular E. coli tryptophan mutants, Dr. Ames worked with strains of Salmonella typhimurium LT2, defective in the biosynthetic pathway of the essential amino acid histidine. These strains were obtained either spontaneously or after chemical or radiation treatment (Hartman et al., 1971). He soon discovered that there were speciﬁc markers in the histidine operon that were ‘‘hotspots’’ for reversion to histidine prototrophy by chemical mutagens. Ames and Whitﬁeld (1966) reported on the induction of frameshift mutations in a number of Salmonella strains. In 1971, Dr. Ames reported on the use of a set of histidine mutants with one of the following genetic markers: hisG46 (base pair substitution mutation), hisC3076 (a þ 1 frameshift mutation), and hisD3052 (a 1 frameshift mutation) to screen large numbers of chemicals for mutagenicity (Ames, 1971). These strains were also deﬁcient in the accurate DNA repair system via a deletion mutation through the uvrB-biotin genes. Elimination of the accurate DNA repair system allows more lesions to be repaired by the error-prone repair system (SOS repair). The extension of the deletion through the biotin genes makes the Salmonella tester strains dependent on biotin. In 1973, Dr. Ames and colleagues (Ames et al., 1973a) reported on a new and simple procedure for testing chemicals, namely the plate incorporation assay, now referred to as the standard plate incorporation assay. The new procedure combined the test article, a mammalian activation system, and bacteria in a small volume of top agar which was then poured on the minimal glucose agar plates. This assay was more sensitive than the spot test (Ames, 1971) and allowed for quantitative testing. In another publication (Ames et al., 1973b) the inclusion of an exogenous metabolic activation system was described. It provided a means to metabolize chemicals via the oxidative cytochrome P450 system that is absent in bacteria but present in humans and other mammals. Additional mutations/genetic manipulations that were introduced to make the Salmonella assay more sensitive are listed below: . Selection of the rfa mutation (Ames et al., 1973a). The rfa mutation leads to a defective lipopolysaccharide (LPS) layer, making the bacteria more permeable to bulky chemicals. . Selection of a new histidine marker hisD6610/hisO1242 (Levin et al., 1982a). . Introduction of the mutagenesis-enhancing plasmid pKM101 (McCann et al., 1975b). This plasmid confers resistance to ampicillin, which is a convenient marker to detect the presence of the

384

MORTELMANS AND RUPA

plasmid. Plasmid pKM101 is a derivative of the multidrug resistant plasmid R46, also referred to as R-Brighton, and was isolated by one of the authors (Mortelmans and Stocker, 1979). . Introduction of an additional histidine marker hisG428 with A:T base pairs at the mutation site, which was incorporated in the multi-copy plasmid pAQ1 (Levin et al., 1982b). Since 1975 the Ames test has been and is still being used by commercial companies and government agencies all over the world. Today it remains the most widely used in vitro short-term test, and consequently it has generated the largest database for chemical mutagens. Over the years, a few modiﬁcations to the original test design have been made, but the basic principles behind the test have remained intact. It is now recognized that the initial claim of a 90% concordance (proportion of correct identiﬁcations of carcinogens as mutagens) is dependent on what classes of chemicals are analyzed and how many chemicals of each class are used in the study. Some classes of carcinogens—that is, electrophylic alkylating agents—are readily detected as mutagens in the Ames test while others, such as hormones, are generally nonmutagenic (IARC, 1997). The lowest concordance was reported by Zeiger (1998), while the highest concordance, 93%, was reported by Purchase et al. (1978). 2. The Standard Plate Incorporation Assay The plate incorporation assay (Ames et al., 1975; Maron and Ames, 1983; Mortelmans and Zeiger, 2000) is depicted in Fig. 1. In brief, to a sterile 13-mm 100-mm test tube placed in a 43 C heating block are added in the following order: (1) 2 ml of 0.6% top agar supplemented with 0.6% NaCl, 0.05 mM biotin, and 0.05 mM histidine, (2) 0.05 to 0.1 ml of indicator organisms (about 108 bacteria), (3) 0.05 ml of a solution of the test article, and (4) 0.50 ml of metabolic activation mixture or buffer. The contents of each tube are gently mixed and then poured onto plates containing 25 ml of minimal glucose agar. After the top agar has set, the plates are incubated for 2 days at 37 C. The number of histidine revertant colonies is counted and compared with the number of the colonies on the negative control plates (bacteria only). Usually up to 5 doses are used spanning 3 logs with a high dose of 5–10 mg/plate for nontoxic chemicals. For toxic chemicals, an appropriate dose range is selected based on an initial toxicity experiment. The results are usually expressed as number of revertants/plate. If there is an increase in the number of revertant colonies per plate in a doserelated manner, the test article is considered a mutagen. Each strain has

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

385

FIG. 1. Schematic presentation of the Salmonella plate incorporation assay.

a characteristic number of spontaneously induced revertant colonies, as well as revertant colonies induced by positive control chemicals that are included in each assay. A mutagenic response in just one tester strain, whether in the presence or absence of a metabolic activation, is sufﬁcient to determine that a chemical is mutagenic. 3. The Tester Strains Table I summarizes the genotype of the most commonly used strains. Except for strains TA102 and TA104, which carry an ochre mutation at the mutation site that is rich in A:T base pairs, all strains carry the site of reversion in G:C rich regions. The strains of choice are, in descending order: TA100, TA98, TA1535, TA1537 or TA97, TA104 and/or TA102, and TA1538. Usually between two and ﬁve strains are used. To detect cross-linking mutagens that require an intact uvrB repair gene, the DNA repair proﬁcient strain TA102 is used. E. coli tryptophan mutant(s) are sometimes used in lieu of strains TA102 or TA104,

386

MORTELMANS AND RUPA TABLE 1 GENOTYPES OF THE MOST COMMONLY USED SALMONELLA TESTER STRAINS

Strain

Mutation

DNA target

uvrB-biotin

LPS defect

Plasmid

TA1535

hisG46

-G-G-G-base pair substitution

Deletion

rfa

None

TA100

hisG46

-G-G-G-base pair substitution

Deletion

rfa

pKM101

TA1538

hisD3052

-C-G-C-G-Cframeshift (1)

Deletion

rfa

None

TA98

hisD3052

-C-G-C-G-Cframeshift (1)

Deletion

rfa

pKM101

TA1537

hisC3076

Near –C-C-C-run frameshift (þ1)

Deletion

rfa

None

TA97

hisD6610

-C-C-C-C-C-Cframeshift (þ1)

Deletion

rfa

pKM101

TA104

hisG428

TAA (ochre)

Deletion

rfa

pKM101

TA102

hisG428

TAA (ochre)

Wild-type

rfa

pKM101 pAQ1

Adapted from Mortelmans and Zeiger (2000).

because the mutation in the tryptophan gene is also at an A:T base-pair. The tester strains usually include E. coli WP2 (pKM101), which is a repair proﬁcient strain, and E. coli (uvrA, pKM101), which is deﬁcient in the accurate repair system. For a review of the historical aspects of the tryptophan reverse mutation assay with E. coli WP2, see Mortelmans and Riccio (2000). The principle of the test is the same as that used in the Salmonella assay, with reversion at the tryptophan locus giving rise to prototrophy. A limited amount of tryptophan is added to the agar plates in lieu of histidine. Some researchers prefer to work with the E. coli tryptophan strains instead of Salmonella strains TA102 or TA104, both of which have a high spontaneous background. This has been recognized by regulatory agencies. Consequently, the bacterial mutation test is often referred to as the Salmonella/E. coli microsome assay. Because the existing mutations of the his tester strains are well characterized and speciﬁc mutations (i.e., base-pair substitution or frameshift) are required for their reversion to wild-type, the type of mutation induced by mutagenic test articles can be elucidated by examining the pattern of reversion response in several tester strains. This valuable information can be obtained from the Salmonella test but not from other routine genetic toxicology tests without DNA sequence analysis.

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

387

4. Modifications to the Ames Test A number of modiﬁcations have been introduced in the Ames test that enhance the detection of certain classes of chemicals. The most commonly used ones are: . The preincubation assay, which consists of preincubating the mixture of chemical, bacteria and buffer, or metabolic activation system, for 20 minutes prior to plating (Yagahi et al., 1975). . The desiccator assay for volatile liquids and gaseous compounds (Araki et al., 1994; Hughes et al., 1987; Simmon et al., 1977). . The Kado microsuspension assay for testing complex mixtures, including urine (Kado et al., 1983). . The spiral mutagenicity assay, which uses an automated approach to bacterial mutagenicity testing (Houk et al., 1989). . A miniscreen assay for use in the pharmaceutical industry (Burke et al., 1996). 5. Special Conditions for the Ames Test under Anaerobic Conditions When the Ames test is performed under highly reduced levels of oxygen, there is a reduction in spontaneous and induced revertant colonies. A complete absence of revertant colonies is observed when the testing is performed in a strict anaerobic environment with prereduced and anaerobically sterilized medium. The reduction or absence of revertant colonies is due to the fact that growth inhibitor(s) are formed by the histidine-dependent bacteria when they enter stationary phase on depletion of the available histidine (Mortelmans and Cox, 1992). These growth inhibitor(s) prevent the histidine revertant bacteria from forming colonies. The inhibitor(s) is (are) formed after 6–8 hours of plating and after incubation at 37 C. Therefore, removing the plates from the anaerobic environment at the time just prior to when the histidine-dependent bacteria enter stationary phase, and incubating the plates in air at 37 C for an additional 40–42 hours, should provide uncompromised results. B. IN VITRO MOUSE LYMPHOMA TKþ/ GENE MUTATION ASSAY The mouse lymphoma assay is the most commonly used in vitro mammalian mutation test. The test detects gene mutations (point mutations) and chromosomal events (deletions, translocations, mitotic recombination/gene conversion, and aneuploidy) that are known to be important in human genetic disorders, including cancer (Applegate

388

MORTELMANS AND RUPA

et al., 1990; Hozier et al., 1992; Moore et al., 1985; Sawyer et al., 1985, 1989). It employs an established cell line consisting of mouse lymphoma cells, L5178Y, with the heterozygous thymidine kinase gene (TKþ/) marker. Cells lacking the thymidine kinase (TK) activity are resistant to the pyrimidine analog triﬂuorothymidine (TFT) and are able to proliferate in the presence of TFT. This proliferation gives rise to large and small mutant colonies. Large mutant colonies appear to represent single gene mutations affecting the TK gene. Small colonies are believed to result from chromosomal damage to the TK and adjacent genes (Blazak et al., 1989; Clive et al., 1979; Combes et al., 1995). TK competent cells are sensitive to TFT, which causes the inhibition of cellular metabolism and halts further cell division. The assay was developed by Clive et al. (1971) and was later validated and standardized (Caspary et al., 1988; Clive et al., 1979; Moore et al., 1981). In this assay, cells are grown in THMG medium containing thydimine, hypoxanthine, methotrexate and glycine to eliminate cells lacking thymidine kinase activity. The cells are then transferred to THG medium (THMG without methotrexate) for 2 days. About 6 105 cell/ml) are exposed in ﬂasks for 3–6 hours with the test article in the presence or absence of a mammalian metabolic activation (e.g., rat liver homogenate). After washing, the cells are allowed to grow in nonselective culture medium for about 2 days. For expression of mutations, the cells are then grown in medium with and without TFT in soft agar in Petri dishes for 11 to 12 days in a humidiﬁed atmosphere containing 5% CO2, at which time colonies are counted. A modiﬁcation of this plating method is to clone and quantify mutants in multiwell microtiter plates (Cole et al., 1990; Oberly et al., 1997). This method appears to be advantageous over the plating method in Petri plates because of its increased cloning efﬁciency and the sensitivity of the assay. For nontoxic chemicals, the highest concentration is usually 5 mg/ml followed by up to 14 lower concentrations spanning 4 logs. For toxic chemicals, the highest dose should be based on cytotoxicity. In the range-ﬁnding experiment, cytotoxicity is usually determined by measuring the relative suspension growth of the cultures after the treatment period. In the mutagenicity experiment, the mutation frequency is calculated from the number of mutant colonies in cloning medium with and without TFT. Cloning efﬁciency is determined in cloning medium without TFT. Concurrent positive and negative (solvent or vehicle) controls, both with and without metabolic activation, are included in each experiment. Statistical analyses are used for data evaluation. The results are expressed as frequency of TFT-resistant colonies/

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

389

106 cells. There are several criteria for determining a positive result, such as a concentration-related or a reproducible increase in mutant frequency. If the test article is positive, colony sizing should be performed on at least one of the test cultures, preferably the culture that was exposed to the highest concentration that induced a positive response, and on the negative and positive controls. If the test substance is negative, colony sizing should be performed on the negative and positive controls only. C. IN VITRO MAMMALIAN CHROMOSOMAL ABERRATION ASSAY Chromosomal aberration assays are used to detect chemicals (i.e., clastogens) that induce chromosome damage, including chromatid and chromosome breaks, and complex chromosome changes such as exchanges, rings, and dicentrics. The induction of chromosomal aberrations in cells may play an important role in the neoplastic development of certain tumors (Yunis, 1983) and is believed to be a relevant biomarker for cancer risk in humans (Bonassi et al., 2000; Hagmar et al., 1998a, b; Lando et al., 1998; Liou et al., 1999). The test procedure usually employs cultured Chinese hamster ovary (CHO) cells by using a protocol that was developed by Galloway et al. (1985). In brief, cell cultures are exposed to the test article in T25 ﬂasks usually for 3–6 hours in the presence or absence of metabolic activation (e.g., rat liver homogenate). The cells are treated with a metaphasearresting substance for 2–4 hours (e.g., colcemid or colchicine) before the harvest approximately 1.5 times the cell cycle, usually 21 hours. Cells are ﬁxed in ﬁxative (methanol: acetic acid, 3:1), dropped on slides, air dried, and stained. Metaphase cells are analyzed microscopically for the presence of chromosomal aberrations. The cell culture conﬂuency and mitotic index are evaluated in an initial range-ﬁnding experiment to measure the cytotoxic effect. In the chromosomal aberration experiments, at least 200 well-spread metaphases should be scored microscopically per concentration and per negative and positive control. The number of metaphases to be examined can be reduced when high numbers of aberrations are observed. Chromosome analyses are performed by using the standard criteria (Savage, 1975). Though the purpose of the test is to detect chemicals that induce structural chromosomal aberrations, it is important to record polyploidy and endoreduplications when these events are seen. An increase in polyploidy may indicate that a chemical has the potential to induce numerical aberrations. However, the guidelines are not designed to measure numerical aberrations and are not routinely used

390

MORTELMANS AND RUPA

for this purpose. Different types of structural chromosomal aberrations should be listed with their numbers and frequencies for experimental and control cultures. Gaps are recorded separately and reported but generally are not included in the total aberration frequency. Concurrent positive controls should be included with each assay. For nontoxic chemicals, the highest concentration used is usually 5 mg/ml. An additional four lower doses are included in the test spanning 4 logs. Negative controls, consisting of solvent or vehicle alone in the treatment medium and treated in the same way as the treatment cultures, should be included for every harvest time. There are several criteria for determining a positive result, such as a concentration-related increase or a reproducible increase in the number of cells with chromosomal aberrations. Biological relevance of the results should be considered ﬁrst. Statistical methods may be used as an aid in evaluating the test results. However, a statistical signiﬁcance should not be the only determining factor for a positive response. D. IN VIVO MAMMALIAN ERYTHROCYTE MICRONUCLEUS TEST The rodent micronucleus test is the most widely used in vivo assay for the detection of clastogens and agents that induce aneuploidy. It provides valuable information about the chemical’s ability to disrupt mammalian chromosome structure and functions. Most human carcinogens are positive in mammalian micronucleus tests. Micronuclei arise from chromosome fragments as a result of chromosome breaks, or chromosome lag during anaphase. These fragments fail to become incorporated into daughter cell nuclei during cell division. It has been established that essentially all agents that cause double-strand chromosome breaks induce micronuclei. The bone marrow of rodents is routinely used in this test, since polychromatic erythrocytes are produced in this tissue. Erythrocytes arise from ‘‘stem cells’’ in the bone marrow and are produced by a series of divisions in a precursor cell population. The constant, rapid turnover of precursor cells makes erythrocytes an ideal cell type for a micronucleus test. Another unique feature of the erythrocyte is that immediately after formation of the fully differentiated erythrocyte, the nucleus is pushed out of the cell. Erythrocytes are the only mammalian cell type without a nucleus, and therefore the differentiated erythrocyte cannot further divide. Thus the bone marrow stem cells are continuously producing new erythrocytes to replace the ones that eventually die. If a stem cell is damaged by a chemical, and a micronucleus is formed as a consequence of this damage, the micronucleus remains in

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

391

the cell after the main nucleus has been pushed out. These micronuclei are very easy to observe microscopically. If there is evidence that the test substance, or a reactive metabolite, will not reach the target tissue, it is not appropriate to use this test. The test was initially developed by Schmid in 1976. It was later standardized and validated by Hayashi et al. (1983) and MacGregor et al. (1987). In brief, animals such as mice or rats are exposed either acutely or chronically to a test article. At predetermined times after exposure (usually 24 and 48 hrs), the animals are sacriﬁced and the bone marrow is extracted, smeared on slides, and stained. The frequency of micronucleated cells among the newly formed (RNA-containing) erythrocytes is determined. The newly formed erythrocytes are identiﬁed by staining the residual RNA that remains in the newly formed cells for about 2 days after enucleation. Cells that stain uniformly positive for RNA are referred to as polychromatic erythrocytes (PCEs). Cells that do not stain positively for RNA are referred to as normochromatic erythrocytes (NCEs). An increase in the frequency of micronucleated PCEs relative to the vehicle control group indicates that the test substance induced structural chromosomal damage or lagging chromosomes in the nucleated erythrocytes. Positive controls should produce micronuclei at exposure levels expected to give a detectable, statistically signiﬁcant increase over background. The highest concentration used for nontoxic chemicals is 2000 mg/kg body weight. This is followed by four additional lower dose levels spanning a total of 4 logs. The origination of the micronuclei, whether by chromosome breaks or by nondisjunction, can be distinguished immunochemically from the presence or absence of the kinetochore proteins in the micronuclei (Gudi et al., 1990) or by in situ DNA hybridization of the centromeric sequence in the micronuclei (Hayashi et al., 1994). The presence of kinetochore proteins, or the centromeric DNA sequence, indicates that the micronuclei were derived from whole chromosomes as a result of aneuploidy induction. These assays are not required by the regulatory agencies. IV. Regulatory Genetic Toxicology Guidelines All national and international regulatory agencies require a battery of genetic toxicology tests, since it is assumed that various combinations of bacterial and mammalian short-term tests should improve the predictability of rodent carcinogenicity. However, it is of interest to note that analysis of two different databases revealed that the Ames test by

392

MORTELMANS AND RUPA

itself was the most reliable assay for predicting rodent carcinogenicity, with a 60–70% predictability (Zeiger, 1998; Zeiger et al., 1990). Over the years, numerous regulatory guidelines have been established in the United States, Canada, Europe, the United Kingdom, Australia, Japan, and the Nordic countries (Choy, 2001). The minimum requirements of the major guidelines are similar, but there are some differences depending on the mission of the regulatory agency. The different guidelines have been streamlined, an effort that has resulted in the establishment of two major guidelines that are applicable on an international level. They are the International Conference on Harmonization (ICH) guidelines (ICH, 1996, 1997) speciﬁc for pharmaceutical products (Mu¨ller et al., 1999) and the Organization for Economic Cooperation and Development (OECD) guidelines (OECD, 1997), which were co-developed with the U.S. Environmental Protection Agency (USEPA), mostly for environmental and agricultural chemicals. To ease the burden faced by industry, the USEPA has also developed a series of harmonized genetic toxicology guidelines under the Ofﬁce of Prevention, Pollution and Toxic Substances (OPPTS) to aid in the evaluation of toxic substances and pesticides. This harmonization provides a single set of guidelines that meet the data requirements under regulatory legislation, such as the Toxic Substances Control Act (TSCA) and the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA). A review of the regulation of toxic substances and pesticides regulated under TSCA and FIFRA is provided by Cimino (2001). Following is a short history and some important details of these two major international guidelines. A. INTERNATIONAL CONFERENCE ON HARMONIZATION GUIDELINES The Technical Guidelines for Registration of Pharmaceuticals for Human Use, formulated by the ICH, were established by individuals from regulatory authorities and experts from the pharmaceutical industries in Europe, Japan, and the United States. The purpose for the internationally harmonized guidelines is to assure the quality, safety, and efﬁcacy of pharmaceutical products. The co-sponsors of ICH are as follows (Mu¨ller et al., 2001): . Commission of the European Union (EU) . European Federation of Pharmaceutical Industries Associations (EFPIA) . Ministry of Health and Welfare, Japan (MHW) . Japan Pharmaceutical Manufacturers Association (JPMA)

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

393

. U.S. Food and Drug Administration (FDA) . Pharmaceutical Research and Manufacturers of America (PhRMA) The International Federation of Pharmaceutical Manufacturers Association (IFPMA) serves in the capacity of coordinator and makes up the ICH secretariat. The World Health Organization, the European Free Trade Area, and Canada have been given an observer status in the ICH process. A steering committee consisting of individuals from the different parties of the three international regions oversees and guides the harmonization process (Mu¨ller et al., 2001). Each regulatory party in the three different regions agrees to accept ICH established guidance as regulatory policy in their respective regions. Expert working groups (EWGs) are then convened to develop a guidance plan(s) in their expert ﬁelds. The EWGs consist of six representatives, three each from regulatory agencies and pharmaceutical industries, from the six co-sponsors of ICH. These experts are responsible for creating draft guidance documents for speciﬁc assays, which ultimately will lead to their implementation in the three ICH regions. There are several steps involved in the ICH guidance development process as indicated below (Adapted from Mu¨ller et al., 2001): Step 1: Preparation of draft guidance document by joint regulatory/ industry EWGs based on scientiﬁc consensus. Step 2: Review of draft guidance document by Steering Committee with approval for its release for wider distribution. Step 3: Consolidation of comments on draft guidance document by individuals from regulatory agencies in the three regions. Step 4: Adoptation of harmonized ICH guidelines by regulators. Step 5: Implementation of the guidelines in the three ICH regions. Guidances are available from the internet at http://www.ifpma.org/. 1. Historical Note The ICH harmonization process for genetic toxicology testing was initiated in 1991 in Brussels, Belgium. The ﬁrst meeting of the Expert Working Group on genetic toxicology testing was held in 1992 in Tokyo, Japan. The EWG identiﬁed more than 60 strategic and technical issues that differed greatly between the regulatory agencies in the European Union, Japan, and the United States. The ﬁrst guidance on ‘‘Speciﬁc Aspects of Regulatory Genotoxicity Tests for Pharmaceuticals’’ (ICH S2A) was ﬁnalized by the EWG in July 1995. A second guidance on ‘‘A Standard Battery for Genotoxicity Testing of Pharmaceuticals’’ (ICH S2B) was ﬁnalized by the EWG in July 1997. A review and

394

MORTELMANS AND RUPA

comprehensive presentation of the ICH genetic toxicology guidance and its history have been published by Mu¨ller et al. (1999). The importance of ICH guidances is that they provide 1) detailed and speciﬁc recommendations for what constitutes a valid genetic toxicology test, 2) approaches to testing strategies, and 3) direction in the interpretation of results. 2. The ICH Standard Battery of Tests The standard battery of genetic toxicology tests speciﬁed in ICH guidance S2B consists of the following three tests: (1) a bacterial gene mutation test (e.g., the Salmonella/E. coli test), (2) an in vitro chromosomal aberration test with cultured mammalian cells (e.g., Chinese hamster ovary chromosomal aberration test) or an in vitro mammalian gene mutation test (e.g., the mouse lymphoma test), and (3) an in vivo test for chromosomal damage (e.g., mouse micronucleus test). The standard three-test battery serves as the primary screen for evaluating the mutagenic potential of pharmaceutical compounds. According to the ICH guidance, negative results in all of the tests in the standard battery are sufﬁcient to conclude with a high level of conﬁdence that there is an absence of genotoxicity. However, in some instances additional tests may be required if there is evidence that there are limitations associated with a particular target, cells, or the test system itself. Examples of limitations associated with any one of the tests are excess toxicity, insolubility of compound, and metabolic limitations. Additional tests may include the use of a modiﬁed test procedure, the use of a modiﬁed metabolic activation system, or an alternative validated test.

B. OECD GUIDELINES To ease the burden faced by industry with international markets, the USEPA accepts results from genetic toxicology tests produced in accordance with guidelines of the OECD. The OECD is an international organization that serves as a forum wherein governments of member nations discuss, develop, and coordinate issues related to environmental health safety. Activities include harmonizing chemical testing and hazard assessment, developing principles of Good Laboratory Practices (GLP), and cooperating in the investigation of existing chemicals. Currently the OECD comprises 29 member nations. Most are European countries, with three North America countries (Canada,

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

395

Mexico, and the United States), and four are Asian or Paciﬁc countries (Japan, Australia, New Zealand, and Korea). There are also a number of countries that are informally afﬁliated with the OECD (Cimino, 2001): (1) partners in transition (the Republic of Slovakia), (2) other nonmember countries (Brazil, Argentina, Chile, and Malta), (3) Center for European Economies in Transition, and (4) Dynamic Asian Economies in Transition (Hong Kong, Malaysia, Singapore, Taiwan, and Thailand). Member countries can express their views into a formal OECD Council Act, a document that is submitted to and reviewed at the highest level of the OECD. There are two types of Council Acts: the Decision of Council, which is legally binding, and the Council Recommendation, which may just include strong recommendations. 1. Historical Note In 1971 the OECD developed its ﬁrst guidelines, ‘‘Guidelines for the Testing of Chemicals,’’ to evaluate the safety of chemicals. These guidelines were to be used by government, industry, and independent laboratories for the safety evaluation of new and existing chemicals, as well as pesticides, pharmaceuticals, and food additives. In addition to covering tests for human health effects, these guidelines also cover tests for physical-chemical properties, environmental effects, and degradation and accumulation in the environment. After the initial adoption of the OECD guidelines in 1981, periodic updates were performed to keep up with the evolving state of sciences, which culminated in the adoption of seven OECD guidelines in genetic toxicology in 1997. An important aspect of the OECD guidelines is that the member countries are committed to the Mutual Acceptance of Data (MAD), which means that results generated in accordance with OECD guidelines and GLPs will be accepted in other member countries (Cimino, 2001). Since the OECD is not a government entity, copies of the OECD guidelines must be purchased. However, drafts of new or updated guidelines are available on the Internet (www.oecd.org/ ehs/testlist.htm). Information on the purchase is available on the OECD Web site (www.oecd.org). The following seven genetic toxicology guidelines were recently updated by the OECD: bacterial reverse mutation test, in vitro mammalian cell gene mutation test, in vitro mammalian chromosome aberration test, mammalian erythrocyte micronucleus test, mammalian bone marrow chromosome aberration test, mammalian spermatogonial chromosome aberration test, and the rodent dominant lethal assay.

396

MORTELMANS AND RUPA

2. The OECD Standard Battery of Tests For most toxic substances and pesticides that are regulated under TSCA and FIFRA in the United States, the battery of tests consists of a three-test system. The required tests are similar to those included in the ICH battery system, namely, (1) a bacterial gene mutation test, (2) an in vitro mammalian assay for chromosomal aberrations, and (3) an in vivo test for chromosomal effects. There is a speciﬁc requirement that pesticides be tested in the mammalian mouse lymphoma mutation assay. In general, there is an overlap between ICH and OECD guidelines. However, OECD guidelines tend to describe individual test methodologies and do not speciﬁcally address testing strategies and overall assessment issues. When negative results are obtained in the initial three-test system, no further testing is required. However, there are instances where additional in vitro and in vivo tests may required, including cancer bioassays, and possibly testing for induction of heritable mutations. Such scenarios may occur when there is concern of exposure routes and structure-activity relationships with known carcinogens. V. Concluding Remarks and Outlook Genetic toxicology tests are an integral part of the toxicological evaluation of chemicals for adverse human health effects, especially cancer. The most commonly used regulatory genetic toxicology tests are a series of well-characterized mutagenicity tests, which include in vitro bacterial and mammalian assays, as well as in vivo mammalian assays. They have a high predictive value for rodent carcinogenicity. However, these tests should not be used as a substitute for rodent carcinogenicity studies. They should be used to determine the spectrum of genetic effects of speciﬁc chemicals and to aid in the interpretation of genetic toxicity test results. There is a constant effort by researchers and regulators to make genetic toxicology a more predictive tool for carcinogenicity. There are mechanisms in place for introducing new test methods or modiﬁcations of test methods to regulatory agencies. In 1993 the Interagency Coordinating Committee on the Validation of Alternative Methods (ICCVAM) was established (NIEHS, 1997). There are three promising methods that have recently emerged: (1) the Comet assay (single cell gel electrophoresis, or SCG) for the detection of single and double DNA strand breaks in mammalian cells (Collins et al., 1997; Tice et al., 2000), (2) ﬂow cytometric analysis of micronuclei (Torous et al., 2003), and (3) in vitro micronucleus assay with cytochalasin B, which is faster and cheaper

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

397

for prescreening chemicals that need to be tested in the standard in vitro metaphase assay (Fenech, 1993). Other tests that look promising but need validation are (1) ﬂuorescence in situ hybridization (FISH), which uses DNA probes to detect speciﬁc chromosome aberrations (Eastmond and Pinkel, 1990; Trask and Pinkel, 1990; Tucker et al., 1995) and (2) the transgenic mouse model, which provides an opportunity to study in vivo gene mutations. It also gives insight into the complex mechanism of carcinogenesis and therefore has a great potential for genotoxicity testing (Burki et al., 1995; Rudolph et al., 1999; Sullivian et al., 1993). REFERENCES Ames, B. N. (1971). The detection of chemical mutagens with enteric bacteria. In ‘‘Chemical Mutagens, Principles and Methods for Their Detection’’ (A. Hollaender, ed.), Vol. 1, pp. 267–282. Plenum, New York. Ames, B. N., Lee, F. D., and Durston, W. E. (1973a). An improved bacterial test system for the detection and classiﬁcation of mutagens and carcinogens. Proc. Natl. Acad. Sci. USA 70, 782–786. Ames, B. N., Durston, W. E., Yamasaki, E., and Lee, F. D. (1973b). Carcinogens are mutagens: a simple test system combining liver homogenates for activation and bacteria for detection. Proc. Natl. Acad. Sci. USA 70, 2281–2285. Ames, B. N., McCann, J., and Yamasaki, E. (1975). Methods for detecting carcinogens and mutagens with the Salmonella/mammalian-microsome mutagenicity test. Mutat. Res. 31, 347–364. Ames, B. N., and Whitﬁeld, H. J., Jr. (1966). Frameshift mutagenesis in Salmonella. Cold Spring Harbor Symp. Quant. Biol. 23, 221–225. Araki, A., Noguchi, T., Kato, F., and Matsushima, T. (1994). Improved method for mutagenicity testing of gaseous compounds by using a gas sampling bag. Mut. Res. 307, 335–344. Applegate, M. L., Moore, M. M., Broder, C. B., Burrell, A., Juhn, G., Kasweck, K. L., Lin, P. F., Wadhams, A., and Hozier, J. C. (1990). Molecular dissection of mutations at the heterozygous thymidine kinase locus in mouse lymphoma cells. Proc. Natl. Acad. Sci. USA 87, 51–57. Blazak, W. F., Los, F. J., Rudd, C. J., and Caspary, W. J. (1989). Chromosome analysis of small and large L5178Y mouse lymphoma cell colonies: Comparison of triﬂuorothymidine-resistant and unselected cell colonies from mutagen-treated and control cultures. Mutat. Res. 224, 197–208. Bonassi, S., Hagmar, L., Stromberg, U., Montagud, A. H., Tinnerberg, H., Forni, A., Heikkila, P., Wanders, S., Wilhardt, P., Hansteen, I. L., Knudsen, L. E., and Norppa, H. (2000). Chromosomal aberrations in lymphocytes predict human cancer independently of exposure to carcinogens. European Study Group on Cytogenetic Biomarkers and Health. Cancer Res. 60, 1619–1625. Bridges, B. A., Dennis, R. E., and Munson, R. J. (1967). Differential induction and repair of ultraviolet damage leading to true reversions and external suppressor mutations of an ochre codon in Escherichia coli/B/r WP2. Genetics 57, 8897–8908. Bridges, B. A., Mottershead, R. P., Green, M. H., and Gray, W. J. H. (1973). Mutagenicity of dichlorvos and methyl methane sulphonate for Escherichia coli WP2 and some derivatives deﬁcient in DNA repair. Mutat. Res. 19, 295–303.

398

MORTELMANS AND RUPA

Burke, D. A., Wedd, D. J., and Burlinson, B. (1996). Use of the Miniscreen assay to screen novel compounds for bacterial mutagenicity in the pharmaceutical industry. Mutagenesis 11, 201–205. Burki, K., and Ledermann, B. (1995). Transgenic animals as pharmacological tools. Adv. Drug Res. 26, 143–177. Caspary, W. J., Daston, D. S., Myhr, B. C., Mitchell, A. D., Rudd, C. J., and Lee, P. S. (1988). Evaluation of the L5178Y mouse lymphoma cell mutagenesis assay: Interlaboratory reproducibility and assessment. Environ. Mol. Mutagen. 12(Suppl. 13), 195–229. Choy, W. N. (2001). Regulatory Genetic Toxicology Tests. In ‘‘Genetic Toxicology and Cancer Risk Assessment’’ (W. N. Choy, ed.), pp. 93–114. Marcel Dekker, Inc., New York. Cimino, M. C. (2001). New OECD Genetic Toxicology Guidelines and Interpretation of Results. In ‘‘Genetic Toxicology and Cancer Risk Assessment’’ (W. N. Choy, ed.), pp. 223–248. Marcel Dekker, Inc., New York. Clive, D., Flamm, W. G., Machesko, M. R., and Bernheim, N. J. (1971). A mutational assay system using the thymidine kinase locus in mouse lymphoma cells. Mutat. Res. 16, 7–87. Clive, D., Johnson, K. O., Spector, K. F. S., Batson, A. G., and Brown, M. M. M. (1979). Validation and characterization of the L5178Y/TKþ/ mouse lymphoma mutagen assay system. Mutat. Res. 59, 61–108. Cole, J., McGregor, M., Fox, M., Thacker, J., and Garner, R. C. (1990). Gene mutation assays in cultured mammalian cells. In ‘‘Basic Mutagenicity Tests’’ (D. J. Kirkland, ed.), pp. 87–114. Cambridge University Press, Avon. Collins, A. R., Dobson, V. L., Dusinska, M., Kennedy, G., and Stetina, R. (1998). The Comet assay: What can it really tell us? Mutat. Res. 375, 183–192. Combes, R. D., Stopper, H., and Caspary, W. J. (1995). The use of L5178Y mouse lymphoma cells to assess the mutagenic, clastogenic and aneugenic properties of chemicals. Mutagenesis 10, 403–408. Eastmond, D. A., and Pinkel, D. (1990). Detection of aneuploidy and aneuploidyinducing agents in human lymphocytes using ﬂuorescence in situ hybridization with chromosome-speciﬁc DNA probes. Mutat. Res. 234, 303–318. Fenech, M. (1993). The cytokinesis-block micronucleus technique: A detailed description of the method and its application to genotoxicity studies in human populations. Mutat. Res. 285, 35–44. Galloway, S. M., Bloom, A. D., Resnick, M., Margolin, B. H., Nakamura, F., Archer, P., and Zeiger, E. (1985). Development of a standard protocol for in vitro cytogenetic testing with Chinese hamster ovary cells: Comparison of results for 22 compounds in two laboratories. Environ. Mutagen. 7, 1–52. Gudi, R., Sandhu, S. S., and Athwal, R. S. (1990). Kinetochore identiﬁcation in micronuclei in mouse bone-marrow erythrocytes: An assay for the detection of aneuploidyinducing agents. Mutat. Res. 234, 263–268. Hagmar, L., Bonassi, S., Stromberg, U., Brogger, A., Knudsen, L. E., Norppa, H., and Reuterwall, C. (1998a). Chromosomal aberrations in lymphocytes predict human cancer: A report from the European Study Group on Cytogenetic Biomarkers and Health (ESCH). Cancer Res. 58, 4117–4121. Hagmar, L., Bonassi, S., Stromberg, U., Mikoczy, Z., Lando, C., Hansteen, I. L., Montagud, A. H., Knudsen, L., Norppa, H., Reuterwall, C., Tinnerberg, H., Brogger, A., Forni, A., Hogstedt, B., Lambert, F., Mitelman, I., Nordenson, S., Salomaa, S., and Skerfving, S. (1998b). Cancer predictive value of cytogenetic markers used in occupational health surveillance programs: Recent results. Cancer Res. 154, 177–184.

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

399

Hartman, P. E., Hartman, Z., Stahl, R. C., and Ames, B. N. (1971). Classiﬁcation and mapping of spontaneous and induced mutations in the histidine operon of Salmonella. Adv. Genet. 16, 1–34. Hayashi, M., Marke-Paakkanen, J., Tanabe, H., Honma, M., Suzuki, T., Matsuoka, A., Mizusawa, H., and Sofuni, T. (1994). Isolation of micronuclei from mouse blood and ﬂuorescence in situ hybridization with a mouse centromeric DNA probe. Mutat. Res. 307, 245–251. Hayashi, M., Sofuni, T., and Ishidate, M., Jr. (1983). An application of acridine orange ﬂuorescent staining to the micronucleus test. Mutat. Res. 120, 241–247. Hill, R. F. (1965). Ultraviolet-induced lethality and reversion to prototrophy in Escherichia coli strains with normal and reduced dark repair ability. Photochem. Photobiol. 4, 563–568. Hollstein, M., McCann, J., Angelosanto, F., and Nichols, W. (1979). Short-term tests for carcinogens and mutagens. Mutat. Res. 65, 133–226. Houk, V. S., Schalkowsky, S., and Claxton, L. D. (1989). Development and validation of the spiral Salmonella assay: An automated approach to bacterial mutagenicity testing. Mutat. Res. 223, 49–64. Hozier, J., Applegate, M., and Moore, M. M. (1992). In vitro mammalian mutagenesis as a model for genetic lesions in human cancer. Mutat. Res. 270, 201–209. Hughes, T. J., Simmons, D. M., Monteith, L. D., and Claxton, L. D. (1987). Vaporization technique to measure mutagenic activity of volatile organic chemicals in the Ames/ Salmonella Assay. Environ. Mutagen. 9, 421–441. International Agency for Research on Cancer (1997). ‘‘IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Overall Evaluation of Carcinogenicity to Humans.’’ Lyons, France. International Conference on Harmonization (1996). Guidance on speciﬁc aspects of regulatory genotoxicity tests for pharmaceuticals. Fed. Reg. 61, 18198–18202. International Conference on Harmonization (1997). Guidance on speciﬁc aspects of regulatory genotoxicity tests for pharmaceuticals. Fed. Reg. 62, 62472–62475. Kado, N. Y., Langley, D., and Eisenstadt, E. (1983). A simple modiﬁcation of the Salmonella liquid incubation assay: Increased sensitivity for detecting mutagens in human urine. Mutat. Res. 122, 25–32. Knudson, A. G. (1993). Antioncogenes and human cancer. Proc. Natl. Acad. Sci. USA 90, 11914–11921. Lando, C., Hagmar, L., and Bonassi, S. (1998). Biomarkers of cytogenetic damage in humans and risk of cancer. The European Study Group on Cytogenetic Biomarkers and Health (ESCH). Med. Lav. 89, 124–131. Levin, D. E., Yamasaki, E., and Ames, B. N. (1982a). A new Salmonella tester strain for the detection of frameshift mutagens: A run of cytosines as a mutational hotspot. Mutat. Res. 94, 315–330. Levin, D. E., Hollstein, M. C., Christman, M. F., Schwiers, E. A., and Ames, B. N. (1982b). A new Salmonella test strain (TA102) with A:T base pairs at the site of mutation detects oxidative mutagens. Proc. Natl. Acad. Sci. USA 79, 7445–7449. Liou, S. H., Lung, J. C., Chen, Y. H., Yang, T., Hsieh, L. L., Chen, C. J., and Wu, T. N. (1999). Increased chromosome-type chromosome aberration frequencies as biomarkers of cancer risk in a blackfoot endemic area. Cancer Res. 59, 1481–1484. Lohman, P. H. M., Mendelsohn, M. L., Moore, D. H., II, Waters, M. D., Brusick, D. J., Ashby, J., and Lohman, W. J. A. (1992). A method for comparing and combining short-term genotoxicity test data: The basic system. Mutat. Res. 266, 7–25.

400

MORTELMANS AND RUPA

MacGregor, J. T., Heddle, J. A., Hite, M., Margolin, B. H., Ramel, C., Salamone, M. F., Tice, R. R., and Wild, D. (1987). Guidelines for the conduct of micronucleus assays in mammalian bone marrow erythrocytes. Mutat. Res. 189, 103–112. Maron, D., and Ames, B. N. (1983). Revised methods for the Salmonella mutagenicity assay. Mutat. Res. 88, 343–350. McCalla, D. R., and Voutsinos, D. (1974). On the mutagenicity of nitrofurans. Mutat. Res. 26, 3–16. McCann, J., and Ames, B. N. (1976). Detection of carcinogens in the Salmonella/ microsome test: Assay of 300 chemicals: Discussion. Proc. Natl. Acad. Sci. USA 73, 950–954. McCann, J., Choi, E., Yamasaki, E., and Ames, B. N. (1975a). Detection of carcinogens in the Salmonella/microsome test: Assay of 300 chemicals. Proc. Natl. Acad. Sci. USA 72, 5135–5139. McCann, J., Spingarn, N. E., Kobori, J., and Ames, B. N. (1975b). Detection of carcinogens as mutagens: Bacterial tester strains with R factor plasmids. Proc. Natl. Acad. Sci. USA 72, 979–983. Mitelman, F. (1988). ‘‘Catalog of Chromosome Aberrations in Cancer.’’ Liss Inc., New York. Moore, M. M., Clive, D., Howard, B. E., Batson, A. G., and Johnson, K. O. (1981). The utilization of triﬂuorothymidine (TFT) to select for thymidine kinase-deﬁcient (TK/) mutants from L5178Y/TKþ/ mouse lymphoma cells. Mutat. Res. 85, 363–378. Moore, M. M., Clive, D., Hozier, J. C., Howard, B. E., Batson, A. G., Turner, N. T., and Sawyer, J. (1985). Analysis of triﬂuorothymidine-resistant (TFT) mutants of L5178Y/ TKþ/ mouse lymphoma cells. Mutat. Res. 151, 161–174. Mortelmans, K., and Cox, M. (1992). Evidence that inhibitor(s) are formed which may interfere with the growth of revertant colonies in the Ames Salmonella and the E. coli tryptophan reverse mutations assays when strict anaerobic conditions are used. Environ. Mol. Mutagen. 20, 44–52. Mortelmans, K., and Riccio, E. S. (2000). The bacterial tryptophan reverse mutation assay with Escherichia coli WP2. Mutat. Res. 455, 61–69. Mortelmans, K., and Stocker, B. A. D. (1979). Segregation of the mutator property of plasmid R46 from its ultraviolet-protection properties. Mol. Gen. Genet. 167, 317–327. Mortelmans, K., and Zeiger, E. (2000). The Ames Salmonella/microsome mutagenicity assay. Mutat. Res. 455, 29–60. Mu¨ller, L., Kikuchi, Y., Probst, G., Schechtman, L., Shimada, H., Sofuni, T., and Tweats, D. (1999). ICH-harmonized guidelines on genotoxicity testing of pharmaceuticals: Evolution, reasoning and impact. Mutat. Res. 436, 195–225. Mu¨ller, L., Kasper, P., and Schechtman, L. (2001). ICH guidances on genotoxicity and carcinogenicity: scientiﬁc background and regulatory practice. In ‘‘Genetic Toxicology and Cancer Risk Assessment’’ (W. N. Choy, ed.), pp. 187–221. Marcel Dekker, Inc., New York. National Institutes of Environmental Health Sciences (1997). Validation and Regulatory Acceptance of Toxicological Test Methods: A Report of the ad hoc Interagency Coordinating Committee on the Validation of Alternative Methods. NIH Publication No. 97–3981. NIEHS, Research Triangle Park, NC. Oberly, T. J., Yount, D. L., and Garriot, M. L. (1997). A comparison of the soft agar and microtiter methodologies for the L5178Y TKþ/ mouse lymphoma assay. Mutat. Res. 388, 59–66.

CURRENT ISSUES IN GENETIC TOXICOLOGY TESTING

401

OECD. (1997). Organization for Economic Cooperation and Development. ‘‘OECD Guidelines for the Testing of Chemicals.’’ TG nos. 471–486. Adopted 1984, 1986, 1997. Paris, France. Purchase, I. F. H., Longstaff, E., Ashby, J., Styles, J. A., Anderson, D., Lefevre, P. A., and Westwood, F. R. (1978). An evaluation of 6 short term tests for detecting organic chemical carcinogens. Br. J. Cancer 37, 873–959. Rudolph, U., and Mohler, H. (1999). Genetically modiﬁed animals in pharmaceutical research: Future trends. Euro. J. Pharmacol. 375, 327–337. Savage, J. R. K. (1975). Classiﬁcation and relationships of induced chromosomal structural changes. J. Med. Genet. 12, 103–122. Sawyer, J. R., Moore, M. M., Clive, D., and Hozier, J. C. (1985). Cytogenetic characterization of the L5178Y TKþ/ 3.7.2C mouse lymphoma cell line. Mutat. Res. 147, 243–253. Sawyer, J. R., Moore, M. M., and Hozier, J. C. (1989). High resolution cytogenetic characterization of the L5178Y TKþ/ mouse lymphoma cell line. Mutat. Res. 214, 181–193. Schmid, W. (1976). The micronucleus test for cytogenetic analysis. In ‘‘Chemical Mutagens, Principles and Methods for Their Detection’’ (A. Hollaender, ed.), Vol. 4, pp. 31–53. Plenum Press, New York. Simmon, V. F., Kauhanen, K., and Tardiff, R. G. (1977). Mutagenic activities of chemicals identiﬁed in drinking water. In ‘‘Progress in Genetic Toxicology’’ (D. Scott, B. A. Bridges, and F. H. Sobels, eds.), pp. 249–258. Elsevier/North Holland, Amsterdam. Sullivian, N., Gatehouse, D., and Tweats, D. (1993). Review: Mutation, cancer and transgenic models-relevance to the toxicology industry. Mutagenesis 8, 167–174. Tice, R. R., Agurell, E., Anderson, D., Burlinson, B., Hartmann, A., Kobayashi, H., Miyanae, Y., Rogas, E., Ryu, J. C., and Sasaki, Y. F. (2000). Single cell gel/comet assay: Guidelines for in vitro and in vivo genetic toxicology testing. Environ. Molec. Mutagen. 35, 206–222. Torous, D. K., Hall, N. E., Murante, F. G., Gleason, S. E., Tometsko, C. R., and Dertinger, S. D. (2003). Comparative scoring of micronucleated reticulocytes in rat peripheral blood by ﬂow cytometry and microscopy. Toxicol. Sci. 74, 309–314. Trask, B., and Pinkel, D. (1990). Fluorescence in situ hybridization with DNA probes. Methods Cell Biol. 33, 383–400. Tucker, J. D., Lee, D. A., and Minkler, J. L. (1995). Validation of chromosome painting. II. A detailed analysis of aberrations following high doses of ionizing radiation in vitro. Int. J. Radiat. Biol. 67, 19–28. Witkin, E. M. (1956). Time, temperature and protein synthesis: A study of ultravioletinduced mutations in bacteria. Cold Spring Harbor Symp. Quant. Biol. 21, 125–138. Yahagi, T., Degawa, W., Seino, Y., Matsushima, T., Nagao, M., Sugimura, T., and Hashimoto, Y. (1975). Mutagenicity of carcinogenic azo dyes and their derivatives. Cancer Lett. 1, 91–97. Yunis, J. J. (1983). The chromosomal basis of human neoplasia. Science 221, 227–236. Zeiger, E. (1998). Identiﬁcation of rodent carcinogens and noncarcinogens using genetic toxicity tests: Premises, promises, and performance. Regul. Toxicol. Pharmacol. 28, 85–95. Zeiger, E. (2001). Genetic toxicity tests for predicting carcinogenicity. In ‘‘Genetic Toxicology and Cancer Risk Assessment’’ (W. N. Choy, ed.), pp. 29–45. Marcel Dekker, Inc., New York. Zeiger, E., Hasman, J. K., Shelby, M. D., Margolin, B. H., and Tennanant, R. W. (1990). Evaluation of four in vitro genetic toxicity tests for predicting rodent carcinogenicity: Conﬁrmation of earlier results with 41 additional chemicals. Environ. Mol. Mutagen. 16, 1–14.

This Page Intentionally Left Blank

INDEX

A ACC deaminase, see 1-Aminocyclopropane-1-carboxylate deaminase Actinorhodin, production activation with ribosome engineering and mechanism, 157–62 Allergy animal model of type I allergy, 16–17 indigenous microﬂora role in immune response, 15–16 recombinant lactic acid bacteria modulation overview, 14–15 prophylaxis and therapy, 17–18 Ames test, see Salmonella/ mammalian-microsome test amf genes, Streptomyces development role, 73–77 1-Aminocyclopropane-1-carboxylate deaminase 1-aminocyclopropane-1-carboxylate metabolism overview, 291–92 ethylene in plant stress response, 293–95 legume nodulation promotion, 303–5 microbial genes and expression, 292–93 plant growth promotion by bacterial expression, 291 plant stress reduction applications with bacterial expression drought, 302–3 ﬂooding, 295–97 ﬂower wilting, 301–2 metals, 298–99 organic toxicants, 297–98 pathogen biocontrol, 300–1 salt stress, 303–4 prospects for agricultural application, 307

transgenic plant expression for stress reduction, 306–7 Antisense RNA, phage defense utilization for lactic acid bacteria, 355–62 Appressorium, see Ustilago maydis Arsenic, Trichoderma enhancement of extraction and degradation, 318–19

B Bacteriophage infection, lactic acid bacteria anti-receptor identiﬁcation, 352–53 comparative genomics studies, 351–52 dairy industry impact, 333 discovery of phages, 336 life cycles, 337–39 native defense systems abortive infection, 349–51 adsorption blocking, 344–45 plasmid encoding, 344 restriction and modiﬁcation systems, 345–49 prevention bacteriophage insensitive mutants, 342–43 engineered defense systems antisense RNA, 355–62 origin-derived phage-encoded resistance, 362–66 overview, 353–54 phage-triggered suicide systems, 368 subunit poisoning, 366–68 superinfection exclusion and immunity, 354–55 phage-inhibitory media, 340 prospects, 368–69 rotation of strains, 340–42 sanitation and manufacturing process improvements, 339–40

403

404

INDEX

Bacteroides-Flavobacterium-Cytophagia phylum, polysaccharide utilization in gut, 99–100 Biﬁdobacterium, polysaccharide utilization in gut, 105–6 bld mutants, Streptomyces, 70–72

C Carbohydrate binding module, see Clostridium glycosyl hydrolases Celesticetin, see Lincosamides Cellulose degradation, see Clostridium glycosyl hydrolases; Gut microbial fermentation Cellulosome, see Clostridium glycosyl hydrolases CFB phylum, see BacteroidesFlavobacterium-Cytophagia phylum Chromosomal aberration assays, genetic toxicology testing, 389–90 Clindamycin, see Lincosamides Clostridial cluster XIVa, polysaccharide utilization in gut, 104–5 Clostridium glycosyl hydrolases biomass conversion, 215–16 carbohydrate binding module and substrate binding, 219–22 characterization cellulose degradation cellulosome-producing bacteria, 231–32 Clostridium stercorarium soluble hydrolases, 233–35 Clostridium thermocellum cellulosomes, 235–43 Clostridium thermocellum soluble hydrolases, 243–44 enzyme features, 231 mesophylic clostridia cellulosomes, 245–47 substrate features, 229, 231 enzyme systems, 227–28 hemicellulose degradation, 247–51 starch degradation, 228–29 substrates, 225–26 supernatant analysis, 226–27 xylanases, 247–51

ﬁbronectin type II module structure and function, 224–25 industrial applications, 216–17, 251 modular structure, 217–18 surface-layer homology module structure and function, 223–24 codY, mutants for ribosome engineering Comparative genomics, lactic acid bacteria, 33–35, 351–52 Corn smut, see Ustilago maydis Cyanide, Trichoderma enhancement of extraction and degradation, 321–23

D Diarrhea, lactic acid bacteria in prevention, 9–10 DivIVA, Streptomyces development role, 69 Dye efﬂuents biodegradation aerobic biodegradation bacteria, 193–97 bioreactors, 201 fungi, 197–201 mechanisms of decolorization, 194, 198 anaerobic biodegradation, 201–3 combined anaerobic/aerobic degradation, 203–4 biosorption, 191–93 chemical treatment coagulation, 190 electrochemical removal, 191 overview, 189 oxidation, 190–91 photochemical degradation, 190–91 dye classiﬁcation and structures, 186–87 enzyme treatment, 206 physical removal adsorption, 188 irradiation, 190 membrane ﬁltration, 188, 190 overview, 189 prospects for decolorization, 206 textile industry efﬂuent features, 186 toxicity, 185

405

INDEX E egl1, Ustilago maydis pathogenesis role, 280 Erythrocyte micronucleus test, genetic toxicology testing, 390–91 eshA, mutants for ribosome engineering, 176–77 Ethylene chemicals in plant reduction, 295 legume nodulation inhibition, 303–5 microbial degradation, see 1-Aminocyclopropane1-carboxylate deaminase plant stress response, 293–95

F Fermentation, see Gut microbial fermentation Fibrobacter, polysaccharide utilization in gut, 101 Fibronectin type II module, structure and function in Clostridium glycosyl hydrolases, 224–25

G GBS, see Group B Streptococcus Genetic toxicology chromosomal aberration assays, 389–90 erythrocyte micronucleus test, 390–91 historical perspective, 381–82 mouse lymphoma assay, 387–89 Salmonella/mammalian-microsome test anaerobic testing, 387 history of development, 382–84 modiﬁcations, 387 standard plate incorporation assay, 384–85 tester strains, 385–86 testing overview, 379–80 prospects, 396–97 regulation International Conference on Harmonization guidelines, 392–94

Organization for Economic Cooperation and Development guidelines, 394–96 streamlining, 391–92 Glycosyl hydrolases, see Clostridium glycosyl hydrolases gpa3, Ustilago maydis pathogenesis role, 278 Group B Streptococcus, lactic acid bacteria for vaccine delivery, 14 Gut microbial fermentation biotechnology applications of microorganisms, 108–9 cross-feeding of breakdown products, 94 human dietary carbohydrates and health effects, 90–91 hydrogen fate, 92 manipulation enzymes, 106–7 prebiotics, 107–8 probiotics, 108 microbial diversity in rumen and small intestine, 92–93 polysaccharide utilization Bacteroides-FlavobacteriumCytophagia phylum, 99–100 Biﬁdobacterium, 105–6 clostridial cluster XIVa, 104–5 eukaryotes, 106 Fibrobacter, 101 Ruminococcus, 101, 103–4 stages attachment, 94–96 degradation, 96–98 transport, 98 prospects for study, 109–10 reductive acetogenesis, 92–94 rumen anatomy, 90

H hda1, Ustilago maydis pathogenesis role, 280 hgl1, Ustilago maydis pathogenesis role, 278 High-performance liquid chromatography, lincosamides, 128–29 HIV, see Human immunodeﬁciency virus HPV, see Human papillomavirus

406

INDEX

Human immunodeﬁciency virus, lactic acid bacteria for vaccine delivery, 11, 13–14 Human papillomavirus, lactic acid bacteria for vaccine delivery, 13 Hyphae, see Ustilago maydis

I IAA, see Indole-3-acetic acid IBD, see Inﬂammatory bowel disease ICH, see International Conference on Harmonization Indole-3-acetic acid bacteria synthesis, 291 corn smut pathogenesis role, 281 Inﬂammatory bowel disease, recombinant lactic acid bacteria management effector molecules for secretion, 22–23, 44 inﬂammation reduction, 20–22 intestinal barrier function role of lactic acid bacteria, 19–20 overview, 18 International Conference on Harmonization, genetic toxicology testing guidelines, 392–94

K Kinesin, Ustilago maydis pathogenesis role, 275

L LAB, see Lactic acid bacteria Laccases, dye efﬂuent treatment, 205 Lactic acid bacteria, see also Streptococcus thermophilus allergy modulation with recombinant bacteria animal model of type I allergy, 16–17 indigenous microﬂora role in immune response, 15–16 overview, 14–15 prophylaxis and therapy, 17–18

bacteriophage infection, see Bacteriophage infection, lactic acid bacteria engineering for safe use in humans biological containment systems, 32–33 food-grade systems for plasmid maintenance and chromosomal insertion, 28–32 overview, 28 regulations, 44 food industry uses, 2, 332–34 genome sequencing, 33–34 infectious disease prevention applications gastrointestinal tract infections, 9–11 overview, 3–4 recombinant bacteria for vaccine delivery, 4–7, 43 respiratory tract infections, 7–9 urogenital tract infections, 11–14 inﬂammatory bowel disease management with recombinant bacteria effector molecules for secretion, 22–23, 44 inﬂammation reduction, 20–22 intestinal barrier function role of lactic acid bacteria, 19–20 overview, 18 pharmaceutical protein production in recombinant bacteria advantages, 23–24 expression vector, 24–26 overview, 23 propagation, fermentation, and initial downstream processing, 26–28 probiotics, 2 prospects for study comparative genomics, 33–35 gene expression analysis and behavior in host, 35–39 host response, 39–42 Lincomycin, see Lincosamides Lincosamides analytical techniques, 128–29 antibiotic mechanism of action, 121, 135–37 biological activity and applications anaerobic bacteria, 143–45

407

INDEX Gram-negative bacteria, 141–43 Gram-positive bacteria, 139–41 pathogenic bacteria sensitivity, 138–39 prospects, 145–46 protozoa and parasites, 144–45 chemical modiﬁcation, 127–28 demethylation, 128 lincomycin biosynthesis amino acid moiety, 130 condensation, 133 genetic control, 133–34 N-methylation, 133 overview, 130–31 sugar moiety, 130, 132–33 production in microorganisms species, 121 strains for production, 124–27 sulfoxide formation, 128 resistance, 137–38 structures, 122–24, 126 lmb genes, lincosamide biosynthesis, 132 lmr genes, lincosamide resistance role, 134

M MAPK, see Mitogen-activated protein kinase Mass spectrometry, lincosamides, 129 MCP, see 1-Methylcyclopropene 1-Methylcyclopropene, ethylene stress reduction in plants, 295 mfa, Ustilago maydis mating role, 268 Mitogen-activated protein kinase, Ustilago maydis mating role, 269–70, 272 Mouse lymphoma assay, genetic toxicology testing, 387–89

Organization for Economic Cooperation and Development, genetic toxicology testing guidelines, 394–96

P PAHs, see Polycyclic aromatic hydrocarbons Peroxidases, dye efﬂuent treatment, 205–6 Phage-encoded resistance, bacteriophage defense utilization for lactic acid bacteria, 362–66 Phytoremediation, deﬁnition, 296 PKA, see Protein kinase A Polycyclic aromatic hydrocarbons phytoremediation, 297 Trichoderma enhancement of extraction and degradation, 324, 326 Polyphenols, Trichoderma enhancement of extraction and degradation, 323–24 ppGpp plant physiology, 177–79 regulation of antibiotic synthesis, 164–65 pra, Ustilago maydis mating role, 268 Prebiotics, gut microbial fermentation manipulation, 107–8 Probiotics, see also Lactic acid bacteria gut microbial fermentation manipulation, 108 Streptococcus thermophilus, 335–36 Protein kinase A, Ustilago maydis signaling, 273, 277–78 Pseudomonas, chemical tolerance induction, 171–72

R N Nitrates, Trichoderma enhancement of extraction and degradation, 319–21

O OECD, see Organization for Economic Cooperation and Development

ram genes, Streptomyces development role, 73–77 Recombination-based in vivo expression technology, lactic acid bacteria, 36–39 Respiratory tract infection, lactic acid bacteria in prevention, 7–9 Restriction and modiﬁcation systems, phage defense in Streptococcus thermophilus, 345–49

408

INDEX

Ribosome engineering cell-free translation systems, 179 combined drug-resistance mutations Streptomyces albus production strain, 173–75 Streptomyces coelicolor as model system, 172–73 drug-resistant mutant production, 156–57 ppGpp in plant physiology, 177–79 prospects for study codY mutants, 177 eshA mutants, 176–77 novel gene utilization, 175 Pseudomonas chemical tolerance induction, 171–72 rationale, 155–56 rpoB mutations antibiotic overproduction mechanisms, 165–67 neotrehalosadiamine, 169–70 ppGpp regulation of antibiotic synthesis, 164–65 enzyme overproduction, 169 rpsL mutation for antibiotic overproduction actinorhodin production activation and mechanism, 159–62 antibiotic-resistant mutations, 158–59 site-directed mutagenesis, 162–64 str mutation for antibiotic overproduction, 167–69 RNA polymerase, see rpoB rpoB, mutations antibiotic overproduction mechanisms, 165–67 neotrehalosadiamine, 169–70 ppGpp regulation of antibiotic synthesis, 164–65 str mutation, 167–69 enzyme overproduction, 169 rpsL, mutations for antibiotic overproduction actinorhodin production activation and mechanism, 159–62 antibiotic-resistant mutations, 158–59 site-directed mutagenesis, 162–64 Rumen, see Gut microbial fermentation Ruminococcus, polysaccharide utilization in gut, 101, 103–4

S S12, see rpsL Salmonella/mammalian-microsome test anaerobic testing, 387 history of development, 382–84 modiﬁcations, 387 standard plate incorporation assay, 384–85 tester strains, 385–86 SALPs, see SsgA-like proteins SapB, Streptomyces development role, 77–78 Short-chain fatty acids, gut fermentation products, 90 SIT, see Speciﬁc immunotherapy Small intestine, see Gut microbial fermentation Speciﬁc immunotherapy, allergy, 15 ssgA, Streptomyces development role, 79, 81 SsgA-like proteins, Streptomyces development role, 78–79, 82–83 ssgB, Streptomyces development role, 82 Starch, Clostridium degradation, 228–29 Streptococcus thermophilus bacteriophage infection, see Bacteriophage infection, lactic acid bacteria dairy product fermentation, 332–34 overview of characteristics, 334 probiotic use, 335–36 proteinases, 335 Streptomyces developmental signaling aerial mycelium formation, 70–71 amf genes, 73–77 carbon source effects, 71–72 novel gene discovery, 72–73 overview, 69–70 prospects for study, 83 ram genes, 73–77 SapB role, 77–78 sporulation regulators ssgA, 79, 81

409

INDEX SsgA-like proteins, 78–79, 82–83 ssgB, 82 submerged sporulation, 70 lincosamide synthesis, see Lincosamides liquid culture, 68–69 model species for study, 65–66 mutants, 66 mycelial growth models, 67–68 ribosome engineering, see Ribosome engineering sporulating colonies, 66 Subunit poisoning, phage defense utilization for lactic acid bacteria, 366–68 Surface-layer homology module, see Clostridium glycosyl hydrolases Synbiotic, deﬁnition, 108

T T-cells, host response to lactic acid bacteria, 40–41 Teliospore, see Ustilago maydis Trichoderma rhizosphere competence and co-metabolism, 315–16 root growth promotion, 316, 326 root symbiosis beneﬁts, 314–15 soil substrates, 313 toxicant extraction and biodegradation enhancement arsenic, 318–19 cyanide, 321–23 nitrates, 319–21 overview, 316–18 polycyclic aromatic hydrocarbons, 324, 326 polyphenols, 323–24 prospects, 326–27

U ubc1, Ustilago maydis pathogenesis role, 278 Ustilago maydis corn smut disease cycle, 263–64 economic impact, 263 gene expression studies fungus in planta, 281–82 maize, 282–84 host–pathogen interaction appressorium response, 272–73 culture growth comparison, 275–76 degradative enzymes, 273–74 gall induction genes, 277–78, 280 hormonal response, 280–81 infection sites, 271–72 nutrient acquisition, 276 penetrating hyphae, 271, 275–77 phases of infection, 274–75 protein kinase A signaling, 273, 277–78 teliosporogenesis genes, 277–78, 280 mating and pathogenicity, 267–71 prospects for study, 284–85 saprophyte budding, 265–66 ﬁmbriae, 265 germination regulation, 266 meiosis, 264–65 teliospores, 264 V Vaccine, recombinant lactic acid bacteria for delivery, 4–7, 43 X Xylanases, Clostridium, 247–51

This Page Intentionally Left Blank

CONTENTS OF PREVIOUS VOLUMES

Volume 41 Microbial Oxidation of Unsaturated Fatty Acids Ching T. Hou Improving Productivity of Heterologous Proteins in Recombinant Saccharomyces cerevisiae Fermentations Amit Vasavada Manipulations of Catabolic Genes for the Degradation and Detoxiﬁcation of Xenobiotics Rup Lal, Sukanya Lal, P. S. Dhanaraj, and D. M. Saxena Aqueous Two-Phase Extraction for Downstream Processing of Enzymes=Proteins K. S. M. S. Raghava Rao, N. K. Rastogi, M. K. Gowthaman, and N. G. Karanth Biotechnological Potentials of Anoxygenic Phototrophic Bacteria. Part I. Production of Single Cell Protein, Vitamins, Ubiquinones, Hormones, and Enzymes and Use in Waste Treatment Ch. Sasikala and Ch. V. Ramana Biotechnological Potentials of Anoxygenic Phototrophic Bacteria. Part II. Biopolyesters, Biopesticide, Biofuel, and Biofertilizer Ch. Sasikala and Ch. V. Ramana INDEX

Volume 42 The Insecticidal Proteins of Bacillus thuringiensis P. Ananda Kumar, R. P. Sharma, and V. S. Malik

Microbiological Production of Lactic Acid John H. Litchﬁeld Biodegradable Polyesters Ch. Sasikala The Utility of Strains of Morphological Group II Bacillus Samuel Singer Phytase Rudy J. Wodzinski and A. H. J. Ullah INDEX

Volume 43 Production of Acetic Acid by Clostridium thermoaceticum Munir Cheryan, Sarad Parekh, Minish Shah, and Kusuma Witjitra Contact Lenses, Disinfectants, and Acanthamoeba Keratitis Donald G. Ahearn and Manal M. Gabriel Marine Microorganisms as a Source of New Natural Products V. S. Bernan, M. Greenstein, and W. M. Maiese Stereoselective Biotransformations in Synthesis of Some Pharmaceutical Intermediates Ramesh N. Patel Microbial Xylanolytic Enzyme System: Properties and Applications Pratima Bajpai Oleaginous Microorganisms: An Assessment of the Potential Jacek Leman INDEX

411

412

CONTENTS OF PREVIOUS VOLUMES

Volume 44 Biologically Active Fungal Metabolites Cedric Pearce Old and New Synthetic Capacities of Baker’s Yeast P. D’Arrigo, G. Pedrocchi-Fantoni, and S. Servi Investigation of the Carbon- and Sulfur-Oxidizing Capabilities of Microorganisms by Active-Site Modeling Herbert L. Holland Microbial Synthesis of d-Ribose: Metabolic Deregulation and Fermentation Process P. de Wulf and E. J. Vandamme Production and Application of Tannin Acyl Hydrolase: State of the Art P. K. Lekha and B. K. Lonsane Ethanol Production from Agricultural Biomass Substrates Rodney J. Bothast and Badal C. Saha Thermal Processing of Foods, A Retrospective, Part I: Uncertainties in Thermal Processing and Statistical Analysis M. N. Ramesh, S. G. Prapulla, M. A. Kumar, and M. Mahadevaiah Thermal Processing of Foods, A Retrospective, Part II: On-Line Methods for Ensuring Commercial Sterility M. N. Ramesh, M. A. Kumar, S. G. Prapulla, and M. Mahadevaiah

Formation of Flavor Compounds in Cheese P. F. Fox and J. M. Wallace The Role of Microorganisms in Soy Sauce Production Desmond K. O’Toole Gene Transfer Among Bacteria in Natural Environments Xiaoming Yin and G. Stotzky Breathing Manganese and Iron: Solid-State Respiration Kenneth H. Nealson and Brenda Little Enzymatic Deinking Pratima Bajpai Microbial Production of Docosahexaenoic Acid (DHA, C22:6) Ajay Singh and Owen P. Word INDEX

Volume 46 Cumulative Subject Index Volume 47 Seeing Red: The Story of Prodigiosin J. W. Bennett and Ronald Bentley Microbial=Enzymatic Synthesis of Chiral Drug Intermediates Ramesh N. Patel Recent Developments in the Molecular Genetics of the Erythromycin-Producing Organism Saccharopolyspora erythraea Thomas J. Vanden Boom

INDEX

Bioactive Products from Streptomyces Vladisalv Behal

Volume 45

Advances in Phytase Research Edward J. Mullaney, Catherine B. Daly, and Abdul H. J. Ullah

One Gene to Whole Pathway: The Role of Norsolorinic Acid in Aﬂatoxin Research J. W. Bennett, P.-K. Chang, and D. Bhatnagar

Biotransformation of Unsaturated Fatty Acids of industrial Products Ching T. Hou

CONTENTS OF PREVIOUS VOLUMES Ethanol and Thermotolerance in the Bioconversion of Xylose by Yeasts Thomas W. Jeffries and Yong-Su Jin Microbial Degradation of the Pesticide Lindane (-Hexachlorocyclohexane) Brajesh Kumar Singh, Ramesh Chander Kuhad, Ajay Singh, K. K. Tripathi, and P. K. Ghosh Microbial Production of Oligosaccharides: A Review S. G. Prapulla, V. Subhaprada, and N. G. Karanth INDEX

Volume 48 Biodegredation of Nitro-Substituted Explosives by White-Rot Fungi: A Mechanistic Approach Benoit Van Aken and Spiros N. Agathos

Volume 49 Biodegredation of Explosives Susan J. Rosser, Amrik Basran, Emmal R. Travis, Christopher E. French, and Neil C. Bruce Biodiversity of Acidophilic Prokaryotes Kevin B. Hallberg and D. Barrie Johnson Laboratory Birproduction of Paralytic Shellﬁsh Toxins in Dinoﬂagellates Dennis P. H. Hsieh, Dazhi Wang, and Garry H. Chang Metal Toxicity in Yeasts and the Role of Oxidative Stress S. V. Avery Foodbourne Microbial Pathogens and the Food Research Institute M. Ellin Doyle and Michael W. Pariza Alexander Flemin and the Discovery of Penicillin J. W. Bennett and King-Thom Chung

Microbial Degredation of Pollutants in Pulp Mill Efﬂuents Pratima Bajpai

INDEX

Bioremediation Technologies for Metal-Containing Wastewaters Using Metabolically Active Microorganisms Thomas Pumpel and Kishorel M. Paknikar

Paleobiology of the Archean Sherry L. Cady

The Role of Microorganisms in Ecological Risk Assessment of Hydrophobic Organic Contaminants in Soils C. J. A. MacLeod, A. W. J. Morriss, and K. T. Semple The Development of Fungi: A New Concept Introduced By Anton de Bary Gerhart Drews Bartolomeo Gosio, 1863–1944: An Appreciation Ronald Bentley INDEX

413

Volume 50

A Comparative Genomics Approach for Studying Ancestral Proteins and Evolution Ping Liang and Monica Riley Chromosome Packaging by Archaeal Histones Kathleen Sandman and John N. Reeve DNA Recombination and Repair in the Archaea Erica M. Seitz, Cynthia A. Haseltine, and Stephen C. Kowalczykowski Basal and Regulated Transcription in Archaea Jo¨rg Soppa Protein Folding and Molecular Chaperones in Archaea Michel R. Leroux

414

CONTENTS OF PREVIOUS VOLUMES

Archaeal Proteasomes: Proteolytic Nanocompartments of the Cell Julie A. Maupin-Furlow, Steven J. Kaczowka, Mark S. Ou, and Heather L. Wilson Archaeal Catabolite Repression: A Gene Regulatory Paradigm Elisabetta Bini and Paul Blum INDEX

Volume 51 The Biochemistry and Molecular Biology of Lipid Accumulation in Oleaginous Microorganisms Colin Ratledge and James P. Wynn Bioethanol Technology: Developments and Perspectives Owen P. Ward and Ajay Singh Progress of Aspergillus oryzae Genomics Masayuki Machida Transmission Genetics of Microbotryum violaceum (Ustilago violacea): A Case History E. D. Garber and M. Ruddat Molecular Biology of the Koji Molds Katsuhiko Kitamoto Noninvasive Methods for the Investigation of Organisms at Low Oxygen Levels David Lloyd The Development of the Penicillin Production Process in Delft, The Netherlands, During World War II Under Nazi Occupation Marlene Burns and Piet W. M. van Dijck Genomics for Applied Microbiology William C. Nierman and Karen E. Nelson INDEX

Volume 52 Soil-Based Gene Discovery: A New Technology to Accelerate and Broaden Biocatalytic Applications Kevin A. Gray, Toby H. Richardson, Dan E. Robertson, Paul E. Swanson, and Mani V. Subramanian The Potential of Site-Speciﬁc Recombinases as Novel Reporters in Whole-Cell Biosensors of Pollution Paul Hinde, Jane Meadows, Jon Saunders, and Clive Edwards Microbial Phosphate Removal and Polyphosphate Production from Wastewaters John W. McGrath and John P. Quinn Biosurfactants: Evolution and Diversity in Bacteria Raina M. Maier Comparative Biology of Mesophilic and Thermophilic Nitrile Hydratases Don A. Cowan, Rory A. Cameron, and Tsepo L. Tsekoa From Enzyme Adaptation to Gene Regulation William C. Summers Acid Resistance in Escherichia coli Hope T. Richard and John W. Foster Iron Chelation in Chemotherapy Eugene D. Weinberg Angular Leaf Spot: A Disease Caused by the Fungus Phaeoisariopsis griseola (Sacc.) Ferraris on Phaseolus vulgaris L. Sebastian Stenglein, L. Daniel Ploper, Oscar Vizgarra, and Pedro Balatti The Fungal Genetics Stock Center: From Molds to Molecules Kevin McCluskey Adaptation by Phase Variation in Pathogenic Bacteria Laurence Salau¨n, Lori A. S. Snyder, and Nigel J. Saunders

CONTENTS OF PREVIOUS VOLUMES What Is an Antibiotic? Revisited Ronald Bentley and J. W. Bennett An Alternative View of the Early History of Microbiology Milton Wainwright The Delft School of Microbiology, from the Nineteenth to the Twenty-ﬁrst Century Lesley A. Robertson

415

LuxS and Autoinducer-2: Their Contribution to Quorum Sensing and Metabolism in Bacteria Klaus Winzer, Kim R. Hardie, and Paul Williams Microbiological Contributions to the Search of Extraterrestrial Life Brendlyn D. Faison INDEX

INDEX

Volume 54 Volume 53 Biodegradation of Organic Pollutants in the Rhizosphere Liz J. Shaw and Richard G. Burns Anaerobic Dehalogenation of Organohalide Contaminants in the Marine Environment Max M. Ha¨ggblom, Young-Boem Ahn, Donna E. Fennell, Lee J. Kerkhof, and Sung-Keun Rhee Biotechnological Application of Metal-Reducing Microorganisms Jonathan R. Lloyd, Derek R. Lovley, and Lynne E. Macaskie Determinants of Freeze Tolerance in Microorganisms, Physiological Importance, and Biotechnological Applications An Tanghe, Patrick Van Dijck, and Johan M. Thevelein Fungal Osmotolerance P. Hooley, D. A. Fincham, M. P. Whitehead, and N. J. W. Clipson Mycotoxin Research in South Africa M. F. Dutton Electrophoretic Karyotype Analysis in Fungi J. Beadle, M. Wright, L. McNeely, and J. W. Bennett Tissue Infection and Site-Speciﬁc Gene Expression in Candida albicans Chantal Fradin and Bernard Hube

Metarhizium spp.: Cosmopolitan InsectPathogenic Fungi – Mycological Aspects Donald W. Roberts and Raymond J. St. Leger Molecular Biology of the Burkholderia cepacia Complex Jimmy S. H. Tsang Non-Culturable Bacteria in Complex Commensal Populations William G. Wade l Red-Mediated Genetic Manipulation of Antibiotic-Producing Streptomyces Bertolt Gust, Govind Chandra, Dagmara Jakimowicz, Tian Yuqing, Celia J. Bruton, and Keith F. Chater Colicins and Microcins: The Next Generation Antimicrobials Osnat Gillor, Benjamin C. Kirkup, and Margaret A. Riley Mannose-Binding Quinone Glycoside, MBQ: Potential Utility and Action Mechanism Yasuhiro Igarashi and Toshikazu Oki Protozoan Grazing of Freshwater Bioﬁlms Jacqueline Dawn Parry Metals in Yeast Fermentation Processes Graeme M. Walker Interactions between Lactobacilli and Antibiotic-Associated Diarrhea Paul Naaber and Marika Mikelsaar

416

CONTENTS OF PREVIOUS VOLUMES

Bacterial Diversity in the Human Gut Sandra MacFarlane and George T. MacFarlane Interpreting the Host-Pathogen Dialogue Through Microarrays Brian K. Coombes, Philip R. Hardwidge, and B. Brett Finlay The Inactivation of Microbes by Sunlight: Solar Disinfection as a Water Treatment Process Robert H. Reed INDEX

Volume 55 Fungi and the Indoor Environment: Their Impact on Human Health J. D. Cooley, W. C. Wong, C. A. Jumper, and D. C. Straus Fungal Contamination as a Major Contributor to Sick Building Syndrome De-Wei LI and Chin S. Yang Indoor Moulds and Their Associations with Air Distribution Systems Donald G. Ahearn, Daniel L. Price, Robert Simmons, Judith Noble-Wang, and Sidney A. Crow, Jr. Microbial Cell Wall Agents and Sick Building Syndrome Ragnar Rylander The Role of Stachybotrys in the Phenomenon Known as Sick Building Syndrome Eeva-Liisa Hintikka Moisture-Problem Buildings with Molds Causing Work-Related Diseases Kari Reijula Possible Role of Fungal Hemolysins in Sick Building Syndrome Stephen J. Vesper and Mary Jo Vesper

The Roles of Penicillium and Aspergillus in Sick Building Syndrome (SBS) Christopher J. Schwab and David C. Straus Pulmonary Effects of Stachybotrys chartarum in Animal Studies Iwona Yike and Dorr G. Dearborn Toxic Mold Syndrome Michael B. Levy and Jordan N. Fink Fungal Hypersensitivity: Pathophysiology, Diagnosis, Therapy Vincent A. Marinkovich Indoor Molds and Asthma in Adults Maritta S. Jaakkola and Jouni J. K. Jaakkola Role of Molds and Mycotoxins in Being Sick in Buildings: Neurobehavioral and Pulmonary Impairment Kaye H. Kilburn The Diagnosis of Cognitive Impairment Associated with Exposure to Mold Wayne A. Gordon and Joshua B. Cantor Mold and Mycotoxins: Effects on the Neurological and Immune Systems in Humans Andrew W. Campbell, Jack D. Thrasher, Michael R. Gray, and Aristo Vojdani Identiﬁcation, Remediation, and Monitoring Processes Used in a Mold-Contaminated High School S. C. Wilson, W. H. Holder, K. V. Easterwood, G. D. Hubbard, R. F. Johnson, J. D. Cooley, and D. C. Straus The Microbial Status and Remediation of Contents in Mold-Contaminated Structures Stephen C. Wilson and Robert C. Layton Speciﬁc Detection of Fungi Associated With SBS When Using Quantitative Polymerase Chain Reaction Patricia Cruz and Linda D. Stetzenbach INDEX