Giardia and Cryptosporidium from Molecules to Disease

GIARDIA AND CRYPTOSPORIDIUM From Molecules to Disease

Acknowledgement We are most grateful to Arturo Pérez-Taylor for his excellent informatics assistance during the preparation of this book.

GIARDIA AND CRYPTOSPORIDIUM From Molecules to Disease

Edited by

Guadalupe Ortega-Pierres Department of Genetics and Molecular Biology, CINVESTAV-IPN, México City, Mexico

Simone M. Cacciò Department of Infectious, Parasitic and Immunomediated Diseases, Istituto Superiore di Sanità, Rome, Italy

Ronald Fayer Environmental Microbial Safety Laboratory, Animal and Natural Resources Institute, ARS-USDA, Beltsville, MD, USA

Theo G. Mank Department of Parasitology, Public Health Laboratory, Haarlem, the Netherlands

Huw V. Smith Scottish Parasite Diagnostic Laboratory, Stobhill Hospital, Glasgow, UK and

R.C. Andrew Thompson WHO Collaborating Centre for the Molecular Epidemiology of Parasitic Infections, School of Veterinary and Biomedical Sciences, Murdoch University, WA, Australia

CABI is a trading name of CAB International CABI Head Office Nosworthy Way Wallingford Oxfordshire OX10 8DE UK Tel: +44 (0)1491 832111 Fax: +44 (0)1491 833508 E-mail: [email protected] Website: www.cabi.org

CABI North American Office 875 Massachusetts Avenue 7th Floor Cambridge, MA 02139 USA Tel: +1 617 395 4056 Fax: +1 617 354 6875 E-mail: [email protected]

© CAB International 2009. All rights reserved. No part of this publication may be reproduced in any form or by any means, electronically, mechanically, by photocopying, recording or otherwise, without the prior permission of the copyright owners. A catalogue record for this book is available from the British Library, London, UK. Library of Congress Cataloging-in-Publication Data Giardia and cryptosporidium : from molecules to disease / edited by Guadalupe Ortega-Pierres . . . [et al.]. p. cm. Includes bibliographical references and index. ISBN 978-1-84593-391-3 (alk. paper) 1. Giardia. 2. Cryptosporidium. 3. Giardiasis–Prevention. I. Ortega-Pierres, Guadalupe. II. C.A.B. International. [DNLM: 1. Giardia. 2. Cryptosporidiosis–prevention & control. 3. Cryptosporidium. 4. Giardiasis–prevention & control. 5. Water Pollution–prevention & control. QX 70 G4346 2008] QL368.D65G53 2008 616.9′3601–dc22 2008025358 ISBN-13: 978 1 84593 391 3 Typeset by AMA Dataset, Preston, UK. Printed and bound in the UK by the MPG Books Group, Bodmin. The paper used for the text pages in this book is FSC certified. The FSC (Forest Stewardship Council) is an international network to promote responsible management of the world's forests.

Contents

Contributors 1

The Impact of Giardia on Science and Society R.C. Andrew Thompson

2

Cryptosporidium in Cattle: From Observing to Understanding Ronald Fayer, Monica Santín and James M. Trout

ix 1

12

TAXONOMY, NOMENCLATURE AND EVOLUTION 3

Names Do Matter Dwight D. Bowman

4

Centenary of the Genus Cryptosporidium: From Morphological to Molecular Species Identification Jan Šlapeta

25

31

MOLECULAR EPIDEMIOLOGY AND TYPING 5

6

7

Molecular Epidemiology of Human Cryptosporidiosis in Developing Countries Lihua Xiao

51

Molecular Epidemiology and Typing of Non-human Isolates of Cryptosporidium Una M. Ryan and Lihua Xiao

65

Insights Into the Molecular Detection of Giardia duodenalis: Implications for Epidemiology Simone M. Cacciò, Marco Lalle, Relja Beck and Edoardo Pozio

81

v

vi

Contents

ZOONOTIC, HUMAN AND ANIMAL HEALTH ISSUES 8

Wildlife with Giardia: Villain, or Victim and Vector? Susan J. Kutz, R.C. Andrew Thompson and Lydden Polley

9

The Role of Livestock in the Foodborne Transmission of Giardia duodenalis and Cryptosporidium spp. to Humans Brent R. Dixon

10 The Risk of Zoonotic Genotypes of Cryptosporidium spp. in Watersheds Hussni O. Mohammed and Susan E. Wade

94

107

123

CLINICAL AND MOLECULAR EPIDEMIOLOGY 11 Clinical Presentation in Cryptosporidium-infected Patients Laetitia M. Kortbeek 12 Molecular Epidemiology of Cryptosporidium and Giardia Infections Paul R. Hunter

131

138

ADVANCES IN DIAGNOSIS 13 Advances in Diagnosis: is Microscopy Still the Benchmark? 147 Rachel M. Chalmers TREATMENT OF DRINKING WATER 14 Control of Cryptosporidium and Giardia in Surface Water by Disinfection Thomas M. Hargy, Jennifer L. Clancy and Lynn P. Landry

158

15 Towards Methods for Detecting UV-induced Damage in Individual Cryptosporidium parvum and Cryptosporidium hominis Oocysts by Immunofluorescence Microscopy Huw V. Smith, B.H. Al-Adhami, Rosely A.B. Nichols, John R. Kusel and J. O’Grady

179

CONTROL IN WATER 16 Effect of Environmental and Conventional Water Treatment Processes on Waterborne Cryptosporidium Oocysts 198 Brendon King, Alexandra Keegan, Chris Saint and Paul Monis 17 Methods for Genotyping and Subgenotyping Cryptosporidium spp. Oocysts Isolated During Water and Food Monitoring Huw V. Smith, Rosely A.B. Nichols, Lisa Connelly and Christopher B. Sullivan

210

Contents

vii

18 Intervention in Waterborne Disease 227 Gordon Nichols, Iain R. Lake, Rachel M. Chalmers, Graham Bentham, Florence C.D. Harrison, Paul R. Hunter, Sari Kovats, Chris Grundy, Steve Anthony, Hester Lyons, Maureen Agnew and Chris Proctor OTHER WATERBORNE PROTOZOA 19 Occurrence and Control of Naegleria fowleri in Drinking Water Wells Charles P. Gerba, Barbara L. Blair, Payal Sarkar, Kelly R. Bright, R.C. Maclean and Francine Marciano-Cabral 20 Environmental Factors Influencing the Survival of Cyclospora cayetanensis Ynes R. Ortega

238

248

BASIC BIOLOGY 21 Recent Advances in the Developmental Biology and Life Cycle of Cryptosporidium 255 Nawal Samih Hijjawi, Annika C. Boxell and R.C. Andrew Thompson 22 Basic Biology of Giardia lamblia: Further Studies on Median Body and Funis 266 Marlene Benchimol 23 Giardia intestinalis: a Microaerophilic Parasite with Mitochondrial Ancestry Gloria León-Avila, José Manuel Hernández and Jorge Tovar

284

METABOLOMICS AND TRANSCRIPTOME 24 Cytoskeleton-based Lipid Transport in a Parasitic Protozoan, Giardia lamblia Cynthia Castillo, Yunuen Hernandez, Sukla Roychowdhury and Siddhartha Das 25 Signalling During Giardia Differentiation Tineke Lauwaet and Frances D. Gillin

292

309

GENOMICS 26 Preliminary Analysis of the Cryptosporidium muris Genome 320 Giovanni Widmer, Eric London, Linghui Zhang, Guangtao Ge, Saul Tzipori, Jane M. Carlton and Joana C. da Silva PROTEOMICS 27 Proteomic Analyses in Giardia Daniel Palm and Staffan G. Svärd

328

viii

Contents

28 Proteomic and Genomic Approaches to Understanding the ‘Power Plant’ of Cryptosporidium Lorenza Putignani, Sanya J. Sanderson, Cristina Russo, Jessica Kissinger, Donato Menichella and Jonathan M. Wastling

344

BIOCHEMISTRY AND PHYSIOLOGY 29 Energy Metabolism and Carbon Flow in Cryptosporidium parvum Guan Zhu

360

30 The Surface Protein Repertoires of Cryptosporidium spp. and Other Apicomplexans Thomas J. Templeton

369

31 Giardan: Structure, Synthesis, Regulation and Inhibition Keriman S ¸ener, Harry van Keulen and Edward L. Jarroll

382

CELL BIOLOGY AND SIGNALLING 32 Protein Kinase C in Giardia duodenalis: a Family Affair M. Luisa Bazán-Tejeda, Raúl Argüello-García, Rosa M. BermúdezCruz, Martha Robles-Flores and Guadalupe Ortega-Pierres 33 Secretory Granule Biogenesis and the Organization of Membrane Compartments via SNARE Proteins in Giardia lamblia Eliana V. Elías, Natalia Gottig, Rodrigo Quiroga and Hugo D. Luján 34 Molecular Mechanisms of Cryptosporidium-induced Host Actin Cytoskeleton Dynamics Steven P. O’Hara, Xian-Ming Chen and Nicholas F. LaRusso

398

409

418

PATHOGENESIS AND HOST–PARASITE RELATIONSHIP 35 Pathogenic Mechanisms in Giardiasis and Cryptosporidiosis Andre G. Buret

428

36 Interferon-gamma (IFN-γ) in Immunological Control of Cryptosporidial Infection Naheed Choudhry, Mona Bajaj-Elliott and Vincent McDonald

442

37 Immune Response to Giardia Infection: Lessons from Animal Models Steven M. Singer and Joel Kamda

451

DRUG TREATMENT AND NOVEL DRUG TARGETS 38 Drug Treatment and Novel Drug Targets Against Giardia and Cryptosporidium Jean-François Rossignol

463

Index

483

Contributors

Maureen Agnew, School of Environmental Sciences, University of East Anglia, Norwich, UK. B.H. Al-Adhami, CFIA, Centre for Animal Parasitology, Saskatoon Laboratory, 116 Veterinary Road, Saskatoon, Saskatchewan, Canada S7N 2R3. Steve Anthony, ADAS Consulting, Woodthorne, Wergs Road, Wolverhampton WV6 8TQ, UK. Raúl Argüello-García, Department of Genetics and Molecular Biology, Centro de Investigación y de Estudios Avanzados del IPN, México City, Av. Instituto Politécnico Nacional No. 2508, Col. San Pedro Zacatenco, México, D.F. 07360, Mexico. Mona Bajaj-Elliott, UCL Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK. M. Luisa Bazán-Tejeda, Department of Genetics and Molecular Biology, Centro de Investigación y de Estudios Avanzados del IPN, México City, Av. Instituto Politécnico Nacional No. 2508, Col. San Pedro Zacatenco, México, D.F. 07360, Mexico. Relja Beck, Department for Parasitology and Parasitic Diseases, University of Zagreb, Heinzelova 55, 10000, Zagreb, Croatia. Marlene Benchimol, Universidade Santa Úrsula, Laboratório de Ultraestrutura Celular, Rua Jornalista Orlando Dantas 59, Botafogo, Rio de Janeiro, R.J., Brazil, CEP 222-31-010. Email: [email protected] Graham Bentham, School of Environmental Sciences, University of East Anglia, Norwich, UK. Rosa M. Bermúdez-Cruz, Department of Genetics and Molecular Biology, Centro de Investigación y de Estudios Avanzados del IPN, México City, Av. Instituto Politécnico Nacional No. 2508, Col. San Pedro Zacatenco, México, D.F. 07360, Mexico. Barbara L. Blair, Department of Soil, Water and Environmental Science, The University of Arizona, Tucson, AZ 85721, USA. ix

x

Contributors

Dwight D. Bowman, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, Ithaca, NY 14850, USA. Email: [email protected] Annika C. Boxell, Division of Veterinary and Biomedical Sciences, Murdoch University, Murdoch, WA 6150, Australia. Kelly R. Bright, Department of Soil, Water and Environmental Science, The University of Arizona, Tucson, AZ 85721, USA. Andre G. Buret, Department of Biological Sciences, Inflammation Research Network, University of Calgary, 2500 University Drive N.W., Calgary, Alberta, Canada T2N 1N4. Email: [email protected] Simone M. Cacciò, Department of Infectious, Parasitic and Immunomediated Diseases, Istituto Superiore di Sanità, Viale Regina Elena 299, Rome 00161, Italy. Email: [email protected] Jane M. Carlton, New York University, Department of Medical Parasitology, 550 First Avenue, New York 10016, USA. Cynthia Castillo, Naturopathic Medicine, Bastyr University, 14500 Juanita Drive NE, Kenmore, WA 98028, USA. Rachel M. Chalmers, UK Cryptosporidium Reference Unit (CRU), National Public Health Service for Wales, Microbiology Swansea, Singleton Hospital, Sketty, Swansea SA2 8QA, UK. Email: [email protected] Xian-Ming Chen, Department of Medical Microbiology and Immunology, Creighton University School of Medicine, 2500 California Plaza, Omaha, NE 68178, USA. Naheed Choudhry, Barts and the London School of Medicine and Dentistry, Centre for Gastroenterology, Queen Mary College, University of London, Turner Street, London E1 2AD, UK. Jennifer L. Clancy, Clancy Environmental Consultants, Inc., 20 Mapleville Depot, PO Box 314, Saint Albans, VT 05478, USA. Lisa Connelly, Scottish Parasite Diagnostic Laboratory (SPDL), Stobhill Hospital, 133 Balornock Road, Glasgow G21 3UW, UK. Siddhartha Das, Infectious Diseases and Immunology Unit, The Border Biomedical Research Center, Department of Biological Sciences, University of Texas at El Paso, 500 W. University Avenue, El Paso, TX 79968, USA. Email: [email protected] Brent R. Dixon, Microbiology Research Division, Banting Research Centre, 251 Sir Frederick Banting Driveway, PL 2204A2, Ottawa, Ontario, Canada K1A 0K9. Email: [email protected] Eliana V. Elías, Mercedes & Martin Ferreyra Institute for Medical Research, INIMEC-CONICET and School of Medicine, Catholic University of Cordoba, Parque Velez Sarsfield, CP 5000, Cordoba, Argentina. Ronald Fayer, Environmental Microbial Safety Laboratory, Animal and Natural Resources Institute, Agricultural Research Service, United States Department of Agriculture, Building 173, BARC-East, 10300 Baltimore Avenue, Beltsville, MD 20705, USA. Email: [email protected] Guangtao Ge, Department of Computer Sciences, School of Engineering, Tufts University, 200 Westboro Road, North Grafton, MA 01536, USA.

Contributors

xi

Charles P. Gerba, Department of Soil, Water and Environmental Science, The University of Arizona, Tucson, AZ 85721, USA. Email: [email protected]. edu Frances D. Gillin, Department of Pathology, Division of Infectious Diseases, University of California at San Diego, San Diego, CA, USA. Email: [email protected] Natalia Gottig, Mercedes & Martin Ferreyra Institute for Medical Research, INIMEC-CONICET and School of Medicine, Catholic University of Cordoba, Parque Velez Sarsfield, CP 5000, Cordoba, Argentina. Chris Grundy, Public and Environmental Health Research Unit, London School of Hygiene and Tropical Medicine, Keppel Street, London, WC1E 7HT, UK. Thomas M. Hargy, Clancy Environmental Consultants, Inc., 20 Mapleville Depot, PO Box 314, Saint Albans, VT 05478, USA. Florence C.D. Harrison, School of Environmental Sciences, University of East Anglia, Norwich, UK. José Manuel Hernández, Department of Cell Biology, Centro de Investigación y de Estudios Avanzados del IPN, Av. IPN 2508, San Pedro, Zacatenco, CP 07300, México D.F., Mexico. Yunuen Hernandez, Laboratory of Parasitic Diseases, National Institutes of Allergy and Infectious Diseases, Bldg 4/Rm16, 4 Center Drive, Bethesda, MD 20892, USA. Nawal Samih Hijjawi, Department of Medical Laboratory Sciences, Faculty of Allied Health Sciences, The Hashemite University, PO Box 150459, Zarqa 13115, Jordan. Paul R. Hunter, School of Medicine, Health Policy and Practice, University of East Anglia, Earlham Road, Norwich NR4 7TJ, UK. Email: Paul.Hunter@ uea.ac.uk Edward L. Jarroll, Department of Biology, Northeastern University, Boston, MA 02115, USA. Email: [email protected] Joel Kamda, Department of Biology and Center for Infectious Diseases, Georgetown University, Washington, DC 20057, USA. Alexandra Keegan, The Co-operative Research Centre for Water Quality and Treatment, Australian Water Quality Centre, SA Water Corporation, Salisbury, South Australia 5108, Australia. Brendon King, The Co-operative Research Centre for Water Quality and Treatment, Australian Water Quality Centre, SA Water Corporation, Salisbury, South Australia 5108, Australia. Jessica Kissinger, Center for Tropical and Emerging Global Diseases and Department of Genetics, University of Georgia, 500 D.W. Brooks Drive, Athens, GA 30602, USA. Laetitia M. Kortbeek, Centre for Infectious Disease Control Netherlands, Laboratory for Infectious Diseases and Perinatal Screening (LIS), National Institute of Public Health and the Environment, PO Box 1, 3720 BA Bilthoven, the Netherlands. Email: [email protected] Sari Kovats, Public and Environmental Health Research Unit, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK.

xii

Contributors

John R. Kusel, Division of Infection and Immunity, Institute of Biomedical and Life Sciences, University of Glasgow, Glasgow G12 8QQ, UK. Susan J. Kutz, Faculty of Veterinary Medicine, University of Calgary, 3330 Hospital Drive NW, Calgary, Alberta, Canada T2N 4N1. Email: [email protected] Iain R. Lake, School of Environmental Sciences, University of East Anglia, Norwich, UK. Marco Lalle, Department of Infectious, Parasitic and Immunomediated Diseases, Istituto Superiore di Sanità, Viale Regina Elena 299, Rome 00161, Italy. Lynn P. Landry, Utility Analysis and Environmental Management, Policy and Planning Department, Metro Vancouver, Greater Vancouver Regional District, 4330 Kingsway, Burnaby, BC, Canada V5H 4G8. Nicholas F. LaRusso, Miles and Shirley Fiterman Center for Digestive Diseases, Division of Gastroenterology and Hepatology, Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905, USA. Email: larusso. [email protected] Tineke Lauwaet, Department of Pathology, Division of Infectious Diseases, University of California at San Diego, San Diego, CA, USA. Gloria León-Avila, Department of Parasitology, Escuela Nacional de Ciencias Biológicas, IPN, Carpio y Plan de Ayala, Casco de Santo Tomás, CP 11340, México, D.F., Mexico. Email: [email protected] Eric London, Tufts Cummings School of Veterinary Medicine, Division of Infectious Diseases, Tufts University, 200 Westboro Road, North Grafton, MA 01536, USA. Hugo D. Luján, Mercedes & Martin Ferreyra Institute for Medical Research, INIMEC-CONICET and School of Medicine, Catholic University of Cordoba, Parque Velez Sarsfield, CP 5000, Cordoba, Argentina. Email:hlujan@ immf.uncor.edu Hester Lyons, ADAS Consulting, Woodthorne, Wergs Road, Wolverhampton WV6 8TQ, UK. R.C. Maclean, Department of Microbiology and Immunology, Medical College of Virginia Campus, Virginia Commonwealth University, Richmond, VA 23298, USA. Francine Marciano-Cabral, Department of Microbiology and Immunology, Medical College of Virginia Campus, Virginia Commonwealth University, Richmond, VA 23298, USA. Vincent McDonald, Barts and the London School of Medicine and Dentistry, Centre for Gastroenterology, Queen Mary College, University of London, Turner Street, London E1 2AD, UK. Email: [email protected] Donato Menichella, Unit of Microbiology and Virology, Bambino Gesù Hospital, Scientific Institute, Piazza Sant’Onofrio 4, 00146, Rome, Italy. Email: [email protected] Hussni O. Mohammed, Department of Population Medicine and Diagnostic Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853, USA. Email: [email protected] Paul Monis, The Co-operative Research Centre for Water Quality and Treatment, Australian Water Quality Centre, SA Water Corporation, Salisbury, South Australia 5108, Australia. Email:[email protected]

Contributors

xiii

Gordon Nichols, Environmental and Enteric Diseases Department, Communicable Disease Surveillance Centre, Health Protection Agency (HPA) Centre for Infections, 61 Colindale Avenue, London NW9 5EQ, UK. Email: [email protected] Rosely A.B. Nichols, Scottish Parasite Diagnostic Laboratory (SPDL), Stobhill Hospital, 133 Balornock Road, Glasgow G21 3UW, UK. J. O’Grady, Strathclyde Institute of Pharmacy and Biomedical Sciences, University of Strathclyde, 27 Taylor Street, Glasgow G4 0NR, UK. Steven P. O’Hara, Miles and Shirley Fiterman Center for Digestive Diseases, Division of Gastroenterology and Hepatology, Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905, USA. Ynes R. Ortega, Center for Food Safety, University of Georgia, 1109 Experiment Street, Griffin, GA 30223, USA. Email: [email protected] Guadalupe Ortega-Pierres, Department of Genetics and Molecular Biology, Centro de Investigación y de Estudios Avanzados del IPN, México City, Av. Instituto Politécnico Nacional No. 2508, Col. San Pedro Zacatenco, México, D.F. 07360, Mexico. Email: [email protected] Daniel Palm, Centre for Microbial Preparedness, Swedish Institute for Infectious Disease Control, SE-171 82 Solna, Sweden. Lydden Polley, Western College of Veterinary Medicine, University of Saskatchewan, 52 Campus Drive, Saskatoon, SK, Canada S7N 5B4. Email:lydden. [email protected] Edoardo Pozio, Department of Infectious, Parasitic and Immunomediated Diseases, Istituto Superiore di Sanità, Viale Regina Elena 299, Rome 00161, Italy. Chris Proctor, ADAS Consulting, Woodthorne, Wergs Road, Wolverhampton WV6 8TQ, UK. Lorenza Putignani, Unit of Microbiology and Virology, Bambino Gesù Hospital, Scientific Institute, Piazza Sant’Onofrio 4, 00146, Rome, Italy. Email: [email protected] Rodrigo Quiroga, Mercedes & Martin Ferreyra Institute for Medical Research, INIMEC-CONICET and School of Medicine, Catholic University of Cordoba, Parque Velez Sarsfield, CP 5000, Cordoba, Argentina. Martha Robles-Flores, Departamento de Bioquímica, Facultad de Medicina, Universidad Nacional Autónoma de México, Apdo. Postal 70-159, México City, D.F. 04510, Mexico. Jean-François Rossignol, Division of Gastroenterology and Hepatology, Department of Medicine, Stanford University School of Medicine, 300 Pasteur Drive, Stanford, CA 94305, USA. Email: [email protected] Sukla Roychowdhury, Infectious Diseases and Immunology Unit, The Border Biomedical Research Center, Department of Biological Sciences, University of Texas at El Paso, 500 W. University Avenue, El Paso, TX 79968, USA. Cristina Russo, Unit of Microbiology and Virology, Bambino Gesù Hospital, Scientific Institute, Piazza Sant’Onofrio 4, 00146, Rome, Italy. Email:russocri@ opbg.net Una M. Ryan, School of Veterinary and Biomedical Sciences, Murdoch University, South Street, Murdoch, WA 6150, Australia. Email: Una.Ryan@ murdoch.edu.au

xiv

Contributors

Chris Saint, The Co-operative Research Centre for Water Quality and Treatment, Australian Water Quality Centre, SA Water Corporation, Salisbury, South Australia 5108, Australia. Sanya J. Sanderson, Department of Pre-Clinical Veterinary Science and Veterinary Pathology, Faculty of Veterinary Science, University of Liverpool, Crown Street, Liverpool L69 7ZJ, UK. Email: [email protected] Monica Santín, Environmental Microbial Safety Laboratory, Animal and Natural Resources Institute, Agricultural Research Service, United States Department of Agriculture, Building 173, BARC-East, 10300 Baltimore Avenue, Beltsville, MD 20705, USA. Payal Sarkar, Department of Soil, Water and Environmental Science, The University of Arizona, Tucson, AZ 85721, USA. Keriman S¸ener, Department of Biology, Northeastern University, Boston, MA 02115, USA. Email:[email protected] Joana C. da Silva, Institute for Genome Sciences, University of Maryland School of Medicine, 20 Penn Street, Baltimore, MD 21201, USA. Steven M. Singer, Department of Biology and Center for Infectious Diseases, Georgetown University, Washington, DC 20057, USA. Email: sms3@ georgetown.edu Jan Šlapeta, Faculty of Veterinary Science, McMaster Building B14, University of Sydney, NSW 2006, Australia. Email: [email protected] Huw V. Smith, Scottish Parasite Diagnostic Laboratory (SPDL), Stobhill Hospital, 133 Balornock Road, Glasgow G21 3UW, UK. Email: huw.smith@ northglasgow.scot.nhs.uk Christopher B. Sullivan, Scottish Parasite Diagnostic Laboratory (SPDL), Stobhill Hospital, 133 Balornock Road, Glasgow G21 3UW, UK. Staffan G. Svärd, Department of Cell and Molecular Biology, BMC, Uppsala University, Box 596, SE-751 24 Uppsala, Sweden. Email: staffan.svard@ icm.uu.se Thomas J. Templeton, Department of Microbiology and Immunology, Weill Medical College of Cornell University, 1300 York Avenue, Box 62, NY 10021 USA. Email: [email protected] R.C. Andrew Thompson, WHO Collaborating Centre for the Molecular Epidemiology of Parasitic Infections, Division of Veterinary and Biomedical Sciences, Murdoch University, Murdoch, WA 6150, Australia. Email: a.thompson@ murdoch.edu.au Jorge Tovar, School of Biological Sciences, Royal Holloway University of London, Egham TW20 0EX, UK. James M. Trout, Environmental Microbial Safety Laboratory, Animal and Natural Resources Institute, Agricultural Research Service, United States Department of Agriculture, Building 173, BARC-East, 10300 Baltimore Avenue, Beltsville, MD 20705, USA. Saul Tzipori, Tufts Cummings School of Veterinary Medicine, Division of Infectious Diseases, Tufts University, 200 Westboro Road, North Grafton, MA 01536, USA. Harry van Keulen, Department of Biological, Geological and Environmental Sciences, Cleveland State University, Cleveland, OH 44115, USA.

Contributors

xv

Susan E. Wade, Department of Population Medicine and Diagnostic Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853, USA. Jonathan M. Wastling, Department of Pre-Clinical Veterinary Science and Veterinary Pathology, Faculty of Veterinary Science, University of Liverpool, Crown Street, Liverpool L69 7ZJ, UK. Email: [email protected] Giovanni Widmer, Tufts Cummings School of Veterinary Medicine, Division of Infectious Diseases, Tufts University, 200 Westboro Road, North Grafton, MA 01536, USA. Email: [email protected] Lihua Xiao, Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Building 22, Mail Stop F-12, 4770 Buford Highway, Atlanta, GA 30341, USA. Email: [email protected] Linghui Zhang, Tufts Cummings School of Veterinary Medicine, Division of Infectious Diseases, Tufts University, 200 Westboro Road, North Grafton, MA 01536, USA. Guan Zhu, Department of Veterinary Pathobiology, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, 4467 TAMU, College Station, TX 77843, USA. Email: [email protected]

This page intentionally left blank

1

The Impact of Giardia on Science and Society R.C.A. THOMPSON WHO Collaborating Centre for the Molecular Epidemiology of Parasitic Infections, Murdoch University, WA, Australia

Abstract Although Giardia has a long history, it is only recently that the clinical impact of Giardia in children has been recognized. Similarly, the emergence of Giardia as a frequent parasite of companion animals, livestock and wildlife raises questions about the clinical and zoonotic significance of such infections. Transmission patterns have already been described but the frequency of interaction between cycles of transmission is only just beginning to be addressed in molecular epidemiological studies. The application of molecular tools has provided information which sets the scene for revising the taxonomy of Giardia. With the completion of the Giardia genome sequencing project, it is hoped that it will soon be possible to compare genome and phenome and provide information of practical value for the control of Giardia, as well as increasing our understanding of the evolution and phylogenetic relationships of Giardia.

Introduction The protozoa that collectively comprise the genus Giardia have intrigued biologists and clinicians for over 300 years, ever since Antony van Leeuwenhoek first discovered the organism (Meyer, 1994). This enigmatic protozoan possesses a number of unusual characteristics including the presence of two similar, transcriptionally active diploid nuclei, the absence of mitochondria and peroxisomes, and a unique attachment organelle – the ventral sucking disc (Thompson and Monis, 2004; Morrison et al., 2007). Phylogenetic relationships are controversial, with one school of thought suggesting that Giardia is a basal eukaryote and the other that Giardia comprises one of many divergent eukaryotic lineages that adapted to a microaerophilic lifestyle rather than diverging before the endosymbiosis of the mitochondrial ancestor (Thompson and Monis, 2004; Morrison et al., 2007). Despite its long history, our understanding of the pathogenesis of Giardia infections and its relationship with its host is limited, and we do not know why clinical disease occurs in some individuals but may not be apparent in others © CAB International 2009. Giardia and Cryptosporidium: From Molecules to Disease (eds G. Ortega-Pierres et al.)

1

2

R.C.A. Thompson

(Savioli et al., 2006). There are no known virulence factors or toxins, and variable expression of surface proteins may allow evasion of host immune responses and adaptation to different environments (Morrison et al., 2007). For many years after van Leeuwenhoek first described Giardia, there was controversy as to whether it was a parasite that caused disease or a commensal. It was not until the 1920s that influential proponents such as Dobell and Miller gave support to the link between the presence of Giardia and disease and, more importantly, a link with malabsorption syndromes (Cox, 1998). Thirty years later, Rendtorff established infections with Giardia in humans following the oral inoculation of cysts (Rendtorff, 1954). Today, there is no doubt about the clinical significance of Giardia infections in humans, and particularly the impact of giardiasis in children and its association with failure to thrive and wasting syndromes (Thompson, 2008).

Taxonomy The recent application of molecular, PCR-based tools has enabled the genetic relationships of a range of morphologically identical ‘strains’ of Giardia to be determined (Thompson and Monis, 2004; Cacciò et al., 2005; Traub et al., 2005; Smith et al., 2006). As a consequence, a large number of species and genotypes of Giardia are now recognized that differ principally in their host range. The current taxonomy of Giardia is summarized in Table 1.1 and has been extensively reviewed (Thompson and Monis, 2004; Cacciò et al., 2005; Thompson et al., 2007). The nomenclature most widely accepted at the present time for the genotypes that have been characterized is ‘assemblage’, although a revised taxonomy has been proposed (Thompson and Monis, 2004; Cacciò et al., 2005). Some species and genotypes/assemblages appear to be restricted to particular species or types of hosts (e.g. Giardia assemblages C/D (G. canis) and E (G. bovis) in dogs and livestock, respectively; see Table 1.1) whereas others have broad host ranges, including humans, (e.g. G. duodenalis assemblages A and B; see Table 1.1) and are therefore of zoonotic significance. Giardia duodenalis (syn. G. intestinalis, G. lamblia) is the only species found in humans. Table 1.1 summarizes the proposed revised taxonomy for assemblages of Giardia. Such a formal nomenclature will avoid confusion and enhance communication.

Clinical Impact Humans In developed countries, infections with Giardia are most common in children, especially in daycare centres and among travellers, and a rising incidence in such settings has led to the designation of giardiasis as a re-emerging infectious disease in the developed world (Thompson, 2000, 2004; Eckmann, 2003; Thompson and Monis, 2004). The World Health Organization (WHO) has given consideration to intestinal protozoa for many years, but because of their very different disease dynamics they did not initially form part of the ‘Neglected Diseases Initiative’.

The Impact of Giardia on Science and Society Table 1.1.

3

Established and proposeda species/assemblages of Giardia.

Species/assemblage

Host

G. duodenalis/assemblage A

Humans and other primates, dogs, cats, livestock, rodents and other wild mammals Humans and other primates, dogs

G. duodenalis/assemblage B (G. enterica)a G. agilis G. muris G. psittaci G. ardeae G. duodenalis/assemblage C/D (G. canis)a G. duodenalis/assemblage F (G. cati)a G. duodenalis/assemblage E (G. bovis)a G. duodenalis/assemblage G (G. simondi )a aSee

Amphibians Rodents Birds Birds Dogs Cats Cattle and other hoofed livestock Rats

Thompson and Monis (2004), Cacciò et al. (2005).

However, since they all have a common link with poverty, the current view is to take a comprehensive approach to all these diseases. In September 2004, Giardia was included in WHO’s Neglected Diseases Initiative (Savioli et al., 2006). In developing countries, particularly in Asia, Africa and Latin America, about 200 million people have symptomatic giardiasis with some 500,000 new cases being reported each year (WHO, 1996). Children living in communities are most commonly infected in developing countries, particularly among disadvantaged groups living in isolated communities, such as Australian Aborigines (Thompson, 2000; Hesham et al., 2005; Savioli et al., 2006). These children are most at risk from the chronic consequences of Giardia infection. Although animals may serve as reservoirs of Giardia infection that under certain circumstances may spill over to humans; from a clinical viewpoint, direct human-tohuman transmission is of most significance, particularly in situations where the frequency of transmission is high. Human-to-human transmission of Giardia can occur indirectly through the accidental ingestion of cysts in contaminated water or food, or directly in environments where hygiene levels may be compromised, such as daycare centres or disadvantaged community settings, where the frequency of transmission is high and/or conditions are conducive to direct person-to-person transfer (Thompson, 2000; Hesham et al., 2005). Under such circumstances, children may be at constant risk of infection, even though chemotherapeutic interventions may be instituted (Thompson et al., 2001; Savioli et al., 2006). If children are constantly exposed they will be re-infected rapidly, since anti-giardial agents have no residual activity. The fact that children in such endemic settings do not appear to develop resistance to Giardia infection may be due to suboptimal immunological competence and/or infection with different ‘strains’/subgenotypes of Giardia (Hopkins et al.,

4

R.C.A. Thompson

1999). It might be expected that competitive interactions would result in the predominance of particular genotypes of Giardia and the exclusion of others (Thompson et al., 2001), but this does not appear to be the case, perhaps due to the suboptimal nutritional and immunological environments to which the parasites are exposed in such communities. Symptomatic infection in humans may not be evident in a significant proportion of infected individuals (Troeger et al., 2007), and represents only a fraction (20–80%) of all stool-positive Giardia infections (Nash et al., 1987; Flanagan, 1992; Rodriguez-Hernandez et al., 1996). Symptoms are highly variable but include persistent, usually short-term, diarrhoea, epigastric pain, nausea, vomiting and weight loss (Thompson et al., 1993; Eckmann, 2003). Symptoms typically occur 6–15 days after infection and last for 2–4 days. As such, infection is assumed to be self-limiting in more than 85% of cases (indicating that effective host defences exist), although chronic cases occur occasionally in the absence of apparent immunodeficiencies (Nash et al., 1987; Flanagan, 1992). The risk factors for clinical giardiasis, particularly in humans, have yet to be resolved, but clearly involve host and environmental factors as well as the ‘strain’/ genotype/assemblage of the parasite (Buret, 2007). However, a distinction needs to be made between the effects of a single infection, which may give rise to the ‘classical’ short-term episode of diarrhoea, and the long-term effects of persistent Giardia infection, particularly in children, in environments where the frequency of transmission is high. Here the picture is very different. In endemic foci where the frequency of transmission is high and often enhanced by poor hygiene and environmental contamination, children are at particular risk from the more serious and long-term consequences of Giardia infection that are associated with malnutrition, micronutrient deficiency and failure to thrive, iron deficiency anaemia and poor cognitive function (Hesham et al., 2004; Savioli et al., 2006; Gbakima et al., 2007; Gonen et al., 2007). Clearly, the impact of Giardia in such circumstances will be exacerbated by poor/suboptimal nutrition and concurrent infections with other enteric parasites such as Hymenolepis nana, Entamoeba coli and Blastocystis. Longitudinal studies on the impact of enteric parasites on childhood growth and mental development in such endemic areas is urgently required (Savioli et al., 2006). The children who are infected in such environments, particularly in developing countries and among disadvantaged groups, represent the most important group in terms of the clinical impact of Giardia (Thompson, 2000; Hesham et al., 2005; Savioli et al., 2006). In such circumstances, it is debatable whether the regular use of drugs is of any benefit. This is in contrast to the situation with gastrointestinal helminths such as hookworm, where regular mass chemotherapy has been shown to have great benefit in control (Reynoldson et al., 1998; Thompson et al., 2001).

Domestic animals Livestock, wildlife, and companion animals are frequently infected with Giardia and are susceptible to host-adapted and zoonotic species of Giardia. Although infections with Giardia are common in dogs and cats, most infected animals

The Impact of Giardia on Science and Society

5

remain asymptomatic. If clinical disease occurs, it is usually associated with young animals and those in kennel or cattery situations (Robertson et al., 2000), where the effects of overcrowding, weaning and nutritional deficiency, may cause stress and exacerbate the effects of an infection (Thompson, 2004). The most consistent clinical sign of giardiasis in dogs and cats is small-bowel diarrhoea, which may be acute or chronic, and self-limiting, intermittent or continuous in nature. Giardia infection in ruminants is often asymptomatic but may also be associated with the occurrence of diarrhoea and ill-thrift in calves (O’Handley et al., 1999; Geurden et al., 2006). The importance of giardiasis as a cause of diarrhoea in ruminants is unclear, especially given that diarrhoea in ruminants is often multifactorial, with more than one pathogen detected (O’Handley and Olson, 2006). Nevertheless, the significance of Giardia infection in ruminants warrants further investigation, particularly with regard to production loss. In cattle, Giardia is commonly found alone or in combination with other pathogens as a cause of calf diarrhoea, which can have economic significance (O’Handley et al., 1999; Olson et al., 2004). In two studies in sheep, Giardia reduced the rate of weight gain, impaired feed efficiency and decreased carcass weight (Olson et al., 1995; Aloisio et al., 2006).

Wildlife It is often a common ‘knee-jerk’ reaction when parasites with zoonotic potential are found in wildlife that they represent a threat to public health as a reservoir and potential source of infection for humans (Thompson, 2004). Indeed, this was the case when WHO initially listed the common enteric protozoan parasite Giardia as a zoonosis over 25 years ago as a result of epidemiological observations suggesting that giardiasis in campers in Canada was caused by drinking stream water contaminated with Giardia from beavers (Thompson, 2004). No one thought to ask the question of where the beavers got their Giardia infections from until only beavers living downstream from a sewage works were found to be infected. With the subsequent application of molecular tools, it has been confirmed that beavers are susceptible to zoonotic strains of Giardia (see Thompson, 2004). The question now is: are they victim or villain with respect to human giardiasis? A similar situation has been reported in non-human primates for which there is a growing literature of the invasion of human pathogens into wild populations (Sleeman et al., 2000; Graczyk et al., 2001, 2002). For example, it was suggested that the discovery of Giardia and the cohabiting enteric protozoan Cryptosporidium in mountain gorillas in the Bwindi Impenetrable National Park, Uganda, was thought to indicate enhanced contact with humans and/or domestic livestock. This was confirmed when rangers and their cattle were found to be infected with Giardia and that the genotype was the same as that recovered from the gorillas (Graczyk et al., 2002). Muskoxen (Ovibos moschatus) are indigenous to the arctic tundra of Canada and Greenland and have been translocated to areas in Alaska, the USA, Russia, Norway and Sweden. These animals are well adapted to their northern environment, and tend to have a relatively simple parasite fauna. Recent surveys on the

6

R.C.A. Thompson

biodiversity and impacts of parasites in Arctic ungulates described Giardia duodenalis, assemblage A, the zoonotic genotype, in muskoxen (Kutz et al., 2008). This unexpected finding (a novel strain, or the livestock strain, was predicted) raises many interesting questions regarding the origin and epidemiology of this parasite in humans and wildlife in this Arctic ecosystem. In particular, is this a pathogen initially introduced to muskoxen by humans? Is Giardia now maintained as a sylvatic cycle in muskoxen (or other wildlife species in the locality) independent of humans? Does the Giardia from muskoxen spill back into humans? The impact of Giardia on the health and production (body condition, fecundity and pelage) of free-ranging ungulates, including muskoxen, remains unknown. In Australia, marsupials are commonly infected with Giardia (Kettlewell et al., 1998) but until recently it was not known to what species or strain(s) of Giardia they were susceptible. Studies on the quenda (Isoodon obesulus), a common widespread species of bandicoot in southern Australia, demonstrated that they were infected with a novel, genetically distinct, form of Giardia, so different from what has been described from humans and other animals that it probably represents a distinct species (Adams et al., 2004). The Giardia isolates genotyped from quenda in their natural habitats have all proved to be the novel strain. However, when quenda were trapped and examined on a farm, they were found to be infected with ‘domestic’ strains of Giardia normally found in livestock and humans (from assemblages A and E; see Table 1.1). Presumably, this reflects the susceptibility of the quenda to other strains of Giardia, as with the case of beavers in North America. This case study raises questions regarding the pathogenicity of non-host-adapted strains of Giardia in naïve wildlife hosts. Additionally, it also raises the question of competition between cohabiting ‘strains’ of Giardia (Thompson and Monis, 2004) and whether, in this case, and perhaps in other species of wildlife, zoonotic strains of Giardia can out-compete the host-specific wildlife strains.

Molecular Epidemiology Humans may be infected with Giardia genotypes belonging to assemblage A or assemblage B (Thompson and Monis, 2004; Cacciò et al., 2005). There is considerable evidence of phenotypic differences between these two assemblages in characters such as metabolism and growth rate (Thompson and Monis, 2004). It has therefore been proposed that there may be differences in the nature of infection with these two assemblages in humans which may be reflected in duration of infection, drug sensitivity and virulence (Thompson and Monis, 2004). There is growing evidence to support these suggestions but there is a need for more focused molecular epidemiological studies. For example, in tea-growing communities in Assam, India, the proportion of assemblage B and A infections in 18 infected people was 61% and 39%, respectively (Traub et al., 2004). Another study in the UK that examined 35 human clinical samples found that 64% were assemblage B, 27% were assemblage A genetic subgroup II, and the remainder were a mixture of assemblage B and assemblage A genetic group II (Amar et al., 2002).

The Impact of Giardia on Science and Society

7

There is also increasing evidence supporting differences in virulence between genetic groups of G. duodenalis. Several studies have examined the relationship between clinical symptoms and the genetic assemblage of G. duodenalis infecting human patients. Three of these, undertaken in Australia, Bangladesh and Spain, all found a statistically significant correlation between the presence of symptoms, as defined by diarrhoea, and infection with Giardia belonging to assemblage A (Read et al., 2002; Haque et al., 2005; Sahagún et al., 2007). In contrast, it was found, in all studies, that infections with assemblage B were usually asymptomatic. A study in the Netherlands of individuals presenting general practitioners with diarrhoeal complaints found that those infected with assemblage A isolates had intermittent diarrhoea, whereas those infected with assemblage B had acute or persistent diarrhoea (Homan and Mank, 2001). However, this study only sampled patients presenting with diarrhoeal complaints. There is a need for additional large-scale molecular epidemiological surveys of Giardia infections in humans. With the limited data currently available it is not possible to determine the geographical distribution and prevalence of human-infective genotypes. With such data it may be possible to determine the significance of any ‘strain’-related differences in virulence. Although studies on the occurrence of the different genotypes of Giardia serve to emphasize the potential public health risk from domestic dogs and cats, data on the frequency of zoonotic Giardia transmission are lacking (Thompson, 2004; Leonhard et al., 2007). Such information can be obtained from molecular epidemiological studies that genotype isolates of the parasites from susceptible hosts in localized endemic foci of transmission or as a result of longitudinal surveillance and genotyping of positive cases. In the former, recent research in localized endemic foci of transmission has provided evidence in support of the role of dogs in cycles of zoonotic Giardia transmission involving humans and domestic dogs from communities in tea-growing areas of Assam, India, and in temple communities in Bangkok, Thailand (Traub et al., 2004; Inpankaew et al., 2007). In both these studies, some dogs and their owners sharing the same living area were shown to harbour isolates of G. duodenalia from the same assemblage. Other studies have shown that zoonotic genotypes of Giardia may occur frequently in individual pet dogs living in urban areas (for a review, see Leonhard et al., 2007).

Future Perspectives Sex Population genetic studies of Giardia in communities where the frequency of transmission is very high, have found evidence of occasional bouts of genetic exchange in Giardia (Meloni et al., 1995). These authors demonstrated multiple banding patterns in a number of isolates of Giardia by allozyme electrophoresis which, if a true reflection of the underlying genotypes of the isolates, would seem to indicate that G. duodenalis is functionally diploid, and that recombination or sexual reproduction must have occurred at some stage to produce the apparent

8

R.C.A. Thompson

heterozygotes (Tait, 1983; Meloni et al., 1995). These observations have been supported by subsequent population genetic studies (Cooper et al., 2007) and recent genomic studies (Ramesh et al., 2005; Morrison et al., 2007; Teodorovic et al., 2007). The evolutionary advantage this gives to Giardia is the capacity to respond to adversity, for example selection pressures imposed by regular exposure to anti-giardial drugs or competition with cohabiting ‘strains’ in circumstances where the likelihood of mixed infections is common (Hopkins et al., 1999).

Correlating phenotype and genotype Genetic studies and sequencing of the Giardia genome have laid an important foundation for understanding this parasite. As a consequence of delineating genotypic groupings, a clearer picture of transmission patterns and host specificity has been obtained. However, phenotypic differences in virulence, drug sensitivity and infectivity have been reported in isolates of G. duodenalis and there is a need to correlate these observations with genetic differences (Thompson and Monis, 2004; Cacciò et al., 2005). The correlation of phenotype and genotype will provide important information about the host–parasite relationship in Giardia infections. The complexity of any biological system, including that of Giardia, lies at the protein level and genomics alone cannot be used to understand these complexities. Using a proteomic approach to examine differences between the human infective genotypes of Giardia, we have found differences in a number of proteins between the two human infective genetic groups – assemblages A and B. These proteins of difference appear to be associated with virulence and pathogenicity but at what level of functionality remains unclear at this stage (Steuart et al., in press). We have also identified proteins, some of which are novel and Giardia-specific, that appear to be key to its survival and transmission, including growth factors and developmental triggers. Giardia is also of evolutionary and biological significance in terms of understanding the origin of higher animals from bacteria as well as fundamental questions about the parasitic way of life. Thus some of the proteins we have identified may contribute to a better understanding of Giardia’s pivotal position in our understanding of eukaryote biology and evolution.

References Adams, P.J., Monis, P.T., Elliot, A.D. and Thompson, R.C. (2004) Cyst morphology and sequence analysis of the small subunit rDNA and ef1 alpha identifies a novel Giardia genotype in a quenda (Isoodon obesulus) from Western Australia. Infection, Genetics and Evolution 4, 365–370. Aloisio, F., Filippini, G., Antenucci, P., Lepri, E., Pezzotti, G., Cacciò, S.M. and Pozio, E. (2006) Severe weight loss in lambs infected with Giardia duodenalis assemblage B. Veterinary Parasitology 142, 154–158. Amar, C.F.L., Dear, P.H., Pedraza-Díaz, S., Looker, N., Linnane, E. and McLauchlin, J. (2002) Sensitive PCR-restriction fragment length polymorphism assay for detection

The Impact of Giardia on Science and Society

9

and genotyping of Giardia duodenalis in human feces. Journal of Clinical Microbiology 40, 446–452. Buret, A.G. (2007) Mechanisms of epithelial dysfunction in giardiasis. Gut 56, 316–317. Cacciò, S.M., Thompson, R.C.A., McLauchlin, J. and Smith, H.V. (2005) Unravelling Cryptosporidium and Giardia epidemiology. Trends in Parasitology 21, 430–437. Cooper, M.A., Adam, R.D., Worobey, M. and Sterling, C.R. (2007) Population genetics provides evidence for recombination in Giardia. Current Biology 17, 1984–1988. Cox, F. (1998) History of human parasitology. In: Cox, F.E.G., Kreier, J.P. and Wakelin, D. (eds) Microbiology and Microbial Infections. Vol. 5: Parasitology. Arnold, London, pp. 3–18. Eckmann, L. (2003) Mucosal defences against Giardia. Parasite Immunology 25, 259–270. Flanagan, P.A. (1992) Giardia – diagnosis, clinical course and epidemiology: a review. Epidemiology and Infection 109, 1–22. Gbakima, A.A., Konteh, R., Kallon, M., Mansaray, H., Sahr, F., Bah, Z.J., Spencer, A. and Luckay, A. (2007) Intestinal protozoa and intestinal helminthic infections in displacement camps in Sierra Leone. African Journal of Medical Science 36, 1–9. Gonen, C., Yilmaz, N., Yalcin, M., Simsek, I. and Gonen, O. (2007) Diagnostic yield of routine duodenal biopsies in iron deficiency anaemia: a study from Western Australia. European Journal of Gastroenterology and Hepatology 19, 37–41. Geurden, T., Claerebout, E., Dursin, L., Deflandre, A., Bernay, F., Kaltsatos, V. and Vercruysse, J. (2006) The efficacy of an oral treatment with paromomycin against an experimental infection with Giardia in calves. Veterinary Parasitology 135, 241–247. Graczyk, T.K., DaSilva, A.J., Cranfield, M.R., Nizeyi, J.B., Kalema, G.R.N.N. and Pieniazek, N.J. (2001) Cryptosporidium parvum genotype 2 infections in free-ranging mountain gorillas (Gorilla gorilla beringei) of the Bwindi Impenetrable National Park, Uganda. Parasitology Research 87, 368–370. Graczyk, T.K., Bosco-Nizeyi, J., Ssebide, B., Thompson, R.C., Read, C. and Cranfield, M.R. (2002) Anthropozoonotic Giardia duodenalis genotype (assemblage) A infections in habitats of free-ranging human-habituated gorillas, Uganda. Journal of Parasitology 88, 905–909. Haque, R., Roy, S., Kabir, M., Stroup, S.E., Mondal, D. and Houpt, E.R. (2005) Giardia assemblage A infection and diarrhoea in Bangladesh. Journal of Infectious Diseases 192, 2171–2173. Hesham, M.S., Edariah, A.B. and Norhavat, M. (2004) Intestinal parasitic infections and micronutrient deficiency: a review. Medical Journal of Malaysia 59, 284–293. Hesham, M.S., Azlin, M., Nor Aini, U.N., Shaik, A., Sa’iah, A., Fatmah, M.S., Ismail, M.G., Firdaus, M.S.A., Aisah, M.Y., Rozlida, A.R. and Norhayati, M. (2005) Giardiasis as a predictor of childhood malnutrition in Orang Asli children in Malaysia. Transactions of the Royal Society for Tropical Medicine and Hygiene 99, 686–691. Homan, W.L. and Mank, T.G. (2001) Human giardiasis: genotype linked differences in clinical symptomatology. International Journal for Parasitology 31, 822–826. Hopkins, R.M., Constantine, C.C., Groth, D.A., Wetherall, J.D., Reynoldson, J.A. and Thompson, R.C.A. (1999) PCR-based DNA fingerprinting of Giardia duodenalis isolates using the intergenic rDNA spacer. Parasitology 118, 531–539. Inpankaew, T., Traub, R., Thompson, R.C.A. and Sukthana, Y. (2007) Canine parasitic zoonoses and temple communities in Thailand. Southeast Asian Journal of Tropical Medicine and Public Health 38, 247–255. Kettlewell, J.S., Bettiol, S.S., Davies, N., Milstein, T. and Goldsmid, J.M. (1998) Epidemiology of giardiasis in Tasmania: a potential risk to residents and visitors. Journal of Travel Medicine 5, 127–130.

10

R.C.A. Thompson Kutz, S.J., Thompson, R.C.A., Kandola, K., Nagy, J., Wielinga, C., Polley, L. and Elkin, B. (2008) Giardia assemblage A: human genotype in muskoxen in the Canadian Arctic. Parasites and Vectors 1, 32 pp. Leonhard, S., Pfister, K., Beelitz, P., Wielinga, C. and Thompson, R.C.A. (2007) The molecular characterisation of Giardia from dogs in Southern Germany. Veterinary Parasitology 150, 33–38. Meloni, B.P., Lymbery, A.J. and Thompson, R.C.A. (1995) Genetic characterization of isolates of Giardia duodenalis by enzyme electrophoresis: implications for reproductive biology, population structure, taxonomy, and epidemiology. Journal of Parasitology 81, 368–383. Meyer, E.A. (1994) Giardia as an organism. In: Thompson, R.C.A., Reynoldson., J.A. and Lymbery, A.J. (eds) Giardia: From Molecules to Disease. CAB International, Wallingford, UK, pp. 3–15. Morrison, H.G., McArthur, A.G., Gillin, F.D., Aley, S.B., Adam, R.D., Olsen, G.J., Best, A.A., Cande, W.Z., Chen, F., Cipriano, M.J., Davids, B.J., Dawson, S.C., Elmendorf, H.G., Hehl, A.B., Holder, M.E., Huse, S.M., Kim, U.U., Lasek-Nesselquist, E., Manning, G., Nigam, A., Nixon, J.E. J., Palm, D., Passamaneck, N.E., Prabhu, A., Reich, C.I., Reiner, D.S., Samuelson, J., Svard, S. and Sogin, M.L. (2007) Genomic minimalism in the early diverging intestinal parasite Giardia lamblia. Science 317, 1921–1926. Nash, T.E., Herrington, D.A., Losonsky, G.A. and Levine, M.M. (1987) Experimental human infections with Giardia lamblia. Journal of Infectious Diseases 156, 974–984. O’Handley, R.M. and Olson, M.E. (2006) Giardiasis and cryptosporidiosis in ruminants. Veterinary Clinics of North America: Food Animal Practice 22, 623–643. O’Handley, R.M., Cockwill, C., McAllister, T.A., Jelinski, M., Morck, D.W. and Olson, M.E. (1999) Duration of naturally acquired giardiosis and cryptosporidiosis in dairy calves and their association with diarrhea. Journal of the American Veterinary Medical Association 214, 391–396. Olson, M.E., McAllister, T.A., Deselliers, L., Morck, D.W., Cheng, K.J., Buret, A.G. and Ceri, H. (1995) Effects of giardiasis on production in a domestic ruminant (lamb) model. American Journal of Veterinary Research 56, 1470–1474. Olson, M.E., O’Handley, R.M., Ralston, B.J., McAllister, T.A. and Thompson, R.C.A. (2004) Update on Cryptosporidium and Giardia infections in cattle. Trends in Parasitology 20, 185–191. Ramesh, M.A., Malik, S.B. and Logsdon, J.M., Jr (2005) A phylogenomic inventory of meiotic genes: evidence for sex in Giardia and an early eukaryotic origin of meiosis. Current Biology 15, 185–191. Read, C., Walters, J., Robertson, I.D. and Thompson, R.C.A. (2002) Correlation between genotype of Giardia duodenalis and diarrhoea. International Journal for Parasitology 32, 229–231. Rendtorff, R.C. (1954) The experimental transmission of human intestinal protozoan parasites. II. Giardia lamblia cysts given in capsules. American Journal of Hygiene 59, 209–220. Reynoldson, J.A., Behnke, J.M., Gracey, M., Horton, R.J., Spargo, R., Hopkins, R.M., Constantine, C.C., Gilbert, F., Stead, C., Hobbs, R.P. and Thompson, R.C.A. (1998) Efficacy of albendazole against Giardia and hookworm in a remote Aboriginal community in the north of Western Australia. Acta Tropica 71, 27–44. Robertson, I.D., Irwin, P.J., Lymbery, A.J. and Thompson, R.C.A. (2000) The role of companion animals in the emergence of parasitic zoonoses. International Journal for Parasitology 30, 1369–1377. Rodriguez-Hernandez, J., Canut-Blasco, A. and Martin-Sanchez, A.M. (1996) Seasonal prevalences of Cryptosporidium and Giardia infections in children attending day

The Impact of Giardia on Science and Society

11

care centres in Salamanca (Spain) studied for a period of 15 months. European Journal of Epidemiology 12, 291–295. Sahagún, J., Clavel, A., Goñi, P., Seral, C., Llorente, M.T., Castillo, F.J., Capilla, S., Arias, A. and Gómez-Lus, R. (2007) Correlation between the presence of symptoms and the Giardia duodenalis genotype. European Journal of Clinical Microbiology and Infectious Diseases 27, 81–83. Savioli, L., Smith, H. and Thompson, A. (2006) Giardia and Cryptosporidium join the ‘Neglected Diseases Initiative’. Trends in Parasitology 22, 203–208. Sleeman, J.M., Meader, L.L., Mudakikwa, A.B., Foster, J.W. and Patton, S. (2000) Gastrointestinal parasites of mountain gorillas (Gorilla gorilla beringei) in the Parc National des Volcans, Rwanda. Journal of Zoo and Wildlife Medicine 31, 322–328. Smith, H.V., Cacciò, S.M., Tait, A., McLauchlin, J. and Thompson, R.C.A. (2006) Tools for investigating the environmental transmission of Cryptosporidium and Giardia infections in humans. Trends in Parasitology 22, 160–167. Steuart, R.F.L., O’Handley, R., Lipscombe, R.J. and Thompson, R.C.A. (2008) Alpha 2 Giardin is an assemblage A specific protein of human infective Giardia duodenalsis. Parasitology (in press). Tait, A. (1983) Sexual processes in the Kinetoplastida. Parasitology 86, 29–57. Teodorovic, S., Braverman, J.M. and Elmendorf, H.G. (2007) Unusually low levels of genetic variation among Giardia Lamblia isolates. Eukaryotic Cell 6, 1421–1430. Thompson, R.C.A. (2000) Giardiasis as a re-emerging infectious disease and its zoonotic potential. International Journal for Parasitology 30, 1259–1267. Thompson, R.C.A. (2004) The zoonotic significance and molecular epidemiology of Giardia and giardiasis. Veterinary Parasitology 126, 15–35. Thompson, R.C.A. (2008) Giardiasis: modern concepts in control and management. Annales Nestlé 66, 23–29. Thompson, R.C.A. and Monis, P.T. (2004) Variation in Giardia: implications for taxonomy and epidemiology. Advances in Parasitology 58, 69–137. Thompson, R.C.A., Reynoldson, J.A. and Mendis, A.H.W. (1993) Giardia and giardiasis. Advances in Parasitology 32, 71–160. Thompson, R.C.A., Reynoldson, J.A., Garrow, S.J., McCarthy, J.S. and Behnke J.M. (2001) Towards the eradication of hookworm in an isolated Australian community. Lancet 357, 770–771. Thompson, R.C.A., Traub, R.J. and Parameswaran, N. (2007) Molecular epidemiology of foodborne parasitic zoonoses. In: Murrell, K.D. and Fried, B (eds) Food-Borne Parasitic Zoonoses: Fish and Plant-Borne Parasites. Springer, New York, pp. 383–415. Traub, R.J., Monis, P.T., Robertson, I., Irwin, P., Mencke, N. and Thompson, R.C.A. (2004) Epidemiological and molecular evidence supports the zoonotic transmission of Giardia among humans and dogs living in the same community. Parasitology 128, 253–262. Traub, R.J., Monis, P.T. and Robertson, I.D. (2005) Molecular epidemiology: a multidisciplinary approach to understanding parasitic zoonoses. International Journal for Parasitology 35, 1295–1308. Troeger, H., Epple, H.-J., Schneider, T., Wahnschaffe, U., Ullrich, R., Burchard, G.-D., Jelinek, T., Zeitz, M., Fromm, M. and Schulzke, J.-D. (2007) Effect of chronic Giardia lamblia infection on epithelial transport and barrier function in human duodenum. Gut 56, 316–317. WHO (1996) Fighting Disease, Fostering Development: The World Health Report 1996. World Health Organization, Geneva.

2

Cryptosporidium in Cattle: From Observing to Understanding R. FAYER, M. SANTÍN AND J.M. TROUT United States Department of Agriculture, Beltsville, MD, USA

Abstract Cryptosporidium parvum is a zoonotic pathogen transmissible from a variety of animals to humans and is a considerable public health concern. Dairy cattle have been identified in numerous reports as a major source of environmental contamination with this pathogen. However, the vast majority of these reports have been based on microscopic examination of the organism in faeces from cattle. This chapter traces the progress of research on bovine cryptosporidiosis from the first observations of infection to the present understanding of susceptibility on the part of the bovine host and pathogenicity on the part of the parasites that constitute the taxa under the umbrella of the genus Cryptosporidium. It includes information based on molecular typing with the SSU rRNA gene and subtyping with the GP60 gene, which enables epidemiologists and others to trace the sources of Cryptosporidium-related outbreaks.

Introduction Recent archaeological evidence from Egypt suggests that cattle were herded in prehistoric times, possibly from around 12,500 BC. It is not clear whether these cattle originally came from the ‘Fertile Crescent’. Initially, it is presumed, cattle were killed for food, later they were used as draught animals, and for thousands of years cows’ milk was consumed only by calves. As farming methods evolved, cattle became a major source of protein, milk and milk products, and leather. Over time, livestock production intensified to support growing human populations. Larger and larger herds became concentrated in increasingly smaller areas to maximize the efficiency of production. In industrialized countries, as well as less affluent areas of the world, the consequence of this evolution in animal husbandry was seen through the impact of enteric diseases on neonates, of which cryptosporidiosis is but one. Through the millennia of this close association between humans and cattle, enteric pathogens had billions upon billions of opportunities to adapt and to be readily transmitted between these hosts. Those that have are 12

© CAB International 2009. Giardia and Cryptosporidium: From Molecules to Disease (eds G. Ortega-Pierres et al.)

Cryptosporidium in Cattle

13

of concern to public health and among them is the species Cryptosporidium parvum. Cryptosporidium infections have been reported in cattle worldwide. Before the development and application of molecular methods to aid in species determination, numerous publications simply documented the presence of Cryptosporidium oocysts in cattle faeces. Subsequently, many other publications have identified Cryptosporidium species in cattle but with limited information on prevalence and herd management. Cryptosporidiosis, especially in calves, has been associated with a wide range of clinical signs from no apparent ill effects to severe morbidity, resulting in poor performance and production losses and in some instances mortality. From the perspective of human health, cattle have often been implicated as a source of zoonotic Cryptosporidium species. Risk of human infection has been based on physical contact with cattle, contamination of fresh fruits and vegetables with manure, and manure runoff from farms into drinking water supplies. With the goal of producing healthy cattle while protecting food and water supplies, studies have been undertaken to obtain a clear understanding of the species of Cryptosporidium that infect cattle, the prevalence of infection, and the relationship of these species to the age of the animals. There has been much progress in defining the Cryptosporidium species, genotypes, and some subtypes present in cattle. This chapter traces the progress of research on bovine cryptosporidiosis from the first observations of infection in cattle to our present understanding of susceptibility and pathogenicity. It includes those parasites that constitute the taxa under the umbrella of the genus Cryptosporidium. Application of this information can benefit both animal and human health.

Early Reports of Cryptosporidiosis in Cattle The first report of bovine cryptosporidiosis appeared in 1971 and described stages of the parasite in tissue sections of the jejunum from an 8-month-old heifer with chronic diarrhoea (Panciera et al., 1971). This observation was soon followed by others in which diarrhoeic dairy and beef calves, 2 weeks old and younger, had similar infections (Barker and Carbonell, 1974; Meuten et al., 1974; Morin et al., 1976). The association between the parasite and illness became stronger when it was reported that cryptosporidia were probably common enteropathogens of calves (Pohlenz et al., 1978) and that Cryptosporidium was a pathogen in experimentally infected calves (Tzipori et al., 1983). A survey of neonatal dairy calves in Maryland, USA, found Cryptosporidium in healthy as well as in diarrhoeic calves (Leek and Fayer, 1984). This finding led to a study of factors contributing to clinical illness in calves experimentally infected with Cryptosporidium obtained from pooled calves’ faeces (Fayer et al., 1985). In that study, clinical illness was not consistently found in neonatal dairy calves experimentally infected with 3.2–30 × 106 oocysts, except when rotavirus and/or Clostridium perfringens was also present. It appeared that Cryptosporidium required the presence of other pathogens to produce illness. These findings gave rise to the following questions: 1. What was different in this study from reports in which cryptosporidiosis in calves was strongly associated with morbidity and mortality?

14

R. Fayer et al.

2. Were other isolates of the parasite more pathogenic or did other studies simply lack viral and bacterial assessments? 3. Why were some naturally infected calves healthy and others severely ill? At that time there appeared to be a single species of Cryptosporidium that lacked host specificity, and answers to those questions would, in part, require the development of better taxonomic data. In the next few years, cattle were recognized as hosts for two species, C. parvum and C. muris (Anderson, 1987), infecting the intestine and the abomasum, respectively, and represented by small and large oocysts (C. parvum, ~4.5 × 5.5 mm and C. muris, ~5.5 × 7.4 mm). Molecular techniques and cross-species transmission studies eventually identified the abomasal form with the large oocysts as a new species, C. andersoni (Lindsay et al., 2000). Cryptosporidium parvum was then associated with diarrhoea in young calves and C. andersoni with asymptomatic adults. As reports accumulated it became apparent that under farm and field conditions calves acquired infections shortly after birth. The highest prevalence of cryptosporidiosis in cattle was found at 1–3 weeks of age, with oocysts excreted for an average of 12 days and diarrhoea, when present, lasting an average of 8 days (reviewed by Santín and Trout, 2008).

What is the Immune Status of Calves and How Does it Affect Susceptibility to Infection? Cell-mediated immunity had been shown to play an important role in control of C. parvum infections in mouse models, and by the early 1990s it was clear that HIV AIDS patients with low levels of T cells were extremely susceptible to cryptosporidiosis and other opportunistic pathogens, although little was known regarding the immune status of calves. To determine why young calves were so susceptible to infection, studies were designed to identify lymphoid cell populations at the site of infection in the ileum where lymphocytes could respond directly to the parasite (Pasquali et al., 1997; Canals et al., 1998). Intra-epithelial lymphocytes (IEL), and lamina propria lymphocytes (LPL) were collected from C. parvum-infected calves and uninfected control calves; the infected group was inoculated with oocysts at 1–2 days of age. Cells were collected from both groups at 7–9 days of age and analysed for phenotype and cytokine mRNA production. Significant increases in CD2+, CD3+, CD4+ and CD8+ T cells were observed in the IELs of infected versus uninfected calves. These findings were supported by reports of elevated IEL CD8+ T cells in infected versus uninfected calves (Wyatt et al., 1997) and of elevated LPL CD4+, CD8+ T, and g/d T cells in infected versus uninfected calves (Abrahamsen et al., 1997). IELs and LPLs from uninfected calves contained much lower percentages of CD4+ and CD8+ T cells than found in adult cattle (Pasquali et al., 1997) which could explain the age-related susceptibility of neonatal calves while supporting early observations of others that T cell subsets protect against cryptosporidiosis in mice (Ungar et al., 1991; Chen et al., 1993; McDonald et al., 1994) and humans (Flanigan et al., 1992). Correlated with

Cryptosporidium in Cattle

15

the increases in CD4+ and CD8+ IELs was the finding of elevated IFN-g and IL-12 mRNA in IELs and LPLs from the ileum of infected calves. However, no significant increases were detected in mRNA levels for IL-2, IL-4 or IL-10 (Canals et al., 1998). These cytokine and phenotypic cell differences in primary infected versus uninfected neonatal calves indicated that the susceptibility of these young calves to enteric infections had an immunological basis, the absence or extremely low levels of T cells.

What Accounts for the Variability in Severity of Infections? Although pre-weaned calves are highly susceptible, infection with different isolates of C. parvum appears to result in variation in the severity of diarrhoea and number of oocysts produced. Such variation with different isolates has been demonstrated in human infectivity studies and in calves (Pozio et al., 1992; Okhuysen et al., 1999). Clinical cryptosporidiosis could not always be induced in experimental calf infections at inoculating doses of 3.2–30 × 106 C. parvum oocysts per calf (Fayer et al., 1985). Subsequently, inoculation of calves with 1 × 104 oocysts of a bovine C. parvum isolate from Alabama, USA, consistently induced severe clinical disease in experimentally infected calves (Fayer and Ungar, 1986). Another isolate of C. parvum from Auburn, Alabama (AUCP-1) was transmitted from one Cryptosporidium-naïve calf to another (26 calves) over a period of 3 years; these infections consistently resulted in oocyst production rates of 0.3–41.5 × 106/g of faeces for 1 day or more during the patent period, accompanied by profuse diarrhoea (Fayer et al., 1997). Then, unexpectedly, over a period of the next six serial passages through calves, the severity of clinical signs and oocyst output steadily decreased until inoculation with 1.5 × 106 oocysts resulted in no diarrhoea and the recovery of fewer than 1 × 106 oocysts per calf. A similar decrease in productivity and pathogenicity with the same isolate occurred at Colorado State University. The same isolate was also serially transmitted, over a 3-year period, through 40 groups of C57BL/6 mice (four mice per group) immunosuppressed with dexamethasone. The mice showed no clinical signs but oocyst output remained consistently high. When oocysts from the mice were transmitted to a calf, the AUCP-1 isolate once again exhibited the earlier pathogenicity and oocyst production. It was not determined what factors led to the loss of pathogenicity in progressive serial passage through calves or why pathogenicity was unaffected by passage through mice; however, one might speculate that the original isolate contained multiple genetically distinct subpopulations that were selected for or against during passage through a particular host.

Selection In an attempt to use selection to obtain a non-pathogenic precocious strain of C. parvum, an experiment was designed in much the same way as the two live commercial vaccines Paracox and Livacox were selected for in poultry (Fayer, 1994). Ten calves were serially infected with oocysts collected from the previous

16

R. Fayer et al.

calf on the first day of the patent period. Instead of the expected decrease in pathogenicity, as seen with coccidia, the duration and severity of diarrhoea steadily increased from calf 1 through to calf 9, although no pattern was seen in number of oocysts shed per calf. Perhaps it was the ability of C. parvum to autoinfect, unlike the Eimeria species in the coccidiosis vaccines, that invalidated the concept of a precocious non-pathogenic immunizing strain of C. parvum. These observations of different isolates and of serial passage of specific isolates of C. parvum demonstrate that not only do differences in pathogenicity exist among field isolates but also that both the pathogenicity and fecundity of an isolate can change over time.

Double-stranded RNA Virus-like Particles Double-stranded RNA (dsRNA) virus-like particles have been found in Babesia, Trichomonas, Giardia, Leishmania and Eimeria (Hotzel et al., 1995). Two sizes of extrachromosomal dsRNAs were found in the cytoplasm of sporozoites of C. parvum and C. hominis but not in seven other species of Cryptosporidium (Khramtsov et al., 1997; Khramtsov and Upton, 2000). Sequence analysis showed distinctly different dsRNA sequences in isolates (species) from calves versus those from humans (Khramtsov et al., 2000). Small dsRNA sequences of isolates from 23 calves and 38 humans (Xiao et al., 2001) showed that isolates from the same outbreak had identical sequences; eight distinct nucleotide sequences were from cattle (C. parvum) and ten from humans (C. hominis). If any dsRNAs in Cryptosporidium are associated with pathogenicity this has not been demonstrated, but a recent study with our colleagues (M. Jenkins and J. Higgins, USDA, personal communication) has shown an association with fecundity. Calves infected with C. parvum-Beltsville oocysts produced substantially more oocysts than calves infected with C. parvum-Iowa oocysts. Increased fecundity was correlated with levels of C. parvum virus (CPV) as measured by real-time RT-PCR using CPV RNA-specific primers. The CPV signal in C. parvum-Beltsville sporozoites relative to C. parvum-Iowa was 3–4 times greater as measured by RT-PCR. The greater fluorescence intensity of C. parvum-Beltsville sporozoites labelled with antibodies to CPV 40 kDa capsid protein supported this observation. These findings suggest that CPV affects fecundity, which might in turn affect the severity of infection. What other factors affect severity of infection is not known.

Prevalence of Severe Morbidity and Mortality There is no formal reporting system for cryptosporidiosis in cattle. Reports of cases, surveys and outbreaks provide information, but not on the same basis as the public health reports of ProMed, FoodNet and MMWR. At the US Department of Agriculture in Beltsville, Maryland, telephone calls from farmers and veterinarians attributing high morbidity and mortality to cryptosporidiosis in dairy and beef calves were frequent, peaking in the late 1980s through to the

Cryptosporidium in Cattle

17

early 1990s, but steadily decreased in number until in 2006 no such calls were received. Whether the actual number of severe outbreaks declined or whether the association of severe diarrhoea and death of calves with cryptosporidiosis was no longer a ‘new’ phenomenon, and therefore consultation was no longer sought, is not known. Nevertheless, this informal barometer of activity suggests a change in the frequency of severe and economically important outbreaks perhaps due to a time when an exceptionally pathogenic strain of C. parvum was present.

Species Other than C. parvum Cryptosporidium andersoni, which colonizes the gastric glands of the abomasum, has been reviewed by Santín and Trout (2008). Infections, primarily in calves older than 4 weeks of age, produce no diarrhoea or visible clinical signs. Infections are more chronic (sometimes lasting over a year) and oocyst production is less than that of C. parvum. Elevated plasma pepsinogen and decreased milk production have been attributed to infection. For other species and genotypes infecting cattle (Table 2.1), there are no reports of clinical signs, histological data or subclinical pathology.

Molecular Identification of Species in Cattle For decades, microscopy was the sole method for detecting oocysts, first by direct faecal smears routinely stained, and later stained by IFA techniques. Microscopy Table 2.1. Species and genotypes of Cryptosporidium found in cattle and locations where they have been detected (adapted from Santín and Trout, 2008). Cryptosporidium species/genotypes C. parvum C. bovis C. andersoni C. ryanae C. hominis C. suis C. suis-like Pig genotype II C. felis C. canis

Prevalence

Geographical location

Frequent Frequent Frequent Frequent Rare Rare (2 calves) Rare (3 cattle) Rare (1 cow) Rare (1 cow) Experimental infection only

Worldwide Worldwide Worldwide Worldwide Scotland, India, Korea USA and Zambia Denmark Denmark Poland USA

18

R. Fayer et al.

was time-consuming and required highly trained personnel. Unless oocysts differed significantly in size, species identification was impossible. PCR was found to be more sensitive than IFA and, when combined with gene sequencing, could differentiate species and genotypes of Cryptosporidium. The most frequently used gene for identification has been the SSU rRNA (small subunit ribosomal RNA) gene (the 18S gene). Molecular analysis of Cryptosporidium from cattle has identified seven species and three genotypes: C. parvum, C. andersoni, C. bovis, C. canis, C. felis, C. hominis, C. suis, C. suis-like genotype, C. ryanae, and Cryptosporidium pig genotype II (Bornay-Llinares et al., 1999; Fayer et al., 2001, 2006; Santín et al., 2004; Smith et al., 2005; Geurden et al., 2006; Park et al., 2006; Starkey et al., 2006; Feng et al., 2007; Langkjær et al., 2007) (Table 2.1).

Prevalence of Species in Cattle In large-scale studies of cattle, C. parvum, C. andersoni, C. bovis and C. ryanae were found most frequently (Santín et al., 2004; Xiao et al., 2004; Fayer et al., 2006; Geurden et al., 2006; Feng et al., 2007; Langkjær et al., 2007). Other species and genotypes were found infrequently or rarely. Crytosporidium hominis was found in a few cattle in Scotland, India and Korea (Smith et al., 2005; Park et al., 2006; Feng et al., 2007), C. suis was found in one calf in the USA and another in Zambia (Fayer et al., 2006; Geurden et al., 2006), C. suis-like genotype and Cryptosporidium pig genotype II were found in three cattle and one cow in Denmark, respectively (Langkjær et al., 2007), C. felis was found in a cow in Poland (Bornay-Llinares et al., 1999), and C. canis infected calves experimentally but natural infections have not been detected (Fayer et al., 2001) (Table 2.1).

Large-scale Cross-sectional Study and Longitudinal Study of Species Prevalence Related to Age of Cattle A large-scale study involving 1615 cattle was conducted over a period of 4 consecutive years (Santín et al., 2004; Fayer et al., 2006, 2007). Each year 15 commercial dairy farms were included, two or three from each of seven states, ranging from Vermont to Florida along the east coast of the USA. Each year an attempt was made to collect faeces from 30 or more cattle from each farm. During the first 2 years, calves from 5 days to 11 months of age were examined. During the third year, dairy heifers 1–2 years of age were examined. During the fourth year, cows over 2 years of age were examined. Most of the farms visited in the first year were visited again in the following years, but in a few cases, because of management practices, either the required age or the required number of cattle were not available and substitute farms were introduced into the study. Based on SSU rRNA gene sequencing from PCR-positive specimens, virtually all infections in pre-weaned calves (1500 bp (complete or almost complete) is desirable for comparative purposes; COWP1 – oocyst wall protein with type I and type II cysteine-rich repeats, oocyst EB module wall protein (Templeton et al., 2004); HSP70 – 70 kDa heat shock protein, cytosolic form; & – draft genome sequence of C. pestis IOWA (Abrahamsen et al. 2004, AAEE01000000) has five SSU rDNA (some are partial); for the purpose of this table chromosome 7: cgd7_7 sequence serves as reference sequence; LeBlancq et al. (1997) identified the total of five copies and two slightly different types of rDNA units per genome of C. pestis KSU-1: referred to as Type A (AF015772) and Type B (AF308600); && – draft genome sequence of C. hominis TU502 (Xu et al. 2004, AAEL01000000) has six SSU rDNAs (some partial); for the purpose of this table chromosome 7: Chro.rrn016 and chromosome 2: Chro.rrn022 as reference sequences; * – on chromosome 2, cgd2_1375 (299599–301356) is currently erroneously annotated as SSU rDNA, a different non-annotated region on the same contig AAEE01000005 is the true SSU rDNA; ** – cgd8_5425 (1155551–1156706); here several additional nucleotides are added as of the contig; *** – the sequence is concatenated from two overlapping partial sequences; # – the sequence from chromosome 3 (Kim et al., 1992) but genome projects of C. pestis Iowa and C. hominis TU502 both identified this gene on chromosome 5 (Abrahamsen et al., 2004; Xu et al., 2004); ## – the sequence reported by Khramtsov et al. (1995); see also mRNA sequence U69698 – unpublished, GenBank.

43

small subunit rDNA C. pestis IOWA (1749 nt) C. hominis TU502 (1753 nt)

200

150

100

50

0 0

5 -2

0

40

70 60 50 40 30 20 10 0

actin C. pestis IOWA (1131 nt) C. hominis TU502 (1131 nt)

70 kDa heat shock protein C. pestis IOWA (2022 nt) C. hominis TU502 (2034 nt)

1 50

0

51

0 10

0

0

5 12

01

0 15

51

0

01

5 17

0

0 20

51

30 25 20 15 10 5 0 0

0

80

7 12 15 10 17 Sequence length category (nt)

0

5 -7

35

5 -2

oocyst wall protein 1 C. pestis IOWA (4872 nt) C. hominis TU502 (4863 nt)

0 0 0 0 0 0 0 0 0 25 -50 -75 100 125 150 175 200 500 1 01 1- 1- 1- 1- 1- 15 5 75 00 25 50 75 00 2 1 1 2 1 1 Sequence length category (nt)

0

0 -5

1 25

Number of sequences in the length category

J. Šlapeta

0

0

0 -5

1 25

1 50

5 -7

0

0

51

0 10

01

5 12

7 10 Sequence length category (nt)

Number of sequences in the length category

Number of sequences in the length category

Number of sequences in the length category

44

0-

40 35 30 25 20 15 10 5 0

0 0 0 0 0 0 0 0 0 25 -50 -75 100 125 150 175 200 225 1 01 1- 1- 1- 1- 1- 15 2 5 75 00 25 50 75 00 2 1 1 1 1 Sequence length category (nt)

0-

Fig. 4.3. Representation SSU rDNA, COWP1, actin and HSP70 sequences available in primary nucleotide databases (NCBI/EMBL, DDBJ). Individual sequences were categorized according to their length. The full length of individual genes from C. pestis and C. hominis draft genomes is given above the graph.

region by Johnson et al. (1995), CPB-DIAGF and CPB-DIAGR primers or COWP1 by Spano et al. (1997), and Cry-15 and Cry-9 primers (Fig. 4.3). Many of the sequence entries for HSP70 and COWP1 are short in sequence length, in contrast to actin, where many species are almost fully sequenced. Moreover, many of the currently recognized genotypes are difficult to examine further due to the generation of only partial sequences by investigators.

Centenary of the Genus Cryptosporidium

45

It should be stressed that every effort should be made to generate full-length sequences for already existing species, genotypes or new genotypes being described. These sequences represent a permanent record upon which an investigation for biological characters can be based. To avoid the confusion that arises from the rapidly expanding complexity of species names and number of sequences generated in different laboratories since the late 1990s, a website has been developed (‘Taxonomy of the genus Cryptosporidium’ at http://www.vetsci.usyd.edu.au/staff/JanSlapeta/). This website aims to provide a common gateway for analysing Cryptosporidium species and to standardize the terminology of individual isolates. Alignments of the reference sequences from the named species are provided, with further detailed annotation of individual species. Furthermore, currently under development is a webbased analysis (BLAST, multiple sequence alignment, tree building) as well as an in-house annotation database of sequences to streamline and standardize the process for the identification of Cryptosporidium isolates.

Will the Real Cryptosporidium parvum Please Stand Up? When reviewing the large amount of information on the human and bovine derived isolates, it became apparent that the mouse species, C. parvum Tyzzer 1912, described and illustrated by Tyzzer (1912), is different from the commonly encountered species affecting humans and cattle (Šlapeta, 2006). Ernest E. Tyzzer described the parasite from the intestine of tame mice in his Harvard laboratory (Tyzzer, 1912). The species was easily transmissible to mice and was present in great numbers in the intestines of affected mice. In the absence of the type material and with a lack of data about the host specificity of Tyzzer’s original material, we need to subjectively align the available data with the current information. Two studies provided noteworthy details on Cryptosporidium spp. in mice (Mus musculus). The first study, by Klesius et al. (1986), determined up to 30% incidence of cryptosporidiosis in mice at a calving site and susceptibility of calves to some of these oocysts from mice faeces. In the second study, Morgan et al. (1999) genotyped many samples of Cryptosporidium spp. from mice obtained worldwide and genetically characterized a specific ‘mouse genotype’ using several independent nuclear loci distinct from known genotypes. This ‘mouse genotype’ was genetically different from the ‘bovine’ and ‘human’ genotypes previously identified in Australia and Maryland, USA, and lumped under the umbrella name C. parvum (Xiao et al., 2004). Thus, the genotyping of mice isolates provided additional evidence for the existence of the ‘bovine genotype’ in wild mice (Morgan et al., 1999). Consequently one can speculate that it was this ‘bovine genotype’ that was infectious for calves in the study of Klesius et al. (1986). This ‘bovine genotype’ essentially infects only neonatal mice and only affects adult mice transiently (Korich et al., 2000), unlike authentic rodent C. parvum isolates that produce heavy infections in adult laboratory mice (Bednarska et al., 2003). No experimental calf infections were conducted using the true C. parvum ‘mouse genotype’. Despite this fact, cattle and human isolates of Cryptosporidium

46

J. Šlapeta

spp. have attracted considerable attention in recent years, but no ‘mouse genotype’ (C. parvum) has been identified (Xiao et al., 2002; Santín et al., 2004). Therefore, C. parvum (C. parvum ‘mouse genotype’) appears to be host-specific, because either there are no natural calf or human infections or they represent an extremely rare case. It is generally agreed that the species that Tyzzer used for his description is indeed the ‘mouse genotype’ (Šlapeta, 2007; Xiao et al., 2007). An urgent step towards the stability of historical information is to apply Tyzzer’s name to the species he used for the description and to use appropriate names for the two dominant human species. The above findings led to the recent taxonomic treatment and support for the identity of the ‘mouse genotype’ with the original description, type host, experimental transmission and illustrations of C. parvum (Šlapeta, 2006). Cryptosporidium parvum sensu Tyzzer, 1912 is a host-adapted species of Mus musculus with no documented capacity to infect either humans or domestic cattle.

Consensus, Reversibility and Universality Two species which can by supported by morphological characteristics and by localization and development in the host are C. muris and C. parvum. The larger C. muris is found in the stomach, whereas the intestine is the location of the smaller C. parvum. The status quo of only these two species affecting mammals is retained by Upton and Current (1985), and is suggested to be practical unless further evidence shows otherwise. However, a diverse spectrum of hosts and distinct DNA has provided further evidence of multiple lineages with different evolutionary history, and thus provides only preliminary evidence for a species. Older names synonymized under the umbrella names C. parvum and C. muris have been re-erected as full species (i.e. C. felis, C. wrairi). Those for which no name was thought to be available and whose identification was based on DNA sequencing have acquired ‘genotype’ status or have been named as new species. However, mixing the latter with the former has introduced inconsistency to what is C. parvum. Indeed, the name C. parvum has lost its purpose; it is no longer a unique identifier. Genotypes now seem to be much more useful descriptors, i.e. human genotype, mouse genotype, canine genotype, snake genotype, bovine genotype, etc. Nevertheless, this is all work in progress. To reinstate some stability to some of the major and clinically important genotypes, backed up by epidemiological data and experimental information, some were logically named as new species. Thus, the status quo of Upton and Current (1985) was abandoned and the following consensus was practically applied to cattle species. The abomasum is parasitized by C. andersoni and at least two named species are recognized to affect the intestine of cattle, the hostspecific C. bovis and the zoonotic C. pestis (Šlapeta, 2006). This approach logically aligns with the recognition of C. hominis which, along with C. pestis, were formerly classified as human and bovine genotypes of C. parvum.

Centenary of the Genus Cryptosporidium

47

Nevertheless this opinion is capable of reverting to the former conservative opinion, thus justifying the continuity of the names according to the ICZN, i.e. grouping C. hominis, C. parvum and C. pestis under the oldest available name, which is C. parvum, thereby reinstating the terms ‘human genotype’, ‘mouse genotype’ and ‘bovine genotype’, respectively. Underlying genetic differences have the potential to be used to distinguish between multiple taxa. While morphologically it is hard to find any differences, and host specificity provides only a partial picture, the use of DNA provides a virtual but very specific and repeatable label for each individual. If even a minor genetic difference is detected then it may reflect the niche adaptation, and so should be viewed as a practical tag for identification. Hence, those populations causing significant disease will stand out from the crowd and ultimately acquire a unique name. Indeed, the name C. hominis is now used for the dominant species transmitted from human to human, unequivocally typified by the draft genome of TU502 (Xu et al., 2004). Similarly, C. pestis is the name for the already known zoonotic species for which the draft genome of the Iowa strain is available (Abrahamsen et al., 2004). With the acceptance of both C. pestis and C. hominis we are clarifying the identity of the medically and veterinary important species. The synopsis for the three species is as follows: Cryptosporidium parvum Tyzzer, 1912 Syn. Cryptosporidium parvum ‘mouse genotype’ Xiao et al. (2002, 2004) Type host: Laboratory mouse, Mus musculus Type locality: Harvard Laboratory, Massachusetts, USA Cryptosporidium hominis Morgan-Ryan, Fall, Ward, Hijjawi, Sulaiman, Fayer, Thompson, Olson, Lal & Xiao, 2002 Syn. Cryptosporidium parvum ‘human genotype’ Xiao et al. (2002) Type host: Human, Homo sapiens Type locality: Perth, Western Australia, Australia Cryptosporidium pestis Šlapeta, 2006 Syn. Cryptosporidium parvum ‘bovine genotype’ Xiao et al. (2002, 2004) Type host: Domestic cattle, Bos taurus Type locality: Iowa, USA

Future Challenges Historically, two morphologically distinct populations of parasites of the gastrointestinal tract were originally described by Ernest E. Tyzzer from mice (i.e. Cryptosporidium muris and C. parvum). Nowadays, DNA sequences are increasingly seen as primary information sources for species identification in many organism groups, including Cryptosporidium. Such approaches stand on the implicit assumption that the reference databases used for comparison are sufficiently complete and feature-rich, with annotated entries. However, the uncertain taxonomic reliability and lack of annotations in public DNA repositories form a major obstacle to sequence-based species identification. The current taxonomic

48

J. Šlapeta

expansion of the genus Cryptosporidium is important relative to the dominant host, pathogenicity and genetic diversity. Unfortunately only few taxonomically available and named species have so far been molecularly characterized which, however, does not invalidate them from scientific usage. On the other hand, data from molecular epidemiological studies constantly reveal new genetic variants (= genotypes), but the lack of sufficient annotation and additional biological characteristics prohibits their elevation to species and ultimately taxonomic recognition. Nevertheless, descriptions of new genotypes remain an essential part of our understanding of the host specificity, the diversity and the epidemiology of the genus Cryptosporidium. In addition to the poor annotation of sequences in public DNA repositories, a huge gap exists between the number of described names and the number of identified genotypes. The closure of this gap represents a prime challenge for the decades to come.

Acknowledgements I would like to thank the organizers of the Second International Giardia and Cryptosporidium Conference (II IGCC) for their invitation to present the paper which forms the basis for this chapter. Financial support for the preparation of this article and participating in the II IGCC was partly provided by the Faculty of Veterinary Science, University of Sydney. I also thank Professor John Ellis (University of Technology Sydney) for editorial suggestions. I apologize to those authors whose work could not be cited owing to space limitations.

References Abrahamsen, M.S., Templeton, T.J., Enomoto, S., Abrahante, J.E., Zhu, G., Lancto, C.A., Deng, M., Liu, C., Widmer, G., Tzipori, S., Buck, G.A., Xu, P., Bankier, A.T., Dear, P.H., Konfortov, B.A., Spriggs, H.F., Iyer, L., Anantharaman, V., Aravind, L. and Kapur, V. (2004) Complete genome sequence of the apicomplexan, Cryptosporidium parvum. Science 304, 441–445. Awad-el-Kariem, F.M., Robinson, H.A., Dyson, D.A., Evans, D., Wright, S., Fox, M.T. and McDonald, V. (1995) Differentiation between human and animal strains of Cryptosporidium parvum using isoenzyme typing. Parasitology 110, 129–132. Barker, I.K. and Carbonell, P.L. (1974) Cryptosporidium agni sp.n. from lambs, and Cryptosporidium bovis sp.n. from a calf, with observations on the oocyst. Zeitschrift für Parasitenkunde 44, 289–298. Bednarska, M., Bajer, A., Kulis, K. and Sinski, E. (2003) Biological characterisation of Cryptosporidium parvum isolates of wildlife rodents in Poland. Annals of Agricultural and Environmental Medicine 10, 163–169. Bird, R.G. and Smith, M.D. (1980) Cryptosporidiosis in man: parasite life cycle and fine structural pathology. Journal of Pathology 132, 217–233. Casemore, D.P., Sands, R.L. and Curry, A. (1985) Cryptosporidium species a 'new' human pathogen. Journal of Clinical Pathology 38, 1321–1336. Current, W.L. and Reese, N.C. (1986) A comparison of endogenous development of three isolates of Cryptosporidium in suckling mice. Journal of Protozoology 33, 98–108.

Centenary of the Genus Cryptosporidium

49

Duszynski, D.W. (1999) Revisiting the code: clarifying name-bearing types for photomicrographs of protozoa. Journal of Parasitology 85, 769–770. Fall, A., Thompson, R.C.A., Hobbs, R.P. and Morgan-Ryan, U.M. (2003) Morphology is not a reliable tool for delineating species within Cryptosporidium. Journal of Parasitology 89, 399–402. Fayer, R. and Ungar, B.L. (1986) Cryptosporidium and cryptosporidiosis. Microbiological Reviews 50, 458–483. Fayer, R., Santín, M. and Xiao, L. (2005) Cryptosporidium bovis n. sp. (Apicomplexa: Cryptosporidiidae) in cattle (Bos taurus). Journal of Parasitology 91, 624–629. Harris, D. (2003) Can you bank on GenBank? Trends in Ecology and Evolution 18, 317–319. International Commission on Zoological Nomenclature (1999) International Code of Zoological Nomenclature (ICZN), London, 306 pp. Available at: http://www.iczn.org/ iczn/index.jsp Iseki, M. (1979) Cryptosporidium felis sp. n. (Protozoa: Eimeriorina) from the domestic cat. Japanese Journal of Parasitology 28, 285–307. Johnson, D.W., Pieniazek, N.J., Griffin, D.W., Misener, L. and Rose, J.B. (1995) Development of a PCR protocol for sensitive detection of Cryptosporidium in water samples. Applied and Environmental Microbiology 61, 3849–3855. Khramtsov, N.V., Tilley, M., Blunt, D.S., Montelone, B.A. and Upton, S.J. (1995) Cloning and analysis of a Cryptosporidium parvum gene encoding a protein with homology to cytoplasmic form Hsp70. Journal of Eukaryotic Microbiology 42, 416–422. Kim, K., Gooze, L., Petersen, C., Gut, J. and Nelson, R.G. (1992) Isolation, sequence and molecular karyotype analysis of the actin gene of Cryptosporidium parvum. Molecular and Biochemical Parasitology 50, 105–113. Klesius, P.H., Haynes, T.B. and Malo, L.K. (1986) Infectivity of Cryptosporidium sp. isolated from wild mice for calves and mice. Journal of the American Veterinary Medical Association 189, 192–193. Korich, D.G., Marshall, M.M., Smith, H.V., O’Grady, J., Bukhari, Z., Fricker, C.R., Rosen, J.P. and Clancy, J.L. (2000) Inter-laboratory comparison of the CD-1 neonatal mouse logistic dose-response model for Cryptosporidium parvum oocysts. Journal of Eukaryotic Microbiology 47, 294–298. LeBlancq, S.M., Khramtsov, N.V., Zamani, F., Upton, S.J. and Wu, T.U. (1997) Ribosomal RNA gene organization in Cryptosporidium parvum. Molecular and Biochemical Parasitology 90, 463–478. Levine, N.D. (1984) Taxonomy and review of the coccidian genus Cryptosporidium (Protozoa, Apicomplexa). Journal of Protozoology 31, 94–98. Morgan, U.M., Constantine, C.C., O’Donoghue, P., Meloni, B.P., O’Brien, P.A. and Thompson, R.C. (1995) Molecular characterization of Cryptosporidium isolates from humans and other animals using random amplified polymorphic DNA analysis. American Journal of Tropical Medicine and Hygiene 52, 559–564. Morgan, U.M., Sturdee, A.P., Singleton, G., Gomez, M.S., Gracenea, M., Torres, J., Hamilton, S.G., Woodside, D.P. and Thompson, R.C. (1999) The Cryptosporidium “mouse” genotype is conserved across geographic areas. Journal of Clinical Microbiology 37, 1302–1305. Morrison, D.A. (2006) Phylogenetic analyses of parasites in the new millennium. Advances in Parasitology 63, 1–124. Nilsson, R., Ryberg, M., Kristiansson, E., Abarenkov, K., Larsson, K. and Koljalg, U. (2006) Taxonomic reliability of DNA sequences in public sequence databases: a fungal perspective. PLoS ONE 1, e59.

50

J. Šlapeta Ryan, U.M., Xiao, L., Read, C., Zhou, L., Lal, A.A. and Pavlasek, I. (2003) Identification of novel Cryptosporidium genotypes from the Czech Republic. Applied and Environmental Microbiology 69, 4302–4307. Santín, M., Trout, J.M., Xiao, L., Zhou, L., Greiner, E. and Fayer, R. (2004) Prevalence and age-related variation of Cryptosporidium species and genotypes in dairy calves. Veterinary Parasitology 122, 103–117. Šlapeta, J. (2006) Cryptosporidium species found in cattle: a proposal for a new species. Trends in Parasitology 22, 469–474. Šlapeta, J. (2007) Response to Xiao et al.: Further debate on the description of Cryptosporidium pestis. Trends in Parasitology 23, 42–43. Spano, F., Putignani, L., McLauchlin, J., Casemore, D.P. and Crisanti, A. (1997) PCRRFLP analysis of the Cryptosporidium oocyst wall protein (COWP) gene discriminates between C. wrairi and C. parvum, and between C. parvum isolates of human and animal origin. FEMS Microbiology Letters 150, 209–217. Templeton, T.J., Lancto, C.A., Vigdorovich, V., Liu, C., London, N.R., Hadsall, K.Z. and Abrahamsen, M.S. (2004) The Cryptosporidium oocyst wall protein is a member of a multigene family and has a homolog in Toxoplasma. Infection and Immunity 72, 980–987. Tilley, M., Upton, S.J. and Freed, P.S. (1990) A comparative study on the biology of Cryptosporidium serpentis and Cryptosporidium parvum (Apicomplexa: Cryptosporidiidae). Journal of Zoo and Wildlife Medicine 21, 463–467. Tyzzer, E.E. (1907) A sporozoan found in the peptic glands of the common mouse. Proceedings of the Society for Experimental Biology and Medicine 5, 12–13. Tyzzer, E.E. (1910) An extracellular coccidium, Cryptosporidium muris (gen. et sp. nov.) of the gastric glands of the common mouse. Journal of Medical Research 23, 487–509. Tyzzer, E.E. (1912) Cryptosporidium parvum (sp. nov.), a coccidium found in the small intestine of the common mouse. Archiv fur Protistenkunde 26, 394–412. Tzipori, S., Angus, K.W., Campbell, I. and Gray, E.W. (1980) Cryptosporidium: evidence for a single-species genus. Infection and Immunity 30, 884–886. Upton, S.J. and Current, W.L. (1985) The species of Cryptosporidium (Apicomplexa: Cryptosporidiidae) infecting mammals. Journal of Parasitology 71, 625–629. Vetterling, J.M., Jervis, H.R., Merrill, T.G. and Sprinz, H. (1971) Cryptosporidium wrairi sp. n. from the guinea pig Cavia porcellus, with an emendation of the genus. Journal of Protozoology 18, 243–247. Xiao, L., Sulaiman, I.M., Ryan, U.M., Zhou, L., Atwill, E.R., Tischler, M.L., Zhang, X., Fayer, R., Lal, A.A. (2002) Host adaptation and host–parasite co-evolution in Cryptosporidium: implications for taxonomy and public health. International Journal for Parasitology 32, 1773–1785. Xiao, L., Fayer, R., Ryan, U. and Upton, S.J. (2004) Cryptosporidium taxonomy: recent advances and implications for public health. Clinical Microbiological Reviews 17, 72–97. Xiao, L., Fayer, R., Ryan, U. and Upton, S.J. (2007) Response to the newly proposed species Cryptosporidium pestis. Trends in Parasitology 23, 41–42. Xu, P., Widmer, G., Wang, Y., Ozaki, L.S., Alves, J.M., Serrano, M.G., Puiu, D., Manque, P., Akiyoshi, D., Mackey, A.J., Pearson, W.R., Dear, P.H., Bankier, A.T., Peterson, D.L., Abrahamsen, M.S., Kapur, V., Tzipori, S. and Buck, G.A. (2004) The genome of Cryptosporidium hominis. Nature 431, 1107–1112.

5

Molecular Epidemiology of Human Cryptosporidiosis in Developing Countries L. XIAO Centers for Disease Control and Prevention, Atlanta, GA, USA

Abstract Genotyping and subtyping tools have been used to characterize the transmission of human cryptosporidiosis in developing countries. Thus far, five Cryptosporidium spp. – C. hominis, C. parvum, C. meleagridis, C. canis and C. felis – are responsible for most Cryptosporidium infections in both immunocompetent and immunocompromised individuals. In most areas, C. hominis is responsible for over 70% of human cryptosporidiosis cases, with C. parvum accounting for 10–20% of infections. Differences have been observed among endemic areas in the proportion of infections due to each species, and in some areas C. meleagridis is also endemic. Results of subtyping suggest that there is high genetic heterogeneity in C. hominis in developing countries, and in these areas, human infections with C. parvum and other species are mostly the result of anthroponotic rather than zoonotic transmission. There is geographical segregation in C. hominis or C. parvum subtypes. Mixed and sequential infections with different Cryptosporidium species/genotypes and subtypes are common. Differences in oocyst shedding and clinical presentation have been observed among Cryptosporidium species and C. hominis subtypes. These findings reveal the diversity of cryptosporidiosis transmission in endemic areas and highlight the need for more extensive studies of cryptosporidiosis epidemiology in diverse areas with a wide spectrum of socioeconomic and environmental conditions.

Introduction Cryptosporidiosis is prevalent in developing countries, perhaps due to the high intensity of environmental contamination and poor hygiene conditions. Children in these highly endemic areas develop cryptosporidiosis very early in life, with peak infections usually occurring before 2 years of age. There is usually a strong association between the occurrence of cryptosporidiosis and the rainy season in tropical areas, or the cooler months in dry areas. These countries often have less intensive animal husbandry. Thus, the transmission of cryptosporidiosis in developing countries is probably different from that in the industrialized nations. Even though most molecular epidemiological studies of cryptosporidiosis have © CAB International 2009. Giardia and Cryptosporidium: From Molecules to Disease (eds G. Ortega-Pierres et al.)

51

52

L. Xiao

been conducted in the industrialized nations, an increasing number of studies are being done in developing countries. This has significantly improved our understanding of cryptosporidiosis transmission in these areas. This includes a better knowledge of parasite diversity, the roles of various transmission routes in cryptosporidiosis epidemiology, and the significance of parasite genetics in pathogenesis and clinical presentations. These developments have enabled health officials to better educate the public about the risk factors involved in the acquisition of cryptosporidiosis in vulnerable populations.

Molecular Epidemiological Tools Genotyping tools Many genotyping tools have been used to differentiate between Cryptosporidium species in humans. The PCR primers of earlier tools were mostly based on antigenic, structural, housekeeping genes and unknown genomic fragments of C. parvum, and included various formats of detection and differentiation (Xiao and Ryan, 2004; Cacciò, 2005). With few exceptions, most of these techniques can efficiently differentiate C. parvum, C. hominis and perhaps C. meleagridis, but are unlikely to amplify some of the more distant species (such as C. canis, C. felis, C. muris and C. andersoni). Therefore, these genotyping tools are mostly replaced by the genus-specific PCR-RFLP techniques based on the SSU rRNA gene, which have higher sensitivity and allow broad species detection.

Subtyping tools Subtyping tools are increasingly used in epidemiological studies of C. parvum and C. hominis. Several types of genetic targets are used in the development of subtyping tools, including microsatellites and minisatellites, double-stranded (ds) RNA elements, and ITS-2. In developing countries, the most widely used subtyping tool is DNA sequencing of the 60 kDa glycoprotein (GP60) gene. The GP60 gene is similar to a microsatellite sequence by having tandem repeats of the serine-coding trinucleotide TCA/TCG/TCT at the 5′ end of the gene. However, in addition to variations in the number of trinucleotide repeats, there are extensive sequence differences in the non-repeat regions, which categorize C. parvum and C. hominis each to several subtype families (alleles). Some of the common subtype families are Ia, Ib, Id, Ie and If for C. hominis and IIa, IIc, IId and IIe for C. parvum (Fig. 5.1). Members of different subtype families differ from each other extensively in the primary sequences. Within each subtype family, subtypes differ from each other mostly in the number of trinucleotide repeats TCA, TCG and TCT (mostly seen in Ie). It should be kept in mind that GP60 and other subtyping tools (including the MST and MLST tools described below) do not amplify the DNA of C. felis, C. canis and other species distant from C. parvum and C. hominis.

Molecular Epidemiology of Human Cryptosporidiosis in Developing Countries 100

IIi-Akiyoshi-154

100

0.1 substitutions/site

53

Id-AF164497 I-Akiyoshi-90

100

70

II-Kenya-9589R IIe-7870

82

II-Alpaca-13987 IIj-NI-8934 IIf-7490 I-Peru9911

100 98

Ia-AF164502 Ig-NI3986

99

IId-4833 IIb-4736

99 100

IIg-Akiyoshi-66

100

FelisAY700394.1| 99

IIa-4804: Zoonotic C. parvum Ib-4746

100

Ie-5632 67

If-4755

55 IIh-Akiyoshi-112

52

II-kenya-9408R

67

IIc-4742: Anthroponotic C. parvum III-DQ067570

100 98

III-4500 III-295

Fig. 5.1. Phylogenetic relationships between the known subtype families of C. hominis (Ia, Ib, Id, Ie, If, etc.) and C. parvum (IIa, IIc, IId, IIe, etc.) based on a neighbour-joining analysis of the GP60 gene sequences. Sequences starting with III were derived from C. meleagridis. Two major anthroponotic and zoonotic C. parvum subtype families (shown in boxes) are identified.

MLT and MLST The recent genome sequencing of C. parvum and C. hominis has allowed the identification of microsatellite and minisatellite sequences in C. parvum and C. hominis genomes and other targets that are highly polymorphic between C. parvum and C. hominis. They are frequently used in multilocus analysis to increase the typing resolution. Two types of techniques are used in typing. In multilocus typing (MLT), variation in microsatellites and minisatellites is assessed on the basis of length variations, using polyacrylamide gel electrophoresis or the GeneScan technology (Ngouanesavanh et al., 2006). This allows the use of

54

L. Xiao

many targets in the MLT techniques economically. The second kind of typing technique, multilocus sequence typing (MLST), relies on the detection of genetic heterogeneity by DNA sequencing of the amplified PCR products (Gatei et al., 2006a, 2007). Compared to MST, MLST allows the detection of nucleotide substitution and the inclusion of markers with single nucleotide polymorphisms.

Proper use of molecular epidemiological tools Caution should be exercised in interpreting typing results, as accurate diagnosis of species and subtypes has important implications for understanding infection sources. In two studies in India, Cryptosporidium mouse genotype was identified in one child and one HIV-positive adult, based on results of RFLP analysis of the SSU rRNA PCR products (Muthusamy et al., 2006; Ajjampur et al., 2007). The true identity of the parasite was probably C. meleagridis, which has the same RFLP pattern as the mouse genotype in the genotyping technique used. The detection of C. andersoni in a Malawian child was also largely based on RFLP analysis (Morse et al., 2007), which needs to be confirmed by DNA sequencing, as DraI and AseI RFLP was mainly used in species differentiation and C. muris may also share the same pattern. The use of RFLP instead of DNA sequencing in differentiating GP60 subtypes should be avoided, as some subtype families share high sequence homology in one region but not in other regions of the gene, and many new subtype families have been found since the initial development of the RFLP differentiation tool. The sequence mosaic in the gene is also responsible for the polyphyletic nature of both C. parvum and C. hominis in the gene (Fig. 5.1). Thus, the identification of IIa and If subtype families in humans in southern India (Muthusamy et al., 2006) needs to be confirmed by the results of DNA sequencing. The subtype family name Ic is still occasionally used by some researchers to describe C. parvum (Muthusamy et al., 2006; Ajjampur et al., 2007), even though this was the result of an initial misidentification of species at the SSU rRNA locus (Strong et al., 2000), and the subtype family was renamed as IIc some time ago (Alves et al., 2003). IIc should be used to avoid unnecessary confusion. In one study (Ajjampur et al., 2007), the so-called Cryptosporidium mouse genotype GP60 sequence was probably from C. meleagridis, and the C. felis GP60 sequences were probably not from C. felis.

Cryptosporidium Species and Genotypes in Humans In a similar way as for people living in the industrialized nations, five Cryptosporidium spp. are responsible for most human Cryptosporidium infections in developing countries; including C. hominis, C. parvum, C. meleagridis, C. canis and C. felis. They were initially found in otherwise healthy children in Peru in a longitudinal cohort study using a SSU rRNA-based genotyping tool (Xiao et al., 2001), but have also recently been found in diarrhoeic children in Kenya (Gatei et al., 2006b). One large-scale study of Ugandan children reported only

Molecular Epidemiology of Human Cryptosporidiosis in Developing Countries

55

Table 5.1. Distribution of common human-pathogenic Cryptosporidium species in children in developing countries. No. of isolate C. hominis C. parvum C. meleagridis C. felis C. canis Reference

Location China India India Kenya Malawi Malawi Uganda Iran Guatemala Lima, Peru Chile

5 50 58

5 47 47

0 0 7

175 37 43 444

153 35 25 326

15 2 10 85

7

3

4

15 85 4

14 67 2

1 8 2

1?

1 3

1

2

3

1

2

2 5

7

Peng et al. (2001) Gatei et al. (2007) Ajjampur et al. (2007) Gatei et al. (2006b) Peng et al. (2003) Morse et al. (2007) Tumwine et al. (2003) Meamar et al. (2007) Xiao et al. (2004) Xiao et al. (2001) Neira-Otero et al. (2005)

C. hominis, C. parvum and C. meleagridis, but the genotyping tool used is not capable of detecting C. canis, C. felis and other species genetically distant from C. parvum and C. hominis (Tumwine et al., 2003). Other studies examined only small numbers of specimens, which was probably responsible for the low species diversity detected (Table 5.1). These five Cryptosporidium spp. have also been found in immunocompromised people in developing countries (Table 5.2). Studies conducted in Thailand, India and Peru showed the presence of these five species in HIV-positive adults (Gatei et al., 2002b; Cama et al., 2003). Small-scale studies in other areas have also revealed the presence of some of the five species in HIV-positive adults or children (Table 5.2). In Lima, Peru, there is no significant difference in the distribution of the five species between children and HIV-positive adults (Xiao et al., 2001; Cama et al., 2003). It is likely that other Cryptosporidium species can infect humans under certain circumstances. Cryptosporidium muris has been found in human cases in Indonesia (Katsumata et al., 1998), Kenya (Gatei et al., 2002a, 2006b), Peru (Palmer et al., 2003) and India (Muthusamy et al., 2006). Other Cryptosporidium species found in humans in developing countries include C. suis in one HIVpositive adult and Cryptosporidium cervine genotype in a child, both in Lima, Peru (Xiao et al., 2002; V. Cama et al., unpublished). New Cryptosporidium genotypes will probably be found in humans in future, even though these parasites account for a very small proportion of Cryptosporidium infections in humans.

56

Table 5.2.

Distribution of common human-pathogenic Cryptosporidium species in HIV-positive people in developing countries.

Location

Age

Vellore, India Bangkok, Thailand

Adults 4 children, 25 adults Adults Adults Adults Children Children Children Adults Adults Mostly adults Adults Adults

Bangkok, Thailand Taiwan Kenya Malawi Ugandaa South Africa Iran Venezuela Colombia Haiti Lima, Peru a Including

No. of isolate C. hominis C. parvum C. meleagridis C. felis C. canis Reference 48 29

31 24

9

1 + 1 (?) 3

5 1

34 4 24 6 76 21 8 10 6 49 302

17 2 14 6 56 16 1 8 3 31 204

5

7 1 1

3 1

8 14 5 7 1 2 16 34

Muthusamy et al. (2006) Tiangtip and Jongwutiwes (2002) 2

3

1 1? 38

1 1? 10

12

Gatei et al. (2002b) Hung et al. (2007) Gatei et al. (2003) Peng et al. (2003) Tumwine et al. (2005) Leav et al. (2002) Meamar et al. (2007) Certad et al. (2006) Navarro-i-Martinez et al. (2006) Ngouanesavanh et al. (2006) Cama et al. (2003)

nine HIV-negative children.

L. Xiao

Molecular Epidemiology of Human Cryptosporidiosis in Developing Countries

57

Cryptosporidium hominis Infection In developing countries, more than 70% of Cryptosporidium infections in children are caused by C. hominis in most of the areas studied (Table 5.1). In these areas, children acquire cryptosporidiosis very early in life, with peak incidence of infections occurring between 1 and 2 years of age. Under such conditions, the transmission of cryptosporidiosis is probably mostly through a direct person-to-person route rather than an indirect waterborne or foodborne route. This is probably responsible for the high proportion of infections attributable to C. hominis. This predominance of C. hominis in humans in developing countries applies also to HIV-infected adults (Table 5.2). A diarrhoea outbreak caused by C. hominis also occurred in a daycare centre in São Paulo, Brazil (Goncalves et al., 2006). Molecular characterizations of C. hominis isolates have revealed the complexity of cryptosporidiosis epidemiology. Despite the seemingly lower diversity of Cryptosporidium in humans at the species/genotype level in developing countries, the results of GP60 subtyping have revealed the complexity of Cryptosporidium transmission in endemic areas. This is evident from the existence of many C. hominis subtype families in each endemic area. Thus, 3–4 C. hominis subtype families were seen in humans in India, Peru, Kenya, Malawi and South Africa (Leav et al., 2002; Peng et al., 2003; Xiao et al., 2004; Ajjampur et al., 2007; Cama et al., 2007; Gatei et al., 2007). In these areas, the complexity of transmission is frequently reflected by the existence of many subtypes within C. hominis subtype families Ia and Id in one endemic area (Leav et al., 2002; Peng et al., 2003; Cama et al., 2007; Gatei et al., 2007). The high C. hominis heterogeneity in developing countries is probably an indicator of intensive and stable cryptosporidiosis transmission in the area. Four common C. hominis subtype families – Ia, Ib, Id and Ie – have been found in humans in many developing countries (Xiao et al., 2004) (Fig. 5.1). Nevertheless, there are geographical differences in their distribution. For example, Ia, Ib and Id are common in Malawi, South Africa, India and Peru (Leav et al., 2002; Peng et al., 2003; Ajjampur et al., 2007; Cama et al., 2007; Gatei et al., 2007). Children in South Africa were commonly infected with If (erroneously named as Ie in the publication), a subtype family not seen in most other studies (Leav et al., 2002). In contrast, the common subtype family Ie was not seen in cryptosporidiosis in South African children (Leav et al., 2002). Two space–time clusters of Ia infections were identified in children in southern India (Ajjampur et al., 2007), but it is unclear whether each cluster was caused by one Ia subtype, as many subtypes of the Ia subtype family are usually present in an area (see above). Within each subtype family, one subtype is frequently seen in certain areas but not in others. For example, there are only two common subtypes within C. hominis subtype family Ib: IbA9G3 and IbA10G2. The former is commonly seen in Malawi, Kenya and India, whereas the latter is commonly seen in South Africa, Botswana, Jamaica and Peru (Leav et al., 2002; Cama et al., 2007; Gatei et al., 2006a, 2007; L. Xiao et al., unpublished). This geographical segregation of C. hominis subtypes becomes much more obvious when specimens from different countries are genetically compared using a MLST tool (Gatei et al., 2006a).

58

L. Xiao

Cryptosporidium parvum Infection Most developing countries have very low frequency of C. parvum infections, although the burden attributable to this species seems to vary by area. In these countries, C. parvum transmission in humans is frequently anthroponotic. This is supported by results of GP60 subtyping in several regions. Two major GP60 subtype families, the zoonotic IIa and the anthroponotic IIc, are responsible for most C. parvum infections in humans (Fig. 5.1). IIa subtypes are rarely seen in humans in developing countries. Instead, the subtype family IIc is responsible for most human C. parvum infections in these areas (Leav et al., 2002; Peng et al., 2003; Xiao et al., 2004; Akiyoshi et al., 2006; Ajjampur et al., 2007). In some places, such as Lima, Peru and southern India, the IIc subtype family is the only C. parvum in humans, whereas in other developing countries such as Malawi and Kenya, IIe (another anthroponotic C. parvum subtype family) is seen in humans in addition to IIc (Peng et al., 2003; Xiao et al., 2004; Ajjampur et al., 2007; Cama et al., 2007). In Uganda, even though IIc subtypes are the dominant C. parvum in children, several new subtype families have been reported (Akiyoshi et al., 2006). All these are likely indicators of the reduced role of zoonotic transmission of C. parvum in these areas. Nevertheless, a recent study of asymptomatic cryptosporidiosis in Zambian dairy farm workers and their household members has identified C. parvum as being more common than C. hominis (Siwila et al., 2007). Zoonotic infection was also suspected to be one cause of the seemingly wider species diversity in Malawian children living in rural areas (Morse et al., 2007). No subtyping of C. parvum was done in these studies to identify possible infection sources. One major outbreak of diarrhoea largely attributed to cryptosporidiosis occurred in Botswana in early 2006 after heavy rain, causing thousands of infections and several hundreds of deaths in a number of districts. Both C. parvum and C. hominis were identified in infected people, with the former responsible for more cases. As expected, five C. hominis subtype families and six subtypes were identified in a small number of specimens analysed. However, the C. parvum specimens analysed had only IIcA5G3a or IIcA5G3b subtypes of the IIc subtype family (L. Xiao et al., unpublished). Because C. parvum was identified as the major species and some patients in the outbreak had enteropathogenic E. coli, cattle waste was initially suspected as the major source of contamination. The finding of only the C. parvum subtype IIc family and the presence of C. hominis clearly suggests that the major initial source of contamination was human sewage.

Infections with Other Species/Genotypes In most developing countries studied, C. parvum and C. hominis are responsible for more than 90% of human cases of cryptosporidiosis, with the remainder attributable to C. meleagridis, C. canis and C. felis. Some areas, however, have a high prevalence of these unusual species. In Lima (Peru) and Bangkok (Thailand), C. meleagridis is as prevalent in humans as C. parvum, being responsible for

Molecular Epidemiology of Human Cryptosporidiosis in Developing Countries

59

10–20% of human cryptosporidiosis cases (Xiao et al., 2001; Gatei et al., 2002b; Cama et al., 2003). Subtyping tools are largely unavailable for these rare species because most C. parvum- and C. hominis-based subtyping primers are not capable of PCR amplification in these species. GP60 and HSP70-based typing tools, however, are available to subtype C. meleagridis. Multiple C. meleagridis subtypes were seen in the small number of human specimens from several different areas (Glaberman et al., 2001). A recent MLST analysis in a small community in Lima, Peru, has shown the presence of multiple subtypes in children, AIDS patients, and birds, with no apparent host segregation of the subtypes found (L. Xiao et al., unpublished). Like C. parvum, most human infections with the usual species are probably the result of anthroponotic transmission, judged by the concurrent presence of C. hominis in some of the C. meleagridis-, C. canis- or C. felis-infected individuals (see below). Possible transmission of C. canis between two siblings and a dog occurred in a household in a slum in Lima, Peru, but the direction of the transmission was not clear (Xiao et al., 2007).

Mixed Infections and Sequential Infections No matter which PCR tool is used in genotyping Cryptosporidium, all broadly specific tools have the problem that they detect only the dominant genotype in the specimen because of the inherent nature of exponential amplification by PCR and the requirement of a substantial amount of PCR product for them to be visible on an agarose gel. Thus, concurrent infection with mixed Cryptosporidium genotypes is more challenging to diagnose and minor populations of species or genotypes in specimens are probably under-detected by these tools (Reed et al., 2002). A low prevalence of mixed infections of multiple Cryptosporidium genotypes has been reported in children in Uganda (19/444), Malawi (1/43) and India (2/50) (Tumwine et al., 2003; Gatei et al., 2007; Morse et al., 2007). Using genotype-specific primers in combination with other genus-specific primers, Cama et al. (2006) identified concurrent C. hominis, C. parvum or C. meleagridis infections in seven out of 21 C. canis- or C. felis-infected HIV-positive people. Seven out of 55 C. meleagridis-infected children and HIV-positive people in Lima, Peru, had concurrent C. hominis infection in a MLST analysis (L. Xiao et al., unpublished). Thus, the accurate identification of infections with mixed genotypes has important implications for understanding the transmission of so-called zoonotic species and genotypes in humans. In developing countries, children frequently have multiple episode of cryptosporidiosis, although the likelihood of clinical illness decreases with increased infection episodes (Bern et al., 2002). Molecular analysis of longitudinal samples from Peruvian children with multiple cryptosporidiosis episodes indicated that immunity against both homologous and heterologous Cryptosporidium species/ genotypes was short-lived, with time intervals between infections of about 1 year. As might be expected, sequential infections with heterologous Cryptosporidium species were more common than sequential infections with homologous

60

L. Xiao

Cryptosporidium species (Xiao et al., 2001, Cama et al., 2008). Even though some children had the same genotype (C. hominis) in sequential infections, results of subtyping suggested that these were in fact due to heterogeneous subtypes of the parasite (Cama et al., 2008).

Cryptosporidium Species/Genotypes, Subtypes, and Virulence The clinical and epidemiological significance of various Cryptosporidium species and subtypes in humans is not yet clear. Results of recent genotyping studies nevertheless support the theory that C. hominis and C. parvum behave differently in humans. Studies in slums in Peru and Brazil have shown that children infected with C. hominis have higher oocyst shedding intensity and longer duration than those infected with C. parvum and other genotypes (Xiao et al., 2001; Bushen et al., 2007). Children infected with C. hominis had a significantly greater severity of diarrhoea than those infected with other species in southern India (Ajjampur et al., 2007). In AIDS patients in Lima, Peru, only infections with C. canis, C. felis or subtype family Id of C. hominis were significantly associated with diarrhoea, and infections with C. parvum were significantly associated with chronic diarrhoea and vomiting. Infections with C. hominis Ib subtype family were also marginally associated with diarrhoea and vomiting. In contrast, infections with C. meleagridis and Ia and Ie subtype families of C. hominis were usually asymptomatic (Cama et al., 2007). These results demonstrate that different Cryptosporidium genotypes and subtype families are linked to different clinical manifestations. Because of the differences in pathogenicity, C. hominis and C. parvum seemingly have different nutritional effects on infected children. In Brazil, heightfor-age (HAZ) Z-scores showed significant declines within 3 months of infection for children infected with either C. hominis or C. parvum. However, in the 3–6 month period following infection, only C. hominis-infected children continued to demonstrate declining HAZ score and those with asymptomatic infection showed even greater decline (P = 0.01). Thus, C. hominis is associated with greater growth shortfalls, even in the absence of symptoms (Bushen et al., 2007).

Conclusions Molecular epidemiological studies of cryptosporidiosis in developing countries are still in their infancy, but significant progress has already been made towards a better understanding of the transmission of the infection. We are beginning to use second-generation molecular tools to answer epidemiological questions that are difficult to address by traditional methods, such as maintenance of immunity and cross-protection, transmission dynamics in different settings, temporal and geographical variations in Cryptosporidium transmission, and the role of parasite factors in the variability in transmission and clinical spectrum of cryptosporidiosis. More importantly, we are starting to see better collaboration between epidemiologists, clinicians, molecular biologists and parasitologists in well-designed

Molecular Epidemiology of Human Cryptosporidiosis in Developing Countries

61

epidemiological studies. Such an integrated approach will undoubtedly lead to better utilization of the available molecular diagnostic tools and a better understanding of the epidemiology of cryptosporidiosis.

Acknowledgements The findings and conclusions reported in this chapter are those of the author and do not necessarily represent the views of the Centers for Disease Control and Prevention.

References Ajjampur, S.S.R., Gladstone, B.P., Selvapandian, D., Muliyil, J.P., Ward, H. and Kang, G. (2007) Molecular and spatial epidemiology of cryptosporidiosis in children in a semiurban community in South India. Journal of Clinical Microbiology 45, 915–920. Akiyoshi, D.E., Tumwine, J.K., Bakeera-Kitaka, S. and Tzipori, S. (2006) Subtype analysis of Cryptosporidium isolates from children in Uganda. Journal of Parasitology 92, 1097–1100. Alves, M., Xiao, L., Sulaiman, I., Lal, A.A., Matos, O. and Antunes, F. (2003) Subgenotype analysis of Cryptosporidium isolates from humans, cattle, and zoo ruminants in Portugal. Journal of Clinical Microbiology 41, 2744–2747. Bern, C., Ortega, Y., Checkley, W., Roberts, J.M., Lescano, A.G., Cabrera, L., Verastegui, M., Black, R.E., Sterling, C. and Gilman, R.H. (2002) Epidemiologic differences between cyclosporiasis and cryptosporidiosis in Peruvian children. Emerging Infectious Diseases 8, 581–585. Bushen, O.Y., Kohli, A., Pinkerton, R.C., Dupnik, K., Newman, R.D., Sears, C.L., Fayer, R., Lima, A.A. and Guerrant, R.L. (2007) Heavy cryptosporidial infections in children in northeast Brazil: comparison of Cryptosporidium hominis and Cryptosporidium parvum. Transactions of the Royal Society of Tropical Medicine and Hygiene 101, 378–384. Cacciò, S.M. (2005) Molecular epidemiology of human cryptosporidiosis. Parassitologia 47, 185–192. Cama, V.A., Bern, C., Sulaiman, I.M., Gilman, R.H., Ticona, E., Vivar, A., Kawai, V., Vargas, D., Zhou, L. and Xiao, L. (2003) Cryptosporidium species and genotypes in HIV-positive patients in Lima, Peru. Journal of Eukaryotic Microbiology 50 (Suppl.), 531–533. Cama, V., Gilman, R.H., Vivar, A., Ticona, E., Ortega, Y., Bern, C. and Xiao, L. (2006) Mixed Cryptosporidium infections and HIV. Emerging Infectious Diseases 12, 1025–1028. Cama, V.A., Ross, J.M., Crawford, S., Kawai, V., Chavez-Valdez, R., Vargas, D., Vivar, A., Ticona, E., Ñavincopa, M., Williamson, J., Ortega, Y., Gilman, R.H., Bern, C. and Xiao, L. (2007) Differences in clinical manifestations among Cryptosporidium species and subtypes in HIV-infected persons. Journal of Infectious Diseases 196, 684–691. Cama, V.A., Bern, C., Roberts, J., Cabrera, L., Sterling, C.R., Ortega, Y., Gilman, R.H. and Xiao, L. (2008) Cryptosporidium species and subtypes and clinical manifestations in children, Peru. Emerging Infectious Diseases 14, 1567–7154.

62

L. Xiao Certad, G., Ngouanesavanh, T., Hernan, A., Rojas, E., Contreras, R., Pocaterra, L., Nunez, L., Dei-Cas, E. and Guyot, K. (2006) First molecular data on cryptosporidiosis in Venezuela. Journal of Eukaryotic Microbiology 53, S30–S32. Gatei, W., Ashford, R.W., Beeching, N.J., Kamwati, S.K., Greensill, J. and Hart, C.A. (2002a) Cryptosporidium muris infection in an HIV-infected adult, Kenya. Emerging Infectious Diseases 8, 204–206. Gatei, W., Suputtamongkol, Y., Waywa, D., Ashford, R.W., Bailey, J.W., Greensill, J., Beeching, N.J. and Hart, C.A. (2002b) Zoonotic species of Cryptosporidium are as prevalent as the anthroponotic in HIV-infected patients in Thailand. Annals of Tropical and Medical Parasitology 96, 797–802. Gatei, W., Greensill, J., Ashford, R.W., Cuevas, L.E., Parry, C.M., Cunliffe, N.A., Beeching, N.J. and Hart, C.A. (2003) Molecular analysis of the 18S rRNA gene of Cryptosporidium parasites from patients with or without human immunodeficiency virus infections living in Kenya, Malawi, Brazil, the United Kingdom, and Vietnam. Journal of Clinical Microbiology 41, 1458–1462. Gatei, W., Hart, C.A., Gilman, R.H., Das, P., Cama, V. and Xiao, L. (2006a) Development of a multilocus sequence typing tool for Cryptosporidium hominis. Journal of Eukaryotic Microbiology 53 (Suppl. 1), S43–S48. Gatei, W., Wamae, C.N., Mbae, C., Waruru, A., Mulinge, E., Waithera, T., Gatika, S.M., Kamwati, S.K., Revathi, G. and Hart, C.A. (2006b) Cryptosporidiosis: prevalence, genotype analysis, and symptoms associated with infections in children in Kenya. American Journal of Tropical Medicine and Hygiene 75, 78–82. Gatei, W., Das, P., Dutta, P., Sen, A., Cama, V., Lal, A.A. and Xiao, L. (2007) Multilocus sequence typing and genetic structure of Cryptosporidium hominis from children in Kolkata, India. Infection Genetics and Evolution 7, 197–205. Glaberman, S., Sulaiman, I.M., Bern, C., Limor, J., Peng, M.M., Morgan, U., Gilman, R., Lal, A.A. and Xiao, L. (2001) A multilocus genotypic analysis of Cryptosporidium meleagridis. Journal of Eukaryotic Microbiology 48 (Suppl.), 19S–22S. Goncalves, E.M., da Silva, A.J., Eduardo, M.B., Uemura, I.H., Moura, I.N., Castilho, V.L. and Corbett, C.E. (2006) Multilocus genotyping of Cryptosporidium hominis associated with diarrhea outbreak in a day care unit in São Paulo. Clinics (São Paulo, Brazil) 61, 119–126. Hung, C.C., Tsaihong, J.C., Lee, Y.T., Deng, H.Y., Hsiao, W.H., Chang, S.Y., Chang, S.C. and Su, K.E. (2007) Prevalence of intestinal infection due to Cryptosporidium species among Taiwanese patients with human immunodeficiency virus infection. Journal of the Formosan Medical Association 106, 31–35. Katsumata, T., Hosea, D., Wasito, E.B., Kohno, S., Hara, K., Soeparto, P. and Ranuh, I.G. (1998) Cryptosporidiosis in Indonesia: a hospital-based study and a communitybased survey. American Journal of Tropical Medicine and Hygiene 59, 628–632. Leav, B.A., Mackay, M.R., Anyanwu, A., O’Connor, R.M., Cevallos, A.M., Kindra, G., Rollins, N.C., Bennish, M.L., Nelson, R.G. and Ward, H.D. (2002) Analysis of sequence diversity at the highly polymorphic Cpgp40/15 locus among Cryptosporidium isolates from human immunodeficiency virus-infected children in South Africa. Infection and Immunity 70, 3881–3890. Meamar, A.R., Guyot, K., Certad, G., Dei-Cas, E., Mohraz, M., Mohebali, M., Mohammad, K., Mehbod, A.A., Rezaie, S. and Rezaian, M. (2007) Molecular characterization of Cryptosporidium isolates from humans and animals in Iran. Applied and Environmental Microbiology 73, 1033–1035. Morse, T.D., Nichols, R.A., Grimason, A.M., Campbell, B.M., Tembo, K.C. and Smith, H.V. (2007) Incidence of cryptosporidiosis species in paediatric patients in Malawi. Epidemiology and Infection 135, 1307–1315.

Molecular Epidemiology of Human Cryptosporidiosis in Developing Countries

63

Muthusamy, D., Rao, S.S., Ramani, S., Monica, B., Banerjee, I., Abraham, O.C., Mathai, D.C., Primrose, B., Muliyil, J., Wanke, C.A., Ward, H.D. and Kang, G. (2006) Multilocus genotyping of Cryptosporidium sp. isolates from human immunodeficiency virus-infected individuals in South India. Journal of Clinical Microbiology 44, 632–634. Navarro-i-Martinez, L., da Silva, A.J., Garces, G., Montoya Palacio, M.N., del Aguila, C. and Bornay-Llinares, F.J. (2006) Cryptosporidiosis in HIV-positive patients from Medellin, Colombia. Journal of Eukaryotic Microbiology 53 (Suppl.), S37–S39. Neira-Otero, P., Munoz-Saldias, N., Sanchez-Moreno, M. and Rosales-Lombardo, M.J. (2005) Molecular characterization of Cryptosporidium species and genotypes in Chile. Parasitology Research 97, 63–67. Ngouanesavanh, T., Guyot, K., Certad, G., Fichoux, Y.L., Chartier, C., Verdier, R.I., Cailliez, J.C., Camus, D., Dei-Cas, E. and Banuls, A.L. (2006) Cryptosporidium population genetics: evidence of clonality in isolates from France and Haiti. Journal of Eukaryotic Microbiology 53, S33–S36. Palmer, C.J., Xiao, L., Terashima, A., Guerra, H., Gotuzzo, E., Saldias, G., Bonilla, J.A., Zhou, L., Lindquist, A. and Upton, S.J. (2003) Cryptosporidium muris, a rodent pathogen, recovered from a human in Peru. Emerging Infectious Diseases 9, 1174–1176. Peng, M.M., Matos, O., Gatei, W., Das, P., Stantic-Pavlinic, M., Bern, C., Sulaiman, I.M., Glaberman, S., Lal, A.A. and Xiao, L. (2001) A comparison of Cryptosporidium subgenotypes from several geographic regions. Journal of Eukaryotic Microbiology 48 (Suppl.), 28S–31S. Peng, M.M., Meshnick, S.R., Cunliffe, N.A., Thindwa, B.D., Hart, C.A., Broadhead, R.L. and Xiao, L. (2003) Molecular epidemiology of cryptosporidiosis in children in Malawi. Journal of Eukaryotic Microbiology 50 (Suppl.), 557–559. Reed, C., Sturbaum, G.D., Hoover, P.J. and Sterling, C.R. (2002) Cryptosporidium parvum mixed genotypes detected by PCR-restriction fragment length polymorphism analysis. Applied and Environmental Microbiology 68, 427–429. Siwila, J., Phiri, I.G., Vercruysse, J., Goma, F., Gabriel, S., Claerebout, E. and Geurden, T. (2007) Asymptomatic cryptosporidiosis in Zambian dairy farm workers and their household members. Transactions of the Royal Society of Tropical Medicine and Hygiene 101, 733–734. Strong, W.B., Gut, J. and Nelson, R.G. (2000) Cloning and sequence analysis of a highly polymorphic Cryptosporidium parvum gene encoding a 60-kilodalton glycoprotein and characterization of its 15- and 45-kilodalton zoite surface antigen products. Infection and Immunity 68, 4117–4134. Tiangtip, R. and Jongwutiwes, S. (2002) Molecular analysis of Cryptosporidium species isolated from HIV-infected patients in Thailand. Tropical Medicine and International Health 7, 357–364. Tumwine, J.K., Kekitiinwa, A., Nabukeera, N., Akiyoshi, D.E., Rich, S.M., Widmer, G., Feng, X. and Tzipori, S. (2003) Cryptosporidium parvum in children with diarrhea in Mulago Hospital, Kampala, Uganda. American Journal of Tropical Medicine and Hygiene 68, 710–715. Tumwine, J.K., Kekitiinwa, A., Bakeera-Kitaka, S., Ndeezi, G., Downing, R., Feng, X., Akiyoshi, D.E. and Tzipori, S. (2005) Cryptosporidiosis and microsporidiosis in Ugandan children with persistent diarrhea with and without concurrent infection with the human immunodeficiency virus. American Journal of Tropical Medicine and Hygiene 73, 921–925. Xiao, L. and Ryan, U.M. (2004) Cryptosporidiosis: an update in molecular epidemiology. Current Opinion in Infectious Diseases 17, 483–490.

64

L. Xiao Xiao, L., Bern, C., Limor, J., Sulaiman, I., Roberts, J., Checkley, W., Cabrera, L., Gilman, R.H. and Lal, A.A. (2001) Identification of 5 types of Cryptosporidium parasites in children in Lima, Peru. Journal of Infectious Diseases 183, 492–497. Xiao, L., Bern, C., Arrowood, M., Sulaiman, I., Zhou, L., Kawai, V., Vivar, A., Lal, A.A. and Gilman, R.H. (2002) Identification of the Cryptosporidium pig genotype in a human patient. Journal of Infectious Diseases 185, 1846–1848. Xiao, L., Fayer, R., Ryan, U. and Upton, S.J. (2004) Cryptosporidium taxonomy: recent advances and implications for public health. Clinical Microbiological Reviews 17, 72–97. Xiao, L., Cama, V.A., Cabrera, L., Ortega, Y., Pearson, J. and Gilman, R.H. (2007) Possible transmission of Cryptosporidium canis among children and a dog in a household. Journal of Clinical Microbiology 45, 2014–2016.

6

Molecular Epidemiology and Typing of Non-human Isolates of Cryptosporidium U.M. RYAN1 AND L. XIAO2 1Murdoch

University, WA, Australia; 2Center for Disease Control and Prevention, Atlanta, GA, USA

Abstract Cryptosporidium has been reported in a wide variety of hosts, with C. parvum and C. hominis being responsible for most human infections. Until recently, it has been assumed that farm animals and wild animals are important zoonotic reservoirs for human cryptosporidiosis. However, recent molecular analysis has revealed a wide range of Cryptosporidium species and genotypes infecting both domestic and wild animals, and the epidemiology of cryptosporidiosis is clearly more complicated than was previously thought.

Introduction Cryptosporidium has been reported in a wide variety of vertebrate hosts (Fayer et al., 2000) and, at present, 16 species of Cryptosporidium are regarded as valid on the basis of differences in oocyst morphology, site of infection, vertebrate class specificity and genetic differences: C. muris in rodents; C. andersoni and C. bovis in cattle and sheep; C. suis in pigs; C. parvum in cattle, humans and other mammals; C. meleagridis in birds and humans; C. hominis in humans; C. baileyi and C. galli in birds; C. serpentis and C. saurophilum in snakes and lizards; C. molnari and C. scophthalmi in fish; C. wrairi from guinea pigs; C. felis in cats; and C. canis in dogs (Fayer et al., 2000, 2001; Alvarez-Pellitero and Sitjà-Bobadilla, 2002; Morgan-Ryan et al., 2002; Ryan et al., 2003a, 2004a; Alvarez-Pellitero et al., 2004). Cryptosporidium parvum and C. hominis are responsible for most human infections (Morgan-Ryan et al., 2002), and it has been assumed that the majority of Cryptosporidium infections in farmed animals that had oocysts in the size range of 4–6 mm were due to C. parvum (cattle genotype) and that farm animals and wild animals represented an important zoonotic reservoir for human cryptosporidiosis. However, recent molecular analysis has revealed a wide range of Cryptosporidium species and genotypes infecting both domestic and wild animals, © CAB International 2009. Giardia and Cryptosporidium: From Molecules to Disease (eds G. Ortega-Pierres et al.)

65

66

U.M. Ryan and L. Xiao

and the epidemiology of cryptosporidiosis is clearly more complicated than was previously thought.

Cryptosporidium Species and Genotypes in Sheep Cryptosporidium has been reported in sheep worldwide; however, most studies on Cryptosporidium in sheep have been based on microscopy, with the reported prevalence ranging from 2.6% to 82% (see Ryan et al., 2005). Recent molecular characterization of Cryptosporidium in sheep suggests that the majority of sheep are not infected with zoonotic genotypes. In Australia, a survey of 1647 sheep faecal samples reported a prevalence of 26.2%. The most common Cryptosporidium genotypes identified were the cervid genotype (55%) and C. bovis (23%). Low levels of C. andersoni, C. suis, marsupial genotype, pig II genotype, C. hominis and a novel previously unidentified genotype were also detected (Ryan et al., 2005). Cryptosporidium parvum was not detected in any of the samples. More recently, in the USA, sequence analysis of 57 specimens corresponding to 8 ewes and 24 lambs that were positive for Cryptosporidium identified the cervid genotype (84%), C. bovis-like genotype (~12%) and C. parvum (3.5%) (Santín et al., 2007). In the UK, C. parvum and a novel genotype were reported in sheep (Chalmers et al., 2002); however, RFLP analysis of the COWP locus was used (Chalmers et al., 2002) and recent studies have shown that this analysis can erroneously identify isolates as C. parvum due to the extent of conservation at the COWP locus (Hamnes et al., 2007).

Cryptosporidium Species and Genotypes in Pigs Cryptosporidium was first reported in pigs in the USA in 1977 (Kennedy et al., 1977). Subsequent studies have been undertaken worldwide, including Australia, Japan, Germany, Spain and Ireland, with prevalences ranging from 1.4% to 100% in pigs aged 1 week to adulthood (Quílez et al., 1996; Izumiyama et al., 2001; Wieler et al., 2001; Maddox-Hyttel et al., 2006; Xiao et al., 2006). The prevalence in piglets under 2 months of age has been reported to range between 0% and 59.2%. The age at which Cryptosporidium is most prevalent is in pigs between 6 and 12 weeks old (see Hamnes et al., 2007). The majority of nursing piglets and weanlings that have been positive for Cryptosporidium have been shown to be infected with two Cryptosporidium spp. – C. suis and pig genotype II (Guselle et al., 2003; Ryan et al., 2003a, 2004a; Vitovec et al., 2006; Xiao et al., 2006; Hamnes et al., 2007; Langkjær et al., 2007) (see Table 6.1). Cryptosporidium suis is adapted to porcine hosts, but poorly infective for cattle (Enemark et al., 2003) and not infective for nude mice and neonatal BALB/c mice (Morgan et al., 1999a; Ryan et al., 2003a; Vitovec et al., 2006). Cryptosporidium suis has been identified in humans (Xiao et al., 2002a) but is not a common human pathogen. Little is known about the pathogenicity or zoonotic potential of pig genotype II but it has not been reported in humans to date and is unlikely to be a major threat to public health. Phylogenetic analysis

Molecular Epidemiology and Typing of Non-human Isolates Table 6.1.

67

Cryptosporidium genotypes infecting pigs.

Country Denmark Weaners Piglets Norway Australia

No. genotyped

C. suis

Pig genotype II

170 13 9 12

24% 71% ~66% 60%

76% 29% ~33%

Other species

Reference Langkjær et al. (2007)

C. parvum (40%)

Australia Weaners Piglets Australia Weaners Piglets Ireland

14 14

50% 50%

50% 50%

27 18 25

24% 100% 57%

76% 0% 39%

Czech Republic

13

100%

Hamnes et al. (2007) Morgan et al. (1999a) Ryan et al. (2003a)

Johnson et al. (2008)

C. muris (3.5%)

Xiao et al. (2006) Vitovec et al. (2006)

at multiple loci has confirmed the species status of C. suis and provides strong evidence that pig genotype II is also a valid species (see Ryan et al., 2003a; Xiao et al., 2004a). In Denmark, sequence analysis of 183 Cryptosporidium-positive pig isolates from sows, weaners and piglets from various herds identified C. suis in 24% of weaners and 71% of piglets, and pig genotype II in 76% of weaners and 29% of piglets (Langkjær et al., 2007). Higher oocyst concentrations were observed in samples genotyped as C. suis than in samples genotyped as pig genotype II. There also appeared to be an age-related change in the species/genotypes infecting pigs, as pig genotype II was more prevalent in weaners whereas the majority of piglets were infected with C. suis (Langkjær et al., 2007). This is in contrast with a previous Australian study in which 28 Cryptosporidium isolates from pigs were genotyped, revealing equal numbers of C. suis and pig genotype II (Ryan et al., 2003a). No correlation between genotype and host age was found, but some pigs infected with pig genotype II seemed to have a high excretion of oocysts (Ryan et al., 2003a). However, a more recent study in Australia reported that pig genotype II was more prevalent in weaners (71%) than C. suis (17%) (Johnson et al., 2008). One of the first genotyping studies on pigs identified C. suis but also C. parvum (cattle genotype) by sequence analysis of the 18S rRNA gene in outdoor pigs suffering from diarrhoea in Western Australia (Morgan et al., 1999a). Interestingly, C. parvum has not been detected in pig isolates since then and it is likely that C. parvum is responsible for only occasional infections in pigs. In the Czech Republic, 0/135 sows, 193/3368 (5.7%) pre-weaned and 201/835 (24.1%) post-weaned piglets were positive for Cryptosporidium infection. Only C. suis was identified; however, only a few isolates were genotyped

68

U.M. Ryan and L. Xiao

(Vitovec et al., 2006). The authors reported that the C. suis strains obtained were larger (6.2 × 5.5 mm) than the oocyst sizes given in the original C. suis description (4.6 × 4.2 mm) (Ryan et al., 2004a). However, studies have shown that there is substantial morphological variation in oocyst size within individual species and that morphology alone is not a reliable indicator for delimiting species (Fall et al., 2003). In Norway, 31 (31%) herds and 57 (8.3%) litters from 684 litters of suckling piglets from 100 indoor swine herds from all regions tested positive for Cryptosporidium. Molecular characterization of nine Cryptosporidium isolates demonstrated both C. suis and Cryptosporidium pig genotype II (Hamnes et al., 2007). A study in Ireland reported that the spreading of pig slurry onto pasture or crops may lead to increases in the occurrence of Cryptosporidium in water, as C. suis, Cryptosporidium pig genotype II, and C. muris were identified in 25 of 56 pig slurry samples from 33 Irish farms (Xiao et al., 2006). The finding of C. muris in pig slurry is unusual and could have originated from either infected pigs or rodents. Even though C. muris infection has never been reported in pigs, this species does have the widest host range of all Cryptosporidium species, and this range includes humans (Iseki et al., 1989).

Cryptosporidium Species and Genotypes in Cattle Since the late 1980s, cattle have been identified as being one of the main reservoir host for the zoonotic C. parvum; however, recent studies in the USA suggest that cattle are infected with at least four Cryptosporidium parasites: C. parvum, C. bovis, C. andersoni and the Cryptosporidium deer-like genotype (Santín et al., 2004; Fayer et al., 2006, 2007; Feng et al., 2007; Xiao et al., 2007). Recent studies in the USA indicated that the occurrence of these Cryptosporidium spp. in cattle is age-related (Santín et al., 2004), as the zoonotic C. parvum was responsible for about 85% of the Cryptosporidium infections in pre-weaned calves but only 1% of the Cryptosporidium infections in post-weaned calves and heifers (Santín et al., 2004). Post-weaned calves were mostly infected with C. bovis, C. andersoni and the deer-like genotype (Santín et al., 2004) (see Table 6.2). These findings clearly demonstrate that neonatal calves are an important source of zoonotic cryptosporidiosis in humans. Neonatal calves are also the age group of cattle mostly affected by cryptosporidiosis in terms of prevalence of infection and the associated morbidity and mortality (Fayer et al., 1997). Studies on heifers aged 12–24 months of age on dairy farms in Pennsylvania, Vermont, New York, Maryland, Virginia, North Carolina and Florida reported a much lower prevalence of C. parvum and reported that C. suis, C. parvum, the deer-like genotype, C. bovis and C. andersoni accounted for 1%, 6%, 15%, 35% and 43%, respectively (Fayer et al., 2006). A study on mature dairy cattle identified an even lower prevalence of Cryptosporidium and reported that C. parvum, C. bovis and C. andersoni were found infecting 0.4%, 1.7% and 3.7% of the 541 cows, respectively (Fayer et al., 2007). The overall lower prevalence of Cryptosporidium in these cows was very highly significant (P < 0.0001) compared

Cryptosporidium genotypes infecting cattle.

Country USA Pre-weaned Post-weaned USA Heifers USA Milking cows New York state Georgia, USA Pre-weaned Post-weaned Milking cows China India Denmark Cows Older calves Young calves Hungary Calves Portugal Calves Adults aC.

No. Genotyped

C. parvum

C. bovis

C. andersoni

Deer-like genotype

278

85% 1%

9% 55%

1% 13%

5% 31%

68

6%

35%

43%

15%

31 115

6.5% 61% (70/115)

29% 37% (42/115)

23 6 3 6 12

26% 0% 0%

3 61 90

100% 4% 82%

22

95%

63 7

100% 71.5%

Other species

Reference Santín et al. (2004)

Fayer et al. (2006) 1%a Fayer et al. (2007) 64.5% 3% (3/115)

Starkey et al. (2006) Feng et al. (2007)

8.4%

39% 66% 66% 83% 91.6% – 73% 14%

22% 16% 0%

33%b

16%

Feng et al. (2007) Feng et al. (2007) Langkjær et al. (2007)

– 14% 3%

– 4%c Plutzer and Karanis (2007) Mendonça et al. (2007)

5% 14.25%

Molecular Epidemiology and Typing of Non-human Isolates

Table 6.2.

14.25%d

suis; b1 mixed C. parvum + C. andersoni isolate; cC. suis-like genotype + 3.0

Consultants (CEC) using direct spectrophotometry (Spectronic Genesys 10UV). Split samples of these waters were ozonated with an applied ozone dose of 2.4 mg/l, and UV absorbances were measured again. Pre- and post-ozone split samples of each of these turbidity levels were shipped chilled overnight to the Department of Civil and Environmental Engineering at Duke University and direct transmittance was again measured using a Cary 300 spectrophotometer. Finally, an integrated sphere device was incorporated to allow measurement of UV absorbance without interference from light scattering by particles. The results of these analyses are shown graphically in Figs 14.9 to 14.12. Figures 14.9 and 14.10 present the pre- and post-ozonated direct UVT data for Coquitlam samples 1 and 2, as measured by CEC and Duke University, and indicate the change in direct UVT as a result of ozonation across the range of turbidities tested. Generally, pre-ozone baseline UVT values were 86%, and decreased with turbidity to as low as 52% at 50 NTU. These figures indicate a significant improvement in UVT post-ozonation in both waters across the turbidity range, with UVT improving in the baseline water by as much as 8 percentage points to 94%.

Control of Cryptosporidium and Giardia in Surface Water

173

Coquitlam 1 direct UV 254% transmittance, pre- and post-ozone, two laboratories 100

CEC DUKE

UV 254%T 1 cm

90 80

Direct %T, Post-ozone

70 Direct %T, Pre-ozone

60 50 0

10

20

30

40

50

Turbidity (NTU)

Fig. 14.9. The pre- and post-ozonated direct UVT data for Coquitlam sample 1, as measured by Clancy Environmental Consultants (CEC) (direct spectrophotometry; Spectronic Genesys 10UV) and Duke University (Cary 300 spectrophotometer). Coquitlam 2 direct UV 254% transmittance, pre- and post-ozone, two laboratories

UV 254%T, 1 cm

100

CEC DUKE

90

Direct %T, Post-ozone 80 70 Direct %T, Pre-ozone 60 50 0

10

20 30 Turbidity (NTU)

40

50

Fig. 14.10. The pre- and post-ozonated direct UVT data for Coquitlam sample 2, as measured by Clancy Environmental Consultants (CEC) (direct spectrophotometry; Spectronic Genesys 10UV) and Duke University (Cary 300 spectrophotometer).

Figures 14.11 and 14.12 compare the integrated sphere (true) pre- and postozone UVT of Coquitlam 1 and 2 waters, respectively. As was noted with direct UVT measurements, ozonation was a beneficial pretreatment for UV. The advantage widens with turbidity in Coquitlam 1, but narrows in Coquitlam 2.

Discussion UV Disinfection of MS2 and Cryptosporidium The sensitivity of MS2 to UV in baseline Coquitlam water was in line with the findings of the numerous studies evaluated in USEPA’s Ultraviolet Disinfection

174

T.M. Hargy et al. Coquitlam 1 integrated sphere (true) UV 254% transmittance, pre- vs post-ozone

UV 254%T, 1 cm

100 90

True %T, Post-ozone

80 70 True %T, Pre-ozone 60 50 0

10

20

30

40

50

Turbidity (NTU)

Fig. 14.11. The integrated sphere (true) pre- and post-ozone UVT of Coquitlam 1 water. Coquitlam 2 integrated sphere (true) UV 254% transmittance, pre- vs post-ozone

UV 254%T, 1 cm

100 True %T, Post-ozone

90 80 True %T, Pre-ozone

70 60 50 0

10

20 30 Turbidity (NTU)

40

50

Fig. 14.12. The integrated sphere (true) pre- and post-ozone UVT of Coquitlam 2 water.

Guidance Manual (UVDGM) (2006). As shown in Table 14.8, all data points generated with baseline Coquitlam water fell well within the recommended limits of the UVDGM. A dose of 83 mJ/cm2 was found to achieve 4-log inactivation of MS2 seeded into the first Coquitlam sample, while 92 mJ/cm2 was required in the second test water. The difference between the curves from the two sample dates is within the range of variability typically seen between collimated beam curves, and is not thought to be indicative of any difference in treatability of the two waters. This interpretation is supported by the results of the test wherein the 4-log inactivation doses (83 and 92 mJ/cm2) were applied to elevated turbidity reactor volumes. In the case of Coquitlam 1, 4-log inactivation was achieved fairly uniformly (range: 3.8–4.1 log) in waters of 0 to 50 NTU. In the second water test, the application of 92 mJ/cm2 uniformly achieved well over 4-log inactivation (range 4.2–4.7).

Control of Cryptosporidium and Giardia in Surface Water

175

Table 14.8. Comparison of MS 2 UV dose response in Coquitlam waters with UVDGM recommendations (USEPA, 2006). UV dose (mJ/cm2) 20 40 60 80 100

Coquitlam 1

Coquitlam 2

UVDGM lower bounds

UVDGM upper bounds

1.2 2.1 3.0 3.7 4.7

1.1 2.0 2.7 3.4 4.3

0.9 1.7 2.4 3.0 3.5

1.5 2.8 4.1 5.2 6.2

This suggests that the 92 mJ/cm2 value determined in the baseline dose response curve for Coquitlam water 2 was high. This is further confirmed by the fact that these higher inactivations achieved in elevated Coquitlam 2 turbidity would be expected by a dose of about 89–100 mJ/cm2 from the first Coquitlam baseline curve. The inactivation data shown in Figs 14.4 and 14.5 indicate that there was no strong negative impact to the inactivation of coliphage MS2 seeded overnight in Coquitlam water at elevated turbidities. That slight shielding did occur is indicated by the results of UV-exposed samples that were put through a dissociation regimen prior to assay. As this process released viable MS2 from the treated matrix, increased counts per ml were realized, decreasing the log inactivation achieved. This exercise, performed on Coquitlam 2 tests from 10 to 50 NTU, demonstrated actual inactivation to be 0.1-log less in all instances. A remaining uncertainty is whether naturally occurring, naturally shielded viral pathogens might pass unaffected through a UV reactor. The low concentrations of indigenous phage and the dominance of mineral, inorganic particles would suggest this is not likely to be a concern for a UV system treating Coquitlam water. Cryptosporidium This study has corroborated the findings of Clancy et al. (2004) and Shin et al. (2001) that approximately 1 mJ/cm2 is necessary for 1-log inactivation of Cryptosporidium. These two references, along with others, form the basis for the USEPA’s inactivation table in UVDGM LT2 (USEPA, 2006), which takes data variability into account and applies safety factors in granting 1-, 2-, 3- and 4-log inactivation credits for validated UV doses of 2.5, 5.8, 12 and 22 mJ/cm2, respectively. Ozone disinfection of MS2 and Giardia The sensitivity to ozone noted here for MS2 masked any impact by Coquitlam particles, as the steepness of the inactivation kinetics exceeded the experimental resolution of CT determination. The application of a CT only 0.1 mg min/l greater than that found to be necessary for 4-log inactivation resulted in at least 5.5-log inactivation across all turbidities.

176

T.M. Hargy et al.

The ozone CT table in the Canadian Guidelines for Drinking Water Quality (Health Canada, 2005) requires CTs of 1.9 and 2.9 mg min/l at 5°C and 1°C, respectively, for 3-log ozone inactivation of Giardia cysts. Assuming a linear relationship between temperature and CT, the CT for 3-log inactivation at 3.5°C would be 2.5 mg min/l, which is roughly the effective CT determined here for 3-log inactivation in baseline Coquitlam water. The results of testing in Coquitlam waters 1 and 2 indicate that ozone is effective in Coquitlam waters at 3.5°C and across a range of turbidities. That no apparent impact of turbidity on ozone efficacy could be discerned from the results of this study is supported by the findings of Dow et al. (2006) in their investigation of ozone disinfection of relatively ozone-resistant B. subtilis spores under varying water quality conditions, including turbidity.

Effect of turbidity on ozone and UV UVT measurements indicate that ozonation of Coquitlam water with an applied dose of 2.4 mg/l will be a highly beneficial pretreatment for UV disinfection. Comparisons of direct UVT values measured in pre- and post-ozonated waters across a range of turbidities clearly indicate a strong benefit from ozone, with UVT values increasing by as much as 8 percentage points. True UVT values measured by integrated sphere also noted a significant benefit of ozonation, although this diminished with increasing turbidity in Coquitlam 2 water.

Impact of UVT measurement methods In light of the difference between direct and true UVT values, the results reported above need to be revisited. It was found that 4-log inactivation of MS2 and 3-log inactivation of Cryptosporidium could be achieved in 50 NTU water. The exposure times used were calculated incorporating the direct UV absorbance of the turbidity-adjusted water, a value that would be subject to the light scattering error. Substituting the true UVT for the direct UVT, as measured at the time of the collimated beam tests, the UV doses actually applied can be calculated, and are shown in Table 14.9. For the sake of completeness, this same assessment should be made of the UV dose applied to Cryptosporidium in elevated turbidity. The true applied dose would increase by the same proportion, but as Cryptosporidium is so sensitive to UV, the net effect of using direct UV in determining exposure time necessary for a target dose of 3.0 mJ/cm2 was to increase the actual applied dose to the 50 NTU sample up to 3.2 mJ/cm2.

Summary and Conclusions This project involved the seeding of surrogate or actual pathogenic organisms to low turbidity Coquitlam source water, determining their inactivation kinetics, and

Control of Cryptosporidium and Giardia in Surface Water

177

Table 14.9. UV doses applied to coliphage MS2 corrected for true UV absorbance.

Test water Turbidity (NTU) Baseline 5 10 15 20 50

Coquitlam 1 Targeted: 83 mJ/cm2

Coquitlam 2 Targeted: 92 mJ/cm2

True dose applied (mJ/cm2)

True dose applied (mJ/cm2)

83.0 83.0 83.0 83.0 83.6 85.3

92.0 92.0 92.5 93.3 94.0 98.5

The applied dose values obtained after making the correction for true absorbance (and UVT) are only notably different from the targeted dose for those tested at the highest turbidities.

then applying the disinfectant dose necessary for target inactivation to Coquitlam water that been adjusted with native particulate matter to a range of turbidity levels. In these latter tests, an effort was made to allow the test microorganisms to associate, by passive contact, with Coquitlam particulate matter, although no active flocculation steps were included. In addition to the evaluation of direct impacts to UV and ozone disinfection by turbidity, the ability of ozone to enhance the UV treatability of Coquitlam water was assessed as turbidity was artificially elevated. The results of this project have confirmed that GVRD has selected an effective multi-barrier approach for protecting the health of consumers of unfiltered drinking water. However, caution is recommended if any circumstances arise such that turbidity spikes due to some cause that enriches the organic rather than inorganic fraction of the suspended particles. An algal bloom or terrestrial upset, such as the immediate fallout of a forest fire, might have an effect not tested in this study. The protected nature of the watershed makes a nutrient influx fuelling an algal bloom an unlikely scenario, but a natural or human-induced forest fire is possible. Ozone was shown to be a very effective pretreatment of Coquitlam water for UV, regardless of turbidity. The UV transmittance of Coquitlam water, and thus its UV treatability, was uniformly improved by ozonation.

References Adams, M.H. (1959) Bacteriophages. Interscience Publishers, New York. Bolton, J.R. and Linden, K.G. (2003) Standardization of methods for fluence (UV dose) determination in bench-scale UV experiments. ASCE Journal of Environmental Engineering 129, 209–215. Clancy, J.L., Marshall, M.M., Hargy, T. and Korich, D.G. (2004) Susceptibility of five strains of Cryptosporidium parvum oocysts to UV light. Journal of the American Water Works Association 96, 84–93.

178

T.M. Hargy et al. Dow, S.M., Barbeau, B., von Gunten, U., Chandrakanth, M., Amy, G. and Hernandez, M. (2006) The impact of selected water quality parameters on the inactivation of Bacillus subtilis spores by monochloramine and ozone. Water Research 40, 373–382. Health Canada (2005) Canadian Guidelines for Drinking Water Quality. Available at: http://www.hc-sc.gc.ca/ewh-semt/pubs/water-eau/doc_sup-appui/protozoa/ chap_12_e.html. Hoff, J.C., Rice, E.W. and Schaefer, F.W., III (1985) Comparison of animal infectivity and excystation as measures of Giardia muris cyst inactivation by chlorine. Applied and Environmental Microbiology, 50, 1115–1117. Meng, Q.S. and Gerba, C.P. (1996) Comparative inactivation of enteric adenovirus, poliovirus and coliphages by ultraviolet irradiation. Water Research 30, 2665–2668. Shin, G.-A., Linden, K.G., Arrowood, M.J. and Sobsey, M.D. (2001) Low-pressure UV inactivation and DNA repair potential of Cryptosporidium parvum oocysts. Applied and Environmental Microbiology 67, 3029–3032. USEPA (1992) Consensus Method for Determining Groundwaters under the Direct Influence of Surface Water using Microscopic Particulate Analysis (MPA). EPA 910992029. USEPA Environmental Services Division, Port Orchard, WA. Government Printing Office, Washington, DC. USEPA (2005) Method 1623: Cryptosporidium and Giardia in Water by Filtration/IMS/ FA. EPA 815-R-05-002. United States Environmental Protection Agency, Office of Water, Washington, DC. USEPA (2006) Ultraviolet Disinfection Guidance Manual for the Final Long Term. 2. Enhanced Surface Water Treatment Rule. Office of Water (4601) EPA 815-R-06-007, United States Environmental Protection Agency, Washington, DC. Wilson, B.R., Roessler, P.F., Van Dellen, E., Abbaszadegan, M. and Gerba, C.P. (1992) Coliphage MS-2 as a UV water disinfection efficacy test surrogate for bacterial and viral pathogens. In: Proceedings of the American Water Works Association Water Quality Technology Conference, Toronto, Canada. American Water Works Association, Denver, CO, USA.

15

Towards Methods for Detecting UV-induced Damage in Individual Cryptosporidium parvum and Cryptosporidium hominis Oocysts by Immunofluorescence Microscopy

H.V. SMITH1, B.H. AL-ADHAMI1, R.A.B. NICHOLS1, J.R. KUSEL2 AND J. O’GRADY3 1Scottish

Parasite Diagnostic Laboratory, Glasgow, UK; 2Institute of Biomedical and Life Sciences, Glasgow, UK; 3Strathclyde Institute of Pharmacy and Biomedical Sciences, University of Strathclyde, UK

Abstract Water is an important transmission route for cryptosporidiosis, with at least 165 waterborne outbreaks of cryptosporidiosis documented. Cryptosporidium can be controlled in water treatment by physical removal and UV disinfection, and a method that can determine whether individual oocysts in a routine sample exposed to UV irradiation have been disinfected is of benefit as it offers increased confidence to water operators. The major effects of UV radiation on cell membranes are alterations of proteins, particularly protein crosslinking. UV-B radiation progressively inhibits protein synthesis. Specific free radical scavengers protect cells against killing and inhibition of protein synthesis by UV-B. UV light also crosslinks the complementary strands of DNA and causes the formation of single strand breaks and pyrimidine dimers. The major lesions induced are cyclobutyl pyrimidine dimers (CPDs; also known as thymine dimers, TD). UV-induced DNA lesions in living cells and in some microorganisms can be repaired by the enzyme-dependent nucleotide excision repair (NER), also named dark repair, and the light-dependent reaction known as photoreactivation (PHR). Dark repair and PHR enable UV-inactivated microorganisms to recover and may reduce the efficiency of UV inactivation. Cryptosporidium parvum oocysts are inactivated at 3–40 mJ/cm2 using medium- and low-pressure UV light. Cryptosporidium parvum can undertake photoreactivation and dark repair at the genomic level and NER repair genes have been identified in C. parvum and C. hominis. However, UV inactivation of Cryptosporidium oocysts is irreversible, despite the presence of the UV repair genes. © CAB International 2009. Giardia and Cryptosporidium: From Molecules to Disease (eds G. Ortega-Pierres et al.)

179

180

H.V. Smith et al. We investigated the following hypotheses for developing a method to demonstrate UV inactivation of Cryptosporidium oocysts: (i) UV disinfection induces the production of reactive oxygen species (ROS); (ii) UV disinfection induces apoptosis; and (iii) UV disinfection causes damage to DNA, which is detectable using fluorogenic DNA reporters. We developed assays for determining UV inactivation of Cryptosporidium oocysts which would remain compatible with, and integrated as much as possible with, existing UK and USA detection methodologies. Our development of suitable methods which can detect UV damage in individual organisms reliably and reproducibly was driven by a search for fluorogenic reporters which could enter and stain UV-killed and damaged Cryptosporidium oocysts. Antioxidants reduce ROS production and the antioxidant glutathione (GSH) plays a significant role in inhibiting the generation of mutagens by ionizing radiation. The presence of GSH, as a putative reporter of UV damage caused by the production of ROS, was investigated in intact, untreated C. parvum oocysts and sporozoites within intact UV-irradiated oocysts (40 mJ/cm2) using the fluorogenic vital dye monochlorobimane (MCB) to detect both GSH levels and activity in oocysts. MCB fluorescence localized GSH in purified, intact, recently excreted and aged C. parvum oocysts, at several nuclear and cytoplasmic sporozoite foci (n = 2–6). We did not demonstrate the function of GSH as an endogenous free radical scavenger in UV-irradiated oocysts, and other free radical scavengers are more active than GSH in UV-treated C. parvum oocysts. MCB is unlikely to be useful as a surrogate for detecting UV damage in UV-treated Cryptosporidium oocysts. The DNA intercalating dye YO-PRO1 (YP) has been used to determine apoptosis and was used to investigate the role of UV irradiation in inducing programmed cell death/ apoptosis. YP detected DNA damage in UV-treated (40 mJ/cm2) C. parvum oocysts. YP was incorporated into sporozoite DNA of intact, irradiated oocysts (possibly apoptotic) which exhibited no apparent oocyst wall damage. However, control oocysts did not exclude YP entirely. YP is unlikely to provide a reliable estimate of the possible apoptotic changes that can occur in irradiated oocysts. An antibody raised against TDs (α-TD) was used to identify changes induced by UV light in C. parvum sporozoites and oocysts, and its nuclear location was validated by co-localization with DAPI. A freeze-thawing (five cycles) procedure improved α-TD antibody labelling within irradiated C. parvum oocysts. No α-TD localization was seen in non-irradiated oocysts. Both C. parvum and C. hominis oocysts exposed to different doses of UV light (range 10–40 mJ/cm2) demonstrated TD lesions following irradiation. We conclude that an immunofluorescence assay using α-TD antibodies which, for C. parvum, has been validated against a neonatal mouse infectivity assay, is suitable for detecting thymine dimers in air-dried oocysts and air-dried sporozoites of C. parvum and C. hominis oocysts, and that the α-dsDNA antibody is a good candidate for a positive control for the assay.

Introduction Cryptosporidium oocysts are frequent contaminants of water with contributions from infected human and non-human hosts, livestock and agricultural practices, and infected feral and transport hosts (Smith and Rose, 1990, 1998; Smith et al., 1995). Recent genetic analyses have raised doubt about the validity of the current classification of the genus Cryptosporidium and reveal that more than one species of Cryptosporidium can infect susceptible, immunocompetent

Towards Methods for Detecting UV-induced Damage

181

human hosts (Table 15.1). There are 16 ‘valid’ Cryptosporidium species and a further 40+ genotypes, which differ significantly in their molecular signatures but, as yet, have not been ascribed species status (Smith et al., 2007). Seven described Cryptosporidium species (C. hominis, C. parvum, C. meleagridis, C. felis, C. canis, C. suis and C. muris) (Table 15.1) and two undescribed species of Cryptosporidium (cervine and monkey) infect immunocompetent and immunocompromised humans (Xiao et al., 2004; Cacciò et al., 2005), but C. hominis and C. parvum are the most commonly detected (Cacciò et al., 2005). Water is an important transmission route, with at least 165 waterborne outbreaks of cryptosporidiosis documented (Girdwood and Smith, 1999; Fayer et al., 2000; Slifko et al., 2000; Karanis et al., 2007). Cryptosporidium can be controlled in water treatment plants by physical removal and disinfection processes. Physical characteristics, such as size, settling velocities and surface charge affect the behaviour of oocysts in physical treatment processes (Smith et al., 1995). Essentially, oocysts behave as inert, discrete particles in water. Current data indicate that Cryptosporidium is controllable in filtration processes and the performance of individual treatment plants in removing oocysts may be expected to parallel their effectiveness for turbidity control. The major factors that influence a plant’s performance for turbidity control should be expected to affect the efficiency of removal of Cryptosporidium similarly. Cryptosporidium oocysts are insensitive to commonly used disinfectants (chlorine, chlorine dioxide, ozone, etc.), but UV disinfection is effective. Clearly, a method that can determine whether all oocysts in a routine sample exposed to UV irradiation have been disinfected is of benefit, as it offers increased confidence to water operators. Here, we focus on UV disinfection of Cryptosporidium oocysts. Oocysts occur at low densities in water (Smith and Rose, 1990, 1998; Smith et al., 1995) and methods which can detect and determine the UV sensitivity of small numbers of organisms reliably and reproducibly from water concentrates are required. Little can be inferred about the likely impact of oocysts detected in water concentrates on public health without knowing whether they are viable or not. The conventional techniques of animal infectivity and excystation in vitro are not applicable Table 15.1.

Some differences among Cryptosporidium species infecting humans.

Species C. hominis C. parvum C. suis C. felis C. canis C. meleagridis C. muris C. andersoni

Oocyst dimensions (µm)

Site of infection

Major host

4.5 × 5.5 4.5 × 5.5

Small intestine Small intestine

5.05 × 4.41 4.5 × 5.0 4.95 × 4.71 4.5–4.0 × 4.6–5.2 5.5 × 7.4 5.6 × 7.4 (5.0–6.5 × 8.1–6.0)

Small intestine Small intestine Small intestine Intestine Stomach Stomach

Humans Neonatal mammalian livestock, humans Pigs Cats Dogs Turkeys Rodents Cattle

182

H.V. Smith et al.

to the small numbers of organisms found in water concentrates, and much effort has been expended on the development of surrogate techniques that can address, accurately, the viability of individual oocysts. Fluorogenic vital dye methods are based on observing whether specific fluorogenic vital dyes are included into or excluded from Cryptosporidium oocysts as a measure of their viability.

The Usefulness of Existing Fluorogenic Vital Reporters for Determining Disinfection Capability is Compromised A variety of fluorescent dyes have been used to determine Cryptosporidium oocyst viability. In a joint UK/USA initiative to test in vitro (maximized in vitro excystation, fluorogenic vital dyes assay (4′,6-diamidino-2-phenylindole (DAPI) and propidium iodide (PI); DAPI-PI), SYTO9 and SYTO59) and in vivo (neonatal CD-1 mouse infectivity) surrogates (Clancy et al., 2000), high levels of variability were detected with both sets of surrogates for untreated, aged and disinfectant-treated controls. Variability in in vitro surrogates was probably associated with subsampling errors (enumeration of only 100 oocysts from a large oocyst stock), while variability in the in vivo surrogate was attributed to delivery of a standard dose by gavage and the use of outbred animals. In vitro assays overestimated oocyst viability compared with neonatal mouse infectivity, particularly following pulsed UV disinfection. Overall, maximized in vitro excystation and SYTO9 produced closer association with mouse infectivity than DAPI-PI. SYTO59 produced the greatest disparity between in vitro and in vivo assays (Clancy et al., 2000). In the UV treatment studies, SYTO9 and SYTO59 consistently demonstrated higher oocyst viabilities than maximized in vitro excystation and DAPI-PI, even though mouse infectivity failed to demonstrate infectious oocysts. The majority of published data indicate that while fluorogenic vital dyes can be useful predictors of oocyst viability, they also underestimate the degree of oocyst inactivation following chemical disinfection (Bukhari et al., 1999; Clancy et al., 2000; O’Grady and Smith, 2002). Thus, studies evaluating the effectiveness of chemical disinfectants using in vitro viability procedures may have reported higher than necessary disinfection requirements due to the overestimation of viability compared with in vivo procedures.

UV Light Disinfection UV light is classified according to wavelength: UV-A, 315–400 nm; UV-B, 280–315 nm; and UV-C, 100–280 nm; and has been used for drinking water disinfection since the beginning of the 20th century. UV light disinfection does not generate significant disinfection by-products. It does not cause a significant increase in assimilable organic carbon, nor does it convert nitrates to nitrites, nor bromide to bromines or bromates. UV disinfection is relatively insensitive to

Towards Methods for Detecting UV-induced Damage

183

temperature and pH differences. Disadvantages include its unsuitability for water with high levels of suspended solids, turbidity, colour, or soluble organic matter, as they can react with UV radiation, reducing disinfection performance. Turbidity reduces the water penetration of UV radiation and there is no disinfection residual.

Mode of action of UV light disinfection The modes of action of UV light on uni- and multicellular organisms involve oxidative stress resulting from the attack by free radicals of several cellular targets (proteins, DNA and lipids). They are primarily twofold, as follows: 1. The major effects of UV radiation on cell membranes are alterations of proteins, particularly protein crosslinking (detected as high molecular weight protein >200 kDa) (Kochevar, 1990). UV light can also crosslink exposed collagen, making it more resistant to enzymic digestion and less elastic (Lee et al., 2001). UV-B radiation progressively inhibits protein synthesis and kills Staphylococcus aureus. The OH- and 1O2-free radical scavengers protect cells against killing and inhibition of protein synthesis by UV-B, suggesting that such radicals mediate the effects of UV-B on this organism. A similar protective effect using a ferric ion chelator suggests an important role for metallic ions in UV-B lethality (El-Adhami et al., 1994). Additional effects of pulsed UV light damage include increased concentration of eluted proteins, increased cell membrane damage, and structural changes in yeast cells (Takeshita et al., 2003). 2. UV light crosslinks the complementary strands of DNA. It also causes the formation of single strand breaks and pyrimidine dimers (Takeshita et al., 2003). DNA conformation and/or flexibility governs the phenomenon of crosslinking. (GA).(TC) suppresses the crosslink formation in DNA more than any dinucleotide composed of only G and C. (CTAG).(CTAG) promotes crosslinking much more than any other tetranucleotide, including (TATA).(TATA), whereas the closely related (CATG).(CATG) belongs among the tetranucleotides that most suppress the UV-light-induced crosslinks between the complementary strands of DNA (Nejedly et al., 2001a). Nejedly et al. (2001b) also identified that the (ATTTTATA).(TATAAAAT) octamer is a candidate for the hotspot of UV-light-induced crosslinking between the complementary strands of DNA. UV light damages DNA, and the major lesions induced are cyclobutyl pyrimidine dimers (CPDs; also known as thymine dimers, TD) (Mitchell, 1988). The UV-induced DNA base modifications lead to the production of reactive oxygen species (ROS), which can damage cellular elements. The antioxidant glutathione (GSH) plays a significant role in inhibiting the generation of ROS (Fischer-Nielsen et al., 1994). UV-induced DNA lesions in living cells (Roza et al., 1991) and in some microorganisms (Oguma et al., 2001) can be repaired by one or more mechanisms. Such mechanisms include the enzyme-dependent nucleotide excision repair (NER), also named dark repair, and the light-dependent reaction

184

H.V. Smith et al.

known as photoreactivation (PHR). Dark repair and PHR enable UV-inactivated microorganisms to recover and may reduce the efficiency of UV inactivation. Some microorganisms found in water exhibit light and/or dark repair.

UV light and disinfection of Cryptosporidium oocysts The potential of UV irradiation for disinfecting drinking water has been recognized for many years. UV irradiation is effective for killing microorganisms, including bacteria, fungi and oocysts. Various studies have shown that parasitic protozoa are more sensitive to UV irradiation than viruses, but less sensitive than most bacteria (reviewed in Hijnen et al., 2006). Cryptosporidium parvum oocysts can be inactivated using pulsed UV light (Slifko et al., 1999) and low- (Oguma et al., 2001; Shin et al., 2001) and medium-pressure UV (Craik et al., 2001; Rochelle et al., 2004). Pulsed light consists of intense flashes of broad-spectrum white light containing wavelengths from 200 nm (UV) to 1000 nm in the near infrared region (Takeshita et al., 2003). Low-pressure mercury lamps emit peak output at 254 nm, whereas medium-pressure mercury lamps emit radiation at several peaks between 248 and 295 nm. Most studies on C. parvum oocyst inactivation have used low- and medium-pressure lamps (Rochelle et al., 2005). No differences in UV inactivation of C. parvum are apparent using either low- or medium-pressure UV lamps (Craik et al., 2001; Rochelle et al., 2004). Table 15.2 identifies some studies that indicate that UV light is an effective disinfectant for waterborne Cryptosporidium oocysts, particularly when mediumand low-UV-pressure lamps are used. The majority of data on the usefulness of UV disinfection have been accrued using sources which deliver continuous UV irradiation; however, the possibility that pulsed UV light may be more effective for water disinfection has also been put forward (Takeshita et al., 2003).

Assays for Determining UV Inactivation of Cryptosporidium Oocysts A variety of assay methods using a variety of C. parvum oocyst isolates have been used. In one of the earliest published works, C. parvum oocyst inactivation by UV radiation was assessed using maximized in vitro excystation and the fluorogenic vital dyes assay. Oocysts were inactivated (>99%, 2-log10 reduction) using a high dose of UV irradiation (up to 8748 mJ/cm2) (Campbell et al., 1995). Morita et al. (2002) showed that C. parvum oocysts exhibited high resistance to UV irradiation, requiring a dose of 230 mJ/cm2 for a 99% reduction in excystation. Animal infectivity and in vitro cell culture studies indicated that C. parvum oocysts were much more sensitive to UV irradiation than previously identified: the UV dose required for a 99% reduction in mouse infectivity was 1.0 mJ/cm2, whereas a dose of approximately 200 mJ/cm2 was required to achieve 99% reduction by excystation (Morita et al., 2002). Rochelle et al. (2004) showed that 99.9% inactivation (0.001% relative mouse infectivity) was achieved with a UV dose of 7.5 mJ/cm2. Using the CD-1 neonatal mouse infectivity assay, 99.9%

Towards Methods for Detecting UV-induced Damage

185

Table 15.2. UV disinfection of Cryptosporidium oocysts determined by oocyst inactivation. UV pressure

UV dose

Time/mode

Log reduction

Reference

Low

120 mJ/cm2

4 gpm/water device

Drescher et al. (2001)

Not stated

15,000 mW/sec = 15,000 mJ/ cm2

150 min/ bench scale

Medium and low

0.8–119 mJ/cm2

Bench scale

Medium

41 mJ/cm2

Bench scale

Low

20–10 mJ/cm2

Bench scale

5.4 with neonatal mouse infectivity >2 with neonatal mouse infectivity Max. 3.4–4.9 with neonatal mouse infectivity >4 with neonatal mouse infectivity >3 with in vitro infectivity

LorenzoLorenzo et al. (1993) Craik et al. (2001)

Bukhari et al. (1999)

Bukhari and LeChevallier (2003)

inactivation at a UV dosage of 19 mJ/cm2 (Bukhari et al., 1999) and 99.0–99.9% inactivation at 10–25 mJ/cm2 (Craik et al., 2001) could be achieved. Shin et al. (2001) utilized a cell culture infectivity assay combined with epifluorescence detection to determine UV inactivation of C. parvum. They reported 99.9% inactivation at a dosage of 3 mJ/cm2 using a low-pressure UV lamp. Both cell culture and animal infectivity assays gave comparable results with a 99.9% inactivation at a dosage of 3 mJ/cm2. Rochelle et al. (2004) assessed the inactivation of different isolates of C. parvum exposed to low- or medium-pressure UV lamps, using cell culture (monolayers of HCT-8 cells) and a reverse transcriptase polymerase chain reaction (RT-PCR) assay. An average dose of 7.6 mJ/cm2 resulted in 99.9% inactivation of oocysts in five different isolates. Using the same culture-based methods, Johnson et al. (2005) demonstrated that C. hominis display similar levels of infectivity in cell culture and have a similar sensitivity to UV light as C. parvum. Cryptosporidium parvum can undertake photoreactivation and dark repair at the genomic level (Oguma et al., 2001; Morita et al., 2002). In addition, NER repair genes have been identified in C. parvum and C. hominis. However, UV inactivation of Cryptosporidium oocysts is irreversible, despite the presence of the UV repair genes (Rochelle et al., 2004).

186

H.V. Smith et al.

Application to Waterborne Oocysts In order to remain compatible with, and to integrate as much as possible with, existing UK and USA detection methodologies, our development of suitable methods that can detect UV damage in individual organisms reliably and reproducibly was driven by a search for fluorogenic reporters which will enter and stain UV killed and damaged Cryptosporidium oocysts. This was guided by the following hypotheses: 1. UV disinfection induces the production of reactive oxygen species (ROS). 2. UV disinfection induces apoptosis. 3. UV disinfection causes damage to DNA, which is detectable using fluorogenic DNA reporters.

Hypothesis 1: UV induces the production of reactive oxygen species (ROS) Ultraviolet light A (UV-A) radiation induces the production of ROS in cells, which can damage cellular elements. Oxidative modifications to DNA nucleotides (e.g. 8-hydroxyguanine (8-oxoG)) are mutagenic (Cheng et al., 1992). Antioxidants reduce ROS production and GSH plays a significant role in inhibiting the generation of 8-oxoG by ionizing radiation (Fischer-Nielsen et al., 1994). GSH reacts with various ROS and is a cofactor for the H202-removing enzyme, glutathione peroxidase. In cultured mammalian cells, GSH depletion and thermal stress increase endogenous oxidative damage, but the addition of thiols to the medium does not reduce the level of oxidative damage caused by GSH depletion and thermal stress (Will et al., 1999). GSH is a low-molecular-weight tripeptide (gamma-glutamylcysteinylglycine) which is synthesized intracellularly. It plays a critical role in the detoxification of several drugs and xenobiotics (Meister and Anderson, 1983) and in cellular defence against agents that cause oxidative stress (Anderson, 1998). A very well-established technique to measure GSH is to add the cell permeant, fluorogenic vital dye monochlorobimane (MCB) to detect both GSH levels and activity in cells. MCB does not fluoresce, but on reacting with GSH yields GSH-bimane adducts which fluoresce. The enzyme glutathioneS-transferase (GST) exclusively mediates the intracellular conjugation of GSH and MCB. The rate of conjugation between GSH and MCB (which produces the fluorescence signal) is dependent on the abundance of GST (Haugland, 2005). As a putative reporter of UV damage caused by the production of ROS, we investigated the presence of the antioxidant GSH in C. parvum sporozoites in intact, untreated oocysts and in sporozoites within intact UV-irradiated oocysts (40 mJ/cm2). Purified C. parvum oocysts were purchased from Bunch Grass Farm (Idaho, USA; Iowa isolate BGF06-1, 10 days old) and stored between 4°C and 8°C until used. The percentage viability and excystation rate of this isolate using our optimized, fluorogenic vital dyes (DAPI-PI; Campbell et al., 1992) and our maximized in vitro excystation protocols (Robertson et al., 1993) were 95.2 ± 2.3 and 96.0 ± 1.0, respectively.

Towards Methods for Detecting UV-induced Damage

187

UV irradiation was delivered using a high-intensity, low-pressure UV lamp with an output at 254 nm. The procedure was adapted from the method described by Rochelle et al. (2004). Short-wave UV irradiation from a UVGL-58 Mineralight lamp was utilized. The intensity of the UV light, measured using a digital UVX radiometer, was (on average) 350 µW/cm2 at 254 nm. A rig was set up, and 10 cm below the lamp a position was marked where the UV intensity was maximal (350 µW/cm2). The UV dose was then determined from: UV dose (mJ/cm2) = Irradiance (mW/cm2) × Exposure time (s) In all experiments, 1 × 106 oocysts were suspended in 5 ml of Hanks’ balanced salt solution (HBSS). Samples were placed in Petri dishes (diameter 36 mm) which were constantly mixed using a magnetic stirrer during exposure to UV light. To achieve different UV dosages (mJ/cm2), oocysts were exposed to UV light for varying times at a constant distance (10 cm) from the constant intensity UV source. For each experiment, control oocysts were kept under the same conditions without irradiation. Statistical analysis was performed using analysis of variance (ANOVA) with P < 0.05 as the criterion for significance using MINITAB version 11 software. MCB fluorescence localized GSH in intact oocysts and the number of distinctly labelled foci in intact oocysts varied between 2 and 6 both in recently excreted (10 days old) and aged (6 months old) oocysts (Al-Adhami et al., 2006). MCB-labelled excysted sporozoites retained the fluorogenic dye at several intrasporozoite foci. GSH distribution in sporozoites was granular in the apical and posterior (nuclear) regions (Al-Adhami et al., 2006). We used buthionine sulphoximine (BSO), a potent and specific inhibitor of GSH (Meister and Anderson, 1983), to determine whether GSH is synthesized in BSO-treated oocysts, by labelling treated oocysts with MCB. When oocysts were depleted of GSH using BSO for 24 h at room temperature (RT), a significant decrease in fluorescence (~50%) was observed, indicating that MCB binds sporozoite GSH. We enumerated oocysts exposed to 10, 20 or 40 mJ/cm2 doses of UV light for MCB inclusion or exclusion, and quantified their fluorescence (using ANALYSIS software for scientific imaging and calibrated image measurements; Olympus, UK). The percentage inhibition of irradiated versus control oocysts was not significantly different (irradiated group mean ± SD 19.8 ± 10.5, 17.2 ± 12.8, 20.6 ± 8.0 at 10, 20 or 40 mJ/cm2, respectively; control group mean ± SD 15.5 ± 9.6, P > 0.05). No significant differences in fluorescence intensity or distribution of MCB occurred in irradiated or control oocysts (irradiated group mean ± SD mJ/cm2 = 152.3 ± 15.9, 20 mJ/cm2 = 149.6 ± 12.2, 40 mJ/ cm2 = 160.2 ± 18.5; control group mean ± SD 165.1 ± 20.3, P > 0.05). Oocysts subjected to GSH depletion showed significant decrease in GSH-bimane adducts (mean ± SD 62.3 ± 13.4) when compared to the control group (mean ± SD 165.1 ± 20.3, P = 0) as demonstrated by fluorescence quantitization.

Determining oocyst viability using MCB and propidium iodide staining The procedure developed was similar to that used in the fluorogenic vital dyes (DAPI-PI) assay of Campbell et al. (1992). The proportions of MCB+

188

H.V. Smith et al.

PI−, MCB− PI+, MCB− PI− and empty oocysts were quantified by enumerating 100 oocysts in triplicate samples. Oocysts were considered viable if they did not include PI but were stained with MCB (MCB+ PI−). Also, intact oocysts which did not include MCB or PI (MCB− PI−) but contained sporozoites under differential interference contrast (DIC) microscopy were considered viable. MCB incorporation correlated well with DAPI+ PI− oocyst staining. The results of DAPI-PI and MCB-PI staining were not significantly different when labelled oocysts were enumerated (mean ± SD 84.7 ± 4.2 versus 73.3 ± 8.7, P = 0.05, respectively). Results using the maximized in vitro excystation assay were not significantly different from DAPI-PI staining (mean ± SD 89.7 ± 2.1 versus 84.7 ± 4.2, P = 0.1) but were significantly different from MCB-PI staining (mean ± SD 89.7 ± 2.1 versus 73.3 ± 8.7, P = 0.01). These data, based on the MCB labelling of intact C. parvum oocysts, identify the presence of glutathione both in nuclear and cytoplasmic foci of sporozoites, which can be specifically depleted by BSO. Its function as an endogenous free radical scavenger in UV-irradiated oocysts was not demonstrated. Thus it is likely that other free radical scavengers are more active than GSH in UV-treated C. parvum (e.g. cysteine, ascorbic acid). MCB is unlikely to be useful as a surrogate for detecting UV damage in UV-treated Cryptosporidium oocysts.

Hypothesis 2: UV disinfection induces apoptosis The role of UV irradiation in inducing programmed cell death/apoptosis was investigated using DNA intercalating dyes. Currently, there are no reports available regarding UV-irradiation-induced apoptotic changes in Cryptosporidium oocysts. YO-PRO1 (YP) binds strongly to nucleic acids and has been used to detect apoptosis in mammalian cells without interfering with cell viability (Idziorek et al., 1995). Apoptosis is a highly regulated physiological process that is linked to pathological events such as oxidative stress in which the reactive oxygen species (ROS) play a key role in the initiation of the process (Plantin-Carrenard et al., 2003). The early stages of apoptosis are characterized by chromatin condensation, nuclear fragmentation and mitochondrial clustering, but with no loss of membrane integrity (Cohen, 1993). The plasma membrane becomes slightly permeable during apoptosis, allowing the uptake of YP but not PI, the impermeant dead-cell stain (Gilbert and Knox, 1997). Apoptosis is accompanied by the activation of an endonuclease enzyme that cleaves DNA initially into large fragments (50–300 kb) which yield internucleosomic fragments of 180–200 base pair multimers (Cohen et al., 1994). YP nucleic acid stain forms the basis of an important assay for apoptotic cells and is compatible with both epifluorescence microscopy and flow cytometry. Selective uptake of YP by apoptotic cells of a dexamethasone-treated population of thymocytes, an irradiated peripheral blood mononuclear cell population, and a growth-factor-depleted tumour B cell line was confirmed by cell sorting (Idziorek et al., 1995). Apoptotic cells take up YP, while viable cells exclude it (Haugland, 2005).

Towards Methods for Detecting UV-induced Damage

189

We focused our investigations on the usefulness of YP to detect DNA damage in UV-treated (40 mJ/cm2) C. parvum oocysts. Three nucleic acid stains were used to identify viable (DAPI, blue nuclei; in the absence of PI, red nuclei) oocysts, apoptotic (YP, green nuclei) oocysts and dead (PI, red nuclei) oocysts. UV disinfection procedures were as previously described. Epifluorescence microscopy, flow cytometry and Nomarski differential interference contrast (DIC) microscopy were used to assess outcomes. Oocysts exposed to UV disinfection, those subjected to apoptosis-inducing drug treatment changes, and controls were analysed using this three-colour assay. DAPI (indicator of viability in the absence of PI staining) was taken up by ~95% of the untreated population, while YP stained 35% of the same population. PI was excluded from all DAPI+ YP+ oocysts. PI stained the remaining 5% of the population. This result was consistent with that obtained with the DAPI-PI assay. YP was used in combination with PI to assess the population structure of irradiated, drug-treated and untreated oocysts by epifluorescence microscopy and fluorescence activated cell sorting (FACS). YP was incorporated into sporozoite DNA of intact, irradiated oocysts (possibly apoptotic) which exhibited no oocyst wall damage as determined by DIC microscopy. Apoptosis was induced in oocysts treated with dexamethasone (1 mM) or etoposide (10 µM) prior to labelling with fluorescent probes. Drug treatment induced increased YP signals compared with controls. However, control oocysts did not exclude YP entirely, showing low-level staining. DIC microscopy revealed alterations in the morphology of drug-treated oocysts, compared with irradiated and control (untreated) oocysts. When oocysts were identified as apoptotic (YP+ DAPI+ PI−), no significant differences were detected between irradiated and drug-treated oocysts, compared with controls (Table 15.3). Unlike our epifluorescence microscopy data, FACS revealed that 71.9% of irradiated and 33.7% of drug-treated populations were YP-positive compared with 4.2% of the control population. Despite the increase in YP-positive oocysts by FACS, it is unlikely that YP can provide a reliable estimate of the possible apoptotic changes that can occur in irradiated oocysts or sporozoites, for two main reasons: (i) the staining pattern resulting from the simultaneous use of these Table 15.3. Epifluorescence microscopy and flow cytometric analyses of recently excreted C. parvum oocysts subjected to different treatments, then stained with YP, DAPI and PI for epifluorescence microscopy or YP and PI for flow cytometry. % of labelled oocysts Assay Epifluorescence microscopy (YP+DAPI+PI−) (YP−DAPI+PI−) (YP−DAPI−PI+) Flow cytometry

Irradiated

Etoposide

Dexamethasone

Control

42.2 54.0 3.8 71.9

35.6 54.1 10.3 ND

ND ND ND 33.7

39.3 57.3 3.4 4.2

ND: Not determined. Note: YP stains apoptotic oocysts, PI stains dead oocysts.

190

H.V. Smith et al.

three dyes makes it impossible to distinguish normal from apoptotic oocysts by standard sets of filters used in epifluorescence microscopy; and (ii) differences in the intensity and the rapid quenching of YP make it very difficult to count and photograph YP-labelled oocysts.

Hypothesis 3: UV disinfection causes damage to DNA, which is detectable using fluorogenic DNA reporters The major DNA lesion caused by UV, the cyclobutyl pyrimidine dimer (TD), is responsible for UV-induced cytotoxicity and mutagenicity in living cells and microorganisms (Mitchell, 1988), and the formation of such lesions in genomic DNA inhibits normal replication and transcription of DNA and results in the inactivation of cells. Two known antibody reporters of nuclear UV damage were investigated. Two different antibodies raised against nucleic acids (α-double sDNA (α-dsDNA) and α-TD) were used to identify changes induced by UV light in C. parvum sporozoites and oocysts. Sporozoite nuclei were also stained with DAPI to validate the co-localization of α-dsDNA and α-TD (Al-Adhami et al., 2007). An improved α-TD antibody labelling procedure within irradiated C. parvum oocysts was established following freeze-thawing (five cycles), based on that of Nichols and Smith (2004). No α-TD localization was seen in non-irradiated oocysts. α-dsDNA antibody bound to the nuclei of both irradiated and non-irradiated sporozoites. Both C. parvum and C. hominis oocysts exposed to different doses of UV light (range 4–40 mJ/cm2) were tested using a standardized set of parameters: oocysts dried onto slides prior to freeze-thawing (five cycles), then fixed in methanol and labelled with either α-TD-Ab, a commercially available, fluorescein-labelled, monoclonal antibody reactive with surface exposed epitopes on Cryptosporidium oocysts (FITC-C-mAb) and DAPI stain or α-TD-Ab and DAPI. Our data indicate that UV irradiation at doses ranging from 10 to 40 mJ/cm2 can be detected using α-TD-Ab and DAPI (Al-Adhami et al., 2007). While the combination of α-TD-Ab, FITC-C-mAb and DAPI produced positive outcomes only with high levels of UV irradiation (40 mJ/cm2), if we replaced FITC-C-mAb with Texas Red (TR)-C-mAb and used the combination of α-TD-Ab, TR-C-mAb and DAPI, DNA damage to C. parvum and C. hominis sporozoites within intact oocysts could be detected at a lower limit of 10 mJ/cm2, but not at 4 mJ/cm2. Currently, the combination of α-TD-Ab, TR-C-mAb and DAPI can be used to detect damage in nuclei of oocysts of C. parvum and C. hominis exposed to UV light (range 10–40 mJ/cm2). Validation by comparison with infectivity in neonatal CD-1 mice All procedures and manipulations performed were as described by Korich et al. (2000), and the experimental design is presented in Table 15.4. Mice infected with untreated oocysts (standard curve, Table 15.4) showed varying levels of infection. In the standard curve group, the percentage infection ranged from 5.2% in mice infected with 30 oocysts to 80% in mice infected with 300 oocysts.

Towards Methods for Detecting UV-induced Damage Table 15.4.

191

Experimental design of animal infectivity experiment.

UV dose (mJ/cm2) 10 10 20 20 None (control) None (control) None (control)

No. of animals 20 20 20 20 20 20 20

Dose (No. of oocysts/10 µl HBSS) 300/10 µl (low) 3000/10 µl (high) 300/10 µl (low) 3000/10 µl (high) 30/10 µl (low) 150/10 µl (medium) 300/10 µl (high)

UV-irradiated oocysts of C. parvum at 10 and 20 mJ/cm2 (Table 15.4) failed to cause infection in neonatal CD-1 mice, despite the fact that large doses of oocysts (up to 3000 per os, which is greater than 44 times the ID50 for this strain of mouse; Korich et al., 2000) were given to the animals. We conclude that an immunofluorescence assay using α-TD antibodies, which, for C. parvum, has been validated against a neonatal mouse infectivity assay, is suitable for detecting thymine dimers in air-dried oocysts and air-dried sporozoites of C. parvum and C. hominis oocysts and that the α-dsDNA antibody is a good candidate for a positive control for the assay.

Genotyping Cryptosporidium Oocysts by PCR Following UV Disinfection The species of waterborne Cryptosporidium oocysts recovered from routine Cryptosporidium monitoring of water sources can be determined by molecular methods (see Smith et al., Chapter 17, this volume). One pertinent issue that arises is whether oocysts which have been disinfected by UV light can be amplified by PCR, as UV disinfection damages DNA and the apoptopic cascade activates an endonuclease enzyme that cleaves DNA initially into large fragments (50–300 kb), which yield internucleosomic fragments of 180–200 base pair multimers (Cohen et al., 1994). Statements to the effect that UV light at up to 40 mJ/ cm2 damages DNA to such an extent that the molecular tools used to determine the species of small numbers of oocysts (nested 18S rRNA loci) cannot be used, have been made on numerous occasions. UV irradiation damage and repair is detectable by PCR in mammalian cultured cells and assay sensitivity is dependent on the size of the DNA fragment amplified. Long PCRs (6–24 kb) are required to ensure sensitive and unequivocal detection of DNA-induced lesions. UV irradiation induces DNA helix distortion, caused by the formation of pyrimidine dimmers, and DNA synthesis is impaired both in vivo and in vitro. Wang et al. (2003) demonstrated damage and repair of the p53 gene in human cells by a multiplex long quantitative PCR, designed to co-amply a 7 kb fragment of the gene and a 500 bp fragment control to increase the reliability of the assay. The lesion frequency detected in this gene

192

H.V. Smith et al.

was 0.63 lesions/10 kb/10 J/m2. In most applications, the UV dosage required to induce detectable lesions by PCR inhibition is much higher than the dosage used for Cryptosporidium inactivation; furthermore, the PCR fragment size is much larger than is used in most PCR applications. However, less PCR product was observed after amplifying the 1.7 kb 16S rDNA gene from cyanobacteria after UV-B radiation-induced damage (Kumar et al., 2004). Genomic DNA, extracted from bacteria exposed to 2.5 W/m2 of irradiation (up to 90,000 mw/s/ cm2 = 90,000 mJ/cm2), was tested by random amplification of polymorphic DNA (RAPD) analysis and direct PCR, and a marked decrease in amplification products, which was directly proportional to exposure time, occurred with both assays. Most PCR applications on the Cryptosporidium 18S rRNA gene used for genotyping oocysts from water do not exceed ~1000-bp fragment length for the first PCR amplification and vary from ~400 to 800 bp for the secondary amplification of nested assays. By extrapolation, it is unlikely that UV irradiation of water will have an adverse effect on the PCRs used for detecting oocysts; however, experimental confirmation of this hypothesis must be sought.

Conclusions Water is an important transmission route for cryptosporidiosis, with at least 165 waterborne outbreaks documented. Cryptosporidium can be controlled in water treatment by physical removal, yet Cryptosporidium oocysts are insensitive to commonly used disinfectants, with the exception of UV. Clearly, a method that can determine whether the small numbers of oocysts found in routine samples exposed to UV irradiation have been disinfected is of benefit, as it offers increased confidence of the success of UV disinfection to water operators. Neonatal animal infectivity, in vitro infectivity, excystation in vitro and current fluorogenic vital dye methods cannot be used to address, accurately, the viability of individual UV disinfected oocysts. UV light is classified according to wavelength: UV-A, 315–400 nm; UV-B, 280–315 nm; and UV-C, 100–280 nm; has been used for drinking water disinfection since the beginning of the 20th century, and is relatively insensitive to temperature and pH differences. Turbidity reduces its water penetration radiation and there is no disinfection residual. The modes of action of UV light on uni- and multicellular organisms involve oxidative stress resulting from the attack by free radicals of several cellular targets. The major effects of UV radiation on cell membranes are alterations of proteins, particularly protein crosslinking. UV-B radiation progressively inhibits protein synthesis. Specific free radical scavengers protect cells against killing and inhibition of protein synthesis by UV-B. Additional effects of pulsed UV light damage include increased concentration of eluted proteins, increased cell membrane damage and structural changes in yeast cells. UV light also crosslinks the complementary strands of DNA and causes the formation of single strand breaks and pyrimidine dimers. The major lesions induced are cyclobutyl pyrimidine dimers (CPDs; also known as thymine dimers, TD).

Towards Methods for Detecting UV-induced Damage

193

UV-induced DNA lesions in living cells and in some microorganisms can be repaired by enzyme-dependent nucleotide excision repair (NER), also named dark repair, and the light-dependent reaction known as photoreactivation (PHR). Dark repair and PHR enable UV-inactivated microorganisms to recover and may reduce the efficiency of UV inactivation. C. parvum oocysts are inactivated at 3–40 mJ/cm2 using medium- and low-pressure UV light. C. parvum can undertake photoreactivation and dark repair at the genomic level, and NER repair genes have been identified in C. parvum and C. hominis. However, UV-inactivation of Cryptosporidium oocysts is irreversible, despite the presence of the UV repair genes. We developed assays for determining UV inactivation of Cryptosporidium oocysts which would remain compatible with, and integrated as much as possible with, existing UK and USA detection methodologies. Our development of suitable methods which can detect UV damage in individual organisms reliably and reproducibly was driven by a search for fluorogenic reporters which could enter and stain UV-killed and damaged Cryptosporidium oocysts, and was guided by the following hypotheses: 1. UV disinfection induces the production of reactive oxygen species (ROS). Antioxidants reduce ROS production and the antioxidant glutathione (GSH) plays a significant role in inhibiting the generation of mutagens by ionizing radiation. We investigated the presence of GSH, as a putative reporter of UV damage caused by the production of ROS, in intact, untreated C. parvum oocysts and sporozoites within intact UV-irradiated oocysts (40 mJ/cm2), using MCB to detect both GSH levels and activity in cells. In purified, intact, recently excreted and aged C. parvum oocysts, MCB fluorescence localized GSH in intact oocysts at several nuclear and cytoplasmic sporozoite foci (n = 2–6). The function of GSH as an endogenous free radical scavenger in UV-irradiated oocysts was not demonstrated. Other free radical scavengers are more active than GSH in UV-treated C. parvum oocysts, and MCB is unlikely to be useful as a surrogate for detecting UV damage in UV-treated Cryptosporidium oocysts. 2. UV disinfection induces apoptosis. The DNA intercalating dye YO-PRO1 (YP) was chosen to investigate the role of UV irradiation in inducing programmed cell death/apoptosis. YP binds strongly to nucleic acids and has been used to detect apoptosis in mammalian cells without interfering with cell viability. We focused our investigations on the usefulness of YP to detect DNA damage in UVtreated (40 mJ/cm2) C. parvum oocysts. YP was incorporated into sporozoite DNA of intact, irradiated oocysts (possibly apoptotic) which exhibited no apparent oocyst wall damage. Dexamethasone- or etoposide-induced apoptosis was localized at higher levels, compared with controls; however, control oocysts did not exclude YP entirely. No significant differences were detected between irradiated and drug-treated oocysts, compared with controls by fluorescence microscopy, but differences were apparent by FACS. Despite the increase in YP-positive oocysts by FACS, YP is unlikely to provide a reliable estimate of the possible apoptotic changes that can occur in irradiated oocysts or sporozoites. 3. UV disinfection causes damage to DNA, which is detectable using fluorogenic DNA reporters. The major DNA lesion caused by UV, the cyclobutyl

194

H.V. Smith et al.

pyrimidine dimer (TD), is responsible for UV-induced cytotoxicity and mutagenicity. It inhibits normal replication and transcription of DNA and results in the inactivation of cells. We used an antibody raised against TDs (α-TD) to identify changes induced by UV light in C. parvum oocysts, and validated its nuclear location by co-localization with the nuclear fluorogen, DAPI. To improve α-TD antibody labelling within irradiated C. parvum oocysts, a freeze-thawing (five cycles) procedure was developed. No α-TD localization was seen in non-irradiated oocysts. Replacing FITC-C-mAb with Texas Red (TR)-C-mAb and using the combination of α-TD-Ab, TR-C-mAb and DAPI, DNA damage to C. parvum and C. hominis sporozoites within intact oocysts could be detected at a lower limit of 10 mJ/cm2, but not at 4 mJ/cm2. Currently, the combination of α-TD-Ab, TR-CmAb and DAPI can be used to detect damage in nuclei of oocysts of C. parvum and C. hominis exposed to UV light (range 10–40 mJ/cm2). We conclude that an immunofluorescence assay using α-TD antibodies, which, for C. parvum, has been validated against a neonatal mouse infectivity assay, is suitable for detecting thymine dimers in air-dried oocysts and air-dried sporozoites of C. parvum and C. hominis oocysts, and that the α-dsDNA antibody is a good candidate for a positive control for the assay.

Acknowledgements This work was funded by the Environmental and Rural Affairs Department, Agricultural and Biological Research Group, Scottish Executive, Scotland, UK. We thank Dr D. Reid, Drinking Water Quality Unit, Scottish Executive, Edinburgh, for managing the project.

References Al-Adhami, B.H., Nichols, R.A.B., Kusel, J.R., O’Grady, J. and Smith, H.V. (2006) Cryptosporidium parvum sporozoites contain glutathione. Parasitology 133, 555–563. Al-Adhami, B.H., Nichols, R.A.B., Kusel, J.R., O’Grady, J. and Smith, H.V. (2007) Detection of UV-induced thymine dimers in individual Cryptosporidium parvum and Cryptosporidium hominis oocysts by immunofluorescence microscopy. Applied and Environmental Microbiology 73, 947–955. Anderson, M.E. (1998) Glutathione: an overview of biosynthesis and modulation. Chemico-Biological Interactions 24, 1–14. Bukhari, Z. and LeChevallier, M. (2003) Assessing UV reactor performance for treatment of finished water. Water Science and Technology 47, 179–184. Bukhari, Z., Hargy, T.M., Bolton, J.R., Dussert, B. and Clancy, J.L. (1999) Mediumpressure UV for oocyst inactivation. Journal of the American Water Works Association 91, 86–94. Cacciò, S.M., Thompson, R.C.A., McLauchlin, J. and Smith, H.V. (2005) Unravelling Cryptosporidium and Giardia epidemiology. Trends in Parasitology 21, 430–437. Campbell, A.T., Robertson, L.J. and Smith, H.V. (1992) Viability of Cryptosporidium parvum oocysts: correlation of in vitro excystation with inclusion or exclusion of fluorogenic vital dyes. Applied and Experimental Microbiology 58, 3488–3493.

Towards Methods for Detecting UV-induced Damage

195

Campbell, A.T., Robertson, L.J., Snowball, M.R. and Smith, H.V. (1995) Inactivation of oocysts of Cryptosporidium parvum by ultraviolet irradiation. Water Research 29, 2583–2586. Cheng, K.C., Cahill, D.S., Kasai, H., Nishimura, S. and Leob, L.A. (1992) 8-Hydroxyguanine, an abundant form of oxidative DNA damage, causes G-T and A-T substitution. Journal of Biological Chemistry 267, 166–172. Clancy, J.L., Bukhari, Z., McCuin, R., Clancy, T.P., Marshall, M.M., Korich, D.G., Fricker, C.R., Sykes, N., Smith, H.V., O’Grady, J.E., Rosen, J.P., Sobrinho, J. and Schaefer, F.W., III (2000) Cryptosporidium: Viability and Infectivity Methods. American Water Works Association Research Foundation / American Water Works Association / UK Drinking Water Inspectorate, pp.137 Cohen, G.M., Sun, X.M., Fearnhead, H., MacFarlane, M., Brown, D.G. Snowden, R.T. and Dinsdale, D. (1994) Formation of large weight fragments of DNA is a key committed step of apoptosis in thymocytes. Journal of Immunology 153, 507–516. Cohen, J.J. (1993) Apoptosis. Immunology Today 14, 126–130. Craik, S.A., Weldon, D., Finch, G.R., Bolton, J.R. and Belosevic, M. (2001) Inactivation of Cryptosporidium parvum oocysts using medium- and low-pressure ultraviolet radiation. Water Research 35, 1387–1398. Drescher, A.C., Greene, D.M. and Gadgil, A.J. (2001) Cryptosporidium inactivation by low-pressure UV in a water disinfection device. Journal of Environmental Health 64, 31–35. El-Adhami, W., Daly, S. and Stewart, P.R. (1994) Biochemical studies on the lethal effects of solar and artificial ultraviolet radiation on Staphylococcus aureus. Archives of Microbiology 161, 82–87. Fayer, R., Morgan, U. and Upton, S.J. (2000) Epidemiology of Cryptosporidium: transmission, detection and identification. International Journal for Parasitology 30, 1305–1322. Fischer-Nielsen, A., Jeding, I.B. and Loft, S. (1994) Radiation-induced formation of 8-hydroxy-2-deoxyguanosine and its prevention by scavengers. Carcinogenesis 15, 1609–1612. Gilbert, M. and Knox, S. (1997) Influence of Bcl-2 overexpression on Na+/K+-ATPase pump activity: correlation with radiation-induced programmed cell death. Journal of Cellular Physiology 171, 299–304. Girdwood, R.W.A. and Smith, H.V. (1999) Cryptosporidium. In: Robinson, R., Batt, C. and Patel, P. (eds) Encyclopaedia of Food Microbiology, Vol. 1. Academic Press, London and New York, pp. 487–497. Haugland, R.P. (2005) Probes for cell adhesion, chemotaxis, multidrug resistance and glutathione. In: Spence, M.T.Z. (ed.) The Handbook: A Guide to Fluorescent Probes and Labelling Technologies. Molecular Probes Inc., Oregon, USA, pp. 767–776. Hijnen, W.A.M., Beerendonk, E.F. and Medema, G.J. (2006) Inactivation credit of UV irradiation for viruses, bacteria and protozoan (oo)cysts in water: a review. Water Research 40, 3–22. Idziorek, T., Estaquier, J., De Bels, F. and Ameisen, J.-C. (1995) YOPRO-1 permits cytofluorometric analysis of programmed cell death (apoptosis) without interfering with cell viability. Journal of Immunological Methods 185, 249–258. Johnson, A.M., Linden, K., Ciociola, K.M., De Leon, R., Widmer, G. and Rochelle, P.A. (2005) UV inactivation of Cryptosporidium hominis as measured in cell culture. Applied and Environmental Microbiology 71, 2800–2802. Karanis, P., Kourenti, C. and Smith, H. (2007) Waterborne transmission of protozoan parasites: a worldwide review of outbreaks and lessons learnt. Journal of Water Health 5, 1–38.

196

H.V. Smith et al. Kochevar, I.E. (1990) UV-induced protein alterations and lipid oxidation in erythrocyte membranes. Photochemistry and Photobiology 52, 795–800. Korich, D.G., Marshall, M.M., Smith, H.V., O’Grady, J., Bukhari, Z., Fricker, C.R., Rosen, J.P. and Clancy, J.L. (2000) Inter-laboratory comparison of the CD-1 neonatal mouse logistic dose-response model for Cryptosporidium parvum oocysts. Journal of Eukaryotic Microbiology 47, 294–298. Kumar, A., Tyagi, M.B. and Jha, P.N. (2004) Evidences showing ultraviolet-B radiation-induced damage of DNA in cyanobacteria and its detection by PCR assay. Biochemical and Biophysical Research Communications, 318, 1025–1030. Lee, J.E., Park, J.C., Hwang, Y.S., Kim, J.K., Kim, J.G. and Sub, H. (2001) Characterization of UV-irradiated dense/porous collagen membranes: morphology, enzymatic degradation, and mechanical properties. Yonsei Medical Journal 42, 172–179. Lorenzo-Lorenzo, M.J., Ares-Mazas, M.E., Villacorta-Martinez de Maturana, I. and DuranOreiro, D. (1993) Effect of ultraviolet disinfection of drinking water on the viability of Cryptosporidium parvum oocysts. Journal of Parasitology 79, 67–70. Meister, A. and Anderson, M.E. (1983) Glutathione. Annual Review of Biochemistry 52, 711–760. Mitchell, D.L. (1988) The relative cytotoxicity of (6-4) photoproducts and cyclobutane dimers in mammalian cells. Photochemistry and Photobiology 48, 51–57. Morita, S., Namikoshi, A., Hirata, T., Oguma, K., Katayama, H., Ohgaki, S., Motoyama, N. and Fujiwara, M. (2002) Efficacy of UV irradiation in inactivating Cryptosporidium parvum oocysts. Applied and Environmental Microbiology 68, 5387–5393. Nejedly, K., Kittner, R., Pospisilova, S. and Kypr, J. (2001a) Crosslinking of the complementary strands of DNA by UV light: dependence on the oligonucleotide composition of the UV irradiated DNA. Biochimica et Biophysica Acta 1517, 365–75. Nejedly, K., Kittner, R. and Kypr, J. (2001b) Genomic DNA regions whose complementary strands are prone to UV light-induced crosslinking. Archives of Biochemistry and Biophysics 388, 216–224. Nichols, R.A.B. and Smith, H.V. (2004) Optimisation of DNA extraction and molecular detection of Cryptosporidium parvum oocysts in natural mineral water sources. Journal of Food Protection 67, 524–532. O’Grady, J.E. and Smith, H.V. (2002) Methods for determining the viability and infectivity of Cryptosporidium oocysts and Giardia cysts. In: Ziglio, G. and Palumbo, F. (eds) Detection Methods for Algae, Protozoa and Helminths. John Wiley and Sons, Chichester, UK, pp. 193–220. Oguma, K., Katayama, H., Mitani, H., Morita, S., Hirata, T. and Ohgaki, S. (2001) Determination of pyrimidine dimers in Escherichia coli and Cryptosporidium parvum during UV light inactivation, photoreactivation, and dark repair. Applied and Environmental Microbiology 67, 4630–4637. Plantin-Carrenard, E., Bringuier, A., Derappe, C., Pichon, J., Guillot, R., Bernard, M., Foglietti, M.J., Feldmann, G., Aubery, M. and Braut-Boucher, F. (2003) A fluorescence microplate assay using Yopro-1 to measure apoptosis: application to HL60 cells subjected to oxidative stress. Cell Biology and Toxicology 19, 121–133. Robertson, L.J., Campbell, A.T. and Smith, H.V. (1993) In vitro excystation of Cryptosporidium parvum. Parasitology 106, 13–29. Rochelle, P.A., Fallar, D., Marshall, M.M., Montelone, B.A., Upton, S.J. and Woods, K. (2004) Irreversible UV inactivation of Cryptosporidium spp. despite the presence of repair genes. Journal of Eukaryotic Microbiology 51, 553–562. Rochelle, P.A., Upton, S.J., Montelone, B.A. and Woods, K. (2005) The response of Cryptosporidium parvum to UV light. Trends in Parasitology 21, 81–87.

Towards Methods for Detecting UV-induced Damage

197

Roza, L., De Gruijl, F.R., Bergen Henegouwen, J.B.A., Guikers, K., Van Weelden, H., Van Der Schans, G.P. and Baan, R.A. (1991) Detection of photorepair of UV-induced thymine dimers in human epidermis by immunofluorescence microscopy. Journal of Investigative Dermatology 96, 903–907. Shin, G.-A., Linden, K.G., Arrowood, M.J. and Sobsey, M.D. (2001) Low-pressure UV inactivation and DNA repair potential of Cryptosporidium parvum oocysts. Applied and Environmental Microbiology 67, 3029–3032. Slifko, T.R., Huffman, D.E. and Rose, J.B. (1999) A most-probable-number assay for enumeration of infectious Cryptosporidium parvum oocysts. Applied and Environmental Microbiology 65, 3936–3941. Slifko, T.R., Smith, H.V. and Rose, J.B. (2000) Emerging parasite zoonoses associated with food and water. International Journal for Parasitology 30, 1379–1393. Smith, H.V. and Rose, J.B. (1990) Waterborne cryptosporidiosis. Parasitology Today 6, 8–12. Smith, H.V. and Rose, J.B. (1998).Waterborne cryptosporidiosis: current status. Parasitology Today 14, 14–22. Smith, H.V., Robertson, L.J. and Ongerth, J.E. (1995) Cryptosporidiosis and giardiasis: the impact of waterborne transmission. Aqua – Journal of Water Supply: Research and Technology 44, 258–274. Smith, H.V., Cacciò, S.M., Cook, N., Nichols, R.A.B. and Tait, A. (2007) Cryptosporidium and Giardia as foodborne zoonoses. Veterinary Parasitology 149, 29–40. Takeshita, K., Shibato, J., Sameshima, T., Fukunaga, S., Isobe, S., Arihara, K. and Itoh, M. (2003) Damage of yeast cells induced by pulsed light irradiation. International Journal of Food Microbiology 85, 151–158. Wang, Y.-C., Lee, P.-J., Shih, C.-M., Chen, H.-Y., Lee, C.-C., Chang, Y.-Y., Hsu, Y.-T., Liang, Y.-J., Wang, L.-Y., Han, W.-H. and Wang, I.-C. (2003) Damage formation and repair efficiency in the p53 gene of cell lines and blood lymphocytes assayed by multiplex long quantitative polymerase chain reaction. Analytical Biochemistry 319, 206–215. Will, O., Mahler, H., Arrigo, A. and Epe, B. (1999) Influence of glutathione levels and heat-shock on the steady-state levels of oxidative DNA base modifications in mammalian cells. Carcinogenesis 20, 333–337. Xiao, L., Fayer, R., Ryan, U. and Upton, S.J. (2004) Cryptosporidium taxonomy: recent advances and implications for public health. Clinical Microbiological Reviews 17, 72–97.

16

Effect of Environmental and Conventional Water Treatment Processes on Waterborne Cryptosporidium Oocysts

B. KING, A. KEEGAN, C. SAINT AND P. MONIS Australian Water Quality Centre, Salisbury, Australia

Abstract Cryptosporidium oocysts are prevalent in surface waters as a result of anthroponotic activity and native animal faecal contamination. A sound understanding of the impact of environmental and water treatment processes on the survival of oocysts is required to properly risk-assess the threat posed to public health by the presence of oocysts in water storages. This chapter provides an overview of recently completed work studying the impact of temperature, predation, sunlight and conventional water treatment processes on oocyst survival in water.

Background Cryptosporidium oocysts are frequently found in surface waters (Smith and Rose, 1990; Rose et al., 1997) and are extremely resistant to chlorine and monochloramine at the concentrations used to disinfect water for potable use (Korich et al., 1990; Finch et al., 1993). In addition to being resistant to commonly used disinfectants, it is generally thought that oocysts can persist for several months or more in the aquatic environment (Robertson et al., 1992; Johnson et al., 1997; Medema et al., 1997). These factors, combined with the relatively low infectious dose demonstrated for some isolates of Cryptosporidium parvum (Messner et al., 2001), mean that Cryptosporidium represents a challenge to water utilities responsible for the provision of safe drinking water to the public. Water utilities require accurate data on the fate and transport of oocysts in water and their response to water treatment processes, in order to effectively riskassess and risk-manage the threat posed by the presence of oocysts in source waters. A number of studies have examined oocyst survival, using techniques such as in vitro excystation or vital dye staining (Robertson et al., 1992; Chauret et al., 1998), but such methods are only indicators of viability and are known to overestimate infectivity (Black et al., 1996; Bukhari et al., 2000). As a consequence, 198

© CAB International 2009. Giardia and Cryptosporidium: From Molecules to Disease (eds G. Ortega-Pierres et al.)

Waterborne Cryptosporidium Oocysts

199

there is a lack of knowledge regarding the inactivation of oocysts in environmental waters, particularly with regard to the effect of biotic and abiotic processes on the retention of oocyst infectivity.

Objectives In order to address knowledge gaps with respect to the survival of oocysts in response to environmental and water treatment conditions, studies were conducted using a cell culture infectivity model (Keegan et al., 2003) to examine the response of oocysts in water to: ● ● ● ● ● ●

Temperature. Microbial activity. Natural light. Chemical flocculation (aluminium sulphate). Dissolved air flotation. Disinfection (chlorination, chloramination).

Approach Details of the methodologies used for the temperature, sunlight and treatment inactivation studies have been described in detail elsewhere (Keegan et al., 2003; King et al., 2005, 2007). Fresh or aged oocysts of Cryptosporidium parvum (Swiss cattle C26) that had been passaged through mice were used for all experiments, and infectivity was assessed using a cell culture infectivity/Taqman PCR assay (CC-PCR) (Keegan et al., 2003). A schematic diagram of the assay is presented in Fig. 16.1. For environmental survival experiments, microcosms containing fresh oocysts and either reagent-grade water, tap water or reservoir water were established in either sterile polycarbonate chambers (for temperature experiments) or methylacrylate chambers (for sunlight inactivation experiments). Oocyst infectivity was measured in triplicate using 10,000 oocysts, and log inactivation was calculated by comparing quantitative PCR results for the treated samples with the non-treatment controls (which represent maximum infectivity). Candidate predators of oocysts were enriched from Hope Valley reservoir water sediment by selecting for organisms with a food preference in the Cryptosporidium oocyst size-range, using cysts of the amoeba Rosculus spp. and baker’s yeast (Saccharomyces cerevisiae). Organisms were collected using micromanipulation and placed in an individual well of a 24-well Cellstar tissue culture plate (Greiner Bio-one, Frickenhausen, Germany) containing 1 ml of autoclaved Hope Valley reservoir water, supplemented with yeast dissolved in Ringer’s solution. Oocysts were pre-stained with a fluorescent antibody (AusFlow Cry 104; BTF, Sydney, Australia) and 4′,6-diamidino-2-phenylindole (DAPI) prior to feeding experiments. Feeding experiments were conducted using 8-well Lab-Tek II Chamber Slides (Nalgene Nunc International, Naperville, Illinois) for amoebae or 1.5 ml microcentrifuge tubes for other organisms in a total volume of 150 µl

200

B. King et al.

Treat to promote release of sporozoites, apply oocysts to cell monolayer

Incubate 48 hours to allow establishment of infection

Wash monolayer, extract DNA from monolayer, conduct quantitative real-time PCR

Analyse data, compare treatments versus untreated controls

Fig. 16.1. Overview of a Cryptosporidium infectivity assay that combines cell culture and real-time PCR.

of sterile reservoir water containing 5 × 104 oocysts per chamber. Samples were fixed by the addition 150 µl of sodium acetate acetic acid formalin (SAF). Organisms were viewed using an Olympus BX60 system microscope (Olympus, Oakleigh, Australia) and images were captured using an Olympus DP50 digital camera. The parameters used for conventional 2 l jar testing (to assess the effect of coagulation on oocyst infectivity) are listed in Table 16.1. Between 4 × 104 and 1 × 105 oocysts were used per jar, with doses of alum ranging from 0 to 100 mg/l,

Waterborne Cryptosporidium Oocysts Table 16.1.

201

Experimental jar test conditions.

Jar test parameters Oocyst mixing Flash mixing Slow mixing Settling Volume Raw water chemistry pH UV254 Temperature Turbidity Colour

5 min @ 200 rpm 1 min @ 200 rpm 14 min @ 20 rpm 15 min 2 l or 400 ml 8.13–8.40 0.148–0.195/cm 16.8 ± 1.2°C 2.95–3.25 NTU 21–24 HU

which is within the range of concentrations used at conventional water treatment plants in South Australia. For samples further treated by dissolved air flotation (DAF), a 15 min flotation step with 12% recycle at 70 psi was added to the end of the standard jar test procedure. For disinfection, contact times for chlorine and chloramine ranged from 300 to 1200 mg min/l.

Outcomes Temperature inactivation Replicate in vitro experiments were performed in reagent-grade MilliQ water and Hope Valley raw water (with or without sterilization) to assess oocyst removal and inactivation rates at temperatures ranging from 4°C to 25°C. Infectivity for each time point was measured using 10,000 oocysts in triplicate, with inactivation determined by comparison with the level of infectivity observed in the 4°C control at each time point. The inactivation of the 4°C sample over time was calculated using the 4°C sample for t = 0 as the reference for maximum infectivity. The rate of inactivation was related to the temperature, with more rapid inactivation observed at the higher temperatures (Fig. 16.2). Additional experiments were conducted to investigate the mechanism responsible for oocyst inactivation by temperature. Oocysts were stored in reagent-grade water at temperatures between 4°C and 37°C and measurements were made for infectivity and oocyst ATP concentrations to determine whether there was any effect on oocyst energy stores. Changes in ATP concentration closely corresponded with changes in infectivity, with a good correlation observed between these two parameters for all temperatures (Fig. 16.3).

Log inactivation

202

B. King et al. Reagent grade water

0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0 -4.5

4 degrees 15 degrees 20 degrees 25 degrees

Log inactivation

0

2

4 6 8 Time (weeks)

10

12

Hope Valley reservoir water

0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0 -4.5 0

2

4 6 8 Time (weeks)

10

Hope Valley reservoir water (sterile)

0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0 -4.5 12

0

2

4 6 8 Time (weeks)

10

12

Fig. 16.2. Inactivation of Cryptosporidium oocysts in water at different temperatures.

Predation Monitoring of oocyst numbers during the initial temperature inactivation study revealed changes in oocyst numbers for some treatments. In the case of the reagent water and autoclaved Hope Valley reservoir water treatments, the oocyst counts remained constant throughout the incubation period for all four temperatures in all replicate experiments. Furthermore, the morphology of the oocysts remained consistent, with a regular shape, good staining with the fluorescent antibody and no clumping observed. However, for one of the replicate experiments conducted on Hope Valley raw reservoir water, a gradual decrease in oocyst numbers was observed for the 15°C treatment and more rapid decreases for both 20°C and 25°C over the 12-week incubation period (Fig. 16.4). Oocysts examined in these samples were often clumped together with variable fluorescent staining and deformations in shape (as many as 18 in an individual clump, data not shown). Oocysts not in clumps exhibited typical staining and morphology, and exhibited infectivity consistent with oocysts in sterile water for the same temperature exposure.

Waterborne Cryptosporidium Oocysts

203

1E-10

0.5

ATP concentration (M) 1E-9

2E-9 1.8E-9 1.6E-9

1 0.0 -0.5

Log inactivation

-1.0

1.4E-9

-1

1.2E-9

-2

1E-9

-3

8E-10

-1.5 -4

6E-10

-2.0 -2.5

4E-10

ATP concentration (M)

Log survival

0

-3.0 CC-PCR -3.5 ATP -4.0 2E-10 -20

0

20

40

60

80

100

120

140

160

180

Time (hours) incubation at 37°C

Fig. 16.3. Comparison of oocyst inactivation measured using cell culture infectivity with ATP concentration, measured during incubation at 37°C. Inset: correlation of oocyst ATP concentration and log inactivation using the combined data from all temperature treatments (15, 20, 25, 30, 37°C) and time points.

Attempts to replicate the oocyst removal in further experiments were variable between replicates, but decreases in oocyst numbers were observed. Reservoir water filtered using an 0.8 micron filter did not exhibit any reduction in oocyst numbers. To further investigate the loss of oocysts, potential predators of oocysts were enriched from the reservoir water and tested to determine whether they could ingest oocysts. A range of ciliates, amoebae and rotifers were identified that could ingest antibody-labelled oocysts, as determined by fluorescent microscopy and DAPI staining, as well as a single platyhelminth and gastrotrich (Table 16.2). In the case of Paramecium, it appeared that oocysts were being digested in food vacuoles (Fig. 16.5). In the case of Blepharisma, it was not possible to view the FITC-labelled oocysts due to autofluorescence, but the oocysts were clearly visible by DAPI staining.

Solar inactivation Outdoor tank experiments and a cell culture infectivity assay were used to measure the effect of solar irradiation on C. parvum oocysts in water. Initial experiments

204

B. King et al. 100 90 80

Oocyst count

70 4°C

60

15°C 50

20°C 25°C

40

4°C autoclaved 30

15°C autoclaved

20

20°C autoclaved 25°C autoclaved

10 0

2

4 6 8 Incubation time (weeks)

10

12

Fig. 16.4. Comparison of oocyst counts in Hope Valley reservoir water (±sterilization) incubated at different temperatures over time.

Table 16.2.

List of phagotrophs capable of ingesting Cryptosporidium oocysts.

Phagotroph group

Taxon

Ciliates

Paramecium Euplotes Blepharisma Oxytricha Holosticha Mayorella Microchlamys Thecamoeba Willaertia Cochliopodium Unidentified rhabdocoel Lepadella Unidentified bdelloid (×3 types) Dicranophorus Chaetonotus

Amoebae

Platyhelminth Rotifers

Gastrotrich

Waterborne Cryptosporidium Oocysts (A)

205

(B)

(C)

20

Fig. 16.5. Paramecium sp. containing ingested oocysts of Cryptosporidium parvum. (A) Oocysts are clearly visible in the food vacuoles (arrow) when viewed under differential interference contrast (DIC) optics. (B) The sporozoite nuclei within oocysts (arrows) can be seen with DAPI staining. (C) Fluorescent antibody-stained oocysts are visible (small arrow), with oocysts with decreased staining or signs of degradation indicated by a larger arrow. Bar = 20 µm.

Table 16.3. index days. UV indexa 1 3 4 4 7 10 11 12

Inactivation of Cryptosporidium oocysts in tap water on different UV

S90 (kJ/m2)b

T90 (h)c

13,200 8,700 5,300 6,200 11,900 4,400 5,900 76

6.4 3.7 1.8 1.7 2.6 0.9 1.0 0.4

a UV

index = estimated dose at solar noon, 1 = 25 mW/m2 UV between 290 and 400 nm. = the insolation necessary to achieve a 90% reduction in cell culture infectivity. c T90 = the time necessary to achieve a 90% reduction in cell culture infectivity. b S90

used tap water and assessed solar irradiation on days with a different UV index. Days with a UV index of 10 or greater resulted in rapid inactivation (Table 16.3). Dark controls (samples wrapped in aluminium foil), kept under the same conditions as the light-exposed oocysts, exhibited no change in infectivity compared with oocysts stored at 4°C, demonstrating that temperature did not affect oocyst infectivity within the time frame of these experiments.

206

B. King et al.

Solar inactivation was next assessed for reservoir waters that varied in turbidity, colour and dissolved organic carbon (DOC) (Table 16.4). Waters with high DOC required higher doses of solar irradiation to achieve oocyst inactivation (Table 16.5). Long pass filter experiments were conducted to determine the biologically active components of solar radiation. The component responsible for the majority of inactivation was found to be UV-B, with a small amount of inactivation attributed to UV-A. Visible light had no measurable effect, as determined by the cell culture infectivity assay. The level of UV-induced DNA damage was assessed using an antibody directed against cyclobutane thymine dimers and also by quantitative sequence detection using real-time PCR. The amount of damage detected was considerably less than expected considering the level of inactivation observed when compared to similar levels of oocyst inactivation using UV-C (254 nm), suggesting that other mechanisms in addition to DNA damage are responsible for oocyst inactivation.

Water treatment Conventional water treatment processes remove Cryptosporidium oocysts through coagulation, flocculation, sedimentation and filtration. Little is known regarding the effect of these processes on oocyst infectivity or whether exposure Table 16.4. ments.

Characteristics of reservoir water used for solar inactivation experi-

Water type Bolivar tap Warragamba Prospect Lake Hope Valley Myponga

pH

Dissolved organic carbon (mg/l)

7.80 7.45 7.15 7.78 7.55

2.8 2.6 3.6 8.0 12.3

Turbidity (NTU)

Colour (HU)

0.238 2.94 2.26 7.42 4.12

2 29 47 38 77

Table 16.5. Inactivation of Cryptosporidium oocysts in reservoir waters on different UV index days. Winter: UV index 3 Water type Bolivar tap Warragamba Prospect Hope Valley Myponga

S90 (kJ/m2) 9,300 7,900 13,300 14,751 27,933

T90 (h) 3.2 2.6 4.5 4.88 10.75

Summer: UV index 10 S90(kJ/m2) 4,500 5,800 5,200 13,500 32,599

T90 (h) 0.85 1.1 1.0 2.5 6.5

Waterborne Cryptosporidium Oocysts

207

to these processes makes oocysts more susceptible to standard disinfection. The effect of the water treatment chemical alum on Cryptosporidium oocyst infectivity was assessed using cell culture infectivity. Coagulation by doses of alum ranging from 40 mg/l to 100 mg/l had no effect on oocyst infectivity (either on oocysts left in suspension or oocysts recovered from flocs). Dissolved air flotation of oocysts following alum coagulation also had no effect on oocyst infectivity. There was no change in the sensitivity of oocysts to chlorine or monochloramine following alum coagulation. The effect of age and temperature on oocyst sensitivity to chlorine was assessed (up to 6 months at 4°C or 15°C). A small increase in oocyst sensitivity to chlorine (Ct = 1200 mg min/l), resulting in 0.5 0 1 log inactivation, was observed for oocysts that had been stored at either 4°C or 15°C for 20 weeks or more (Fig. 16.6).

Summary and Future Directions While environmentally tough and resistant to common disinfection chemicals, Cryptosporidium oocysts are not invincible and are susceptible to environmental stresses. The data produced from this work need to be incorporated into fate and transport models, as current models use inaccurate data for temperature inactivation, and solar inactivation data have either been absent from such models or derived from other organisms. This will not only allow prediction of the location of oocysts in a reservoir but may also allow some prediction of infectivity if environmental conditions (temperature, light, DOC) are known. Importantly we also

1 week 0

week 8

week 12

week 16

week 20

week 24

Log survival

0 -1 -2 -3 -4 -5 Incubation period No treatment 4

JT 4

JT+cl2

No treatment 15

JT 15

JT+cl2 15

Fig. 16.6. Inactivation of oocysts by chlorine as a function of oocyst age and exposure to different temperatures.

208

B. King et al.

identify the need for further research effort directed towards biotic inactivation and removal. Questions which need to be answered include: ● ● ●

Are excreted oocysts still infectious? How does predation affect oocyst resistance to common disinfectants? Can such oocysts still be efficiently recovered by standard immunomagnetic separation (IMS) concentration?

Such work will provide valuable information for determining the relative risks associated with Cryptosporidium oocysts in water.

Acknowledgements We acknowledge the financial support received from the Co-operative Research Centre for Water Quality and Treatment, Water Services Association Australia, Australian Water Quality Centre, and South Australian Water Corporation. We thank David Daminato, Stella Fanok, Bret Robinson, Daniel Hoefel and Kylie Harvey for their excellent technical expertise and fruitful discussions.

References Black, E.K., Finch, G.R., Taghi-Kilani, R. and Belosevic, M. (1996) Comparison of assays for Cryptosporidium parvum oocysts viability after chemical disinfection. FEMS Microbiology Letters 135, 187–189. Bukhari, Z., Marshall, M.M., Korich, D.G., Fricker, C.R., Smith, H.V., Rosen, J. and Clancy, J.L. (2000) Comparison of Cryptosporidium parvum viability and infectivity assays following ozone treatment of oocysts. Applied and Environmental Microbiology 66, 2972–2980. Chauret, C., Nolan, K., Chen, P., Springthorpe, S. and Sattar, S. (1998) Aging of Cryptosporidium parvum oocysts in river water and their susceptibility to disinfection by chlorine and monochloramine. Canadian Journal of Microbiology 44, 1154–1160. Finch, G.R., Black, E.K., Gyürék, L. and Belosevic, M. (1993) Ozone inactivation of Cryptosporidium parvum in demand-free phosphate buffer determined by in vitro excystation and animal infectivity. Applied and Environmental Microbiology 59, 4203–4210. Johnson, D.C., Enriquez, C.E., Pepper, I.L., Gerba, C.P. and Rose, J.B. (1997) Survival of Giardia, Cryptosporidium, poliovirus and Salmonella in marine waters. Water Science and Technology 35, 261–268. Keegan, A.R., Fanok, S., Monis, P.T. and Saint, C.P. (2003) Cell culture-Taqman PCR assay for evaluation of Cryptosporidium parvum disinfection. Applied and Environmental Microbiology 69, 2505–2511. King, B.J., Keegan, A.R., Monis, P.T. and Saint, C.P. (2005) Environmental temperature controls Cryptosporidium oocyst metabolic rate and associated retention of infectivity. Applied and Environmental Microbiology 71, 3848–3857. King, B.J., Hoefel, D., Daminato, D.P., Fanok, S. and Monis, P.T. (2007) Solar UV reduces Cryptosporidium parvum oocyst infectivity in environmental waters. Journal of Applied Microbiology 104, 1311–1323.

Waterborne Cryptosporidium Oocysts

209

Korich, D.G., Mead, J.R., Madore, M.S., Sinclair, N.A. and Sterling, C.R. (1990) Effects of ozone, chlorine dioxide, chlorine, and monochloramine on Cryptosporidium parvum oocyst viability. Applied and Environmental Microbiology 56, 1423–1428. Medema, G.J., Bahar, M. and Schets, F.M. (1997) Survival of Cryptosporidium parvum, Escherichia coli, faecal enterococci and Clostridium perfringens in river water: influence of temperature and autochthonous microorganisms. Water Science and Technology 35, 249–252. Messner, M.J., Chappell, C.L. and Okhuysen, P.C. (2001) Risk assessment for Cryptosporidium: a hierarchical Bayesian analysis of human dose response data. Water Research 35, 3934–3940. Robertson, L.J., Campbell, A.T. and Smith, H.V. (1992) Survival of Cryptosporidium parvum oocysts under various environmental pressures. Applied and Environmental Microbiology 58, 3494–3500. Rose, J.B, Lisle, J.T. and LeChevallier, M. (1997) Waterborne cryptosporidiosis: incidence, outbreaks, and treatment strategies. In: Fayer, R. (ed.) Cryptosporidium and Cryptosporidiosis. CRC Press, Boca Raton, Florida, pp. 93–109. Smith, H.V. and Rose, J.B. (1990) Waterborne cryptosporidiosis. Parasitology Today 6, 8–12.

17

Methods for Genotyping and Subgenotyping Cryptosporidium spp. Oocysts Isolated During Water and Food Monitoring

H.V. SMITH, R.A.B. NICHOLS, L. CONNELLY AND C.B. SULLIVAN Scottish Parasite Diagnostic Laboratory, Glasgow, UK

Abstract Cryptosporidium oocysts are frequent contaminants of water, with contributions from infected human and non-human hosts, livestock and agricultural practices, and infected feral and transport hosts. Numerous waterborne outbreaks of cryptosporidiosis have been documented and as oocysts occur at low densities in water, methods which can detect and determine the genotype and subgenotype of small numbers of organisms reliably and reproducibly from water and food concentrates are required. Drinking water quality is also an important component of food production. Oocysts can enter the food chain from livestock and agricultural practices and from sewage effluent. Sensitive molecular methods are required for determining Cryptosporidium species, genotypes and subgenotypes in water and in/on foods. Sensitivity of detection can be increased by amplifying loci on multi-copy genes and polymerase chain reaction (PCR) amplification of loci in the 18S rRNA gene is considered to be the most suitable approach, as they can provide information about more species than single-copy loci, and have been more widely accepted worldwide. Representatives of the 16 valid Cryptosporidium species and the 44+ genotypes can be found in environmental, water and food concentrates, everywhere in the world, which raises significant issues regarding approaches to determine their presence, particularly if they are present as mixtures. Primer specificity should be evaluated against organisms that are common contaminants of water and food matrices, particularly those that are closely related to Cryptosporidium phylogenetically. Typing and subtyping systems used for human and non-human samples should also be used for environmental samples, particularly for source and disease tracking and risk assessment, in order to avoid any confusion arising from using different systems for human and non-human hosts and environmental samples for veterinary and public health investigation of disease outbreaks. 210

© CAB International 2009. Giardia and Cryptosporidium: From Molecules to Disease (eds G. Ortega-Pierres et al.)

Methods for Genotyping and Subgenotyping Cryptosporidium spp.

211

Introduction Cryptosporidium is an apicomplexan, obligate protozoan parasite that causes enteritis in human hosts. First described in the intestines of laboratory mice (Tyzzer, 1912), Cryptosporidium gained importance in the 1970s, when it was found to cause diarrhoea in young calves and humans. Since then it has emerged as a parasite of worldwide importance and has acquired special relevance as an opportunistic pathogen of people with acquired immune deficiency syndrome (AIDS). Recent genetic analyses have raised doubt about the validity of the current classification of the genus Cryptosporidium and reveal that more than one species of Cryptosporidium can infect susceptible human hosts (Table 17.1). There are 16 ‘valid’ Cryptosporidium species and a further 40+ genotypes, which differ significantly in their molecular signatures but, as yet, have not been ascribed species status (Smith et al., 2007). Seven described Cryptosporidium species (C. hominis, C. parvum, C. meleagridis, C. felis, C. canis, C. suis and C. muris) (Table 17.1) and two undescribed species of Cryptosporidium (cervine and monkey) infect immunocompetent and immunocompromised humans (Xiao et al., 2004; Cacciò et al., 2005), but C. hominis and C. parvum are the most commonly detected (Cacciò et al., 2005). Human cryptosporidiosis is characterized by profuse, watery diarrhoea, ’flu-like illness, malaise, abdominal pain, anorexia, nausea, flatulence, malabsorption, vomiting, mild fever and weight loss. In immunocompetent individuals, cryptosporidiosis is self-limiting, but it can become chronic in immunocompromised individuals. There is no recognized effective drug treatment for human cryptosporidiosis. The C. parvum ID50 for seronegative adult human volunteers is isolate-dependent and ranges from 9 to 1042 oocysts (Okhuysen et al., 1999), while the C. hominis ID50 (isolate TU502) is 10–83 oocysts (Chappell et al., 2006).

Table 17.1.

Some differences among Cryptosporidium species infecting humans.

Species

Oocyst dimensions (µm)

Site of infection

Major host

4.5 × 5.5 4.5 × 5.5

Small intestine Small intestine

C. suis

5.05 × 4.41

Small intestine

Humans Neonatal mammalian livestock, humans Pigs

C. felis

4.5 × 5.0

Small intestine

Cats

C. canis

4.95 × 4.71

Small intestine

Dogs

4.5–4.0 × 4.6–5.2

Intestine

Turkeys

5.5 × 7.4

Stomach

Rodents

Stomach

Cattle

C. hominis C. parvum

C. meleagridis C. muris C. andersoni

5.6 × 7.4 (5–6.5 × 8.1–6.0)

212

H.V. Smith et al.

Worldwide, the incidence of human cryptosporidiosis is higher in underdeveloped or developing countries, in children, and in immunocompromised individuals. The incidence of human cryptosporidiosis has been estimated at 3.3 cases per 100,000 individuals in 12 European countries, and in the USA Cryptosporidium infections are estimated to cause disease in 300,000 individuals annually. The incidence of waterborne and foodborne cryptosporidiosis worldwide is difficult, if not impossible, to estimate accurately, since the source of infection for sporadic cases, which constitute the majority of human cryptosporidiosis cases, is very seldom determined. Transmission can occur by direct contact with an infected person or animal, or indirectly via contaminated water or food. Most commonly, zoonotic transmission occurs as a consequence of cryptosporidiosis in neonatal calves and lambs, which can excrete up to 109 oocysts/g faeces (the environmentally resistant and infective form of the parasite). Once in the environment, infectious oocysts can contaminate water sources and food.

Oocyst Contributions to Water and Food Cryptosporidium oocysts are frequent contaminants of water, with contributions from infected human and non-human hosts, livestock and agricultural practices, and infected feral and transport hosts (Smith et al., 1995; Smith and Grimason, 2003). Water is an important transmission route, with at least 165 waterborne outbreaks of cryptosporidiosis documented (Girdwood and Smith, 1999; Fayer et al., 2000; Slifko et al., 2000; Karanis et al., 2007). Oocysts occur at low densities in water (Smith and Rose, 1990, 1998; Smith et al., 1995; Smith and Grimason, 2003), and methods which can detect and determine the genotype and subgenotype of small numbers of organisms reliably and reproducibly from water and food concentrates are required. Oocysts are resistant to numerous disinfectants normally used in water treatment, and water is an important component of food production from crop irrigation, harvesting, sorting, to storage and distribution. In the food industry, potable water, uncontaminated with infectious oocysts, is needed for the preparation of ready-to-eat food and for the dilution of beverages. Oocysts can enter the food chain from agricultural practices such as muck-spreading and slurry spraying of oocyst-contaminated, animal-derived faecal material onto land used for cultivation and from animals pasturing near crops intended for human consumption. The use of contaminated, untreated sewage (and waste stabilization pond) effluents and untreated water for crop irrigation can also contaminate crops. Runoff from, and percolation through, contaminated pasture and soils can contaminate adjacent waterbodies, and oocysts transported into rivers and marine estuaries can lead to the contamination of shellfish, which are frequently eaten raw or lightly cooked. Infectious oocyst contamination of drinking water and foods that are to be eaten raw or with minimum heating is associated with a high risk of transmitting cryptosporidiosis. The number of oocysts which a method can detect should realistically be below its infectious dose. Given that the ID50 for C. parvum and

Methods for Genotyping and Subgenotyping Cryptosporidium spp.

213

C. hominis isolates ranges from 9 to 1042 oocysts, sensitive methods with consistent recovery efficiencies are required for reliable risk assessment. Current Cryptosporidium detection methods include extraction of oocysts from water and food matrices, concentration of oocysts from the eluted material, and detection by epifluorescence and Nomarski differential interference contrast microscopy (Anonymous, 1999a, 1999b, 2005; USEPA, 2001; Smith and Cook, 2008). Experimental recovery efficiencies from water and food matrices range from 1% to 59% (Smith et al., 2001; Nichols and Smith, 2002).

Methods for Water and Foods Standardized methods are available for isolating and enumerating Cryptosporidium oocysts in water, many of whose component parts have been developed in the authors’ laboratory (e.g. Smith et al., 1989, Grimason et al., 1994; Campbell and Smith, 1997). Immunomagnetic separation (IMS) has standardized and increased oocyst recoveries from water concentrates (Smith et al., 2001). The benefits of IMS, in capturing oocysts from crude samples and concentrating and processing them in a buffer free of PCR inhibitors, increases the sensitivity of detection (Smith, 1996), and this approach has been used to genotype oocysts in water and foods (e.g. Xiao et al., 2000, 2001; Nichols et al., 2002, 2003; Jiang et al., 2005). There is no standardized method for isolating and enumerating Cryptosporidium oocysts from foods. Nichols and Smith (2002) reviewed published methods for isolating and enumerating oocysts isolated from various liquid and solid foodstuffs and identified a broad range (1–59%) in the recovery efficiencies of these methods. Some used IMS while others did not. Commercial IMS kits for concentrating oocysts from water concentrates have also been used for concentrating oocysts from food concentrates (Robertson and Gjerde, 2000, 2001; Robertson et al., 2002), but kits optimized for water concentrates are not necessarily optimized for food matrices (Cook et al., 2006a; Smith and Cook, 2008). Cook et al. (2006a) developed and validated (Cook et al., 2006b) a method with the aim of producing standard protocols for detecting Cryptosporidium and Giardia in foods. Component parts were adapted from methods developed for detecting Cryptosporidium and Giardia in water, but specific new sample treatments had to be developed for primary extraction of oocysts and cysts from food samples. Importantly, these treatments were optimized, by varying the physicochemical parameters, to produce sample extracts which were compatible with the IMS and microscopy materials. The optimization was monitored through the recovery efficiencies achieved through parameter variation (Smith and Cook, 2008). The validated method, developed to determine oocyst contamination on lettuce and raspberries, has been used to genotype oocysts on a variety of foods. For C. parvum, the lowest median infectious dose for adult human volunteers is nine oocysts (Okhuysen et al., 1999). To detect this level of contamination, a minimum of one oocyst should be recovered from the extract and end up on the microscope slide where it can be seen. This would necessitate a recovery

214

H.V. Smith et al.

efficiency of at least 11%. The method of Cook et al. (2006a, 2006b) has a higher recovery efficiency, being 59.0 ± 12.0% for lettuce and 41.0 ± 13.0% for raspberries.

What’s Out There? Molecular identification methods can be performed on samples which are suitable for microscopic evaluation, and DNA extraction can be performed either before oocysts are dissociated from magnetizable beads or following their enumeration on microscope slides. One advantage of using microscope slides where oocysts have been detected by microscopy is that both intact and empty oocysts can be observed. Empty oocysts cannot be amplified by molecular methods as they do not contain sporozoite DNA, and this will lead to an underestimation of (i) oocyst contamination and (ii) the efficiency of physical removal processes used for water and food. Since not all Cryptosporidium species that can contaminate waters and foods are infectious to humans, Cryptosporidium species, genotype and subgenotype identification using polymerase chain reaction (PCR)-based methods augments risk assessment. The viability/infectivity of oocysts is also of importance for assessing risk. Currently, there is no accepted method for determining the infectivity/viability of C. parvum oocysts recovered from water or food. Investigation into the extent of the occurrence of different species/genotypes of Cryptosporidium in the environment and in/on foods is only now being addressed. Previously, we showed that oocyst staining with both commercially available fluorescein isothiocyanate (FITC)-labelled monoclonal antibodies that recognize exposed epitopes on oocyst walls and the nuclear fluorogen 4′,6diamidino-2-phenylindole (DAPI), which are used in the standardized methods for water and methods for foods (see below), are amenable to PCR amplification (Nichols et al., 2003), validating the use of these samples. This approach has been adopted in many laboratories investigating environmental contamination by Cryptosporidium. Discordant results between PCR and microscopy were observed during analysis of storm water samples: 10 microscopy-negative samples were PCRpositive (36 of 42 samples were PCR-positive). This occurred within duplicates of the same subsample. Other variables that affected PCR positivity included the volume of template used per reaction and the number of species/genotypes present in the sample (see below). Samples with more than one species/genotype had a PCR positivity rate of 73% compared with 34% for those containing a single species/genotype (Xiao et al., 2006). Methods targeting 18S rRNA gene loci are the most widely used for identifying Cryptosporidium species/genotypes in human and non-human hosts and environmental (water and food) samples, and are advocated here. As there are 20 copies of the Cryptosporidium ribosomal DNA gene present per oocyst (LeBlancq et al., 1997), PCR amplification of multi-copy genes is a practical approach to sensitive molecular detection. Furthermore, as the complete sequences of the SSU rRNA gene from different Cryptosporidium species, originating from a variety

Methods for Genotyping and Subgenotyping Cryptosporidium spp.

215

of hosts, are available from the GenBank database, this gene has been exploited the most for species identification. Much of the earlier data were derived from PCR-restriction fragment length polymorphism (RFLP) analysis, but it has become clear that this approach often has insufficient discrimination for some of the recognized Cryptosporidium species/genotypes found in environmental samples, and that DNA sequencing is the preferred option. Where possible, a multilocus approach to characterizing Cryptosporidium isolates is essential for accuracy, and various 18S and other loci are available for species (and subtype) determination; however, single copy loci may not have sufficient sensitivity for detecting the small numbers (>10) of oocysts frequently found in environmental samples.

Cryptosporidium and Water The analysis of US storm water samples revealed the presence of Cryptosporidium spp. in 27 of 29 samples, mainly wildlife Cryptosporidium genotypes (Xiao et al., 2001). The most common species/genotypes found in surface waters were C. parvum, C. hominis and C. andersoni, with C. andersoni reported to be the most commonly found in wastewater (eight samples). Storm waters are expected to carry greater microbial loads, including larger numbers of Cryptosporidium oocysts (Kistemann et al., 2002). In an analysis of 121 water samples from storm events in three New York area watersheds (Jiang et al., 2005), 107 contained amplifiable Cryptosporidium DNA, and, of the 22 Cryptosporidium species and genotypes identified, only 11 were of known species or genotypes. Reliable detection of oocysts in storm waters using EPA method 1623 (0.5 ml packed pellet volume after filtering 20l samples) depends on the analysis of repeated samples and subsamples, both by microscopy and PCR. Limited subsampling can lead to an underestimation of both the number of oocysts and the number of species/genotypes present (Xiao et al., 2006). An analysis of oocyst-positive water concentrates on microscope slides from four UK water companies/water utilities revealed the presence of C. muris or C. andersoni, C. parvum or C. hominis, C. meleagridis or Cryptosporidium cervine, ferret or mouse genotypes, all of which can be discriminated by further restriction endonuclease digestion (Nichols et al., 2006a). Of a total of 33 slides analysed, 32 produced PCR products, but 10 were in poor condition, making microscopic re-evaluation impossible. All positive slides generated PCR-RFLP patterns consistent with published data. Two slides contained larger (~8 × 6 µm) oocysts, and PCR-RFLP analysis revealed the presence of C. andersoni DNA (Nichols et al., 2006a). In a further study of 1039 oocyst-positive slides, 56.1% of which were from Scottish final water samples and 44% from Scottish raw waters, 601 (57.7%) samples were positive at one or two 18S rRNA gene loci. The most common sample contained one oocyst, which accounted for 45.1% of all samples, and the majority of samples contained between one and five oocysts (81% of the dataset). In 71 of the Scottish final water samples, human-infectious species/genotypes were detected: C. parvum in 67.6%, C. hominis in 18.3%, C. meleagridis in 1.4% and the Cryptosporidium cervine genotype in 30.9% of samples (H.V. Smith et al., unpublished). Other species/genotypes detected

216

H.V. Smith et al.

included C. muris, C. andersoni, C. baileyi, C. bovis, Cryptosporidium muskrat genotypes I and II, Cryptosporidium cervine genotype, Cryptosporidium opossum genotype I, Cryptosporidium deer-like genotype, and six previously undescribed Cryptosporidium genotypes.

Cryptosporidium and Food A preliminary analysis of Cryptosporidium contamination of vegetables in Polish local markets involved analysing 21 samples of local produce (cabbage (various), leek, spinach, lettuce, green onions, broccoli, celery, cauliflower). Samples were purchased from markets in locations based on the number of homesteads in the area which had >30 cows in one herd. Using the method of Cook et al. (2006a), six of the 21 samples were oocyst-positive. By plotting where samples were purchased against the number of homesteads which had >30 cows in one herd, we determined that all oocyst-positive samples were purchased in local markets which had the highest density of homesteads with >30 cows in one herd (range 41–99 homesteads). Oocyst-negative samples were purchased in local markets where the density of homesteads with >30 cows in one herd ranged from 0 to 99 (H.V. Smith and A. Rzezutka, unpublished). The number of oocysts detected ranged from 2 to 35, and two samples contained between 1 and 2 excysted oocysts. Of these six positive samples, three were PCR-positive, and RFLP and sequencing revealed them to contain C. parvum DNA. Cryptosporidium parvum subgenotyping, at the GP60 locus and by multilocus genotyping using mini- and microsatellites (Mallon et al., 2003a, 2003b), is under way. Clearly, these data identify the breadth of species and genotypes detected, and it is likely that representatives of the 16 valid species and the 44+ genotypes can be found in environmental, water and food concentrates worldwide, which raises significant issues regarding approaches to determine their presence. Furthermore, where mixtures of species/genotypes occur, both PCR-RFLP and sequencing options can be compromised. Prior to the routine adoption of molecular methods for investigating Cryptosporidium contamination of the environment, both the variability between methods and the recognized difficulties in amplifying nucleic acids from environmental specimens by PCR must be overcome.

Towards Standardizing Approaches for Determining Cryptosporidium Oocyst Contamination of the Environment, Water and Foods Methods that can genotype small numbers of Cryptosporidium oocysts reliably and reproducibly are required to determine which species/genotypes/ subgenotypes are present, and with what frequency, in water and in/on foods. A prerequisite to identifying sources of human infection and transmission routes is the requirement to define human infective parasites. Environmental matrices contain many inhibitory substances in varying quantities, which will decrease the sensitivity of detection. This demands more effective methods both for

Methods for Genotyping and Subgenotyping Cryptosporidium spp.

217

neutralizing inhibitory effects and for extracting nucleic acids. Much effort has been directed towards the efficient extraction of Cryptosporidium DNA, selection of Cryptosporidium-specific primers and overcoming the effects of inhibitors such as clays, humic and fulvic acids, polysaccharides and other organic compounds, salts and heavy metals, etc. (Sluter et al., 1997; Nichols et al., 2003; Nichols and Smith, 2004; Sunnotel et al., 2006).

DNA extraction DNA extraction is at the centre of efficient PCR amplification and the detection of small numbers of oocysts by molecular methods. A standard, maximized method for DNA extraction from Cryptosporidium oocysts is essential both for detecting small numbers of oocysts and for evaluating the sensitivity of detection by PCR using different primers. Disruption of the robust oocyst wall is a prerequisite for releasing sporozoite nuclei and effective DNA extraction, while the liberation of DNA from bound protein is essential both for efficient primer annealing and for successful PCR amplification. Oocyst wall disruption following freezing by immersion in liquid nitrogen and thawing is the preferred method for the release of sporozoite DNA from small numbers of intact oocysts; however, the optimum temperature reported in the literature for thawing and liberation of oocyst contents varies from 37°C to 100°C (Johnson et al., 1995; Mayer and Palmer, 1996; Rochelle et al., 1997; Sluter et al., 1997; Chung et al., 1998). Furthermore, the number of freeze–thaw cycles and the medium for DNA extraction also vary (Laberge et al., 1996; Leng et al., 1996; Deng et al., 1997; Rochelle et al., 1997; Sluter et al., 1997; Chung et al., 1998; Kaucner and Stinear, 1998). Not only have different C. parvum isolates been used by these researchers, but also different protocols, which underlines the need for maximizing and standardizing oocyst disruption and DNA extraction protocols, especially when dealing with small numbers of oocysts. Nichols and Smith (2004) described a method that maximizes DNA extraction reliably from small numbers of partially purified or purified oocysts. Sporozoite DNA was liberated from C. parvum oocysts by 15 cycles of freezing (liquid nitrogen) and thawing (65°C) in lysis buffer containing sodium dodecyl sulphate (SDS). The inhibitory effects of sodium dodecyl sulphate were abrogated by the addition of Tween 20 to the PCR reaction. Seven different C. parvum oocyst isolates were tested, and the method detected fewer than five oocysts following direct PCR amplification of a segment of the 18S rRNA gene. Older oocysts, which were more refractory to freeze–thawing, were disrupted effectively. These authors recommended 15 cycles of freeze–thawing to maximize oocyst disruption and DNA extraction, particularly when isolate history and oocyst age were unknown.

Sensitivity of 18S rRNA gene loci for genotyping oocysts from water and food In addition to efficient DNA extraction, an optimized protocol for genotyping Cryptosporidium oocysts present on microscope slides must include the use of

218

H.V. Smith et al.

genetic loci that maximize sensitivity. No ‘standard’ genetic locus exists for determining species identity, but RFLP or sequencing of 18S rRNA gene loci provide information about more species than single-copy loci for small numbers of oocysts. For detecting small numbers of oocysts (