Advances in Parasitology Volume 58

Advances in PARASITOLOGY

VOLUME 58

Editorial Board M. Coluzzi, Director, Istituto de Parassitologia, Universita` Degli Studi di Roma ‘La Sapienza’, P. le A. Moro 5, 00185 Roma, Italy C. Combes, Laboratoire de Biologie Animale, Universite´ de Perpignan, Centre de Biologie et d’Ecologie Tropicale et Me´diterrane´enne, Avenue de Villeneuve, 66860 Perpignan Cedex, France D.D. Despommier, Division of Tropical Medicine and Environmental Sciences, Department of Microbiology, Columbia University, 630 West 168th Street, New York, NY 10032, USA J.J. Shaw, Instituto de Cieˆncias Biome´dicas, Universidade de Sa˜o Paulo, av. Prof. Lineu Prestes 1374, 05508-900, Cidade Universita´ria, Sa˜o Paulo, SP, Brazil K. Tanabe, Laboratory of Biology, Osaka Institute of Technology, 5-16-1 Ohmiya Asahi-Ku, Osaka, 535, Japan P. Wenk, Falkenweg 69, D-72076 Tu¨bingen, Germany

Advances in PARASITOLOGY Edited by

J.R. BAKER Royal Society of Tropical Medicine and Hygiene, London, England

R. MULLER London School of Hygiene and Tropical Medicine, London, England and

D. ROLLINSON The Natural History Museum, London, England

VOLUME 58

Amsterdam Boston Heidelberg London New York Oxford Paris San Diego San Francisco Singapore Sydney Tokyo

ELSEVIER B.V. Sara Burgerhartstraat 25 P.O. Box 211, 1000 AE Amsterdam, The Netherlands

ELSEVIER Inc. 525 B Street Suite 1900, San Diego CA 92101-4495, USA

ELSEVIER Ltd The Boulevard Langford Lane, Kidlington Oxford OX5 1GB, UK

ELSEVIER Ltd 84 Theobalds Road London WC1X 8RR UK

ß 2004 Elsevier Ltd. All rights reserved. This work is protected under copyright by Elsevier Ltd, and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier’s Rights Department in Oxford, UK: phone (þ44) 1865 843830, fax (þ44) 1865 853333, e-mail: [email protected]. Requests may also be completed on-line via the Elsevier homepage (http://www.elsevier.com/locate/permissions). In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (þ1) (978) 7508400, fax: (þ1) (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (þ44) 20 7631 5555; fax: (þ44) 20 7631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of the Publisher is required for external resale or distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier’s Rights Department, at the fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. First edition 2004

British Library Cataloguing in Publication Data A catalogue record is available from the British Library. ISBN: 0-12-031758-3 ISSN: 0065-308X The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Printed in Great Britain.

CONTRIBUTORS TO VOLUME 58 J.-C. ANTOINE, Unite´ d’Immunophysiologie et Parasitisme Intracellulaire, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris cedex 15, France J. BETHONY, Cellular and Molecular Immunology Laboratory, ‘‘Rene´ Rachou’’ Research Centre FIOCRUZ, Barro Preto – CEP 30190-002, Belo Horizonte, Brazil S. BROOKER, Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK N. COURRET, Unite´ d’Immunophysiologie et Parasitisme Intracellulaire, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris cedex 15, France A. J. DAVIES, School of Life Sciences, Kingston University, Penrhyn Road, Kingston upon Thames, Surrey, KT1 2EE, UK B. FRIED, Department of Biology, Lafayette College, Easton, Pennsylvania 18042, USA T. K. GRACZYK, The W. Harry Feinstone Department of Molecular Microbiology and Immunology, and Department of Environmental Health Sciences, Bloomberg School of Public Health, Johns Hopkins University, 615 N. Wolfe Street, Baltimore, Maryland 21205, USA P. J. HOTEZ, Department of Microbiology and Tropical Medicine, The George Washington University, Washington DC 20037, USA T. LANG, Unite´ d’Immunophysiologie et Parasitisme Intracellulaire, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris cedex 15, France P. T. MONIS, Australian Water Quality Centre, South Australian Water Corporation and Cooperative Research Centre for Water Quality and Treatment, Private Mail Bag 3, Salisbury, SA 5108, Australia

ADVANCES IN PARASITOLOGY VOL 58 ISSN: 0065-308X DOI: 10.1016/S0065-308X(04)58006-5

v

vi

CONTRIBUTORS TO VOLUME 58

E. PRINA, Unite´ d’Immunophysiologie et Parasitisme Intracellulaire, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris cedex 15, France N. J. SMIT, Department of Zoology, Rand Afrikaans University, P.O. Box 524, Auckland Park 2006, South Africa R. C. A. THOMPSON, WHO Collaborating Centre for the Molecular Epidemiology of Parasitic Infections, Veterinary and Biomedical Sciences, Murdoch University, Murdoch, WA 6150, Australia

PREFACE

The opening paper in this volume concerns the intricate interactions between Leishmania and antigen-presenting cells of the mammalian host. Jean-Claude Antoine, Eric Prina, Nathalie Courret and Thierry Lang from the Institut Pasteur, Paris provide a detailed overview of how Leishmania spp. interact with two cell types, macrophages and dendritic cells, and describe some of the strategies used by Leishmania spp. to survive in these inducible or antigen-presenting cells. This is a fascinating account of the complex interactions that can occur between host and parasites. The authors highlight a number of questions and challenges in need of further research. In the next paper, Andrew Thompson of the University of Melbourne, Australia and Paul T. Monis from the Australian Water Quality Centre, Salisbury consider the variation observed in Giardia and the implications for taxonomy and epidemiology. Giardia is an intestinal parasite often encountered in humans, which can cause acute or chronic diarrhoea, dehydration, abdominal pain, nausea, vomiting and weight loss. Awareness of the parasite goes back a long time; indeed Giardia might have been observed as far back as 1681 by Antonie van Leeuwenhoek. It is interesting to read how the story has unfolded over the years and to appreciate the considerable ongoing debate that has concerned Giardia especially relating to the taxonomy, phylogeny and host specificity. The application of new molecular tools for identification and diagnosis are helping to unravel the mysteries of the transmission and host specificity of this parasite. Undoubtedly the findings have relevance to the control of giardiasis. The authors propose that this new information be reflected in the redesignation of several species of Giardia described previously. ADVANCES IN PARASITOLOGY VOL 58 ISSN: 0065-308X DOI: 10.1016/S0065-308X(04)58007-7

vii

viii

PREFACE

Bernard Fried at Lafayette College, Pennsylvania and Thaddeus Graczyk of Johns Hopkins University, both in the USA, continue the series of reviews on echinostomes (previous reviews in volumes 29, 38 and 49 of Advances in Parasitology). The 10 species of Echinostoma considered in the present review do not include the most important medical or veterinary parasites, although they can play a significant role in causing disease in waterfowl and aquatic mammals. Some species are also widely used as experimental models since the complete life cycles can be conveniently maintained in the laboratory. This has enabled them to be used to help elucidate many aspects of trematode biology including physiological, biochemical, immunological and molecular studies. These aspects, as well as systematic and descriptive studies, are comprehensively reviewed. Human hookworm infection is extremely common with estimates of over 700 million cases in the tropics and subtropics. Often occurring together with other intestinal helminths, hookworm infection remains an important public health problem. Indeed there has been a gradual realization that the effects of infection are greater than had been assumed in the past. In this review, Simon Brooker from the London School of Hygiene and Tropical Medicine, UK, Jeffrey Bethony from the ‘‘Rene´ Rachou’’ Research Centre FIOCRUZ, Brazil and Peter Hotez from The George Washington University, USA provide an extensive overview of current knowledge highlighting recent advances in our understanding of the biology, immunology, epidemiology and public health significance of hookworm infections. It is extremely encouraging that large-scale treatment campaigns are under way around the world and the authors consider the advantages of regular population-based chemotherapy. Nico Smit, of Rand Afrikaans University in South Africa, and Angela Davies, of Kingston University in the UK, complete the volume with an account of the relatively little-known but fascinating gnathiid isopods. These small crustacea have free-living, non-feeding adults and parasitic juveniles, comprising several larval stages, which feed on the blood and tissue fluids of fishes. Apart from the sometimes considerable pathogenic effects to the fish of this parasitism, at least one genus of gnathiid (Gnathia) serves as a vector

PREFACE

ix

of the apicomplexan protozoan Haemogregarina bigemina, a widespread parasite of teleosts. Smit and Davies suggest that further investigation of the capacity of gnathiids to act as vectors of other parasitic groups is warranted. John Baker Ralph Muller David Rollinson

This Page Intentionally Left Blank

CONTENTS CONTRIBUTORS TO VOLUME 58 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v vii

Leishmania spp.: on the Interactions They Establish with Antigen-Presenting Cells of their Mammalian Hosts Jean-Claude Antoine, Eric Prina, Nathalie Courret and Thierry Lang Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introductory Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. The Ms as Host Cells for Leishmania spp. . . . . . . . . . . . . . . . . . . . 3. The Ms as Cells Presenting Leishmania Antigens . . . . . . . . . . . . . 4. Ability of Infected Ms to Destroy Leishmania Parasites They Harbour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. The DCs as Cells That Can Also Shelter Leishmania spp. . . . . . . 6. Role of DCs in the Presentation of Leishmania Antigens to Naive and Activated Specific T Lymphocytes . . . . . . . . . . . . . . . 7. The Potential of Infected APCs or APCs Loaded with Leishmania Antigens as Vaccines or Therapeutic Agents . . . . . . . . . . . . . . . . . . . 8. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2 3 17 24 32 37 41 50 53 55 56

Variation in Giardia: Implications for Taxonomy and Epidemiology R.C.A. Thompson and P.T. Monis Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Current Taxonomy and Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . .

ADVANCES IN PARASITOLOGY VOL 58 ISSN: 0065-308X DOI: 10.1016/S0065-308X(04)58009-0

70 70 75

xi

xii

CONTENTS

3. 4.

Phenotypic Evidence for Variation in G. duodenalis . . . . . . . . . . . . 85 Genetic Characterization of Giardia and Prospects for a Revised Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 5. Epidemiology and Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 6. Perspectives for the Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

Recent Advances in the Biology of Echinostoma species in the ‘‘revolutum’’ Group Bernard Fried and Thaddeus K. Graczyk

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma caproni . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma trivolvis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma paraensei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma revolutum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma friedi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma miyagawai . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma echinatum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma parvocirrus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma luisreyi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Echinostoma jurini . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

140 140 142 152 165 169 174 175 176 177 177 178 178 179 180

Human Hookworm Infection in the 21st Century Simon Brooker, Jeffrey Bethony and Peter J. Hotez

1. 2. 3. 4. 5. 6. 7.

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Immune Responses to Hookworm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Epidemiology and Transmission Dynamics . . . . . . . . . . . . . . . . . . . . Public Health Consequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Global Distribution and Disease Burden . . . . . . . . . . . . . . . . . . . . . . . Strategies for Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

198 199 201 211 220 233 242 250

xiii

CONTENTS

8.

Future Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263

The Curious Life-Style of the Parasitic Stages of Gnathiid Isopods N.J. Smit and A.J. Davies Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gnathiid Phylogeny, Taxonomy and Morphology . . . . . . . . . . . . . Life Cycles of Gnathiids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Adaptation of Juveniles to Blood Feeding . . . . . . . . . . . . . . . . . . . . . Pathology and Transmission of Infection Associated with Blood Feeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Behaviour of Gnathiid Juveniles and the Role of Cleaner Fishes in their Removal from Clients . . . . . . . . . . . . . . . . . . 7. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. 2. 3. 4. 5.

289 290 303 317 334 358 369 373 377 378

INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393 CONTENTS OF VOLUMES IN THIS SERIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 The colour plate section appears between pages 50 and 51.

This Page Intentionally Left Blank

Leishmania spp.: on the Interactions They Establish with Antigen-Presenting Cells of their Mammalian Hosts Jean-Claude Antoine*, Eric Prina, Nathalie Courret and Thierry Lang Unite´ d’Immunophysiologie et Parasitisme Intracellulaire, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris cedex 15, France

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introductory Remarks . . . . . . . . . . . . . . . . . . . . . . . . . 1.1. The Life Cycles of Leishmania spp. . . . . . . . . . . . . . . 1.2. ‘‘Classical and Natural Experimental Models’’ of Leishmaniases . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3. Cells Containing Leishmania or Leishmania Antigens in Infected Mice . . . . . . . . . . . . . . . . . . . 1.4. Background on Ms and DCs: The Sentinels of the Body 1.5. Background on the Biosynthesis of MHC Class I and II Molecules by APCs and Formation of Peptide–MHC Molecule Complexes . . . . . . . . . . . . . . . . . . . . . . 2. The Ms as Host Cells for Leishmania spp. . . . . . . . . . . . . 2.1. Binding and Internalization of Promastigotes and Amastigotes . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. The Formation of PVs After Promastigote or Amastigote Phagocytosis and the Adaptation of Parasites to These Intracellular Niches . . . . . . . . . . . . . . . . . . . . . . . 3. The Ms as Cells Presenting Leishmania Antigens . . . . . . . . 3.1. PVs and the Ag Presentation Machinery . . . . . . . . . . . 3.2. MHC I Ag Presentation by Infected Ms . . . . . . . . . . 3.3. MHC II Ag Presentation by Infected Ms . . . . . . . . . . 3.4. Expression of Co-stimulatory Molecules by Infected Ms 3.5. In vivo Data . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

2 3 3

. .

6

. . . .

8 10

. . . .

14 17

. .

17

. . . . . . .

19 24 24 25 26 29 31

. . . . . . .

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

ADVANCES IN PARASITOLOGY VOL 58 ISSN: 0065-308X DOI: 10.1016/S0065-308X(04)58001-6

Copyright ß 2004 Elsevier Ltd. All rights of reproduction in any form reserved

2

J.-C. ANTOINE ET AL. 4. Ability of Infected Ms to Destroy Leishmania Parasites They Harbour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. The Different Pathways Leading to the Development of Leishmanicidal Properties . . . . . . . . . . . . . . . . . . . . 4.2. Mechanisms of Leishmania Killing . . . . . . . . . . . . . . . 4.3. How Leishmania Evade the Killing Mechanisms? . . . . . . . 4.4. In Susceptible Mice, Ms Can Follow an Alternative Activation Pathway Leading to Uncontrolled Parasite Growth . . . . . . . . . . . . . . . . . . . . . . . . . . 5. The DCs as Cells That Can Also Shelter Leishmania spp. . . . . . 5.1. Binding and Phagocytosis of Promastigotes and Amastigotes . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Phagosomal Compartments Housing Parasites in DCs. . . . 6. Role of DCs in the Presentation of Leishmania Antigens to Naive and Activated Specific T Lymphocytes . . . . . . . . . . . 6.1. Ability of Leishmania spp. to Induce DC Maturation and Migration . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. MHC I and MHC II Ag Presentation by DCs Put in Contact with Parasites or Leishmania Ags . . . . . . . . . . . 6.3. Possible Mechanisms used by Leishmania spp. to Limit Ag Presentation by DCs . . . . . . . . . . . . . . . . 6.4. DCs and the Polarization of Leishmania-Specific CD4 T Lymphocytes. . . . . . . . . . . . . . . . . . . . . . . . . . . 7. The Potential of Infected APCs or APCs Loaded with Leishmania Antigens as Vaccines or Therapeutic Agents . . . . . . . . . . . . . 7.1. APCs as Vaccines . . . . . . . . . . . . . . . . . . . . . . . . . 7.2. APCs as Therapeutic Agents . . . . . . . . . . . . . . . . . . . 8. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

32

. . .

32 35 36

. .

37 37

. .

37 40

.

41

.

41

.

43

.

48

.

49

. . . . . .

50 50 53 53 55 56

ABSTRACT Identification of macrophages as host cells for the mammalian stage of Leishmania spp. traces back to about 40 years ago, but many questions concerning the ways these parasites establish themselves in these cells, which are endowed with potent innate microbicidal mechanisms, are still unanswered. It is known that microbicidal activities of macrophages can be enhanced or induced by effector T lymphocytes following the presentation of antigens via MHC class I or class II molecules expressed at the macrophage plasma membrane. However,

LEISHMANIA SPP.

3

Leishmania spp. have evolved mechanisms to evade or to interfere with antigen presentation processes, allowing parasites to partially resist these T cell-mediated immune responses. Recently, the presence of Leishmania amastigotes within dendritic cells has been reported suggesting that they could also be host cells for these parasites. Dendritic cells have been described as the only cells able to induce the activation of naive T lymphocytes. However, certain Leishmania species infect dendritic cells without inducing their maturation and impair the migration of these cells, which could delay the onset of the adaptive immune responses as both processes are required for naive T cell activation. This review examines how Leishmania spp. interact with these two cell types, macrophages and dendritic cells, and describes some of the strategies used by Leishmania spp. to survive in these inducible or constitutive antigen-presenting cells.

1. INTRODUCTORY REMARKS 1.1. The Life Cycles of Leishmania spp. Leishmania spp. are heteroxenous, digenetic protozoan parasites and as such they live successively in two hosts, namely hematophagous insect vectors known as sand flies and some mammals playing the role of reservoirs, from which these infectious agents can be transmitted to other organisms of the same species or of a different species, including humans (for a review see Peters and Killick-Kendrick, 1987a; Schnur and Greenblatt, 1995). In female sand flies, Leishmania spp. exist extracellularly in the lumen of the digestive tract where they adopt a flagellated, elongated promastigote form and go through several differentiation stages. After a differentiation process called metacyclogenesis, promastigotes infective for mammals, termed metacyclic promastigotes, accumulate in the anterior parts of the digestive tract, from where they can be inoculated into the dermis of mammals during a blood meal (Rogers et al., 2002). In mammals, Leishmania spp. are obligate intracellular parasites. Indeed, after the bite of an infected sand fly, at least some of the injected metacyclics are rapidly engulfed

4

J.-C. ANTOINE ET AL.

by resident dermal phagocytic cells or cells rapidly recruited from the epidermis or the blood. During the early stages of the infection, a large part of the cells internalizing parasites appears to be macrophages (Ms), inside which promastigotes differentiate into egg-shaped amastigotes devoid of the external flagellum. This process takes several days and occurs within organelles named parasitophorous vacuoles (PVs), the morphology of which, and at least certain properties vary with different Leishmania species (Antoine et al., 1998; Courret et al., 2001). The life cycle is completed when a sand fly takes a blood meal on a parasitized mammal. During this process, the vector can be infected by free amastigotes or by infected mammalian cells located in the skin dermis. In the gut of the insect, the amastigotes differentiate rapidly into promastigotes. As an example, the cycle of L. amazonensis is presented in Figure 1. Humans can also be infected by numerous Leishmania species, but for most of them they are accidental hosts. About 12 million people distributed in 88 countries are suffering from leishmaniasis in the world, and it is estimated that 2 million new cases arise each year. In Europe, Africa and Asia, L. donovani, L. infantum, L. major, L. tropica and L. aethiopica are the main species infecting humans, whereas in South and Central America mainly L. chagasi, L. mexicana, L. amazonensis, L. guyanensis and L. braziliensis are responsible for leishmaniases. According to the Leishmania species initiating infection and their genetic/immunologic status, humans can remain asymptomatic or display more or less severe pathologic processes. Four major forms of human leishmaniases have been described: cutaneous, diffuse cutaneous, mucocutaneous and visceral. Cutaneous leishmaniases are generally benign. Parasites develop locally in the skin at the sites where infected sand flies have inoculated metacyclic promastigotes. In contrast, visceral leishmaniases are fatal in the absence of treatment. In these forms, parasites develop mainly in the liver, the spleen and bone marrow (for a review see Peters and Killick-Kendrick, 1987b; Schnur and Greenblatt, 1995). As to the wild mammalian reservoirs, which in many Leishmania life cycles are rodents, they are generally asymptomatic after infection or develop mild pathologies (Lainson and Shaw, 1979).

LEISHMANIA SPP.

5 Lutzomyia flaviscutellata

Metacyclic promastigotes in the mouthparts

Dividing promastigotes in the abdominal midgut

insect vector

mammalian host

Proechymis (natural reservoir)

Human (accidental host)

Phagocytosis of metacyclics Parasites begin their differen- Amastigotes multiply within parasitophorous vacuoles by dermal macrophages tiation into amastigotes

Life cycle of Leishmania amazonensis Phagocytosis of free amastigotes by new macrophages allows maintenance of infection

Figure 1 Life cycle of Leishmania amazonensis. This Leishmania species lives alternatively in the sand fly Lutzomyia flaviscutellata and in several rodents, especially in Proechymis spp. that are considered as the main reservoirs. L. amazonensis is endemic in 17 countries of South and Central America and is present mainly in wet forests of the Amazonian Basin. Humans can be incidentally infected by this Leishmania species when they penetrate forest areas but they play no role in the maintenance of the natural transmission cycle. See the text for the description of the different steps of the cycle.

6

J.-C. ANTOINE ET AL.

1.2. ‘‘Classical and Natural Experimental Models’’ of Leishmaniases 1.2.1.

‘‘Classical experimental models’’

The parasite infections due to Leishmania spp. have been the object of numerous and detailed studies for at least 15 years. Although mice are not natural hosts for Leishmania, inbred laboratory mice inoculated with these parasites have been widely used as experimental hosts for elucidating and characterizing the immunoparasitic processes involved in cutaneous and visceral leishmaniases. However, the ‘‘classical models’’ of leishmaniases are quite different from the situation in nature. They have typically relied on the inoculation of a high number of parasites (106–108 promastigotes, usually in stationary phase and thus heterogeneous, or sometimes lesionderived amastigotes, which are not the life cycle stage introduced into the mammalian host by the sand fly bite) into subcutaneous sites (generally the footpad) or intravenously. These protocols are very far from the natural infections where it has been estimated that vector sand flies inoculate 10 to 1000 metacyclic promastigotes into a dermal site (ear, tail, top of the foot). Host genetic factors controlling both the innate and adaptive immune responses, and consequently the infection outcomes have been discovered using these ‘‘classical models’’. For example, L. major parasites induce progressive cutaneous disease in BALB/c mice that are unable to control parasite expansion at the site of inoculation and in the draining lymph node. As a result, BALB/c mice are considered susceptible to L. major. In contrast, other strains (C57BL/ 6 or B10.D2) are considered resistant because the transient lesions that develop at the site of inoculation heal, parasite multiplication is controlled and they develop protective immunity as reflected by resistance to reinfection. However, a low number of amastigotes persist in mice clinically cured and immune to re-infection. Localization of these parasites is still a matter of debate but it seems clear that they are involved in the maintenance of long-lasting immunity. It has clearly been demonstrated that resistance of mice

LEISHMANIA SPP.

7

to primary infection with L. major is linked to the IL-12 driven activation/expansion of Leishmania-reactive CD4 Th1 cells also known as inflammatory CD4 T cells. This lymphocyte subset produces IFN- , which contributes to the activation of macrophages (Ms) and the nitric oxide (NO) dependent-killing and/or stasis of the intracellular amastigotes. On the other hand, mouse susceptibility to L. major has been correlated with the development of a Th2 type response (mediated by CD4 Th2 cells also known as helper CD4 T cells) and the inability to generate a sufficiently potent Th1 type response. In these susceptible mice, the production of cytokines such as IL-4, IL-13, IL-10, TGF- preventing and/or down regulating the IFN- -dependent macrophage activation is thought to participate in sustained disease development (for a review see Reiner and Locksley, 1995; Sacks and Noben-Trauth, 2002). It is important to stress that the scenario described above does not take into account the role of parasite genetic factors in determining the disease outcome. There is evidence that it does not apply to all Leishmania infections and events may differ according to the infecting Leishmania species. For instance, most inbred strains of mice, otherwise resistant to L. major, are susceptible to L. amazonensis, a situation that, for at least some mouse strains, results from an impaired Th1 type response rather than an enhanced Th2 type response (Afonso and Scott, 1993). Although this point is still being debated, a role for CD8 T lymphocytes in the resolution of primary murine L. major or L. amazonensis infections has been reported repeatedly (for a review see Milon et al., 1995). After their activation/differentiation, these cells acquire cytotoxic properties but they are also the source of cytokines and especially of IFN- .

1.2.2. ‘‘Natural experimental models’’ Recently, more ‘‘natural models’’ of Leishmania infections mimicking as much as possible the natural infections of wild mammals acting as reservoirs were set up in C57BL/6 and BALB/c mice. They rely on the

8

J.-C. ANTOINE ET AL.

inoculation of 10 to 1000 purified metacyclics of L. major or L. amazonensis into the ear dermis (Belkaid et al., 2000; Courret et al., 2003; Lang et al., 2003) or subcutaneously into the footpad (Lira et al., 2000). One of the characteristics of these models is that a parasite amplification occurs at the site of the inoculation before detection of clinical signs. For instance, in BALB/c mice infected in the ears with 1000 L. major metacyclic promastigotes, no lesion could be observed until about 3 weeks post-inoculation in spite of the fact that the parasite population increases at the site of infection. During this period, no production of cytokines (IFN- , IL-4) by lymph node CD4 T lymphocytes can be detected (Lang et al., 2003). Likewise, in B6 mice infected intradermally with L. major, the early parasite growth takes place without any detectable inflammatory process. During this phase, the presence of IL-4-producing cells is noted in the epidermis, and only very low amounts of IL-4 are produced by draining lymph node cells. This phase is followed by the development of a transient lesion concomitantly with the onset of IFN- production by lymph node and skin cells and a decrease in parasite load. Interestingly, after the clinical cure, a small number of living parasites persist permanently in the host (Belkaid et al., 2000). In this model, protective immunity is mediated by both IFN- -producing CD4 and CD8 T lymphocytes (Belkaid et al., 2002b) but at the same time, IL-10-secreting CD4 and CD25 regulatory T cells counterbalance the activity of these effector cells through physical contact or the production of cytokines, thus allowing persistence of parasites in an otherwise immune host (Belkaid et al., 2002a).

1.3. Cells Containing Leishmania or Leishmania Antigens in Infected Mice The different CD4 and CD8 T lymphocytes described earlier require to be primed or re-stimulated specialized cells known as antigenpresenting cells (APCs) including dendritic cells (DCs), Ms and B lymphocytes. APCs ensure the so-called antigen (Ag) presentation process that consists in the expression, at their cell surface, of

LEISHMANIA SPP.

9

complexes of peptides and MHC class I or class II molecules recognized by CD8 and CD4 T cells, respectively. These peptides are derived from exogenous or endogenous Ags that undergo an intracellular degradation known as Ag processing, after which they bind to MHC molecules either in compartments of the endocytic pathway (peptides bound to class II) or in the rough endoplasmic reticulum (RER) (peptides bound to class I). Recent data indicate, however, that under certain conditions, MHC class I–peptide complexes can be generated within phagosomal compartments (see later in this chapter). Additionally, APCs express on their cell surface co-stimulatory and adhesion molecules required for T lymphocyte activation. These cells can also process lipid Ags and express them in association with MHC class I-like molecules (CD1 molecules). These complexes are involved in the activation of a subset of T lymphocytes, the NKT cells, which express receptors of natural killer cells and T cells. However, the possible role of this lymphocyte subset during Leishmania infections remains to be documented. Several studies have examined the cells that are able to present Ags to Leishmania-specific naive T lymphocytes belonging to the different subsets described above, or cells that can reactivate already primed T lymphocytes. So far, Leishmania parasites/Ags have been detected in five types of cells, namely Ms, DCs, polymorphonuclear leukocytes, eosinophil granulocytes and fibroblasts (Grimaldi et al., 1984; Moll et al., 1997; Bogdan et al., 2000; Moll, 2000; Laskay et al., 2003). Among these cells, only Ms and DCs were clearly identified as APCs in vitro or in vivo. We will examine here the experimental data illustrating the roles that Ms and DCs play (i) in the parasite growth, (ii) in the induction, shaping and maintenance of the immune responses, and (iii) in parasite containment. Special attention will be paid to the Ag presentation capacity of these cells. In the ‘‘classical model’’ of leishmaniasis, based on the injection of a large number of L. major promastigotes, both Ms and DCs are rapidly infected after parasite inoculation (Moll et al., 1997; Moll, 2000; Muraille et al., 2003b). However, the lineages of cells encountered by the parasites in the early stages postinoculation may depend heavily upon the initial number of parasites

10

J.-C. ANTOINE ET AL.

and their infectiousness. For instance, a large number of inoculated Leishmania parasites can target a variety of host cells/host cell lineages, a point that could have an important impact on the disease development and progression. In this context, it is important to note that in the ‘‘natural model’’ of leishmaniasis based on the infection of B6 mice with a low number of L. major metacyclics, only Ms appear infected at the inoculation site during the parasite amplification phase preceding the onset of immunity (Belkaid et al., 2000), suggesting that the absence of anti-Leishmania immune response characterizing this silent phase could be linked to the lack of infected DCs. Likewise, it has been shown in a recent study that after inoculation of 1000 L. major metacyclics in the ear dermis of C57BL/6 or BALB/c mice, the only infected cells emigrating from ear skin explants prepared between day 3 and day 21 post-inoculation seem to be Ms (Baldwin et al., 2004). It is of interest that the absence of infected DCs has also been described in ‘‘classical experimental models’’ of visceral leishmaniasis that are characterized by an impaired T cell-mediated immune response in the spleen. Thus, after intravenous inoculation of 2  107 L. donovani stationary phase promastigotes into BALB/c mice, parasites present in the spleen 4 to 8 weeks later were localized mainly in MOMA-2 þ - (a monoclonal antibody that strongly reacts with mouse monocytes and Ms), macrosialin/CD68 þ - (a lysosomal glycoprotein of the lamp/lgp family strongly expressed in Ms) Ms, most of them expressing MHC class II molecules, but not in DEC205/CD205 þ - (a C-type lectin expressed by several DCs subsets) DCs (Lang et al., 2000) (Figure 2; see colour plate section).

1.4. Background on Ms and DCs: The Sentinels of the Body Before describing the interplay between Leishmania and Ms and DCs, the characteristics and properties of these two host cells will be briefly summarized. Ms and DCs originate from bone marrow cells and they are key players in both innate and adaptive immunity. As such, they are

LEISHMANIA SPP.

11

Figure 2 Characterization of the cells harbouring parasites in the spleen of a BALB/c mouse infected with L. donovani. Thirty days after parasite inoculation, the spleen was processed for immunohistochemistry. This section of the red pulp has been labelled with the monoclonal antibody MOMA-2, which binds to Ms, and then with an appropriate alkaline phosphatase conjugate (pink staining in the colour plate). After incubation with enzyme substrate, the section was counterstained with haematoxylin to visualize the parasite and mouse cell nuclei. Arrows point to infected cells. Bar, 20 mm. (See Plate 1.2 in colour plate section.)

located in almost every tissue/organ and are particularly numerous at the entry doors of foreign molecules/bodies and especially infectious agents. They display complementary and also partially overlapping functions. 1.4.1. Ms Ms constitute a family of heterogeneous tissue-resident cells and they are considered professional phagocytes. Being present mainly in

12

J.-C. ANTOINE ET AL.

lymph nodes, spleen, liver, lungs, gastrointestinal tract, serous cavities, bones, synovium, skin and central nervous system, they are not only involved in the internalization and killing of infectious agents, but also in the removal of altered cells and modified molecules, working in this manner in the maintenance of homeostasis. As other cells of the innate immune system, they recognize microorganisms owing to the expression at their cell surface of nonclonal receptors referred as pattern recognition receptors (PRRs) that specifically recognize invariant structures called PAMPs (for pathogen-associated molecular patterns) produced by microorganisms and likely metazoan pathogens (it must be stressed that PRRs could also have endogenous ligands such as matrix degradation products and heat shock proteins, released, for instance, during an inflammatory reaction). According to their microenvironment and the signals they receive, Ms can follow different pathways of activation (for instance, after recognition of microbial products or through the action of IFN- or IL-4/IL-13, or else, after receiving signals requiring cognate interactions with T lymphocytes), or can acquire a deactivated phenotype (for instance through the action of IL-10 or TGF- ). After their stimulation by IFN- or by other cytokines like IL-4/IL-13, they are able to generate MHC molecule–peptide complexes also known as ligands for T cell receptors (TCRs), especially ligands for CD4 T cells, and to activate T cells if they also express co-stimulatory molecules. Through their ability to express TCR ligands and to secrete a large panel of cytokines and other mediators, Ms are involved in the shaping and regulation of the adaptive immune responses. Killing of intracellular microorganisms by Ms is mainly due to the activation or induction of two effector mechanisms. Thus, rapid stimulation of Ms after the recognition of PAMPs by PRRs or activation by IFN- (produced for instance by Th1 cells after recognition of TCR ligands expressed at the M surface), in synergy with other molecules, leads to the production of reactive oxygen intermediates (ROI) and reactive nitrogen intermediates (RNI) by the phagocyte NADPH oxidase (phox) and the inducible nitric oxide synthase (iNOS or NOS2), respectively. Phox catalyses the reduction

LEISHMANIA SPP.

13

of molecular oxygen to superoxide O2 that is rapidly converted to hydrogen peroxide (H2O2) and hydroxyl radicals (OH ) whereas NOS2, as other NO synthases, generates nitric oxide NO from L-arginine and molecular oxygen. NO is then converted to various reactive oxides of nitrogen called RNI. In contrast, after activation of Ms by IL-4/IL-13, ROI production is not up regulated and expression of L-arginase I rather than that of iNOS is induced. L-Arginase I competes with iNOS for the same substrate, namely L-arginine and catalyses the production of L-ornithine that is used for polyamine biosynthesis. Ms are also endowed with numerous other functions that are beyond the scope of this review (for a review see Burke and Lewis, 2002; Gordon, 2003).

1.4.2. DCs DCs form a family of heterogeneous migratory cells and they are considered to be professional APCs involved in the permanent surveying of the body. In their immature state, they also display moderate phagocytic capacity compared to the high capacity of Ms. Spleen, lymph nodes, thymus, skin epidermis, mucosae of the respiratory, gastrointestinal and urogenital tracts, and interstitial spaces of vascularized organs such as heart and kidney are particularly rich in DCs. Like Ms, DCs are specialized in the recognition of PAMPs, which after binding to specific receptors can lead to DC activation. DCs also play critical roles in the adaptive immune system since they ensure the transport of Ags that they capture from the peripheral tissues to the draining secondary lymphoid organs, and they are the only APCs able to prime naive specific T lymphocytes. Depending on their state of differentiation and/or their lineage/subset and/or the microbial or endogenous signals they receive, they are involved in the induction of different types of T cell-dependent immune responses or in the induction of T cell tolerance. For example, immature, non-activated DCs loaded with Ags have been shown to induce anergy of specific T cells or the development of regulatory T cells that suppress the activation of

14

J.-C. ANTOINE ET AL.

effector T cells. At this developmental stage, DCs express on their plasma membrane low amounts of MHC class II and no or very few co-stimulatory molecules. In contrast, mature, activated, Agcontaining DCs, which display on the cell surface high levels of MHC class II and co-stimulatory molecules, have been described as potent inducers of T cell immunity. Interestingly, a third differentiation stage called semi-mature has been identified recently. These DCs have a surface phenotype very similar to that of mature DCs but, in contrast to the latter, they do not produce IL-12 (for a review see Liu et al., 2001; Lotze and Thomson, 2001).

1.5. Background on the Biosynthesis of MHC Class I and II Molecules by APCs and Formation of Peptide–MHC Molecule Complexes In this chapter, the different intracellular pathways leading to the formation and expression of complexes between peptides and MHC class I or MHC class II molecules will be briefly described. 1.5.1.

MHC I Ag presentation

Several mechanisms contribute to the generation of MHC I–peptide complexes, namely the so-called conventional and alternate mechanisms that lead to the formation of complexes between class I molecules and peptides derived from cytosolic Ags and exogenous phagocytosed Ags, respectively. This last process is also known as Ag cross-presentation. The conventional pathway is largely dependent upon newly synthesized class I molecules located in the RER. MHC I molecules are composed of a transmembrane MHC-encoded  chain and an invariant polypeptide, the 2 microglobulin, which are assembled with the help of various chaperone molecules such as calnexin and calreticulin. They are then loaded with peptides coming from the cytosol and translocated into the RER lumen via the transporter for Ag presentation (TAP). These peptides derive from proteins synthesized by the cells such as viral proteins, self

LEISHMANIA SPP.

15

proteins, tumour Ags or from microorganisms entering into the cytosol and they are generated by several proteolytic systems including the proteasome that has been most widely studied. The proteasome is a large complex of polypeptides and proteases localized in the cytosol and involved in the degradation of ubiquitinated cytosolic proteins. In APCs, the proteasome acquires specialized subunits allowing the generation of immunogenic peptides (immunoproteasome). MHC I–peptide complexes then follow the constitutive secretory pathway to reach the cell surface (for a review see Brodsky et al., 1999; Ramachandra et al., 1999). An alternate MHC I Ag presentation pathway has been found in phagocytic cells. It is very likely that the term alternate pathway encompasses several distinct mechanisms all leading to the formation of complexes between MHC I molecules and exogenous Ags, but in the context of this review we will discuss only that recently documented by M. Desjardins’ group (for a review see Desjardins, 2003) and resting on the transient formation of an endoplasmic reticulum (ER)–phagosome mix compartment. Electron microscopic studies demonstrate that nascent phagosomes can fuse with ER elements. Proteomic analyses of early phagosomes corroborate these results, since many proteins of the ER are found in these compartments, such as calnexin, calreticulin, MHC class I molecules, TAP and the translocon SEC61, a membrane protein complex involved in the translocation of nascent or newly synthesized polypeptide chains from the cytosol to the lumen of the ER and also mediating the retrotranslocation of proteins from the ER lumen to the cytosol. Furthermore, proteasome subunits and polyubiquitinated proteins have been localized on the cytoplasmic side of the phagosomes. Accordingly, it has been proposed that phagocytosed Ags could reach the cytosol using SEC61, could be cleaved by the phagosome-associated proteasome and the resulting peptides could return to the phagosome lumen through the TAP complex to be bound by the few phagosome-associated class I molecules. The existence of such a pathway has been recently proven in DCs and Ms using latex beads coated with proteins as model particles (Guermonprez et al., 2003; Houde et al., 2003).

16

1.5.2.

J.-C. ANTOINE ET AL.

MHC II Ag Presentation

As for the generation of MHC I–peptide complexes, several pathways involved in the formation of MHC class II–peptide complexes have been described. After their synthesis in the RER of APCs, MHC class II molecules that are composed of two transmembrane chains  and reach the Golgi apparatus and then the trans-Golgi network in association with transmembrane invariant chains Ii. Ii chains are endowed with important functions since they play a chaperone role for class II molecules, they block the peptide-binding groove of the latter and they allow their targeting from the trans-Golgi network to early or late endosomes. A small part of the class II–Ii complexes also reach the cell surface using the constitutive secretory pathway. These complexes then enter the endocytic compartments (early endosomes, late endosomes and lysosomes). In APCs, the late endosomes and lysosomes, crossed by newly synthesized class II molecules are also called MHC class II compartments (MIIC). In these sites, Ii chains are progressively degraded by endosomal/lysosomal proteases and notably by cathepsin S, cathepsin L and cathepsin F except for a small portion of these chains called CLIP (for class II-associated invariant chain peptide) that remains associated for a while with the peptide binding site of class II molecules. After this, class II molecules associate with MHC molecules H-2M that catalyse the removal of CLIP and favour the formation of high stability complexes between MHC class II molecules and peptides derived mainly from internalized antigens and generated by the action of lysosomal proteases. However, the organelles used by the class II–peptide complexes to reach the cell surface are largely unknown. More recently, studies have demonstrated that phagosomes formed in DCs and Ms are involved in the formation of peptide–MHC II complexes. Indeed, after their fusion with endocytic compartments and especially with late endosomes and lysosomes, phagosomal compartments acquire all the molecules required for the formation of these complexes, namely proteases, MHC class II and H-2M molecules. A third pathway called recycling/early endosome pathway using endocytosed MHC class II molecules recycling back to the cell surface has also been described.

LEISHMANIA SPP.

17

It requires less extensive Ag processing than the classical pathway and is thus more favourable to epitopes highly sensitive to proteolysis. Formation of these complexes occurs in early endocytic compartments and is Ii chain- and H-2M independent (for a review see Brodsky et al., 1999; Ramachandra et al., 1999).

2. THE MS AS HOST CELLS FOR LEISHMANIA SPP. 2.1. Binding and Internalization of Promastigotes and Amastigotes Both promastigotes and amastigotes can be phagocytosed by Ms through an actin-mediated phagocytic process. Recent studies show that in most cases, phagocytosis of both L. major and L. amazonensis metacyclic promastigotes proceeds first by the engulfment of the parasite cell body followed by the progressive internalization of the flagellum. Presence of actin around one part of the parasite but rarely around the entire parasite suggests that actin polymerization occurs sequentially all along the parasite until its complete internalization (Courret et al., 2002; N. Courret, S. Henri and E. Handman, unpublished observation). Although it has been previously claimed that initial interactions of promastigotes with M plasma membrane occur through the tip of the flagellum, the recent generation of L. amazonensis promastigotes devoid of external flagella has demonstrated that in vitro infectivity of this parasite stage for Ms is not dependent upon the presence of this organelle (Cuvillier et al., 2000, 2003). Numerous M receptors are involved in promastigote binding and phagocytosis including the mannose/fucose receptor, CD35 also known as CR1 (which binds the complement components C3b and C4b), CR3 (the integrin CD11b/CD18 that binds the complement component iC3b, ICAM-1 and extracellular matrix proteins) and the receptor for the C-reactive protein (an acute phase protein belonging to the pentraxin protein family) (for a review see Alexander et al., 1999). These receptors have been shown to bind either directly to parasite cell surface glycoconjugates such as the

18

J.-C. ANTOINE ET AL.

lipophosphoglycan (LPG, a complex glycolipid forming a dense glycocalyx), the leishmanolysin/GP63 (a family of zinc metalloproteases), the glycoinositolphospholipids of low molecular weight structurally related to LPG and a GPI-anchored membrane proteophosphoglycan, or indirectly through opsonization by complement components C3b and iC3b and the C-reactive protein (for a review see Handman, 2000). Interestingly, CR1 and CR3 are not involved in the release of inflammatory mediators and their use could thus facilitate parasite establishment in the Ms. Parasite establishment may also be facilitated by down regulation of several signalling events involved in the activation of microbicidal mechanisms, as described for Ms infected with L. major or L. donovani promastigotes. This may be due, at least in part, to the activation by Leishmania molecules of the M phosphotyrosine phosphatase SHP-1, a known negative regulator of numerous signalling pathways (Forget et al., 2001; Nandan et al., 2002) and possibly to the rapid expression of SOCS3 (suppressor of cytokine signalling 3) mRNA (Bertholet et al., 2003) encoding a protein involved in the negative regulation of several signalling pathways. The interaction of L. major, L. amazonensis or L. braziliensis promastigotes with Ms has also been shown to lead to the secretion of IL-10 and TGF- endowed with anti-inflammatory and M deactivating properties, and the secretion of IFN-/ that under certain conditions inhibit some M functions and make Ms more permissive to parasite growth (Mattner et al., 2000; Sacks and Sher, 2002). However, not all signalling cascades are down regulated by Leishmania promastigote infection. For instance, whereas IL-12 synthesis is inhibited in infected Ms, production of other proinflammatory cytokines is not blocked. Binding and phagocytosis of amastigotes are also mediated by several receptors such as the mannose/fucose receptor, CR3 and Fc Rs (in vivo, amastigotes released by infected Ms are very likely to be rapidly coated with Leishmania-specific antibodies, which can promote parasite internalization). The latter, which constitute a family of receptors belonging to the immunoglobulin superfamily, bind the Fc piece of IgGs and display either activating or inhibitory

LEISHMANIA SPP.

19

properties. The amastigote molecules involved directly or indirectly (through opsonins) in the binding and phagocytosis of this Leishmania stage remain to be identified. In this regard, it is interesting to note that in the amastigotes of most Leishmania species, there is a strong down regulation of the plasma membrane-associated high molecular weight glycoconjugates like the LPG and the leishmanolysin. Therefore, parasite molecules involved in amastigote internalization are at least in part different from those allowing promastigote phagocytosis. Inhibition of some but not all M signalling pathways has also been described in Ms infected with amastigotes of several Leishmania species. For instance, IFN- -induced IL-12 synthesis is blocked in infected Ms, whereas up regulation of MHC class II molecule synthesis by IFN- still occurs in Ms infected with certain Leishmania species (Antoine et al., 1998; Sacks and Sher, 2002). Infectious stages of Leishmania spp. thus seem to selectively modulate M functions establishing an intracellular niche compatible with their survival.

2.2. The Formation of PVs After Promastigote or Amastigote Phagocytosis and the Adaptation of Parasites to These Intracellular Niches Very quickly after internalization, promastigotes start their differentiation into amastigotes, a process that will not be completed before at least 5 days and which includes morphological changes (the external flagellum is lost 2 to 5 h after phagocytosis), an almost immediate change of the plasma membrane composition, biochemical modifications (Courret et al., 2001). In parallel with this differentiation process, the phagocytic compartments harbouring the parasites are the object of numerous remodelling events (Figure 3; see colour plate section). Shortly after internalization, L. major or L. amazonensis metacyclic promastigotes are located in very long phagosomes, the membrane of which tightly follows the outline of the parasites including the flagellum (Courret et al., 2002). The origin of this initial large amount of early phagosomal membrane is still a

20

J.-C. ANTOINE ET AL.

Figure 3 Biogenesis of PVs harbouring L. amazonensis in mouse bone marrow-derived Ms. After their internalization, metacyclic promastigotes are located in long, narrow phagosomes that are exactly the same shape as the parasites (300 ). From experiments conducted with L. donovani, a contribution of the rough endoplasmic reticulum (RER) in the formation of these organelles is suspected but has not yet been documented for this Leishmania species. At this time, most of the parasite-harbouring compartments have already acquired lysosomal membrane glycoproteins, contain

LEISHMANIA SPP.

21

matter of debate. While there is no doubt that the invaginated plasma membrane is involved in the formation of phagosomes, recent data indicate that, in Ms, the ER is the main source of phagosomal membrane surrounding several internalized microorganisms and notably L. donovani (Gagnon et al., 2002). The early Leishmania-harbouring organelles rapidly acquire phagolysosomal properties through fusion with late endocytic compartments, as shown by their acidification, the presence of late endosomal/ lysosomal glycoproteins in their membrane and the presence of several lysosomal hydrolases in their lumen (Alexander et al., 1999; Courret et al., 2002) (Figures 3 (see colour plate section) and 4 and Table 1). Likewise, phagosomes formed after internalization of L. donovani, L. mexicana or L. amazonensis amastigotes have been shown to rapidly acquire a competence to fuse with late endosomes/ lysosomes (Antoine et al., 1998; Alexander et al., 1999; Dermine et al., 2000). It must be stressed, however, that M. Desjardins et al. obtained different results about the biogenesis of PVs depending on the parasite life cycle stage used for infection. Indeed, these authors have shown that after the phagocytosis of L. donovani or L. major promastigotes, the parasites are transiently located in phagosomes with poor fusogenic properties towards late endocytic compartments whereas, after phagocytosis of the amastigote stage of these species, early phagosomes rapidly fuse with late endocytic compartments. Such a behaviour of the early phagosomes formed after promastigote several cathepsins and are surrounded by rab7p. If late endosomes/ lysosomes of the Ms are loaded with fluorescein dextran (Fdex) or horseradish peroxidase (HRP) before parasite phagocytosis, 300 phagosomes already contain these tracers clearly indicating that late endocytic compartments rapidly fuse with them. Later on (2–5 h), parasites lose their flagella and phagolysosomes shrink. Very likely, the excess of phagolysosomal membrane is removed by recycling vesicles. At 12 h, phagolysosomes that are still individual begin to expand and at 18 h they fuse together to form huge PVs. At this stage, PVs are still coated with rab7p and exhibit phagolysosomal properties. Although differentiation of the parasites is not yet achieved, they have already adopted the amastigote morphology. The involvement of high molecular weight proteophosphoglycans (PPG) secreted by the parasites in the PV expansion has been illustrated. (See Plate 1.3 in colour plate section.)

22

J.-C. ANTOINE ET AL.

Figure 4 Ultrastructural aspect of a bone marrow-derived M infected 18 h before with L. amazonensis metacyclic promastigotes. Parasites that have already started their differentiation into amastigotes are located in a huge communal PV. One hour before fixation, cells were incubated with horseradish peroxidase (HRP, 25 mg/ml) to load endosomes and lysosomes. Cells were then fixed and processed for peroxidase cytochemistry before embedding. HRP-containing endosomes/lysosomes (arrows) fuse with PV as indicated by the electron dense product present in the PV lumen. Presence of HRP at a parasite apex, a part of the microorganism that is involved in its attachment to the PV membrane, is also noted suggesting that this area is a centre of fusions with endocytic compartments of the host cell (arrowhead). Bar, 1 mm (micrograph from C. Fre´hel, N. Courret and J.-C. Antoine).

internalization would allow parasites to start differentiation in environmental conditions milder than those encountered in phagolysosomes. These effects may be linked to: (i) the expression of LPG by this parasite stage (Dermine et al., 2000; Spa¨th et al., 2003a), (ii) an impaired recruitment of rab7p (a small GTPase involved in the fusion of late endosomes/lysosomes, in the maturation of phagosomes and in the spatial organization/retrograde movement of these different compartments) (Scianimanico et al., 1999), (iii) a long-lasting accumulation of filamentous actin around the phagosomes (Holm et al., 2001) and (iv) the absence of translocation from the cytosol to the phagosome membrane of the protein

LEISHMANIA SPP.

23

Table 1 Characteristics of the PVs present in Ms and DCs infected 24 to 72 h ago with L. amazonensis metacyclic promastigotes or amastigotes

Morphology Parasite burden

pH Cathepsins (B, H, L, D) lamp-1 lamp-2 Macrosialin MOMA-2 Ag Iic MHC class II H-2M rab7p

Ms

Immature DCsa

Mature DCsa

strongly distended high number of parasites/PV; most PVs are communal acidicb þ

tight or slightly distended low number of parasites/PV; PVs are individual or communal acidicb þ/

tight or slightly distended low number of parasites/PV; PVs are individual or communal acidicb þ/

þ þ þ þ

þ þ þ þ þ/ þ þ þ/

þ þ þ þ NDe þ/ þ þ/

d

þd þd þ

a In all experiments summarized in this table, immature DCs were put in contact with parasites not coated or coated with specific antibodies. Twenty-four to 72 h later, most DCs have still an immature phenotype or have acquired a mature phenotype, respectively. b Quantitative determinations were done only for PVs present in Ms (pH 4.7 to 5.3). c Invariant chains involved in the trafficking of MHC class II molecules. d Ms treated with IFN- . e Not determined.

kinase C, an enzyme known to participate in phagosome maturation (Holm et al., 2001, 2003). The origin of the discrepancy between the results of Courret et al. and the latter ones is unclear, but the fact that different Leishmania strains and host cells were used in these studies is an important point to consider. In any case, the generality of the model proposed by M. Desjardins et al. was recently questioned by several new findings. First, it was found that, at least for L. mexicana, glycoconjugates with phosphoglycan repeats expressed on the promastigote cell surface including LPG are not determining factors for the differentiation of promastigotes into amastigotes (Ilg, 2000a; Garami et al., 2001; Ilg et al., 2001). Additionally, after their phagocytosis by mouse bone marrowderived Ms, both wild type and lpg null L. mexicana promastigotes

24

J.-C. ANTOINE ET AL.

have been found in phagocytic compartments that rapidly acquire the lysosomal glycoprotein lamp-1 (in less than 2 h), a type I membrane protein that is mainly expressed in late endosomes/ lysosomes (N. Courret and E. Handman, unpublished results). Finally, recent data show that after phagocytosis of L. major promastigotes, parasite survival is not linked to the transient inhibition of phagosome–lysosome fusion (Spa¨th et al., 2003a). Leishmania, and especially its amastigote stage, have evolved mechanisms to cope with the harmful conditions in PVs. Indeed, these microorganisms are acidophilic and resistant to acid hydrolases, a property that, until recently, had been linked to the strong expression of low molecular weight glycoinositolphospholipids at this parasite stage. However, construction of L. mexicana and L. major mutants unable to synthesize these molecules has led to the demonstration that they are not required for the survival of amastigotes within mouse Ms (Garami et al., 2001; Zufferey et al., 2003). On the other hand, PVs represent a good source of nutrients (Burchmore and Barrett, 2001) where the parasites have access to the various molecules internalized by the host cells, as well as to the metabolites derived from these molecules by the action of the lysosomal hydrolases. Recent data have also shown that amastigotes can gain access to molecules originating in the cytosol of the host cells and translocating into the PVs. Thus, small anionic molecules can be translocated into the PV lumen by a transporter present in the PV membrane, and larger cytosolic molecules can reach PVs via fusion of the latter with autophagosomes (Schaible et al., 1999).

3. THE MS AS CELLS PRESENTING LEISHMANIA ANTIGENS 3.1. PVs and the Ag Presentation Machinery Infected Ms can be potential APCs since they express MHC class I molecules and, after IFN- treatment, MHC class II and H-2M molecules (Antoine et al., 1998). While MHC class I molecules have

LEISHMANIA SPP.

25

not been detected in PVs, both MHC class II and H-2M molecules are generally easily detected in the membrane of PVs harbouring different Leishmania species (Antoine et al., 1998, 1999). This is true also for PVs of Ms isolated from spleens of L. donovani-infected mice, indicating the physiological relevance of the presence of class II molecules in these compartments (Lang et al., 2000). Although these molecules are in direct contact with the parasites, whether complexes between them and parasite Ags are generated in these organelles is still unknown. However, complexes between class II molecules and host cell Ags are detected in PVs (Antoine et al., 1998, 1999) indicating that PVs are engaged in the formation of such complexes or at least in their intracellular trafficking. Interestingly, MHC class II molecules, H–2M molecules as well as class II–peptide complexes present in PVs can be internalized by L. mexicana and L. amazonensis. The complexes accumulate in the parasite lysosomal organelles called megasomes where they are most likely degraded by parasite lysosomal proteases, especially cysteine proteases (Antoine et al., 1998, 1999). Both amastigotes and intermediate stages in the transformation between promastigotes and amastigotes (from 12 h post-phagocytosis of metacyclic promastigotes) can internalize molecules involved in the class II-restricted Ag presentation. This process is apparently specific, since other PV membrane molecules tested so far, namely the lysosomal glycoproteins lamp-1, lamp-2 and macrosialin, the MOMA-2 Ag and rab7p are not affected (Antoine et al., 1998; Courret et al., 2001). Of course, it is tempting to see in this phenomenon an escape mechanism from the adaptive immune response but the formal demonstration of such a possibility remains to be provided. Nevertheless, it is likely that internalization of PV membrane molecules by Leishmania is a means evolved by the parasites to modulate their intracellular niche. 3.2. MHC I Ag Presentation by Infected Ms Although studies have shown that CD8 T lymphocytes can play a protective role against Leishmania infections through their cytotoxic properties and IFN- production, the role of Ms and especially

26

J.-C. ANTOINE ET AL.

infected Ms in the activation/reactivation of these cells has not been thoroughly examined. Furthermore, the few studies bearing on this topic have provided contradictory results. Thus, Ms of various origin (bone marrow-derived, resident peritoneal, inflammatory) incubated with a parasite extract or infected with L. major amastigotes appear unable to reactivate primed, Leishmania-specific CD8 T cells purified from infected mice developing a protective immune response (Belkaid et al., 2002b). Using defined Ags, it has been demonstrated on the one hand that Ms infected with L. mexicana promastigotes producing Escherichia coli -galactosidase are unable to activate -galactosidase-specific CD8 T cell lines even after the killing of the intracellular parasites (Lo´pez et al., 1993). In contrast, L. amazonensis-infected Ms can present the promastigotespecific plasma membrane protein gp46 in the context of MHC class I molecules to specific CD8 T cell lines (McMahon-Pratt et al., 1998). Data from McMahon-Pratt and colleagues also suggested that gp46 reaches the host cell cytosol by an undefined mechanism, is processed by the proteasome, and the resulting peptides bind to newly synthesized MHC class I molecules present in the RER. Interestingly, recent proteomic analyses of the phagosomal compartments have led M. Desjardins’ group (Desjardins, 2003) to propose new mechanisms for the formation of complexes between MHC class I molecules and Ags of intracellular microorganisms residing in such organelles. The model is based on the fact that phagosomes in formation rapidly fuse with ER making the mix compartment at least transiently competent for the formation of MHC I–peptide complexes (see Section 1.5.1). Whether Leishmania Ags coming from phagocytosed parasites could follow this pathway to be eventually presented to CD8 T lymphocytes is not known. Clearly, studies on the role of Leishmania-infected Ms in the class I–restricted presentation of parasite Ags are required to settle this question. 3.3. MHC II Ag Presentation by Infected Ms More studies have been devoted to the presentation, by Leishmaniainfected Ms, of Ags associated with MHC class II molecules, and

LEISHMANIA SPP.

27

the conclusions from these studies are much clearer. It was first demonstrated that nonparasite Ags are less efficiently presented by Ms infected initially by promastigotes or amastigotes than by uninfected Ms, and that this deficiency increases with the parasite burden. In contrast, presentation of immunogenic peptides by infected and then fixed Ms is not altered or is even more efficient than their presentation by uninfected Ms (for a review, see Overath and Aebischer, 1999). On the whole, these data suggested that either the processing of native Ags, or the formation of complexes between class II and peptides derived from native Ags, or the trafficking of these complexes towards the plasma membrane is modified by the presence of the parasites, leading to a lower expression of cell surface MHC class II–peptide complexes. Alternatively, a parasite-induced alteration of the distribution rather than of the expression of plasma membrane-associated MHC class II–peptide complexes, which in uninfected Ms are partly localized in lipid rafts or tetraspan microdomains, has been recently proposed to explain these findings (Meier et al., 2003). Recall that lipid rafts are membrane microdomains enriched in cholesterol, glycosphingolipids and GPIanchored proteins and that tetraspan microdomains are membrane domains made up of tetraspan proteins such as CD9, CD63, CD81, CD82, which can interact to form clusters and bind other membrane proteins like integrins or MHC class II molecules. The role of these domains in Ag presentation and in the formation of immunological synapses between APCs and CD4 T cells has been recently illustrated (for a review see Vogt et al., 2002; Poloso and Roche, 2004). The possibility of competition between peptides derived from Ags expressed at the parasite surface or released by live parasites with peptides derived from exogenous Ags was also put forward to explain the infected M deficiency described above, but has not been supported by subsequent studies. First, after infection of Ms with amastigotes of L. amazonensis, no clear increase of SDS-resistant, stable peptide–class II molecule complexes can be detected, suggesting that amastigotes are not an important source of antigenic peptides (Antoine et al., 1999). Second, it has also been shown that Ms infected with the amastigotes of different Leishmania species are

28

J.-C. ANTOINE ET AL.

unable to present several parasite Ags expressed in different sites, namely a cysteine proteinase located in megasomes, the cytosolic protein LACK (Leishmania homologue of receptors for activated C kinase), a membrane-bound acid phosphatase associated with parasite organelles, the plasma membrane-associated Ags gp46 and P-8 (for a review, see McMahon-Pratt et al., 1998; Overath and Aebischer, 1999). In contrast, Ms infected with the promastigote stage can transiently present some Leishmania Ags such as LACK, gp46 and P-8 and the level of presentation correlates, at least for the Ag LACK, to the degree of virulence of the phagocytosed promastigotes (log phase, stationary phase, metacyclic), the less virulent, namely the log phase promastigotes, being the best for the generation of class II–peptide complexes (McMahon-Pratt et al., 1998; Courret et al., 1999). These data are consistent with the idea that only Ms containing parasites that are rapidly killed after internalization are able to present some Leishmania Ags, including Ags expressed at the parasite plasma membrane. However, parasite killing is not always sufficient for Ag presentation by infected Ms. For instance, partial or complete killing of intracellular amastigotes at different times post-phagocytosis does not allow presentation of LACK (Courret et al., 1999). Thus, parameters other than the resistance to killing must be involved in the weak presentation or in the lack of presentation of parasite Ags characterizing Ms initially infected with metacyclics or amastigotes. Factors such as the nature/ the degree of maturation of the parasite-harbouring organelles that can vary with the sheltered parasite form, or the expression/set up by the virulent stages of some molecules/mechanisms interfering with the Ag presentation machinery may be important (Courret et al., 1999) such as the secretion of cysteine proteases (Duboise et al., 1994) or high molecular weight polyanionic proteophosphoglycans (for a review see Ilg, 2000b). Taken together, these results suggest that virulent stages of Leishmania have evolved strategies to avoid or minimize their recognition by naive and effector CD4 T lymphocytes (Figure 5; see colour plate section). At this point, it is however important to emphasize that infected Ms are able to present parasite Ags provided that they come from the external milieu. In that

LEISHMANIA SPP.

29

case, efficiency of presentation depends upon the APC parasite load, as noted above for the presentation by infected Ms of exogenous nonparasite Ags. So, in vivo, it is quite possible that infected Ms take up parasite Ags released by parasitized cells or originating from live or killed extracellular parasites allowing them to reactivate CD4 T lymphocytes more or less efficiently according to the level of MHC class II expression and the number of parasites they harbour. It has been reported that salivary gland extracts of the insect vector Lutzomyia longipalpis can decrease the ability of Ms infected with L. major stationary phase promastigotes to reactivate already primed T cells but whether this effect is due to an alteration of the Ag presentation process has not been investigated (Theodos and Titus, 1993). In any case, these results suggest that vector saliva could synergize with parasite strategies to enhance parasite infectivity.

3.4. Expression of Co-stimulatory Molecules by Infected Ms The data discussed above indicate that the sensu stricto class IIrestricted Ag presentation process is at least partially altered in Leishmania-infected Ms and their ability to stimulate T cells is compromised. The capacity of APCs to activate CD4 T lymphocytes and especially naive ones is also dependent on other cell surfaceassociated molecules namely adhesion molecules (e.g. ICAM-1, ICAM-2, LFA-1, LFA-3) and co-stimulatory molecules like B7-1 (CD80) and B7-2 (CD86). These are involved in the transient binding of T cells to APCs and in signalling cascades leading to T cell activation, respectively. In the absence of co-stimulatory signal delivered by APCs, interacting naive T cells are inactivated and acquire a state known as anergy, or die by apoptosis. In contrast, effector T cells are less dependent for their activation upon the presence of co-stimulatory molecules at the surface of APCs or target cells. It was thus important to examine the expression level of these molecules in Leishmania-infected Ms to define their role in the

30

J.-C. ANTOINE ET AL. 7

MIIC

6 2 MHC class II molecule

5

2

megasomes flagellar pocket

H-2M Ii chain fragment peptide parasite Ags PPG host cell protease soluble parasite protease membrane-bound parasite protease

1

3 PV

4

fusion lipid raft

MIIC

Figure 5 Possible mechanisms by which Leishmania spp. evade or interfere with the MHC II-restricted Ag presentation pathway in infected Ms. Infection of Ms does not impede the synthesis and expression of MHC class II and H-2M molecules induced by IFN- . These molecules accumulate in MIIC that fuse with nascent or mature PVs. Presence of parasites in these compartments can physically interfere with the intracellular trafficking of class II molecules en route to the cell surface (1). With some Leishmania species, PVs enlarge considerably, which could further alter the MHC II pathway since the different molecules involved in the Ag presentation process are very likely diluted in these huge compartments (1). In PVs harbouring certain Leishmania species (L. amazonensis, L. mexicana), MHC class II molecules and H-2M molecules are often confined to small areas corresponding to the sites of parasite attachment to the PV membrane, which could impede the access of MHC class II molecules to exogenous peptides (2). Internalization of MHC class II molecules, H-2M molecules as well as complexes between class II and self peptides by amastigotes and intermediate stages between promastigotes and amastigotes of L. amazonensis and L. mexicana have been also illustrated. These molecules accumulate in megasomes where, very likely, they are degraded. By removing class II molecules, the chaperone molecules H-2M and peptide–MHC II complexes (possibly complexes between class II and parasite peptides) from PVs, this process could allow parasites to escape Ag presentation (3), just as the very low release of immunogenic molecules and the very weak expression of plasma membrane proteins by the parasites (4), as well as the secretion or the expression at the parasite cell surface of proteases able to alter the Ag processing (5). In fact, the only product abundantly secreted by amastigotes of some Leishmania species (L. mexicana, L. amazonensis) is a poorly immunogenic proteophosphoglycan

LEISHMANIA SPP.

31

activation/reactivation of parasite-specific T cells. Several studies using mouse or dog Ms as host cells have demonstrated that L. donovani or L. infantum promastigotes or amastigotes do not induce the expression of B7 and CD24 molecules (CD24 is also a costimulatory molecule) or even down regulate their spontaneous expression whereas that of adhesion molecules (ICAM-1, VCAM-1) is marginally affected (Kaye et al., 1994; Saha et al., 1995; Pinelli et al., 1999). It is also considered that signalling cascades leading to the up regulation of B7 molecules may be blocked in Leishmaniainfected Ms since these cells are insensitive to stimuli which normally induce the synthesis of such molecules (Kaye et al., 1994). These findings suggest that infected Ms are probably not involved in the activation of naive Leishmania-specific T cells, in line with the fact that L. major-infected Ms are unable to stimulate in vitro primary T cell responses (Konecny et al., 1999) or are endowed with a poor priming potential compared to DCs (Shankar and Titus, 1997). 3.5. In vivo Data Most of the conclusions concerning the Ag presentation capacity of Leishmania-infected Ms were drawn from in vitro experiments and their physiological relevance can thus be questioned. There are however a few in vivo/ex vivo studies that corroborate the in vitro findings. First, Ms present in lymph nodes of L. major-infected C57BL/6 mice, including parasitized Ms are unable to reactivate parasite-specific T cells (Moll et al., 1995). Furthermore, the lack of (PPG) unable to elicit a CD4 T cell response (Aebischer et al., 1999). The last step of MHC class II intracellular trafficking leading the MHC II– peptide complexes from the PVs to the plasma membrane could be slowed down (6). Finally, the distribution of MHC–peptide complexes displayed at the M cell surface could be modified in infected Ms. It has been shown that the concentration of MHC–peptide complexes in lipid rafts of the cell surface allows more efficient presentation of Ags to T cells especially when the number of these complexes is limiting. Presence of parasite molecules like glycolipids could disrupt these membrane microdomains (7). (See Plate 1.5 in colour plate section.)

32

J.-C. ANTOINE ET AL.

CD24 expression and the very weak B7-1 expression have been documented for parasite-containing Kupffer cells (resident Ms of the liver located in the sinusoids) of L. donovani-infected BALB/c mice. Otherwise, these cells express MHC class II molecules. Interestingly, uninfected Kupffer cells show higher levels of costimulatory molecules suggesting that the infected cell phenotype described above is due to the presence of intracellular parasites (Kaye et al., 1994). In apparent conflict with the preceding data, down regulation of T cell-dependent immune responses involving costimulatory molecules and their counter-receptors have also been recently documented in murine L. donovani and L. chagasi infections. Thus, it has been demonstrated that during the chronic phase of these diseases, interactions between B7 molecules (mainly B7-1) and CTLA-4 (an alternative receptor for B7 molecules expressed by activated T lymphocytes) block IFN- and NO production and parasite killing. These effects promote parasite growth and may be linked to the secretion of TGF- by CD4 T lymphocytes following CTLA-4 engagement (Gomes and DosReis, 2001). Otherwise, this cytokine has been described as an inducer of L-arginase I synthesis in Ms, an enzyme involved in the production of polyamines that are required for parasite proliferation (see Section 4.4). However, the cells expressing B7 molecules have not been clearly identified in these studies (infected Ms? Leishmania Ag-loaded Ms?).

4. ABILITY OF INFECTED MS TO DESTROY LEISHMANIA PARASITES THEY HARBOUR 4.1. The Different Pathways Leading to the Development of Leishmanicidal Properties Despite the fact that parasites have developed strategies to down regulate signalling pathways leading to enhancement of M leishmanicidal properties, it is well known that in L. major-resistant mice, which control the parasite multiplication after a short period of parasite amplification, Ms acquire an activated microbicidal

LEISHMANIA SPP.

33

phenotype. How this phenotype is acquired during Leishmania infection is, however, not very well understood. On the one hand, it is generally assumed that activation is a consequence of direct contact between infected cells and Leishmania-specific effector T cells involving TCR ligand–TCR interactions. On the other hand, several studies have shown that presentation of parasite Ags by amastigoteharbouring Ms is strongly affected or does not occur (see Sections 3.2 and 3.3). It is thus possible that infected Ms are activated by a by-stander effect independent of Ag presentation, but this explanation does not seem compatible with all the available experimental findings. In any case, in vitro experiments and the use of transgenic and gene knockout mice have established that IFN- produced by effector CD4 Th1 cells and CD8 T lymphocytes plays a key role in M activation. These experiments have also shown that this cytokine must act in synergy with other molecules released/expressed by the T lymphocytes, or by the Ms themselves for these to acquire parasiticidal properties (Sacks and Noben-Trauth, 2002). Whereas in vitro experiments have clearly demonstrated that tumour necrosis factor (TNF) can play this potentiating role, some in vivo experiments have led some authors to propose another function for this cytokine during L. major infections. They suggested that TNF may be involved more in the control of the inflammatory process that occurs at the site of parasite inoculation than in the parasite killing (Kanaly et al., 1999; Chakour et al., 2003). Otherwise, cognate interactions between CD40 ligand (CD40L/CD154, a member of the TNF family of cytokines and cell surface molecules) expressed by effector T lymphocytes and M-associated CD40 (a member of the TNF receptor family) may contribute, in the presence of IFN- , to the acquisition by Ms of a leishmanicidal state (Kamanaka et al., 1996; Soong et al., 1996). Interestingly, this pathway appears to be altered in L. major-infected Ms from susceptible mice (Awasthi et al., 2003). Finally, recently, another pathway involving Fas (CD95, member of the TNF receptor family) expressed by IFN- -stimulated Ms and the counter-receptor Fas ligand (member of the TNF family) expressed by CD4 Th1 cells and CD8 T lymphocytes has been

34

J.-C. ANTOINE ET AL.

ILIL-10 IL-6 ILIL-4 ILIFNIFN-α/β

ILIL-13 amastigotes

NADPH

NADP + H + + e-

killed amastigotes

H2O2 O2

apoptosis of infected MΦs

TGFTGF-β

O2OH-

parasite molecules involved in the protection towards leishmanicidal mechanisms: LPG, PPG, GIPLs, trypanothione, SOD, catalase

Fas FasL L-arginine + O2 effector T lymphocyte

CD40L

NO

RNI

?

non replicative amastigotes NO synthase 2

NO

CD40

NADPH oxidase activation inhibition

MIF

TNFTNF-α

ILIL-7 IFNIFN-α/β

IFN-γ

Figure 6 Mechanisms of Leishmania killing developed by Ms and how parasites can evade them. Several cytokines produced by Ms themselves (bent arrows in the lower part of the figure) or by other cells such as effector CD4 Th1 and CD8 T lymphocytes can activate or induce several leishmanicidal mechanisms (black arrows) while other cytokines synthesized by Ms (bent arrows in the upper part of the figure) or by other cells such as CD4 Th2 lymphocytes can block the activation or the induction of these same mechanisms (black T). According to recent studies, binding of Massociated CD40 and Fas to CD40L and FasL expressed by effector T lymphocytes can also activate Ms to a leishmanicidal state, which could imply that infected Ms express MHC-parasite peptide at their cell surface. These different pathways lead to the activation and synthesis of NADPH oxidase and NO synthase 2, respectively. Activation of NADPH oxidase, an enzyme involved in electron transport, results from the binding of cytosolic enzyme subunits to membrane-bound enzyme subunits. Once completely assembled, NADPH oxidase catalyses the transfer of electrons from NADPH to molecular oxygen and thus the production of superoxide O2 that is released in the extracellular medium or, after phagocytosis, in the lumen of phagosomes. The acidic pH of this compartment leads to rapid conversion of O2 into H2O2, OH and other toxic reactive oxidants (ROI) that can kill amastigotes. Parasites have evolved different mechanisms to resist or destroy ROI. Molecules involved in these mechanisms are listed in the right inset. As to NO synthase 2, it catalyses the formation of NO from

LEISHMANIA SPP.

35

reported as taking part in the elimination of L. major (Conceic¸aoSilva et al., 1998; Huang et al., 1998). Fas–Fas ligand interaction can induce apoptosis of Fas-expressing cells. So, this pathway could lead infected Ms towards apoptosis and thus deprive amastigotes of their host cells (Conceic¸ao-Silva et al., 1998). Additionally, it has also been shown that Fas–Fas ligand interaction can result in the killing of intracellular parasites, when it occurs in cooperation with IFN- (Chakour et al., 2003). Although here Fas–Fas ligand interaction appears essential for the parasite destruction, it should be pointed out that in another experimental model of L. major infection, Fas is not required for parasite control (Kanaly et al., 1999). The use of different pathways of M activation according to the infecting Leishmania strain or the degree of parasite virulence could account for these apparently conflicting results (Figure 6).

4.2. Mechanisms of Leishmania Killing Stimulation of mouse Ms through IFN- and co-stimulators leads to the induction or the activation of several innate and inducible leishmanicidal mechanisms culminating in the production of ROI by the phagocyte NADPH oxidase phox and the production of RNI by the inducible nitric oxide synthase NOS2. The fact that both mechanisms are truly involved in the control of parasite multiplication was recently demonstrated by two groups using C57BL/6 or C57BL/6  129/Sv mice lacking either gp91 phox (one of the subunit of the NADPH oxidase) or NOS2. After infection of these mice with L. donovani amastigotes, it was shown that both mechanisms act to limit parasite growth in the liver during the first two weeks, but also that NOS2 alone is sufficient for the almost complete clearance of the L-arginine and molecular oxygen. NO and the derivatives known as RNI are toxic for amastigotes. Possibly, NO could also inhibit the multiplication of amastigotes and lead them towards a quiescent stage. However, this last mechanism, which could participate in the persistence of live parasites even in immune animals, has not yet been formally demonstrated. Fas–FasL interactions can also lead to apoptosis of infected Ms.

36

J.-C. ANTOINE ET AL.

liver-associated parasites (Murray and Nathan, 1999). In another study carried out with L. major-infected mice, it was reported that whereas NOS2 plays a major role in the control of parasite number in the skin lesions and draining lymph nodes during the acute phase of the disease, both NOS2 and phox are required in these sites during the chronic phase of the infection to maintain the parasite population at a low level. In contrast, in the spleens of these mice, control of parasite growth requires a functional phox whatever be the stage of the infection (Blos et al., 2003) (Figure 6). Most of the studies reported above have been carried out with L. major or L. donovani as infectious agents. Of course, extrapolation of the results to other species is risky as it is known that Leishmania species display different levels of resistance towards M-mediated microbicidal mechanisms. As a proof, it has even been recently demonstrated that Ms stimulated with IFN- are more permissive for the growth of L. amazonensis amastigotes than unstimulated Ms (Qi et al., 2004).

4.3. How Leishmania Evade the Killing Mechanisms? Leishmania have developed numerous strategies to avoid ROI and RNI or to neutralize/degrade these compounds. First, parasite binding and phagocytosis do not trigger significant production of ROI and do not induce the synthesis of RNI. Furthermore, once inside the Ms, the parasites can at least partially inhibit the host cell activation by IFN- and co-stimulators, which leads to a lower production of O2 , H2O2 and NO (Maue¨l, 1996). Interestingly, inhibition of activation can be induced even if the parasites are killed after phagocytosis. The phosphoglycans expressed at the cell surface are not essential for this effect to be observed, at least for the L. major species (Spa¨th et al., 2003a, b). On the other hand, parasites are endowed with antioxidant molecules such as LPG and trypanothione (unique thiol apparently present only in the trypanosomatids) able to detoxify ROI, or molecules such as catalase and superoxide dismutase able to degrade them (Maue¨l, 1996). Recent generation

LEISHMANIA SPP.

37

of L. major mutants deficient in the synthesis of LPG or of cell surface-associated and secreted phosphoglycans has confirmed that these molecules protect parasites against oxidative stress (Spa¨th et al., 2003a, b). Interestingly, mutants unable to synthesize plasma membrane-associated and secreted phosphoglycans can persist apparently indefinitely but at a very low level in infected BALB/c mice without inducing clinical signs suggesting that despite their inability to survive in Ms, they can find shelter within cells not involved in the production of ROI (Spa¨th et al., 2003b) (Figure 6).

4.4. In Susceptible Mice, Ms Can Follow an Alternative Activation Pathway Leading to Uncontrolled Parasite Growth There is now increasing evidence that in the presence of Th2 type cytokines, especially IL-4, Ms undergo an alternative activation program leading to the synthesis of L-arginase I instead of NOS2. Likewise, the so-called deactivating cytokines IL-10, TGF- and prostaglandin E2 too induce L-arginase I synthesis (for a review see Gordon, 2003). It has been demonstrated that under this cytokinic environment, L. major- and L. infantum-infected Ms also express L-arginase I (Iniesta et al., 2001, 2002; Kropf et al., 2004a, b), which through the production of L-ornithine from L-arginine, participates in parasite expansion. Indeed, L-ornithine is used for the biosynthesis of polyamines that are key compounds for parasite growth (Figure 7).

5. THE DCS AS CELLS THAT CAN ALSO SHELTER LEISHMANIA SPP. 5.1. Binding and Phagocytosis of Promastigotes and Amastigotes Studies of interactions between various Leishmania species and mouse or human DCs have been the object of several investigations

38

J.-C. ANTOINE ET AL.

IL-10 IL-13 IL-4

L-arginine

L-ornithine + urea

polyamine biosynthesis essential for parasite growth amastigotes

TGF-β β

L-arginase I activation

PGE2

Figure 7 Alternative activation of Leishmania-infected Ms. In the presence of IL-4/IL-13 (for example, in mice developing a Th2 type immune response towards Leishmania), or in the presence of the deactivating cytokines IL-10 or TGF- (for example, in mice affected by a visceral leishmaniasis), infected Ms could be engaged in the synthesis of L-arginase I that degrades L-arginine into L-ornithine. The amino acid L-ornithine is an essential compound for biosynthesis of the polyamines spermidine and putrescine. Polyamines can be produced by the Ms and then conveyed into parasites through transporters or by amastigotes themselves since they also have the enzymes required for this metabolic pathway (for a review see Burchmore and Barrett, 2001; Mu¨ller et al., 2001). As polyamines are key components for parasite proliferation, Ms activated in that manner are highly permissive for parasite expansion.

LEISHMANIA SPP.

39

in the recent years (Moll et al., 1997; Moll, 2000; Udey et al., 2001; Sacks and Sher, 2002). There is no doubt from in vitro experiments that DCs can phagocytose Leishmania but whether both parasite stages can be efficiently phagocytosed by DCs is still a debated question. Thus, some authors claim that only amastigotes can be phagocytosed by DCs or that they are much more efficiently captured by DCs than promastigotes (Blank et al., 1993; von Stebut et al., 1998, 2000) whereas others clearly demonstrate that promastigotes too can be internalized by these cells (Konecny et al., 1999; Marovich et al., 2000; Bennett et al., 2001; Qi et al., 2001; Amprey et al., 2004). Likewise, results bearing on the parasite survival in DCs show great variability. Some authors report that L. major metacyclics are unable to survive within mouse spleen DCs (Konecny et al., 1999) whereas others conclude that both L. amazonensis metacyclics and amastigotes can establish within mouse bone marrow-derived DCs (Qi et al., 2001). Whether the apparent discrepancies noted between these studies are due to the type of DCs used [monocyte-derived DCs, bone marrow-derived DCs, spleen DCs, Langerhans’ cells (DCs found in the epidermis), foetal skinderived DCs], to the Leishmania species (L. major, L. tropica, L. donovani, L. mexicana, L. pifanoi, L. amazonensis) or the origin of the parasites (culture-derived promastigotes, axenic amastigotes, lesionderived amastigotes) is unclear and deserves further investigation. In a recent study from our laboratory, we have shown that both metacyclic promastigotes and amastigotes of L. amazonensis can be captured by mouse bone marrow-derived DCs and that the parasite uptake can be increased by some opsonins such as Leishmaniaspecific IgGs, but not by others such as the complement C3 component. This study has also demonstrated that metacyclics can differentiate into amastigotes, and that parasites can survive and grow in DCs (Prina et al., 2004). So far, on the host side, only a role for the integrin CR3 (CD11b/ CD18) has been demonstrated for the uptake of L. major amastigotes by mouse Langerhans’ cells (Blank et al., 1993). The C-type lectin DC-specific ICAM-3-grabbing nonintegrin (DC-SIGN, CD209) has been shown to be involved in the binding of promastigotes and axenic

40

J.-C. ANTOINE ET AL.

amastigotes of L. pifanoı¨ and L. infantum to human monocytederived DCs (Colmenares et al., 2002, 2004), and the Fc receptor(s) have been implicated in the uptake of antibody-coated L. amazonensis metacyclics and amastigotes by mouse bone marrowderived DCs (Prina et al., 2004). On the parasite side, except for the surface-bound opsonins (C3, IgGs), little is known about the molecules which can be recognized by DC receptors. If the binding of purified L. mexicana LPG to DC-SIGN has been demonstrated (Appelmelk et al., 2003), the attachment of L. pifanoı¨, and L. infantum promastigotes and axenic amastigotes to human monocytederived DCs appear, however, to be independent of LPG expression (Colmenares et al., 2004).

5.2. Phagosomal Compartments Housing Parasites in DCs Once internalized by DCs, Leishmania settle in organelles sharing at least some properties with PVs of infected Ms stimulated with IFN- (Flohe´ et al., 1997; Henri et al., 2002; Bennett et al., 2003; Prina et al., 2004). Hereafter, these compartments will also be called PVs. The most complete study of these organelles, which has been carried out on DCs infected initially with L. amazonensis promastigotes or amastigotes (Prina et al., 2004), indicates that they are acidic and bounded by a membrane containing late endosomal/lysosomal glycoproteins namely lamp-1, lamp-2, macrosialin, the MOMA-2 Ag as well as MHC class II and H-2M molecules (Figure 8; see colour plate section). Furthermore, 40 to 70% of these compartments are coated with the small GTP-binding protein rab7p. However, 24 to 48 h post-phagocytosis, PVs located in DCs differ from PVs found in Ms: they are smaller, the level of associated rab7p is lower, and cathepsins are very hard to detect in the lumen of these PVs (Table 1). Clearly, a more extensive characterization of these organelles is necessary to determine whether PVs in DCs are really different from those in infected Ms, or if the differences noted could be explained by slower kinetics of biogenesis. Transient or intrinsic differences between the PVs present in Ms and DCs could have, of course, important

LEISHMANIA SPP.

41

Figure 8 Distribution of MHC class II and H-2M molecules in a mouse bone marrow-derived DC infected with L. amazonensis amastigotes purified from a nude mouse. After fixation, cells were processed for the detection of MHC class II (left micrograph, green staining in the colour plate) and H-2M (right micrograph, red staining in the colour plate) molecules by fluorescence confocal microscopy. Molecules were visualized using as primary antibodies the monoclonal antibody 2G9 and purified rabbit antibodies directed against the H-2M 1 cytosolic tail, respectively, and appropriate fluorochrome conjugates. The analysis of a single cell section is presented. MHC class II and H-2M molecules are co-localized in numerous endocytic compartments including a PV (arrows). Arrowheads point to the weak class II staining of the plasma membrane. n, DC nucleus. Bar, 5 mm. (See Plate 1.8 in colour plate section.)

consequences for the respective capacity of these cells to present Leishmania Ags. 6. ROLE OF DCS IN THE PRESENTATION OF LEISHMANIA ANTIGENS TO NAIVE AND ACTIVATED SPECIFIC T LYMPHOCYTES 6.1. Ability of Leishmania spp. to Induce DC Maturation and Migration It has been demonstrated in the pioneering work of Moll and collaborators that, after inoculation of mice with a large number of L. major promastigotes, skin DCs rapidly take up Leishmania and/or

42

J.-C. ANTOINE ET AL.

Leishmania Ags and migrate to the draining lymph nodes (Moll et al., 1997; Moll, 2000). It is suspected that at least some of these migrating DCs undergo maturation that makes them capable of activating naive Leishmania-specific CD4 and CD8 T lymphocytes. Several in vitro studies have been devoted to the ability of various Leishmania species to induce DC maturation but, once again, apparently conflicting results have been obtained from different laboratories. Whether this variability is due to variations in the experimental conditions or truly reflects complex species-specific or strain-specific interactions between Leishmania and DCs is not known, but the fact that Leishmania spp. express highly polymorphic cell surface molecules leads us to think that the cross-talk between DCs and these parasites could vary greatly according to the Leishmania species examined. Most studies indicate that L. major (amastigotes and/or promastigotes) activates mouse and human DCs, leading to the up regulation of MHC class II and co-stimulatory molecules, and to the secretion of low amounts of IL-12 p40 (under some experimental conditions, very low amounts of IL-12p70 are also released by infected DCs) (Flohe´ et al., 1998; von Stebut et al., 1998, 2000; Konecny et al., 1999; Henri et al., 2002; McDowell et al., 2002). However, other Leishmania species (L. tropica, L. donovani, L. mexicana) are unable to up regulate class II and co-stimulatory molecules in DCs and/or to induce the production of IL-12p40 and IL-12p35 (Bennett et al., 2001; McDowell et al., 2002). An important role for some parasiteassociated opsonins in the maturation of mouse DCs infected with L. amazonensis was recently highlighted (Prina et al., 2004). Thus, after internalization of unopsonized promastigotes, C3-coated promastigotes or amastigotes derived from nude mice, the majority of infected DCs remain phenotypically immature (Figure 9; see colour plate section). In contrast, internalization of antibody-opsonized promastigotes or amastigotes induces rapid DC maturation (Figure 9; see colour plate section). It is tempting to propose that in vivo, in the absence of specific antibodies (e.g. shortly after infecting naive mammals), infected DCs may remain immature or semi-mature, meaning that they are unable to elicit an efficient anti-Leishmania T cell response. This, of course, could represent a strategy evolved by some

LEISHMANIA SPP.

43

Leishmania species to allow their establishment and amplification in mammals before the onset of immune responses. Furthermore, recent data suggest that by interfering with the motility/the migratory properties of DCs, Leishmania could prevent, at least transiently, the establishment of T cell-mediated immunity. Thus, L. major promastigotes or molecules they secrete/release like the LPG can partially inhibit the motility of mouse spleen DCs and the migration of Langerhans’ cells from mouse skin explants (Ponte-Sucre et al., 2001; Jebbari et al., 2002). Impaired migration of spleen DCs purified from L. donovani-infected mice in response to chemokines has also been recently documented. This defect is apparently linked to the down regulation of CCR7 and could explain why, in spleens of mice with chronic L. donovani infection, DCs fail to migrate from the marginal zone to the white pulp (Ato et al., 2002). In this case, however, CCR7 down regulation and the subsequent impaired migration of DCs is apparently not due to the direct action of parasite molecules but due to the effect of cytokines abundantly produced during L. donovani infection, namely TNF- and IL-10 (Figure 9; see colour plate section). In apparent contrast with the results of the previous study, those of De Trez et al. (2004) show that intravenous inoculation of promastigotes of several Leishmania species including L. donovani, L. major, L. mexicana and L. braziliensis induces both splenic DC migration from marginal zones to T cell areas and splenic DC maturation. However, the two studies examine the pattern of splenic DC distribution at very different times after parasite injection, namely 9 to 48 h (De Trez et al., 2004) vs. 14 to 42 days (Ato et al., 2002), which prevents doing strict comparisons of the results. In any case, the fact that in the study of De Trez et al. (2004), DC migration and maturation are observed only after injecting a huge number of promastigotes (108 to 5  108) makes the results questionable. 6.2. MHC I and MHC II Ag Presentation by DCs Put in Contact with Parasites or Leishmania Ags Accurate studies of the sensu stricto Ag presentation processes occurring in Leishmania-infected DCs are still scarce. According to

44

J.-C. ANTOINE ET AL.

A or

class II molecules H-2M molecules class II + H-2M molecules CD40, CD54, CD80, CD86, OX40L antibody C3

B epidermis

white pulp of a spleen infected with L. donovani

skin dermis

DCs L. major promastigotes

marginal zone B cell area

parasite molecules lymph node

T cell area

T cell area B cell area

DC

DC TNF-α, IL-10

Figure 9 Some mechanisms employed by different Leishmania species to avoid or limit Ag presentation by DCs. (A) Lack of maturation or weak maturation of DCs after internalization of some Leishmania species. The left part of the figure shows that immature DCs express low amounts of MHC class II, co-stimulatory and adhesion molecules on their cell surface.

LEISHMANIA SPP.

45

Bennett et al. (2003), mouse bone marrow-derived DCs initially infected with L. mexicana stationary phase promastigotes or axenic amastigotes are as efficient as uninfected DCs to: (i) process an exogenous Ag, in this case pigeon cytochrome C, (ii) generate complexes between class II molecules and a cytochrome C-derived peptide and express them at the plasma membrane, and (iii) present these complexes to a specific CD4 T cell hybridoma. These data are thus in striking contrast to those obtained when infected Ms are used as APCs (see Section 3.3), but the origin of the functional differences exhibited by the two types of infected APCs has not yet been identified. As to the ability of infected DCs to process and present parasite Ags to Leishmania-specific T lymphocytes, it remains to be carefully examined. In any case, several in vitro experimental protocols have demonstrated that mouse DCs incubated with either Numerous endocytic compartments containing both MHC class II and H-2M molecules are scattered in their cytoplasm. After phagocytosis of unopsonized or C3-opsonized metacyclic promastigotes of L. amazonensis (upper arrow) by DCs, most of them (about 70% at 72 h post-phagocytosis) keep an immature phenotype. Parasites are detected in MHC class II þ - and H-2M þ -PVs. Presence of these molecules within parasites can also be observed (intraparasite yellow spot in the colour plate). In contrast, if promastigotes are opsonized with antibodies (lower arrow), infected and also a part of uninfected DCs rapidly mature. This appears by an increase of the MHC class II, co-stimulatory and adhesion molecule expression at the cell surface. Intracellular MHC class II molecules including those located in PVs are much less abundant or even undetectable. H-2M molecules are however still present in the PV membrane as well as in some endocytic compartments clustered in the cell center. (B) Parasites can directly or indirectly affect migration of DCs towards T cell areas of secondary lymphoid organs. The left schema shows that molecules released by L. major promastigotes including LPG can reduce the efflux of DCs from the skin and thus their access to T cell areas in draining lymph nodes, perhaps by modulating the expression of traffic molecules involved in their homing characteristics. The right schema shows that in the spleen of mice chronically infected with L. donovani, DCs fail to migrate from the marginal zones to the T cell areas known as periarteriolar lymphoid sheaths. In this case, impairment of DC migration is linked to the down regulation of CCR7, a receptor for the lymphoid chemokines CCL19 and CCL21, induced by IL-10 and TNF- abundantly produced during this infection. (See Plate 1.9 in colour plate section.)

46

J.-C. ANTOINE ET AL.

L. major Ags/peptides or live promastigotes or amastigotes are able to induce primary T cell responses (Shankar and Titus, 1997; Konecny et al., 1999; Filippi et al., 2003). They are also able to reactivate several T cell subsets including a Leishmania-specific CD4 T cell clone, T cells purified from mice immunized with parasite Ags (Moll et al., 1997; Moll, 2000), as well as IFN- -producing CD4 and CD8 T lymphocytes and IL-10-producing CD4 CD25 regulatory T lymphocytes purified from lymph nodes and/or lesions of C57BL/6 mice initially infected with a low number of L. major metacyclics (Belkaid et al., 2002a, 2002b). Similarly, human DCs derived from monocytes of Leishmania-infected patients and incubated with L. major promastigotes can reactivate autologous T cells (Marovich et al., 2000). Interestingly, whereas primed, Leishmania-specific CD4 T cells can be as efficiently reactivated by mouse DCs previously incubated with either parasite extracts or live parasites, DCs incubated with live parasites are stronger stimulators for primed, Leishmania-specific CD8 T cells (Belkaid et al., 2002b) suggesting that formation of phagosomes or of the recently described endoplasmic reticulum–phagosome mix compartments favours the presentation of Leishmania Ags via the MHC class I molecules (Desjardins, 2003; Guermonprez et al., 2003; Houde et al., 2003). Otherwise, DCs purified from lymph nodes of L. major-infected, susceptible or resistant mice as early as 16 h to 6 days after inoculation of a large number of promastigotes can present parasite Ags including the Ag LACK to CD4 T cell clones or hybridomas (Moll et al., 1997; Moll, 2000; Filippi et al., 2003; Misslitz et al., 2004; Ritter et al., 2004) and can activate naive LACK-specific CD4 T lymphocytes purified from transgenic mice clearly indicating that DCs can prime Leishmania-specific CD4 T cells (Filippi et al., 2003). DCs seem also to play a key role all along the infection process as shown by the fact that DCs harbouring intact Leishmania or containing parasite Ags can still be detected in lymph nodes of C57BL/6 mice infected with L. major but clinically cured. These cells are able to reactivate Leishmania-specific T cells (Moll et al., 1995). It is thus suspected that Leishmania-containing DCs are involved in the maintenance of T cell memory/activity.

LEISHMANIA SPP.

47

While it has been documented that in L. major-infected mice, some DCs at the sites of parasite inoculation and in the draining lymph nodes contain Leishmania Ags (Moll et al., 1997; Moll, 2000; Filippi et al., 2003; Ritter et al., 2004; Misslitz et al., 2004), there is little in vivo data showing the presence of live parasites in DCs (Muraille et al., 2003b; Baldwin et al., 2004; Misslitz et al., 2004). The respective roles of Leishmania Ag-loaded DCs and infected DCs in the induction and shaping of T cell immune responses are thus unclear. To approach this issue, the kinetics of recruitment of infected DCs in lesions and lymph nodes is an important parameter to determine. Two recent studies have assessed this parameter in L. major-infected BALB/c and C57BL/6 mice that had received either a huge number of stationary phase promastigotes (2  107) into the footpads (Misslitz et al., 2004) or a low number of metacyclic promastigotes (103) into the ear dermis (Baldwin et al., 2004). Data drawn from the first study show that DCs containing parasite Ags but not live parasites appear first in the lymph nodes from 4 h post-inoculation, peak at 16 to 24 h and then rapidly disappear. These cells, which are much more numerous in susceptible BALB/c mice than in resistant C57BL/6 mice, would be involved in the initiation of the adaptive immune response directed against Leishmania. Infected DCs appear at later times, from 24 h post-inoculation. Unfortunately, the role of the latter in the shaping of the immune response has not been examined in this study (Misslitz et al., 2004). In the second study mimicking the natural conditions of parasite inoculation, infected DCs including ‘‘lymphoid’’ DCs, ‘‘myeloid’’ DCs, Langerhans cells and plasmacytoid DCs appear in the draining lymph nodes but not before 3 weeks post-inoculation. Such cells are present in lymph nodes of BALB/c mice all over the time period under study (12 weeks). In contrast, the presence of infected DCs in the lymph nodes of C57BL/6 mice is very transient (Baldwin et al., 2004). Altogether, these data point to a role of Leishmania Ag-containing DCs in the initiation of the adaptive immune response, especially in ‘‘classical’’ murine leishmaniasis. The contribution of infected DCs in this process is less clear. Nevertheless, an important point must be emphasized, namely that their kinetics of appearance in the

48

J.-C. ANTOINE ET AL.

lymph nodes is strongly dependent upon the size of the parasite inoculum.

6.3. Possible Mechanisms used by Leishmania spp. to Limit Ag Presentation by DCs The ability of Leishmania-infected DCs to present parasite Ags in the context of MHC class I or class II molecules as well as the general impact of Leishmania infection on the Ag presentation capacity of these cells thus remain to be carefully assessed using well-defined in vitro systems allowing quantitative measurements. In this respect, it is interesting to note that, as in PVs of IFN- -stimulated Ms, Leishmania in DCs are in direct contact with MHC class II and H-2M molecules (Flohe´ et al., 1997; Henri et al., 2002; Prina et al., 2004), but whether PVs of these cells are involved in the loading of class II molecules with parasite peptides remains to be determined. On the other hand, several in vitro and in vivo findings suggest that the Ag presentation and activation capacities of DCs could be impaired by the presence of intracellular parasites. For example, in DCs infected with L. amazonensis or L. mexicana as in Ms infected with these Leishmania species (see Section 3.1), parasites internalize MHC class II and H-2M molecules, which could be a means to limit the expression of parasite peptide-class II complexes on the DC cell surface (Figure 9; see colour plate section) (Bennett et al., 2003; Prina et al., 2004); this process appears however to be of lower amplitude in DCs than in Ms (Bennett et al., 2003; J.-C. Antoine and E. Prina, unpublished results). Moreover, parasite-harbouring DCs present in lymph nodes of both BALB/c (susceptible) and C57BL/6 (resistant) mice infected with L. major express only low levels of MHC class II molecules on their cell surface and no B7-2 (Muraille et al., 2003b), which could be due to a parasite-induced down regulation of these molecules. Alternatively, these last findings could be explained by the lack of maturation of Leishmania-infected DCs present in lymph nodes. Finally, a very recent study indicates that human monocytederived DCs infected with L. donovani promastigotes display an

LEISHMANIA SPP.

49

inhibition of CD1a, b and c expression (at both mRNA and protein levels) (Amprey et al., 2004). Recall that these molecules have been shown to present lipids and glycolipids of microorganisms to specific T cells. Inhibition of CD1 expression by Leishmania-infected DCs could thus represent a parasite strategy to limit Ag presentation capacity of these cells. In these same cells, no down regulation of MHC class I and class II expression could be noted (Amprey et al., 2004).

6.4. DCs and the Polarization of Leishmania-Specific CD4 T Lymphocytes It is now clear that DCs play an important role in the polarization of the CD4 T cell responses against various pathogens including parasites, but unfortunately very few studies have been performed in this field using Leishmania (Sher et al., 2003). Studies of L. major infections in knockout mice lacking MyD88 (an adaptor protein expressed by different cells including DCs, and involved notably in the signalling cascades activated after the recognition of microbial molecules by Toll-like receptors) have indicated that MyD88 is an essential component for the production of IL-12 (very likely by DCs) and the subsequent appearance of Leishmania-specific CD4 Th1 effector cells. In the absence of MyD88, genetically resistant mice become susceptible and develop a CD4 Th2 type immune response (Debus et al., 2003; de Veer et al., 2003; Muraille et al., 2003a). These data suggested that L. major Ags can be recognized by Toll-like receptors and LPG has been shown to be a possible ligand (Becker et al., 2003; de Veer et al., 2003). Of course, many other explanations can be put forward because MyD88 is involved in the expression of many genes and in the basal activity of the transcription factor NFB (Shi et al., 2003). Whether, in the absence of MyD88, DCs do not mature when put in contact with L. major, or whether they follow an activation pathway leading to a Th2 type immune response is not yet clear. In this respect, it is interesting to note that bone marrowderived DCs prepared from mice exhibiting either a mixed Th1/Th2 type response (BALB/c mice) or a Th1 type response (C3H/HeJ mice)

50

J.-C. ANTOINE ET AL.

after infection with L. amazonensis mature normally when they are incubated with amastigotes of this Leishmania species (very likely covered with antibodies) as shown by the up regulation of cell-surface-associated MHC class II and co-stimulatory molecules but only the former produce IL-4 in response to CD40 engagement [a member of the TNF receptor family that binds CD154 (CD40 ligand), a member of the TNF family expressed by activated T cells] (Qi et al., 2001). These studies may indicate that in infected mice developing a Th2 type immune response, DCs follow an alternative activation pathway after parasite contact leading to the production of cytokines like IL-4 or IL-10. Interestingly, the propensity of DCs purified from various mouse strains to induce the differentiation of CD4 T cells into Th1 or Th2 effector cells after in vitro or in vivo loading with Leishmania Ags appears to be linked at least in part to some intrinsic properties (Filippi et al., 2003). Whether this is due to differential capacities to secrete polarizing cytokines (IL-12, IL-4, IL1 , IL-1) (Qi et al., 2001; Filippi et al., 2003; von Stebut et al., 2003) or to express cell surface molecules like co-stimulatory molecules [CD80, OX40L (a member of the TNF family that binds OX40 (CD134), a member of the TNF receptor family expressed by activated T cells)] (Akiba et al., 2000; Filippi et al., 2003) or PRRs (Toll-like receptors) (Liu et al., 2002) remains to be established.

7. THE POTENTIAL OF INFECTED APCS OR APCS LOADED WITH LEISHMANIA ANTIGENS AS VACCINES OR THERAPEUTIC AGENTS 7.1. APCs as Vaccines The potential use of DCs or Ms loaded with Leishmania Ags or infected as vaccines against L. donovani or L. major has been recently tested in BALB/c mice. In the visceral leishmaniasis model, intravenous inoculation of bone marrow-derived DCs pulsed with a promastigote lysate can partially protect mice from infection with promastigotes, as shown by the significant decrease in the parasite