Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author.

Epidemiological Studies of Cryptosporidiosis

A thesis presented in partial fulfilment of the requirements for the degree of

Doctor of Philosophy

in Veterinary Pathology

Massey University, Palmerston North, New Zealand

Alejandro Grinberg

2009

i i

ABSTRACT

A. G rinberg (2009). Doctoral thesis, Massey U niversity, Palmerston North, New Zealand.

An i nterpretive overview of the l iterature on intest inal cryptosporidiosis i n humans and domestic

mammals (Chapter 1 ) is fol lowed by two studies of the popu latio n genetic structure of the

protozoan parasites Cryptosporidium parvum and Cryptosporidium hominis (Chapter 2 ), f ive

epidemiological studies of cryptosporidiosis in foals, calves and humans in New Zealand

(Chapter 3) , and an i nvestigation of a serendipitous outbreak of cryptosporidiosis among a class

of veterinary students, which occurred at the end of 2006 (Chapter 4 ) .

The analysis o f the popu lation genetic structure of C. parvum and C. hominis i nd icates the

existence of a s ign if icant genetic segregat ion of geograph ically separated parasite populations,

consistent with al lopatry. The resu lts do not conform to a simpl istic model that considers al l C.

parvum as multi -host anthropozoonotic agents, and provide statistical support to the idea of the

occurrence of anthroponotic cycles that do not involve cattle. Rather than conform ing to a r igid

parad igm of either a clonal or a pan m ictic species, data are consistent with the eo-occurrence of

clonal and recombinatorial diversif ication in C. hominis, and perhaps C. parvum.

The resu lts of the epidemiological studies in New Zealand suggest cryptosporidiosis caused by

C. parvum is relatively common i n young foals and calves . I n the ti me and space frames

underlying these studies, humans, calves, and foals were i nfected with a genetically

homogeneous C. parvum population . This feature is in accordance with previous reports that

have indicated C. parvum as the dominant species in humans duri ng the peaks of i ncidence of

cryptosporidiosis in winter and spring , and support the view that the peaks are i n large part

attributable to di rect and/or ind i rect zoonotic transmission of C. parvum.

Final ly, the outbreak of cryptosporidiosis among a class of veterinary students h igh l ighted the

potential hazard for explosive large-scale outbreaks in New Zealand. The results of the

investigation were consistent with point-source exposure and zoonotic transm ission of a rare C.

parvum subtype through di rect contact with calves during a practicum .

i i i

PREFACE

With the advent of HIV-AIDS and the re-emergence of the importance of i nfectious diseases at

the end of the 20 1h century, several microorganisms not previously known to cause d isease,

i ncluding members of the genus Cryptosporidium, were recogn ised as novel aetiolog ical agents

in humans and an i mals. In the i naugural issue of the journal Emerging Infectious Diseases i n

1 995, D r . David Satcher, Di rector of the US Centers for Disease Control and Prevention,

i ncluded Cryptosporidium i n a g roup of m icroorgan isms which i n h is opin ion were the "major

etio logic agents" identified since 1 973.

Cryptosporidium i s a genus of protozoan parasites first described in animals in 1 907. Despite

the early description of these parasites, cryptosporidiosis, that is, the disease caused by

members of the genus, was i n it ia l ly described in 1 97 1 i n a heifer and then in 1976 i n a 3-year

old g i rl from a farm, both in the US. The most com mon cl in ical presentation of cryptosporidiosis

in developed countries is of self- l i m it ing d iarrhoea! i l l ness. However, in H I V-positive or otherwise

immunocompromised patients, cryptosporidiosis may man ifest as a chronic debi l itat ing o r

fu lm inant disease. Conversely, i n many deve lopi ng reg ions o f the world the infections with

Cryptosporidium are associated with persistent ch i ldhood diarrhoea, h igh mortality,

m alnour ishment, and stunted g rowth.

Cryptosporidiosis is a notif iable d isease I n New Zealand. According to a recent Publ ic Health

Su rveillance Report, the rate of notifications of cryptosporidiosis in the tr imester October­

December 2008 was of >2 cases per 1 0 ,000 population, s im i lar to the rate of salmonellosis and

g iardiasis over the same period (http://www.surv.esr.cr i . nz/PDF _survei llance/, accessed Apri l

200 9). I n add it ion to the endem ism of cryptosporidiosis, Cryptosporidium parasites pose a

s ign ificant hazard due to the resistance of the oocysts to chlorination and the potential to cause

-massive-water=borne-disease-outbreaks;-stJeh-as-the- epieemie--iA-Mi lwaltk.ee,--W-isc-e-AsiA-iA-

1 993 , when an estimate of 400,000 people acqui red an infection through the i ngestion of

contami nated mun icipal water supply. Thus, ensur ing preparedness against outbreaks of

cryptosporidiosis i s critical.

Many Cryptosporidium taxa can infect both humans and an imals, and infections in humans are

often acquired zoonotical ly, through di rect or i ndirect contact with the faeces of i nfected

an imals. Thus, understand ing the biology and epidemiology of cryptosporidiosis in different host

species is important, to both enhance the health of animals, and also to devise strategies for the

control of the zoonotic spread of the parasites.

Th is PhD thesis i ncludes an interpretive overview of the relevant l iterature , and accounts of

e ight epidem io log ical studies of Cryptosporidium i nfections i n humans and an imals undertaken

between 2002 and 2007 by the author. A s imple search in Medl ine between 1 995-2008 using

the keyword 'Cryptosporidium' returned 2343 citations. Therefore, the overview of the l i terature

iv

reports on ly those articles that in the author's view shaped the th ink ing i n each particular area;

h owever, further articles are cited in the sections for the individual studies. Six of the eight

studies have been publ ished i n peer-reviewed journals. To integrate the i ndividual studies i nto

the whole picture and avoid corrupti ng the publ ished manuscripts, each study is preceded by an

I ntroduction . Some successive studies have been published in d i fferent years. Thus, the

l iterature cited m ay differ between studies, main ly as a reflection of the progressive

accumulation of publ ications on the same topics. The published m anuscripts have been s l ightly

m od if ied. Modif icat ions included the addit ion of cross-references and m aterial and methods that

- for conciseness - could not be included in the published m anuscripts. In addition , the

bibl iography was re-formatted according to the requirements of the New Zealand Veterinary

Journal and the relevant raw data were included i n Appendices.

V

ACKNOWLEDGMENTS

I express my frank appreciation to m y supervisors Prof. Bi l l Pom roy, Dr . N icolas Lopez­

Vi l la lobos , and Prof. G iovanni Widmer, for helping me to complete the studies and write this

thesis with enjoyment. I hope this has been a gratifyi ng endeavour for them too. I am grateful to

Prof. G rant Gui lford, former Head of IVABS, Prof. Kev in Stafford, D i rector of Post-graduate

Studies, Prof. Hugh Bla i r , Di rector of Research, and all staff at IVABS, for creating a su itable

research environment that faci l itated my work.

The names of friends and colleagues who he lped me to perform the studies have been i nc luded

in the l ist of authors of the relevant publ ished papers. P rof . Andy Tait, Wellcome Centre for

Mo lecu lar Parasito logy, Glasgow, Un ited Kingdom , generously donated the raw data used in

the study presented i n Section 2 . 1 . Prof. Anne Chao, N ational Tsing Hua Un iversity, Taiwan,

calculated the 95% confidence intervals of the Chao 1 and ACE1 richness esti mates in the same

study. The molecu lar characterisat ion of the Cryptosporidium isolates used in the study

presented in Section 2.2 was performed by Dr. Sultan Tanriverd i , Divis ion of Infectious

Diseases, Cummings School of Veter inary Medic ine, Tufts Un iversity, MA, USA. The studies

performed in Chapters 3 and 4 uti l ised several molecular methods developed over the years at

the Protozoa Research Un it, Massey Un iversity, Palmerston North , and compi led in the Manual

of Methods for Genotyping Cryptosporidium Oocysts from Faecal Specimens, written i n

November 2002 by M rs K im Ebbett. Mr. Errol Kwan, Anthony Pita and the late J im Learmonth

performed some laboratory analyses reported i n Sect ions 3 . 1 -3 .4 . Assistance in the laboratory

was also provided by Mr . Yi Shi, An i m al Health Monitori n g and Disease P revent ion Uni t , U rumqi

City, X inj iang, People's Republ ic of Ch ina , dur ing h is stay in New Zealand as a vis it ing scholar

under te author's supervision in 2007-2008 . Other peop le to whom I am very g rateful are Mr.

Graham Young , former bacterio log ist at Gribbles Diagnostic Laboratories, Ham ilton , Jan Bird ,

M icrobiology Section Leader of Path Lab Waikato , and Chris Pickett , Hamilton I'VIeCITCal

Laboratory, for provid ing Cryptosporidium-positive specimens from cattle and humans. Dr .

lsobel G ibson , New Zealand Veterinary Pathology Ltd . , kindly provided Cryptosporidium­

positive specimens from foals. The sym pathetic New Zealand farmers who al lowed me to take

samples from anim als under thei r care and the veterinary students who provided faecal

specimens and responded to the questionnaires reported in Chapter 4 are also thanked. Final ly ,

I thank Prof. Sau l Tzipor i and the staff of the Divis ion of Infectious Diseases, Cumm ings School

of Veter inary Medicine, Tufts Un iversity, for hosti ng me for two months in 2008 .

The Lewis Filch Veterinary Research Fund, McGeorge Research Fund, G raham Chalmers Al ien

Memorial Veter inary Scholarsh ip and the New Zealand Min istry of Health provided funding for

the projects. Addit ional funding was obtained from unrelated professional consu ltancy

assignments over the years. The studies presented in Section 3 .4 and Chapter 4 have been

approved by the An imal and Human Eth ics Committees of Massey Un iversity.

vi

For my beloved wife, Vicky

v i i

not by bread alone does man survive Deuteronomy 8:3

vi i i

LIST OF CONTENTS

1 INTESTINAL CRYPTOSPORIDIOSIS IN HUMANS AND DOMESTIC MAMMALS:

AN INTERPRETIVE OVERVIEW

1 .1 Cryptosporidiosis in the historical perspective

1 .2 Impact of cryptosporidiosis on human and animal health

1 .3 The taxonomic classification of genus Cryptosporidium

1 .4 The life cycle of the intestinal Cryptosporidium parasites

1 .5 Pathogenesis of intestinal cryptosporidiosis

1 .6 Infections with Cryptosporidium in domestic mammals

1 .6 . 1 Infections with Cryptosporidium in cattle

1 .6 .2 Infections with Cryptosporidium in small ruminants

1 .6.3 Infections with Cryptosporidium in horses

1 .6.4 Infections with Cryptosporidium in cervids

1 .6 .5 I nfections with Cryptosporidium in dogs and cats

1 .6.6 I nfections with Cryptosporidium in pigs

1 .7 Genetic typing of Cryptosporidium

1 .8 Cryptosporidium parvum and C. hominis population genetic structure

1 .9 Zoonotic cryptosporidiosis

1 .1 0 Concluding remarks

1 . 1 1 R eferences

2 STUDIES OF THE POPULATION GENETIC STRUCTURE OF

CRYPTOSPORIDIUM PARVUM AND CRYPTOSPORIDIUM HOMINIS

2

4

8

1 2

1 3

1 3

1 7

1 9

20

2 1

22

23

24

25

3 1

3 1

STUDIES O F THE POPULATION GENETIC STRUCTUR E O F CRYPTOSPOR/0/UM PARVUM 51

AND CRYPTOSPOR/0/UM HOMIN/S

2 . 1 Host-shaped segregation of the Cryptosporidium parvum multilocus genotype repertoire 52

2 . 1 .1 Summary 52

2 . 1 .2 Intra uct1on --&2--

2 . 1 .3 Materials and methods 53

2.1 .4 Results 54

2 . 1 .5 Discussion 57

2.2 Inferences about the global population structure of Cryptosporidium parvum and 60

Cryptosporidium hominis

2.2 . 1 Summary 60

2.2.2 Introduction 60

2.2.3 Materials and methods 61

2.2 .4 Results 64

2 .2 .5 Discussion 70

2.3 Concluding remarks 72

2.4 References 73

ix

3 EPIDEMIOLOGICAL STUDIES OF CRYPTOSPORIDIOSIS IN DOMESTIC

ANIMALS IN NEW ZEALAND

EPIDEMIOLOGICAL STUDIES OF CRYPTOSPORID IOSIS I N DOMESTIC

AN I MALS IN N EW ZEALAND

79

3.1 Identification of Cryptosporidium parvum 'cattle' genotype from a severe outbreak of neonatal 80

foal diarrhoea

3.1 . 1 Summary 80

3 . 1 .2 Introduction 80

3.1 .3 Materials and Methods 8 1

3. 1 .4 Resu lts 83

3 . 1 .5 Discussion 85

3.2 Genetic diversity and zoonotic potential of Cryptosporidium parvum causing foal diarrhoea 87

3.2. 1 Summary 87

3.2.2 Introduction 87

3.2.3 Materials and Methods 88

3.2.4 Resu lts 90

3.2.5 Discussion 92

3.3 A study of neonatal cryptosporidiosis of foals in New Zealand 94

3.3 . 1 Summary 94

3.3.2 I ntroduction 94

3.3.3 Materials and Methods 95

3.3.4 Results 97

3.3 .5 Discussion 1 0 1

3.4 The occurrence of Cryptosporidium parvum, Campylobacter and Salmonella in newborn 1 03

calves in the Manawatu region of New Zealand

3.4. 1 Summary 1 03

3.4.2 I ntroduction 1 03

3.4.3 Materials and Methods 1 05

-- 3-44---Hes�Jits. 1 07

3.4.5 Discussion 1 08

3.5 Persistent dominance of two GP60 !la al leles in diarrhoeagenic Cryptosporidium parvum from 1 1 2

man and cattle in the Waikato region of New Zealand

3 .5 . 1 Summary 1 1 2

3.5.2 Introduction 1 1 2

3.5.3 Materials and Methods 1 13

3.5.4 Results 1 1 4

3 .5 .5 Discussion 1 1 6

3.6 Conclud ing remarks 1 1 7

3. 7 References 1 1 8

X

4 AN OUTBREAK OF CRYPTOSPORIDIOSIS AMONG A COHORT OF YOUNG

ADULTS: MOLECULAR AND DESCRIPTIVE EPIDEMIOLOGY AND RISK FACTOR

ANALYSIS

4 . 1 An outbreak of cryptosporidiosis among a cohort of young adults: molecular and descriptive 1 33

epidemiology and risk factor analysis

4. 1 Summary 1 33

4.2 I ntroduction 1 33

4.3 Materials and Methods 1 34

4.4 Results 1 37

4 .5 Discussion 1 43

4.6 References 1 46

5 GENERAL DISCUSSION 1 5 1

APPENDICES 1 56

x i

LIST OF TABLES

Table 2 .1 Distribution of C. parvum multi locus g enotypes ( MLGs) i n Scotland, stratified by reg ion (Aberdeenshire or Dumfriessh i re), and host species (human or bovine C. 56 parvum)

Table 2.2 Rarefaction, Chao 1 and ACE 1 r ichness est imators, by comparison 56

Table 2.3 C. parvum and C. hominis l i nkage disequi l ibriu m and double-banded genotype 70 stat ist ics accordi ng to country of orig in

Table 3.1 Restriction fragment length polymorphisms between C. parvum 'catt le' and 84 'human' genotypes

Table 3.2 S ing le nucleotide polymorphisms and deduced amino acid changes i n the 60-kDa g lycoprote in of C. parvum isolates from foals, as compared with the or ig inal 91 sequence reported by Strong et a l . (2000)

Table 3.3 B i locus sequence types (BLST) of foal , human , and bovi ne C.parvum in New 92 Zealand

Table 3.4 H aematology, biochemistry and venous blood gas results from a hospital ised 99 foal affected with cryptosporidiosis

Table 3.5 Prevalence of C. parvum and Campylobacter spp among newborn calves from 1 08 24 dai ry farms i n the Manawatu region of N ew Zealand

Table 4.1 Demographic characteristics of the o utbreak of cryptosporidiosis among a class 1 40 of veterinary students reported in Section 4. 1

Table 4.2 Risk factor analysis of the outbreak of cryptosporidiosis reported i n Section 4 . 1 1 43

x i i

LIST OF FIGURES

Figure 1 . 1 The intestina l and extraintestinal life cycle of C.parvum 1 1

Figure 1 .2 Least square means of the log10 of ( 1 +number of C. parvum oocyst per gram 1 7

of faeces) i n 20 newborn calves affected with cryptosporidosis in a dairy farm i n Israel

Figure 1 .3 Nucleotide polymorphisms (in yel low) between the 1 8S rRNA gene 20 sequences of the so cal led "horse genotype", C. parvum, and C. wrairi

Figure 1 .4 Upper g raph: The n um ber of cases of cryptosporidiosis notified in New Zealand between 1 997 and 2007 ; lower graph : the seasonal shifts between the number 28 of C. parvum and C. hominis isolates identified in New Zealand between 2000 and 2003

Figure 1 .5 Dendrogram showing "human only sub-groups" of C. parvum multilocus genotypes in Scotland, as reported by Mal lon et al. (20 03) (above) and sing le and double 30 locus variant networks (SDLVN) of the same multilocus genotytpes (be low)

Figure 2.1 Rarefaction curves, by comparison 57

Figure 2.2 Single locus variant eBU RST networks for C. parvum and C. hominis 66

Figure 2.3 C. parvum and C. hominis M LG rank abundance plots 67

Figure 2.4 C. parvum and C. hominis analytical rarefaction curves 68

Figure 3.1 Fused and atrophic duodenal vi l li (arrow) and mildly increased numbers of 84 mononuclear inflammatory cel ls within the lamina propria

Figure 3.2 PCR-restriction fragment length polymorphism analysis of Cryptosporidium !3- 8 5 tubu lin , poly T , R N R and COWP genes

Figure 3.3 Foals with cryptosporidiosis. 1 00

Figure 4.1 Duration of i l l ness (upper g raph) and distribution of symptoms ( lower graph) in the outbreak of cryptosporidiosis, as elicited from the responses to questionnaire 01 141

Figure 4.2 Epidemic curve of the outbreak of cryptosporidiosis as elicited by the 1 42 responses to 01

xiii

AIDS

BLST

bp

Cl

CIN

COWP

GP60

HAART

HIV

HSP70

lgA

lgG

I gM

IVABS

LATU

MLG

MU

OPG

PAS

PCR

poly T

RFLP

RNR

SIA

SNP

UPGMA

uv XLD

ZN

18S rRNA

LIST OF ABBREVIATIONS

acqu i red immu nodeficiency syndrome

Bi locus seq uence type

base-pairs

confidence interval/confidence l im it

cefsu lodin , i rg asan and novobiocin agar

Cryptosporidium oocyst wall protein

Cryptosporidium surface G P45/15 g lycoprotei n (or 60-kDA g lycoprote in)

High ly Active Antiretroviral Therapy

human i m munodeficiency v i rus

70 kDa Heat Shock Protei n gene

Immunoglobu l i n A

Immunoglobu l i n G

Im munoglobu l in M

Institute of Veterinary, An i m al and Biomedical Sciences

Large An imal Teaching Un i t

mu lti locus genotype

Massey U n iversity

oocysts per g ram of faeces

periodic acid-Schiff stain

polymerase chain react ion

polythreon i ne repeat

restriction frag ment length polymorph ism

ribonuclease reductase

standardized i ndex of association

s ingle nucleotide polymorphism

Unweighted Pair Group Method with Arithmetic mean

u ltra violet

Xylose Lys ine-dehoxycolate

Ziehl Neelsen

smal l subu nit 18S ribosomal RNA

xiv

CHAPTER 1

INTESTINAL CRYPTOSPORIDIOSIS IN HUMANS AND DOMESTIC

MAMMALS: AN INTERPRETIVE OVERVIEW

1.1 CRYPTOSPORIDIOSIS IN THE HISTORICAL PERSPECTIVE

Cryptosporidium is a genus of protozoan parasites of amphibians, f ish, repti les, birds,

m ammals, and humans, f i rst described by Ernest Edward Tyzzer (1875-1 965) i n 1907 (Tyzzer

1907). Using l ight microscopy, this eminent infectious disease researcher at Harvard Un iversity,

Massachussetts, observed coccidian- l ike parasites in the gastric and intestinal cells of m ice,

which he then named Cryptosporidium muris and Cryptosporidium parvum, respectively (Tyzzer

1910, 19 1 2) . He proposed the name Cryptosporidium (from Greek Kpum6c;, kryptos : h idden)

because sporocysts were not seen in the oocysts (Tyzzer 1 9 1 0).

Despite the early recognition of these new parasites by Tyzzer, the f i rst descript ion of the

disease cryptosporid iosis in any host was recorded about 60 years later in a calf (Panciera

1 971 ) , and later in a 3 year-old g i r l from a farm who had symptoms of abdominal pain and

d iarrhoea (N ime et al. 1976). Unti l the 1980s Cryptosporidium parasites were not considered to

be important pathogens in immunocompetent humans (B i rd and Sm ith 1980) , as

cryptosporidiosis was mostly seen in immunodeficient patients. This was due to the fact that

the diag nosis depended upon h istopathological examination of i ntestinal t issues, a procedure

that was performed only in the most severe cases ( Meisel et al. 1976). With the emergence of

the H IV-AIDS epidemic in the 1 980s, h istopatholog ical tests were required more f requently to

diagnose the disease in patients affected with A IDS (Casemore et al. 1984ab, 1985ab).

Eventually, s imple methods for the m icroscopic detection of the oocysts in faeces, for example

those based on the cold Ziehl Neelsen stai n i ng of faecal smears, were adopted by human

diag nostic laborator ies (Garcia et al. 1 983 ; asemore e a . ""9-s-4ab . nrere"Stingly; these-

methods were f i rst devised for the diagnosis of cryptosporidiosis in calves ( Poh lenz et al . 1978 ;

Henriksen and Pohlenz 1 981 ). The i ntroduct ion of such simple, non- i nvasive and cost effective

diagnostic procedu res allowed cryptosporidiosis to also be included in the differential d iagnosis

of infectious d iarrhoea in immunocompetent humans and in calves. Consequently, the rate of

identification of Cryptosporidium i ncreased, to eventually become one of the m ost common

agents identified from cases of diarrhoea in humans and calves in more than 1 00 countries

( reviewed by Fayer 2008).

The interest in Cryptosporidium research increased significantly after the realisation that these

parasites can cause explosive water-borne diarrhoea! d isease epidem ics. The Mi lwaukee,

Wisconsin, epidemic in 1993, in which more than 400,000 people are believed to have acqu i red

the agent from a s ingle dri nking water plant source ( MacKenzie et al. 1 994) , was a particular

event that focused the attention on Cryptosporidium. This, and other outbreaks, prompted the

Director of the US Centers for Disease Control and Prevention to include Cryptosporidium in the

l ist of agents defined by him as "major etiologic agents identif ied since 1973" (Satcher 1995) .

1 .2 1MPACT OF CRYPTOSPORIDIOSIS ON HUMAN AND ANIMAL HEALTH

The most common clin ical presentation of cryptosporidiosis in imm unocompetent patients in

developed cou ntries is of self- l imit ing d iarrhoea. Some exceptional cases of gastrointesti nal

i l lness lasting up to four months have been described (Hunter and N ichols 2002; Warren and

Guerrant 2008). Subcl in ical carriage of Cryptosporidium has been reported in humans (Siwi la

et al. 2007), and i m m u nocompetent ind ividuals can excrete oocysts i n the faeces for up to

seven weeks after infection (Steher-G reen et al. 1 987) . However, in H IV-AIDS or otherwise

immunocompromised patients cryptosporidiosis may be a chronic debi l itat ing , or even a

fulm inant d isease, and these patients might also suffer from extrai ntesti nal infections (Chappel l

and Okhuysen 2002 ; Hunter and Nichols 2002; Tzipori and Ward 2002; Huang et al. 2004).

Specific anti- Cryptosporidium treatments are not avai lable. Thus, cryptosporidosis was, u nt i l

recently, a major debi l itating disease and cause of mortality in H IV-positive patients.

Encourag ing ly, the recent i ntroduction of High ly Active Anti retroviral Therapy (HAART) protocols

decreased the incidence, severity and mortality of cryptosporidiosis as well as other

opportun istic infections in these patients ( Maggi et al. 2000; M iao et al . 2000; Pozio and

Morales 2005). However, the burden of cryptosporidiosis remains serious in immunosuppressed

patients in countries that cannot afford HAART (Tzipori and Widmer 2008) .

In many developing reg i ons of the world cryptosporid iosis is associated with persistent

childhood d iarrhoea, m alnourishment, stunted g rowth and high mortality (these aspects of

human cryptosporidiosis have been reviewed by Di l l ing ham et al. 2002, and Snel l ing et al.

2007). The reasons for the difference in the spectrum of severity of cryptosporidiosis between

developed and deve loping regions are not understood because mu ltifactorial epidemiological

comparisons between reg ions are lacking. The h igher prevalence of H IV-positive people in

some deve loping reg ions could i ncrease the overall r isk of chronic cryptosporidiosis as

compared with reg ions with lower H IV prevalence. For exam ple, 67/76 (88%) of

Cryptosporidium-posit ive children adm itted with chronic d iarrhoea to Mu lago hospital, Kampala,

Uganda, were H IV-positive (Tumwine et al. 2005). The cycle of m alnourishment leading to

immunosuppression cou ld also explain the difference in the spectrum of severity of

cryptosporidiosis between developing and developed regions. However , the effect of

malnou rishment is d ifficu lt to assess as recent observations in an imal models indicate that

malnou rishment might be an effect, rather than a cause of chronic cryptosporidiosis (Coutinho

et al. 2008). In one epidemiological study in Bangladesh , there was a significant d ifference in

the lgA and lgM levels in patients with persistent diarrhea caused by C. parvum, as compared

with those with acute diarrhea, which had g reater lgA and lgM levels. The authors hypothesised

2

that a d imin ished m ucosal lgA response may contribute to the persistence of the infections

( Khan et al. 2004).

A h igh incidence of Cryptosporidium superinfections, as expected in heavi ly contaminated

environments, can also provide a plausible explanation for the h igh rate of chronic i nfections

observed in some reg ions where hygiene is poor. I ndeed, as will be seen later in this overview,

Cryptosporidium may diversify by recombinatio n in the gastrointestinal tract of the host. I n

theory, superinfections with genetical ly heterogeneous parasites m ay faci l itate t h e emergence

of recombi nant parasites, which may differ antigen ically from their parental cel ls, a feature that

could enable the evasion of the imm une response of the host and enhance chronic infect ions.

The study presented in Section 2.2 explores the rate of recombinat ion in natural C. parvum and

C. hominis populations.

Stud ies addressing the deg ree of protection conferred by previous exposure against re-i nfection

with Cryptosporidium parasites in humans provided conflicting results. In one study with human

volunteers, Okhuysen and col leagues (1999) found a partial resistance to re-infection in

previously exposed persons. Conversely, in a different study, volunteers with pre-existing anti­

Cryptosporidium antibodies shed less oocysts in the faeces than the seronegative, but on the

other hand manifested a more severe d isease (Chappel l et al. 1999). lt is plausible that the

immune response developed in the course of cryptosporidiosis is stronger against homologous

than hetero logous species , or genotypes. This idea has been recently corroborated by Sheoran

and colleagues (2008). These researchers showed that the immunity developed against C.

hominis provided on ly partia l protection agai nst subsequent C. parvum i nfection in the

gnotobiotic pig model. The effect of previous exposu re on the susceptibi l ity to C. parvum

infection wil l be further explored in the study reported in Chapter 4.

A wide range of Cryptosporidium taxa have been identified i n many host spec1es y

O'Donoghue 1 995 and Fayer 2008). However , so far a causative role in disease has been

establ ished on ly for C. parvum in calves and to a lesser extent lambs and kids, and perhaps for

Cryptosporidium andersoni i n juven ile and adu lt cattle.

Cryptosporidium parvum is an important causative agent of neonatal calf diarrhoea worldwide.

Causality in calves has been confi rmed by numerous observational studies, experimental

infections, and therapeutic trials ( Pohlenz et al. 1978; Tzipori et al. 1980b, 1983 ; Fayer and E l l is

1993 ; Naciri et a l . 1993, 1999 ; O'Handley et a l . 1 999; Castro-Hermida et al. 2002b; Gr inberg et

al. 2002; Sevinc et al. 2003). Sig n ificantly, resu lts of recent molecular epidemiolog ical studies

ind icated that Cryptosporidium bovis, which was described in 2005 (Fayer et al. 2005) , is also

widespread in young cattle (Fayer et al. 2007; Feng et al. 2007). The finding that C. bovis is

h ig hly prevalent i n cattle is of considerable public health relevance, as th is parasite, which is

phenotypical ly-s im i lar and thus, easi ly confused with C. parvum, has never been identif ierd from

3

cl in ically overt i nfections in cattle o r humans, and is therefore not considered either pathogenic

o r zoonotic.

Cryptosporidium andersoni i nfects the gastric glands of juven i le and adult cattle caus ing m i ld

lesions of an u ncertain pathogenic significance (see Section 1.6). lt produces oocysts different

from C. parvum in shape and s ize (Lindsay et al. 2000) , and has been reported o nce in

humans, in an H IV-positive person (Guyot et al. 2001 ).

1.3 THE TAXONOMIC CLASSIFICATION OF GENUS CRYPTOSPORID/UM

There is lack of consensus on what constitutes a Cryptosporidium taxon. Thus, in th is overview

the specific terms of 'species' or 'genotype' will be used for those taxa which in the author' s

opinion are widely accepted as such by the scientific com munity. Other Cryptosporidium genetic

variants wi l l be referred to by us ing the term 'taxon'. The term 'isolate' wil l be used to describe

any Cryptosporidium-positive faecal specimen f rom an individual human or an imal host, o r

genomic DNA extracted from such specimens.

Cryptosporidium a re intracel lu lar protozoan parasites that can be classified biological ly as either

intest inal (that is , parasites with a biological cycle that is completed in the cel ls of the i ntestinal

tract of the host) , or gastric (where the cycle is completed in the gastric cel ls) taxa. Some

authors have indicated that there is h igh phylogenetic relatedness between taxa with in each

g roup, regard less of the host-range (Xiao and Ryan 2008) . This h igh l ights the shortcom ings of

the nomenclature based on the host species i nst igated by Tyzzer when proposing the name C.

muris to the parasites he fou nd in m ice (Tyzzer 1910, 191 2) and widely used in the 1970s

(Barker and Carbonell 1974; Tzipori 1988). The sequencing of the genomes of C. parvum and

C. hominis, the major intest inal Cryptosporidium of man and animals, has been completed

(Abrahamsen et a l. 2004; Xu et al. 2004) , and that of the gastric C. muris is sti l l in progress.

major advance, as it will enable the genomic comparison

i ntestinal species (Tzipori and Widmer 2008).

Many opinion leaders endorse the taxonomy indicating Cryptosporidium as a genus with in the

Phylum Apicomplexa, Class Coccidea, Order Eucoccidiorida, Family Cryptosporidi idae (Tzipori

and Ward 2002 ; Fayer 2008). However, based on the nucleotide sequence of the small subunit

ribosomal RNA gene, Cryptosporidium and the G regarina appear to form a s ing le clade, which

is separated from the coccidia and the other members of the Phylum Apicomplexa (Carreno et

al. 1999). In agreement with the apparent phylogenetic separation of Cryptosporidium from the

coccidia, there are conspicuous phenotypic differences between these taxa, which had been

reviewed by Barta (2007). T hese are the epicel lu lar localization of Cryptosporidium with in the

cells, the apparent lack of a plastid, the endogenous sporulation , resistance to folate pathway

antimicrobials, and lack of susceptibi l ity to anticoccidial chemotherapeutic agents.

4

As said , in the 1970s the taxonomic nomenclature with in the genus Cryptosporidium was largely

based on the host-species from which the oocysts were recovered. However, as the oocysts of

many intestinal Cryptosporidium are m orphological ly sim i lar, and i n m any instances no host­

specificity was observed, a l l i ntestinal cryptosporidium were subsequently considered to be long

to a s ing le species named C. parvum (Tzipori et a l . 1 980a). This view started to change in the

1 990s, when the fi rst papers report ing on the existence of genetic heterogeneity in C. parvum

were publ ished. Currently, the taxonomic classification within the genus Cryptosporidium is

main ly based on the genotype, rather than on phenotype.

The f i rst report of genetic heterogeneity in C. parvum isolates found in the literature was

published in 1 99 1 . The paper described a study performed using southern blot hybridization of

chromosomal restriction endonuclease digests (Ortega et al. 1991 ). The f i rst applicat ion of the

PCR techn ique to amplify Cryptosporidium gene sequences was publ ished in the same year

( Laxer et al. 1 991 ). This work demonstrated the feas ib i l ity of using the PCR technique for

Cryptosporidium research , which eventually resu lted i n the discovery of numerous g enetic

variants. The smal l subu n it ribosomal RNA (188 rR NA) gene sequence is currently the most

popular locus used for the taxonomic typ ing of Cryptosporidium isolates. I n 1996, Zamani et al.

(1996) ind icated that the 1 88 rRNA gene of C. parvum is present in one heterogeneous and

fou r identical copies in the genome. Heterogeneous copies of the 188 rRNA gene seem to

occur also in other Cryptosporidium taxa, such as Cryptosporidium felis (Xiao et al. 1 999). This

locus is particularly usefu l for taxonomic typ ing , as it comprises a species-specific reg ion. In

add ition , it is present in four identical copies in the genome of C. parvum and C. hominis, which

should - at least in theory - enhance PCR sensitiv ity (Pevsner 2003; R i ley 2004). Therefore, the

188 r R NA gene is usual ly among the f irst to be characterised when new Cryptosporidium

parasites are found, and Caccio and col leagues (2005) recommended the inclusion of th is gene

in an molecular identif ication scheme for Crvptosporidium. Other coding or non-coding genes,

such as the �-tubul in gene, the Cryptosporidium o ocyst wal l protein (COWP) gene, a

polythreonin repeat (poly T) , a thrombospondin-re lated protein , the ribonuclease reductase

( R N R ) gene and the so-called LI B13 marker, have also been used for taxonomic identif ication

purposes (Carraway et a l. 1 996, 1997; Peng et al. 1 997; Spano et al. 1998 ; Caccio et a l . 1999,

2000 , 2005 ; Tanriverdi et al . 2003) . I ntuit ively, sequences of taxonomically informative loci

should be conserved with in a taxon, but differ between taxa. The d ifferences between the taxa

can be revealed by means of restriction fragment length polymorphism analysis (RFLP) or

sequence analysis of the PCR ampl icons.

So far, evidence for recombi natio n between Cryptosporidium taxa in nature is lacking.

Therefore, the same combination of al leles at al l loci with taxonomic information content should

be present i n iso lates belonging to the same taxon. H owever, for m any Cryptosporidium taxa

the nucleotide sequences of many loci are sti l l u n known, and so identification schemes

performed using several taxonomic loci can lead to am biguous conclusions. For example , Elwin

5

and Chalmers recently reported on the f inding of Cryptosporidium isolates in sheep showing the

R FLP pattern of the 188 rRNA gene of C. bovis, and the COW P pattern of the Cryptosporidium

'cervi ne genotype' ( E iwin and Chalmers 2008}. The authors concluded that the isolates were

mixtures of both taxa , and that preferential PCR amplif ication of the C. bovis 188 r RNA gene

had occurred. In this author's view, such a conclusion is not supported as the COW P gene

sequence of C. bovis is not known.

D ifferent l ists of Cryptosporidium taxa have been published in recent years (X iao and R yan

2004 ; Fayer 2004; Fayer 2008; X iao and Fayer 2008}. However , it is apparent that the

taxonomy within genus Cryptosporidium is far from being resolved, and there is st i l l a long l ist of

genetic variants of uncertain taxonomic standing and bio logical s ign ificance. At the 61h Meet ing

on M olecu lar Epidemiology and Evolutionary Genetics of Infectious Diseases held in 2002, it

was concluded that the followi ng aspects should be described when suggest ing a new

Cryptosporidium species : ( 1 } , morphometric data on oocysts; (2} , a genetic characterisation of

the proposed species; (3}, demonstrate natural, and when feasible experimenta l , host

specificity; and (4) , com ply with I nternational Code for Zoological Nomenclature (Fayer 2008} .

Notwithstanding th is proposal there is sti l l uncertainty and confusion as to what constitutes a

valid Cryptosporidium taxon. Conf l ict ing l ists of "accepted", " recogn ised", or "val id

Cryptosporidum species" have been recently published by different authors (Snel l ing et a l .

2007 ; X iao and Feng 2008; Fayer 2008 ; Xiao and Fayer 2008} , and there are discrepancies

between the l ists suggested by the same authors in different papers. For example, in one

recent paper, the number of species was set at 16 "recorded species" (Snel l ing et al . 2007), and

in the same year other authors proposed 16 "val id" species and "over 30 genotypes" (Santi n

and Fayer 2007) . A year later, X iao and Feng (2008) i ndicated there are 1 6 "accepted"

Cryptosporidium species and "about 50 genotypes", but in another paper Xiao i nd icated the

number of "val id' species to be 1 8 , and that there are "over 40 genotypes" (Xiao and Fayer

2008}. Whi le in 2008, Fayer suggested the existence of 16 species and more

genotypes (Fayer 2008} , in another paper this author indicated 16 species and o n ly 11

genotypes (Xiao and Fayer 2008). Furthermore, the number of genetic variants and taxa

recogn ised cont inues to increase. For i nstance, most recently, Ryan et al. (2008} proposed the

new species name of C. fayeri (in honour of Fayer) for Cryptosporidium genetic variants isolated

from the red kangaroo (Macropus rufus), and in the same year, Fayer et al. proposed the

species name of C. ryanae ( in honour of Ryan) , as a new species name for the so-cal led

Cryptosporidium deer- l ike genotype (Fayer et al. 2008) .

Accord ing to Fayer , a genotype " is not a taxon", but rather a part ia l and temporary descriptor

(Fayer 2008}. In fact, the term 'genotype' has been used to acknowledge the uniqueness but

also the incomplete knowledge on the new variant. Occasional ly, new genotype names have

been designated based on l imited evidence. Such is the case of the Cryptosporidium 'horse

genotype'. This descriptor was proposed by Xiao and Feng (2008) and Fayer and Xiao (2008)

6

based on a previous report of a new, partial ( -4 70 base-pairs long) 1 8S rRNA gene sequence

identified in o ne i solate from a Przewalski 's wi ld horse ( Equus przewa/sk1) i n a zoo ( Ryan et al.

2003) . The gene sequence of the 'horse genotype' has never been identified again in any other

host, but nonetheless, it has been repeated ly clai med that horses are "wel l known to be i nfected

with the horse genotype" (X iao and Feng 2008; Fayer and Xiao 2008). The idea of the

occurrence of a 'horse genotype' is further d iscussed in Section 1 .5.4 and in the studies

reported in Sections 3.2 and 3 .3.

There are a number of objective reasons for the lack of a clear consensus of what constitutes a

Cryptosporidium taxon. First, i n many cases there is sti l l uncertaintly on how to d iscern between

the intra-taxo n and inter-taxon genetic variatio n (Fayer 2008) . Second, the requisite of a

biological species, which implies a natural population reproductively isolated from other

populat ions, is difficult to assess in Cryptosporidium. I ndeed, although in vitro Cryptosporidium

g rowth has been achieved (Current and H aynes 1984 ; H ijjawi et al . 2002, 2004) , systems

capable of yield ing high numbers of Cryptosporidium oocysts in vitro have not been been

developed. In addition, experimental animal m odels which support the g rowth of d ifferent

species are often lacking . Consequently, the only experimental unit for cross-breedi ng

experiments are the Cryptosporidum ' iso lates', which are composed of a f in ite number of

oocysts orig i nat ing from one, natural ly i nfected an imal. However, as wi l l be seen in the section

describ ing the Cryptosporidium l i fe cycle, the oocysts derived from an individual an imal cannot

be considered a single clone or strain , because they may represent m ixtures deriving from

assemblages of taxa infect ing the same host, recombinants, or perhaps even reproductively

separated l i neages of the same taxon (Xiao et al . 2004) . Unfortunately, the genetic identificat ion

using PCR-based gene ampl ification can not resolve such a complex genet ic m ake-up of the

isolates, due to the problem of PCR ampl ificatio n bias (Suzuki and Giovannoni 1996, Rochel le

et al . 2000). This is somewhat different to the s ituation with bacteria, in which a single cell can

be isolated and g rown and its progenies considered as a si ngle clonal l ineage. Thus, the i ntra­

iso late genetic d iversity of Cryptosporidium is st i l l unknown , and th is l im its the possibi l ities to

meaningful ly interpret the results of genetic cross-breed ing studies using the isolates as the

operational taxonomic un it . Lastly, as in vivo passages are needed for the propagation of

Cryptosporidium, there is a g reat potential for cross contamination of the iso lates, which is

d iff icult to pred ict and monitor. Th is problem was wel l i l lustrated by the displacement of the

or ig i nal parasites of the widely used ' IOWA' C. parvum isolate with exogenous variants found i n

calves (Cam a e t al. 2006) .

Of the several Cryptosporidium taxa identified i n humans, C. parvum and C. hominis accou nt for

the majority of cases of cryptosporidiosis worldwide (reviewed by X iao 2007, and N ichols 2008) .

Notwithstanding the problems concerning the taxonomy of genus Cryptosporidium, the

differentiat ion of C. parvum and C. hominis i nto two different species seems to have reached a

consensus. The species name C. hominis was initially proposed by Morgan-R yan et al. (2002),

7

to differentiate the C. parvum ' human genotype' (synonym: 'type 1 ' o r 'genotype H') , which

cycles i n humans, from the C. parvum 'cattle genotype' (synonym : 'genotype C' , ' type 2' , or

'cattle'/'bovine' genotype), i nfecting both humans and animals. The new proposed nomenclature

preserved the species name of C. parvum to the 'cattle' genotype, whi le the new name of C.

hominis was g ive n to the 'human genotype'. The separation of these genotypes i nto two species

is consistent with the conspicuous genetic and epidemiological d ifferences exist ing between

these taxa, and the apparent absence of recombinant genotypes, which is consistent with two

reproductively isolated popu lat ions ( Spano et al. 1 998; Mclaughl in et a l . 2000) . The species

names of C. parvum and C. hominis wi l l therefore be used throughout th is thesis, except in the

study presented in Section 3.2, which reports the resu lts of a study performed in 2002, when the

old name of C. parvum 'catt le' genotype was sti l l widely use.

1 .4 THE LIFE CYCLE OF THE INTESTINAL CRYPTOSPOR/D/UM PARASITES

The present dissertation of the life cycle focuses on issues of genetic exchange and population

genetic structure of intestinal Cryptosporidium, as these are further explored i n the

epidem iolog ical study presented in Section 2 .2 .

The biological cycle of intest inal Cryptosporidium parasites was described before the d ifferent

species could be d ifferentiated based on the genotype. Except for some comparative notes

(Tzipori 1988; Fayer 2008) , authoritative descriptions of the cycle i n the different host species,

or of the different taxa, were not found i n the scientific l iteratu re. I n early descript ions of the life

cycle, the name C. parvum was used to describe the cycle of both C. parvum and C. hominis

(Tzipori and Gr iffiths 1 998; Tzipori and Ward 2002). I n a recent d issertation of the l ife cycle,

Fayer (2008) used the general descriptor of Cryptosporidium spp.

The entire biological cycle of i ntesti nal Cryptosporidium parasites is completed i n the intestinal

tract of the host, and cu lm inates with the excretion of sporu lated and fu l ly i nfectious oocysts in

the faeces (Figure 1 . 1 ). The oocysts are coated with a hard protective wal l and are h igh ly

resistant in the envi ronment ( Ranucci et al. 1 993; Spano et al. 1 997 ; reviewed by Fayer 2008).

The complex physical and chemical factors governing the decay in the i nfectivity of the excreted

oocysts have been widely studied in the laboratory, and their description is beyond the scope of

the present dissertat ion ( reviewed by Fayer 2008 and Peng et al. 2008). However, it is assu med

that some Cryptosporidium oocysts can remai n i nfectious for long periods of t ime, even in what

has been defined by Fayer as "harsh envi ronments" (Fayer 2008).

I nfections with Cryptosporidium are typically acqu i red through the i n gestion of oocysts in water,

food, or other i ngesta. Accord ing to some authors, the cycle m ay also commence by an

i nhalation of the oocysts (Tzipori 1 988; Fayer 2008). Each oocyst conta ins four haploid

sporozoites, which are the basic replicat ing Cryptosporidium cel ls. E ight chromosomes have

been described in C. parvum and C. hominis ( Blunt et al. 1997; Xu et a l . 2004). In the small

8

i ntestine, the sporozoites excyst a long a suture at one of the poles of the oocyst, and the free

sporozoites in fect the enterocytes. l t is thought that the i nvasion of the enterocytes is in itiated by

an adhesion of sporozoite surf ace lect ins to the intestinal m ucus fol lowed by an interact ion of

l igand surface proteins with receptors present on the surface of the enterocytes, and f inal ly, an

internalisat ion of the sporozoite ( Riggs et al . 1 997; Cevallos et al . 2000 ; Smith et al. 2005a;

Feng et al . 2006). As in other Apicomplexa, i t is bel ieved that the select ion of the host cel l , and

contact and penetration of the sporozoite to the cel l , i s mediated by structures and

m acromolecules p resent in an apical complex (reviewed by Borowski et al. 2008) .

Feng and eo-workers (2006) reported that b i le salts, which are normal ly present in the gut,

s ign ificantly enhanced the i n i t ia l i nvasion of cel ls by both C. parvum and C. hominis in vitro.

Once in the cel l , the sporozoite develops with in a parasitophorous vacuole located with in the

m icrovi l l i of the brush border. The cell membranes of the sporozoite and the enterocyte fuse at

the basal zone of the parasitophorous vacuole, forming the feeder organel le (Theodos et al .

1 998) , and the sporozoite g radual ly transforms into a trophozoite. Then, in a process referred to

as merogony (or schizogony) the trophozoite's nucleus undergoes mu ltiple asexual divis ions, to

form a mult inucleate meront (o r schizont) , which subsequently segregates into 6-8 uninucleate

merozoites. The mature merozoites bu rst f rom the meront and abandon the cel l , to infect new

epithelial cel ls , i n which they u ndergo a second round of merogony. Two types of meronts have

been described for C. parvum: Type 1 meront, which is composed of of 6-8 merozoites, and

Type 1 1 meront, with only fou r merozo ites. l t i s bel ieved that Type 11 meronts either differentiate

i nto an intracel lu lar microgamont, which releases several extracel lu lar haploid microgametes, or

into a macrogamont, which remains in the cel l. By a poorly understood process, the

m icrogamete penetrates f i rst the host's cel l and then the m acrogamont membranes, to ferti l ise

the macrogamont, with the formation of a d iploid zygote. Subsequently, the zygote undergoes a

reductional divis ion into two haploid progenies, which then divide again , conservatively, to

generate four haploid sporozoites, whi le at the same time the oocyst cell wall i s being formed. In

this process, cons idered by m any as a meiosis, chromosomal remodel ing by recombination m ay

occur if the gamonts are genetically-different, generat ing two pairs of progenies which differ

genetically from each parental gamete.

Whi le diversificat ion by recom bination has been documented in C. parvum in exper imental

infections (Tanriverdi et al. 2007) , the contribution of this process to the genetic diversification of

C. parvum and C. hominis in n ature is not well understood. According to the above cycle, each

oocyst may contain two pairs of genetical ly-heterogeneous sporozoites, which differ from the

parental cells due to recombination. Moreover, the Cryptosporidium iso lates, which are

composed of a f in ite number of oocysts recovered from a single host, m ay themselves be

composed of an assemblage of genetical ly-heterogeneous oocysts. With in-oocyst and between­

oocyst heterogeneity are therefore the two possible sources of genetic variation with in

Cryptosporidium isolates. I nterestingly, these sources of variation are not usual ly taken i nto

9

accou nt, o r discussed in molecular epidemiological studies of cryptosporidiosis, in which

t raditional ly the operational taxonomic units of the analysis have been the 'isolates'. As said in

Section 1.3, the intra-isolate genetic variation is difficul t to assess using the PCR due to the

potential for preferential amplification of the templates ( Suzuki and Giovan noni 1996; Rochel le

et al. 2000). The genotyping of sing le oocysts, as recently reported (Hashimoto et al. 2006) ,

might in part address this problem, but cannot resolve the intra-oocyst genetic variation, which

requires the use of PCR-free sequencing technologies that are just now becoming accessible.

At the end of the cycle, a fu l ly infectious oocyst contai ning four naked sporozoites is released

from the cel l (Smith et al. 2005a). Some observations have suggested that Cryptosporidium

oocysts exist in two forms, a thin walled and a thick wal led variant (Current and Reese 1986;

Tzipori 1 988) , leadin g to the belief that the thin walled oocysts may excyst in the intestina l tract

of the same host causing autoinfections ( Ridley and O lsen 1991; Templeton et al. 2004). As

discussed in Section 1 .4, these autointections may provide a plausible explanation for the

occurrence of chronic i nfections in i m m unosuppressed i ndividuals (Current 1988 ; Di l lingham et

al. 2002).

10

small s<>orozoileslcysll

f

.. � Merozoite �

causing: Villous blunting,

mild Inflammation, prolonged diarrhea

and malnutrition

llmeron� nuclear divisions

4 merozmtes)

J M�metocyte (� � ·c::J v +

Macrogametocyte ® \

Figure 1 . 1 : The intestinal and extraintestinal life cycle of Cryptosporidium parvum. (source: Dillingham et

al. 2002). This figure will be incorporated in the thesis published online upon receipt of the copyright

permission from the publisher.

1 1

1.5 PATHOGENESIS OF INTESTINAL CRYPTOSPORIDIOSIS

I ntesti nal Cryptosporidium are obl igate i ntrace l lu lar parasites of the enterocytes. The i nvasion of

the cell is bel ieved to be i n it iated by an i nteraction between l igand macromolecu les on the

sporozoite's su rface and its apical complex and the surface of the enterocyte, fol lowed by the

i n ternalisat ion of the sporozoite . A number of putative l igands have been described in C.

parvum sporozoites ( reviewed by Tom ley and Soldati 2001, and Borowski et al. 2008) .

Antibodies d i rected agai nst some epitopes of these l igands were able to neutral ise C. parvum

i nfectivity both i n m ice and in vitro (R iggs et a l . 1 997, Tzipori and Ward 2002). One of the most

w idely studied C. parvum and C. hominis putative l igand prote in is the 60 kDA surface

g l ycoprotei n (Cevallos et al. 2000; Strong et al. 2000) , also known as the 'G P60' prote in . As

expected for a molecule under immunolog ical pressure , the gene encoding the GP60 is h igh ly

polymorph ic and comprises numerous non-synonymous polymorphisms (Strong et al . 2000) .

S uch a polymorphism reduces the appeal of the GP60 as a potential immun is ing agent, but on

the other hand enhances i ts usefu lness as an epidemiological marker. Chapters 2 and 3 of th is

thesis report epidemiological studies using subtyping of the G P60 gene.

Within the enterocyte, the sporozoite carves a niche in a un ique i ntracel lular but

extracytoplasmic location. Un l ike other apicomplexan organ isms, such as Toxoplasma,

Plasmodium, Eimeria and Cyclospora, which develop in the cytoplasm surrounded by a

parasitophorous vacuole, the Cryptosporidium sporozoite remains anchored to the host's cell

m embrane with in the remnant of the m icrovi l lus (Griffiths et al. 1 998) . it has been hypothesised

that in th is location it can develop whi lst protected from the host's cell cytoplasm , the immune

s ystem, and m any chemotherapeutics (Griff iths et al. 1998 ; Theodos et al. 1 998).

T he patholog ical changes induced by C. parvum and C. hominis vary in severity, but are fair ly

s im i lar in the various hosts. H istolog ically, using G iemsa stain the i nfect ing organisms can be

v isualised as small basoph i l ic bodies embedded in the m icrovi l l i . Other changes i nc lude the

p resence of blunt, fused i ntestinal vi l l i combi ned with m ucosal hyperplasia accompanied by a

variable i nflam matory response consist ing of lymphoid cel ls , macrophages, and neutroph i ls

inf i ltrat ing the lamina propria. The lesions are often found throughout the smal l i ntesti ne, but

can also appear in the large intestine. They tend to be more severe in the d istal jejunum and

i leum (Laurent et al. 1999; Tzipori and Ward 2002; Stewart and Penzhorn 2004). Hyperaemia

o f the affected segments and stunting and fus ion and/or cross bridg ing of the adjacent v i l l i are

often observed in calves, and have also been described in the horse (K im 1990).

I n H IV-positive patients, the morphologic changes seem to be correlated with the number of

o rganisms present in the t issues (Genta et al . 1 993). In immunocompetent hosts, i ntestinal

cryptosporidiosis is typical ly a self- l im it ing disease last ing a few days. The short duration is

g enerally presumed to be due to the mount ing of an im mune response, but perhaps also to the

biolog ical 'program ming' of a l im ited number of merogony cycles, as on ly two types of meronts

1 2

are known to exist (Tzipori 1 988) . However, as previously stated, it has been hypothesised

that the th in-walled oocysts excyst in the intest inal tract of the same host causing autoinfections.

Autoinfect ions have been impl icated as a leading cause of chronic debi litating cryptosporidiosis

i n malnourished chi ldren (D i l l i ngham et al. 2002) . While i n an imals, chronic C. parvum

i n fections have been experimentally produced in mice (Ungar et a l . 1 990 ; Perrym an and

Bjorneby 1 991 ) , in nature they have only been described in immunodefic ient Arabian horses

(Snyder et al. 1 978; Gibson et al. 1 983) .

The variabil ity in infectivity between different Cryptosporidium iso lates has been assessed in

h uman volu nteers (Okhuysen et al. 1 999) . However, the interpretation of the results of such

studies is difficult , because Cryptosporidium oocysts do not survive freez ing, and need to be

passed in vivo in order for thei r viabi l ity to be maintained. Such passages may modify the

genetic m akeup of the isolate due to cross-contam i nation with wild oocysts or even i nternal

recombinat ion , which are d ifficult to assess or mon itor.

1.6 INFECTIONS WITH CRYPTOSPORIDIUM IN DOMESTIC MAMMALS

1.6.1 Infections with Cryptosporidium in cattle ( 8os taurus)

The most prevalent Cryptosporidium taxa found in cattle are C. parvum, C. andersoni and C.

bovis. Other taxa, such as C. hominis (Smith et al. 2005b), the Cryptosporidium 'cervine'

genotype and the 'deer-l ike genotype' (Fayer et al. 2005, 2006; Trotz-Wi l l iams et a l . 2006 ; Feng

et al. 2007; Feltus et al. 2008) , C. suis and 'su is-l ike genotype' (Geurden et al. 2006), and C.

felis (Bornay L l inares et al. 1 999) , have on ly been reported sporadical ly in cattle, but their

i mpact on bovi ne health is unknown.

Cryptosporidium andersoni is a gastric species that infects the gastric epithel ial cells of juveni le

and mature cattle and causes m i ld di lat ion of the pyloric g lands, hypertrophy of the gastric

m ucosa, and th i n ni ng of the epithelial l i n ing , with little or no inf lammation (Oison et al. 1 997;

Kvac et al. 2008) . Animals infected with C. andersoni excrete oval oocysts 4x7 11m in d iameter

( Lindsay et al. 2000; Kvac et al. 2008) . Some authors postulated that infections with C.

andersoni m ay impair prote in d igestion and decrease productivity ( Esteban and Anderson

1 995) . C. andersoni is not considered zoonotic, although it has been isolated from one H IV­

positive patient (Guyot et al. 2001 ) . No reports of C. andersoni i nfections in cattle in New

Zealand were retrieved in the scientific literature consulted.

Conversely, C. parvum and C. bovis are considered intestinal parasites producing

phenotypical ly s im ilar, round oocysts 4-6 11m i n d iameter (Fayer et al. 2005; Fayer 2008) . Whi le

C. parvum is a well-known pathogen of cattle, no reports of cl inical ly overt i nfections with C.

bovis were found in the scientific literature consu lted. However, it should be remembered that C.

bovis was on ly described for the f i rst t ime in 2005 and consequently, much of the information

about infections with this parasite is sti l l unknown.

1 3

The f i rst descript ion of an i nfection with Cryptosporidium in cattle dates from 1971 (Panciera et

a l . 1971 ) . A number of years later, the lesions in calves were described in more detai l (Morin et

al. 1976 ; Pohlenz et al. 1978) . At the begin ning of the 1980s, the aetiologic role of

Cryptosporidium as a frank pathogen of cattle was debated due to the frequent eo- infect ions

with other enteropathogen s and the m i ld h istopathological lesions found i n the course of

infections (deGraaf et al . 1999). I ndeed, Angus, f rom the Morendun Research I nstitute,

Scotland, argued that " lt seems probable that cryptosporid ial infections represent a ser ious

compl icat ion of virus-induced e nterit is, particu larly in you ng calves." (Angus 1983).

The first outbreak of calf d iarrhoea in which Cryptosporidium was identified as the sole agent

was published in 1980 (Tzipori et al. 1 980b). Although 85% of the calves were affected, no

mortality was recorded. Koch 's postulates were fulfi l led by Tzipori and eo-workers in 1983

(Tzipori et al. 1 983) . Final ly , in 1985, Upton and Current suggested the species name of C.

parvum as a descriptor for the parasites producing 'small oocysts ' , distinguish ing it from the

' large oocyst' type invading the gastric m ucosa of cattle and at that time known as C. muris, but

later re-classified as C. andersoni (Upton and Current 1985 ; L indsay et a l . 2000).

I n the years that fol lowed, the aetiologic role of Cryptosporidium was repeated ly corroborated by

the results of experimental infections, observational studies, and therapeutic tr ials (Moore and

Zeman 1991 ; Brenner et al . 1993 ; Fayer and El l is 1 993; Naciri et al . 1 993, 1999; Moore et al .

2003; Grinberg et al . 2002 ; Joachim et al. 2003 ; Sevinic et al. 2003). Currently, C. parvum i s

considered among the com monest aetio log ical agents of neonatal calf diarrhoea worldwide

(Oison et al. 2004; Fayer 2008). Conversely, the prevalence of C. parvum i n post-weaned,

juven i le , and adult cattle is low (Atwill et al . 1 999; Atwi ll and DasPerei ra 2003 ; Fayer et al. 2006 ;

Santin and Trout 2008; Santin et a l . 2004, 2008) . Moreover , because the oocysts of C. parvum

and C. bovis are morpholog ical ly s im i lar, it is l ikely that many parasites observed in the past i n

t h e faeces o f juven i le and adu lt cattle were C. bovis, rather than C. parvum. In recent years ,

studies us ing some form of genotyping indicated that C. parvum is the com monest

Cryptosporidium species of unweaned calves, while C. bovis is found at a greater prevalence i n

post weaned calves (Santin e t a l . 2004, 2008 ; Fayer et al. 2006; Geurden et a l . 2006).

I nfections with C. parvum in calves are typically acqui red perinatally. The prepatent period

ranges between 3 and 11 days (Anderson 1 981, 1982 ; Fayer et al. 1998 ; Uga et al. 2000 ;

G rinberg et a l . 2002). In u ncompl icated cases of cryptosporidiosis, the cl in ical presentat ion is

fair ly predictable and characterised by a profuse self- l im it ing diarrhoea! disease lasti ng a

number of days (O' Handley et al. 1999; Uga et al . 2000 ; Gr inberg et al. 2002) . I n additio n to C.

parvum, other v i ral, bacterial and parasitic pathogens, such as rotavi rus, enterotox igenic

Escherichia coli, and coronavirus, are prevalent in calves during the f i rst weeks of life . Some

authors believe that eo-infect ions with these agents increase the severity of cryptosporidiosis

14

(deGraaf et al . 1999; Olson et al. 2004) , but no epidemiological stud ies support ing th is v iew

were found i n the scientif ic l iterature consulted.

The faecal oocyst excret ion cu rve in calves is predictable and bell-shaped. I n general, i nfect ions

are perinatal and the oocysts reach detectable numbers i n the faeces 3-5 days post infection ,

concomitantly with the onset of d iarrhoea. Their numbers peak a few days later, then rapidly

decrease to undetectable levels (Anderson and Bulg in 1981; Fayer et al. 1998; Atwi l l et a l .

1999 ; Uga et al . 2000; Cast ro-Hermida et a l . 2002b; Gri nberg et al. 2002; Figure 1.2). Oocysts

numbers as h igh as 10 7 oocysts per g/m l (OPG) of faeces have been reported at the peak of

excretion ; d iarrhoea, if it occurs, is general ly concom itant with the shedding of oocysts (Fayer et

al. 1998; Uga et al. 2000; G rinberg et al. 2002; F igure 1.2).

In neonatal calf cryptosporidiosis rem ission general ly occu rs, and death rates seem to be very

low in wel l -m anaged farms (O'Handley et al. 1999 ; Uga et al. 2000; G ri nberg et al. 2002).

Consu latat ion of the scientific l iterature revealed only one report of h igh mortality in calves with

cryptosporidiosis in o ne farm (Sanford et al. 1982) . Evidence of differences in breed

susceptibi l ity to cryptosporidiosis is l im ited. Some opin ion leaders have suggested that "when

infection occurs in beef calves, it is usual ly more severe in these calves than in dai ry" (Oison et

al. 2004) , although neither supporting data or references are provided. The study in New

Zealand reported i n Sect ion 2.1, which was published i n 2005, suggests a possible effect of the

breed on the susceptib i l ity of calves to Cryptosporidium i nfections. In terestingly , a s im i lar effect

has been suggested by the results of a subsequent molecu lar epidemio log ical study i n the USA,

in which none of the Jersey cattle surveyed were found to be carrying C. parvum (Starkey et al .

2006).

Unl ike coccidian parasites , Cryptosporidium oocysts do not requ i re part icular environmental

conditions to become infectious, as they are excreted ful ly i nfectious in the faeces. Each

infected calf excretes hu ndreds of m i l l ions of oocysts during the patent period, readi ly

contaminati ng the farm environment with oocysts that are fai r ly resistant to phys ical and

chem ical inactivants (Gri nberg et al. 2002; reviewed by Fayer 2008) . As a resu lt,

cryptosporidiosis can become a permanent problem on farms with anecdotal evidence

suggest ing that clean i ng and chemical dis infection of the facilities provide little rel ief. I ndeed,

disease incidence r isks of up to 100% have been recorded in herds with year round calving

(Uga et al. 2000; G rinberg et al. 2002). The economic losses associated with calf

cryptosporidiosis are main ly due to the i ncreased labour needed to treat calves and the costs of

diagnostic test ing. Convalescent calves are believed to develop a protective immunity (Fayer

1998). I nteresting ly, chronic infections and stunt ing have not been recorded in calves, and there

appear to be no sign ificant studies providi ng evidence that cryptosporidiosis has a long-term

effect on performance as measured , for instance, by body weight at wean ing.

15

Since there is overwhelm ing evidence indicating a patent period of on ly a few days between the

f i rst and third week of l ife, purposive sampling of an imals of this age is needed in o rder to

assess the occurrence of C. parvum in catt le. In addit ion, the epidemiological term of

'prevalence' seems inappropriate to describe the rate of occu rrence of C. parvum i n cattle, as

the i nfections are short l ived. A more useful indicator m ig ht be farm-level prevalence, that is, the

number of infected farms in a reg ion , which can be measured by testing many calves between

the f i rst and th i rd week of life on farms, as in the study presented in Sectio n 3.4 .

Cross sectional studies of calf cryptosporidiosis i n different reg ions have shown a variable

prevalence of infected farms, and of calves with in farms (Genchi et al. 1 984; Harp and

Woodmansee 1 989 ; Garber et al . 1 994 ; Santin et al. 2004, Winkworth et al. 2008; Pau l et al.

2008, Coklin et al. 2007). A few long itudinal studies from overseas applying repeated sampl ing

on the same farms reported 1 00% farm- level prevalence (Castro-Hermida et al. 2002b ; Trotz­

Wi l l iams et al. 2005). At the an imal- level , an infectio n i ncidence of 1 00% has been occasional ly

documented in longitud inal studies of calves on individual farms (Uga et a l . 2000 ; Gri nberg et al.

2002, Santin et al. 2008) , and there is a general bel ief that most calves acqu i re C. parvum

i nfections during the f i rst month of l ife. Th is idea m ig ht be tested by the applicat ion of repeated

sam pl ing of cohorts of calves on a s ign ificant number of farms, but such studies are labour­

intensive and were not found in the scientific literature, so the cumu lative incidence of C.

parvum i nfections in calves is not known .

Some farm management characteristics have been evaluated as potential risk factors for

Cryptosporidium i nfections, with conflict ing results. For example, some studies reported g reater

farm-prevalence in dairy herds than in open range cow-calf beef un its (Oison et al. 1 997; Kvac

et al. 2006). Converse ly, in Tennessee (USA) , a g reater prevalence of cryptosporid iosis was

found in farms where calves were al lowed to nurse with dams (Qu igley et al. 1 994) , wh ich was

typical for open range cow-calf farms. In another study in the State of New York, a decreased

risk of infection was found in farms with artificial feed ing of calves (Moham med et al. 1 999). In a

study in Spain , calf m anagement practices had n o effect on the prevalence of C. parvum

i nfection (Castro-Hermida et al. 2002a) , but in anothe r study, a positive association between the

herd size and the farm- level prevalence of Cryptosporidium was found (Garber et al. 1 994).

Col lectively, these contrast ing resu lts indicate a complex m u ltifactorial epidem iology

compounded by regional factors.

Despite the wide variety of genetic variants of C. parvum found in nature, no studies com paring

the effects or the i nfectious dose of different var iants in cattle were found in the scientific

l iterature. In one study using exper imental infection , the duration of oocyst shedding was

associated with the chal lenge dose, with larger doses leading to longer du ration of the sheddi ng

of oocysts (Moore et al. 2003). lt should be emphasised that the view that C. parvum is the on ly

intestinal Cryptosporidium parasitising young cattle can no longer be un iversal ly appl ied, due to

1 6

the recent recognit ion of a wisdespread distribution of C. bovis in th is host species. P revious

est imates of the C. parvum prevalence should therefore be reassessed using genetic

identification tools. Such a reassessment is important, as C. bovis is phenotypically

indistingu ishable from C. parvum but has never been found in humans, which challenges the

idea that al l Cryptosporidum i nfecti ng young cattle are potential ly zoonotic.

Up u nt i l 2005, bovine cryptosporidiosis had on ly been occasionally reported in New Zealand. I n

a letter to the editor of the New Zealand Veterinary Journal , Townsend and Lance ( 1 987)

reported that 206 out of 550 (37%) calf diagnostic faecal specimens subm itted to the R uakura

Ani m al Health Laboratory from Ju ly t i l l December 1984 - 1 986 were positive for

Cryptosporidium, with the h ighest rate of infection seen in specimens f rom 4- 1 4 days o ld calves.

I n 2003, Learmonth et al. reported the occurrence of Cryptosporidium in 7% of faecal

specimens from cows (n=354) and calves (n=304) on 36 herds in the Waikato ( Learmonth et al.

2003) . However, the ages of the calves and the farm-level prevalence were not reported.

Sect ions 3.4 and 3.5 of this thesis reports two epidemiolog ical stud ies of Cryptosporidium i n

young cattle in New Zealand .

1 4-00 � � + &: 12 -00

Q .2 1 0-00

0 m 8-00 c <ll

� 6-00 m � <ll

4-00 :::J CT m 1ii 2-00 <ll Q)

....J

0

• Treated • Untreated

5 7 9 1 1 1 3 1 5

+ Observation day 1 8 21

Figure 1 .2. Least square means of the log1 0 (1 + number of C. parvum oocyst per gram of faeces) in 20

newborn calves affected with cryptosporidosis in a dairy farm in Israel. Red bars: 1 0 untreated calves;

green bars: 1 0 calves treated with paromomycin sulphate between Days 1 and 9 of life. OPG+ 1 = oocyst

per gram of faeces + 1 ; The calves' age in days are indicated on the X axis (from Grinberg et a l . 2002}.

This f igure will be i ncorporated in the thesis pubished on line upon receipt of the copyright permission from

the publishers.

1 .6.2 Infections with Cryptosporidium in small rumi nants

Knowledge about the pathogenesis of Cryptosporidium i nfections in lambs and kids is scarce,

and the impact of cryptosporidiosis on the health of smal l rum inants is not we l l def ined. In a

literature review, de Graaf suggested that C. parvum is an important pathogen of lambs and

17

kids (1999) , but no support ing data were provided. S imi larly, the public health s ign i f icance of the

i solates isolated from sheep and goats is not understood , as wide variat ion is reported in the

prevalence of potential ly zoonot ic taxa in these host species.

Cl in ically-overt Cryptosporidium i nfections in small ruminants were f i rst described in 1- 3 week

o ld lambs ( Barker and Carbonel l 1974) , and subsequently in a 2-week old kid with d iarrhoea

(Mason et al . 1981 ) . I nteresting ly, Koch's postu lates were fulfi l led in specific-pathogen-free

lambs using a calf isolate (Angus et al. 1982) . Although the natural h istory of cryptosporidiosis

in lambs and kids is not wel l u nderstood, i t appears to be sim i lar to the d isease observed in

calves. Documented c l in ical s igns i nclude d iarrhoea, depression and anorexia , accompanied by

the excret ion of faecal oocysts (Anderson 1982 ; Angus et al . 1982 ; Tzipori et al. 1982 ;

Thamsborg 1990 ; Ortega-Mora and Wright 1994) .

A variable prevalence of Cryptosporidium oocysts i n faecal specimens f rom sheep has been

reported in different studies, and were recently summarised by Santin and Trout (2008) .

However, there are confl ict ing reports about the genetic m akeup and zoonotic potential of the

Cryptosporidium parasites i nfect ing lambs and kids. In 1998, Morgan et al . reported on the

identification of C. parvum (the "calf g roup" in that paper) in a small number of goats and one

lamb (Morgan et al. 1998) . In a recent extensive molecular epidemiological study, C. parvum

was the only species identif ied i n 137 diarrhoeic lambs and 17 goat kids. Al l were under 21 days

of age, and located on 71 s heep and 7 goat farms in the north-eastern reg ion of Arag6n , Spain

(Quilez et al. 2008) . In support of this f inding , Muel ler- Doblies and colleagues (2008) reported

that C. parvum was the dom inant species isolated from d iarrhoeic lambs i n the United Ki ngdom ,

but C. bovis and the 'cerv ine genotype' were also identif ied i n some specim ens. By contrast, the

"cervine genotype" was the predominant genetic variant in a random sample of faecal

specimens from subcl i n ical ly i nfected lambs on ten farms in Belg ium, whi le on ly C. parvum was

identified in kids in the same region (Geurden et al. 2008). Conversely, Chalmers et al. reported

a novel 'sheep genotype' in subcl in ical ly i nfected lambs (Chalmers et al. 2002). In a recent

report, Pritchard and eo-workers reported on the identif ication of C. parvum in 43 out of 48

oocyst-positive faecal specimens from lambs subm itted to diagnostic laboratories in E ng land

and Wales for post morten examination ( Pritchard et al. 2008) . Other species sporadical ly

isolated from lambs were C. bovis (Pritchard et al. 2008) and C. hominis ( Ebeid et al. 2003;

Gi les et al. 2009). I nterest ing ly, preweaned lambs in Western Austral ia were recently found to

be infected with C. bovis (n = 52) , the 'cervine genotype' (n = 1 0) , and C. parvum (n = 2) when

a genetic identification scheme using the 18S rRNA gene was used for the identif icat ion.

However, when the same sam ple of faecal specimens was typed target ing a second locus (a C­

type lectin-encoding gene which, accord ing to the authors, is C. parvum-specif ic) , 63 C. parvum

were identif ied (Yang et a l . 2008). lt is unclear how the authors could d ifferentiate between C.

bovis and C. parvum us ing the above C-type lectin-encoding gene, as the sequence of this

18

gene in C. bovis is sti l l unknown. Recently, Paoletti et al. (2009) identified C. parvum i n 26/26

PCR-positive faecal specimens from lambs on six farms in central Italy.

1.6.3 Infections with Cryptosporidium in horses

Equ ine cryptosporidiosis was f i rst reported in immunodeficient Arabian foals us ing m icroscopic

parasito logical methods, fol lowed by a few descriptions of overt infections also in

immunocompetent foals (Snyder et a l . 1978 ; G ibson et a l . 1983 ; Gajadhan et al. 1985 ; Coleman

et al. 1989). A num ber of surveys indicate subcl in ical Cryptosporidium i nfect ions are re latively

com mon in horses (Tzipori and Campbell 1 981; Netherwood et al. 1994 ; Xiao and Herd 1994 ;

Cole et al. 1998 ; Chalm ers et al. 2005). However, reports conf i rming the causative role of these

parasites in diarrhoea of horses are scant.

Early unsuccessfu l attempts to produce exper imental disease in foals using calf

Cryptosporidium isolates in the 1980s (Tzipori 1983) induced some authors to believe that

horses are infected with unique Cryptosporidium variants (Saul Tzipori , personal com mu nication

to the author from 2002) . This idea has been recently reiterated by other authors , who have

suggested the existence of a Cryptosporidium 'horse genotype' (see Section 1.3). However, the

Cryptosporidium ' horse genotype' has only been reported in one isolate from a Przewalski 's wild

horse ( Equus przewalskit) i n a zoo in the Czech Republic (Ryan et a l . 2003) .The orig ina l paper

reporti ng this novel genetic variant i ncluded a number of discrepancies. For instance, whi le it

was concluded that the isolate from the E. przewalskii was "most related" to Cryptosporidium

wrairi based on its 18S rRNA gene sequence, the dendrogram presented in the same paper

showed that the isolate clustered also with C. parvum ( Ryan et al. 2003). S imi larly, whi le it was

acknowledged that the 70 kDa heat shock protein gene of the isolate f rom the Przewalski's wild

horse was not determined (see Table 1 in Ryan et al. 2 003) , the authors concluded that a "h igh

degree of sequence identity" with C. wrairi was also at th is locus. In order to further investigate

these d iscrepancies , th is author retrieved the 18S rRNA gene sequence of the isolate f rom the

Przewalski's wild horse from Genbank, dubbed the 'horse genotype' by Xiao and Feng (2008) ,

and al igned it with the 18S rRNA gene sequence of C. parvum and C. wrairi publ ished in the

same database by R yan et al. The a l igned nucleotide sequence of C. parvum, C. wrairi, and the

isolate f rom the Przewalski's wild horse, are shown in Figure 1 .3. lt is notable that there is a

very smal l , s im i lar number of polymorphisms between the sequence of the 'horse genotype' and

C. parvum and C. wrairi, which is inconsistent with the idea that the former was mostly related to

C. wrairi.

The f i rst known outbreak of cryptosporid iosis in domestic foals i ncorporating cl in ical ,

epidemio log ical and pathological data, as wel l as the identification of the outbreak iso lates as C.

parvum 'cattle' genotype, is reported in Section 2.1 of this thesis. Fol low up studies are

reported in Sections 2.2 and 2.3.

19

C . p a r vum - - - - - - - - - - - - -T CGAT T C C G G AGAGGGAGCC TGAGAAAC G G CTACCACATC

C . w r a i r i - - - - - - - - - - - - TCGATT C C G G AGAGGGAGCC TGAGAAAC G G CTACCACATC

H o r s e G e n o t ype - - - - - - - - CGATT C C G G AGAGGGAGCC TGAGAAAC G G CTAC CACATC

C . p a r vum TAAGGAAGGC AGCAGGCGCG CAAATTACCC AATCCTAATA CAGGGAGGTA

C . w r a i r i TAAGGAAGGC AGCAGGCGCG CAAATTACCC AATCCTAATA CAGGGAGGTA

H o r s e G e n o t ype TAAGGAAGGC AGCAGGCGCG CAAATTACCC AATCCTAATA CAGGGAGGTA

C . p a r vum G T GACAAGAA ATAACAATAC AGGACTTTTT GGTTTT GTAA TTGGAATGAG

C . wr a i r i G T GACAAGAA ATAACAATAC AGGAC TTTTT GGTTTTGTAA TTGGAATGAG

H o r s e G e n o t ype G T GACAAGAA ATAACAATAC AGGAC TTTTT GGTTTTGTAA TTGGAATGAG

C . p a r vum TTAAGTATAA ACCCC TT TAC AAGTATCAAT TGGAGG GCAA GTCTGGTGCC

C . w r a i r i TTAAGTATAA AC C C C T T TAC AAGTATCAAT T G GAGGGCAA GTCTGGTGCC

H o r s e G e n o t ype TTAAGTATAA A C C C C T T TAC AAGTATCAAT TGGAGGGCAA GTCTGGTGCC

C . pa rvum GCAGCCGCG GTAATTC CAG CTC CAATAGC GTATATTAAA GTTGTTGCAG

C . wr a i r i AGCAGCCGCG GTAATT C CAG CTC CAATAGC GTATATTAAA GTTGTTGCAG

H o r s e G e n o t ype AGCAGCCGCG GTAATTC CAG CTC CAATAGC GTATATTAAA GTTGTTGCAG

C . p a r vum TTAAAAAGCT CGTAGTT GGA TTTCTG TTAA TAATTTATAT AAAATAT T T T

C . w r a i r i TTAAAAAGCT CGTAGTT GGA TTTCTGTTAA TAATTTATAT TAATATTTT

H o r s e G e n o t ype TTAAAAAGCT CGTAGTTGGA TTTCTGTTAA TAATTTATAT AAAATAT T TT

C . p a r vum G AJGAATATT TATATAATAT TAACATAATT CATATTACTA TATA TTT . A

C . w r a i r i GA . AAATATT TATATAATAT TAACATAATT CATATTAC TA ��TAT T TT T

H o r s e G e n o t ype GAAAAATATT TATATAATAT TAACATAATT CATAT TACT G A TA�TTT

C . p a r vum GTATATGAAA TTTTACTTTG AGAAAATTAG AGTGCT TAAA GCAGGCATAT

C . wr a i r i GTATATGAAA TTTTACTTTG AGAAAATTAG AGTGCT TAAA GCAGGCATAT

H o r s e G e n o t ype GTATATGAAA TTTTACTTTG AGAAAATTAG AGTGCT TAAA GCAGGCATAT

C . p a r vum G C C T T GAATA CTCCAGCATG GAATAATATT AAAGAT T T T T ATCTTTCTTA

C . w r a i r i G C C T T GAATA CT C CAGCATG GAATAATATT AAAGATTTTT ATCTTTCTTA

H o r s e G e n o t ype G C C T TGAATA CTCCAGCATG GAATAATATT AAAGATTTTT ATCTTTCTTA

C . p a r vum TTGGTT CTAA GATAAGAATA ATGAT TAATA GG GACAGTTG GGGGCA

C . w r a i r i T T G G TTCTAA GATAAGAATA ATGATTAATA G GGACAG TT G GGGGCA

H o r s e G e n o t ype T T G G TTCTAA GATAAGAATA ATGATTAATA G G . ACAG T T G GGGGCA

Figure 1 .3 N ucleotide polymorphisms (in yellow) between the 1 88 rRNA gene sequence of the so-called

'horse genotype' (Genbank accession number AY273770) , C. parvum (Genbank accession number

AF093490), and C. wrairi (Genbank accession number AF1 1 5378). The Cryptosporidium 'horse genotype'

gene sequence is reported in ful l , as originally reported in GenBank. For C. parvum and C. wrairi, only the

segments overlapping the 'horse genotype' sequence are reported.

1 .6.4 Infections with Cryptosporidium in cervids

Little is known about the epidem iology of Cryptosporidium infections in deer. Tzipori et al.

reported an outbreak of diarrhoea in young red deer associated with the presence of

Cryptosporidium oocysts in their faeces, and h istopathological lesions consistent with

cryptosporidiosis (Tzipori et al. 1 98 1 a} . I n 2002, a n 1 8S rRNA gene sequence o f a

Cryptosporidium 'deer genotype' (synonym 'cervi ne genotype') was reported i n deer (Genbank

accession number AY1 209 1 0) , without specifyi ng the species of deer (Xiao et al . 2002} . The

same genetic variant was later identified i n a number of host species (summarised by Santi n

and Fayer 2007) i ncluding humans (N ichols 2008) , and specifically, one person in New Zealand

(Learmonth et al. 2004} . I nterest ing ly, a Cryptosporidum "deer- l ike genotype" was also

20

described, but th is variant has on ly been reported in calves. The descriptor "deer- l ike" was

proposed to account for the s im ilarity between the 1 8S rRNA gene sequence of th is variant and

the above 'deer genotype'. Recently, the species name "Cryptosporidium ryanae" was

proposed for the Cryptosporidium "deer- l ike genotype" (Fayer et al. 2008}. The authors based

the assig nment of a species status to the Cryptosporidium "deer-like genotype" on the size of

the oocysts (which m easure 2.9-4 x 2.9-3.6 1-Jm and are thus smaller than C. parvum), and the

genetic d ifferences from other known Cryptosporidium taxa at multiple loci.

In an epidemiolog ical study of four farms in China, two out of 1 24 faecal specimens from farmed

sika deer (Cervus nippon Temminck) contained Cryptosporidium oocysts (Wang et al. 2008}.

The isolates were assigned to the 'cerv ine genotype' , although thei r 18S rRNA gene sequence

differed sl ightly from the publ ished sequence of th is genotype. Subcl in ical Cryptosporidium

i nfect ions in captive wh ite-tailed deer ( Odocoileus virginianus) have been reported ; the ocysts

recovered from the infected deer were infectious to neonatal mice and calves (Fayer et al .

1996). Asymptomatic sheddi ng of Cryptosporidium oocysts has been reported in I reland

(Skerrett and Holland 2001 ), but the oocysts have not been identified. Cryptosporidiosis i n

farmed deer calves has been reported in New Zealand (Orr et al. 1985). Although there is

anecdotal evidence of cryptosporidiosis being a sig nif icant clinical problem in young farmed

deer in New Zealand, no reports were found in the veterinary literature. Only one report of an

i nfection with a conf irmed C. parvum in a deer was found in the scientific literature, but the

species of deer was not specified (Sulai man et al . 1998).

1 .6.5 Infections with Cryptosporidium in dogs and cats

I n 1 979, lseki proposed the name Cryptosporidium felis as a descriptor for the parasites found

in cats in Japan (from Fayer 2008}. Later, Sargent et al. (1998) described a Cryptosporidium

"smal l oocyst type" i n subcl in ically infected cats, and complemented the phenotypic f indings

with genotyping, resu lt ing in the def in ition of a novel genotype in th is host species. The fi rst

paper suggesting C. felis as a genetical ly-distinct species was published in 1998 (Morgan et al.

1998). As noted i n Section 1 . 6. 1 , C. felis was also identified in one cow (Bornay Ll inares et al.

1 999}. Santin and colleagues recently found C. felis and C. muris in cats in Bogota, Colombia

(Sant in et al. 2006}. No other Cryptosporidium taxa are known to cycle i n fel ine popu lations. In a

literature review of 58 cases of human C. felis i nfect ion reported in different parts of the world ,

more than 80 % of cases occurred i n H IV-positive patients (Raccurt 2007).

The f i rst evidence of the cycl ing of Cryptosporidium parasites in dogs was provided by Tzipori

and Campbell (1981 } , who reported the presence of anti- Cryptosporidium antibodies i n 1 6/20

dogs. In 2000, Morgan and col leagues ( 2000) identified a novel Cryptosporidium variant in e ight

dogs from Austral ia and the US, which they dubbed the "dog genotype". This genotype was

elevated to the species status by Fayer et al. one year later ( Fayer et al. 2001 ).

21

Surveys aimed at assessing the prevalence of Cryptosporidium in canine and fel ine popu lat ions

have been performed in different countries, with contrasting resu lts (recently reviewed by Sant in

and Trout 2008) . In a recent study, C. canis and C. fe/is were the main variants found in

Australian dogs and cats, respectively ( Palmer et al. 2008). I n another survey in the USA

Cryptosporidium oocysts were found in 30/250 domestic cats and al l the successfu l ly

genotyped iso lates were ident if ied as C. felis based on the sequence of the 1 8S rRNA gene

( Bal lweber et al . 2009). As C. canis, C. felis and C. muris have on ly occasional ly been found in

humans, avai lable epidemiological evidence would preclude dogs and cats as s ign ificant

sources of zoonotic cryptosporidiosis in immunocompetent humans. However, the zoonotic

species C. parvum has been sporadically identified in dogs Italy (Giangaspero et al. 2006), the

USA ( Fayer et al. 2001 ) , and Czeck Republic (Hajdusek et a l . 2004). Furthemore, the zoonotic

potential of Cryptosporidium f rom cats has been suggested after one successful attem pt to

infect 6-weeks-old kittens by o ral inoculation of an isolate recovered from an immunodeficient

person (Current et al. 1 983) , and from one cat-to-human transmission event inferred in one

case of cryptosporidiosis i n a ch i ld fol lowing exposure to an i nfected cat ( Egger et a l . 1990). In

addition , X iao and colleagues reported an infection with C. canis i n one g i r l , her brother and dog

l iv ing i n L ima, Peru. Both chi ldren had diarrhea, but the dog was asymptomatic (Xiao et al.

2007). No data on the pathogenicity of Cryptosporidium parasites in dogs and cats were found

in the scientific l iterature.

1 .6.6 Infections w ith Cryptosporidium in pigs

Although both C. parvum and C. hominis have been propagated in pigs (Widmer et al . 2000;

Pereira 2002; Akiyoshi et al . 2003; Ebeid et al . 2003), knowledge about the prevalence and

genetic makeup of these parasites in farmed pigs is scarce. Cl in ical s igns of depression ,

diarrhoea and vom iting have been observed i n pig lets experimentally infected with oocysts

origi nat ing from calves (Tzipori et al. 1981 b). However, no clear association between i nfection

and disease is known to exist, as surveys have reported subcl in ical i nfections in pigs (Gusel le et

al. 2003, Qui lez et al. 1996).

In 2004, the species name of C. suis was suggested for a genetic variant found in pigs ( R yan , et

a l. 2004). Recently, a 22% infection rate with C. suis and the 'pig genotype 1 1 ' was reported in

289 p igs located i n four Western Austral ian faci l it ies (Johnson et al. 2008). Xiao and colleagues

(2008) found DNA sequences of C. suis, the 'pig genotype 1 1 ' , and C. muris, i n 25 out of 56 pig

slurry samples f rom 33 I rish farms. The Cryptosporidium 'pig genotype 1 1 ' was found in 33/33

pigs from one faci l ity in Canada (Guselle et al. 2003). Because the most common

Cryptosporidium taxa found in pigs were on ly occas ional ly found in humans, some authors

hypothesised that "domestic pigs do not pose a sign ificant public health risk" (Johnson et al.

2008). However , the zoonotic C. parvum has also been identified in the Czech Republic (Kvac

et al. 2008) . Furthermore, p ig lets cou ld be experimental ly i nfected using isolates from calves

(presumably C. parvum) (Tzipori et al. 1981 b), suggesting there is no biological restriction for

22

the cycl ing of zoonotic C. parvum i n pigs. No data about the pathogen icity of Cryptosporidium

taxa occurring in pigs was found in the consu lted l iteratu re.

1.7 GENETIC TYPING OF CRYPTOSPORID/UM

Despite the recent i mprovements of in vitro cult ivation (H ijjawi et al. 2002, 2004) ,

Cryptosporidium are st i l l hard to g row in vitro and thei r propagation depends on animal

inocu lation. Therefo re, disease diagnosis depends on the di rect detection of the oocysts or

antigens in the faeces by microscopic techniques, or commercial kits, rather than the isolation of

the agent. Because the oocysts of many i ntesti nal Cryptosporidium are s im i lar , and the

phenotypic identif icatio n of the taxon of an isolate is not feasible, the generic term of

cryptosporidiosis is commonly used in diagnostic practice. Fortunately, the genetic typabi lity of

isolates does not seem to be severely compromised if faeces are refrigerated , either with or

without preservatives, which has enabled retrospective molecu lar epidemiological studies to be

performed using extracted DNA from such faeces.

Although Cryptosporidium taxa are genetical ly s im i lar , m any polymorphic genetic loci , which can

be used for the typi ng of cl in ical or environmental parasites, exist. As stated above, the f i rst

genetic polymorph isms between Cryptosporidium isolates were detected in the early 1 990s

(Ortega et al. 1991 ). Later, techn iques involv ing the PCR am plification of genes fol lowed by

either an analysis for the presence of restriction f rag ment length polymorphisms ( RFLP) , or

sequencing of the amplicons, were developed (Bonnin et al. 1 996 ; Leng et a l . 1996; Carraway

et al. 1997). The h igh reproducibi l ity and cost effectiveness of the PCR enabled the extensive

use of molecu lar tools i n epidem iolog ical surveys, which resu lted i n the identification of a long

l ist of Cryptosporidium taxa.

The f i rst step of any genetic characterisation of a Cryptosporidium iso late is its ass ignment to a

taxon. Genetic loci that are h igh ly conserved with in taxa, but on the other hand sufficiently

polymorphic to allow differentiation between taxa, are used for this purpose. As stated above,

Caccio and colleagues (2005) recommended the use of the 188 rRNA gene for all genetic

identification schem es for Cryptosporidium. The advantages of us ing this gene for taxonomic

purposes have been described in Section 1 .3 .

Once an isolate has been assigned to a taxon, genotyping beyond this level is not necessary

un less further specific detai ls are requ i red. Conversely, if a differentiat ion between genetic

variants with in the same taxon is needed (for instance, to infer transmission routes or sources of

infect ions) , further subgenotyping (synonym : subtyping) at loci that are polymorphic within the

taxa, such as m icro and minisatel l ite repeats, is requi red. Micro and m in isatell ite repeats are

sequence repeats of variable length found throughout the genomes of eukaryotes. They are

usual ly composed of a variable num ber of repeats of 1 -6 basepairs (microsatel l ites) or 10-500

basepai rs (minisatel l i tes) (Pevsner 2003). M icro and m inisatell ite sequences may exhibit length

23

polymorph isms due to either loss or gain of repeat un its, presumably due to sl ippage during

DNA repl ication (Jeffreys et a l. 1985). As a resu lt, alleles of different s izes can often be

observed i n eukaryotic organisms at m in i and m icrosatel l ite loci, enabl ing the identification of

unique genetic f ingerprints. These loci are advantageous for epidemiological i nference, thanks

to their h igh degree of polymorphism.

M icrosatel l ite and m in isate l l ite length polymorphism analysis was fi rst used for subtyping of

Cryptosporidium i n the early 2000s (Feng et al. 2000 ; Caccio et al. 200 1 ). Because the

genomes of C. parvum and C. hominis iso lates have now been sequenced (Abrahamsen et al.

2004; Xu et al. 2004} , it is poss ible to exam i ne the genomic databases of these species for the

presence of repeat sequences that are l ikely to show polymorph isms. The 70 kDa heat-shock

protein (HSP70) gene of C. parvum and C. hominis contains such a polymorphic repeat

(Khramtsov et al. 1995; Mallon et al. 2003}. Another usefu l polymorph ic protei n-coding repeat is

the polyser ine repeat reg ion of the Cryptosporidium 60 kDA surface g lycoprotein gene. This

gene codes a 60 kDa precursor protein that is cleaved i nto two sub-u nits, G P15 and G P40, and

so it is also known as the 'G P60', 'G P40/15' , o r 'G P45/15' gene (Ceval los et al. 2000 ; Strong

et al. 2000).

Both s ing le- locus and mu lt i - locus subtypi ng schemes have been used for the subtyp ing of

Cryptosporidium. The s ing le l ocus subtyping approach i nvolves the characterisation of each

isolate at a sing le polymorphic locus, whereas multi locus subtyping schemes use two or more

loci. The s ing le locus approach is useful for source tracking studies, and is advantageous, as

compared with mu lti locus subtyping schemes, due to its s implicity. Indeed, a big disadvantage

of the multi locus approach is that fai lure to characterise an isolate at all the loci could preclude it

from the analysis. On the other hand, the mu lt i locus approach is, i ntu itively, more discrim inatory

than any of the corresponding single locus schemes. In add ition , as C. parvum and C. hominis

undergo sexual recombinat ion within the species as part of their l ife cycle (Tanriverdi et al.

2007} , s ing le-locus subtyping schemes are l im ited in their abi l ity to capture phylogenetic

relations between isolates.

1 .8 CRYPTOSPOR/DIUM PA R VUM AND C. HOMINIS POPULATION G ENETIC STRUCTURE

The genetic structures of local C. parvum and C. hominis populations have been studied i n

several locations using s ing le o r multiple genetic markers. Observations based on s ing le loci

(Peng et al. 1997; Widmer et al. 1998 ; Caccio et al. 200 1 ) contributed to the proposal of

elevat ing the genetically-distinct, anthroponotically transmitted C. parvum "human genotype" to

a new species, which was named C. hominis ( Morgan-Ryan et al. 2002}.

Due to its extensive sequence polymorphism , the G P60 gene has been widely used for studying

C. parvum and C. hominis populations ( Leav et al . 2002; Alves et al. 2003, 2006; Sulaiman et

al. 2005 ; Akiyosh i et al. 2006; Misic and Abe 2006; Xiao et al . 2006, Waldro n et al. 2009} . Th is

24

appoach has led to the identification of what has been designated as Cryptosporidium G P60

"subtypes", or "fam i l ies" (Sulaiman et al. 2005). Mu lt i locus subtyping schemes us ing un l inked

loci (that is, loci that do not eo-segregate duri ng the meiotic process) are amenable to the study

of the population genetic structure using al lel le l i nkage disequi l ibriu m analysis , which is not

possible using the s ing le locus approach. Such schemes have been applied to the study of C.

parvum and C. hominis populat ions in Scotland ( Mal lon et al. 2003) , India (Gatei et a l . 2007a,b) ,

and the Middle East (Tanriverdi et a l . 2006). However, l ittle is known about the population

structure of C. parvum and C. hominis on a g lobal scale.

Five molecular epidem iological studies of Cryptosporidium using s ing le or mu lt i locus subtyping

approaches are reported in Chapters 2 -4 of th is thesis.

1 .9 ZOONOTIC CRYPTOSPORIDIOSIS

As stated above, in the 1 980s, intest inal Cryptosporidium parasites were sti l l considered to be a

s ingle species known as C. parvum (Tzipori et al. 1980a) , and cryptosporidiosis was mainly

considered a zoonotic disease (Schultz 1 983). However, at about the same time, the

observation of human -to-human transmission events i nd icated a more complex epidem iology

for human cyptosporidiosis (Casemore and Jackson 1984) . Final ly, i n 1998, Awad-ei-Kariem

used isoenzyme e lectrophoresis to demonstrate the existence of genetical ly d istinct "human

and an imal populations of C. parvum", a pattern later repeatedly confi rmed genotypical ly us ing

PC R , and eventual ly substantiated by the dist inction into the species of C. parvum and C.

hominis.

Since then , seven Cryptosporidium species and a n umber of other taxa have been identified in

hum ans. Of these, C. hominis and C. parvum account for the vast majority of infections in

im munocompetent and immunocompromised people, and other taxa have on ly been found

sporadically, mainly in immunocompromised individuals. The taxa sporad ically i nfecting humans

have been repeatedly reviewed and are C. meleagridis, which infects mainly avian species, C.

canis, found i n dogs, C. fe/is, found in cats, C. muris, found in rodents , the Cryptosporidium

'cervi ne genotype' , or iginal ly found in deer but also from sheep and cattle, the Cryptosporidium

suis- l ike genotype, Cryptosporidium andersoni and the andersoni-l ike genotype, the 'ch ipmunk

genotype' , the 'skunk genotype' and Cryptosporidium 'monkey genotype' (Xiao and Fayer 2008 ;

Nichols 2008 ; Xiao and Ryan 2008) , and most recently the Cryptosporidium 'rabbit genotype'

(Chalmers et al. 2009). Xiao and Fayer (2008) argued that, as the rare genetic variants found in

humans were identified from m icroscopically-positive faecal specimens from clin ical ly-affected

individuals, they reflect active i nfections rather than a passive transit of oocysts i n the

gastrointestinal tract.

Although C. hominis is believed to cycle primari ly among hum ans, it has also been reported in a

dugong (Dugong dugon) ( Morgan-Ryan et al. 2002) , in non-human primates (Akiyoshi et al.

25

2003) , cattle (Smith et al. 2005b) , lambs (G i les et al . 2009), and one goat (Gi les et al. 2009). I n

addition , C. hominis was propagated in g notobiotic and conventionally reared piglets (Widmer et

al. 2000; Ebeid et al. 2003) , one lamb (G i les et al. 200 1 ) and Mongol ian gerbi ls (Meriones

unguiculatus) (Baishanbo et al . 2005) .

Conversely, C. parvum cycles in hum ans and other anim als, in part icular young cattle, and it is

considered potential ly zoonotic. There are reports of C. parvum infections in a wide spectrum of

mammalian hosts, including horses. However, most studies have been based on m icroscopy,

with no genetic characterisation of the isolates, and basically, the host-range of C. parvum is

sti l l unresolved.

Thus far, natural infections with C. parvum in animals have been conf i rmed by means of

molecu lar too ls in rum inants (cattle, sheep, goats, s ika deer) , pigs, dogs (see Section 1 .5 ) , two

pet leopard geckos ( Eublepharis macularius) (Pedraza-D iaz et a l . 2009), and tortoises (one

Testudo graeca and two Testudo hermannt) (Traversa et al . 2009). However, it should be noted

that the genetic identificat ion in the tortoises was performed by sequence analysis of the COWP

gene, and the makeup of this gene in many Cryptosporidium taxa is unknown. The studies

presented in Sections 2.1 and 2.3 of this thesis are the fi rst known reports reports of c l in ical ly­

overt infectio ns with C. parvum in horses.

The majority of reports of C. parvum i nfections in an imals concern cattle. Whi le there is a large

amount of evidence indicati ng C. parvum is the most prevalent species in pre-weaned calves, it

has rare ly been found in post-weaned calves. Conversely, the prevalence of C. bovis and C.

andersoni, which seem to have little zoonotic significance, tends to increase in post-weaned

and juven i le cattle (Sant in et al. 2004 ; Fayer et al. 2006; Sant in et a l. 2008). lt is widely

accepted that young calves are mainly infected with C. parvum, and that these animals are a

major source of zoonotic C. parvum. The possibi l i ty of a di rect zoon otic transm ission of C.

parvum from cattle is supported by the resu lts of transm ission studies in human vol unteers

using bovine isolates, and from numerous case and case-control studies i n people in several

countries, i ncluding New Zealand ( Pohjola et al. 1 986; Reif et al. 1 989; M i l lard et al. 1 994;

DuPont et al. 1 995; Chappel and Okhuysen 200 1 ; Stefanog iannis et a l . 2001 ) . However, the

relative contribution of the oocysts or ig inating from cattle as a source of human

cryptosporid iosis is diff icu lt to assess, as human-to-human and bovine-to-human transm issions

can eo-occur. Perhaps the most convinci ng data support ing a wide zoonotic transm ission

through the contamination of the ecosystem with bovine C. parvum oocysts was the sharp

decrease in the number of notifications of h uman cryptosporidiosis in E ngland and Wales during

the 200 1 foot and mouth disease epidemic. This decrease has been attributed to the restrict ions

on human m ovement i n ru ral areas and the extensive cul l ing of animals imposed dur ing the

epidemic , which reduced human exposure to livestock (Smerdon 2003 ; Sopwith et al . 2005) .

Interest ingly , further epidemiological evidence support ing the widespread zoonotic transm ission

26

of C. parvum origi nated from New Zealand. Here, human i nfections with C. parvum fol low a

seasonal pattern, with the number of notifications peaking every year in sprin g and early

summer, soon after the calving season. Using faecal speci mens submitted to diag nostic

laboratories, Learmonth et al. (2003, 2004, 2005) elegantly showed that this peak is

accompanied by a virtual substitution of the anthroponotic C. hominis, which is seen year round,

wi th the potent ial ly zoonotic C. parvum (F igure 1.4) . Thus, it is possible that the synch ronous

presence of m i l l ions of h igh ly susceptible newborn calves determines a massive amplification of

C. parvum, contributing to the cattle-to human transm ission of cryptosporidiosis at that time.

Unt i l recently, a h igh prevalence of C. parvum i nfections i n humans i n a given reg ion , as

compared to C. hominis, was general ly viewed as evidence of a widespread zoonotic

transmission of cryptosporidiosis i n the reg ion (Mclaug h l in et al. 2000; Learmonth et al. 2003,

2004, 2005 ; Hunter and Thompson 2005). However, some researchers rejected th is view, and

suggested a more complex m odel, which contemplates the existence of anthroponotic C.

parvum parasites that do not cycle in cattle (Mal lon et al . 2003; Xiao et al. 2004). This concept

is of considerable public health i nterest, as it i m plies that any barrier across the livestock­

human interface would be ineffective in reg ions where anthroponotic C. parvum cycl ing prevai ls.

The anthroponotic model of transmission of C. parvum is supported by two l i nes of evidence.

One l ine, proposed by Xiao et al. i n a review of the l iterature (2004) , is based on the observation

of C. parvum isolates from humans carrying the G P60 al lele fami ly " le " (now nominated " lie"

fol lowing the acceptance of C. parvum and C. hominis as different species) , which were never

found in cattle. According to this argument, C. parvum carryi ng lie G P60 alleles are

anthroponotic and do not cycle in cattle. However , the literature support ing this model reveals a

more complex picture.

27

300

250

200

150

lOO

50 \. \ � 0 t.�l I ! I I 1 1 I I I ! I I ! I I I! I 1 1 1 1 1 1 1 I l l \ I I l l I I l l I I l l ! 1 1 1 1 1 1 I l l I f I! I l l ! I I 1 1 I I ! I I l l 1 1 I ! ��� 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I I I l l I I l l I ! I I I I! I l l I 1 1 1 1 1 1 1

L-t 997__L_ 1 99&--L- ! 99�200�200 1__L_200�200�200�200s--L-200&--L-2007_J Month I Year

i 0 .!!! '0 ... .! e ::J

z

40

35

30

25

20

1

1 0

5

0

Oat.e

o C. patvum l>ot�lne • C. homtnls

Figure 1 .4 Upper graph : The number of cases of cryptosporid iosis notified in New Zealand between 1 997

and 2007 (from: New Zealand Public Health Surveillance Report, Institute of Environmental Science and

Research Ltd. http://www.nzpho.org.nz/NotifiableDisease.aspx, accessed October 2008); lower graph: the

seasonal sh ifts between the number of C. parvum and C. hominis isolates identified in New Zealand

between 2000 and 2003 (from : Learmonth et al . 2004). These figures will be incorporated in the thesis

publ ished on l ine upon receipt of the copyright permission from the publishers.

While it is indeed accurate that /le al leles have been identified on a number of occasions in

human C. parvum i n some geograph ical reg ions (Leav et al. 2002; Alves et al. 2003, 2006; Xiao

et al. 2004b; Gatei et al. 2007b; Waldron et al . 2009) , thei r absence in the respective cattle

populations could not be ru led out, as on ly a few or no bovine isolates from these regions were

subtyped in the same studies. Moreover, most human C. parvum isolates carrying G P60 /le

alleles or ig inated from H IV-positive pat ients, and so the general isatio n to the population as a

whole, is difficu lt. lt shou ld be also remembered that C. parvum is a genetically superdiverse

species consist ing of a large number of rare genetic variants (Mal lon et al. 2003) and so a

significant sampl ing effort would be needed in order to declare the bovine population as being

f ree of the rare C. parvum GP60 /le allele family.

28

A different l ine of evidence support ing the existence of anthroponotic C. parvum was provided

by Mallon and eo-workers (2003) . These authors subtyped C. parvum iso lates from humans and

cattle in Scotland at seven polymorphic loci, and by doing so were able to def ine the mu lt i locus

genotype of each isolate. Then, the authors generated a pairwise genetic d istance m atrix of

seven possible values ( 1 -to-6 loci difference) between the multi locus genotypes, which was

used to construct a dendrogram based on the Unweighted Pair Group Method with Arithmetic

mean (UPGMA) . Based on a simple visual i nspection of the UPGMA dendrogram ( reproduced

in Figure 1 .5 ) , the authors concluded the presence of "human-on ly" mu lti locus genotype "sub­

g roups" in the sample. However, a re-analysis of the raw data provided by Mal lon et al. by th is

author revealed numerous ties in proximity between the mu lti locus genotypes (that is , mu lt i locus

genotypes that are equidistant f rom multiple m u lti locus genotypes) . These ties determi ned the

presence of severe distort ions in the dendrog ram , which were probably not seen by the or ig inal

authors (the concept of t ies in proximity in cluster analysis and how they may distort

dendrogram s is wel l explained in MacCu ish et al. 2001 , and Arnau et al. 2005) . For i nstance,

mu lti locus genotype 21 (from the "human-only sub-group C") is a double locus variant of the

other members of the same sub-group, but mu lti locus genotypes 1 9 , 20, and 1 8 from the same

"sub-group C" have double locus variants also in "sub-group B", some of which or ig inated f rom

bovine iso lates (Figure 1 .5) . In addition, although the dendrogram would suggest the "human­

on ly sub-group A" is a wel l -defined cluster, its components (m ulti locus genotypes 2, 3 and 57)

are only weakly clustered. I ndeed, the s ing le and double locus variant networks of mu lti locus

genotypes show that mu lt i locus genotype 57 does not cluster with 2 and 3 . Further exploration

of the raw data provided by M allon et al. i nd icated that mu lti locus genotype 57 is on ly a four­

loci variant of 2 and 3.

Together, these resu lts suggest that the concept of the existence of anthroponotic C. parvum

requi res corroborat ion. A study further addressing the questio n of the existence of anthroponotic

C. parvum by a re-analys is of the data presented by Mallon et al . (2003) using a different

analytical approach is reported in Section 2 . 1 . Four studies addressing the zoonotic impact of

cattle and horses in New Zealand are presented i n Chapters 3 and 4.

29

J\-1.£. .Mnllon et al. /Jnfecnon. GPm>nrr and Emhmon 3 (.1003) :!07-11 8 r-r======:::;::=::=������� �7 � c � ·::::.:�A 3 btun.'Ul only 5 1 55 ..--c==� '---c== � r--e==� t--c==�

54 --t---48 47 40 10 1 1

'---1--- � 56 r-c==== l3 14 '-----c== �

r::::;::::::= 15 16 48

'---c== :; ..--c== �

_..t--t:==== �� --c==�� L� 26 4 1 43 28

29

,..------� 1 -"'i--- 18 19

Sub-group B Hwn .. m.

00..-incand

1...--------------c== �� I C.pa1,.mll monkey gc11otypc

, 57 , I

5 1

�----- � r------ � C. pnm1111 Type I 37 ------ 35 36

�' 38

, J•

, J

C. pm,·um Type2

, l

Fig. 1 .5 Dendrogram showing the "human only sub-groups" of C. parvum multilocus genotypes in

Scotland reported by Mallon et al. (2003) (above) , and single and double locus variant networks ( SDLVN)

of the mult i locus genotytpes (MLG), generated by the author using the data provided by Mallon et al .

(below) . I n the networks, the mu ltilocus genotypes are represented by dots l inked to their single locus

variants (purple l ines) and double locus variants (blue lines). The numbers represent the MLG identifiers

as defined by Mallon et al . Note the presence of numerous ties in proximity between MLGs of "sub-group

C" and "sub-group B" in the dendrogram. The SOL VN were constructed using eBURST software (see

Section 2.2). The upper f igure wi l l be incorporated in the thesis publ ished on l ine upon receipt of the

copyright permission from the publishers.

30

1.10 CONCLUDING REMARKS

Cryptosporid iosis is a disease of significant impact on both human and bovi ne health i n both

developed and d eveloping countries. Ernest Edward Tyzzer fi rst real ised that whi le some

Cryptosporidium p arasites i nvaded the gastric epithel ium, others were found i n the enterocytes.

He proposed the species name of C. muris tor the gastric parasites, and C. parvum tor the

intestinal parasites of mice (Tyzzer 1 9 10 , 1912) , form ing the basis for the phenotypic taxonomic

nomenclature wh ich was accepted unt i l recently. With the advent of cost effective PCR

equipment and reagents in the 1990s, many laboratories around the world started genotyping

isolates. As a consequence, a large amount of evidence ind icati ng an extensive genetic

diversity with in genus Cryptosporidium rapidly accumulated. Nonetheless, the taxonomy with in

genus Cryptosporidium is far f rom being resolved and there is sti l l no fu l l consensus o n what

constitutes a Cryptosporidium taxon. These developments forced researchers to re-assess

some of their establ ished views, and perform new epidemio logical research us ing molecular

tools. Such a rapid evolution of thought wi l l be apparent in some of the epidemiolog ical stud ies

that fol low.

1 . 1 1 REFERENCES

Abrahamsen MS, Templeton T J, Enomoto S, Abrahante JE, Zhu G, lancto CA, Deng M, Liu C,

Widmer G, Tzipori S, Buck GA, Xu P, Bankier AT, Dear PH, Konfortov BA, Spriggs HF, lyer L,

Anantharaman V, Aravind l , Kapur V. Complete genome sequence of the apicomplexan,

Cryptosporidium parvum. Science 304, 441 -5, 2004

Akiyoshi DE, Mor S, Tzipori S. Rapid displacement of Cryptosporidium parvum type 1 by type 2 in mixed

infections in piglets . Infection and Immunity 71 , 5765-71 , 2003 .

Akiyoshi DE, Tumwine JK, Bakeera-Kitaka S, Tzipori S. Subtype analysis of Cryptosporidium isolates

from children in Uganda. Journal of Parasitology 92: 1 097-1 00, 2006

Alves M, Xiao l, Antunes F, Matos 0. Distribution of Cryptosporidium subtypes in humans and domestic

and wild ruminants in Portugal . Parasitology Research 99: 287-92, 2006

Alves M, Xiao l, Sulaiman I, lal AA, Matos 0, Antunes F. Subgenotype analysis of Cryptosporidium

isolates from humans, catt le , and zoo ruminants in Portugal. Journal of Clinical Microbiology 4 1 , 2744-47,

2003

Anderson BC. Patterns of shedding of cryptosporidial oocysts in I daho calves. Journal of the American

Veterinary Medical Association 1 78, 982-84, 1 981

Anderson BC. C ryptosporidiosis in Idaho lambs: natural and experimental infections. Journal of the

American Veterinary Medical Association 8 1 , 1 5 1 -53, 1 982

Anderson BC, Bu lgin MS. Enteritis caused by Cryptosporidium in calves. Veterinary Medicine and Small

Animal Clinician 76, 865-68, 1 98 1

3 1

Angus KW. Cryptosporidiosis in man, domestic animals and b i rds: a review. Journal of the Royal Society

of Medicine 76, 62-70, 1 983

Angus KW, Tzipori S, Gray EW. Intestinal lesions in SPF lambs associated with Cryptosporidium from

calves with d iarrhoea. Veterinary Pathology 1 9, 67-8, 1 982

Arnau V, Mars S, Marin I . Iterative Cluster Analysis of Protei n Interaction Data. Bioinformatics 2 1 , 364-

78, 2005

Atwi l l ER, das Pereira M. Lack of detectable shedding of Cryptosporidium parvum in periparturient dairy

cattle. Journal of Parasitology 89, 1 234-6, 2003

Atwill ER, Johnson E, Kl ingborg DJ, Veserat GM, Markegard G, Jensen WA, Pratt DW, Delmas RE,

George HA, Forero LC, Phi l ips RL, Barry SJ, McDougald NK, Gildersleeve RR, Frost WE. Age,

geograph ic, and temporal d istribution of fecal shedding of Cryptosporidium parvum oocysts in cow-calf

herds. American Journal of Veterinary Research 60, 420-25, 1 999

Awad-EI-Kariem FM, Robinson HA, Petry F, McDona ld V, Evans D, Casemore D. Differentiation

between human and animal isolates of Cryptosporidium parvum using molecular and biological markers.

Parasitology research 84, 297-301 , 1 998

Baishanbo A, Gargala G, Delaunay A, Francrois A, Ballet JJ, Favennec L. I nfectivity of

Cryptosporidium hominis and Cryptosporidium parvum Genotype 2 Isolates in lmmunosuppressed

Mongolian Gerbi ls. Infection and Immunity 73, 5252-55, 2005

Ballweber LR, Panuska C , Huston CL, Vasilopulos R , Pharr GT, Mackin A. Prevalence of and risk

factors associated with shedding of Cryptosporidium felis in domestic cats of Mississippi and Alabama.

Veterinary Parasitology 1 60, 306-1 0, 2009

Barker I K, Carbonell PL. Cryptosporidium agni sp. n. from lambs and Cryptosporidium bovis sp. n. from a

calf with observations on the oocyst. Zeitschrift fur Parasitenkunde 44, 289-98, 1 974

Barta JR. Evolutionary Origins of Cryptosporidium species (Myzozoa, Apicomplexa) - Unique Coccidium

or Vertebrate G regarine? P roceedings of the 1 1 International Conference on Giardia and Cryptosporidium,

Morel ia, Michoacan, Mexico, May 1 3 - 1 8, 2007. Taxonomy, Nomenclature and Evolution S ession,

abstract 5, 2007

Bird RG, Smith MD. Cryptosporidiosis in man: parasite l i fe cycle and fine structural pathology. Journal of

Pathology 1 32, 2 1 7-33, 1 980

B lunt OS, Khramtsov NV, Upton SJ , Montelone BA. Molecular Karyotype Analysis of Cryptosporidium

parvum: Evidence for Eight Chromosomes and a Low-Molecular-Size Molecule. Clinical and Diagnostic

Laboratory Immunology 4, 1 1 -3, 1 997

32

Bonnin A, Fourmaux MN, Dubremetz J F, Nelson RG, Gobet P, Harly G , Buisson M, Puygauthier­

Toubas D, Gabriei-Pospisil G, Naciri M, Camerlynck P. Genotyping human and bovine isolates of

Cryptosporidium parvum by polymerase chain reaction restriction fragment length polymorphism analysis

of a repetitive DNA sequence. FEMS Microbiology letters 1 37, 207-1 1 , 1 996

Bornay-llinares FJ , da Si lva AJ, Moura IN, Myjak P, Pietkiewicz H, Kruminis-Lozowska W, G raczyk

TK, Pieniazek NJ. I dentification of Cryptosporidium felis in a cow by morphologic and molecular methods.

Applied and Environmental Microbiology 65, 1 455-8, 1 999

Borowski H, Clode PL, Thompson RCA. Active invasion and/or encapsulat ion? A reappraisal of host-cell

parasitism by Cryptosporidium. Trends in Parasitology 24, 509- 1 6, 2008

Brenner J, Elad D, Markowics A, Grinberg, A, Trainin Z. Epidemiological study of neonatal calf

diarrhoea in I srael - a one-year survey of faecal samples. Israel Journal of Veterinary Medicine 48, 1 1 3-

1 6, 1 993

Caccio S, Homan W, van Dijk K, Pozio E . Genetic polymorphism at the 13-tubul in locus among human

and animal isolates of Cryptosporidium parvum. FEMS Microbiology Letters 1 70, 1 73-79, 1 999

Caccio S, Homan W, Camill i R, Traldi G , Kortbeek T, Pozio E. A microsatellite marker reveals

population heterogeneity within human and animal genotypes of Cryptosporidium parvum. Parasitology

1 20 , 237-44, 2000

Caccio S, Spano F, Pozio E. Large sequence variation at two microsatel l ite loci among zoonotic

(genotype C) isolates of Cryptosporidium parvum. International Journal of Parasitology 3 1 , 1 082-86, 2001

Caccio SM, Thompson RCA, McLauchl in J , Smith HV. Unravell ing Cryptosporidium and Giardia

epidemiology. Trends in Parasitology 2 1 , 430-37, 2005

Cama V, Arrowood M, Ortega V, Xiao, L . Molecular characterization of the Cryptosporidium parvum Iowa

isolate kept in different laboratories. Journal Eukaryotic Microbiology. 53: 840-2, 2006

Carraway M, Tzipori S, Widmer G. Identification of Genetic H eterogeneity in the Cryptosporidium parvum

ribosomal repeat. Applied and Environmental Microbiology 62, 7 1 2-1 6, 1 996

Carraway M, Tzipori S, Widmer G. A new restriction length polymorphism from Cryptosporidium parvum

identifies genetically heterogeneous parasite populations and genotypic changes following t ransmission

from bovine to human hosts. Infection and Immunity 65, 3958-960, 1 997

Carreno RA, Martin OS, Barta JR. Cryptosporidium is more closely related to the gregarines than to

coccidia as shown by phylogenetic analysis of apicomplexan parasites inferred using small-subunit

ribosomal RNA gene sequences. Parasitology Research 85, 899-904, 1 999

Casemore DP, Armstrong M, Jackson FB. Screening for Cryptosporidium i n stools. Lancet 1 984; 1 :734-

735, 1 984

33

Casemore DP, Armstrong M , Sands RL. Laboratory diagnosis of cryptosporidiosis. Journal of Clinical

Pathology 38, 1 337-4 1 , 1 985a

Casemore DP, Jackson FB. Hypothesis : cryptosporidiosis in human beings is not primarily a zoonosis.

Journal of Infection 9 , 1 53-6, 1 984

Casemore DP, Sands RL, Curry A. Cryptosporidium species a "new" human pathogen . Journal of

Clinical Pathology 38, 1 321 -36, 1 985b

Castro-Hermida JA, Gonza les-Lozada YA, Ares-Mazas E. Prevalence of and risk factors involved in the

spread of neonatal bovine cryptosporidiosis in Galicia (Spain) . Veterinary Parasitology 1 06, 1 - 1 0, 2002a

Castro-Hermida JA, Gonzales-Lozada Y A, Mezo-Menendez M, Ares-Mazas E. A study of

cryptosporidiosis in a cohort of neonatal calves. Veterinary Parasitology 1 06, 1 1 -7, 2002b

Ceval los AM, Zhang X, Waldor MK, Jaison S, Zhou X, Tzipori S, Neutra MR, Ward HO. Molecular

cloning and expression of a gene encoding Cryptosporidium parvum glycoproteins G P40 and gp1 5.

Infection and Immunity 68, 4 1 08-1 6 , 2000

Chalmers RM, Elwin K, Reilly WJ, lrvine H, Thomas AL, Hunter PR. Cryptosporidium in farmed

animals: the detection of a novel isolate in sheep. International Journal for Parasitology 32, 21 -6, 2002

Chalmers RM, Robinson G, E lwin K, Hadfield SJ, Xiao L, Ryan U, Modha D, Mal laghan C.

Cryptosporidium sp. rabbit genotype, a newly identified human pathogen. Emerging I nfectious Diseases

1 5, 829-30, 2009

Chalmers RM, Thomas AL, Butler BA, Morel MC. Identification of Cryptosporidium parvum genotype 2

in domestic horses. Veterinary Record 1 56, 49-50, 2005

Chappel CL, Okhuysen PC. C ryptosporidiosis in health adu lt volunteers. I n : Cryptosporidiosis, the

analytical challenge. M. Smit h and KC Thompson (eds). The Royal Society of Chemistry, Cambridge, UK ,

Pp 62-72, 2001

Chappell CL, Okhuysen PC. Cryptosporidiosis. Current Opinion in Infectious Diseases 1 5, 523-7, 2002

Chappell CL, Okhuysen PC, Sterl ing CR, Wang C, Jakubowski W, Dupont HL. I nfectivity of

Cryptosporidium parvum i n healthy adults with pre-existing anti-C. parvum serum immunoglobul in

American Journal of tropical Medicine and Hygiene 60, 1 57-64, 1 999

Coklin T, Farber J , Parrington L, Dixon B. Prevalence and molecular characterization of Giardia

duodena/is and Cryptosporidium spp. in dairy cattle in Ontario, Canada. Veterinary Parasitology 1 50 , 297-

305, 2007

34

Cole DJ, Cohen NO, Snowden K, Smith R. Prevalence of and risk factors for fecal shedding of

Cryptosporidium parvum oocysts in horses. Journal of the American Veterinary Medical Association 21 3,

1 296-302, 1 998

Colema n SU, Klei TR, French DD, C hapman MR, Costvet RE. Prevalence of Cryptosporidium sp in

equ ids in Louisiana. American Journal of Veterinary Research 50, 575-77, 1 989

Coutinho BP, Oria RB, Vieira CM, Sevilleja JE, Warren CA, Maciel JG, Thompson MR, Pinkerton RC,

Lima AA, Guerrant RL. Cryptosporidium infection causes under-nutrit ion , and conversely, weanling

under-nutrition intensifies infect ion. Journal of Parasitology 94 , 1 225-32, 2008

Current WL. The biology of Cryptosporidium. American Society for Microbiology News 54, 605- 1 1 , 1 988

Current WL, Haynes TB. Complete development of Cryptosporidium in cel l cul ture. Science 224, 603-05,

1 984

Current WL, Reese NC. A comparison of endogenous development of three isolates of Cryptosporidium

i n suckling mice. Journal of Protozoology 33, 98- 1 08, 1 986

Current WL, Reese NC, Ernst JV, Bai ley WS, Heyman MB, Weinstein WM. Human cryptosporidiosis in

immunocompetent and immunodeficient persons. New England Journal of Medicine 308, 1 252-57, 1 983

De Graaf DC, Vanopdenbosch E, Ortega-Mora LM, Abbassi H, Peeters JE . A review of the importance

of cryptosporidiosis in farm animals. International Journal for Parasitology 29, 1 269-87, 1 999

Dil l ingham RA, L ima AA, Guerrant RL. Cryptosporidiosis : epidemiology and impact. Microbes and

Infection 4, 59-66, 2002

DuPont HL, Chappel C L, Sterl ing CR, Okhuysen PC, Rose JB, Jakubowsky W. The infectivity of

Cryptosporidium parvum in healthy vol unteers. New England Journal of Medicine 332, 855-59, 1 995

Ebeid M, Mathis A, Pospischi l A, Deplazes P. Infectivity of Cryptosporidium parvum genotype in

conventionally reared piglets and lambs. Parasitology Research 90, 232-35, 2003

Egger M, Nguyen XM, Schaad UB, Krech T. I ntestinal cryptosporidiosis acqu ired from a cat. Infection 1 8,

1 77-8, 1 990

Elwin K, Chalmers RM. Contemporary identification of previously reported novel Cryptosporidium isolates

reveals Cryptosporidium bovis and the cervine genotype in sheep ( Ovis aries). Parasitology Research 1 02,

1 1 03- 1 1 05, 2008

Esteban E, Anderson BC. Cryptosporidium muris: prevalence, persistency and detrimental effects on mi lk

product ion in a dry lot dairy. Journal of Dairy Science 78, 1 068-72, 1 995

Fayer R. Cryptosporidium: a water-borne zoonotic parasite. Veterinary Parasitology 1 26, 37-56, 2004

35

Fayer R. Biology. I n : Fayer, R . , Xiao, L. (Eds. ) , Cryptosporidium and cryptosporidiosis, second ed. CRC

Press and lW A Publ ishing, Boca Raton, FL , pp . 1 -42, 2008

Fayer R, Ellis W. Paromomycin is effective as prophylaxis for cryptosporidiosis in dairy calves. Journal of

Parasitology 79, 771 -4, 1 993

Fayer R, Fischer JR, Sewel l CT, Kavanaugh OM, Osborn DA. Spontaneous cryptosporidiosis in captive

white-tailed deer (Odocoi leus virginianus) . Journal of Wildlife Diseases 32, 61 9-22, 1 996

Fayer R, Gasbarre L, Pasquali P, Canals A, Almeria S, Zarlenga D. Cryptosporidium parvum infection

in bovine neonates: dynamic cl inical , parasitic and immunologic patterns. International Journal for

Parasitology 28 :49-56, 1 998

Fayer R, Santin M, Trout JM. Prevalence of Cryptosporidium species and genotypes in mature dairy

cattle on farms in eastern Un ited States compared with younger cattle from the same locations. Veterinary

Parasitology 1 45, 260-66, 2007

Fayer R, Santin M, Trout JM. Cryptosporidium ryanae n. sp. (Apicomplexa: Cryptosporidiidae) in cattle

(8os taurus). Veterinary Parasitology 1 56, 1 9 1 -8 . 2008

Fayer R, Santin M, Trout JM, G reiner E. Prevalence of species and genotypes of Cryptosporidium found

in 1 -2-year-old dairy cattle in the eastern United States. Veterinary Parasitology 1 35, 1 05- 1 2, 2006

Fayer AM, Santin M, Xiao L. Cryptosporidium bovis n. sp. (Apicomplexa: Cryptosporidiidae) in catt le (8os

taurus) . Journal of Parasitology 9 1 , 624-29, 2005

Fayer R, Trout JM, Xiao L , Morgan UM, Lal AA*, Dubey JP. Cryptosporidium canis N.Sp. from domestic

dogs. Journal of Parasitology 87, 1 41 5-22, 200 1

Feltus DC, Giddings CW, Khaitsa ML, McEvoy JM. High prevalence of Cryptosporidium bovis and the

deer- l ike genotype in calves compared to mature cows in beef cow-calf operations. Veterinary

Parasitology 1 5 1 , 1 9 1 -95, 2008

Feng H, Nie W, Sheoran A, Z hang Q, Tzipori S. Bile acids enhance invasiveness of Cryptosporidium

spp. into cultured cel ls. Infection and Immunity 74, 3342-46, 2006

Feng Y, Ortega Y, He G, Das P, Xu M, Zhang X, Fayer R, Gatei W, Cama V, Xiao L. Wide geographic

distribution of Cryptosporidium bovis and the deer- l ike genotype in bovines. Veterinary Parasitology 1 44,

1 -9 , 2007

Feng X, Rich SM, Akiyoshi D, Tumwine JK, Kekitiinwa A, Nabukeera N, Tzipori S, Widmer G.

Extensive polymorphism in Cryptosporidium parvum identified by mult i locus m icrosatel l i te analysis.

Applied and Environmental Microbiology 66, 3344-49, 2000

36

Gajadhan A, Caron JP, Alien JR. Cryptosporidium in two foals. Canadian Veterinary Journal 26, 1 32-34,

1 985

Garber LP, Salman MD, Hurd HS, Keefe T, Schlater J. Potential risk factors for Cryptosporidium

infection in dairy calves. Journal of the American Veterinary Medical Association 205, 86-9 1 , 1 994

Garcia LS, Bruckner DA, Brewer TC, S himizu RV. Techniques for the recovery and identification of

Cryptosporidium oocysts from stool specimens. Journal of Clinical Microbiology 1 8, 1 85-90, 1 983

Gatei W, Barrett 0, Lindo JF, Cama V, Xiao L. Extensive homozygosity in Cryptosporidium parasites

identified by multilocus sequence typing in Jamaica. Proceedings of the 1 1 International Conference on

Giardia and Cryptosporidium, Morelia, Michoacan, Mexico, May 1 3 - 1 8, Molecular Epidemiology and

Typing, Session 2, Abstract 1 1 , 2007a

Gatei W, Das P, Dutta P, Sen A, Cama V, Lal AA, Xiao L. Multi locus sequence typing and genetic

structure of Cryptosporidium hominis from children in Kolkata, India. Infection Genetics and Evolution 7,

1 97-205, 2007b

Genchi C, Hermon J, Sangali G, Traldi G. La criptosporidiosi nel vitel lo, fattore determinante nella

diarrea neonatale. (Calf cryptosporid iosis, a determinant factor of neonatal diarrhoea). Praxis Veterinaria

3, 4, 1 984

Genta RM, Chappel l C L, White AC, Kimball KT, Goodgame RW. Duodenal morphology and intensity of

infection in AI DS-related intestinal cryptosporidiosis. Gastroenterology 1 05, 1 769-75, 1 993

Geurden T, Goma FY, Siwila J , Ph iri IGK, Mwanza AM, Gabriel S, Claerebout E, Vercruysse J.

Prevalence and genotyping of Cryptosporidium in three cattle husbandry systems in Zambia. Veterinary

Parasitology 1 38, 2 1 7-22, 2006

Geurden T, Thomas P, Casaert S, Vercruysse J, Claerebout E. Prevalence and molecular

characterisation of Cryptosporidium and Giardia in lambs and goat kids in Belgium. Veterinary

Parasitology 1 55, 1 42-5, 2008

G iangaspero A, lorio R, Paoletti 8, Traversa 0, Cappel l i G. Molecular evidence for Cryptosporidium

infection in dogs in Central Italy. Parasitology Research 99, 297-99, 2006

Gibson JA, Hi l l MWM, Huber MJ. Cryptosporidium in Arabian foals with severe combined

immunodeficiency. Australian Veterinary Journal 60, 378-9, 1 983

G iles M, Chalmers R, Pritchard G , Elwin K, Mueller-Doblies 0, Clifton-Hadley F. Cryptosporidium

hominis in a goat and a sheep in the UK. Veterinary Record 1 64 , 24-5, 2009

G iles M, Webster KA, Marshal! JA, Catchpole J, Goddard TM. Experimental infection of a lamb with

Cryptosporidium parvum genotype 1 . Veterinary Record 1 49, 523-25, 2001

37

Griffiths JK, Balakrishnan R, Widmer G, Tzipori S. Paromomycin and geneticin inhibit intracellular

Cryptopsporidium parvum without trafficking through the host-cell cytoplasm : impl ications for d rug del ivery.

Infection and Immunity 66, 387 4-83, 1 998

Grinberg A, Markovics A, Gal indez J, Lopez-Vil lalobos N, Kosak A, Tranquil lo. VM. Contro l l ing the

onset of natural cryptosporidiosis in calves with paromomycin sulphate. Veterinary Record 1 5 1 , 606-8,

2002

Guselle NJ, Appelbee AJ , Olson ME. Biology of Cryptosporidium parvum in p igs : from weaning to

market. Veterinary Parasitology 1 1 3 , 7- 1 8, 2003

Guyot K, Fol let-Dumoulin A, Lel ievre E, Sarfati C, Rabodonirina M, Nevez G, Cai l l iez JC, Camus D,

Dei-Cas E. Molecular characterization of Cryptosporidium isolates obtained from humans in France.

Journal of Clinical Microbiology 39, 3472-80, 200 1

Hajdusek 0, Ditrich 0, Slapeta J. Molecular identification of Cryptosporidium spp. i n animal and human

hosts from the Czech Republic. Veterinary Parasitology 1 22, 1 83- 1 92, 2004

Harp JA, Woodmansee DB, Moon HW. Effect of colostral antibody on susceptibility of calves to

Cryptosporidium parvum in fection. American Journal of Veterinary Research 50, 2 1 1 7- 9, 1 989

Hashimoto A, Sugimoto H , Sh igemitsu M, Tsuyoshi H . Genotyping of single Cryptosporidium oocysts

in sewage by semi-nested PCR and di rect sequencing. Water Research 40, 2527-32, 2006

Henriksen SA, Pohlenz J. Staining of cryptosporidia by a modified Ziehi- Neelsen technique. Acta

Veterina Scandinavica 22:594-5, 1 98 1

Hijjawi N S , Meloni BP, Ng'anzo M, Ryan UM, O lson M E , Cox PT, Monis PT, Thompson R C . Complete

development of Cryptosporidium parvum in host cell-free culture. International Journal for Parasitology 34,

769-77, 2004

Hijjawi NS, Meloni BP, Ryan UM, Olson ME, Thompson RC. Successful in vitro cultivation of

Cryptosporidium andersoni: evidence for the existence of novel extracellular stages in the l i fe cycle and

implications for the classificat ion of Cryptosporidium. International Journal for Parasitology, 32 : 1 7 1 9-25,

2002

Huang DB, Chappell C, Okhuysen PC. Cryptosporidiosis in chi ldren . Seminars in pediatric infectious

diseases 1 5, 253-59, 2004

Hunter PR, Nichols G. Epidemiology and cl in ical features of Cryptosporidium i nfection in

immunocompromised patients. Clinical Microbiology Reviews 1 5, 1 45-54, 2002

Hunter PR, Thompson RCA. The zoonotic transmission of Giardia and Cryptosporidium. International

Journal for Parasitology 35, 1 1 8 1 -90, 2005

38

Jeffreys AJ , Wi lson V, Thein SL. Hypervariable 'minisatellite' regions in human DNA. Nature 3 1 4, 67-73,

1 985

Joachim A, Krull T, Schwarzkopf J , Daugschies A. Prevalence and control of cryptosporidiosis in

German dairy herds. Veterinary Parasitology 1 1 2, 277-88, 2003

Johnson J, Buddle R, Reid S, Armson A, Ryan UM. Prevalence of Cryptosporidium genotypes in pre

and post-weaned pigs in Australia. Experimantal Parasitology 1 1 9 , 4 1 8-21 , 2008

Khan WA, Rogers KA, Karim MM, Ahmed S, Hibberd PL, Calderwood SB, Ryan ET, Ward HO.

Cryptosporidiosis among Nangaladeshi chi ldren with diarrhoea: a prospective, matched case-control study

of c linical features, epidemiology and systemic antibody responses. American Journal of Tropical Medicine

and Hygiene 7 1 , 4 1 2-1 9, 2004

Khramtsov NV, Ti l ley MDS, B lunt BA, Montelone BA, Upton, SJ . Cloning and analysis of a

Cryptosporidium parvum gene encoding a protein with homology to cytoplasmic form HSP70. Journal of

Eukaryotic Microbiology 42, 41 6-22, 1 995

Kim CW. Cryptosporidiosis in pigs and horses. In : Cryptosporidiosis of man and animals 1 05-1 1 . J P

Dubey C A Speer and R Fayer (eds). C R C Press inc . , Boca Raton, Florida, USA, 1 990

Kvac M, Kouba M, V1tovec J. Age-related and housing-dependence of Cryptosporidium infection of

calves from dairy and beef herds in South Bohemia, Czech Republ ic. Veterinary Parasitology 1 37 , 202-

09, 2006

Kvac M, Sak 8, Hanzlikova D, Kotilova J, KviHonova D. Molecular characterization of Cryptosporidium

isolates from pigs at slaughterhouses in South Bohemia, Czech Republic. Parasitology Research

1 04:425-28, 2009

Kvac M, Sak B, Kvetonova D, Ditricha 0, Hofmannovad L, Modrya D, Vitovec J, Xiao L. Infectivity,

pathogenicity, and genetic characteristics of mammalian gastric Cryptosporidium spp. in domestic

ruminants. Veterinary Parasitology 1 53, 363-67, 2008

Laurent F, McCole D, Eckmann L , Kagnoff MF. Pathogenesis of Cryptosporidium parvum infection.

Microbes and Infection 1 , 1 4 1 -8, 1 999

Laxer MA, Timbl i n BK, Patel RJ. DNA sequences for the specific detect ion of Cryptosporidium parvum

by the polymerase chain reaction. American Journal of Tropical Medicine and Hygiene 45, 688-94, 1 99 1

Learmonth JJ, lonas G, Ebbet KA, Kwan ES. Genetic characterisation and transmission cycles of

Cryptosporidium species isolated from humans in New Zealand. Applied and Environmental Microbiology

70, 3973-8, 2004

39

Learmonth J, lonas G , Pita A, Cowie R. Identification and genetic characterisation of Giardia and

Cryptosporidium strains i n humans and dairy cattle in the Waikato Region of New Zealand. Water Science

and Technology 4 7, 2 1 -6, 2003

Learmonth J, lonas G, Pita A, Cowie R. Seasonal shift in Cryptosporidium parvum transmission cycles in

New Zealand. Journal of Eurkaryotic Microbiology 48, 34s-35s, 2005

Leav BA, Mackay MR, Anyanwu A, RM OC, Cevallos AM, Kindra G, Rol l ins NC, Bennish Ml, N elson

RG, Ward HO. Analysis of sequence d iversity at the h ighly polymorphic Cpgp40/1 5 locus among

Cryptosporidium isolates from human immunodeficiency virus-infected chi ldren in South Africa. Infection

and Immunity 70,3881 -90, 2002

Leng X, Mosier DA, Oberst RD. Differentiation of Cryptosporidium parvum, C. muris, and C. baileyi by

PCR-R FLP analysis of the 1 88 rRNA gene. Veterinary Parasitology 62, 1-7, 1 996

Lindsay OS, Upton SJ, Owens DS, Morgan UM, Mead JR, Blagburn BL. Cryptosporidium andersoni n.

sp. (Apicomplexa: Cryptosporiidae) from cattle , Bos taurus. Journal of Eukaryotic Microbiology 47, 91 -5,

2000

MacCuish J, Nicolaou C, MacCuish NE. Ties in proximity and clustering compounds. Journal of

Chemical Information and Modeling 4 1 , 1 34 -46, 2001

Mac Kenzie WR, Hoxie NJ, Proctor ME, Gradus MS, Blair KA, Peterson D E, . Kazmierczak JJ,

Addiss DG, Fox KR, Rose JB, Davis JP. A massive outbreak in Milwaukee of Cryptosporidium i n fection

transmitted through the public water supply. New England Journal of Medicine 331 , 1 6 1 -7 ; 1 994

Maggi P, Larocca AMV, Quarto M, Sergio G, Brandonisio 0, Angarano G, Pastore G. Effect of

antiretroviral therapy on cryptosporidiosis and microsporidiosis in patients infected with human

immunodeficiency virus type 1 . European Journal of Clinical Microbiology and Infectious Diseases 1 9 :21 3-

1 7, 2000

Mclaughlin J , Amar C, Pedraza-Diaz S, Nichols GL . Molecular epidemiological analysis of

Cryptosporidium spp. in the United K ingdom: Results of genotyping Cryptosporidium spp. in 1 705 fecal

samples from humans and 1 05 fecal samples from livestock animals. Journal of Clinical Microbiology 38,

3984-90, 2000

Mallon M, Macleod A, Wastling J, Smith H, Tait A. Multi locus genotyping of Cryptosporidium parvum

Type 2 : population genetics and sub-structuring. Infection, Genetics and Evolution 3, 207- 1 8, 2003

Mason, RW, Hartley WJ, Tilt L. I ntestinal cryptosporidiosis in a kid goat. Australian Veterinary Journal 57,

386-88, 1 981

Meisel JL, Perera OR, Meligro C, Rubin CE. Overwhelming watery diarrhea associated with a

Cryptosporidium in an immunosuppressed patient. Gastroenterology 70: 1 1 56-60, 1 976

40

Miao YM, Awad-ei-Kariem FM, Franzen C, El l is OS, Muller A, Counihan HM, Hayes PJ, Gazzard BG.

Eradication of cryptosporidia and microsporidia fol lowing successful antiretroviral therapy. Journal of

Acquired Immunodeficiency Virus 25 : 1 24-29, 2000

Mil lard PS, Gensheimer KF, Addiss DG, Sosin OM, Beckett GA, Houck-Jankoski A, Hudson A. An

outbreak of cryptosporidiosis from fresh-pressed apple cider. Journal of the American Medical Association

272, 1 592-96, 1 994

Misic Z, Abe A. Subtype analysis of Cryptosporidium parvum isolates from calves on farms around

Belgrade, Serbia and Montenegro, using the 60 k Da glycoprotein gene sequences. Parasitology 1 -8 , 2006

Mohammed HO, Wade SE, Schaaf S. Risk factors associated with Cryptosporidium parvum infection in

dairy cattle in southeastern New York State. Veterinary Parasitology 83, 1 -1 3, 1 999

Moore DA, Atwi l l ER, Kirk JH, Brahmbhatt D, Alonso LH, Hou LL, Singer MD, Mi l ler TO. Prophylactic

use of decoquinate for infections with Cryptosporidium parvum in experimentally challenged neonatal

calves. Journal of the American Veterinary Medical Association 223, 839-45 2003

Moore DA, Zeman OH. Cryptosporidiosis in neonatal calves: 277 cases ( 1 986- 1 987). Journal of the

American Veterinary Medical Association 1 98, 1 969-71 , 1 991

Morgan UM, Sargent KO, Deplazes P, Forbes DA, Spano F, Hertzberg H, El l iot A, Thompson RCA.

Molecular characterisation of Cryptosporidium from various hosts. Parasitology 1 1 7 :3 1 -7 , 1 998

Morgan UM, Xiao L, Monis P, Fal l A, lrwin PJ, Fayer R, Denholm KM, Limor J, Lal AA, Thompson

RCA. Cryptosporidium spp. in domestic dogs: the "Dog" genotype. Applied and environmental

microbiology 66, 2220-23, 2000

Morgan-Ryan UM, Fall A, Ward LA, Hijjawi N, Sulaiman I, Fayer R, Thompson RCA, Olson M, Lal A,

Xiao, L. Cryptosporidium hominis n. sp. (Apicomplexa: Cryptosporid i idae) from Homo sapiens. Journal of

Eukariotic Microbiology 49, 433-40, 2002

Morin M, Lariviere S, Lalier R. Pathological and m icrobiological observations on spontaneous cases of

acute neonatal calf diarrhoea. Canadian Journal of Comparative Medicine 40, 228, 1 976

Mueller-Doblies D, Gi les M, Elwin K, Smith RP, Cl ifton-Hadley FA, Chalmers RM. Distribution of

Cryptosporidium species in sheep in the UK. Veterinary Parasitology 54, 2 1 4-9, 2008

Naciri M, Mancassola R, Yvore P, Peeters JE. The effect of halofuginone lactate on experimental

Cryptosporidium parvum infections in calves. Veterinary Parasitology 45, 1 99-207, 1 993

Naciri M, Lefay MP, Mancassola R, Poirier P , Chermette R. Role of Cryptosporidium parvum as a

pathogen in n eonatal diarrhoea complex in suckl ing and dairy calves in France. Veterinary Parasitology

85, 245-57, 1 999

41

Netherwood T, Wood JLN, Townsend HGG, Mumford JA. Foal diarrhoea between 1 99 1 and 1 994 in

the Un ited Kingdom associated with Clostridium perfringens, rotavirus, Strongyloides westeri and

Cryptosporidium spp. Epidemiology and Infection 1 1 7, 375-83, 1 994

N ichols G. I n : Fayer, R . , Xiao, L. (Eds . ) , Cryptosporidium and cryptosporidiosis, second ed. CRC Press

and lW A Publishing, Boca Raton , FL, pp. 79- 1 1 8, 2008

N ime FA, Burek JD, Page DL. Acute enterocolitis in a human being infected with the protozoan

Cryptosporidium. Gastroenterology 70, 592-8, 1 976

Okhuysen PC, Chappell CL, Crabb JH, Sterling CR, DuPont HL. Virulence of three d istinct

Cryptosporidium parvum isolates for healthy adults. Journal of Infectious Diseases 1 80, 1 275-8 1 , 1 999

Olson ME, O'Handley RM, Ralston BJ, McAIIister TA, Thompson RCA. Update on Cryptosporidium

and Giardia i nfections in cattle. Trends in Parasitology 20, 1 85- 1 9 1 , 2004

Olson ME, Thorlakson CL, Deselliers L, Morck DW, McAI I ister TA. Giardia and Cryptosporidium i n

Canadian farm animals. Veterinary Parasitology 68, 375-8 1 , 1 997

Orr M B, MacKintosh CC, Sutie JM. Cryptosporidiosis in deer calves. New Zealand Veterinary Journal 33:

1 5 1 - 1 53, 1 985

Ortega YR, Sheehy RR, Cama VA. Restriction-fragment length polymorphism analysis of

Cryptosporidium parvum isolates of bovine and human origin. Journal of Protozoology 38, S40-S41 , 1 991

Ortega-Mora LM, Wright SE . Age-related resistance in ovine cryptosporidiosis: patterns o f infection and

humoral immune response. Infection and Immunity 62, 5003-09, 1 994

O'Donoghue PJ. Cryptosporidium and cryptosporidiosis in man and animals. International Journal of

Parasitology 25, 1 39-95, 1 995

O'Handley RM, Cockwil l C, McAIIister TA, Jelinski M, Morck DW, Olson ME. Durat ion of naturally

acquired giardiasis and cryptosporidiosis in dairy calves and their association with diarrhea. Journal of the

American Veterinary Medical Asscociation 21 4, 39 1 -6, 1 999

Palmer CS, Traub RJ, Robertson ID, Devlin G, Rees R, Thompson RC. Determining the zoonotic

signif icance of Giardia and Cryptosporidium i n Australian dogs and cats. Veterinary Parasitology 1 54 , 1 42-

7, 2008

Panciera RJ, Thomassen RW, Garner FM. Cryptosporidial infection in a calf. Veterinary Pathology 8,

479-84, 1 971

Paoletti B, Giangaspero A, Gatti A, lorio R, Cembalo D , Mi l i l lo P , Traversa D. lmmunoenzymatic

analysis and genetic detection of Cryptosporidium parvum in lambs from Italy. Experimantal Parasitology

in p ress, doi : 1 0. 1 0 1 6/j .exppara.2009.05.006, 2009

42

Paul S, Chandra D, Ray DD, Tewari AK, Rao JR, Banerjee PS, Baidya S, Raina OK. Prevalence and

molecular characterization of bovine Cryptosporidium isolates in I ndia. Veterinary Parasitology 1 53 , 1 43-6,

2008

Pedraza-Diaz S, Ortega-Mora LM, Carrion BA, Navarro V, G6mez-Bautista M. Molecu lar

characterisation of Cryptosporidium isolates from pet repti les. Veterinary Parasitology 1 60, 204- 1 0, 2009

Peng X, Murphy T, Holden NM. Evaluation of temperature on d ie-off 1 rate of Cryptosporidium parvum

oocysts in water, soi ls and faeces. Applied and Environmental Microbiology, Published on l ine,

doi: 1 0. 1 1 28/AEM.01 442-08, 2008

Peng MM, Xiao L, Freeman AR, Arrowood MJ, Escalante AA, Weltman AC, Ong CSL, Mac Kenzie

WR, Lal AA, Beard CB. Genetic polymorph ism among Cryptosporidium parvum isolates: evidence of two

distinct h uman transmission cycles. Emerging Infectious Diseases 3, 567-3, 1 997

Pereira SJ . Pathogenesis of h uman and bovine Cryptosporidium parvum in gnotobiotic pigs.

Journal of Infectious Diseases 1 86 , 71 5-8, 2002

Perryman LE, Bjorneby JM. I mmunotherapy of cryptosporidiosis in immunodeficient animal models.

Journal of Protozoology 38, 98S-1 OOS, 1 99 1

Pevsner J . Bioinformatics and functional genomics Pp. 543-50. John Wi ley and Sons Inc. publ ications,

Hobokem New Jersey, USA, 2003

Pohlenz J, Moon HW, Chevi l le NF, Bermick WJ. Cryptosporidiosis as a probable factor in neonatal

diarrhoea of calves. Journal of the American Veterinary Medical Association 1 72, 452, 1 978

Pohjola S, Jokipii AM, Jokipii L. Sporadic cryptosporidiosis in a rural population is asymptomatic and

associated with contact to catt le. Acta Veterinaria Scandinavica 27, 9 1 -1 02, 1 986

Pozio E, Morales MAG. The impact of H IV-protease inhibitors on opportunistic parasites. Trends in

Parasitology 21 , 58-63, 2005

Pritchard GC, Marshal! JA, Gi les M, Mueller-Doblies D, Sayers AR, Marshal! R N , Elwin K, Chalmers

RM. Cryptosporidium species in lambs submitted for diagnostic postmortem examination in England and

Wales. Veterinary Record 1 63, 688-9, 2008

Quigley JD, Martin KR, Semis DA, Potgieter LND, Reinemeyer CR, Rorhbach BW, Dowlen HH,

Lamar KC. Effects of housing and colostrum feeding on the prevalence of selected infectious organisms in

feces of Jersey calves. Journal of Dairy Science 77, 31 24-31 , 1 994

Quilez J, Torres E, Chalmers RM, Hadfield SJ, Del Cacho E, Sanchez-Acedo C. Genotype and

subtype characterisation of Cryptosporidium in lambs and goat kids in Spain. Applied and Environmental

Microbiology, published ahead of print , doi : 1 0. 1 1 28/AEM.00606-08, 2008

43

Quilez J, Sanches-Acedo C, Clavel A, dei-Cacho E, Lopez-Bernal F. Prevalence of Cryptosporidium

infection in pigs i n Aragon (northestern Spain) . Veterinary Parasitology 67, 83-88, 1 996

Raccurt CP. Worldwide human zoonotic cryptosporidiosis caused by Cryptosporidium felis. (in French,

abstract avai lable in English at http ://www.ncbi .n lm.n ih .gov/pubmed/1 7432054). Parasite 1 4, 1 5-20, 2007

Ranucci L, Mu ller HM, La Rosa G, Reckmann I , Morales MA, Spano F, Pozio E, Crisanti A.

Characterization and immunolocalization of a Cryptosporidium protein containing repeated amino acid

motifs. Infection and Immunity 6 1 , 234 7-56, 1 993

Rid ley RK, Olsen RM. Rapid diagnosis of bovine cryptosporidiosis with a modified commercial

acid-fast staining procedure. Journal of Veterinary Diagnostic Investigation 3, 1 82-3, 1 991

Reif JS , Wimmer L , Smith JA, Dargatz DA, Cheney JM. Human cryptosporidiosis associated with an

epizootic in calves. American Journal of Public Health 79, 1 528-30, 1 989

Riggs MW, Stone AI, Vaunt PA, Langer RC, Arrowood MJ, Bentley DL. Protective monoclonal

antibody defines a ci rcumsporozoite-like glycoprotein exoantigen of Cryptosporidium parvum sporozoites

and merozoites. Journal of immunology 1 58, 1 787-95, 1 997

Riley LW. I n : Molecu lar Epidemiology of I nfectious Diseases, pp. 49. ASM Press, Wash ington DC, USA,

2004

Rochelle PA, De Lean R, Atwi l l ER. Intra-isolate heterogeneity and reproducibility of PC R-based

genotyping of Cryptosporidium parvum using the -tubul in gene. Quantitative Microbiology 2, 87- 1 0 1 , 2000

Ryan UM, Monis P , Enemark H L, Sulaiman I , Samarasinghe 8, Read C, Buddle R, Robertson I ,

Zhou L, Thompson RCA, Xiao L . Cryptosporidium suis N . Sp. (Apicomplexa:Cryptosporidi idae) in pigs

(Sus Scrota) . Journal of Parasitology 90, 769-73, 2004

Ryan UM, Power M, Xiao L. Cryptosporidium fayeri n. sp. (Apicomplexa: Cryptosporidiidae) from the red

kangaroo (Macropus rufus) . Journal of Eukaryotic Microbiology 55, 22-6, 2008

Ryan U, Xiao L, Read C, Zhou L, Lal AA, Pavleski I. Identification of novel Cryptosporidium genotypes

from the Czech Republic. Applied and Environmental Microbiology 69, 4302-07, 2003

Sanford SA, Josephson GKA. Bovine cryptosporidiosis: c l inical and pathological findings in forty-two

scouring neonatal calves. Canadian Veterinary Journal 23, 340-3, 1 982

Santm M, Fayer R. l ntragenotypic variations in the Cryptosporidium sp. cervine genotype from sheep with

impl ications for public health. Journal of Parasitology 93, 668-72, 2007

Santin M, Trout JM. Livestock. In: I n : Fayer, R . , Xiao, L. (Eds. } , Cryptosporidium and cryptosporidiosis,

second ed. CRC Press and lW A Publ ishing, Boca Ratan , FL, pp. 451 -83, 2008

44

Santin M, Trout JM. Companion Animals. In : In : Fayer, R . , Xiao, L. ( Eds.} , Cryptosporidium and

cryptosporidiosis, second ed. C RC Press and lW A Publ ishing, Boca Raton, FL, pp. 437·49, 2008

Santin M, Trout JM, Fayer R. A longitudinal study of cryptosporidiosis in dairy cattle from birth to 2 years

of age. Veterinary Parasitology 1 55 , 1 5-23, 2008

Santin M, Trout JM, Vecino JAC, Dubey J P, Fayer R. Cryptosporidium, Giardia and Enterocytozoon

bieneusi in cats from Bogota (Colombia) and genotyping of isolates. Veterinary Parasitology 1 4 1 , 334-9,

2006

Santin M, Trout JM, Xiao L, Zhoub OL, G reiner E, Fayer R. Prevalence and age-related variation of

Cryptosporidium species and genotypes in dai ry calves. Veterinary Parasitology 1 22 , 1 03-1 7 , 2004

Sargent KD, Morgan UM, El l iot A, Thompson RCA. Morphological and genetic characterisation of

Cryptosporidium oocysts f rom domestic cats. Veterinary Parasitology 77, 221 -7, 1 998

Satcher D. Getting ahead of the curve. Emerging Infectious Diseases 1 , 1 -6, 1 995

Schultz, MG. Emerging zoonoses. New England Journal of Medicine 308, 1 285-86, 1 983

Sevinc F, lrmak K, Sevinc M . The prevalence of Cryptosporidium parvum infection in diarrhoeic and non­

diarrhoeic calves. Revue de Medecine Veterinaire 1 54, 357-6 1 , 2003

Sheoran AS, Jeong K, Wiffin A, Chapman-Bonofiglio, Tzipori S. Cryptosporidium hominis-specific

immunity provides incomplete protection against C. parvum in the gnotobiotic piglet model . International

Workshop on Opportunistic P rotists ( IWOP- 1 0}, May 28-3 1 , 2008, Boston MA, USA (Abstract P063}

Siwila J, Phiri IG, Vercruysse J, Goma F, Gabriel S, C laerebout E, Geurden T. Asymptomatic

cryptosporidiosis in Zambian dairy farm workers and their household members. Transactions of the Royal

Society of Tropical Medicine and Hygiene 1 01 , 733-4, 2007

Skerrett HE, Holland CV. Asymptomatic shedding of Cryptosporidium oocysts by red deer hinds and

calves. Veterinary Parasitology 94, 239-46, 2001

Smerdon WJ, Nichols T, C halmers RM, Heine H, Reacher MH. Foot and mouth d isease in l i vestock

and reduced cryptosporidiosis in humans, England and Wales. Emerging Infectious Diseases 9 , 22- 8,

2003

Smith HV, Nichols RAB, G rimason AM. Cryptosporidium exystation and invasion: getting to the guts of

the matter. Trends in Parasitology 2 1 , 1 33-42, 2005a

Smith HV, Nichols RAB, Mallon M, Macleod A, Tait A, Rei l ly WJ, Browning LM, Gray D, Reid SWJ,

Wastl ing JM. Natural Cryptosporidium hominis infections in Scottish cattle. Veterinary Record 1 56, 7 1 0-

1 1 , 2005b

45

Snell ing WJ, Xiao L, Ortega-Pierres G , Lowery CJ, Moore JE, Rao JR, Smyth S, Mi llar BC, Rooney

PJ, Matsuda M, Kenny F, Xu J, Dooley JSG. Cryptosporidiosis in developing countries. Journal of

Infection in Developing Countries 1 , 242-56, 2007

Snyder SP, England JJ, McChesney AE. Cryptosporidium in immunodeficient Arabian foals. Veterinary

Pathology 1 5, 1 2-7, 1 978

Sopwith W, Osborn K, Chalmers R, Regan M. The changing epidemiology of cryptosporidiosis in North

West England. Epidemiology and Infection 1 33, 785-93, 2005

Spano F, Puri C, Ranucci L, Putignani LA, Crisanti. Cloning of the entire COW P gene of

Cryptosporidium parvum and ultrastructural localization of the protein during sexual parasite development.

Parasitology 1 1 4 , 427-37, 1 997

Spano F, Putignani LA, Crisanti A, Sall icandro P, Morgan U , Le Blancq SM, Tchack L, Tzipori S.,

Widmer G. Multilocus genotypic analysis of Cryptosporidium parvum isolates from different hosts and

geographical origins. Journal of Clinical Microbiology 36, 3255-59, 1 998

Starkey SR, Zeigler PE, Wade SE. Factors associated with shedding of Cryptosporidium parvum versus

Cryptosporidium bovis among dairy cattle in New York State. Journal of the American Veterinary Medical

Association 229, 1 623-6. 2006

Stefanogiannis N , McLean M, Van Mi l H. Outbreak of cryptosporidiosis l inked with a farm event. New

Zealand Medical Journal 1 1 4, 51 9-21 , 2001

Steher-Green J, McCaig l, Remsen HM, Rains CS, Fox M, Juranek DD. Shedding of oocysts in

immunocompetent individuals infected with Cryptosporidium. American Journal of Tropical Medicine and

Hygiene 338-42, 1 987

Stewart CG, Penzhorrn Bl . Cryptosporidiosis. in : I nfectious Diseases of Livestock, 2nd edition. Pp. 334.

Authors: JAW Coetzer and RC Tustin. Oxford University press, Oxford UK. 2004

Strong WB, Gut J , Nelson RG. C loning and sequence analysis of a highly polymorphic Cryptosporidium

parvum gene encoding a 60-kilodalton glycoprotein and characterization of its 1 5- and 45-kilodalton zoite

surface antigen products. Infection and Immunity 68, 4 1 1 7-34, 2000

Sulaiman IM, Hira PR, Zhou L, AI-Aii FM, AI-Shelahi F A, Shweiki HM, lqbal J, Khal id N, Xiao L.

Unique endemicity of cryptosporid iosis in chi ldren in Kuwait. Journal of Clinical Microbiology 43, 2805-09,

2005

Sulaiman IM, Xiao L, Yang C, Escalante L, Moore A, Beard CB, Arrowood MJ, Lal AA. Differentiating

human from animal Isolates of Cryptosporidium parvum. Emerging Infectious Diseases 4, 681 -5, 1 998

Suzuki MT, G iovannoni SJ . Bias caused by template annealing in the amplification of m ixtures of 1 6S

rRNA Genes by PCR. Applied and Environmental Microbiology 62, 625-30, 1 996

46

Tanriverdi S, Arslan MO, Akiyoshi DE, Tzipori S, Widmer G . Identification of genotypical ly mixed

Cryptosporidium parvum populations in humans and calves. Molecular and Biochemical Parasitology 1 30,

1 3-22, 2003.

Tanriverdi S, Blain JC, Deng B, Ferdig MT, Widmer G. Genetic crosses in the apicomplexan parasite

Cryptosporidium parvum define recombination parameters. Molecular Microbiology 63, 1 432-9, 2007

Tanriverdi S, Markovics A, Arslan MO, ltik A, Shkap V, Widmer G. Emergence of distinct genotypes

of Cryptosporidium parvum in structured host populations. Applied and Environmental Microbiology 72:

2507- 1 3, 2006

Templeton TJ, Lancto CA, Vigdorovich V, Liu C, London NR, Hadsal l KZ, Abrahamsen MS. The

Cryptosporidium oocyst wall protein is a member of a multigene family and has a homolog in Toxoplasma.

Infection and Immunity 72, 980-7, 2004

Thamsborg SM, Jorgensen RJ, Henriksen SA. Cryptosporidiosis in kids of dairy goats. Veterinary

Record 1 27, 627-28 , 1 990

Theodos CM, Griffiths JK, D'onofrio J, Fairfield A, Tzipori S. Efficacy of nitazoxan ide against

Cryptosporidium parvum in cell culture and in animal models. Antimicrobial Agents and Chemotherapy 42,

1 959-65, 1 998

Tomley FM, Soldati DS. Mix and match modules: structure and function of microneme proteins in

apicomplexan parasites. Trends in Parasitology 1 7, 8 1 -8, 200 1

Townsend KG, Lance DM. Cryptosporidiosis in calves. New Zealand Veterinary Journal 35, 2 1 6-7, 1 987

Traversa D, lorio R, Otranto D, Modryc D, Slapeta J. Cryptosporidium from tortoises: Genetic

characterisation, phylogeny and zoonotic implications. Molecular and Cellular Probes 22, 1 22-8, 2008

Trotz-Wil liams LA, Jarvie BD, Martin SW, Leslie KE, Peregrine AS. Prevalence of Cryptosporidium

parvum infection in southwestern Ontario and its association with diarrhoea in neonatal dairy calves.

Canadian Journal of Veterinary Medicine 46, 349-51 , 2005

Trotz-Wi l liams LA, Martin DS, Gatei W, Cama V, Peregrine AS, Martin SW, Nydam DV, Jamieson F,

Xiao L. Genotype and subtype analyses of Cryptosporidium isolates from dai ry calves and humans in

Ontario. Parasitology Research 99, 346-52, 2006

Tumwine JK, Kekitiinwa A, Bakeera-Kitaka S, Ndeezi G, Downing R, Feng X, Akiyoshi DE, Tzipori

S. C ryptosporidiosis and microsporid iosis in Ugandan children with persistent diarrhea with and without

concurrent infection with the human immunodeficiency vi rus. American Journal of Tropical Medicine and

Hygiene 73, 921 -5, 2005

Tyzzer EE. A sporozoan found in the pepetic gland of the common mouse. Proceedings of the Society for

Experimental Biology 5, 1 2-3, 1 907

47

Tyzzer EE. An extracellular coccid ium, Cryptosporidium muris (ge n . et sp. nov. ) , of the gastric glands of

the common mouse. Journal of Medical Research 23, 487-509 , 1 9 1 0

Tyzzer EE. Cryptosporidium parvum (sp. nov . ) , a coccidium found i n the small intestine of the common

mouse. Arch. Protistenkd. 26, 394-4 1 2 , 1 9 1 2

Tzipori S. C ryptosporidiosis in animals and humans. Microbiological Reviews 47, 84-96, 1 983

Tzipori S. C ryptosporidiosis in perspective. Advances in Parasitology 27, 63- 1 29, 1 988

Tzipori S, Angus KW, Campbell I , Gray EW . Cryptosporidium: evidence of a single species genus.

Infection and Immunity 30, 884-86, 1 980a

Tzipori S, Angus KW, Campbell I , Sherwood D. Diarrhea in young red deer associated with infection

with Cryptosporidium. Journal of Infectious Diseases 1 44, 1 70-5, 1 98 1 a

Tzipori S, Campbell I. Prevale nce of Cryptosporidium antibodies i n 1 0 animal species. Journal of Clinical

Microbiology 1 4, 455-6, 1 98 1

Tzipori S , Campbell I , Sherwood D, Snodgrass OR, Whitelaw A . A n outbreak of calf d iarrhoea

attributed to cryptosporidial infection . Veterinary Record 1 07, 579-80, 1 980b

Tzipori S, G riffiths JK. Natural h istory and biology of Cryptosporidium parvum. Advances in Parasitology

40, 5-36, 1 998

Tzipori S, Larsen J , Smith M, Lue R. Diarrhea in goat kids attributed to Cryptosporidium i n fection.

Veterinary Record 1 1 1 , 35-6, 1 982

Tzipori S, McCartney E, Lawson GH, Rowland AC, Campbell I . Experimental infection of piglets with

Cryptosporidium. Research in Veterinary Sciences 3 1 , 358-361 , 1 98 1 b

Tzipori S, Smith M, Halpin C, Angus KW, Sherwood D, Campbell I. Experimental cryptosporidiosis i n

calves: cl i n ical manifestations a n d pathological findings. Veterinary Record 1 1 2 : 1 1 6-20, 1 983

Tzipori S, Ward H. Cryptosporid iosis : Biology, pathogenesis and disease. Microbes and Infection 4 ,

1 047-58, 2002

Tzipori S, Widmer G. A hundred-year retrospective on cryptosporidiosis. Trends in Parasitology 24, 1 84-

9 2008

Uga S, Matsuo J, Kono E, Kimura K, lnoue M, Rai SK, Ono K. Prevalence of Cryptosporidium parvum

infection and pattern of oocyst shedding in calves in Japan. Veterinary Parasitology 94, 27-32, 2000

Ungar BL, Burris JA, Quinn CA, Finkelman FD. New mouse models for chronic Cryptosporidium

infection in immunodeficient hosts. Infection and Immunity 58, 961 -9, 1 990

48

Upton SJ, Current W. The species of Cryptosporidium (Apicomplexa: Cryptosporidiidae) infecting

mammals. Journal of Parasitology 7, 625-29 , 1 985

Waldron LS, Ferrari BC, Power ML. Glycoprotein 60 d iversity in C. hominis and C. parvum causing

human cryptosporidiosis in NSW, Australia. Experimental Parasitology 1 22 1 24-1 27, 2009

Wang R, WangJ, Sun M, DangH, Feng Y, Ning C, Jian F, Zhang L, Xiao L. Molecular

characterization of the Cryptosporidium cervine genotype from a sika deer ( Cervus nippon Temminck) in

Zhengzhou, China and l iterature review. Parasitology Research 1 03 , 865-9, 2008

Warren CA, Guerrant RL. In: Fayer, R. , Xiao, L. (Eds. ) , Cryptosporidium and Cryptosporidiosis, second

ed. CRC Press and IWA Publishing, Boca Raton, FL, p. 235, 2008

Widmer G, Akiyoshi D, Buckholt MA, Feng X, Rich SM, Deary KM, Bowman CA, Xu P, Wang Y ,

Wang X , Buck GA, Tzipori S. Animal propagation and genomic survey of a genotype 1 isolate of

Cryptosporidium parvum. Molecular and Biochemical Parasitology 1 08 : 1 87-1 97, 2000

Widmer G, Tchack L, C happell C L, Tzipori S. Sequence polymorphism in the beta-tubul in gene reveals

heterogeneous and variable population structures in Cryptosporidium parvum. Applied and Environmental

Microbiology 64 : 44 77-81 , 1 998

Winkworth CL, Matthaei CD, Townsend CR. Prevalence of Giardia and Cryptosporidium spp in calves

from a region in New Zealand experiencing intensification of dairying. New Zealand Veterinary Journal 56,

1 5-20, 2008

Xiao L. Molecular epidemiology of endemic cryptosporidiosis in humans. Proceedings of the 11

International Conference on Giardia and Cryptosporidium, Morelia, Michoacan, Mexico, May 1 3 - 1 8, '

2007. Molecular Ep idemiology and Typing session 2, Abstract 7, 2007

Xiao L, Bern C, Sulaiman IM, Lal AA. Molecular epidemiology of human cryptosporidiosis. In R. C. A.

Thompson (ed.) , Cryptosporidium: from molecules to disease 227-62. Elsevier, Amsterdam, The

Netherlands, 2004a

Xiao L, Cama VA, Cabrera L, Ortega Y, Pearson J, Gi lman RH. Possible transmission of

Cryptosporidium canis among chi ldren and a dog in a household. Journal of Clinical Microbiology 45,

201 4-1 6, 2007

Xiao, L., Fayer, R. Molecular characterisation of species and genotypes of Cryptosporidium and Giardia

and assessment of zoonotic transmission. International Journal for Parasitololgy 38, 1 239-55, 2008

Xiao L, Fayer R, Ryan U, Upton SJ. Cryptosporidium taxonomy: recent advances and implications for

publ ic health. Clinical Microbiology Reviews 1 7, 72-97, 2004b

Xiao L, Feng Y. Zoonotic cryptosporidiosis (Minireview) . FEMS Immunology and Medical Microbiology 52,

309-23, 2008

49

Xiao L, Herd RP. Epidemiology of equine Cryptosporidium and Giardia infections. Equine Veterinary

Journal 26, 1 4-7, 1 994

Xiao L, Moore JE, Ukoh U, Gatei W, Lowery CJ, Murphy TM, Dooley JSG, M il lar BC, Rooney PJ, Rao

JR. Prevalence and Identity of Cryptosporidium spp. in pig slurry. Applied and Environmental Microbiology

72, 446 1 -63, 2008

Xiao L, Ryan UM. Cryptosporidiosis: an update in molecular epidemiology. Current Opinion in Infectious

Diseases 1 7, 483-90, 2004

Xiao L, Ryan UM. Molecular epidemiology. I n : Fayer, R . , Xiao, L. (Eds . ) , Cryptosporidium and

Cryptosporidiosis, second ed. C RC Press and lW A Publishing, Boca Ratan, FL, Pp. 1 1 9- 1 63, 2008

Xiao L, Limor JR, Li L, Morgan U, Thompson RC, Lal AA. Presence of heterogeneous copies of the

small subunit rRNA gene in Cryptosporidium parvum human and marsupial genotypes and

Cryptosporidium felis, Journal of Eukaryiotic Microbiology 46, 44S-45S, 1 999

Xiao L, Zhou L, Santin M, Yang W, Fayer R. Distribution of Cryptosporidium parvum subtypes in calves

in eastern United States. Parasitology Research 1 00, 701 -06, 2006

Xiao L, Sulaiman IM, Ryan UM, Zhou L, Atwil l ER, Tischler ML, Zhang X, Fayer R, Lal AA. Host

adaptation and host-parasite eo-evolution in Cryptosporidium: implications for taxonomy and public health,

International Journal of Prasitology 32. 1 773-85, 2002

Xu P, Widmer G, Wang V, Ozaki LS, Alves JM, Serrano MG, Puiu 0, Manque P , Akiyoshi 0, Mackey

AJ, Pearson WR, Dear PH, Bankier AT, Peterson DL, Abrahamsen MS, Kapur V, Tzipori S, Buck GA.

The genome of Cryptosporidium hominis. Nature 431 , 1 1 07-1 2, 2004

Yang R, Jacobson C, Gordon C, Ryan U. Prevalence and molecular characterisation of Cryptosporidium

and Giardia species in pre-weaned sheep in Australia. Veterinary Parasitology, On line,

doi : 1 0. 1 01 6/j.vetpar.2008. 1 2.021 , 2008

Zamani F, Upton SJ, LeBianc SM. Ribosomal RNA gen e o rganisation in Cryptosporidium parvum,

(abstract 455). In: Program and abstracts of the 45th annual meeting of the American Society of Tropical

Medicine and Hygiene 1 996, Baltimore MD p 250

50

CHAPTER 2

STUDIES OF THE POPULATION G ENETIC STRUCTURE OF

CRYPTOSPORIDIUM PARVUM AND CRYPTOSPORIDIUM HOM/N/5

This chapter contains two studies of the populat ion genetic structure of C. parvum and C.

hominis. The fi rst study (Sectio n 2 . 1 ) is a re-analysis of the raw data previously publ ished by

Mal lon et al . {2003a} (see also Section 1 .9} us ing a different analytical approach based on

diversity statistical tests. The study was possible due to the avai labi l ity of unpubl ished post-code

data generously supplied by Prof. Andy Tait, Wel lcome Centre for Molecular Parasitology,

G lasgow, Un ited Kingdom . These data remain confidential . The second study ( Section 2 .2) is

an or ig inal analysis of mu lti locus genotype data for C. parvum and C. hominis from seven

countries provided by Drs. S. Tarn iverdi and G . Widmer, Department of Infectious Diseases,

Cummings School of Veter inary Medicine, Tufts Un iversity, MA, U SA. All the laboratory work

described in Sect ion 2.2 was performed at Tufts. The author provided C. parvum DNA from New

Zealand, performed the analys is of the data and wrote the manuscript, together with G. Widmer.

The above studies were publ ished as:

Section 2 . 1 : Grinberg A, Lopez-Vi l la lobos N , Pom roy W , Widmer G, Smith H , Tait A. Host

shaped segregation of the Cryptosporidium parvum mult i locus-genotype repertoire.

Epidemiology and Infection 1 36 , 273-78, 2008

Section 2 .2 : Tannverdi S, G ri nberg A, Chalmers RM , Hunter P R , Petrovic Z, Akiyosh i DE,

London E , Zhang L , Tzipori S , Tumwine J K Widmer G . I nferences about the g lobal population

structure of Cryptosporidium parvum and Cryptosporidium homin is . Applied and Environmental

Microbiology, 27, 7227-34, 2008

Both Sections are presented as published, except that their format has been modified as

indicated i n the Preface.

5 1

2.1 HOST-SHAPED SEGREGATION OF THE CRYPTOSPORID/UM PAR VUM M U LTILOCUS

GENOTYPE REPERTOIRE

2.1 . 1 Summary

Cattle are among the m ajor reservoirs of C. parvum in nature. However, the relative contribution

of C. parvum oocysts or ig i nating from cattle to human d isease burden is d ifficult to assess, as

various transmission pathways - i nclud ing the human to human route, can eo-occur. In this

study, m u lti locus genotype richness of representative samples of human and bovine C. parvum

are compared statistically using analyt ical rarefaction and nonparametric taxonomic richness

est imators. Results indicate that in the t ime and space frames underlyi ng the analysed data

humans were infected with sign ificantly wider spectra of C. parvum genotypes than catt le, and

consequently, a signif icant fraction of human infect ions m ay have not orig inated from the local

bovine reservoirs .

Th is study provides statistical support to the emerging idea of the existence of disti nct

anthroponotic C. parvum cycles that do not involve catt le.

2.1 .2 I ntroduction

The m u lt ihost protozoan parasite Cryptosporidium parvum (formerly C. parvum 'Type 2') is a

major cause of diarrhoea i n humans and newborn calves, worldwide. Due to the h igh incidence

of early calfhood infect ions and the large numbers of oocysts shed with faeces dur ing natural

infections (Ongerth and Stibbs 1 989; Uga et al. 2000), newborn calves are considered among

the m ost efficient am pl if iers of C. parvum i n nature. Direct calf-to-human C. parvum

transm ission has been repeatedly i nferred from numerous case and case-control stud ies

(Pohjo la et al . 1 986ab; Reif et al . 1 989 ; S h ield et al . 1 990 ; Mi l lard et a l 1 994; Stefanogiann is et

al . 200 1 ; Hunter et al. 2004; Hunter and Thompson 2005; Xiao et al . 2004) . Yet, the relative

contribut ion of the environmental dispersal of C. parvum oocysts orig inat ing from cattle to

overall human morbidity is difficult to assess, as various transmission pathways , i nclud ing the

human-to-human route , can eo-occur.

Based on molecular epidem iological data, some authors have argued for the existence of

anthroponotic C. parvum that do not cyc le in cattle (Xiao et al . 2004) . I n support of th is idea,

Alves et a l . recently observed that HIV-positive humans in Portugal were infected with a wider

spectrum of C. parvum genetic l ineages than cattle (Aives et al. 2003, 2006) . Such i nference is

of considerable biologic and publ ic health interest, and chal lenges the generally held view that

disease control measures should target l ivestock, in particu lar cattle, as the- main reservoir for

human infections. Yet, thus far this m odel is supported by nonstatistical inferences, and from

the genetic characterisation of C. parvum isolates recovered from H IV-positive patients over

long periods of t ime (Aives et al. 2003, 2006) , and thus, its general val idity needs to be

corroborated.

52

In a study published in 2003, M allon et a l . applied a h ighly discrim inatory mu lti locus genotyping

scheme on a large battery of C. parvum cl inical iso lates from humans and cattle i n the Scottish

regions of Aberdeenshire and Dumfriesshire (Mallon et al . 2003a,b) . Forty eig ht C. parvum

mu lti locus genotypes (MLGs) were described, i ndicating an extensive genetic d iversity of this

parasite. Whereas a number of ubiqu itous and h ighly abundant M LGs caused the majority of

infections in both humans and cattle, there were many low abundance M LGs which were seen

in one or both hosts or reg ions, featuring a superdiverse M LG distribution . Based on a

dendrogram generated using the unweighted pair group method with arithmetic mean

(U PGMA), the authors hypothesised that some C. parvum i nfecting humans m ig ht not cycle in

cattle. Here, the resu lts of an analysis of the MLG abundance data generated by Mallon et al .

are presented. The analysis applies taxonomic diversity stat istical methods to test the

hypothesis that humans are i nfected with a wider spectrum of C. parvum M LGs than catt le. The

results are discussed in an epidem iolog ic and publ ic health context.

2. 1 .3 Materials and Methods

In this study, the C. parvum M LG abundance ( i .e . , the number of isolates in each M LG) of

Aberdeenshire and Dumfriesshi re or iginal ly reported by Mallon et al . (2003a,b) were used. The

or ig inal data f rom Orkney and Thurso were not i ncluded, as no human isolates were original ly

typed i n these regions. The a im of the analysis was to test the hypothesis that humans were

i nfected with a wider spectrum of C. parvum M LGs than catt le. Hence, the M LG richness ( i .e . ,

the total nu mber of MLGs) of the human and bovine C. parvum M LG assemblages were

compared us ing established taxonomic diversity statistics, based on the work ing assumption

that the isolates are independent (Krebs 1 989). To conform to this assumption, it was necessary

to remove the bovine duplicates with the same M LG, or ig inating from the same farm , as within­

farm enzootic C. parvum has been repeatedly documented us ing molecular too ls (Peng et a l .

2003 ; Tanriverdi et al . 2006 ; Trotz-Wil l iams et a l . 2006) and such dupl icates could have biased

the results. Therefore, the isolates' post-codes (most l i kely corresponding to the farm of or ig in)

were retrieved, and a new dataset that inc luded only one isolate per M LG-post-code

combination was generated. Hence, data were subjected to the following comparisons:

1 - Com parison between human and bovine C. parvum MLG richness with no reference to the

region of orig in .

2 - Comparison between human and bovine C. parvum M LG richness i n Aberdeenshire.

3 - Comparison between human and bov ine C. parvum M LG richness in Dumfriesshire.

4 - Comparison between M LG richness of bovine C. parvum from Aberdeenshire and

Dumfriesshi re .

5 - Comparison between M LG rich ness of human C. parvum from Aberdeensh i re and

Dumfriesshi re .

6 - Comparison between M LG richness of Aberdeenshi re and Dumfriessh i re , with no reference

to the host species.

53

M LG richness was compared by means of analytical rarefaction and the total r ichness

estim ators Chao 1 and ACE 1 . Rarefact ion is a stat istical method fi rst described by Sanders ( in

Hughes et a l . 200 1 ) for estimating the number of taxa expected to be present in a random

sample of any size taken from a given collect ion. The approach is useful to compare observed

taxonomic r ichness among env i ronments that have been unequally sam pled. I ndeed, observed

taxonomic r ichness can fluctuate stochastical ly due to sampl ing variat ion and is sam ple-size

dependant (Hughes and Bohannan 2005) . In essence, the difference between taxonomic

r ichness of samples taken from homogeneous (non-partitioned) populations should on ly reflect

the combined effect of sampl ing variat ion and sample s ize difference . In our case, rarefaction

answered the question : What i s the expected number of MLGs - and variance - in a random

sample of the size of the smal l subsample taken from the large sub-sample of each

comparison?

Richness estimations by analytical rarefaction were calcu lated us ing PAST software (Hammer

et a l . 2005) , which applies variance est imates g iven by Heck et al . ( 1 975) . Rarefaction cu rves of

the sub-sam ples in each comparison were constructed increasing the sam ple size by one, each

time , using the 'step by 1 ' procedure of the rarefaction menu of PAST. In addition , MLG

richness of the human and bovine samples - and the 95% confidence intervals (Cl ) - were

compared using the nonparametric total r ichness estimator Chao 1 (Chao 1 984) and the

abundance coverage estimator ACE 1 (Chao 1 992) , which return theoretical esti mates of the

total popu lation richness, includ ing u nseen M LGs.

2.1 .4 Results

Overal l , 1 1 bovine dupl icates were e l im inated from the dataset. There were no m issing post­

codes of bovine C. parvum from Dumfriesshire, and 1 4 missing post-codes from Aberdeenshire ,

which proportional ly correspond to the possible presence of on ly 2-3 bovine C. parvum

dupl icates for that reg ion . No dupl icates were seen in the human sample; there were 23 m issing

post-codes of human isolates. Yet, as wi l l be discussed later, the possible presence of human

C. parvum dupl icates does not alter the inferences of this study.

The f inal dataset, which consists of 1 67 isolates, is shown in Table 2 . 1 . Twenty f ive M LGs are

represented in humans only, six in cattle on ly, and 1 7 MLGs were shared. Overal l , and in each

reg ion separately, the human C. parvum sub-samples are larger than the bovine C. parvum

sub-samples.

Nominal results of the analytical rarefact ion, Chao1 and AC E 1 total r ichness est imates and their

95% C l , are reported in Table 2.2 and Figure 2 . 1 . Notice that, by rarefaction , the human sub­

samples have greater MLG richness than the bovine subsamples, overa l l , and in each individual

region (comparisons H B, HBA, and HBD as defined in Table 2.2). These features are not l ikely

to be the result of stochastic sampl ing variat ion because the 95% lower confidence l im its of the

54

M LG richness of the rarefied human sub-samples do not encompass the observed rich ness of

the corresponding bovine sub-samples. Conversely, the lower 95% boundary of the calcu lated

richness of the rarefied large subsamples in comparisons BB and HH largely overlap the

observed richness of the small subsamples, i ndicati ng that there is no substantial d ifference in

MLG r ichness between the reg ions i n the human or bovine sub-samples (Table 2.2) . The

rarefaction curves are shown in Figure 2 . 1 . N otice that at a sample s ize of 64 in comparison HB,

the rarefaction curve for the bovine sample a lmost reaches the asymptote, whereas the curve

for the human samples is st i l l steep. This suggests that there i s a s ignificant number of unseen

human C. parvum M LGs, but at the same t ime, bovi ne C. parvum MLGs were re latively well

sampled, i . e . a further increase in the size of the bovine sample would not be expected to

greatly increase the number of new MLGs. I nterest ing ly, the rarefaction curves in contrast BB,

HH, and AD largely overlap, which indicates that with in each host, MLG richness does not d i ffer

between regions, nor does it differ among reg ions.

Chao 1 and ACE1 total r ichness est imators of the human sub-sample are greater than the

estimators for the bovine sub-sample. I nterest ingly, the 95% C ls of the Chao 1 estimate of the

human and bovine C. parvum samples do not overlap, and the Cls of ACE 1 esti mator overlap

sl ightly.

55

Human sample

Bovine sample

Total Isolates

Aberdeenshire M LGs

2( l ), 3 ( I ) , 4 ( l ), 5( I ), 6(8), 7(3) ,

8( 1 4), 9( 1 ) , 1 0(3), 1 1 ( 1 ) , 1 3(2) ,

1 5( 1 ), 1 7( 1 ) , 1 8( 1 ), 1 9( 1 ) , 20( 1 ) ,

2 1 ( I ), 22(3) , 23( 1 ), 24(6), 25 (5 ) ,

26( 1 ) , 27( I ) , 28( I ), 29( 1 ) , 30(3 ) . TOTAL ISOLATES: 64

6(9), 7( 4 ) , 8(7) , 9(2), I 0(2), 1 1 ( I ) , 1 2(2) , 1 3( 1 ) , 1 4( 1 ), 1 6( 1 ) , 2 2(4),

23( 1 ), 24( 1 ) 30( 1 ), 3 1 ( 1 ) . TOTAL ISOLATES: 38

102

Dumfriesshire M LG s

6(8 ) , 7( 1 ) , 8(6), 9(2), 1 0( 1 ),

1 1 (3) , 2 2(4), 24( 1 ) , 2 5( 1 ), 46( 1 ) ,

47( 1 ), 50( 1 ), 5 1 ( 1 ) , 52( 1 ) , 5 3(3) , 54( 1 ) , 56( 1 ), 57( I ), 58( I ) .

TOTAL I SOLATES: 39

6(4), 7( 1 ) , 8(7), 9(3), 1 1 ( 1 ) ,

22( 4 ), 23( I ), 24(2), 27( I ), 48( I ), 55( I ).

TOTAL I SOLATES: 26

65

Total isolates

103

64

167

Table 2.1 Distribution of C. parvum multilocus genotypes (MLGs) in Scotland, stratif ied by region

(Aberdeenshi re or Dumfriessh i re) , and host species ( human or bovine C. parvum) . MLGs are represented

with numbers, as in the original papers by Mallon et al. {2003a,b) . MLG abundances are in parentheses.

Comparison

HB

HBA HBD

BB HH AD

MLG richness in the small sub-sample

1 8

1 5

1 1

1 1

1 9

23

Richness (95% Cl) of large samples rarefied

at the respective small­sample size

26.8 (2 1 . 8-30. 1 )*

1 8.67 ( 1 5 , 2-22 .08)*

1 4.4 ( 1 1 . 7- 1 7 . 2 )*

1 2 . 1 (9.8- 1 4 .5 )

1 9 .9 ( 1 5 .5-2 2 . 3 )

23 .5 ( 1 9 .6-26.9)

Chao1 and ACE1 estimators

(95% Cl)

Chao l H: 324 (87- 1 65 3 )

Chao l B : 2 6 (20 - 5 6 )

A CEJ H: I SO (72 - 396)

A CE J B : 36 (22 - 96)

NC NC NC NC NC

Table 2.2 Rarefaction, Chao1 and ACE 1 richness estimators, by comparison. HH, h uman versus bovine

C. parvum; HBA, bovine versus human C. parvum i n Aberdeenshire; HBD, human versus bovine C.

parvum in Dumfriesshi re ; BB, bovine Aberdeenshire C. parvum versus bovine Dumfriesshire C. parvum;

HH, human Aberdeenshire C. parvum versus h uman Dumfriesshi re C. parvum; AD, C. parvum from

Aberdeenshire versus C. parvum from Dumfriesshire. The 95% confidence intervals not encompassing

the observed r ichness of the respective small samples are indicated by asterisks. H : human C. parvum

sample; B: bovine C. parvum sample. C l : confidence i nterval; NC: not calculated.

56

40 30 35 30 25 25 20 20 1 5 1 5 1 0 1 0 5 5 0 0

6-l 3

20 1 6 I -I

1 5 1 2 1 0

1 0

5

0 26 26

30 25 20 1 5 H H 1 0 5 0

65 I 39

Figure 2.1 Rarefaction curves, by comparison. Sample sizes (starting from 1 ) are on horizontal axes and

estimated multi locus genotype (MLG) richness on vertical axes. HB , human versus bovine C. parvum;

H BA, human versus bovine C. parvum in Aberdeensh ire; HBD, human versus bovine C. parvum in

Dumfriesshire; BB, bovine Aberdeenshire C. parvum versus bovine Dumfriesshire C. parvum; AD, C.

parvum from Aberdeenshire versus C. parvum from Dumfriessh ire ; HH , human Aberdeenshire C. parvum

versus human Dumfriesshi re C. parvum. Sub-sample sizes within each comparison are defined in Table 1 .

The long curves in each contrast represent either human or Aberdeensh ire sub-samples. The calculated

rarefied MLG richness are reported nearby the lower 95% confidence interval bars. Rarefaction curves in

contrast HH overlap, so no error bar is provided.

2.1 .5 Discussion

The study reported by Mal lon et al . i n two papers (2003a,b) is one of the most s ign ificant

genetic comparisons between human and bovine C. parvum iso lated from cl in ical ly overt

infections so far publ ished. The orig inal authors explored patterns of population genetic

structure us ing a l lele l inkage statistics and phenetic clustering methods. Here, the M LG

abundances were modelled using the diversity statistical approach , which al lowed an estimation

of total number of MLGs (seen and unseen M LG) as a function of the number of isolates i n the

sample. To comply with the working assumption of the approach, it was necessary to remove

non-i ndependent dupl icates that m ight have inf lated the M LG abundances. The m ost obvious of

such dupl icates were the bov ine isolates of identical M LG , possibly or iginating from the same

57

···· ·· ···· ···

farms . I ndeed, without removing such clusters the d ifference between the M LG richness of the

h u m an and bovine samples wou ld have been g reater. Other levels of spatial clustering of

M LGs, for example clustering due to an imal trade between farms, cou ld not be ruled out. Yet,

such dupl icates are also possible in the human sam ple, as the same M LG may have been

transmitted to different households in the course of point-source outbreaks. Converse ly,

although the presence of post-code dupl icates in the h uman sample were possible (as not all

the human post codes were retrieved) , this does not a lter the resu lts of this study but on the

contrary, if present, such dupl icates on ly increase the sample size of hu man C. parvum without

add ing new M LGs, leading to a m ore stringent statistical test for the comparisons between

h osts .

One of the most important find i ngs of the orig i nal study by Mal lon et al . was that most i nfections

were caused by a relatively smal l number of highly abundant and ubiquitous M LGs that were

s hared by both host species. Our resu lts indicate that the MLG excess seen in the human

sam ple can not be discounted on the basis of sample sizes alone as that it is beyond the

expected stochastic variation determ ined by sample s ize difference. Furthermore, a s imi lar

M LG excess was seen in the human sample in two different regions, but not between the sub­

samples or iginati ng from the same host species but fro m different reg ions. This repetition in the

two sub-samples provides a cross validat ion against random type- 1 error (Hughes and

Bohannan 2005) . We therefore infer that in the t ime and space frames underlying the or ig inal

study, hu mans were infected by a s ign if icantly wider spectrum of M LGs than cattle. These

f indings are in accordance with the inference by Alves et al. based on the genotyping of

Cryptosporid ium surface proteins in iso lates recovered f rom H IV positive patients (Aives et al .

2003, 2006), and support i ts extension to the general populat ion.

The occurrence of an excess of low abundance C. parvum MLGs that did not transcend the

human boundary m ig ht i ndicate that certain M LGs i nfecting humans are not self-sustai n ing in

catt le. Such an idea is in l ine with previous observations (Xiao et a l . 2004) , and with the

hypothesis of the occurrence of 'human-only' M LGs form u lated in the orig inal study based on a

s imple i nspection of a UPGMA dendrogram. Alternatively, it m ight merely reflect a wide

reshuff l ing of the parasite's genetic repertoire across the human ecosystems via complex social

networks, such as trave l . Because this study analysed isolates col lected from cl in ical ly overt

cases, it could be claim ed that some M LGs seen on ly in humans caused on ly subcl inical

infections or mi ld d isease in cattle and thus, were not seen. Yet, such possibi l i ty is d iff icult to

reconci le biologically. Indeed, un less the occurrence of host specificity is assumed, newborn

calves - obviously lacking an acqui red active anti- Cryptosporidium i m m unity - should be

considered more susceptible to Cryptosporidium disease than adult humans, which were widely

represented in this study (data not shown) . Furthermore, this study analysed M LG r ichness

differences, hence, the possibi l i ty that M LG richness of bovine C. parvum was underesti m ated

is equal ly val id for the human sample.

58

In conclusion, in the t ime-space frame underlying th is or ig inal study humans were i nfected with

a wider spectrum of C. parvum genotypes than catt le , i nd icat ing that a s ign ificant fraction of

human i nfections was l ikely to have been caused by parasites that did not or ig inate from the

regional bovine reservoi rs. These results do not provide evidence of the occurrence of host

specificity in C. parvum, which in the author' s view can on ly be obtai ned by cross transm ission

studies in cattle us ing putative human-only parasites. Yet, they do not conform to a s i mpl istic

model that considers all C. parvum as m ult ihost anthropozoonotic agents, and support

statist ical ly the emerging concept of the occurrence of d ist inct cycles that do not involve cattle.

Such a phenomenon should be taken i nto account when assessin g the potential benefits of

var ious barriers across the l ivestock-human interface on public health, as it is l i kely that such

barriers would be i neffective in reg ions where anthroponotic C. parvum cycl ing is common.

Further epidemiological studies i n different geographical regions i n which humans and newborn

cattle s hare the same environment, and in vivo in catt le, a re warranted.

59

2.2 INFERENCES ABOUT THE G LOBAL POPULATION STRUCTURES OF

CRYPTOSPORID/UM PAR VUM AND CRYPTOSPORID/UM HOMINIS

2.2.1 Summary

In the previous study, publ ished multi locus genotype data of Scottish C. parvum from cattle and

humans were analysed to explore for the presence of partitions in the genetic reprtoire of the

parasite among the host species. In th is study, a battery of C. parvum and C. hominis i solates

from seven countries was genotyped using a n ine-locus DNA subtyping scheme. To assess the

existence of geographical partitions, the mu lt i locus genotype data were m ined using a cluster

analysis based on the nearest-neighbor principle. With in each country, the population genetic

structures were exp lored by combining d iversity stat istical tests, l i nkage disequ i l ibr iu m , and

eBU R ST analysis . For both parasite species, a quasi-complete phylogenetic segregat ion was

observed among the cou ntries. Cluster analysis accurately identified recently introduced

isolates. Rather than conforming to a strict paradigm of either a clonal or a panm ictic populat ion

structure, data are consistent with a flexible reproductive strategy characterized by the eo­

occurrence of both propagation patterns. The relative contribution of each pattern appears to

vary between the regions, perhaps dependent on the prevai l ing eco logical determinants of

transmiss ion.

2.2.2 Introduction

Cryptosporidium parvum and C. hominis are two species of ubiqu itous protozoan parasites

responsible for the vast majority of cases of intesti nal cryptosporidiosis in humans.

Cryptosporidium parvum i nfects humans and an imals and is considered zoonotic. I n contrast,

C. hominis is thought to lack significant an imal reservoirs and appears to be transm itted

primari ly among humans. Transmission is typically through the ingestion of water contaminated

with oocysts, or direct contact with faeces. During most of their life cycle C. parvum and C.

hominis are haploid and reside in the intestinal cells of the host, where they undergo two rounds

of asexual mu ltipl icat ion. Thereafter, the differentiation and fusion of gametes leads to a

transient diploid stage, fol lowed by meiotic division . Meiot ic recombinat ion between genetical ly

heterogeneous C. parvum genotypes has been documented i n experimental infect ion

(Tanriverdi et a l . 2007} , but the relative contribution of this process to the diversificatio n of C.

parvum and C. hominis i n nature is not understood.

Ascertain ing the popu lation genetic structure of C. parvum and C. hominis is relevant to

understanding the pathobio logy of cryptosporidiosis and tracking sources of infectio n .

Cryptosporidium parasite popu lations have been studied using PC R-based ampl ification of

single or mu ltiple genetic markers in Italy (Caccio et a l . , 2000, 2001 ), Denmark ( Enemark et a l .

2002) , Scotland (Section 2 . 1 ; Mallon et al . 2003a,b), Ind ia (Gatei et al . 2007} , Israel and Turkey

(Tanriverdi et al . 2006}, England and Wales (Hunter et al. 2007} , France and H aiti

(Ngouanesavanh et al. 2006) , and I ndia, Kenya, USA, and Peru (Gatei et al . 2006) . Some

60

studies have used a gene encod ing the sporozoite surface glycoprotein G P60 (Ceval los et al .

2000 ; Strong et a l . 2000} . This m arker is appeal ing because of its extensive sequence

polymorphism and functional relevance. However, PCR-based studies based on a sing le-locus

are l im ited in their abi l ity to capture the structure of a popu lation , which is better explored using

m u lti locus typing schemes. The mu lti locus approach was used to explore the popu lation genetic

structures of C. parvum and C. hominis in Scotland (Mal lon et al. 2003a,b) , Haiti

( Ngouanesavanh et al. 2006}, and in I ndia, Kenya, U SA, and Peru (Gatei et al. 2006} .

The a im of the present study was to expand our understanding of the popu lation structure of C.

parvum and C. hominis to a g lobal scale. Therefore , a large battery of archived C. parvum and

C. hominis DNA sam ples from seven cou ntries was genotyped at n ine polymorphic loci , and the

m u lt i locus genotype ( M LG) data were m ined us ing cluster analysis , and com bi n i ng d iversity

statistical tests with measures of l inkage disequ i l ibrium .

2.2.3 Materials and Methods

Parasite samples. DNA from 289 hu man C. hominis isolates from Uganda, the US, and the

U K , and from 2 1 7 human and bovine C. parvum iso lates from Uganda, New Zealand, Turkey,

I s rael, Serbia, and the US were i nit ial ly included . DNA purification m ethods varied from country

to country as described below. Archived DNA was stored at -20QC.

I s rael ( I L) : A total of 62 DNA samples from Cryptosporidium parvum-positive faecal specimens

co l lected between March 2004 and April 2005 from 7-1 3 days-old calves on 1 4 farms ( n=60} , a

horse (n= 1 ) and a goat kid (n=1 } , were used . The col lection sites, oocysts purification and DNA

extraction method were previously described (Tanriverdi et al . 2006} .

New Zealand (NZ) : Twenty six isolates from humans (n= 1 9) and calves (n=7) were used. The

iso lates or ig inated from spontaneously reported cases diagnosed between August and

N ovem ber 2006 at med ical or veterinary diagnostic laboratories in the Waikato reg ion , North

I s land ( n= 1 3) , and in the South Is land (n= 1 3) . To purify the oocysts, approxi m ately 5 g of

faeces were suspended in water to a volume of 7 ml and strained with gauze. A vo lume of 3 m l

of diethyl ether was added to extract the fat (Current 1 990) . Oocysts were captured with 50 111

of i m m unomagnetic beads (Dynabeads®, Dynal , Norway). Purified oocysts were suspended in

1 00 111 TE buffer and subjected to three cycles of freeze-thawing. DNA was extracted from the

oocyst lysate usi n g a commercial column DNA pu rification kit (H i Pure™ template preparation

kit, Roche Diagnostics, l nd ianapol is , lnd . ) as previously described (Tanriverdi et al. 2006) , and

the DNA eluted in 50 111 disti l led water.

Serbia (SRB) : C. parvum-positive faecal specimens ( n=50} were col lected between April 2003

and September 2006 f rom calves 8-30 days of age raised on 1 0 farms located 1 8-45 km from

the city of Belgrade . Oocysts were purified by sedimentation on sucrose g radients as previously

described (Abe et al . 2 002} . Oocysts were lysed with three cycles of freeze-thawing , and DNA

purified with DNA purification kits (QIAamp™ DNA M in i Kit, as previously reported ( M isic and

Abe 2006} .

6 1

T urkey (TR) : Cryptosporidium parvum oocysts were isolated between September 2001 and May

2005 from 1 5 faecal specimens from calves on 1 4 farms in the Kars region in northeastern

Turkey. Oocyst purif ication and DNA extraction methods were previously described (Tanriverdi

et a l . 2006} .

Uganda (UG) : Cryptosporidium-positive faecal specimens (n==237; 62 C. parvum, 1 75 C.

hominis) were collected from chi ldren under five years of age diagnosed with persistent

d iarrhoea at the D iarrhoea Treatment Ward of Mulago Hospital, in Kampala, Uganda, as part of

a large cross-sectional survey that was previously described (Tumwine et al. 2003, 2005} . The

specimens were col lected from November 1 999 through January 2001 , and from November

2002 through May 2003. Most speci mens orig inated from patients resid ing in Kampala and its

surroundings. DNA was extracted us ing the FastDNA® SPIN Kit for Soil (Qbiogene I nc . ,

Carlsbad, California) .

Un ited K ingdom (UK) : A total of 1 43 C. hominis DNA s amples were selected from an archive

com posed of over 1 4 ,000 samples extracted from human isolates submitted between 2000 and

2005 to the UK Cryptosporidium Reference Un it by diag nostic laboratories throughout Eng land

and W ales. Of these, 1 1 3 samples were from sporadic cases in i m m unocompetent patients in

the North West of E ngland and Wales. These cases were previously inc luded in a case control

study which ran from November 2000 to February 2002 (Hunter et a l . 2004} . A further 21 C.

hominis samples from sporadic cases were chosen from the national col lection to include H IV­

positive cases, and increase the geographical and age d istribution, and n ine samples were from

cases involved in two outbreaks l inked to water supplies . An outbreak was defined as two or

m ore l inked cases of cryptosporidiosis. Twenty-six sporadic cases had a history of being

outside the UK in the two weeks before the onset of i l lness, of which e ight had been in Pakistan.

Other p laces were defi ned as Mediterranean countries (n== 1 4}, other European desti nations

( n== 1 }, Africa (1 } , New Zealand (n== 1 ) and one unknown. Oocysts were separated from faecal

m aterial by flotation on a saturated salt solut ion, then i ncubated at 1 OOQC for 60 m in , d igested

with proteinase K, and the DNA extracted using DNA extraction kits as above (E iwin et a l .

200 1 } . Of a l l the samples from the UK, 1 05 were re-identified as C. hominis using the L IB1 3

m arker (see below)

United States (US) : Iso lates from the U S included C. parvum (n==1 0) and C. hominis (n==1 1 ) . The

C. parvum isolates were the reference isolate ' IOWA', three laboratory isolates used in human

vo lunteer studies (Okhuysen et al . , 1 999, 2002} , a bovine isolate from the state of Con necticut,

as wel l as isolates f rom spontaneous infections and from HIV positive individuals enrolled in a

cl in ical trial (Widmer et al. 1 998) . The C. hominis isolates were obtained from individuals

enrol led in the same cl in ical trial , f ro m two unrelated H IV-posit ive and two H IV-negative

individuals.

Genotyping. The Lib 1 3 species-specific marker (Tanriverdi et al . 2003} was used to

discri m i nate C. parvum f rom C. hominis and ru le out C. meleagridis, C. muris, and C. andersoni,

except for the isolates from Uganda, which were typed using the COW P restriction fragment

62

length polymorphism ( Spano et al. 1 997) . The UK isolates had all been previously identified at

the COW P locus by PCR-RFLP and were additional ly tested using the Lib 1 3 m arker. Nine m in i­

and m icrosatellite markers (MS) were used to subtype C. parvum and C. hominis as previously

described (Tanriverd i et al . 2006; Tanriverdi and Widmer 2006) . Al l genotyp ing work was

performed at Tufts University. Markers are located on chromosome I (MSA, M SB),

chromosome 11 (MSC) , chromosome I l l ( MSK) , chromosome IV (MSE) , chromosome V ( MS9) ,

chromosome VI (MSG) , chromosome V I I ( 1 887) , and chromosome V I I I (TP 1 4) . M inisatell ite

MS9 and microsatel l i te TP1 4 were developed by Mal lon et al. (2003a) and adapted as

previously described (Tanriverdi and Widmer 2006) . PCR amplifications were carried out in a

real-time PCR mach ine (L ightCycler® Roche Applied Science) using SYBR® G reen I m aster

m ix (Roche Diagnostics) as described (Tanriverdi and Widmer 2006) . Amplicons were

fract ionated on 1 5% n ative polyacrylam ide gels and the bands visually scored in comparison to

a set of representative al le les and to a 1 00-base-pairs ladder. The al lele scoring method was

validated by sizing selected markers on a capi l lary electrophoresis instru ment (CEO™ 8000,

Beckman Coulter, Ful lerto n , CA) . Al leles were numbered with a 3-dig it code i ndicating their s ize

i n base-pairs.

Analysi s of the data. Only isolates that amplif ied the Lib 1 3 or COW P marker and al l the n ine

loci were included in the analysis . The M LGs of isolates which i ncluded double banded loci

were defined by om itti ng the large bands. This conversion was arbitrary, and was necessary

because the species a re h aploid. No genotypes with three or more bands were observed. The

possibi l ity that some M LG s generated by the arbitrary om ission of the large bands cou ld have

biased the inferences was ruled out by repeating the analyses after om itting double banded

isolates from the dataset (see below) . To e l im i nate possible bias introduced by resam pl ing sub­

structured popu lat ions, on ly one representative isolate per MLG-farm combination was left in

the dataset as described in Section 2 . 1 . Th is resu lted i n the removal of 65 C. parvum i so lates.

This procedure was not applied to human C. parvum and C. hominis as residence i nformation

was not available. However, most isolates from the UK originated from sporadic cases of

cryptosporidiosis in patients residing at d ifferent addresses. For the analysis, M LG data were

compi led in categorical m atrixes com posed of 1 1 colum ns , with C. parvum or C. hominis

isolates characterised by a unique code, n ine al leles, and a MLG identifier. Al l the d ifferences

between proport ions were statistical ly assessed using the two-tailed Fisher's exact test.

The n ine-locus subtyp ing scheme used in this study defines a genetic d istance matrix of only

nine possible values. Therefore, dendrograms were not used to assess g lobal substructur ing

due to the occurrence of t ies and potential for severe distortions. Instead , the extent of

substructuring was assessed using a cluster analysis based on the Nearest Neighbor principle

(Cover and Hart 1 967) . The pairwise Ham ming distances (Hamming 1 950) , expressed as the

number of loci by which the MLGs differ, were used as the genetic d istances . For each C.

parvum or C. hominis M LG, its nearest neighbors, def ined as the M LGs at the shortest

63

Hamming distance, were retrieved manual ly. Within each country, the measure of clustering

was the proportio n of M LGs having al l the nearest neighbors i n the same country. The results

of this cluster analysis were explored by inspect ing the s ing le locus variant networks generated

using eBU RST software as described below. As wi l l be seen, the data did not require any

statistical test ing to be performed.

A combination of methods was used to explore the C. parvum and C. hominis popu lation

structure with in each country. The genetic diversity of both species was assessed using M LG

rank-abundance plots (Whittaker 1 965} , with M LG abundances scaled as a precent. In addit ion,

mu lt i locus genotype rich ness of C. parvum and C. hominis were statistically compared among

countries by means of analytical rarefaction (Section 2 . 1 ; Hughes et al . 200 1 ) . Rarefaction is a

statistical method for estimating the number of taxa expected to be present in a random sample

of any size taken from a given col lection. The approach is useful to compare taxonom ic

richness among environments, as the samples' taxonomic richness is dependent on sample

size, and fluctuates stochastical ly due to sampl i ng variation (Section 2.1 ) . Rarefaction was

performed using the individual-based rarefaction option in PAST software (Ham mer et al. 2005) .

To assess whether the arbitrary om ission of the large bands could have biased the shapes of

the rarefaction curves, the analysis was repeated after omitti ng double-banded isolates.

L inkage disequ i l ibr ium across all loci was assessed using the standard ized i ndex of association

( S/A) proposed by Habould and Hudson (Haubo ld and Hudson 2000). This i ndex, a derivation

of the Maynard S m ith's index of association , f luctuates stochastically around zero in complete

pan mixia, but departs further from zero (u nbound and in both d i rections) as l i nkage

disequi l ibr ium i ncreases. The i ndex, and its probabi l ity under a nu l l model of panmixia, were

calcu lated using L IAN software ( http ://aden ine .biz .fh-weihenstephan .de/cg i-b in/l ian/l ian .cg i .p l ,

accessed Ju ly 2007) , with 1 000 allele random isat ions. Linkage disequi l ibrium statistics were not

calcu lated for C. parvum from the USA, Turkey, and Israel , due to the smal l sample sizes.

These tests were repeated after om ission of double-banded isolates from Uganda.

Finally, the population structures were explored by visual assessment of networks of single

locus variants constructed using the eBU RST software (Fei l et al. 2004) . S ing le locus variant

eBURST networks are diagrams composed of M LGs depicted as dots, which are l i nked to their

s ingle locus variants by l i nes. A dot's diameter is proportional to the relative MLG frequency, so

that the largest dots correspond to the most abundant M LGs and the smal lest dots represent

s ingletons.

2.2.4 Results Genotyping. A total of 5 1 6 iso lates were identified as C. parvum (n=227) or C. hominis

(n=289) and genotyped. Out of 4,644 (5 1 6 isolates for each of the n ine loci) PCRs, 230 (5%)

fai led to generate vis ible ampl icons . The proport ion of fai l ing reactions was sign ificantly g reater

64

i n C. hominis (7 .0%) than i n C. parvum (2 .2%) (two-tai led Fisher's exact test; P=0.001 ) . The

same difference was observed between C. parvum and C. hominis iso lates from Uganda, a

country that provided human samples from both species wh ich were processed i n the same

m anner. Therefore, primer site po lymorphisms or a lower C. hominis oocyst output, rather than

different DNA extraction techn iques or host species, may have contributed to this difference.

Among all PCRs, 30 C. parvum ( 1 .4%) and 42 C. hominis ( 1 .6%) isolates showed double

bands. The s izes of the majority of the C. parvum ( 2 1 /27 [78%)) and C. hominis (33/38 [87%))

bands which were found i n double-banded reactions were also found i nd ividual ly i n other

isolates, suggest ing that most double-banded reactions reflected mixed infections rather than

PCR artifacts. The proportion of double-banded isolates was significantly greater in Ugandan C.

hominis isolates, of which 1 8 .2% had at least one double-banded locus, than in C. hominis

isolates from the Un ited Kingdom, of which on ly two (2%) were double-banded (two-tailed

Fisher's exact test; P=O.OO) (Table 2 .3) . The differences between the proportions of double­

banded C. parvum isolates from various countries were also sign ificant. However, there was no

difference between the proportion of double-banded C. parvum and C. hominis iso lates

from Uganda (two-tai led Fisher's exact test; P= 0 .24) . After removing C. parvum replicates and

incomplete C. parvum and C. hominis MLGs, the final data set was composed of 1 23 C. parvum

and 1 90 C. hominis i so lates (Appendix 2 .2) .

Cryptosporidium hominis and Cryptosporidium parvum geographic clustering . Sixty n ine

un ique C. parvum and 73 un ique C. hominis M LGs were included in the cluster analysis

(Appendix 2 .2 ) . No M LGs are shared between the two parasite species or between countries, in

either C. parvum or C. hominis. Remarkably, with the exception of one C. parvum M LG from

Uganda (UG27) , which differs from its nearest neighbor at four loci , and one from Serbia

(SRB46), which has equid istant nearest neighbors in Serbia and Israe l , al l the M LGs have their

nearest neig hbors with i n their own countries, i nd icating that more al leles are shared with i n each

country than between the countries. A marked phylogenetic segregation is also evident in the

single locus variant networks, al l of which are composed of MLGs from only one country (F ig .

2 .2) . Sign ificantly, the C. hominis isolates U K79, U K78, UK8 1 , UK28, and U K29 are not l inked

to the large C. hominis cluster from the Un ited Kingdom . Retrospective analysis of the patients'

data ind icated that al l these M LGs originated from patients that had trave l led to the United

Kingdom from Pakistan less than two weeks pr ior to the iso lation of the parasites. Together, the

clustering of these isolates and the simi lar travel h istories strongly suggested that these isolates

were introduced. I n contrast to other countries , C. hominis and C. parvum M LG s from the United

States did not c luster, which is consistent with the disparate geographic and temporal orig ins of

these samples, and are therefore not shown in Fig . 2 .2.

65

A

IL49

SRB61

SRB62

B

UK31 . UKJO

UK33 • •

• L51

l.4�

ll28

• UG23 o SRBSa

o UGS , SRB59

· SRB46 o UG12

• UG1 1

SRI!eo SRB71

SRB73

. uK2 • UK7B , UK37

, UK79 • UKJ

• UK8 1

e UG2

• UG1 4

• UG20

TR:)l

TR30

TR26 o TR29

• 0024

• 0027

. u= ______. UG 19

0018 NZ� -� _-- NZ35

• UG22 · Nl52 NZ43

• UK29

• UK28 • UK77 • UK82 · UK32

• UG16

• UK4 • UK36

· UK80

. uK20 . uK25

UK22 . • UK26 .!:JK21 -· UG44 ..-

K1 � UK23

• UK2f • UK24

UG71

• UG70

• UG65

, UG74

U013 I ,UG16 ,· -

UG1:; ... . �·�. --U05

·-; UG17 0014 UG12

UG4 1

UG43 UG66 T --- · UG67

�G46 . __ JJG49 u� ,!-JO_OO lK.45: l ,.__ - I tlG40 - . UG39 UG47 .,.

UG7 . .

LJC42 .-

'· ,uG62

GB "- , UGS1

0G56 -r o UGe3 ,uo10

G76

e UG1 1 • UG57

• UG:;.! •--UG6

· 0056

-.. u-;�

- • UG72

, UG64

• UGSO

. UG!>6

UG51

Figure 2.2 Single locus variant eBU RST networks for C. parvum (A) and C. hominis (B) . Each MLG is

represented as a dot and designated by a unique identif ier. Dot diameters are proportional to the number

of isolates. For example, C. hominis UK22 was the most abundant MLG, found in 59 isolates, and the

smallest dots represent singletons. Single- locus variants are connected by l ines. IL , Israel ; NZ, New

Zealand ; SRB, Serbia; TR, Turkey; UG, Uganda; UK, United K ingdom.

66

Population genetic structure. Figure 2 .3 ( left) shows the g lobal MLG rank abu ndance curves

for C. hominis and C. parvum. The plot reveals the presence of a smal l number of highly

abundant M LGs and a large number of s ing letons i n both species, consistent with previous data

from Scotland ( Mallon et al. 2003a,b ; Sect ion 2 . 1 ). I n C. hominis, 59 out of 1 90 (31 %) isolates

belong to the m ost abundant MLG ( U K22) (Appendix 2 .2) , whereas the m ost abundant C.

parvum M LG (NZ41 ) accounts for 1 4/1 23 ( 1 1 %) isolates. A comparison between C. hominis

rank abundance plots from the United Kingdom and Uganda (Fig . 2 .3 , r ig ht) revealed a m ore

even d istribution in Uganda, where the m ost abundant MLG (UG 1 1 ) accounts for only 1 0/85

( 1 2%) of the isolates, compared with 5 1 /74 (69%) in the Un i ted Kingdom ( isolates from patients

reporting trave l excluded) .

� 0

"' u

100

� 10 "0 c: ::> .0 "' "' ·"' iii �

100

-...- UG --o- UK

---+-- C. hominis - -o - C. parvum

10 20 30 40 50 60 70 80 90 10 15 20 25 30 35 40 45 50 rank rank

Figure 2.3 C. parvum and C. hominis MLG rank abundance plots. The relative abundance of each MLG

is shown on the vertical axis as percent of a l l isolates. For each curve, the least abundant MLGs are

singletons. Left: global analysis shows similar rank abundance for both species; R ight: comparison of

Uganda and UK C. hominis showing a more even MLG distribution in Uganda.

Rarefactio n curves were used to compare C. parvum and C. hominis MLG richness levels

among cou ntries (Fig . 2 .4) . The MLG r ichness levels in the C. hominis and C. parvum samples

from Uganda and C. parvum from Turkey and Serbia, countries with a large proportion of

isolates with double-banded loci, are very s im i lar, as indicated by s im i lar rarefaction curves . I n

contrast, M LG richness in the samples from the Un ited K ingdom , Israel , and New Zealand,

countries where only a few double-banded genotypes were observed, is smal ler, as indicated

by lower curves approaching the asymptote. R arefaction curves drawn without double-banded

C. parvum and C. hominis isolates from Uganda and C. parvum isolates from Serbia and Turkey

displayed very s im i lar and overlapping curves ( not shown) , i nd icating that the arbitrary o m ission

of large bands did not bias th is result . As a substantial number of C. parvum and C. hominis

isolates was co l lected from humans in Uganda as part of the same surveys , it was possible to

compare M LG richness levels between C. parvum and C. hominis by using samples orig inat ing

67

c ::J 0 u

80

60

<.9 40 _j :2

20

from the same environment. Rarefied from a sam ple size of 85 to the C. parvum sam ple size of

36, C. hominis M LG richness in Uganda has a value of 24.2, which is the same as the M LG

richness of 24 observed for C. parvum from the same country.

- C. hominis - - - C. parvum

20 40 60 80 100 120 140 1 60 1 80 200 0 10 20 30 40 50 60 70 80 90

isolate count isolate count

Figure 2.4 C. parvum and C. hominis analytical rarefact ion curves. Left, global cu rves; right, country­

specif ic curves. Ch, C. hominis; Cp, C. parvum. Error bars indicate the 95% confidence intervals for global

C. parvum and C. hominis and C. hominis from Uganda and the U K. Isolates from the US were excluded.

Note that the MLG richness in New Zealand is the lowest among the analysed countries, a result that wi l l

be re-addressed in the study presented in Section 3.5 and in Chapter 5.

I n addition to showi ng a clear MLG segregation by country, as discussed above in the context of

the cluster analys is , the eBURST d iag rams (Fig. 2 .2) also reveal differences between countries

with respect to the parasites' population structure. The d iagram of C. hominis from the United

Kingdom has a star-l ike topo logy, with the m ost abundant MLG (UK22) connected to the

g reatest number of single-locus variants. This topology is remin iscent of eBU RST networks of

computer-simu lated popu lations d iversifying at a h igh mutation-to-recombination ratio (Turner et

a l . 2007). In samples drawn from such popu lations, most l inks between the s ing le- locus variants

m i rror true clonal descent relat ions. Converse ly, the eBU R ST diag ram for C. hominis from

Uganda is stragg ly and shows long chai n ing of sing le-locus variants, a feature typical of

computer-simu lated popu lations d iversify ing at a low mutation-to-recombination ratio , i n which a

s ignificant proportion of l inks arise from recombination rather than true clonal descent (Turner et

a l . 2007) . The eBURST diagram for C. parvum from Uganda is also stragg ly and shows M LG

chain ing . The exclusion of double-banded isolates from Uganda did not s ign ificantly affect the

topologies of the eBU RST diagrams ( not shown) , indicating that the om ission of the large bands

for the def in it ion of the MLGs did n ot bias these resu lts .

68

I nterest i ng ly, the M LGs with the greatest number of s ingle-locus var iants ( U K22, UG1 1 , U G 1 5,

I L -44 , N Z4 1 ) were also the most abundant in al l the countries where abundant M LGs were

observed ( Fig . 2 .2 and Appendix 2.2) . This pattern argues against a recent epidemic expansion

of the abu ndant M LGs, as in such cases, al l the M LGs were equal ly l ikely to expand

epidem ical ly and prevail in the sample. Instead, the pattern is consistent with a popu lation

orig i nating from founders , as proposed for bacterial species by Fei l and Spratt (2001 ) .

Accord ing to th is structure, abundant genotypes show a g reater degree of diversif ication than

do rare genotypes, s imply because they undergo more m ultiplications than rare genotypes.

In support of the d ifferences between the eBU RST network topologies and M LG rich ness levels

between Uganda and the Un ited Kingdom , there is strong statistical evidence of al lele l i nkage

disequ i l ibrium in C. hominis iso lates from the United Kingdom . Conversely, although

"statistical ly signif icant," the SIA for C. hominis iso lates f rom Uganda is close to zero (Table

2 .3) . I nteresti ng ly, the proport ion of double-banded isolates of C. hominis is also sign ificantly

greater in Uganda than in the Un ited K ingdom, indicati ng a higher rate of m ixed infect ions in

Uganda, but the SJA did not change m uch after e l im inating the double-banded isolates (not

shown) . To further rule out an epidem ic structure or a b ias due to the presence of imported,

thus reproductively iso lated, M LGs in C. hominis isolates from the Un ited Ki ngdom, l i nkage

disequi l ibr ium ana lysis was repeated by us ing one iso late per MLG, without the iso lates from

patients report ing travel . As expected for a basic clonal popu lation structure (Sm ith et al. 1 993) ,

l i nkage disequi l ibr ium persisted (SIA=0 . 1 786; P= 0 .0 1 ) .

The popu lation structure of C. parvum i s less clear a s t h e sample sizes from Israel , Serb ia and

Turkey are relatively smal l , and no C. parvum samples from the UK were avai lable. Although

there is statistical evidence for al lele l i nkage disequi l ibri um in C. parvum from Uganda, the s la is

only 0 . 1 6 and the eBURST diagram is straggly and shows chain ing of MLGs. On the other

hand , there is no statistical evidence for l inkage disequi l ibr ium in C. parvum from New Zealand

and Serbia (Table 2 .3) . However, the lack of l inkage diseq u i l ibrium in these countries should not

be viewed as evidence for complete panm ixia, as it could be due to an u nderpowered statistical

test due to the small sample s ize, or the presence of only four polymorphic loci in New Zealand

C. parvum (Appendix 2 .2 ) . I ndeed, the eBU RST network for New Zealand C. parvum is star-l ike,

with the m ost abundant M LG ( NZ41 ) in the center, rem in iscent of a clonal population structure .

In contrast, the diagrams from Serbia and Turkey are straggly, consistent with the g reater

proport ion of double-banded C. parvum isolates observed in these countries.

69

Number of SIA Number (percentage) of

isolates with (p-value) double-banded isolatel

complete MLGsa

Uganda (UG)

C. hominis 85 0.06 (p<O.OO I ) 3 2 ( 1 8%)

C. parvum 36 0. 1 6 (p<O.OO I ) 7 ( 1 1 %)

UK

C. hominis 95 0.3 8 (p<O.OO I ) 2 (2%)

New Zealand (NZ)

C.parvum 24 0.03 (p<O. I ) 0

Israel (IL)

C. parvum 20 ND l (2%)

Serbia (SRB)

C. parvum 23 0.0 1 6 (p<0. 1 ) 7 ( 1 3%)

Turkey (TR)

C. parvum l l ND 5 (33%)

Table 2.3 C. parvum and C. hominis l inkage disequi l ibr ium and double-banded genotype statistics

according to country of origin. C. parvum and C. hominis i solates from the United States were excluded .

NO, not determined; SIA, standardised index of association.

a Isolates from USA excluded ; b All typed isolates, including incomplete MLGs, were used to calculate the

p roportion of double-banded isolates.

2.2.5 Discussion

I n this study, a battery of C. parvum and C. hominis isolates col lected from seven countries was

genotyped, enabl ing i nferences on the parasites' g lobal population structures to be made. The

m ain aims were to assess whether C. parvum and C. hominis popu lations are geographical ly

partitioned, and if so, compare the structure of different popu lations.

The results of the cluster analysis and the eBURST diagrams show a quasi-complete

phylogenetic segregation among countries i n both parasite species. Gene flow was therefore

70

not sufficient to erase genetic divergence among these geograph ical ly separated populations,

wh ich basical ly rem ained al lopatric . This result is noteworthy consider ing that both species are

cosmopolitan pathogens and m ay eas i ly travel between countries as asymptom atic infect ions.

Potential ly, the ecology of cryptosporidiosis in different countries may have selected for

phenotypes which are best adapted to each environment. This hypothesis would predict that

im ported parasites would be un l ikely to spread if environmental factors or transmission patterns

are unfavourable. This view is supported by the findi ng of a number of clustered C. hominis

isolates from the U nited Kingdom, which were apparently introduced from Pakistan and did not

c luster with the large network of M LG s from the Un ited Kingdom. Analogous observations were

reported in a recent M LG analysis of Scottish and Peruvian human isolates ( Morrison et a l .

2008) and are in accordance with the results of a study using sequence analysis of the G P60

locus (Chalmers et al . 2008) . The diversity between local and imported M LGs m ay also explain

the putative anthroponotic C. parvum reported by other authors (Mallon et a l . 2003b; Hunter et

al. 2007), as it is possible that "anthropogenic" isolates were in fact recently imported.

Geog raphic segregat ion, however, was not reflected in the C. parvum sample from New

Zealand, as no partit ion between North and South Is land was observed and the isolates

clustered in a s ingle eBU RST network (Figure 2 .2 ) . This interesting result suggests that i ntense

gene flow, as m ight occur through animal and human movements , may mix the genetic

repertoire of C. parvum even in the presence of sig nificant geograph ical barriers, i n this case

the Cook Strait. To further evaluate this model, the same cluster ana lysis was performed with

previously published C. parvum M LG data from Scotland (Mallon et al . 2003b) and no

phylogenetic part i t ion between the noncontiguous regions of Aberdeensh i re and Dumfriessh i re

could be seen (Fig u re 1 . 5) . The lack of partitioning i n the Scottish data may be explai ned by the

fact that the cattle popu lation in Aberdeenshire experiences h igh movement due to the im port of

cattle from elsewhere in the United Kingdom, therefore preventing the establishment of settled

and epidemiological ly isolated herds (Gi lbert et a l . 2005) . In contrast to these f indi ngs , C.

parvum samples col lected from ind ividual farms in Israel and Turkey showed a m arked

segregation of the M LGs by farm and a m icroepidemic structure with i n the farms (Tanriverdi et

al. 2006) , i ndicating both micro- and m acroecological factors impact C. parvum populations.

Because C. hominis is thought to propagate pr imari ly i n hum ans but C. parvum has a wider host

range, it was interesting to compare the population structures of t hese species, to assess

whether the differences in host range affect the population structures. Cryptosporidium

parasites have been defined as be ing clonal or panmict ic based o n the resu lts of l inkage

statistical tests emulating null m odels of random allele association (Mallon et al. 2003a,b;

Ngouanesavanh et a l . 2006) . However, these tests are l im ited i n their abi l ity to describe

popu lations propagating both clonal ly and by recom bination, as i n such cases the test stat istic

wil l be below or above the nu l l -model rejection threshold in function of the contribution of each

propagation pattern to the populat ion structure (Sm ith et al . 1 993) . Therefore, in this study we

used a combinat ion of complementary analytical approaches. In contrast to reports of local

7 1

genetic homogeneity i n C. hominis ( Mal lon et al . 2003a; Ngouanesavanh et al . 2006; Hu nter et

al . 2007) , no evidence of difference in the genetic d iversity of C. parvum and C. hominis was

found in Uganda, a cou ntry that provided samples of both species. Furthermore, the eBU RST

topology and l inkage disequi l ibr ium statistics in Ugandan C. parvum and C. hominis are

consistent with a s im i lar population structure in both species. Conversely, the coherent

differences between C. hominis from Uganda and the UK with respect to M LG richness,

eBU RST topology, and SIA, suggest a g reater rate of recombination in C. hominis i n Uganda

than in C. hominis from the U K . The sample from Uganda showed greater M LG richness, as

ind icated by a sign ificantly steeper rarefaction curve , as would be expected in a popu lation that,

in additio n to accum ulat ing variation by mutation , d iversifies by recombinat ion. Consistent with

this resu lt, there is a greater proportion of doub le-banded isolates in Uganda, suggesting that

many infections were caused by genetical ly heterogeneous paras ites, which enhances the

opportun ity for a d iversif ication by recombi nat ion. On the other hand, the resu lts of M LG

diversity analysis, eBURST network topology, and l inkage disequi l ibrium i n the United Ki ngdom ,

converge to i ndicate a g reater contribution of clonal propagation of C. hominis in th is country,

and perhaps also of C. parvum i n New Zealand.

Therefore, rather than being species-specific, C. parvum and C. hominis popu lations appear to

be shaped by local and host-related factors. The author postu lates that frequent transm ission i n

h igh ly contam inated environments increases the probabi l ity of infections with genetical ly

heterogeneous parasites, favour ing recombination . In countries where the san itary condit ions

are better and HIV is less prevalent ( l ike in the U K) , coi nfections with heterogeneous parasites

orig inat ing from environmental sources may be less f requent, and clonal propagatio n may

prevai l .

I n summary, the resu lts presented here ind icate that although C. parvum and C. hominis are

present worldwide, gene flow does not seem suff icient to erase genetic divergence among

geog raph ical ly separated populat ions, support ing a m odel of al lopatr ic diversification . As a

consequence, mu lt i locus genotyping would enable in m any cases the tracking of parasites to

their country of orig i n . R ather than conforming to a strict paradigm of either clonal or panmictic

species, the data are consistent with the eo-occurrence of both propagation pathways. The

relative contribut ion of each pathway appears to vary according to the prevai l ing ecological

determinants of transm ission, indicative of a h igh adaptive plasticity in C. parvum and C.

hominis.

2.3 CONCLUDING REMARKS

The resu lts of the studies presented in this Chapter indicate the existence of a conspicuous

sub-structuring in space and host-range, of the genetic repertoire of C. parvum and C. hominis.

These features support the idea of the existence of anthroponotic C. parvum parasites that do

not cycle in cattle, and i ndicate the feasibi lity of tracki ng of Cryptosporidium parasites to their

72

country of orig in using molecu lar tools. A com bination of different analytical approaches , rather

than l inkage disequ i l ibrium alone where results are open to mu lt iple explanat ions , was used i n

the study presented in Section 2 . 2 to i nfer the population genetic structures of C. parvum and C.

hominis at a g lobal scale. The results converged in indicat ing that C. hominis, and perhaps C.

parvum, are capable of d iversifying both by m utation and recombination. The rel iabi l ity of

genetic markers as faithful i ndicators of ancestry (and their usefulness as epidemio logical

m arkers) m ay thus vary among regions in function of the relative contribution of recombination

to the genetic d iversification . The cl in ico-epidemiological i m pl ications of the b io-geographical

sub-structuring of C. parvum and C. hominis sti l l need to be establ ished .

2 .4 REFERENCES

Alves M, Xiao L, Antunes F, Matos 0. Distribution of Cryptosporidium subtypes in h umans

and domestic and wild ruminants in Portugal . Parasitology Research 99, 287-92, 2006

Alves, M, Xiao L, Sulaiman I, Lal AA, Matos 0, Antunes F. Subgenotype Analysis of

Cryptosporidium Isolates from Humans, Catt le, and Zoo Rum inants i n Portugal . Journal of

Clinical Microbiology 2744-7, 2003

Caccio S, Homan W, Camil l i R , Traldi G , Kortbeek T, Pozio E. A m icrosatellite m arker

reveals popu lation heterogeneity with in human and animal genotypes of Cryptosporidium

parvum. Parasitology 1 20 , 237-44, 2000

Caccio S, Spano F, Pozio E . Large sequence variation at two m icrosatel l i te loci among

zoonotic (genotype C) isolates of Cryptosporidium parvum. International Journal of Parasitology

3 1 , 1 082-6 , 200 1

Chao A. Non parametric esti m ation of the number of classes in a population . Scandinavian

Journal of Statistics 1 1 , 265-70 , 1 984

Chao A, Lee SM. Estimating the number of classes via sampl ing coverage. Journal of the

American Statistical Association 87, 2 1 0- 1 7, 1 992

Cevallos AM, Bhat N, Verdon R, Hamer DH, Stein B, Tzi pori S, Pereira ME, Keusch GT,

Ward HD. Mediation of Cryptosporidium parvum infection in vitro by m ucin-like glycoproteins

defined by a neutralizing monoclonal antibody. Infection and Immunity 68, 5 1 67-75, 2000

Chalmers RM, Hadfield SJ, Jackson CJ, Elwin K, Xiao L, H unter P. Geograph ic Linkage and

Variation in Cryptosporidium hom inis . Emerging Infectious Disease 1 4, 496-8, 2008

73

Cover TM, Hart PE. Nearest Neighbor Pattern Classification . Ieee Transactions on Information

Theory l t 1 3 , 2 1 -+, 1 967

Elwin K, Chalmers RM, Roberts R, Guy EC, Casemore DP. Modificat ion of a rapid method for

the identification of genespecific polymorph isms in Cryptosporidium parvum and its application

to c l in ical and epidemiological invest igations. Applied and Environmental Microbiology

67:5581 -84, 2001

Enemark HL, Ah rens P, Juel CD, Petersen E, Petersen RF, Andersen JS, Lind P,

Thamsborg SM. Molecular characterizat ion of Danish Cryptosporidium parvum i solates.

Parasitology 1 25, 331 -4 1 , 2002

Feil EJ, Spratt BG. Recombination and the population structures of bacterial pathogens.

Annual Review of Microbiology 55, 561 -90, 200 1

Feil EJ, Li BC, Aanensen OM, Hanage WP, Spratt BG. eBU R ST: Inferring patterns of

evolutionary descent among clusters of related bacter ia l genotypes from mu lti locus sequence

typ ing data. Journal of Bacteriology 1 86, 1 5 1 8-30, 2004

Gatei W, Hart CA, G i lman RH, Das P, Cama V, Xiao L. Development of a m u lti locus sequence

typ ing tool for Cryptosporidium hominis. Journal of Eukaryotic Microbiology 53 Suppl 1 , S43-8,

2006

Gatei W, Das P, Dutta P, Sen A, Cama V, Lal AA, Xiao L. Multi locus sequence typ ing and

genetic structure of Cryptosporidium hominis from chi ldren in Kolkata, I nd ia. Infection Genetics

and Evolution 7, 1 97-205, 2007

Gilbert M, Mitchell A, Bourn D, Mawdsley J, Cl ifton-Hadley R, Wint W. Catt le movements

and bovine tuberculosis in Great Britain . Nature 435, 49 1 -6, 2005

Hammer 0, Harper OAT, Ryan PD. PAST: P alaeontolog ical Statistics, ver . 1 .37.

(http ://folk.uio. no/ohammer/past) . Downloaded on 1 5 January 2005

Hamming RW. Error Detecting and E rror Correct ing Codes. Bell System Technical Journal 29,

1 47-60, 1 950

Heck KL, van Belle G, Simberloff D. Explicit calcu lation of the rarefaction d iversity measurement and the determination of sufficient sample size. Ecology 56, 1 459-6 1 , 1 975

Haubold B, Hudson RR. LIAN 3.0: detect ing l inkage disequi l ibr ium in mu lti locus data.

Bioinformatics 1 6 , 84 7-8, 2000

74

Hughes JB, Bohannan BJM. Application of ecological diversity statistics i n m icrobial ecology.

I n : Kowalchuk GA, de Bruj in FJ, Head IM , Akkermans ADL, van E lsas JD (eds . ) . Molecular

Microbial Ecology Manual, 2nd Ed. Pp. 1 33 1 -44. Kluwer Academic Publ ishers, Dordrecht, The

Netherlands 2005

Hughes JB , Hellmann JJ, Ricketts TH. Counting the uncountable: stat istical approaches to

estimating m icrobial diversity. Applied and Environmental Microbiology 67, 4399-06, 2001

Hunter PR, Hadfield SJ, Wilkinson D, Lake IR, Harrison FC, Chalmers RM. Subtypes of

Cryptosporidium parvum in humans and disease risk. Emerging Infectious Diseases 1 3 , 82-8,

2007

Hunter PR, Hughes S, Wood house S, Syed Q, Verlander NO, Chalmers RM, Morgan K,

Nichols G , Beeching N, Osborn K. Sporadic cryptosporidiosis case-control study with

genotyping . Emerging Infectious Diseases 1 0 , 1 24 1 -9, 2004

Hunter P, Thompson RCA. The zoonotic transmission of Giardia and Cryptosporidium.

International Journal for Parasitology 35, 1 1 8 1 -90, 2005

Krebs CJ. Ecological Methodo logy, 2nd ed . , p. 4 1 5. Add ison Wesley Longman I nc . , Menlo

Park, Ca. USA, 1 989

Mallon ME, Macleod A, Wastl ing JM, Smith H , Tait A. Mult i locus genotyping of

Cryptosporidium parvum Type 2: popu lation genetics and sub-structuri ng . Infection, Genetics

and Evolution 3 , 207-1 8 , 2003a

Mallon MA, Macleod J , Wastling H, Smith H, Reil ly 8, Tait A. Population structures and the

role of genetic exchange in the zoonotic pathogen Cryptosporidium parvum . Molecular Genetics

and Evolution 56: 407-1 7, 2003b

Millard PS, Gensheimer KF, Addiss DG, Sosin OM, Beckett GA, Houck-Jankoski A,

Hudson A. An outbreak of cryptosporidiosis from fresh-pressed apple cider

Journal of the American Medical Association 272 : 1 592-96, 1 994

Misic Z, Abe N. Subtype analysis of Cryptosporidium parvum isolates from calves on farms

around Belgrade, Serbia and Montenegro, using the 60 kDa glycoprotei n gene sequences .

Parasitology, 1 -8 , 2006

Morrison LJ, Mallon ME, Smith HV, Macleod A, Xiao L, Tait A. The popu lation structure of

the Cryptosporidium parvum population in Scotland : A complex picture. Infection, Genetics and

Evolution 8, 1 2 1 -9 2008

75

Ngouanesavanh T, Guyot K, Certad G, Fichoux YL, Chartier C, Verdier RI, Cail l iez JC,

Camus D, Dei-Cas E, Banuls AL. Cryptosporidium Popu lation Genetics: Evidence of Clonality

in I solates from France and Haiti . Journal of Eukaryotic Microbiology 53, S33-S6, 2006

Okhuysen PC, Chappell CL, Crabb JH, Sterling CR, DuPont HL. Viru lence of th ree distinct

Cryptosporidium parvum isolates for healthy adults. Journal of Infectious Diseases 1 80 , 1 275-

8 1 , 1 999

Ongerth JE, Stibbs HH . Prevalence of Cryptosporidium i nfection in dai ry calves in Western

W ash ington . American Journal of Veterinary Research 50, 1 069-71 , 1 989

Peng MM, Mark AE, Wilson L, Holland RE, Steven AE, Meshnick R, Lal AA, Xiao L. Genetic

d iversity of Cryptosporidium spp. in cattle in Mich igan: impl ications for understanding the

transm ission dynam ics. Parasitology Research 90, 1 75-80, 2003

Pohjola S, Oksanen H, Jokipii L, Jokipii AM . Outbreak of cryptosporidiosis among veteri nary

students. Scandinavian Journal of Infectious Diseases 1 8, 73-8, 1 986a

Pohjola S, Jokipi i AM, Jokipii L . Sporad ic cryptosporidiosis in a rural populat ion is

asymptomatic and associated with contact to cattle. Acta Veterinaria Scandinavica. 27, 9 1 - 1 02 ,

1 986b

Reif JS, Wimmer L, Smith JA, Dargatz DA, Cheney JM . Human cryptosporidiosis associated

with an epizootic in calves. American Journal of Public Health 79, 1 528-30, 1 989

Shield J, Baumer JH, Dawson J A, Wilkinson PJ. Cryptosporidiosis - an educational

experience. The Journal of I nfection 2 1 , 297-301 , 1 990

Smith JM, Smith NH, O'Rourke M, Spratt BG. How clonal are bacteria? Procceedings of the

National Academy of Science. USA. 90 :4384-88, 1 993

Spano F, Putignani L, Mclauchl in J, Casemore DP, Crisanti A. PCR-RFLP analysis of the

Cryptosporidium oocyst wal l prote in (COW P) gene d iscrim inates between C. wrairi and C.

parvum, and between C. parvum isolates of human and animal or ig in . FEMS Microbiology

Letters 1 50 , 209- 1 7, 1 997

Stefanogiannis N, Mclean M, Van Mil H. Outbreak of cryptosporid iosis l inked with a farm

event. The New Zealand Medical Journal 1 1 4 :51 9-2 1 , 2001

76

Strong WB, Gut J , Nelson RG. Cloning and sequence analysis of a h igh ly polymorphic

Cryptosporidium parvum gene encoding a 60-ki lodalton g lycoprotei n and characterization of its

1 5- and 45-kilodalton zoite surface antigen products . I nfection and Immunity 68, 4 1 1 7-34,

2000

Tanriverdi S, Arslan MO, Akiyoshi DE, Tzipori S, Widmer G. Identification of genotypical ly

m ixed Cryptosporidium parvum populations in humans and calves. Molecular and Biochemical

Parasitology 1 30 , 1 3-22, 2003

Tanriverdi S, Markovics A, Arslan MO, ltik A, Shka p V, Widmer G. Emergence of d istinct

genotypes of Cryptosporidium parvum in structured host popu lations. Applied and

Environmental Microbiology 72, 2507- 1 3 , 2006

Tanriverdi S, Widmer G. Differential evolution of repetitive sequences in Cryptosporidium

parvum and Cryptosporidium hominis. Infection Genetics and Evolution 6, 1 1 3-22 , 2006

Tanriverdi S, Blain JC, Deng B, Ferdig MT, Widmer G . Genetic crosses in the apicom plexan

parasite Cryptosporidium parvum defi ne recombination parameters. Molecular Microbiology 63,

1 432-9, 2007

Trotz-Wil l iams LA, Martin OS, Gatei W, Cama V, Peregrine AS, Martin SW, Nydam DV,

Jamieson F, Xiao L. Genotype and subtype analyses of Cryptosporidium isolates from dairy

calves and humans i n Ontario. Parasitology Research 99, 346-52, 2006

Tumwine JK, Kekiti inwa A, Nabukeera N, Akiyoshi DE, Rich SM, Widmer G, Feng X,

Tzipori S. Cryptosporidium parvum in chi ldren with d iarrhea in Mu lago Hospital, Kampala,

Uganda. American Journal of Tropical Medicine and Hygiene 68, 71 0-5, 2003

Tumwine JK, Kekiti inwa A, Bakeera-Kitaka S, Ndeezi G , Downing R, Feng X, Akiyoshi DE,

Tzipori S . Cryptosporidiosis and m icrosporidiosis in ugandan chi ldren with persistent d iarrhea

with and without concurrent infection with the human immunodeficiency virus. American Journal

of Tropical Medicine and Hygiene 73, 92 1 -5, 2005

Turner KM, Hanage WP, Fraser C, Connor TR, Spratt BG. Assessing the reliabi lity of

eBU R ST using s imu lated popu lations with known ancestry. BMC Microbiology 7, 30, 2007

Uga S, Matsuo J, Kono E, Kimura K, lnoue M, Rai SK, Ono K. Prevalence of

Cryptosporidium parvum infection and patterns of oocyst shedding in calves in Japan.

Veterinary Parasitology 94, 27-32, 2000

77

Whittaker RH. Dominance and diversity in land p lant com munit ies: N umerical relat ions of

species express the i mportance of competition i n commun ity function and evolution . Science

1 47, 250-60, 1 965

Widmer G, Tzipori S, Fichtenbaum CJ, Griffiths JK. Genotypic and phenotypic

characterization of Cryptosporidium parvum isolates from people with A IDS. Journal of

Infectious Diseases 1 78 , 834-40, 1 998

Xiao L, Fayer R, Ryan U, U pton SJ. Cryptosporidium taxonomy: Recent Advances and

Impl ications for Publ ic Health . Clinical Microbiology Reviews 1 7, 72-97, 2004

78

CHAPTER 3

EPIDEMIOLOGICAL STUDIES OF CRYPTOSPORIDIOSIS IN DOMESTIC

ANIMALS IN NEW ZEALAND

This chapter contai ns three epidemiological studies of cryptosporidiosis in New Zealand on

horses (Sections 3 . 1 - 3 .3) and cattle ( Sections 3 .4 and 3 .5) . The studies on horses were

in i tiated in 2002 with a letter submitted by the author to New Zealand equ ine practit ioners. I n

that letter, the practit ioners were asked t o co-operate i n identifyi ng cases o f cryptosporid iosis in

foals in order to genetically characterise the Cryptosporidium isolates , g iven the lack of data

avai lable. Followin g th is in itiative, a phone cal l from a concerned cl in ic ian in the Waikato (Dr.

Laurinda Oliver) was received, which prompted the outbreak investigat ion reported in Section

3 . 1 . The molecular test ing described in th is section was kindly performed and described by the

late J i m Learmonth . The subsequent studies reported in Section 3.2 and 3.3 were performed i n

2006 a n d 2007. T h e stud ies o n catt le were performed i n 2002 (Section 3.4) and 2006 ( Section

3 .5) . Except for some diagnostic test ing performed in the study in Section 3. 1 , all the work for

these studies was performed at Massey University.

Four of the above studies have been publ ished as :

Sect ion 3 . 1 : Grinberg A, Oliver L, Learmonth JJ, Leyland M, Roe W, Pom roy W E . Identif ication

of Cryptosporidium parvum 'cattle' genotype from a severe outbreak of neonatal foal diarrhoea.

Veterinary Record 1 53, 628-30 , 2003

Sect ion 3 .2 : Grinberg A, Learmonth J , Kwan E, Pom roy W, Lopez Vi l la lobos N , Gibson I ,

Widmer G. Genetic diversity and zoonotic potential o f Cryptosporidium parvum causing foal

diarrhoea. Journal of Clinical Microbiology 46:2396-98, 2008

Section 3 .3 : G rinberg A , Pom roy WE, Carslake H , Sh i Y, Gibson I , Drayton B. A study of

neonatal cryptosporidosis of foals i n New Zealand . New Zealand Veterinary Journal 57, 284-

289, 2009

Sect ion 3.4: G rinberg A, Pom roy W, Weston JF, Ayanegui-Aicerreca A, Kn ight D . The

occurrence of Cryptosporidium parvum, Campylobacter and Salmonella i n newborn dairy calves

in the Manawatu reg ion of New Zealand. New Zealand Veterinary Journal 53, 3 1 5-20, 2005

The studies are presented as published, except that thei r format has been modified as indicated

in the Preface.

79

3.1 IDENTIFICATION OF CRYPTOSPOR/D/UM PARVUM 'CATTLE' GENOTYPE FROM A

SEVERE OUTBREAK O F NEONAT AL FOAL DIAR RHOEA

3.1 . 1 Summary

A severe outbreak of diarrhoea in Thoroughbred foals accompanied by shedding of

Cryptosporidium-l ike structures in the faeces of the affected an imals was i nvestigated. The

outbreak occurred in 2002, i n a com mercial Thoroughbred broodmare farm in the Waikato

region of New Zealand . N ine foals suffered from acute, m i ld, and severe disease accom panied

by dehyd ration and weakness. Despite intensive treatments, two foals died from the d isease

and a th i rd was euthanased d ue to severe condition . Six faecal samples from affected foals

were sent to a veterinary d iag nostic laboratory where they were tested for bacterial and vi ral

enteropathogens and for Cryptosporidium. Five faecal samples were positive for

Cryptosporidium. Postmortem examinations were performed on the euthanased foal and on the

dead animals. Gross f indings i ncluded extreme emaciation and di lated , thin-wal led, f lu id-fil led

intestines with no evidence of i nflam mation in the abdom inal cavity. Tissues and a faecal

specimen from the euthanised foal and three faecal specimens from other foals on the farm

were submitted to Massey Un iversity. Al l four faecal specimens were negative for Salmonella

species, Yersinia species, and group A rotavi rus VP6 protein . Microscopically, rou nd, acid-fast

structures resembl ing Cryptosporidium oocysts were seen in the four specimens.

Histopathological ly, numerous round, PAS-negative, Giemsa-positive, Cryptosporidium- l ike

organ isms, 2-5 11m in diameter, were present at the apical border of vi l lous epithel ia l cel ls

throughout the i ntesti ne.

Cryptosporidium isolates recovered from the faeces of three foals were subjected to a

mu lti locus genetic characterization . Sequence analysis of a 850 base-pair ampl icon from the

1 8S rRNA gene revealed sequences identical to each other and to the C. parvum 'cattle'

genotype sequences publ ished at the GenBank accession number AF093490. The mu lt i locus

genotyping showed restriction fragment length features consistent with C. parvum 'cattle'

genotype at the (3-tubu l in gene, the poly threonine repeat, COW P, and the R N R genes. This is

the f i rst report of an outbreak of cryptosporidiosis in foals which incorporated epidem iological,

cl inical and patho log ical data, as wel l as genetic characterization of the outbreak isolates .

3.1 .2 Introduction

The i ntesti nal protozoan parasite Cryptosporidium parvum has been intensively studied over the

past decade, due to its m ajor impact on human and an imal health. So far, two genotypes of C.

parvum have been extensively characterised : the 'human' genotype, primarily associated with

infections in man, and the 'cattle' genotype, found in human beings and also in domestic

livestock such as catt le, sheep and goats (Morgan et al. 200 1 ; Akiyoshi et al. 2002) . The genetic

divergence between the two genotypes manifests at a number of genomic loci , includ ing the

1 8S rRNA gene (Xiao et al. 2000), the Cryptosporidium oocyst wall protein (COWP) gene

(Spano et al . 1 997) , the ribonuclease reductase ( R N R ) gene (Widmer et al . 1 998) , the

80

polythreonine repeat (poly T) (Carraway et al . 1 997) and the r3-tubul in gene (Caccio et a l . 1 999) .

Extensive with in-genotype genetic heterogeneity also exists (Widmer et al . 2002; Mal lon et a l .

2003ab) . Equ ine cryptosporidiosis was in i t ia l ly described i n immunodeficient Arabian foals us ing

morphometric and morphologic parasitolog ical methods, fol lowed by descript ions of the disease

also in immunocompetent foals (Snyder et a l . 1 978 ; G ibson et al . 1 983; Gajadhan et a l . 1 985;

Coleman et al . 1 989) .

Whereas some surveys i ndicate that infections with Cryptosporidium species in horses are

relatively com mon, reports confirm i ng its role in foal d iarrhoea are scarce and attem pts to

produce experimental disease have been unsuccessfu l (Tzipori and Campbell 1 981 ; Tzipori

1 983; Coleman et a l . 1 989 ; Xiao and Herd 1 994; Netherwood et a l . 1 996; Cole et al. 1 998) .

Moreover, the published data on the genetic m akeup and biology of the equ ine isolates i s scant

and, strictly speaking, even their taxonomy with in the genus of Cryptosporidium is sti l l

unresolved.

This study describes the resu lts of an i nvestigation of a severe outbreak of diarrhoea i n

thoroughbred foals, accom pan ied by the shedding of Cryptosporidium- l ike structu res in the

faeces of the affected ani mals. To the author's knowledge, this is the fi rst report of an outbreak

of cryptosporid iosis in foals, wh ich incorporates epidemiological, cli nical and pathological data,

as well as the genetic characterisation of the outbreak isolates.

3.1 .3 Materials and Methods

Outbreak characteristics. Du ring the peak of the foal ing season in 2002, there was a severe

outbreak of foal diarrhoea in a commercial thoroughbred broodmare farm located in the Waikato

region of New Zealand. According to the practit ioner in charge, the outbreak lasted for

approximately one month and, dur ing that period, nine foals suffered from acute, mi ld to severe

d isease accompanied by dehydration and weakness. Approximately 30 foals were born on the

farm dur ing the same period. The i ndex case and six other foals were aged between four and

n ine days at the onset of d iarrhoea. The other two manifested the disease when they were three

weeks of age, which was seven to 1 0 days after returning to the farm from a regional neonatal

i ntensive care u nit where they were sent soon after birth due to unrelated condit ions. During the

outbreak, two foals died f rom the d isease and a third was euthanised due to its severe c l in ical

condition despite intensive treatments, which included intravenous fluids, broad-spectrum

antibiotics ( including metronidazole) , gastric protectants and anti-ulcer medication. Fresh cow

colostrum was also g iven via a nasogastric tube to some of the foals as an external source of

immunoglobu l ins . None of the affected foals was an orphan or on a foster m are . All mares had

colostrum tested at the t ime of foal ing and had adequate readings on a refractometer, with

colostrum immunoglobul in G ( lgG) levels corresponding to approx imately 60 g/litre. The serum

lgG levels of three affected foals were more than 800 m g/d l , indicat ing normoglobulinaemia . A

fourth foal had an lgG level of 640 mg/d l , attributable to fa i lure of passive transfer or to the

8 1

consumption of antibody over the course of the disease. Beef cattle co-existed with the foals

and adu lt horses on adjacent paddocks of the same farm . Yearl ing cattle arrived every year

from a constant external source and no cows calved on the farm . Some paddocks used by

horses during the outbreak had been previously grazed by cattle. Foals aged up to one week

were normal ly housed in a barn with a h igh an imal turnover but, during the outbreak, all new

foals were subsequently stabled in an alternative bui ld ing and a strict isolation of the sick

an imals was promptly implemented. Six faecal samples from affected foals were sent to a

commercial veterinary diagnostic laboratory where they were tested for bacterial and v i ral

enteropathogens and Cryptosporidium. One of the samples had Gram-posit ive rods resembl ing

Clostridium species, but it was negative for Clostridium perfringens on culture. Test ing for

Clostridium difficile toxin was not performed. Five faecal samples were positive for

Cryptosporidium oocysts. Other bacterial or viral pathogens were not detected. Postmortem

examinations were undertaken on the euthanased foal and on the two that died. Gross f i ndings

i ncluded extreme emaciation and di lated th in-walled , f lu id-f i l led intesti nes with no evidence of

inflam mation in the abdom inal cavity.

Laboratory methods. Pieces of small and large intestines, kidney, lung and liver from the

euthanased foal were col lected in 1 0 per cent formal in and subm itted to Massey University,

along with faecal samples from the euthanased foal and from three other affected foals wh ich

were sti l l al ive. The tissues were processed routinely for h istopathology and al l sections were

stained with haem atoxyl in and eosin . Sections of intestine were also stai ned with periodic acid­

Sch iff (PAS) and Giemsa. The faecal samples were tested for the presence of several

pathogens. For Salmonella species, the faeces were directly inocu lated onto Xylose Lysine­

dehoxycolate (XLD) plates and i ncubated at 37"C for 24 hours, or placed i n selen ite enr ichment

broth for 24 hou rs at 37"C with subsequent subcu ltur ing onto XLD and incubation as before. For

Yersinia species, faeces were directly i nocu lated onto selective-differential media inc luding

cefsu lod in , i rgasan and novobiocin (GIN) and incubated at 37"C for 48 hours. These were

placed in tubes containing phosphate-buffered sal ine (pH 7,3) , and incubated up to 1 4 days at 2

to 4 "C, with subsequent subcultures onto G I N plates , and then incubated as above. Virological

procedures i ncluded test ing for the faecal VP6 protein of g roup A rotavirus with com mercial

immunochromatographic kits, accord ing to the manufacturer's instructions (Rota-Srip ; Goris

BioGoncept, Gembloux, Belg ium) . For Cryptosporidium species, a method based on faecal

oocyst sed imentation by centrifugation, fol lowed by acid-fast stain of smears and screening with

a l ight m icroscope at x400 was applied (the method is described in Section 3.4) .

After completion of the diagnostic tests, the faecal specimens were m ixed with equal volu mes of

2 per cent potassium dichromate and conserved at 2 to 4 "C for further genetic analyses. Two

months later, Cryptosporidium isolates recovered from three faecal specimens were subjected

to a multilocus genetic characterisat ion. Oocysts were concentrated from the faeces and most

of the faecal debris was removed by the formal saline/ether method (Al ien and Rid ley 1 970) .

82

Methanol-fixed smears of the oocyst concentrates were stained with an immunofluorescent

m onoclonal antibody, according to the manufacturer's instructions (Merifluor C/G, Medir idian

Bioscience , C incin nat i , Ohio, USA), and exam i ned with an epifluorescent microscope at x200.

The smears were positive for Cryptosporidium species and negative for Giardia species. To

isolate the oocysts from the rem ain ing faecal debris, an immunomagnetic separation kit was

used according to the manufacturer's instruct ions ( Dynal®, lnvitrogen, Carlsbad, CA, USA) .

DNA was extracted by suspend ing the oocysts in 1 00 !J I of TE buffer ( 1 0mM Tris hydroch loride ,

1 m M disodium ethylenediamine tetra-acetic acid , pH 8·5) containing 1 per cent Non idet P40.

To this, 20 !J I of a 20 per cent suspension of Che lex 1 00 resin (Biorad Laboratories, Hercu les,

CA, USA) was added before f ive freeze/thaw cycles of two m inutes in l iquid n itrogen and two

m inutes in 95 "C water. Ce l l debris was removed by centrifugation at 1 0,000 g for three m inutes

and the supernatant was col lected and stored at 4 "C.

Each isolate was characterised at five genomic loci . A nested PCR-restriction fragm ent length

polymorph ism (PCR- RFLP) analysis was applied at the 1 8S rRNA gene as described by X iao et

al . (2000) , followed by the sequencing in both di rections of the 850 bp f ragment of the

secondary amplicon using an ABI Prism 377 autosequencer (Perkin Elmer, Waltham , MA,

USA) . The 1 8S rRNA gene sequences were al ig ned with other Cryptosporidium 1 8S rRNA

gene sequences in Genbank us ing C lustal W software (Thompson et a l . 1 997).

PCR- RFLP methods were also applied at the �-tubul in gene (Caccio et al . 1 999 ) , COW P gene

(Spano et al . 1 997) , RN R gene (Widmer et a l . 1 998) and a Cryptosporidium poly threonine

repeat (Poly T) (Carraway et a l . 1 997) us ing the restriction enzymes Vspl (Promega, Madison ,

W l , USA), Ode! ( R oche, Auckland, NZ) , Rsal ( l nvitrogen, Carlsbad, CA, USA) , Tsp509 I ( New

England Biolabs, Ipswich, MA, USA) and Rsal ( l nvitrogen, Carlsbad, CA, USA) , respectively,

according to the m anufacturers' instructions. Restriction fragments were resolved on 3 .5 per

cent agarose and the gel was visual ised by ethid ium brom ide stain ing . The DNA from a

previously characterised C. parvum 'human' genotype isolate was i ncluded i n the analysis as a

contro l .

3.1 .4 Results

The four faecal samples were negative for Salmonella spp. Yersinia spp. and g roup A rotavirus

VP6. Microscopical ly, round, acid-fast structures resembl ing C. parvum oocysts were seen .

Histopatho logical ly, v i l l i with i n the duodenum and i leum were moderately atrophic and

occasional groups of vi l l i were fused . There were m i ldly increased numbers of lymphocytes and

plasma cel ls, with occasional neutroph i ls , i n the lamina propria. Numerous round, PAS­

negative, Giemsa-positive, Cryptosporidium-l ike organisms, 2 to 5 !Jm in diameter, were present

at the apical border of vil lous epithelial cells throughout the intest ine (F igure 3 . 1 ). These

organ isms were m ost numerous in the duodenum, where they extended to the base of crypts,

but a few organisms were present in the colon . With in the jejunum there were occasional focal

areas of vacuolation of epithel ia l cells at the t ips of the vi l l i . No other relevant h istolog ical

83

abnormalit ies were noted i n the other t issues examined. The lesions were consistent with

intestinal cryptosporid iosis (K im 1 990) .

Sequence analysis of the 850 bp secondary ampl icon f rom the nested PCR-RFLP of the 1 8S

rRNA gene revealed that the three isolates from the foals were identical to each other and to the

published sequences for the 'catt le' genotype (GenBank accession number AF 093490) . The

mu ltilocus PCR-RFLP analysis showed restriction fragments consistent with C. parvum 'cattle'

genotype at the {3-tubu/in gene, po ly T, COW P , and R N R genes (Table 3 . 1 ; Figure 3 .2) .

Figure 3.1 (a) Fused and atrophic duodenal vi l l i (arrow) and mildly increased numbers of mononuclear

inflammatory cells within the lamina propria. Haematoxyl i n and eosin, x 1 00. (b) Protozoal organisms

among microvil li on the luminal border of a duodenal v i l lus (arrow). Giemsa, x 600

Locus PCR length (in base- Restriction 'Human genotype' 'Bovine genotype' pairs) endonuclease fragments size ( in fragments size ( in

base-pairs) base-pairs)

Poly T 3 1 8 Rsal 3 1 8 45, 273

�-tubul in 592 Ode/ 592 1 78, 4 1 4

R N R 441 Tsp509 / 1 80, 21 0 1 0, 47, 7 1 ' 96, 1 07, 1 1 0

COWP 550 Rsa/ 34, 1 06, 1 25, 285 34, 1 06, 4 1 0 1 8S rRNA 832 Vspl 82, 86, 1 04, 560 82, 1 04, 645

Table 3.1 Restriction f ragment length polymorphisms between C. parvum "cattle' and 'human' genotypes.

The bibl iographic references for each locus are provided in the text.

84

(a) 1 2 3 4 5 6 7 8 9 10 (b) 1 2 3 4 5 6 7 8 9 10

... .. :o.1. � � � � ,_. a p.r ..

� � - ...... �.1 ...

� - - -

- ....

-_-"' :'...._ -

...

• = = � ,.,., - -

- .. ..,.., �i�J -· .... • M:' ...... �<....· . .,:.. ··�(.l•. -

Figure 3.2 PeR-restriction fragment length polymorphism analysis of P-tubul in, poly T, R N R and COWP

genes. (a) Lanes 1 and 6 50 bp ladder, Lanes 2 to 4 P-tubu lin , horse isolates, Lane 5 P-tubulin, human

genotype, Lanes 7 to 9 Poly T, horse isolates, Lane 1 0 Poly T, human genotype. (b) Lanes 1 and 6 50 bp

ladder, Lanes 2 to 4 RNR, horse isolates, Lane 5 R N R , human genotype, Lanes 7 to 9 COWP, horse

isolates, Lane 1 0 COW P, human genotype

3.1 .5 Discussion

The results of this study indicate the possibi l ity of the emergence and circu lation of C. parvum

'cattle' genotype in foals, which m ight cause, or be a co-factor of , severe , l i fe-threateni ng

outbreaks of diarrhoea in apparently immunocom petent, normog lobu l inaemic an imals . This

investigation did not i nclude the whole range of possible causes of diarrhoea i n newborn foals,

and it is impossible to rule out completely eo-morbidity with other causes. Nonetheless, the

epizootological features and the histopathological f indings were most consistent with

cryptosporidiosis.

In this case, as in an outbreak previously reported by Konkle et al . ( 1 997) , there was a

tem poral-spatial l ink between the horses and calves. Human and calf cryptosporidiosis due to

C. parvum 'cattle' genotype is commonly diagnosed i n the Waikato d istrict in New Zealand

( Learmonth et al . 200 1 ) , a d istrict also known for its i ntensive cattle-rearing i ndustry and

relatively h igh density of thoroughbred breed ing premises. The same characteristics probably

exist in other horse-rearing countries, yet, to the authors' knowledge, th is is the first report

confirming cryptosporidiosis in foals with identification of C. parvum 'cattle' genotype. This wou ld

suggest that the disease is being underd iag nosed in the f ield, or that the tests used by most

veterinary diagnostic laboratories, whi le being well establ ished for calves, are inadequate for

diagnosing cryptosporid iosis in foals. Alternatively, it is plaus ible that, at present, the

85

susceptib i l ity of horses to C. parvum 'catt le' genotype is l im ited to a relatively narrow spectrum

of pathogenic horse-adapted subtypes , and that disease would only m anifest fol lowing specif ic

sporadic host-parasite encounters. This m ay also account for previous unsuccessful attempts to

produce experimental d isease in foals with bovine isolates (Tzipori 1 983).

I n order to test this hypothesis, a battery of horse iso lates of C. parvum 'cattle' genotype from

different outbreaks needs to be characterised by mu lti locus genetic typ ing of high discrim inatory

power, which may be fol lowed by analys is of the populat ion genetic structure (Aiel lo et a l . 1 999,

Caccio et al . 2000, Mal lon et al . 2003a,b) . Such a study in horses wou ld a lso further the general

understand ing of the m olecular mechanisms of host adaptation in C. parvum.

86

3.2 GENETIC DIVERSITY AND ZOONOTIC POTENTIAL OF CRYPTOSPORID/UM PAR VUM

CAUSING FOAL DIARRHOEA

3.2.1 Summary

Section 3 . 1 reported the fi rst known report of an outbreak of cryptosporid iosis in foals which

included the identification of the isolates as C. parvum. The author postu lated that the

susceptibi l ity of horses to C. parvum may be l im ited to a relatively narrow spectrum of horse­

adapted subtypes , and disease would on ly m an ifest fol lowing specific sporadic host-parasite

encounters. Therefore , in the study presented in this section Cryptosporidium isolates from

diarrhoeic foals in New Zealand (n=9) collected during the above outbreak and in 2006-2007 ,

were in itial ly aga in identified as C. parvum. Then, the isolates were subtyped at two

polymorphic loci and compared with human (n=45) and bovine (n=8) isolates. Foal C. parvum

were genetical ly-d iverse, m arkedly s im i lar to human and bovi ne isolates , and carried G P60 /la

A1 8G3R1 al leles, i nd icati ng a zoonotic potentia l .

3.2.2 Introduction

Intestinal Cryptosporidium parasites, in part icular Cryptosporidium parvum, are common causes

of diarrhoea in humans and an imals worldwide. C. parvum is also a zoonotic species.

Diarrhoea! d isease is a common cl inical condition in newborn foals (Cohen 1 994, Knottenbelt et

al. 2005 ; Magdes ian 2005 ; Crabbe 2007) . W hi le results of a number of stud ies suggest that

intestinal carriage of Cryptosporidium is relatively com mon in horses (Xiao and Herd 1 994; Cole

et al. 1 998; Hajduseka et al. 2004 ; Majewska et al. 2004 ; Chalmers et al . 2005) , reports

confirming the role of Cryptosporidium in diarrhoea in foals are rare, and their pathogenic

potential in these hosts is sti l l debated (Wilk ins 2004; Crabbe 2007) . The idea of

Cryptosporidium i n foals being of low pathogen icity is difficult to reconci le with our previous

report of an outbreak of foal diarrhoea in which C. parvum was identified as the sole agent

(Section 3 . 1 ). Therefore, the author hypothesised that d isease in foals m ay be underdiagnosed,

or alternatively, may only man ifest following specific encounters with a narrow spectrum of

horse-adapted parasites (Section 3 . 1 ) . However, at that juncture the hypothesis of the

existence of horse-adapted C. parvum could not be corroborated, as h igh ly-discrim inatory

molecular tools for the subtypin g of the isolates were not widely available.

The aim of th is study was to test this hypothesi s using novel molecular tools. Therefore,

Cryptosporidium isolates collected in New Zealand from diarrhoeic foals in 2002, 2006 and

2007 were genetically identified and subtyped by sequencing of the polymorphic reg ions of the

sporozoite 60-kDa g lycoprote in (GP60) and the 70 kDa heat shock prote in (HSP70) genes. To

infer about the possible routes of transmission and the zoonotic potentia l , the i solates from foals

were compared with isolates from humans and cattle also isolated in New Zealand .

87

3.2.3 Materials and Methods

Parasites. Cryptosporidium-positive diagnostic faecal specimens collected in New Zealand

f rom foals (n=9) , humans (n=45) , and cattle ( n=8) , were used in this study. The specimens from

foals were collected dur ing the foal ing seasons of 2002 (n=3) , 2006 ( n=5) , and 2007 (n= 1 ) . The

specimens from 2002 orig inated f rom an outbreak of foal cryptosporid iosis in the W aikato

region, which has been described i n Section 3 . 1 . Four specimens from 2006 or ig i nated f rom an

outbreak of foal d iarrhoea i n a farm located i n the North Island (see Section 3 .3 ) . One of these

specimens orig inated from a foal that was hospital ised due to severe d iarrhoea, and the other

three were col lected a week later by the fi rst author, from 1 -2 weeks-old diarrhoeic foals

presented for examination dur ing a visit to the farm. The fifth specimen from 2006 and the

specimen from 2007 were diag nostic specimens from two and three weeks old d iarrhoeic foals,

and were donated by a com m ercial laboratory operating in the North I s land. No addit ional

i nformation was provided on these sporadic specimens. The human specimens were col lected

between 2001 and 2003 as p art of a different study ( Learmonth et al. 2004) . As only C.

parvum was identif ied in foals (see below) , C. parvum-positive human specimens were used.

The bovine specimens orig inated from diarrhoeic calves on eight farms, and were col lected

between August -October 2006 by a veteri nary laboratory operating in the North Island . Al l

faecal specimens were stored between 2-4 "C, with no preservatives. Faecal speci mens

collected in 2002 were preserved in 2 per cent potassium dichromate (Section 3.1 ) .

Cryptosporidium identificatio n and subtyping . The identification of human C. parvum and

foal C. parvum col lected in 2002 has been previously done and described by the late J im

Learmonth and eo-workers of the Protozoan Research Un it, Massey University (Learmonth et

a l . 2004; Section 3 . 1 ) . The identification of Cryptosporidium from 2006 and 2007 was performed

us ing a nested PCR followed by sequence analysis of a -850 base-pair segment of the 1 8S

rRNA gene (Xiao et al. 2000 ) . Genomic DNA of the specimens from foals from 2006 was

extracted from the oocysts as described in Section 3 . 1 , with a modification consist ing of the use

of three freeze/thaw cycles of one m inute in l iquid nitrogen and water at 95 "C. Genomic DNA of

bovine specimens and the foal specimen from 2007 was extracted using a DNA extraction kit

(DNA Stool Min i Kit , Oiagen, H i lden , GmbH) . The nested PCR was performed in a thermo­

cycler (GeneAmp 9700 Applied Biosystems, Foster City, CA), using the pr imers 5' -GTT AAA

CTG CGA ATG GCT CA -3' (forward) and 5' -CCA TTT CCT TCG AAA GAG GA-3' (reverse)

for the primary amplification of a - 1 325 base-pairs reg ion . The 20 111 reaction m ixtu re consisted

of 2 111 of 1 OX PCR buffer ( ln vitrogen, Carlsbad, CA, USA), 6 m M MgCI2 , 250 mM (each)

deoxyribonucleoside triphosphate, 0.2 mM (each) pr imer, 2.5 un its of P lati num® Taq

( l nvitrogen , Carlsbad, CA, USA) , and 1 �I of DNA template. Thermo-cycl ing consisted of a hot

start at 96"C for two minutes fol lowed by 35 cycles of 94 "C for 30 sec. , 55 ° C for 30 sec., 72° C

for 45 sec. and a f inal extension of 72° C for 5 m in . The prim ary PCR product was di luted 1 : 1 0

with water prior to a secondary amplif ication with the primers 5'-CTC GAG TTT ATG GAA GGG

TTG-3' and 5'- CCT CCA ATC TCT AGT TGG CAT A -3'. Wi th the exception of 3 m M MgCI2

88

the PCR and cycl ing condit ions were the same as the prim ary round. This secondary PCR

product was 850bp in length ( Xiao et al . 2000) . Molecular-grade water and a C. parvum-positive

specimen from a calf were used as negative and positive controls . Amplicons were

electrophoresed in 1 .5% agarose, then stained with eth id ium brom ide and visual ised under UV

lig ht. Products o f the expected size were purified on a column (H igh Pure PCR Product

Purificat ion Kit, Roche D iag nostics, Man n heim , GmbH) and the DNA concentration was

measured us ing a spectrophotometer (NanoDrop 2000, Thermo Fisher Scientif ic, Wi lm ington ,

DE, USA) . Amplicons were then sequenced in both directions using a n ABI 3730 DNA analyser

(Applied Biosystems, Foster City, CA) .

Complementary sequences were alig ned and edited ; m argi nal segments that could not be

accurately determined were tri mmed and the f inal sequences aligned with Cryptosporidium 1 8S

rRNA gene sequences in our database us ing ClustaiX software (Thomson et al . 1 997) . For the

subtyping, genomic DNA was extracted using extraction kits as above. Subtypi ng was ach ieved

by means of sequencing two polymorphic loc i . The f i rst was a - 830 base-pairs region of the

Cryptosporidium sporozoite G P60 gene . Th is locus was ampl ified by Jim Learmonth using the

nested PCR described by G laberman et al. {2000), except that the anneal ing temperature of the

primary PCR of the present assay was 60 "C. The second locus was a - 470 base-pairs reg ion

of the HSP70 gene that comprises a polymorphic repeat (Khramtsov et a l . 1 995; Mal lon et a l .

2003ab) . Ampl ification of th is locus was performed using a reaction developed by Mr . E rrol

Kwan , Protozoa Research U nit, Massey Un iversity.

The PCR mixture included 200�M of the primers 5'- CACCATCCAAGAACCAAAGG (forward) ,

and 5 ' - GCCTAAAGGTAGAGTGTGCTTITC ( reverse) , 1 xPCR buffer ( l nvitrogen, Carlsbad ,

CA) , 1 .5mM MgCI2, 1 m M d NTPs (Fermentas Lifesciences, GmbH) , and 1 un it of taq

polymerase ( Platinum® Taq, l nvitrogen , Carlsbad, CA) , in a final vo lume of 20�1 . Thermo­

cycl ing consisted of 1 cycle of 96"C for 2 m i ns , 572C for 2 mins, and 72 "C for 2 mins, then 40

cycles of 94 "C for 30 sec, 57 "C for 30 sec, and 72 "C for 30 sec, and a f inal elongation step of

72"C for 2 mins . Water was used as a negative control in each batch test ing . G P60 and HSP70

amplicons were electrophoresed, purified, and sequenced as above.

Final sequences were alig ned with publ ished G P60 and H SP70 gene sequences (Khramtsov et

al. 1 995; Strong et al. 2000) . Due to the conserved nature of the HSP70 protein among

eukaryotic organisms, the possibi lity that H SP70 genes f rom other organisms were ampl ified

was checked by clusteri ng the sequences on line with s imi lar sequences deposited in GenBank

using the neig hbor-join ing cluster ing algorithm of B LAST

(http ://www. ncbi . n lm .nih .gov/blast/Biast.cg i , accessed on January 2008) (provided by The

National Center for Biotech nology Information of the National I nstitutes of Health , USA) .

89

I n it ial ly, human and bovine Cryptosporidium were not sequenced at the HSP70 locus. However,

isolates having G P60 sequences identical to the sequences in foal isolates were chosen at

random and subtyped at the HSP70 locus, allowin g a comparison between bi locus sequence

types (BLST) from different hosts.

3.2.4 Results

All the 1 8S rRNA gene sequences of foal and bovine Cryptosporidium were i ndisti ngu ishable

from the C. parvum sequence deposited in GenBank u nder accession number AF093490 (X iao

et a l . 2000) . One foal sequence from 2006 in it ial ly d iffered by 1 base-pair, but re-extracted DNA

yie lded a sequence i ndistingu ishable from AF093490 . S ix foal specimens amplif ied at both the

G P60 and HSP70 loci , one at the G P60 locus on ly, one at the HSP70 locus only, and one cou ld

not be amplified at either locus. The edited G P60 and HSP70 sequences were longer than 71 0

and 420 base-pairs, respectively, and comprised the repeat reg ions of both loci . Al l foal , human

and bovine C. parvum had G P60 /la A 1 8G3R1 al leles accord ing to the nomenclature suggested

by Peng et al. (200 1 ) and Sulaiman et al. (2005) . Th is nomenclature consists of the letter 'A'

fol lowed by the number of TCA codons, the letter 'G' fol lowed by the number of TCG codons,

and the letter 'R ' fol lowed by the number of ACATCA sequences at the end of the polyser ine

repeat.

There were five d ifferent GP60 sequences in foal C. parvum, differing by s ing le nucleotide

polymorphisms (SNP) but exhibiti ng more than 99% simi larity with each other (Tables 3.2 and

3 .3 ) . The foal C. parvum G P60 sequences observed in this study have been deposited in

GenBank under accession numbers E U483074 to E U483080.

There were five different HSP70 sequences C. parvum from foals, which differed with each

other by SNPs and from the sequence reported by Kh ramtsov et al . ( 1 995) by the number of

m inisatell ite repeats and/or SNPs external to the repeat. Using Blast software, these sequences

clustered only with C. parvum sequences. In each outbreak, at least one isolate differed from

others at both loc i . E ight subtyped foal C. parvum defi ned seven genetic variants (Table 3 .3 ) .

Identical GP60 sequences were found in three C. parvum from foals, 4 1 /45 from humans, and

the e ight from calves . Ten human C. parvum and the eight bovine C. parvum with th is G P60

al le le were also sequenced at the HSP70 gene, reveal ing the same BLST in two foal , 1 0 human

and seven bovine C. parvum.

90

Bases Cod on

position* change

4 1 9 - 42 1 TCA t o TCG

425 - 427 TCA to CCA

563-565 TCT to TCC

629-631 CAA to CGA

647-649 ACC to ccc

671 -673 AAA to AGA

698-700 ATG to GTG

9 7 1 -973 ACC to GCC

1 070- 1 072 AGA to AGG

Deduced amino acid

change

serine to serine

serine to prol ine

serine to serine

g lycine to argin ine

t h reonine to prol ine

lysine to argin ine

metion ine to val ine

threon ine to alanine

arg i n i n e to argin ine

C. parvum positive specimen in which the polymorphism was

seen

1 -6 ; 9

5

4

6

6

5

5

4

3

Table 3.2 Single nucleotide polymorph isms and deduced am ino acid changes in the 60-kDa g lycoprotei n

o f C. parvum isolates from foals , a s compared with the seq uence reported b y Strong et a l i n GenBank

accession num ber AF022929. D ifferences i n the number of se rine repeats are not reported.

* Accord i ng to Strong et a l . (2000) .

91

Isolate number*

1 I A (foai!W aikato/02)

2/1 (foai/Waikato/02)

3/2 (foai/Waikato/02)

4/7 4 (foai/Manawatu/06)

5/73 (foai/Manawatu/06)

6/48 (foai/Manawatu/06)

7/72 (foai/Manawatu/06)

8/47 (foai/Waikato/06)

9/569 (foai/W aikato/07)

1 0- 1 9 (human)

20-26 (bovine)

27 (bovi ne)

G P60 allele designation

2

3

4

5

ON

ON

HSP70 allele B LST

designation designation

2 2

3 3

3 4

ON NO

4 5

ON N O

5 N O

2 2

2 2

2 2

N O N O

Table 3.3 Bilocus sequence types (BLST) of foal , human, and bovine Cryptosporidium parvum i n New

Zealand. Arbitrary numbers designate alleles and BLSTs. Isolates 9/569 and 8/47 are from the sporadic

cases. BLST, b i locus sequence type; DN, did not ampl ify; GP60, C. parvum 60-kDa su rface glycoprotei n ;

H SP70, C. parvum heat-shock protein 70 gene ; N O : not determined. * : these isolate numbers were used

a lso in GenBank

3 .2.5 Discussion

This is the fi rst report describing the genetic d iversity of Cryptosporidium parasites caus ing foal

d iarrhoea. The key f indings of this study were the high genetic diversity of foal C. parvum and

their s imi larity with human and bovine isolates.

Based on the identificat ion of a novel partial 1 88 rRNA gene sequence in a Przewalski's wild

horse ( Equus przewalskit) i n a zoo (Ryan et al. 2003), X iao and Feng suggested that horses are

i nfected with a Cryptosporidium 'horse genotype' (Xiao and Feng 2008) , which has never been

identified in other hosts. In addit ion, results of recent molecular studies support the possib i lity of

the existence of host-restriction in C. parvum ( Mal lon et al. 2003a,b; Xiao and Feng 2008) .

92

Thus, the mere identificat ion of C. parvum in the 2002 outbreak did not allow conclusions to be

drawn about the or ig in or zoonotic potential of foal Cryptosporidium. In this study, the isolates

were identified , subtyped, and then compared with human and bovine iso lates.

Conform ing with the superdiverse genetic structure described for C. parvum populations i n

Sect ion 2. 1 , eight foal isolates cou ld b e subdivided into seven genetic variants . Nonetheless,

the dominant /la G P60 allele and BLST were shared by the three host species. G P60 /la al leles

are also highly prevalent in human and bovine C. parvum in other countries (Sula iman et al.

2005; Leoni et al . 2007) . Therefore, the genetic repertoire of foal, bovine and human C. parvum

largely overlap, and so foal C. parvum should be considered potential ly zoonotic.

Lastly, in accordance with the resu lts of waterborne outbreak investigat ions in hu mans (Mathieu

et al . 2004; G laberman et a l . 2000; Leoni et a l . 2007) , we identified different C. parvum al leles

among foals in two outbreaks, suggest ing that genetical ly-heterogeneous parasite assemblages

m ay be involved in outbreaks of cryptosporidiosis. This feature may hamper our abi l ity to track

infectio n sources using molecular tools.

93

3.3 A STU DY OF NEONATAL CRYPTOSPORIDIOSIS OF FOALS IN NEW ZEALAND

3.3.1 Summary

Sections 3 . 1 and 3.2 dealt with the molecu lar characterisation of Cryptosporidium parasites

causing neonatal diarrhoea in foals. The study presented in the present Section deals with the

cl in ico-epidemiological aspects of neonatal cryptosporidiosis in farmed foals in New Zealand.

The Cryptosporidium isolates collected dur ing th is study were included in the study reported in

Section 3 .2 .

To assess the occurrence of Cryptosporidium oocysts i n d iagnostic faecal specimens f rom

foals, selected specimens received by a di agnostic veterinary laboratory i n New Zealand

between 2006 and 2007 were subm itted to Massey Un iversity (MU) and tested microscopical ly

for the presence of Cryptosporidium oocysts. Then , the Cryptosporidium parasites were

genetical ly identified to taxon level . I n addit ion, specimen subm ission data from the

participat ing laboratory for 2005-2007 were examined.

Twelve faecal speci mens from diarrhoeic foals were transferred to MU between 2006 and 2007,

from which three tested posi t ive for C. parvum. One Cryptosporidium-positive specimen

identified i n the course of the study orig inated from a foal that was hospital ised due to severe

diarrhoea. This case tr iggered an on-site investigation of the broodm are farm that i t came from,

which revealed a high incidence of neonatal diarrhoea in foals. Four affected foals were

examined during two visits to the farm . In all four cases, the d isease was self- l imit ing and

man ifested during the second week of l ife, rasembl ing foal heat d iarrhoea. A s imi lar c l in ical

manifestat ion had been observed by the owner and the veteri narian in charge in most of the

foals born on the farm during the foal ing season . Cryptosporidium parvum was the on ly

enteropathogen found i n the faeces of the 4/4 affected foals examined. The oocyst shedding

was of short du ration and intense in al l cases, reminiscent of cryptosporid iosis i n calves.

Specimen submission records for 2005-2007 indicated 67 faecal specimens were tested for

Cryptosporidium by the part ic ipat ing laboratory, and 1 2 ( 1 8%) tested positive.

3.3.2 Introduction

Numerous microorganisms, parasites, and non-infectious factors have been associated with

neonatal d iarrhoea in foals ( Magdesian 2005) . However, the causal role of many of these

agents is u ncertain, and it has been est imated that the aet iological diagnosis of diarrhoea i n

foals remains elusive in about 50% o f cases (Knottenbelt e t a l . 2005) .

Cryptosporidium protozoan parasites, i n particu lar C. parvum, are common causes of diarrhoea

in humans and young calves worldwide. Cryptosporidium parvum is also a zoonotic species

(Fayer 2008). Carriage of Cryptosporidium parasites appears to be relatively common in horses

94

(Tzipori and Campbell 1 98 1 ; Coleman et a l . 1 989; Xiao and Herd 1 994; Cole et al . 1 998;

Hajduseka et a l . 2004; Majewska et a l . 2004; Chalmers et al. 2005) . However, reports

confirm ing the role of Cryptosporidium in d iarrhoea in foals are scarce, and their pathogenic

potential in immu nocompetent an im als has been debated (Wilkins 2004; Crabbe 2007) . The

concept of low pathogenicity for Cryptosporidium i s also supported by early, unsuccessful

attempts to produce disease in foals using isolates from calves (Tzipori 1 983} , and by at least

one study which found no association between the presence of Cryptosporidium oocysts and

d iarrhoea in horses (Xiao and Herd 1 994} . S im i larly, in a case control study performed in the

Un i ted Kingdom, Netherwood and eo-workers ( 1 996) found Cryptosporidium oocysts in 1 7% of

d iarrhoeic foals and 20% of controls , but a statistically sig n ificant association between the

presence of the oocysts and d iarrhoea in only one of the mu ltivariate statistical models

exami ned. Further, data from the United States and the United Kingdom i ndicate that testing for

Cryptosporidium is not widely performed by d iag nostic laboratories, and the agent is not

consistently i ncluded in the differential d iag nosis of diarrhoea in foals in epidemiolog ical studies

(Anonymous 2007a,b,c, 2008a,b; Hol l is et al. 2008 ; Roberts et al. 2008; Woh lfender et al.

2009) .

The concept o f a low pathogenicity o f Cryptosporidium i n foals is difficult to reconcile with the

previous report of an outbreak of d iarrhoea in foals in New Zealand in which C. parvum was

identified as the sole agent (Sect ion 3 . 1 ) . In the present section, the results of a survey of the

occu rrence of Cryptosporidium parasites in d iag nostic faecal specimens from diarrhoeic foals

subm itted to a diagnostic veterinary laboratory in New Zealand are presented. The results of the

survey, and of an investigation of an outbreak of cryptosporidiosis in foals triggered during the

same survey, are discussed in view of the author's previous f indings.

3.3.3 Materials and methods

Study outline. Between 2005 and 2007, a col laborative study between Massey University (MU)

and a diagnostic laboratory operat ing i n the North Island of New Zealand was conducted. The

study a im was to assess the occurrence of Cryptosporidium oocysts in diagnostic faecal

speci mens from foals, and genetical ly characterise Cryptosporidium iso lates. Therefore, a

subset of faecal specimens from foals received in 2006 and 2007 by the participating laboratory

for a m icrobiolog ical analysis for causes of diarrhoea was transferred to MU . The specimens

were transferred at the discretion of the pathologist in charge, provided faecal m aterial was

ava i lable after the completion of the requested tests, and regardless of whether or not the

test ing for Cryptosporidium had been i n it ial ly requested . In addition, the participat ing laboratory

provided specimen submission data for 2005, 2006, and 2007. The results of the genetic

characterisation of the i solates has been reported in Section 3 .2 .

On-site outbreak investigation. I n November 2006, large numbers of Cryptosporidium

oocysts were seen on d i rect faecal smears from one of the specimens transferred to MU . The

95

specimen originated from an 8-day-old Thoroughbred foal ( index case) that had been

hospitalised at MU due to severe diarrhoea. The foal was born on a broodmare farm in the

lower North Is land. The referring veterinarian reported that there had been a high i ncidence of

neonatal d iarrhoea on the farm, which prompted an outbreak investigation .

The farm was visited twice, on 5 and 12 December 2006 . Three diarrhoeic foals (aged 6, 8 , and

1 1 days at the first visit) and their dams, were presented for examination dur ing these visits.

Faecal and environmental specimens were col lected at each visit, and rectal faecal specimens

collected from the foals during the fi rst ( n=3) , and second (n=3} visit to the farm, and from thei r

mares (n=3) dur ing the fi rst visit . Faecal specimens from the hospital ised foal were taken 1 , 4 ,

5 , 6 , 8 , and 1 1 days fol lowing admission (n=6} . In add ition , a sample of bedd ing material from

the foal ing shed, and muddy soil from a paddock used to house mares and foals, were

col lected . All the samples were analysed for Cryptosporidium and other com mon

enteropathogens of foals.

Laboratory methods. In order to increase the sensitivity of the test for the detection of

Cryptosporidium oocysts, the faecal spec imens submitted to MU were tested by means of two

methods, with parallel i nterpretation . The fi rst method included the s imple stain ing of d i rect

faecal smears using both the modified Ziehi-Neelsen (ZN) stain and a bivalent com mercial

immunofluorescence test kit for Cryptosporidium and Giardia (Mer iF iuor Crypto & G iardia;

Merid ian Bioscience, Cincinnati OH, U SA) . The stained smears were examined microscopically

at 400x m agn ification , and were considered positive for Cryptosporidium i f there were one or

more round acid-fast oocysts of 4-6 �m diameter contai n ing internal sporozoites (Ortega and

Arrowood 2003}, or apple-green f luorescent oocysts. The specimens that tested negat ive by

this method were then subjected to an additional test, which d iffered in that it included

separation of the oocysts from faecal debris by sed imentat ion prior to the preparat ion of the

slides as previously described (Grinberg et a l . 2002; see also Section 3 .4} .

The specimens col lected from the hospita l ised foal one day after adm ission and from the foals

during the f i rst visit to the farm , were tested for the presence of Cryptosporidium and Giardia as

described above, and for ova of he lm inths, Salmonella spp. , and Group-A rotavirus. Testing for

Salmonella spp. was performed using an i nocu lation of a loopful of faeces onto xylose-lysine­

deoxycholate agar and into selenite enr ichment broth fol lowed by subcu lture onto xylose-lysine­

deoxycholate (Fort Richards Laboratories, Auckland, NZ) . Media were incubated for 24 h at

37"C in aerobic cond itions . Testing for Group-A rotavi rus was performed using a com m ercial

latex agg lut ination test kit (SerobactRotavirus ; Medvet, Adelaide, Austral ia) . The presence of

ova of helminths in the faeces was assessed m icroscopically after salt flotation by the

technician of the laboratory of parasitology of the Institute of Veterinary, Animal and Biomedical

Sciences, M U . Specimens from mares, subsequent specimens collected from the hospital ised

96

foal on Days 4, 5 , 6 , 8 and 1 1 fol lowing admission, and from the foals during the second visit to

the farm, were on ly tested for Cryptosporidium, as the other agents had already been excluded.

The samples of bedding materia l and soi l were on ly tested for the presence of Cryptosporidium

using the followi ng method : A 5 g sample from each source was suspended i n 50 m l tap water,

sieved through a coarse sieve , and centrifuged at 900g for 1 0 m inutes. The deposit was re­

suspended in 3 m l water, and tr ipl icates of 1 0 1-11 were deposited as drops on sl ides, stained for

imm unofluorescence, and screened microscopical ly as described above .

The taxonomic identificat ion of the Cryptosporidium parasites was performed by means of

sequence analys is of a species-specific reg ion of the Cryptosporidium 1 8S small subunit

ribosomal RNA ( 1 8S rRNA) gene, us ing the oocyst separat ion method, and nested PCR and

sequencing procedures described in Section 3.2 .3 . I n addition , DNA was extracted from the two

environmental samples using a DNA extraction kit ( DNA Stool Min i Kit ; Qiagen, H i lden ,

Germany) , and tested for the p resence of the Cryptosporidium 1 8S rRNA gene us ing the same

nested PCR and sequencing m ethods,

The number of Cryptosporidium o ocysts per ml of faeces was estimated in the first faecal

specimen col lected from the hospitalised foal , and in one specimen col lected from a diarrhoeic

foal during the fi rst visit to the farm. These specimens were thorough ly mixed , and tripl icates of

1 0 IJ I were seria l ly di luted in normal saline to a d i lution of 1 o-2. A 1 0 1-1 1 a l iquot of this d i lution

was deposited as a drop on a s l ide and stained usi ng the immunofluorescence kit. The oocysts

on the entire surface of the drops were counted using the 400x m agn if ication of an

epifluorescent m icroscope, and the number of oocysts per ml of faeces extrapolated by

mu ltiplying the mean count of the tripl icates by 1 o4. An oocyst count cou ld not be performed on

the other specimens from foals on the farm due to the need to preserve the smal l amount of

faecal material obtained for DNA extraction .

3.3.4 Results

Laboratory submission data for 2005-2007. A total of 1 3 1 specimens from foals with a

history of diarrhoea and/or enteritis were received by the participati ng laboratory between 2005-

2007. I n many cases the specimens arrived with a request for testing but with no cl i nical

history ind icated. Routine testin g for Cryptosporidium was only performed if specifically

requested by the submitt ing veterinarian or patholog ist in charge, by means of a ZN sta in of

faecal smears. Only 67 speci mens were tested for Cryptosporidium and 1 2 ( 1 8%) tested

positive. Twelve l iquid faecal specimens, not all i n itia l ly tested for Cryptosporidium by the

participating laboratory, were transferred to MU in 2006 and 2007, and were used in this study

in addition to the specimens col lected during the on-farm investigat ion . Three of these

specimens tested positive for Cryptosporidium at M U . Two of the Cryptosporidium-positive

specimens were col lected from foals in the Waikato aged 2 and 3 weeks, in 2006 and 2007; no

97

addit ional information o n these specimens was avai lable. The th i rd Cryptosporidium-positive

specimen or ig inated from the h ospitalised foal . Sequence analysis revealed that these i so lates,

and the additional isolates col lected during the o n-farm investigat ion (see below) , were C.

parvum (the molecular ident ification of these parasites is reported in Section 3 .2) .

Findings from the on-site investigation. The index case had a normal parturition and nursing

h istory; 700 ml of the m are's colostrum was admin istered by stomach tube to the foal 1 hour

after birth and the foal stood and nursed normally. Twelve hours pr ior to admission, the foal

developed watery, m alodorous, non-haemorrhagic d iarrhoea. No abnormalit ies had been seen

in either the mare or foal prio r to this, and the mare had not shown any recent oestrous

behaviour. The foal developed mi ld colic s igns about 4 hours later, m anifested by rol l i ng and

flank watching. At th is time, lactated R i ngers solution ( 1 L , I V) , f lunixin meglum ine ( 1 OOm g IV) ,

and trimethoprim and su lphad iazine ( 1 440mg combined dose, IM) , were adm in istered by the

referring veterinarian . The foal rem ained bright but did not nurse, and no i mprovement in the

signs of colic was seen . Two hours before admission , hyoscine-n-butylbrom ide (40mg I V) and

dipyrone (5000mg IV) were admin istered to no effect. On admission, the foal was depressed,

and showed frequent f lan k-watching behaviour. Heart and respiratory rates and rectal

temperature were with in normal ranges. Intestinal sounds were norm al, but there were frequent

episodes of watery diarrhoea and faecal sta in ing of the perineum. Ultrasonographic exam ination

of the foal 's abdomen showed m i ld gassy d istension of the large intest ine and hypermoti l ity of

the sm all i ntestine. Umbi l ical rem nants appeared normal . I n itial treatment i ncluded hydration

treatment with IV boluses of lactated R inger's solution supplemented with 2 .5% g lucose, and

oral b ismuth subsalicylate. A faecal specimen submitted for bacteriolog ical culture to the

participating diagnostic laboratory at admission revealed a l ight growth of a mixture of G ram­

positive cocci and Gram-negative rods, but n o pathogenic bacteria were isolated .

One day after adm ission , a faecal specimen was subm itted to M U for an analysis for

Cryptosporidium oocysts . At MU , numerous Cryptosporidium- l ike oocysts were seen on d i rect

smears from this specimen and stained by both the modified ZN and the immunofluorescent

stains. The same specimen was negative for Salmonella spp . , Giardia, G roup-A rotavirus, and

ova of he lminths. Two days after adm ission, isotonic f lu ids were administered via stomach tube,

and omeprazole and l ive yoghurt treatments were started. Over the followi ng 3 days the foal

stabi l ised and faecal consistency gradual ly i mproved . Blood samples taken at adm ission and in

the following days showed anaemia throughout the observation period, the exact cause of

which cou ld not be determined . A mi ld leucopenia, hyperfibri nogenaemia, and hyponatraemic

metabolic acidosis, and e levated levels of aspartate aminotransferase and glutamate

dehydrogenase were observed (Table 3.4) . Addit ional faecal specimens taken on Days 4 , 5, 6,

after admission were positive for Cryptosporidium. The foal was discharged 1 2 days fo l lowing

adm ission, after two consecutive faecal specimens (for Days 8 and 1 1 ) were negative for

98

Cryptosporidium. Skin excoriations as a resu lt of the diarrhoea were present on the perineal

reg ion and haunches on the day of discharge ( Figure 3.3) .

Units

RBC x 1 012

HCT % Hb g/L PLT x 1 09

WBC x1 09

Neutrophils x1 09

Lymphocytes x1 09

Monocytes x1 09

Fibrinogen

lgG

Glucose

CK

AST

GGT

GLDH

Bile acids

Total protein

Albumin

Globulin

Urea

Creatinine

Phosphate

Total Ca

Na

K

Chloride

Venous pH

PVC02

VHC03

Anion gap

g/L

g/dl

mmoi/L

I U/L

I U/L

IU/L

IU/L

JlmOI/L

g/L

g/L

g/L

mmoi/L

JlmOI/L

mmoi/L

mmoi/L

mmoi/L

mmoi/L

mmoi/L

mm Hg

mmoi/L

mmoi/L

At admission

5.45*

2 1 .6*

69* 206

7 . 9

6 .6

0 .8*

0.6*

0.84

6.6*

45

5.7

1 28

3.4

1 02

7.26*

38.3*

1 7.2*

· 1 0*

1 2 h post­

admission

4.79*

1 9.4*

2 1 0

4.4*

3.2*

0.9*

0.3

6.2*

42

4.3

1 3 1

3.6

1 06

7.2 1 *

4 1 .9*

1 6.9*

· 1 1 *

36 h post­admission

4.78*

1 8.9*

232

6.2

5.0

0 .7*

0.5

50

6 days post­

admission

5.6*

22*

274

7. 1

4 . 1

2.4

0.6*

5.0*

1 93

446*

63

308*

1 5

53

28

25

1 .7*

52*

1 .86

2.94

1 3 1

4.2

1 2 days post­

admission

4.5*

1 8*

201

6.6

5.2

1 . 1

0 .3

0.98

Normal rang ea

6.9-1 1 .8

32 - 47

1 02-1 54 1 00 - 350

6 .2-1 2.4

4 . 1 - 9. 1

1 ·3 . 1

0 . 1 - 0.5

1 .5 - 4.2

0 .69- 1 .86

3 . 3 - 5.6

1 41 -4242

69-3 1 6

1 5- 63

1 -8

0 - 20

42 - 66

2 5 - 35

1 3 - 36

4 . 1 - 1 2.5

1 06 - 31 2

1 .2 - 2.2

2 .4 - 3.3

1 3 1 - 1 41

3 .0 - 4.6

93 - 1 05

7 . 36- 7.43

45 - 49

22.3 - 25

+4 - -4

Table 3.4. Haematology, biochemistry and venous blood gas results from a hospitalised foal affected with

cryptosporidiosis. Tests indicated by a dash were not performed.

a Normal ranges are those used by New Zealand Veterinary Pathology Ltd, New Zealand, for

Thoroughbred foals aged 0-3 weeks. Asterisk denote values outside normal range. RBC= red blood

cel ls; HTC= haematocrit; Hb=haemoglobin; PL T = platelets; WBC= white blood cel ls ; C K= Creatine kinase

AST = Aspartate aminotransferase; GGT = Gamma Glutamyl transferase; GLDH= Glutamate

dehydrogenase; PVC02: carbon dioxide partial pressure; VHC03: venous bicarbonate concentration

The broodm are farm was visited twice, on 05 and 1 2 December 2006. The farm was located on

agricu ltural land crossed by a stream . lt was a new business in its f i rst operational foal ing

season, established on land previously used to raise cattle and sheep. A smal l number of sheep

and weaned beef cattle were g razing on the same facility dur ing the visits.

99

Potable water was supplied to the animals, and they also had ful l access to a stream . Pregnant

m ares from different locations were transferred to the facility for foal ing and breeding one month

before the est imated date of foal ing , and returned to thei r farms 2 months after foal ing. Mares

were al lowed to g raze up unt i l 3-4 days before the expected foal ing date , when they were

moved to a foal i ng barn. Mares and neonates were moved back to a paddock as soon as

possible after foal ing . Forty-five foals, the m ajority of which were Thoroughbred , were born on

the faci l ity between August and November 2006. Diarrhoea had affected the majority of them

throughout the season. Most cases were self- l im i ti ng , with an onset at about 6-8 days of age

and a duration of 2-3 days. Some protracted cases were seen, and one foal died in October

after suffering from diarrhoea. Management of diarrhoeic foals included their transfer, with their

dams, to a sectio n of the foal ing shed also used for foal ing , as no other faci l ity for the sick

animals was avai lable (Figure 3 .3) . Treatment of d iarrhoeic foals i ncluded the admin istration of

oral electrolytes and a 3-day course of an oral co-trimoxazole.

Figure 3.3 Foals with cryptosporidiosis . Left: a foal and the mare on the farm ; Right: the hospitalised foal

at the day of discharge . Notice the presence of excoriat ions in the hau nches caused by the faecal soi l ing

d u ring the disease. These pictures were taken by BVSc student Abigai l Deuel

The three diarrhoeic foals presented for exam ination during the vis its to the farm were the last

neonates of the season. They appeared bright but their tails were soi led with faeces. On the

f i rst visit, one foal passed l iquid faeces, one a small amount of semi -l iquid faeces, and the th ird

d id not pass faeces , although a smal l amount of faecal material cou ld be extracted from the

rectum . Their dams passed ful ly formed faeces. The rectal temperature of the 1 1 -day-old foal

was sl ightly elevated (39.4 "C). As the other two foals were restless, their rectal temperatures

were not measured. The same three foals were bright and passed f i rm faeces on the second

vis i t .

Laboratory f indings. The fi rst faecal specimens collected from the foals on the farm were

negative for Salmonella spp . , Giardia spp, G roup-A rotavirus , and ova of helm inths, but had

large numbers of Cryptosporidium oocysts, which were seen on d i rect smears by both the

1 00

modif ied ZN and imm unofluorescent stai ns . There were > 1 o6 oocysts per m l of faeces in the

first faecal specimen collected from the hospital ised foal and in one specimen col lected from a

diarrhoeic foal dur ing the fi rst visit to the farm . Al l the Cryptosporidium isolates were identified

as C. parvum by sequence analysis , as described previously in Sect ion 3 .2 . No oocysts were

seen in the speci mens from mares and foals taken during the second visit to the farm . S imilarly,

no oocysts were seen in the environmental samples. However, a C. parvum 1 8S rRNA gene

sequence was ampl ified from DNA extracted from the soil sample, but not from the bedding

materia l .

3.3.5 Discussion

Cryptosporidiosis in foals is a poorly characterised d isease. In an earl ier outbreak in 2002, the

diagnosis of cryptosporidiosis was based on the iso lat ion of C. parvum as the sole agent, in

conjunction with conclusive histopatholog ical f indings (Section 3 . 1 ). In the outbreak described

here, post-mortem f indings were not avai lable. Nonetheless, the same diagnosis was supported

by the f inding of a h igh incidence of self- l imit ing diarrhoea manifest ing dur ing the f i rst and

second weeks of l ife, and in particular the f inding of the concom itant c l in ical and parasito logical

rem ission in the four foals, remin iscent of bovine cryptosporidiosis (Anderson 1 98 1 ; Uga 2000;

Grinberg et al. 2002) .

Accord ing to the data reported in Sections 3 . 1 and 3 .2 , C. parvum has been identified i n New

Zealand from cases of diarrhoea in foals from two separate reg ions, dur ing three non­

consecutive foal ing seasons. Furthermore, 1 8% of the specimens from foals tested for

Cryptosporidium by the participat ing laboratory in 2005-2007 were positive for this agent.

Col lectively, these resu lts suggest foal cryptosporidiosis caused by C. parvum is not uncommon

in New Zealand.

In New Zealand, m ost registered foals are born on agricultural land in reg ions with large

popu lations of cattle . M oreover, the foa l ing and calvi ng seasons generally overlap (Rogers et al .

2007) , and two independent studies converged i n ind icat ing Cryptosporidium parasites are

present in about 40% of the dai ry farms (Section 3 .4 ; Winkworth et al. 2008) . Th is cou ld

facilitate the transmission of C. parvum of bovine or ig in to foals as compared with other horse

rearing countries. In fact, C. parvum isolates or ig i nating from foals, calves and hum ans in New

Zealand are genetical ly-s imi lar (Sect ion 3 .2) . The f indi n g in the study presented here of a C.

parvum DNA signature in soil is evidence of a contaminated farm environ ment and supports this

view. However, cryptosporidiosis in foals may also be common in other cou ntries, as suggested

by the presence of oocyst- l ike structures in the faeces of foals affected by foal-heat d iarrhoea in

Italy (Sgorbini et al . 2003 ) .

i t i s i ntriguing that cryptosporid iosis is seldom reported in foals. There are remarkable

s imi larit ies , in terms of d isease onset, duration , and severity, between foal-heat d iarrhoea and

1 0 1

the cases of cryptosporidiosis described previously (Section 3 . 1 ) and herein . Foal-heat

d iarrhoea is a common self- l imiti ng condit ion of 1 -2 week-old foals that has a poorly

characterised aetiology (Magdesian 2005 ; Crabbe 2007), and it wou ld seem possible that

cryptosporidiosis in foals is underdiagnosed, or i n many cases managed empi rical ly as foal-heat

d iarrhoea. I ndeed, the smal l number of specimens from foals tested for Cryptosporidium by the

participating laboratory and overseas ind icates the parasite is not consistently considered in the

d ifferential d iag nosis by cl i n ic ians. The short patent period, as observed in th is study in 4/4

foals, could also contribute to the underdiag nosis by i ncreasing the l ikel ihood of the agent not

being present in specimens col lected after any de lay.

The presence of the hospitalised foal al lowed cl in ical and haematological parameters of

cryptosporidosis to be investigated, at least in th is case. In addition to the alterat ion in generic

markers of dehydration and i nflammation , the foal suffered from anaemia throughout the

observation period. However, as a simi lar c l in ical investigation cou ld not be performed in the

other foals, this f inding needs to be corroborated .

Some authors have suggested that horses are i nfected with a so-called ' Cryptosporidium horse

genotype' (Xiao and Fayer 2008 ; Xiao and Feng 2008; Xiao and Ryan 2008; see also Section

1 .6.3) . Based on a s ingle observation , which was not accompanied by any descript ion of the

i nfection with the same genotype i n the horse, the 'horse genotype' has been classified by

Fayer as a 'major' intest inal Cryptosporidium affecting the horse (Fayer 2008) . The putative

occurrence of the Cryptosporidium 'horse genotype' in equine popu lations is of significance, as

un like C. parvum, this genetic variant has not been identified in humans and therefore is not

considered zoon otic. However, whereas C. parvum has been repeatedly identified in the

domestic horse ( Equus cabal/us) in New Zealand and overseas ( Sections 3. 1 and 3 .2 ;

Chalmers et al . 2005; Tanriverdi et a l . 2006) , the 'horse genotype' has on ly been reported in

one Przewalski's wild horse ( Equus przewalskit) i n a zoo (Ryan et al . 2003) . Thus , the idea of

wide cycl ing of a Cryptosporidium ' horse genotype' in the domestic horse needs to be

substantiated, and people hand l ing d iarrhoeic foals should be mindfu l of the zoonotic potential

of cryptosporidiosis in foals caused by C. parvum.

I n conclusion , the results of this study complement previous observations and suggest the

incidence of c l in ically overt C. parvum i nfections in newborn foals in New Zealand m ay be

underest imated. The author postu lates that cryptosporidiosis may be underdiagnosed, perhaps

accounting for a proportion of cases empi rically d iagnosed as foal-heat d iarrhoea. it is advisable

to take hygienic precautions when handl ing diarrhoeic foals, unti l C. parvum, which is a

potential ly zoonotic agent, is ru led out i n the laboratory, preferably by mu ltiple test ing to

enhance the agent's detection .

1 02

3.4 THE OCCURRENCE OF CRYPTOSPORIDIUM PAR VUM, CAMPYL OBA CTER AND

SALMONELLA IN N EWBORN CALVES IN THE M ANAWATU REGION OF N EW ZEALAND

3.4.1 Summary

In 2002, a cross-sectional study was conducted dur ing the winter calv ing season, with the aim

of determin ing the rate of occurrence of Cryptosporidium oocysts in faecal specimens taken

from newborn dairy calves on 24 dairy farms in the M anawatu reg ion of New Zealand.

Faecal specimens were collected from the rectums of 1 85 newborn calves from 24 dairy farms

selected by convenience criteria . The specimens were tested microscopical ly for the presence

of Cryptosporidium parvum oocysts, and bacter iologically for the presence of Campylobacter

spp. and Salmonella spp. The identification of the Cryptosporidium oocysts to taxon level was

not performed, as at that junction C. parvum was considered the only intestinal species i nfect ing

newborn calves (th is m atter wi l l be further discussed in Section 3.5) .

I nfections with C. parvum were identified in 33/1 56 (21 . 1 %) calves from 1 0 farms. More than

1 o6 oocysts/gr of faeces were detected in calves from four farms. Campylobacter spp. were

isolated from 58/1 6 1 (36%) calves from 1 8 farms ; in particu lar, C. jejuni subsp jejuni was

isolated from 1 1 /1 6 1 (6 ,8%) calves from seven farms . Salmonellae were not detected .

These results indicate that despite the short, concentrated calving pattern and the long interval

between calving seasons characterising most dairy farms in New Zealand , C. parvum is

widespread among calves. Campylobacter spp, especial ly C. jejuni, colonise the i ntesti nal tract

of calves, early in l ife.

3.4.2 Introduction

The prevalence of human m icrobial enteropathogens in cattle has been the subject of intensive

research in many countries, due to the risk of an im al-to-human transm iss ion. I n New Zealand,

the protozoan parasite Cryptosporidium parvum and bacteria belonging to the genus

Campylobacter and Salmonella are among the m icrobial pathogens most com monly detected in

cases of human gastrointest inal i nfections (Anonymous 2004) .

A number of Cryptosporidium species have been reported to infect humans, but on ly the

anthroponotic species C. hominis (Morgan- Ryan et al . 2002) (formerly known as C. parvum,

'hu man ' genotype, o r Type I) and the zoonotic C. parvum (formerly known as C. parvum 'cattle'

genotype, or Type 11) are widespread. Other species have on ly been reported sporadical ly in

intestinal and non-intestinal infections in immunocompromised ind ividuals (X iao et al . 2004;

Xiao and Ryan 2008) . Cryptosporidium parvum is an obl igate intestinal parasite and has a wide

host range (Fayer 2004) . The parasite is widespread in l ivestock and it is believed that l ivestock,

especial ly cattle, fu nction as natural reservoi rs for human infect ions. Perhaps the most

convincing data support ing th is belief were the 35% reduction in laboratory notifications of

human cryptosporidiosis during the 2001 foot-and-mouth disease epidemic in Eng land and

1 03

Wales, attributed to a reduction in the catt le populat ion and decreased di rect and i ndirect

exposure of humans to l ivestock (Smerdon et al. 2003} .

I n catt le, infections are predom inantly perinatal, and a typical faecal oocyst shedding period

lasts a number of days fol lowed by the development of resistance to re-infection (Anderson

1 98 1 ; Harp and Woodm ansee 1 990; Peeters et al . 1 993 ; Fayer et al . 1 998} . In calves, C.

parvum causes d iarrhoea, as repeatedly indicated by results of observational stud ies,

experimental i nfections, and therapeutic tr ia ls (Tzipori et a l . 1 983; Heine et al. 1 984; Fayer and

El l is 1 993 ; N aciri et al . 1 993, 1 999 ; Fayer et a l . 1 998; O'Handley et a l . 1 999; Castro-Hermida et

al . 2002; Gr inberg et al . 2002 ; Sevinc et al . 2003}. Faecal counts as high as 1 06 - 1 0?

oocysts/g (OPG) were found in faeces of infected calves during the peak of excretion (Ongerth

and Stibbs 1 989; Fayer et al. 1 998; O'Handley et al . 1 999; Uga et al . 2000; N ydam et al . 2001 ) .

Most studies indicate peak shedding of oocysts during the second week after birth , and that

adu lt catt le on ly sporadical ly shed low nu mbers of oocysts. Hence, 1 - 2-weeks-old calves are

regarded among the most eff icient ampl ifiers of C. parvum in nature , whereas adult cattle are,

perhaps, maintenance hosts ( Harp and Woodmansee 1 990 ; Harp et al. 1 996; Fayer et al. 1 998 ;

Atwi l l e t a l . 1 999, 2003; de la Fuente et al . 1 999, Sischo et a l . 2000; Uga et a l . 2000 ; Hoar et a l .

200 1 ; Nydam et al . 200 1 ; G rinberg et al . 2002; Atwi l l et a l . 2003; Atwi l l and das Pereira 2003;

Sevinc et al . 2003; Fayer 2004}. Given these characteristics, gaug ing the prevalence of C.

parvum prevalence in newborn calves is i mportant i n assessment of the possible zoonot ic

impact of dairy farm ing .

Bacteria belonging to the genus Campylobacter, especial ly C. jejuni subsp. jejuni (C. jejum) , are

important human pathogens causing main ly gastro intestinal i l l ness (Nachamkin 2003) .

Although C. jejuni has caused mastitis exper imental ly and C . jejuni and C. fetus subsp fetus can

sporadical ly cause abort ion , only C. fetus subsp veneralis, the cause of bovine venereal

campylobacteriosis, is considered a true pathogen of catt le. The other Campylobacter species

are generally considered non-pathogenic commensals of the bovine gastrointest inal tract

(Joens 2004) and are of main ly zoonotic importance.

There are about 2,500 serotypes of Salmonella and most of those infect ing m ammals belong to

the species S. enterica subsp enterica (Libby et al. 2004) . In the developed world, human

salmonellosis tends to m an ifest cl in ical ly as a gastrointest inal i l lness accompanied by diarrhoea,

fever, and abdominal cramps. Cases are mostly l inked to the i ngest ion of contam inated food of

animal origin (Bopp et al . 2003) . Bovine salmonellosis m anifests mai nly as enteritis or a

septicaem ia (Libby et a l . 2004} . Salmonella enterica serotype Typhimur ium was, by far, the

serotype most frequently isolated from cases in humans and cattle in New Zealand in 2003 and

2004, and m any of the isolates from both hu mans and cattle belonged to the same phagetype

(Anonymous 2004} . The bovine host-adapted S. enterica serotype Dubl in (Libby et a l . 2004}

was not isolated from cattle in New Zealand in 2003 and 2004 (Anonymous 2004) .

1 04

Although the presence of C. parvum in faecal samples from calves submitted to diag nostic

laboratories in New Zealand has been repeated ly reported (Townsend and Lance 1 987;

Learmonth et a l . 2004; Varney 2004) , the prevalence of th is parasite in dairy cattle in New

Zealand is unknown . Extrapolating C. parvum preva lence data from overseas to cattle in New

Zealand is d ifficult. Most existi ng prevalence data originate from countries where the dairy

i ndustry has a year- round calvi ng pattern, whereas the vast majority of farms in New Zealand

have a concentrated calv ing pattern, mostly in late winter. A sharp increase in the numbers of

immunological ly-na"ive calves, and their virtual absence between calving seasons, m ight have a

d ifferent effect on the epidemio logy of this infectio n in New Zealand as compared to cattle­

rearing countries in which there is un rem itting enzootic cycl ing due to a conti nuous p resence of

susceptible calves.

The aim of the study presented here was to assess the occurrence of C. parvum among

newborn calves i n the Manawatu reg ion of New Zealand, and to provide basel ine data for future

reference. Whi lst assessment of Campylobacter and Salmonella was not the m ain purpose of

the study, i t was envisaged that by including these two zoonotic organisms, th is study would

also provide data on their impact on cattle in New Zealand .

3.4.3 Materials and methods

Study design and sampling strategy. A cross-sectional survey was conducted in the

Manawatu reg ion of New Zealand during the calving season , in August 2002. Dai ry farms were

selected by convenience criter ia from the cl ient l ist of a veter inary practice. Farms were el ig ible

for inclusion in the study if at least 1 50 cows had been m i lked in the previous season, and

approximately 60 farms fulfi l led th is criterion . No information on the occurrence of enteric

pathogens and neonatal calf diarrhoea in previous calving seasons was considered. Farmers

agreeing to participate were asked to allow al l the calves between 9 and 1 5 days of age to be

sampled. This purposive sampl ing was performed i n consideration of the budget available, the

estimated number of calves of this age that would be avai lable for the sampl ing , and mainly

g iven that longitudina l excret ion stud ies in calves i ndicated a peak of prevalence of oocysts in

faeces mostly coincidental with th is age interval (Anderson 1 98 1 ; Harp et al . 1 996a; Fayer et a l .

1 998 ; Atwi l l e t a l . 1 999; Uga et a l . 2000; Grinberg et al . 2002} . Faecal specimens were collected

from the rectums of i ndividual calves and transferred to clean watert ight containers; g loves were

changed between calves. I nformation recorded included gender and breed of the calves, and

the type of flooring or bedding m aterial .

Faecal specimens were categorised as 'sol id ' (when deposited in its container, the specimen

conserved the orig inal shape) , 'semi-sol id' (sam ple spreading across the bottom of the

container but not l iquid) , and ' l iqu id ' (sample had a l iquid consistency) . Speci mens were

transported on ice to the microbiology laboratory and were processed the same day for

Campylobacter and Salmonella, and stored at 4QC for up to 72 h unti l tested for C. parvum.

1 05

All procedu res i nvolvi ng the sampling of the anim als were approved by the Animal Ethics

Com m i ttee of Massey U ni versity, Palme rston North , New Zealand.

Laboratory procedu res for detection of Cryptosporidium parvum. One g of faeces was

suspended in 1 0 ml of tap water and sieved through a 500 11m sieve. Then, oocysts were

concentrated by centrifugation at 900 g and the pel let (of approxi m ately one m l ) was re­

suspended in 4 ml normal sal ine. Ten m i c rolitres of suspension was deposited as a drop on a

sl ide, a i r-dried, and stained using the cold acid-fast stai n i ng tech nique. The enti re area of the

smear was exami ned usi n g an optical m icroscope at 400x magnification , and acid-fast oocyst­

l ike structures stained red were assessed morphological ly and morphometrical ly at 1 OOOx

magnification. Samples were considered positive if they h ad at �east one round acid-fast oocyst

with a d iameter of about 4 - 6 11m , and contai ning i nternal pu rple-stained sporozoites (Ortega

and Arrowood 2003) . Each oocyst detected represented an average of at least 500 oocysts i n

one gram o f faeces.

Laboratory procedu res for detection of Campylobacter. A 500 mg sample of faecal m ateria l

was i noculated i n cefoperazone-am photericin-teicoplanin (CAT) b roth prepared i n -house

(Atabay and Carry 1 998) , and incu bated with loose caps for 48 h at 37 "C in jars co ntain ing

com m ercial kits for microaerobic, capnoph i l ic atmosphere suitable for the growth of

Campylobacter spp. ( M itsubishi Gas Chem ical Co. , Tokyo , Japan). Then, 5 111 were plated onto

modified cefoperazone-ch arcoal-deoxycholate agar ( m C C DA) and on com m ercial CAT agar

plates ( Fort Richards Labo ratories, Auckland , New Zealand), and i ncu bated as before. P lates

were c hecked for g rowth every 48 h for a maximum of 8 days ( E n g berg et a l . 2000) . When

g rowth was seen, one colony was tested for a G ram -stai n reaction , and oxidase reaction using

oxidase-detection str ips (Oxidase strips, Oxoid Ltd . , Hants, UK). Colo n i es of oxidase-positive,

G ram -negative wavy baci l l i resem bl ing cam pylobacters were subcultu red onto a 5% sheep

blood agar plate prepared in-house and i ncubated at 37"C for 48 hours in jars, as before.

Bacteria recovered from blood agar plates were suspended in nutr ient broth conta in ing 1 5%

(Vo l u m eNolume) g lycero l , and frozen at -70 "C for future reference. After 1 2 m onths , one

frozen isolate from each positive sample was cultured for three cycles o n 5% sheep blood agar

plates, as before, and identified using the followi ng tests (Nacha m k i n 2003) : h ippu ricase

activity us ing a fast test, as per the m an ufacturer's instructions (Becton D ickinson Microbiology

System s , Cockeysvi l le M D , USA) ; susceptibi l ity to nal idixic acid and cephalot h i n using the disc

diffusion test; g rowth at 25"C and 42 "C ; catalase reactio n , and production of i ndoxyl acetate as

described by Popovic-Uroic et al . ( 1 990) .

Laboratory procedures for Salmonella. Swabs of faecal material were i nocu lated into tubes

contai n i ng selen ite en rich ment broth ( Fort Richards labo ratories) and i ncubated aerobically at

37"C for 20-24 h. Five 111 of broth were i nocu lated onto xylose-lysi ne-deoxycholate (XLD) plates

1 06

( Fort Richards Laboratories) and incubated as before. Further steps were not needed, as there

was no growth consistent with Salmonella in the samples.

Statistical analysis. Assuming a perfect test, the 95% confidence interval (C l ) of the proportion

of Cryptosporidium-positive farms was calculated using the f reeware Win Episcope 2 .0 (De Bias

and Ortega 2000} , option 'Sample Size to Estim ate Percentage ' , a popu latio n size of 60, level of

confidence set at 0 .05 , and f in i te populat ion correction . Odds ratios (OR) and exact 95% C l , and

two-tailed Fisher's exact tests, were used to assess associations between the presence of

Cryptosporidium infections and l iquid faeces on farms, and at the calf leve l . Calculat ions were

performed using Episheet© 2002 freeware ( Rothman 2002} . The modest sample size precluded

a mu ltivariate analysis.

3.4.4 Results

A total of 1 85 calves from 24 farms were sampled, during 26 farm visits carried out between

August 1 2 and September 3, 2002. Two farms, one test ing positive and the other test ing

negative to Cryptosporidium at fi rst instance, were re-sampled three weeks after the fi rst

sampling i n order to re-asses their Cryptosporidium i nfection status. All the sam ples were

tested for Salmonella, 1 62 for Campylobacter and 1 56 for Cryptosporidium, as some samples

were of an insuff icient quantity for test ing accord ing to the th ree different protoco ls. Most farms

reported sel l ing male calves at the age of about 1 week, hence, only 2 1 /1 85 ( 1 1 .4 %} calves

were male. Management systems on farms included leavin g calves with their mothers for 1 - 4

days, and then g rouping them i nto pens in barns or paddocks, where they were fed colostrum

and/or whole m i lk. Only o ne farm reported feeding calves with milk replacer. Calves of the

same age were mostly kept together. A variety of flooring or bedding materials were reported

(Table 3 .5) .

There were 33/1 56 (21 .2%) Cryptosporidium-positive calves from 1 0/24 (42%) farms (95% Cl=

26.3-56.8} . Most infected farms had more than one calf infected . The within-farm prevalence of

infected calves ranged from on ly one of seven to seven of seven calves test ing positive (Table

3.4) . There was i nterpretive uncertainty concern ing on ly one farm , which was eventual ly

considered posit ive by the presence of one positive sample contain ing only two oocysts on the

entire sl ide. One farm where a l l calves tested negative on the first visit became infected and

showed positive calves on a second visit about 3 weeks later, and was thus considered

Cryptosporidium-positive. The other farm visited twice had infected calves on both occasions

but it was counted as posit ive only once. S ix infected calves from four farms were shedding

more than 1 o6 OPG faeces. I nterest ingly, none of the four farms rearing Jersey cattle had

infected calves (Table 3 .5 ) . Of 1 0 farms where calves had l iquid faecal samples, n ine were

Cryptosporidium-positive. The occurrence of l iquid faeces on farms was positively associated

with the f ind ing of at least o ne Cryptosporidium-positive calf (0R=1 1 7, 95% C l = 4.8 - 5489; P<

1 07

0 .0 1 ) . Ten of 1 6 calves with l iquid faeces were Cryptosporidium-positive , but on ly 1 0/33

Cryptosporidium-positive calves had l iquid faeces. The association between the presence of

l iquid faeces and Cryptosporidium i nfections at the calf-level was significant (0R=8.47, 95% C l =

C l=2 .45-30.8 ; p<0 . 0 1 ) .

Breed Bedding No. C. parvum- No. Campylobacter- Number type positive positive samples/No. of l iquid

samples/No. examined faeces exa mined

F Wood chips 5/8 4/8 2 F Paddock 0/7 5/7 0 F Woodch ips 0/9 1 /9 0 J Sawdust 0/8 7/1 3 0 J Wooden slats 0/3 0/3 0 F Sawdust 017 3/8 1 F Sawdust 0/9 7/1 3 0 c Concrete 1 /7 2/7 1 c Woodch ips 3/8 1 0/1 1 1 nr Saw dust 0/7 2/8 0 nr Paddock 0/5 1 /6 0 J Paddock 0/3 0/3 0 c Wooden slats 0/6 2/6 0 J Woodchips 0/5 0/5 0 c Wooden slats 6/6 5/9 2 F Sawdust 4/1 0 2/1 0 4 c Paddock 0/7 1 /4 0 c Straw 0/7 0/5 0 c Sawdust 3/8 0/6 1 F Sawdust 0/4 3/4 0 F N R 3/6 1 /4 2 F Paddock 1 /4 1 /2 0 c Sawdust 4/6 1 /4 1 F Woodchi�s 3/6 0/6 1

Table 3 .5. Prevalence of Cryptosporidium parvum and Campylobacter spp among newborn calves from 24

dairy farms in the Manawatu region of N ew Zealand (each row represents one farm) . F= Friesian ; J=

Jersey; C= cross-breed; NR= not recorded

Fifty-eig ht (36%) samples from 1 8 farms tested positive for Campylobacter. A wide range of

Campylobacter phenotypes were detected; C. jejuni subsp. jejuni was isolated from 1 1 samples

from 7 farms (Table 3 .5) . The most common phenotype was that of C. sputorum-fecalis, isolated

from 30 samples from 1 1 farms. C. hyointestinalis subsp hyointestinalis phenotype was isolated

in 1 0 samples from seven farms, C. coli phenotypes were isolated fro m three samples, and C.

lari was isolated once. Colonies consistent with Salmonella were not observed on XLD agar

plates.

3.4.5 Discussion

The prevalence of C. parvum i n d airy calves in New Zealand has not been measured

previously. In th is study, a purposive sampling of newborn calves at the theoretical peak of the

1 08

patent period was applied to en hance herd-test sensitivity. More than 3 sam ples were tested in

the m ajority of the Cryptosporidium-negative farms , and 8/1 0 Cryptosporidium-positive farms

were declared positive by the presence of mu ltiple i nfected calves (Table 3.4) . Hence, test

accuracy at the herd level was h igh . The overall farm-prevalence was 42%. Com parable

percentages of Cryptosporidium-infected farms have been reported in countries with year-round

calving (Genchi et al . 1 984 ; Harp and Woodmansee 1 990 ; Garber et al . 1 994; Oui lez et a l .

1 996) . However, a prevalence >90% was reported i n the Canadian states of Quebec and

Ontario in cross-sectional studies, and in Galicia ( Spain ) in a long itudinal study (Ruest et a l .

1 998 ; Castro-Herm ida et a l . 2002 ; McAI I ister et a l . 2005 ; Trotz-Wi l l iams et a l . 2005) . Although

some inadvertent bias due to non-random recruitment of farms is possible, these resu lts

suggest that enzootic C. parvum cycles estab lish in a considerable proport ion of farms, despite

the short and concentrated period of calv ing and long t ime interval between calving seasons

characteristic of most dairy farms in New Zealand. The proportion of C. parvum positive calves

with in farms was variab le, in accordance with resu lts described elsewhere ( Santi n et al. 2004) .

More than 1 o6 OPG faeces were detected i n six samples from four farms. Such h igh numbers

are typical ly found dur ing a smal l fraction of the patent period (Fayer et a l . 1 998; Uga et a l .

2000 ; Gri nberg et a l . 2002) . Hence, it is conceivable that i n a longitudinal study, an excret ion of

th is m agn itude would have been observed, at some stage, in the majority of the positive calves.

G iven the h igh number of New Zealand dairy cattle, and in v iew of the fact that calves p roduce

the vast majority of C. parvum oocysts shed on farms (Atwi l l and das Pereira 2003; Fayer

2004) , these results indicate a massive natural am plif ication of C. parvum during the calving

season . What remains u nknown is whether infections on farms are in it ia l ly acquired from

environmental sources, or from infected reservoi r-cows. Another question is whether or n ot

calves on apparently negative farms become infected later i n life, as calves raised in iso lation

remained susceptible to experimental i nfectio n for at least 3 months (Harp and Woodm ansee

1 990) .

The possibil ity of cattle-to-human cycl ing of C. parvum is supported by the resu lts of

experimental infect ions i n humans with cattle iso lates (DuPont et al . 1 995, Chappel and

Okhuysen 2001 ) , and by studies of the popu lation genetic structure of C. parvum that have

shown extensive sharing of genotypes between humans and cattle ( Mallon et al . 2003ab) . A

number of f indings converge in suggest ing an epidemiolog ical associat ion between the

occurrence of human cryptosporidiosis and the dynamics of the dairy cattle population in New

Zealand: National notificat ions of cases of human cryptosporid iosis from the last 4 years

denoted monthly fluctuat ions , and the nu mber of notified cases peaked after the end of the

calv ing season in spring (Anonymous 2004) . Rivers impacted by dairy farms were found to be

at a h igher risk of being contam inated with faecal i ndicator bacteria ( Donnison and Ross 1 999) ,

and results of a systematic survey indicate a widespread d istribution of Cryptosporidium spp.

oocysts in surface waters in New Zealand (Brown et al . 1 998) . Final ly, mo lecular

characterisat ion of the Cryptosporidium cl in ical isolates recovered from humans in 200 1 and

1 09

2002 in New Zealand showed a disti nctive virtual substitution of the anthroponotic species C.

hominis with the zoonotic C. parvum after the end of the calving season in both years

(Learmonth et a l . 2004) . Our study provides another line of evidence to tentatively explain th is

interest ing sh ift: it is plausible that the sign ificant expansion of the niche represented by the

large increase in numbers of immunolog ical ly-na'lve calves m ight al low a select ive ampl ification

of h ighly-infective clones, followed by their environmental d ispersal v ia calf-related biomasses.

However, humans can acqu i re C. parvum infectio n through a variety of routes, including

ingestion of contami nated food and water, an imal-to-human, and person-to-person (Guerrant

1 997; Xiao et al . 2004) and at present, the idea of a massive cattle-to-human parasite cycl ing is

specu lative. To prove this concept, an analysis of the populat ion genetic structure us ing a large

number of isolates of C. parvum from human and catt le, and a h igh ly discri m inatory typing

method, wou ld be necessary. Such a study is presented in Section 3 .5 .

I n th is study, the probabi l ity o f detecting cryptosporidial i nfections was significantly higher on

farms in which at least one calf was suffering f rom l iquid diarrhoea, and the associat ion between

the presence of l iqu id faeces and an i nfection was significant at the calf-level . However, this

study was not desig ned to study the causes of d iarrhoea, hence, these stat istics are on ly

reported for futu re reference.

A relatively h igh percentage of farms infected with Campylobacter spp, and especial ly with C.

jejuni, the most important human pathogen of th is genus (Nachamkin 2003) , was observed i n

o u r study. Campylobacter jejuni is known to co lonise the intestinal tract o f asymptom atic catt le.

However, most popu lation-based studies were conducted using either samples from adult

cattle, older calves, or without precise age-specification (Garcia et al. 1 985; Meanger and

Marshal ! 1 989; G iacoboni et al. 1 993; Atabay et al. 1 998 ; Stanley et a l . 1 998; Busato et a l .

1 999 ; Hoar et a l . 2001 ; N ielssen 2002 ; Savi l l et a l . 2003 ; Sato et a l . 2004 ; Bae et al . 2005;

Wesley et a l . 2000) , and little is known about co lon isation of the intestines by Campylobacter in

the neonate calf . Campylobacter jejuni was found in a relatively h igh percentage of 4-week-old

calves in an abatto i r study (Grau 1 988). The present study indicates that a wide range of

Campylobacter species, in particular C. jejuni, co lon ise the intestinal tract of cattle early in l ife.

Salmonellae were n ot detected in this study. This is in agreement with reports from other

countries indicat ing a very low rate of detect ion of intest inal carriage of Salmonella among

healthy young calves (Lance et a l . 1 992; Busato et a l . 1 999; Naciri et a l . 1 999, McAI I ister et a l .

2005) and consisten t with the observation of a lower prevalence of carriage of Salmonella in

unweaned calves than in adult an imals observed by Huston et al . (2002) . These low rate of

subcl in ical carriage of Salmonella in newborn calves reinforces the diag nostic value of i solating

of Salmonella fro m the faeces of newborn calves dur ing disease outbreaks.

Calves infected with C. parvum and C. jejuni may be a source of human disease by direct

transmission . Sm ith et al. (2004) concluded that calves were the reservoir of mu ltiple enteric

1 1 0

pathogens, i nclud ing C. parvum and C. jejuni, for chi ldren i n the Un ited States . Stefanogiannis

et al . (200 1 ) l i nked an outbreak of cryptosporidiosis i n ch i ldren in New Zealand to direct contact

with calves during a farm visit . The i nfectivity of C. parvum and C. jejuni i n hum ans is h igh

(Okhuysen et al . 1 999; Chappel and Okhuysen 2001 ; N achamkin 2003) . Thus, care should be

taken to avoid unnecessary exposure of people to calves less than 4 weeks old on farm s. An

epidemiolog ical study in the USA reported that the i ncreased frequency of spreading manure on

fields was a risk factor for detecting Cryptosporidium oocysts in streams (Sischo et al. 2000) . At

present, it seems advisable to avoid spreading calf manure and bedding o n paddocks.

In conclus ion, this study provides a snapshot of the occurrence of C. parvum, Campylobacter

spp and Salmonella spp in the Manawatu region, which perhaps can be generalised to other

regions in New Zealand with comparable cattle-rearing practices. Despite the theoretical

sanit ising effect of t ime between calving seasons in New Zealand, results confi rm that newborn

calves are amplifiers of C. parvum. Although results suggest an important role for C. parvum in

the neonatal calf diarrhoea complex, its attributable impact in New Zealand needs to be

establ ished. Campylobacter spp, in part icular C. jejuni, but not Salmonella, were also found to

be widespread. Studies of alternative systems to reduce the envi ronmental d ispersal of viable

enteropathogens from preweaned calf wastes at the end of the calv ing season are warranted.

1 1 1

3.5 PERSISTENT DOMINANCE OF TWO G P60 //a ALLELES IN DIARRHOEAG ENIC HUMAN

AN D BOVINE CRYPTOSPORIDIUM PAR VUM IN N EW ZEALAND

3.5.1 Summary

The study presented i n Sectio n 3 .4 provided an estim ate of the dairy farm-prevalence of C.

parvum in New Zealand. At that junction , the molecular characterisation of the iso lates was not

deemed necessary as C. parvum was considered the only taxon extensively parasitis ing catt le.

The author has hypothesised that because New Zealand has a un ique dairy enviro nment,

where mi l l ions of susceptible newborn calves are reared withi n a short t ime period dur ing

winter, C.parvum genetic l ineages able to rapidly f i l l the niche represented by the transient

presence of susceptible newborn calves, have become establ ished.

In this study, 68 d iarrhoeagenic human (n=20} and bovine (n=48) Cryptosporidium isolates

collected duri ng the annual peak of morbidity i n wi nter and sprin g of 2006, were genetical ly­

identified at the 1 8S rRNA gene and subtyped at the polymorphic region of the G P60 gene. Al l

the identified isolates were C. parvum. Two dominant G P60 al leles belong ing to the /la

A1 8G3 R 1 and //a A 1 9G4R 1 alel le famil ies, which persisted i n New Zealand as the dominant C.

parvum allele since 2002, were found in bovine and human C. parvum. The author postu lates

that the cyclic domi nance of these subtypes in New Zealand is due to a rapacious phenotype, of

an ability to rapid ly f i l l the niche during the calving season , and survive between seasons.

These results corroborate previous observations on the domi nant role of C. parvum i n humans

during the seasonal peaks of morbidity, and provide the molecular epidem iological evidence to

support the notion of widespread zoonotic transm ission of this parasite dur ing these peaks.

3.5.2 Introductio n

U nti l recently, C . parvum was considered the sole i ntestinal Cryptosporidium species i nfecting

young cattle and due to the h igh incidence of cryptosporidiosis and the large numbers of

oocysts e l im inated in the faeces, i nfected calves were considered among the m ajor amplifiers of

zoonotic C. parvum in nature ( Harp and Woodmansee 1 990; Harp et al . 1 996; G rinberg et al .

2002) . However, this v iew started to change as phenotypical ly-s imi lar Cryptosporidium taxa of

uncertain pathogen icity and zoonotic potential , such as C. bovis, the Cryptosporidium 'deer-l ike

genotype', C. suis, C. hominis, and C. suis-l ike genotype, have been identified in catt le using

molecu lar tools ( reviewed by Sant in and Trout 2008} . Furthermore, results of some molecular

epidemiological studies revealed the possib i l ity of the existence of host-substructuring, and the

occurrence of human C. parvum subtypes that do not cycle in cattle (Mallon et al . 2003a,b ; X iao

and Fayer 2008; Section 2 . 1 ) . Therefore, whi le in the past the mere identificat ion of C. parvum­

l ike oocysts i n cattle was regarded as as compel l ing evidence for the zoonotic potential of these

parasites, current epidemiological practice requ i res the gebetic identification of the isolates and

their comparison with human parasites using molecular tools . This trend is clearly manifested by

1 1 2

the large number of molecular studies of Cryptosporidium in l ivestock using subtyping publ ished

in recent years ( l isted by Santin and Trout 2008).

Many studies have used the gene encoding the Cryptosporidium sporozoite surface

glycoprotein GP60 (Cevallos et a l . 2000 ; Strong et al . 2000) (see Sections 1 .5 and 1 . 7 ) . Th is

locus is appeal ing because of i ts extensive sequence polymorph ism , in part icular the length

polymorphism of a polyserine repeat. Many studies have revealed a variety of G P60 al le l les in

C. parvum from hum ans and cattle with i n different reg ions (Aives et al. 2003, 2006; Thom pson

et a l . 2007; Brog l ia et a l . 2008; Oui lez et al . 2008; Z intl et a l . 2009) . However, it is somehow

surpris ing that on ly a few studies have genetically-compared isolates col lected from h umans

and animals over the same t ime and space frames ( Mal lon et a l . 2003a,b ; Sect ion 2 . 1 ) .

Human cryptosporidiosis has been a notifiable disease i n New Zealand since 1 996 ( Learmonth

et al . 2004) . Since then, a s ignificant associat ion between the rate of notificat ions of

cryptosporidiosis in humans and the dynamics of the l ivestock population have been recorded.

For instance, every year the number of notified cases of cryptosporidiosis has peaked s harply

after the end of the calving season in spring , and then abruptly dropped in the su m mer,

coinciding with a sharp reduction in the number of immunological ly-naive calves p resent

(www.surv.esr.cr i .nz/PDF _survei l lance/Annuai Rpt, accessed on 1 0 February 2009; F igure 1 .4) .

Most interestingly, Learmonth et a l . {2001 , 2004) revealed a virtual substitution of the

anthroponotic species C. hominis with C. parvum dur ing the peaks of morbidity. However,

except in the study presented in Section 3.2, no genetic comparisons between human and

bovine C. parvum isolates were undertaken in New Zealand.

I n the present study, Cryptosporidium-positive faecal specimens from diarrhoeic humans and

calves col lected over the same time period were g enetical ly identif ied and subtyped by

sequence analysis of the 1 8S rRNA gene and the Cryptosporidium sporozoite surface 60 kDA

glycoprotein gene (GP60) , to assess the possible role of cattle as a source of zoonotic

cryptosporidiosis in the Waikato reg ion . The results are discussed in the context of previous

f ind ings and the pecu l iar eco logical conditions prevai l ing in the reg ion .

3.5.3 Materials and Methods

Study location. This study used human and bovine Cryptosporidium-positive faecal specimens

col lected from overt infections in the region of W aikato, North Is land, New Zealand . The

Waikato is a m ixed rural-urban region with ea. 4000 dairy farms and one m i l l ion cows

{2007/2008 New Zealand Dairy Statistics, Livestock Improvement Corporatio n L im ited ,

www. l ic .co.nz, down loaded on 1 0 February 2009) . In 2006, the rate of notif ication of

cryptosporidiosis in the Waikato was between 24.2 - 76.5 cases per 1 00,000 population , well

above the national rate of 1 7.8 per 1 00 ,000 popu lation (Notifiable and other Diseases in New

Zealand, Annual Report 2006, www.surv.esr.cri . nz/PDF _surveillance/Annuai Rpt, downloaded

1 1 3

on 1 0 February 2009) . Results of previous molecu lar epidemiological studies ind icated C.

parvum as the predominant Cryptosporidium species in humans in that reg ion ( Learmonth et a l .

2004) .

Cryptosporidium positive faecal samples. Forty eight bovine and 20 human faecal

specimens were used in th is study. The specim ens were arbitrari ly selected from a col lection of

bovine and human Cryptosporidium-positive speci mens donated by veteri nary and human

diagnostic laboratories i n the W aikato reg ion . The bovine speci mens or ig inated from unweaned

calves and were orig inal ly subm itted by farmers or veteri nary practitioners for analysis of causes

of diarrhoea during the calv ing season between July-October 2006. No i nformation about eo­

infections with other enteropathogens was avai lable. Only one specimen per farm was i nc luded

in this study. The human faecal specimens were or ig inal ly submitted to two diag nostic

laboratories for analysis of causes of diarrhoea between Ju ly-October 2006. No patient- level

inform ation was avai lable. The bovine and human specimens were submitted on ice to M assey

Un iversity and stored between 2 - 4 "C with no p reservatives, unti l analysed.

Cryptosporidium identification and subtyping. Genomic DNA was extracted from the 68

specimens using a commercial DNA extraction kit (QIAamp® DNA Stool M in i Kit , Q iagen,

Hi lden , GmbH) . The identification of Cryptosporidium was performed us ing a nested PCR

fol lowed by sequence analysis of a -850 base-pair segment of the 1 8S rRNA gene descr ibed in

Sections 3 . 1 and 3.3. Mo lecu lar-grade water was used as negative control. Forward and

reverse sequences were a l igned and edited with the aid of chromatogram s. Proximal and d istal

sequence seg ments that could not be accurately determined were trimmed , and the edited

sequences al igned with Cryptosporidium 1 8S rRNA gene sequences in GenBank using

nucleotide a l ignment ClustaiX software (Thom pson et al . 1 997) . Cryptosporidium subtyp ing was

performed by analysis of the sequence of the po lymorphic region of the gene coding for the

sporozoite G P60 gene (Cevallos et a l . 2000 ; Strong et a l . 2000) as described i n Section 3 .2 .

3.5.4 Results

Thirty n i ne out of 48 (81 %) bovine specimens yielded 1 8s rRNA gene amplicons of the

expected s ize in gels. Fourteen of these (36%) yielded sequences that could be unambiguously

edited . The 1 8s rRNA gene sequences of the remain ing 25 ampl icons could not be edited due

to uneditable umbiguities in the chromatog rams, or the absence of sequence s ignals.

Conversely, 1 6 out of 20 (80%) human Cryptosporidium-positive faecal specimens yielded 1 8S

rRNA gene am plicons of the expected sizes o n gels, which could al l be accu rately edited . The

1 6 human sequences and 1 3/ 1 4 bovi ne sequences were identical to the sequence of the C.

parvum 1 8S rRNA gene (Gen Bank accession number AF093490) . One bovine sequence was

identical to the polymorph ic (or "Type B") copy of the same C. parvum gene (LeBiancq et a l .

1 997; see Section 1 .3) .

1 1 4

Unambiguous G P60 sequences could be generated for 32 (67%) bovine iso lates (the G P60

sequences of e ight bovine C. parvum isolates have been reported in Section 3 .2) . Th i rty one out

of 32 GP60 sequences of bovine isolates corresponded to known C. parvum alleles.

Su rpris ing ly, one isolate had a G P60 al lele previously identified several t imes in C. hominis from

humans. This C. hominis allele had been reported seven times in GenBank under the taxon

name C. parvum ' human genotype' (accession numbers AF403 1 69 . 1 ) , and C. hominis

(accession numbers AY382670, AY382673. 1 , AY382669. 1 , EF576980 . 1 , EF591 786. 1 , and

EF59 1 785 . 1 ) . However, DNA extracted from the same iso late a number of months later yielded

a G P60 sequence of C. parvum which was identical to the most abundant sequence found in

this study (see below). Thus, the fi rst C. hominis sequence m ay have been the result of an

isolate identification error. Th i rty e ight out of 48 (77%) bovine isolates could be unambiguously

identif ied as C. parvum at either the 1 8S rRNA gene, the GP60 locus, or both loci and ten

bovine isolates (2 1 %) could not be identified. Seven out of 20 (35%) human isolates yielded

unambiguous G P60 sequences. Al l these human isolates yielded C. parvum 1 8S rRNA gene

sequences.

There were three different GP60 sequences in the sample (Appendix 3 .5) . Twenty-n ine bovine

and six human sequences were identical to each other. Accord ing to the nomenclature

suggested by Sulai m an et al. (2005) , this allele belonged to the C. parvum G P60 al lele fami ly

/la A 1 8G3R 1 . The same sequence was the dominant C. parvum allele identified in New

Zealand from foals and humans between 2001 -2007 (Sect ion 3 .2) . A search in a database

maintained by the Protozoa Research Unit of the I nstitute of Veterinary, Animal and Biomedical

Sciences, Massey U n iversity, New Zealand (PRU) , revealed that the same /la A 1 8G3R1 al lele

has been identified in 1 65/263 (62%) of the iso lates collected between 2002-2008, and

genetical ly characterised by the Un i t ( Errol Kwan, personal comin ication) . A search in GenBank

(website maintained by The Nat ional Center for Biotechno logy I nformation (NCBI) of the

N ational I nstitutes of Health, USA) revealed that this allele had also been identified from calves

in Ontario, Canada (Trotz-Wi l l iams et al. 2006) . Three C. parvum isolates from calves had

G P60 sequences identical with each other and belonging to the C. parvum G P60 al lele fami ly

/la A 1 9G4R 1 . A search in the PRU database indicated th is al lele as the second most frequent

allele seen in human C. parvum in New Zealand , present in 64/263 (24%) isolates . The same

allele has been found in bovine C. parvum in eastern USA (GenBank accession number

006305 1 6) and Canada (accession numbers AF 1 64493. 1 and 001 92504. 1 ) , and also in

h um ans in Brasi l (accession number AF1 64493 . 1 ) . One human C. parvum isolate also had a

G P60 sequence belonging to the G P60 al lele fam i ly /la A 1 8G 3 R 1 which differed from the

sequence present in the other six human isolates at a single nucleotide external to the

polyserine repeat. The d ifference was not l i kely to be an isolated edit ing error as it persisted

after the author re-ed ited the sequence a year later. This allele has never been previously

identified in New Zealand, and an on l ine search in GenBank did not return any identical

sequence .

1 1 5

3.5.5 Discussion

I n the present study, d iagnostic faecal specimens were analysed in order to genetical ly­

compare Cryptosporidium parasites from humans and calves col lected dur ing the peak of

morbidity period i n 2006 i n the Waikato. The avai labil ity of isolates from both humans and cattle

differentiates this study from other molecular epidem iological studies applyi ng s im i lar genetic

typing methods which only used isolates from one host species.

Accumulated epidemiological data f rom a number of countries i ndicated that most

Cryptosporidium i nfect ions i n newborn calves are caused by C. parvum, whereas other taxa are

rarely found in this age g roup (Santin et al. 2004, 2008 ; Trotz-Wil l iams et al. 2006; Fayer et al .

2006; Geurden et a l . 2006) . The results of the present study support these results.

In the study in newborn calves in New Zealand presented in Sect ion 3 .4, the dairy farm

prevalence of C. parvum was estimated to be -40%. However, that study was performed in

2002, and at that juncture the genetic identif ication of the isolates was not carried out, so the

presence of non-parvum taxa could not be established . U n like the above study, the present

study was not designed to estimate the farm-prevalence of C. parvum. Nevertheless, the

identification of this species in 38 different farms in the Waikato supported the conclusion that

there is a widespread presence of this parasite in the dairy econiche in New Zealand expressed

in Section 3 .4 .

The exclusive f ind ing of C. parvum i n hu m ans corroborated previous f ind ings that indicated the

dominant role of C. parvum i n humans during the annual peaks of cryptosporidiosis notifications

in spring (Learmonth et al . 2004) . I n the study presented in Section 2 . 1 , a partit ion of the

multi locus genotype repertoire of C. parvum between humans and cattle was identif ied using

rarefaction analysis , which supported the notion of the existence of anthroponotic C. parvum

cycles in Scotland. The studies presented i n th is sectio n , and in Sect ions 2 .2 and 3.2 , fai led to

identify any genetic host-partition of the genetic repertoire of C. parvum in New Zealand . I n fact,

in the present study C. parvum isolates from humans and calves shared the most abundant

GP60 al leles, and although in the study in Section 2.2 the sample-size of C. parvum was too

small for an analysis usi ng rarefaction, human and bovi ne isolates shared most of the mu lt i locus

genotypes, form ing a si ng le eBURST network (Figure 2.4) . In addit ion, no putative

anthroponotic C. parvum isolates carry ing G P60 /le al leles have been identif ied in this study, or

stored in the database of the PRU. Col lectively, these resu lts do not support the existence of

anthroponotic C. parvum cycles but rather suggest a significant role of the zoonotic

transm ission route in New Zealand .

The avai labi l ity of the resu lts of a study from previous years (Section 3 .2 ) , and the G P60 data

stored by the PRU al lowed an assessment of the temporal variation in the prevalence of the

different C. parvum G P60 subtypes. The fact that the two most abundant al le les in 2006 were

1 1 6

also the dominant al leles i n m u ltiple host species since 2002 was strik ing . Despite the wide

spectrum of C. parvum subgenotypes in c i rculation in humans in New Zealand, two

subgenotypes - belonging to the al lele famil ies //a A1 8G3R 1 and /la A1 9G4 R 1 - pers istently

dom inated the samples. As h ypothesised in the study presented in Section 3.4, the author

postulates that the cyclic dominance of these subgenotypes is due to a rapacious phenotype,

and an abi l ity to rapidly f i l l the n iche during the calving season and survive between seasons .

The fact that th is study analysed d iagnostic specimens was a l imitat ion. Further studies us ing

representative samples of bovi ne C. parvum are needed in order to better captu re the genetic

diversity of this parasite species in cattle in New Zealand.

F inal ly, the pathogenic potential of the non-parvum Cryptosporidium cycl ing in cattle is n ot we l l

understood. Apart from s ix studies analys ing d iag nostic specimens (Mal lon et al . 2003a,b;

Thompson et a l . 2007; Brogl ia et al . 2008; Qui lez et a l . 2008; Soba and Logar 2008) , al l the

molecular studies analysing Cryptosporidium isolates of bovine or ig in analysed specimens

col lected from an imals of a wide age range and/or on a l imi ted number of selected farm s, with

no cl in ical h istory i ndicated (Mclauchlin et al. 2000; Learmonth et al. 200 1 ; Alves et a l . 2003;

Peng et al . 2003; Santin et a l . 2004; Nei ra-Otero et al . 2005; Roy et a l . 2006; Geurden et a l .

2006, 2007; Trotz-Wi l l iams et a l . 2006; Langkjar et al . 2007; Xiao et a l . 2007; Duranti et a l .

2008 ; Feltus et a l . 2008; Santi n et al 2008 ; Soba and Logar 2008) . The results of the present

study are in accordance with the above six studies, i nd icating C. parvum as the sole

d iarrhoeagenic Cryptosporidium species in newborn calves.

3.6 CONCLUDING REMARKS

Chapter 3 reports five epidemio log ical studies of cryptosporidiosis in foals, cattle and humans in

New Zealand. The combined resu lts of stud ies 3 . 1 , 3 .2, and 3.3 ind icate that c l in ical ly overt

infections with C. parvum are relat ively common in foals in New Zealand, and that the condit ion

is l ikely to be underdiagnosed . The eight cases of cryptosporid iosis documented in foals i n

Sections 3 . 1 - 3 .3 had an onset between the fi rst and third week o f l ife. The disease was self­

l im iting and characterised by a short and intense oocyst shedding period, resembl ing neonatal

cryptosporidiosis in calves. The i nfections were caused by C. parvum parasites genetical ly

s im ilar to bovi ne and human isolates, indicat ing the possibi lity of cross-transmission of C.

parvum between these host species. Due to the potential for zoonotic transm ission of C.

parvum, it is advisable to take precautions when handl ing d iarrhoeic foals and calves u nt i l th is

agent has been ruled out by the laboratory.

The results of the studies described in Section 3.4 - 3 .5 indicate newborn calves are sign ificant

amplifiers of potentially zoonotic C. parvum i n New Zealand, despite the short and concentrated

calving pattern characterising dairy farm ing . Biomasses from calf-rearing operations should be

adequately treated in order to avoid the contam i nation of the environment with i nfectious

1 1 7

oocysts. The f inding of two persistently dominant G P60 /la alleles in C. parvum warrants further

invest igat ion .

3.7 REFERENCES

Aiello A E, Xiao L, Limor JR, Liu C, Abrahamsen MS, Lal AA. Microsatell ite analysis of the

human and bovine genotypes of Cryptosporidium parvum. Journal of Eukaryote Microbiology

46, 46S-4 7S, 1 999

Akiyoshi DE, Feng X, Buckholt G, Widmer G, Tzipori S. Genetic analysis of a

Cryptosporidium parvum human genotype 1 isolate passaged throug h d ifferent host species.

Infection and Immunity 70, 5670-75, 2002

Alien AVH, Rid ley OS. Further observations on the formal ether concentration technique for

faecal parasites. Journal of Clinical Pathology 23, 545, 1 970

Alves, M, Xiao L, Sulaiman I , Lal AA, Matos 0, Antunes F. Subgenotype Analysis of

Cryptosporidium Isolates from H umans, Cattle, and Zoo Ruminants in Portugal . Journal of

Clinical Microbiology 2744-7, 2003

Alves M, Xiao L, Antunes F, Matos 0. Distribution of Cryptosporidium subtypes in humans

and domestic and wild rum inants in Portugal . Parasitology Research 99: 287-92, 2006

Anderson BC . Patterns of shedding of cryptosporid ia l oocysts in Idaho calves. Journal of the

American Veterinary Medical Association 1 78, 982-4, 1 98 1

Anonymous. New Zealand Publ ic Health Survei l lance Report, Vo lume 2 , Issue 4 , Institute of

Environmental Science and Research Ltd, Porirua, New Zealand, 2004

Anonymous. Equine disease surveil lance, July to September 2006. Veterinary Record 1 60, 5-

8 , 2007a

Anonymous. Equine disease survei l lance, January to M arch 2007. Veterinary Record 1 6 1 , 79-

82, 2007b

Anonymous. Equine disease survei l lance, April to June 2007. Veterinary Record 1 60, 507- 1 0 ,

2007c

Anonymous. Equi ne d isease survei l lance, Ju ly to September 2007. Veterinary Record 1 62, 70-

4, 2008a

1 1 8

Anonymous. Equine disease survei l lance, Apri l to June 2008. Veterinary Record 1 63, 367-80,

2008b

Atabay HI, Corry JE. The isolation and prevalence of campylobacters from dairy cattle us ing a

variety of methods. Journal of Applied Microbiology 84, 733-40, 1 998

Atwil l ER, das Pereira M. Lack of detectable shedding of Cryptosporidium parvum in

periparturient dairy cattle. Journal of Parasitology, 89, 1 234-6 , 2003

Atwil l ER, Johnson E, Kl ingborg DJ, Veserat GM, Markegard G, Jensen WA, Pratt DW,

Delmas RE, George HA, Forero LC, Phi l ips RL, Barry SJ, McDougald NK, Gildersleeve

RR, Frost WE. Age, geographic and temporal d istribution of fecal sheddi ng of

Cryptosporidium parvum oocysts in cow-calf herds. American Journal of Veterinary Research

60, 420-5, 1 999

Atwil l ER, Hoar B, das Gracas Cabral Pereira M, Tate KW, Rulofson F, Nader G. Improved

quantitative est imates of low environmental loading and sporadic peripartur ient shedd ing of

Cryptosporidium parvum in adu lt beef catt le. Applied and Environmental Microbiology 69, 4604-

1 0 , 2003

Bae W, Kaya KN, Hancock DD, Cal l DR, Park WY, Besser T. Prevalence and antim icrobial

resistance of thermoph i l ic Campylobacter spp. from cattle farms in Washi ngton State. Applied

and Environmental Microbiology 71 , 1 69-74, 2005

Bopp CA, Brenner FW, Fields PI, Wel ls JG, Strockbine NA. Escherichia, Shigella and

Salmonella. I n : PR Murray (ed) . Manual of Clinical Microbiology, 8th edtn . Pp 654-71 , ASM

Press, Washington DC, USA, 2003

Broglia A, Reckinger S, Caccio SM, Nockler K. Distr ibution of Cryptosporidium parvum

subtypes i n calves in Germany. Veterinary Parasitology 1 54 , 8-1 3, 2008

Brown T, lonas G, Learmonth J, Keys L, Mclenachan T. The distribution of Giardia and

Cryptosporidium i n New Zealand waters - a nationwide survey. Water and Wastes in New

Zealand 1 0 1 , 60-63, 1 998

Busato A, Hofer D, Lentze T, Gail lard C, Burnens A. P revalence and infection r isks of

zoonotic entheropatogenic bacterial i n Swiss cow-calf farms. Veterinary Microbiology 69, 25 1 -

63, 1 999

1 1 9

Caccio S, Homan W, Camilli R, Traldi G , Kortbeek T, Pozio E. A microsatellite marker

reveals populat ion heterogeneity with in human and an imal genotypes of Cryptosporidium

parvum. Parasitology 1 20, 237-44, 2000

Caccio S, Homan W, Vand Dijk K, Pozio E. Genetic polymorphism at the beta-tubul in locus

among human and animal isolates of Cryptosporidium parvum. FEMS Microbiology Letters 1 70 ,

1 73-9, 1 999

Carraway M, Tzipori S, Widmer G. A new restriction length po lymorphism from

Cryptosporidium parvum identifies genetical ly heterogeneous parasite popu lations and

genotypic changes fol lowing transmission from bovi ne to human hosts. Infection and Immunity

65, 3958-60, 1 997

Castro-Hermida J A, Gonzales-Lozada Y A, Ares-Mazas E. Prevalence of and risk factors

involved in the spread of neonatal bovine cryptosporidiosis in Gal icia (Spain ) . Veterinary

Parasitology 1 06 , 1 - 1 0 , 2002

Cevallos AM, Bhat N, Verdon R, Hamer OH, Stein B, Tzipori S, Pereira ME, Keusch GT,

Ward HO. Mediation of Cryptosporidium parvum i nfection in vitro by mucin- l ike g lycoproteins

defined by a neutral iz ing monoclonal antibody. Infection and Immunity 68, 5 1 67-75, 2000

Chalmers RM, Thomas AL, Butler BA, Morel MC. Identification of Cryptosporidium parvum

genotype 2 in domestic horses. Veterinary Record 1 56, 49-50, 2005

Chalmers RM, Thomas AL, Butler BA, Morel MZ. Identification of Cryptosporidium parvum

genotype 2 in domestic horses. Veterinary Record 1 56 : 49-50, 2005.

Chappel CL, Okhuysen PC. Cryptosporidiosis in health adult vo lunteers. I n : Sm ith M. and

Thompson KC (eds) , Cryptosporidiosis, the analytical challenge, Pp 62-72. The Royal Society

of Chem istry, Cambridge, UK, 200 1

Cohen, N.D. Causes of and farm m anagement factors associated with d isease and death i n

foals. Journal o f the A merican Veterinary Medical Association 204: 1 644- 1 651 , 1 994

Cole DJ, Cohen NO, Snowden K, Smith R. Prevalence of and risk factors for fecal shedding

of Cryptosporidium parvum oocysts in horses. Journal of the American Veterinary Medical

Association 2 1 3 : 1 296-02, 1 998

1 20

Coleman SU, Klei TR, French DD, Chapman MR, Costvet RE. Prevalence of

Cryptosporidium sp. in equids in Louisiana. American Journal of Veterinary Research 50, 575-

77, 1 989

Crabbe B. The foal. I n : The comprehensive guide to equine veterinary medicine, p . 295.

Ster l ing Publ ishing Ltd . , New York. 2007

De Bias I , Ortega C, WinEpiscore 2.0 Freeware, 2000

de la Fuente R, Luzon M, Ruiz-Santa-Quiteria JA, Garcia A, Cid D, Orden JA, Garcia S,

Sanz R , Gomez-Bautista M. Cryptosporidium and concurrent infections with other m ajor

enterophatogens in 1 to 30-day-old d iarrheic dairy calves in central Spain . Veterinary

Parasitology 80, 1 79-85, 1 999

Donnison AM, Ross CM. Animal and human faecal pol lution in New Zealand rivers. New

Zealand Journal of Marine and Freshwater Research 33, 1 1 9-28, 1 999

DuPont HL, Chappel CL, Sterling CR, Okhuysen PC, Rose JB, Jakubowsky W. The

infectivity of Cryptosporidium parvum in healthy volunteers . New England Journal of Medicine

332, 855-859, 1 995

Duranti A, Caccio SM, Pozio E, Di Egidio A, De Curtis M, Battisti A, Scaramozzino P. Risk

Factors Associated with Cryptosporidium parvum I nfection in Cattle A. Zoonoses and Public

Health, Publ ished on l ine ahead of print, doi : 1 0 . 1 1 1 1 /j . 1 863-2378.2008 . 0 1 1 73.x, 2008

Engberg J , On SLW, Harrington CS, Gerner-Smidt P. Prevalence of Campylobacter,

Arcobacter, Helicobacter and Sutterel/a spp. in human fecal samples as estimated by a

reevaluation of isolation methods for Campylobacter. J ournal of Clinical Microbiology, 38, 286-

91 , 2000

Fayer R. Cryptosporidium: a water-borne zoonotic parasite. Veterinary Parasitology 1 26 , 37-

56, 2004

Fayer R. Biology. I n : Fayer R, Xiao L. (eds) . Cryptosporidium and Cryptosporidiosis, 2nd edtn.

Pp 1 -42. C RC Press and I WA Publ ish ing, Boca Ratan FL, USA, 2008

Fayer R, Gasbarre L, Pasquali P, Canals A, Almeria S, Zarlenga D. Cryptosporidium

parvum infection in bovine neonates: dynamic cl in ical , parasitic and immunologic patterns.

International Journal for Parasitology 28, 49-56, 1 998

1 2 1

Fayer R, Ell is W. Paromomyci n is effective as prophylaxis for cryptosporidios is i n da i ry calves.

Journal of Parasitology 79, 771 - 4, 1 993

Fayer R, Santin M, Trout JM, Greiner E. Prevalence of species and genotypes of

Cryptosporidium found in 1 -2-year-old dairy cattle in the eastern Un ited States . Veterinary

Parasitology 1 35, 1 05-1 2 , 2006

Feltus DC, Giddings CW, Khaitsa ML, McEvoy JM. High prevalence of Cryptosporidium

bovis and the deer-l ike genotype in calves compared to mature cows i n beef cow-calf

operations. Veterinary Parasitology 1 5 1 , 1 9 1 -5, 2008

Gajadhan AA, Caron JP, Al ien JR. Cryptosporidium i n two foals. Canadian Veterinary Journal

26, 1 32-34 , 1 985

Garber LP, Salman MD, Hurd HS, Keefe T, Schlater J. Potential risk factors for

Cryptosporidium i nfection in dairy calves. Journal of the American Veterinary Medical

Association 205, 86-9 1 , 1 994

Garcia MM, Lior H, Stewart RB, Ruckerbauer GM, Trudel JAR, Skljarevsky A. I solation

characterisat ion and setrotyping of Campylobacter jejuni and Campylobacter col i from slaughter

cattle. Applied and Environmental Microbiology 49, 667-72, 1 985

Genchi C, Hermon J, Sangali G, Traldi G . La criptosporid iosi nel vitel lo, fattore determ inante

nel la diarrea neonatale. (Calf cryptosporidiosis, a determ inant factor of neonatal diarrhoea) .

Praxis Veterinaria 3 , 4, 1 984

Geurden T, Berkvens D, Martens C, Casaert S, Verycruysse J, Claerebout E. Molecular

epidemiology with subtype analysis of Cryptosporidium i n calves in Belgiu m . Parasitology 1 34,

1 98 1 -87, 2007

Geurden T, Goma FY, Siwila J, Phiri IGK, Mwanza AM, Gabriel S, Claerebout E,

Vercruysse J. Prevalence and genotyping of Cryptosporidium in three cattle husbandry

systems in Zambia. Veterinary Parasitology 1 38, 2 1 7-22, 2006

Giacoboni, Gl, ltoh K, Hirayama K, Takahash i E, M itsuoka T. Com parison of fecal

Campylobacter in calves and cattle of different ages and areas in Japan . Journal of Veterinary

Medical Science 55, 555-59, 1 993

Gibson JA, Hi l l MWM, Huber MJ. Cryptosporidium in Arabian foals with severe com bined

im munodeficiency. Australian Veterinary Journal 60, 378-79, 1 983

1 22

Glaberman S, Moore JA, Lowerly CJ, Chalmers R , Sulaiman I, Elwin K, Rooney PJ,

Mi llar BC, Dooley J S G, Lal AA, Xiao L. Three dr ink ing-water-associated cryptosporidiosis

outbreak, Northern I reland. Emerging Infectious Diseases 8, 631 -3, 2000

Grau, FH. Campylobacter jejuni and Campylobacter hyointestinalis in the intestinal tract and on

the carcasses of calves and catt le. Journal of Food Protection 5 1 , 857-6 1 , 1 988

Grinberg A, Markovics A, Gal indez J, Lopez-Vil lalobos N, Kosak A, Tranqui l lo VM.

Contro l l ing the onset of natural cryptosporid iosis in calves with paromomycin su lphate.

Veterinary Record 1 5 1 , 606-8, 2002

Guerrant RL. Cryptosporidios is : an emerging, h ighly i nfectious threat. Emerging Infectious

Diseases 3 , 5 1 -7, 1 997

Hajduseka, 0., 0. Ditricha, and J . Slapeta. Molecular ide ntification of Cryptosporidium spp. in

an imal and human hosts from the Czech R epublic. Veterinary Parasitology 1 22, 1 83-92, 2004

Harp J A, Jardon P, Atwi l l ER, Zylstra M, Checel S, Goff JP, De Simone C. Field testi ng of

prophylactic measures agai nst Cryptosporidium parvum i nfections in calves in a Californ ia dairy

herd. American Journal of Veterinary Research 57, 1 586-9 , 1 996

Heine J, Pohlenz JFL, Moon HW, Woode GN. E nteric lesions and diarrhoea in g notobiotic

calves monoinfected with Cryptosporidium species. Journal of Infectious Diseases 1 50 , 768-75,

1 984

Harp JA, Woodmansee DB . Resistance of calves to Cryptosporidium parvum: Effects of age

and previous exposure . Infection and Immunity 58, 2237-40, 1 990

Hoar BR, Atwil l ER, Elmi C, Farver TB. An examinatio n of risk factors associated with beef

cattle shedding pathogens of potential zoonotic concern. Epidemiology and Infection 1 27 , 1 47-

55, 2001

Hol l is AR, Wilkins PA, Palmer JE, Boston RC. Bacteremia in equ ine neonatal diarrhea: A

retrospective study ( 1 990-2007) . Journal of Veterinary Internal Medicine 22, 1 203-9, 2008

Huston CL, Wittum TE, Love BC. Persistent fecal Salmonella shedding in f ive dairy herds .

Journal of the American Veterinary Medical Association 220, 650-5, 2002

1 23

Joens LA. Campylobacter and Helicobacter. I n : G yles L, Prescott JF , Songer JG , Thoen CO

(eds) . Pathogenesis of Bacterial I nfections in An imals, 3rd Edtn , Pp. 353-6 1 . Blackwel l

Publ ish ing , Ames, USA, 2004

Kim CW. Cryptosporidiosis in pigs and horses. I n Cryptosporidiosis of Man and Animals, Pp.

1 05- 1 1 . Eds : J. P. Dubey, C. A. Speer, R . Fayer. Boca Ratan, CRC Press . 1 990

Khramtsov NV, Ti l ley MDS, Blunt BA, Montelone BA, Upton SJ. Cloning and analysis of a

Cryptosporidium parvum gene encoding a protein with homology to cytoplasmic form HSP70.

Journal of Eukaryotic Microbiology 42:4 1 6-422, 1 995

Knottenbelt OK, Holdstockand N , Madigan JE . Neonatal syndromes. I n : Equine

neonatology, Medicine and Surgery, 1 st ed. p . 2 1 9 . E lseiver, London, 2005

Konkle OM, Nelson KM, Lunn DP. Nosocomial transm ission of Cryptosporidium i n a veterinary

hospital. Journal of Veterinary Internal Medicine 1 1 , 340-43, 1 997

Lance SE, Mi l ler GY, Hancock DD, Bartlett PC, Heider LE. Salmonella i nfections in neonatal

dairy calves. Journal of the American Veterinary Medical Association 201 , 864-8, 1 992

Langkjaer RB1 , Vigre H, Enemark HL, Maddox-Hyttel C. Molecu lar and phylogenetic

characterization of Cryptosporidium and Giardia from pigs and catt le in Denmark. Parasitology

1 34, 339-50, 2007

Learmonth JJ, lonas G, Ebbet KA, Kwan ES. Genetic characterisation and transm ission

cycles of Cryptosporidium species isolated from humans in New Zealand. Applied and

Environmental Microbiology 70, 3973-8, 2004

Learmonth JJ, lonas G, Pita A, Cowie RS. Seasonal sh ift in Cryptosporidium parvum

transmission cycles in New Zealand. Journal of Eukaryotic Microbiology Suppl, 34S-35S, 2001

Le Blancq SM, Khramtsov NV, Zamani F, Upton SJ, Wu TW. Ribosomal RNA gene

organization in Cryptosporidium parvum. Molecular and Biochemical Parasitology 90 :463-78,

1 997

Leoni F, Mallon ME, Smith HV, Tait A, McLauchlin J. Mult i locus analysis of Cryptosporidium

hominis and Cryptosporidium parvum isolates from sporadic and outbreak-related human cases

and C. parvum isolates from sporadic l ivestock cases in the U nited Kingdom . Journal of Clinical

Microbiology 45: 3286-94, 2007

1 24

Libby SJ, Hasley TA, Altier C, Potter J, Gyles CL. Salmonella. I n : Gyles L, Prescott JF,

Songer JG, Thoen CO (eds) . Pathogenesis of Bacterial I nfections in An imals, 3rd edtn . Pp 1 43-

67 Blackwel l Publ ish ing , Ames, USA, 2004

Magdesian KG. Neonatal foal d iarrhoea. Veterinary Clinics of North A merica, Equine practice

2 1 : 295-3 1 2 , 2005

Majewska AC, Solarczyk P, Tamang L, Graczyk TK. Equine Cryptosporidium parvum

i nfections i n western Poland . Parasitology Research 93: 274-78, 2004

Mallon M, Macleod J, Wastl ing J, Smith H, Reil ly B, Tait A. Popu lation structures and the

role of genetic exchange in the zoonotic pathogen Cryptosporidium parvum. Molecular Genetics

and Evolution 56: 407-1 7 , 2003a

Mallon M, MacLeod A, Wastling J, Smith H, Tait A. Mult i locus genotyping of Cryptosporidium

parvum Type 2: populatio n genetics and sub-structuring . Infection, Genetics and Evolution 3 ,

207-1 8 , 2003b

Mathieu E, Levy DA, Veverkav F, Parrish M, Sarisky J , Shapiro N, Johnston S, Handzel T,

Hightower A, Xiao L, Lee V, York S, Arrowood M, Lee R, Jones JL. Epidem iolog ic and

environmental investigat ion of a recreational water outbreak caused by two genotypes of

Cryptosporidium parvum i n Ohio in 2000. American Journal of tropical Medicine and Hyg iene

71 , 582-9, 2004

McAI I ister T, Olson ME, Fletch A, Wetzstein M, Entz T. Prevalence of Giardia and

Cryptosporidium in beef calves in southern British Columbia. Canadian Veterinary Journal 46,

47-55, 2005

Mclauchl in J , Amar C, Pedraza-Diaz S, Nichols GL. Molecular Epidemiolog ical Analysis of

Cryptosporidium spp. in the Un ited Ki ngdom : Resu lts of Genotyp ing Cryptosporidium spp. in

1 ,705 Fecal Samples from Humans and 1 05 Fecal Samples from Livestock Ani mals. Journal of

Clinical Microbiology 38, 3984-90, 2000

Meanger JD, Marshal! R B . Seasonal prevalence of thermophi lic Campylobacter i nfections in

dairy cattle and a study of infection of sheep. New Zealand Veterinary Journal 37, 1 8-20, 1 989

Morgan U, Xiao L, Fayer R, Lal AA, Thompson ARC. Molecular epidem iology and

system atics of Cryptosporidium parvum. In Cryptosporidium: the Analytical Chal lenge. Eds M .

Smith, K . C . Thompson. Cambridge, R oyal Society o f Chemistry. p p 44-50, 200 1

1 25

Morgan-Ryan UM, Fall A, Ward LA, Hijjawi N, Sulaiman I, Fayer R, Thompson RCA, Olson

M, Lal A, Xiao L. Cryptosporidium hominis n. sp. (Apicomplexa: Cryptosporid i id ae) from Homo

sapiens. Journal of Eukariotic Microbiology 49, 433-40, 2002

Nachamkin I . Campylobacter and Archobacter. I n : Murray PR (ed) . Manual of C l in ical

M icrobiology, 8th ed. , Pp. 902- 1 4, ASM Press, Washington DC, USA, 2003

Naciri M, Mancassola R, Yvore P, Peeters JE. The effect of halofug inone lactate on

experimental Cryptosporidium parvum i nfections in calves. Veterinary Parasitology 45, 1 99-207,

1 993

Naciri M, Lefay MP, Mancassola R, Poirier P, Chermette R. Role of Cryptosporidium

parvum as a pathogen in neonatal diarrhoea complex in suckl ing and dairy calves in France.

Veterinary Parasitology 85, 245-57, 1 999

Neira-Otero P, Munoz-Sald1as N, Sanchez-Moreno M. Rosales-Lombardo MJ. Molecular

characterization of Cryptosporidium species and genotypes in Ch i le. Parasitology Research 97,

63-7, 2005

Netherwood T, Wood JLN, Townsend HGG, Mumford JA. Foal diarrhoea between 1 99 1 and

1 994 in the Un ited Kingdom associated with Clostridium perfringens, rotavi rus, Strongyloides

westeri and Cryptosporidium spp. Epidemiology and Infection 1 1 7, 375-83, 1 996

Nielssen EM. Occurrence and strain diversity of thermophi l ic Campylobacter i n cattle of

different age g roups in dairy herds. Letters in Applied Microbiology 35, 85-9, 2002

Nydam DV, Wade SE, Schaaf SL, Mohammed HO. Number of Cryptosporidium parvum

oocysts or Giardia spp. cysts shed by dairy calves after natural infection . American Journal of

Veterinary Research 62, 1 6 1 2-5 , 200 1

O'Handley RM, Cockwil l C, McAI I ister TA, Jelinski M , Morck DW, Olson ME. Duration of

naturally acqui red g iardiasis and cryptosporidiosis in dairy calves and their associat ion with

diarrhea. Journal of the A merican Veterinary Medical Association 2 1 43, 39 1 -6 , 1 999

Ongerth J E, Stibbs HH. Prevalence of Cryptosporidium i nfection in dairy calves in Western

Washingto n . American Journal of Veterinary Research 50, 1 069-71 , 1 989

Okhuysen PC, Chappell CL, Crabb JH, Sterling CR, DuPont HL. Viru lence of three distinct

Cryptosporidium parvum isolates for healthy adults. Journal of Infectious Diseases. 1 80 , 1 275-

81 , 1 999

1 26

Ortega YR, Arrowood M. Cryptosporidium, Cyclospora and Isospora. I n : Mu rray P R (ed) .

Manual o f Cl in ical M icrobiology, 8 th edtn. Pp 2008- 1 9 . ASM Press, Washington DC, USA, 2003

Peeters JE, Vi l lacorta I, Naciri M, Vanopdenbosch E. Specific serum and local antibody

responses against Cryptosporidium parvum during medication of calves with halofug inone

lactate. Infection and Immunity 6 1 , 4440-5, 1 993

Peng MM, Mark AE, Wilson L, Holland RE, Steven AE, Meshnick R, Lal AA, Xiao L. Genetic

divers ity of Cryptosporidium spp. in cattle in M ichigan: impl ications for understanding the

transmission dynamics. Parasitology Research 90, 1 75-80, 2003

Peng MM, Matos 0, Gatei W, Das, M. Stantic-Pavl inic P, Bern C, Sulaiman IM, G laberman

S, Lal AA, Xiao L. A comparison of Cryptosporidium subgenotypes from several geographic

reg ions. Journal of Eukaryotic Microbiology, Suppl, 288-3 1 S , 2001

Popovic-Uroic T, Patton CM, Nicholson MA, Kiehl bauch JA. Evalu at ion of the i ndoxyl

acetate hydro lys is test for rapid differentiat ion of Campylobacter, Helicobacter and Wolinella

species. Journal of Clinical Microbiology 28, 2335-39, 1 990

Quilez J, Sanchez-Acedo C, del Cacho E, Clavel A, Causape AC. Prevalence of

Cryptosporidium and Giardia i nfections in cattle in Aragon (northeastern Spai n ) . Veterinary

Parasitology 66, 1 39-46, 1 996

Quilez J, Torres E, Chalmers RM, Robinson G, Dei-Cacho E, Sanchez-Azedo E.

Cryptosporidium species and subtype analysis from dairy calves in Spai n . Parasitology 1 35,

1 6 1 3-20, 2008

Roberts VLH, Knottenbelt DC, Will iams A, McKane SA. Suspected prim ary lactose

intolerance in neonatal foals . Equine Veterinary Education 20, 249-5 1 , 2008

Rogers CW, Gee EK, Firth EC. A cross-Sectional survey of Thoroughbred stud farm

management in the North Island of New Zealand . New Zealand Veterinary Journal 55, 302-7,

2007

Rothman K. E pisheet© freeware . Spreadsheets for the analysis of epidem iolog ical data,

2002 .

Roy S , Sarkar S, Batabyal, Amiya A , Pramanik K. Observations on the epidemiology of

bovine cryptosporidiosis in Ind ia . Veterinary Parasitology 1 41 , 330-33, 2006

1 27

Ruest N, Faubert G, Couture V. Prevalence and geographic distribution of Giardia spp. and

Cryptosporidium spp. in dairy farms in Quebec. Canadian Veterinary Journal 39, 697-700, 1 998

Ryan U, Xiao L, Read C, Zhou L, Lal AA, Pavlasek I. Identification of Novel Cryptosporidium

Genotypes from the Czech Republic. Applied and Environmental Microbiology 69, 4302-07,

2003

Santin M, Trout JM. Livestock. I n : Fayer R, X iao L. (eds ) . Cryptosporidium and

Cryptosporidiosis, 2nd edtn. Pp 451 -83 . CRC Press and lW A Publ ish ing , Boca Ratan FL, USA,

2008

Santm M, Trout JM, Fayer R. A long itud ina l study of cryptosporidiosis in dairy cattle from birth

to 2 years of age. Veterinary Parasitology 1 55 1 5-23, 2008

Santin M, Trout JM, Xiao L, Zhoub OL, Greiner E, Fayer R. Prevalence and age-related

variat ion of Cryptosporidium species and genotypes i n da i ry calves, Veterinary Parasitology 1 22

1 03-1 7 , 2004

Sato K, Barlett PC, Kaneene JB, Downes FP. Com parison of prevalence and antim icrobial

susceptibi l ity of Campylobacter spp. isolates from organic and conventional dairy herds i n

Wiscons in . Applied and Environmental Microbiology 70 , 1 442-7, 2004

Savil l M, Hudson A, Devane M, Garrett N, Gi lp in B, Ball A. Elucidation of potential

transm ission routes of Campylobacter i n New Zealand. Water Science and Technology 47, 33-

8 , 2003

Sevinc F, lrmak K, Sevinc M. The prevalence of Cryptosporidium parvum i nfection in the

diarrhoiec and non-diarrhoeic calves. Revue de Medecine Veterinaire 1 54 (5}, 357-6 1 , 2003

Sgorbini M , Corazza M, D'Agostino C, Perrucci S, G iorgi L. Cryptosporidiosis and foal heat

diarrhoea. lppologia 4, 5- 1 0, 2003

Sischo WM, Atwill ER, Lanyon LE, George J. Cryptosporidia on dairy farms and the role

these farms may have in contam inating surface water supplies in the northeastern Un ited

States . Preventive Veterinary Medicine 43, 253-67, 2000

Smerdon WJ, Nichols T, Chalmers RM, Heine H, Reacher MH. Foot-and-mouth disease in

l ivestock and reduced cryptosporidiosis i n humans, Eng land and Wales . Emerging Infectious

Diseases 9, 22- 8, 2003

1 28

Smith KE, Stenzel SA, Bender J B , Wagstrom E, Soderl u n d D , Leano FT, Taylor CM, Belle­

Isle PA, Dan i la R. Outbreaks of enteric i nfections caused by multiple pathogens associated

with calves at a farm day camp. Pediatric Infectious Diseases Journal. 23, 1 098-1 04, 2004

Snyder SP, England JJ, McCh esney AE. Cryptosporidium in immunodeficient Arabian foals.

Veterinary Pathology 1 5, 1 2-7, 1 978

Soba B, Logar J. Genetic classification of Cryptosporidium isolates from humans and calves in

Slovenia. Parasitology 1 35, 1 263-270, 2008

Spano F, Putignani L, McLauchl in J, Casemore DP, Crisanti A. PCR-RFLP analysis of the

Cryptosporidium oocyst wal l protein (COW P) gene determ ines between C wrairi and C parvum,

and between C parvum isolates of human and animal or ig i n . FEMS Microbiology Letters 1 50,

209-1 7, 1 997

Stanley KN, Wallace JS, Cu rrie JE, Diggle PJ, Jones K. The seasonal variation of

thermophi l ic campylobacters i n beef cattle, dai ry cattle and calves. Journal of Applied

Microbiology 85, 4 72-80 , 1 998

Stefanogiannis N, McLean M , Van Mil H. Outbreak of cryptosporidiosis l i nked with a farm

event. New Zealand Medical Journal 1 1 4, 5 1 9-2 1 , 2001

Strong WB, G ut J, Nelson RG. Clon ing and sequence analysis of a h igh ly polymorphic

Cryptosporidium parvum gene encoding a 60-Ki lodalton g lycoprotein and characterization of Its

1 5- and 45-Ki lodalton zoite surface antigen products. Infection and Immunity 68, 41 1 7-34, 2000

Sulaiman IM, H i ra PR, Zhou L, AI-Ai i FM, AI-Shelahi F A, Shweiki HM, lqbal J, Khal i d N,

Xiao L. Unique endem icity of cryptosporidiosis in ch i ldren in Kuwait. Journal of Clinical

Microbiology 43, 2805-09, 2005

Tanriverdi S, Markovics A, Arslan MO, ltik A, Shkap V, Widmer G . Emergence of dist inct

genotypes of Cryptosporidium parvum i n structured host popu lations. Applied and

Environmental Microbiology 72, 2507- 1 3 , 2006

Thompson H P, Dooley JSG, Ken ny J, M cCoy M, Lowery CJ, Moore J E, X iao L. Genotypes

and subtypes of Cryptosporidium spp. in neonatal calves in Northern I reland. Parasitology

Research 1 00 , 61 9-24, 2007

1 29

Thompson JD, G i bson T J, Plewniak F, Jea n mougin F, Higg ins DG. The ClustaiX windows

interface : flexible strateg ies for mu lt iple sequence a l ignment aided by qual ity analysis tools.

Nucleic Acids Research 24, 4876-82, 1 997

Townsend KG, Lance OM. Cryptosporidiosis in calves. New Zealand Veterinary Journal 35,

2 1 6 , 1 987

Trotz-Wi ll iams LA, Jarvie BD, Martin SW, Leslie KE, Pereg rine AS. Prevalence of

Cryptosporidium parvum i nfection in southwestern O ntario and its association with d iarrhoea i n

neonatal dairy calves . Canadian Journal of Veterinary Medicine 46, 349-5 1 , 2005

Trotz-Wil l iams LA, Martin OS, Gatei W, Cama V, Peregrine AS, Martin SW, Nydam DV,

Jamieson F, Xiao L. Genotype and subtype analyses of Cryptosporidium isolates from dairy

calves and hum ans i n Ontario. Parasitology Research 99, 346-52, 2006

Tzi pori S. Cryptosporidiosis in anim als and humans. Microbiological Reviews 4 7 , 84-96, 1 983

Tzi pori S, Campbe l l I . Prevalence of Cryptosporidium antibodies in 1 0 an imal species. Journal

of Clinical Microbiology 1 4 , 455-56, 1 98 1

Tzipori S , Smith M , Halpin C , Angus KW, S herwood D , Campbell I . Experimental

cryptosporidiosis in calves: cl i nical man ifestations and patho logical f indings. Veterinary Record

1 1 2 , 1 1 6-20, 1 983

Uga S, Matsu o J, Kono E, Kimura K, lnoue M , Rai SK, Ono K. Prevalence of

Cryptosporidium parvum i nfect ion and patterns of oocyst shedding i n calves in Japan.

Veterinary Parasitology 94, 27-32, 2000

Varney K . Quarterly review of d iagnostic cases - April to June 2004. Surveillance 31 (3) , 2 1 -5,

2004

Waldron LS, Ferrari BC, Power ML. Glycoprotei n 60 diversity in C. hominis and C. parvum

causing human cryptosporidiosis i n N SW, Austral ia. Experimental Parasitology 1 22 1 24-1 27,

2009

Wesley IV, Wells SJ, Harmon KM, Green A, Schroeder-Tucker L, G lover M , Siddique I .

Fecal shedding of Campylobacter and Arcobacter spp. in dairy catt le. Applied and

Environmental Microbiology 66, 1 994-2000 , 2000

1 30

Widmer G , Lin L, Kap u r V, Feng X, Abrahamsen MS. Genomics and genetics of

Cryptosporidium parvum: the key to u nderstand ing cryptosporidiosis. Microbes and Infection 4,

1 081 -90, 2002

Widmer G , Tzi pori S, Fitchenbaum CJ, G riffiths JK. Genotypic and phenotypic

characterisation of Cryptosporidium parvum

isolates from people with A I DS. Journal of Infectious Diseases 1 78, 834-840, 1 998

Wilkins PA. Disorders of foals. I n : SM Reed, WM Bayly, DC . Se l lon (eds) . Equine Internal

Medici ne . 2nd Edtn . P 1 4 1 9 . Elsevier, Ph i ladelphia, U SA, 2004

Wi nkworth CL, Matthaei CD, and Townsend C R . Prevalence of Giardia and Cryptosporidium

spp. in calves from a region i n New Zealand experienci ng i ntensification of dairying . New

Zealand Veterinary Journal 56 , 1 5-20, 2008

Wohlfender FD, Barrelet FE, Doherr MG, Straub R , a nd Meier H P. Di seases in neonatal

foals. Part 1 : The 30 day incidence of disease and the effect of prophylactic anti m icrobial drug

treatment during the fi rst th ree days post partum. Equine Veterinary Journal 41 , 1 79-85, 2009

Xiao L, Alderisio K, Limor J, Royer M, Lal AA. Identification of species and sources of

Cryptosporidium oocysts in storm waters with a smal l-subunit rRNA-based diagnostic and

genotyping too l . Applied and Environmental Microbiology 66, 5492-98, 2000

Xiao L, Fayer R. Molecu lar characterisation of species and genotypes of Cryptosporidium and

Giardia and assessment of zoonotic transmission. International Journal of Parasitology 38,

1 239-55, 2008

Xiao L, Fayer R, Ryan U, U pton S. Cryptosporidium taxonomy: Recent advances and

Impl ications for public health. Clinical Microbiology Reviews, 1 7 , 72-97 , 2004

Xiao L, Feng V. Zoonotic cryptosporidiosis. FEMS Immunology and Medical Microbiology 52,

309-23, 2008

Xiao L, Herd RP. Epidem iology of equine Cryptosporidium and Giardia i nfections. Equine

Veterinary Journal 26 , 1 4-1 7 , 1 994

Xiao L, Ryan U. Molecular epidem iology. I n : Fayer R, Xiao L (eds . ) , Cryptosporidium and

Cryptosporidiosis, 2 nd Edtn Pp 1 1 9-63. CRC Press and lW A Publish ing , Boca Ratan FL, USA,

2008

1 3 1

Xiao L, Zhou L, Santin M, Yang W, Fayer R . D istribut ion of Cryptosporidium parvum subtypes

in calves i n eastern United States. Parasitology Research 1 00, 70 1 -06 , 2007

Zintl A, Proctor AF, Read C, Dewaal T, Shahaghy N , Fanning S, M u lcahy G. The prevalence

of Cryptosporid ium species and subtypes i n human faecal samples i n I re land . Epidemiology

and Infection 1 37 , 270-7, 2009

1 32

CH APTER 4

AN OUTBREAK OF CRYPTOSPORIDIOSIS AMONG A COHORT OF YOUNG

ADULTS : MOLECULAR AND DESCRIPTIVE EPIDEMIOLOGY AND RISK

FACTOR ANALYSIS

This chapter describes the i nvestigation of an outbreak of gastrointest inal i l lness amongst a

class of veterinary students at Massey University in 2006. The outbreak was i nvestigated by

combin ing traditional epidem iological and molecular techn iques. Apart from a number of

d iagnostic tests performed by commercial laboratories as acknowledged in the text, all the

laboratory analyses were performed at Massey University.

4.1 Summary

An explosive outbreak of gastroi ntestinal i l l ness, which occu rred in New Zealand in 2006 among

a cohort of 96 veter inary students, instigated laboratory and questionnaire-based

epidemiological investigations . There were 25 c l in ical cases among 80 respondents to the

questionnai re (31 % attack rate) . Cryptosporidium parvum was isolated as the sole

enteropathogen from fou r out of seven faecal specimens analysed. A rare G P60 //a A2 1 G4R 1

al lele was identified in two out of two C. parvum isolates successful ly subtyped , indicating a

poi nt-source of i nfection. The results of the laboratory and epidemiological investigations ruled

out a water-borne cause, leaving di rect contact with newborn calves dur ing a calf-handl ing

practicum as the most l i kely point-source of exposure .The investigation fai led to ident ify C.

parvum i n faecal specim ens taken from calves on the farm that provided animals for the

practicu m , but revealed a DNA sequence very simi lar to the 1 8S ribosom al RNA gene of

Cryptosporidium bovis i n one specimen. Assuming that exposure to C. parvum occurred dur ing

the calf-handl ing practicum , the median i ncubation period was 5 days ( range 0-1 1 days) . Al l the

cases were self- l im it ing and m anifested with diarrhoea, abdominal discomfort, and in a smal l

proport ion of cases, vom it ing. Among the putative r isk factors analysed, or ig i nat ing from a rural

background and previous contact with rum inants, were associated with a decreased risk of

i l lness in m ales, but not in fem ales.

4.2 Introd uction

Protozoan parasites belong ing to the genus Cryptosporidium, i n particular Cryptosporidium

parvum and Cryptosporidium hominis, are common causes of gastrointesti nal d isease i n

humans worldwide (Fayer 2008 ) . Whi lst C . parvum i nfects humans and young l ivestock and can

be transmitted zoonotical ly, C. hominis is thought to cycle primarily among humans. Both

species can be transmitted through the i ngest ion of water or food contaminated with oocysts. As

the oocysts are excreted sporulated and ful ly infectious in the faeces , transmission through

di rect contact with infected hosts is also possible.

1 33

Cryptosporidiosis may m anifest sporadical ly or as epidemic outbreaks. Outbreaks as a resu lt of

person to person transmission, ingest ion of contaminated water supply or food, fol lowing

exposure in swimm ing pools and water parks, and contact with farm animals, including among

veterinary students, have been described (Aipert et al . 1 986 ; Combee et a l . 1 986; Pohjola et al .

1 986; Heijbel et al . 1 987 ; R eif et al . 1 989 ; Shield et a l . 1 990; Miron et a l . 1 99 1 ; Mi l lard et a l .

1 994; Anonymous 2000 ; Hel lard et al . 2000; Stefanogiann is et a l . 2001 ; Preiser et a l . 2003;

Mathieu et al . 2004; Causer et al. 2006 ; Jones et al . 2006 ; Ne i ra-Munoz et al . 2007; Turabel idze

et a l . 2007; Wheeler et a l . 2007; Gait et a l . 2008 ; H ajdu et al. 2008; Beaudeau et a l . 2008;

Hoek et a l . 2008 ; Ethelberg et a l . 2009) . Although in m any countries cryptosporidiosis is a

notifiable d isease, the outbreaks are often protracted and difficult to detect because of delays in

the investigations (M iron et a l . 1 99 1 ; M i l lard et al . 1 994; Hel lard et al . 2000 ; Anonymous 2000;

Preiser et a l . 2003 ; M athieu et al . 2004; Causer et a l . 2006; Ne i ra-Munoz et a l . 2007;

Turabel idze et al . 2007 ; W heeler et a l . 2007 ; Hajdu et a l . 2008 ; Beaudeau et al . 2008,

Valderram a et a l . 2009) . I n addition , most previous cryptosporidiosis outbreak invest igat ions

have dealt with outbreaks in open commun ities (Aipert et a l . 1 986; Pohjola et al . 1 986; Levine et

al . , 1 988 ; Reif et al . 1 989 ; Mi ron et al . 1 99 1 ; M i l lard et a l . 1 994; Anonymous 2000; Hel lard et a l .

2000; Preiser et a l . 2003; M athieu et a l . 2004; Causer et a l . 2006; Neira-Munoz et a l . 2007;

Wheeler et a l . 2007; Beaudeau et al . 2008; Hajdu et a l . 2008; Hoek et al . 2008) , requir ing the

use of case-contro l stud ies, which are m ore prone to u ncontrollable confounding than cohort

studies (Anonymous 2000 ; Crombie et a l . 1 981 ; Vonberg and Gastmeier 2007) . Traditional ly,

the most compell ing reason to investigate the outbreaks has been to determ ine the source of

infection in order to prevent further transmission, and l im ited efforts have been d i rected towards

the collect ion of cl in ico-epidemiolog ical data dur ing outbreaks. Little documented c l in ico­

epidemiolog ical i nformation on infect ions in immunocom petent hosts exists (reviewed by

Warren and Guerrant 2008), and knowledge about the natural history of cryptosporidiosis has

m ain ly been obtained fro m experimental i nfection trials i n human volu nteers (Chappel l et a l .

1 999, 2001 ; Okhuysen et al . 1 999) . Furthermore, apart from immunosuppression ( Hunter and

N ichols 2002) , the constitut ional risk factors for cryptosporidiosis are not understood.

I n 2006, the author investigated an outbreak of gastrointesti nal i l lness in a c lass of 96 veterinary

students, i n which C. parvum was isolated as the sole agent from mu lt iple faecal specimens.

The synchronous exposu re of a cohort of young adu lts to wi ld C. parvum faci l itated the

collect ion of cl in ico-epidemiological data and an assessm ent for the presence of exposure and

constitut ional risk factors for cryptosporidiosis in the context of a retrospective cohort study.

The epidem iological features of the outbreak and the resu lts of a risk factor analysis are

reported in th is section .

4.3 Materials and Methods

Sequence of events. On October 9, 2006 ( Day 1 3) , three students from a class of 96 enrol led

in the second year of the Bache lor of Veterinary Science program of Massey University, New

1 34

Zealand, reported a gastrointestinal i l lness to the reg ional public health service. As a result ,

faecal specimens from these students were subm itted for analysis for Campylobacter and

Cryptosporidium to a diagnostic laboratory associated with the same publ ic body. The

specimens tested negative for Campylobacter and one tested positive for Cryptosporidium. The

presence of a s ingle Cryptosporidium-positive specimen precluded the identification of an

outbreak by the authorit ies at that stage. On October 1 0 (day 1 4) , a concerned student f rom the

same class i nformed a faculty member of the Institute of Veterinary, An imal and Biomedical

Sciences ( IVABS) of Massey U n iversity, about the occurrence of several cases of i l l ness among

students enrol led in the same c lass. The student inferred that the cases appeared to be l inked

to a practicum held on Septem ber 26 ( Day 0) in the Large Animal Teaching Un it (LATU) , i n

which the students handled you ng calves o f less than o n e month of age. Duri ng the practicum ,

groups o f students stayed i n a calf barn for about 20 m inutes to practise thoracic auscu ltations ,

as wel l as measuring the rectal temperature and respiration rate. No g loves were provided, but

the students were able to wash hands at the end of the practicum . No other academic or

recreational act ivities had occurred that could have resu lted in exposure of the class to

enteropathogens, and there were no other known cases of gastroenteritis among students from

other classes during the same period. On October 1 1 ( Day 1 5) , an urgent request to conduct an

outbreak investigation was sol ic ited by the author. The request was approved by Massey

University's H uman Eth ics Committee on October 1 3 (Day 1 7) . The same day, the author

addressed the class and i nvited the students to anonymously submit stool speci mens to the

m icrobio logy laboratory of IVABS during the weekend of October 1 3- 1 5 . The d iagnosis of

cryptosporidosis was released the morning of Monday 1 6 October (Day 1 8) . An environmental

investigation at the LATU was i nstigated on October 1 7 (Day 2 1 ) . On October 1 9 (Day 23) , the

author and a un iversity health and safety officer delivered a quest ionnaire to the class. A

second supplementary questionnaire was del ivered on May 25, 2007.

Environmental i nvestigation. The LATU was visited on October 1 7 . No calves were present

in the faci l ity at the t ime of the v isit. The faci l ity received potable water from a non-artesian bore,

which was m ai ntained by u niversity staff. The depth of the bore was 200 m below ground leve l .

Water sanitat ion consisted o f treatment with utravio let l ight a t the point o f use. A 1 OOL water

sample was col lected at one of the dr inking water taps and fi ltered on-site usin g a commercial

f i lter (Fi ltaMax®, ldexx Laborato ries, USA) . Flow rate was adjusted to 2Umin through the f i lter.

The sample was tested for the p resence of Cryptosporidium oocysts as described below.

The dairy farm from which the newborn calves were obtained was visited the same day in order

to assess whether C. parvum was cycl ing in calves. The farm was situated about 5 km from the

teaching faci l ity. The calves actually used for the practicum had been sold , and thus could not

be traced. As the calv ing season had ended in September, other calves less one month of age

were not present on the farm . H owever, faecal specimens from nine calves older than 30 days

of age were col lected and subm itted on ice to the m icrobiology laboratory of IVABS.

1 35

M ic ro biological analysis. Smears from the human and bovine stoo l specimens were exam ined

m icroscopically for the presence of Cryptosporidium oocysts and Giardia using a com mercial

imm unofluorescence test kit (MeriFiuor C/G , Mediridian B ioscience, C inc in nat i , Ohio, USA). In

addit ion, the human specimens were tested for the presence of Salmonella spp . , g roup A

rotavirus, norovirus genog roups 1 and 2, and Campylobacter spp.

Testing for Salmonella spp. and g roup A rotavirus was performed as previously described

( Section 3 . 1 ). Testing for norovi rus genogroups 1 and 2 was performed by means of reverse

t ranscriptase real-time PCR at the New Zealand Norovirus reference laboratory ( I nstitute of

E nvironmental Science and Research, Porirua City) . Test ing for Campylobacter was performed

o n ly in three human speci mens that had been subm itted to the diagnostic laboratory associated

with the publ ic health service (see above) us ing routine d iagnostic procedures.

Testing for Cryptosporidium oocysts in dr inking water was performed by Mr . Anthony P ita,

Protozoa Research Un it , Massey Un iversity. Briefly, the f i lter was taken apart and the foam

disks were homogenised with elution buffer. The eluant was col lected and centrifuged at 1 500g

or 1 5 m in . The pel let vo lume was recorded and supernatant aspirated from the tube unti l the

volume above the pellet was 5 m l . Water was added so the pellet vo lume was 5% or less in a

1 0 m l sample. An anti- Cryptosporidium and anti -Giardia immunomag netic separat ion kit

(Dynabeads® GC-Combo IMS kit: l nvitrogen , Carlsbad, CA, USA) was used for the separation

of the oocysts. The sample was then dried onto a sl ide, stai ned with F ITC label led anti- Giardia

and anti- Cryptosporidium antibod ies (Aq uaGio™ G/C kit, Waterborne Inc . , New Orleans, LA,

U SA) and screened m icroscopically for the presence of apple-green oocysts us ing an

epif luorescence m icroscope .

In October 2007, genomic DNA was extracted from the Cryptosporidium-positive human

specimens and from the n ine bovine specimens us ing DNA extraction kits as described in

Section 3 . 5 . For the taxonomic identif ication of Cryptosporidium, a fragment of the 1 8S

ribosomal RNA ( 1 8S rRNA) gene was amplified using a nested PCR fol lowed by sequence

analysis . The identified Cryptosporidium were subjected to subtyping by means of PCR

fol lowed by sequencing of the polymorphic reg ions of the sporozoite 60-kDa g lycoprotein

(G P60) and the 70 kDa heat shock protein (HSP70) genes. The PCR and sequenci ng

procedures for the 1 8S r R NA and HSP70 genes have been described in Sections 3 .2 and 3 .5 .

The GP60 sequences were al igned with other s im i lar sequences reported in Genbank.

Questionnaires. Two questionnaires were del ivered to the c lass on different dates and were

anonymously completed i n the classroom by interested students . The fi rst questionnaire (01 ;

Appendix 4 . 1 ) was de livered on 1 9 October 2006. The responses to 01 were used to el icit the

epidemic curve and the demographics and d istribution of the cli nical s igns, and assess the

sign ificance of putative constitutional r isk factors for i l l ness, such as gender, nationality, chronic

1 36

use of m edicat ion, domestic background (rural or u rban) , and the frequency of previous physical

contact with rum inants . A calendar report ing the dates of the main academic activities held

between September 1 8 and October 1 8 was provided , in order to faci l itate the correct

identification of the dates of i l l ness by the affected students . In order to m in im ise response bias,

01 did not i nclude specific questions addressing the calf-handling pract icu m . Thus, i n order to

assess the rate of attendance at the calf-handl ing practicu m and the source of dr inking water

(tap water, other) consumed by the students during the practicum, a second supplementary

questionnai re {02) was n ecessary. Th is question naire was delivered on 24 May 2007and

included on ly three stra ightforward questions (Appendix 4 .2 ) .

Analys i s o f t h e data. Responses to 0 1 and 02 were manipulated us ing an electronic

spreadsheet and used for the description of the demographics of the outbreak, the natural

h istory of d isease, the epidemic curve , and for a risk factor analysis. A case was defi ned as any

student report ing abdom inal discomfort, and/or diarrhoea, and/or vomiti ng between September

1 8 and October 1 8 2006, in 01 .

The statistical s ign ificance of the differences between proportions was assessed using the two­

tailed Fisher's exact test. For the analysis of risk factors, the strength of associat ion between

exposure or constitutional variables and the outcome bi nary variable (case; non-case) was

assessed us ing the relative risk (RR) and its probabi l i ty u nder a nu l l model of R R= 1 , and the

95% confidence interval (C l ) . Variables used were : gender, national ity (New Zealand cit izens;

non-New Zeal_anders) , background (rura l ; urban) , h istory of previous physical i nteraction with

rum inants (no interact ion ; less than one-two interactions/year; more than o ne/two

interact ions/year) , taking chronic medicat ion (yes; no) , attendance at the calf handl ing practicum

(attended ; not attended ) , drinking tap water at the same practicum (yes; no) (see Appendices

4 . 1 -4.3 ) . The presence of bivariate interactions between i ndependent variables was explored

by stratif icat ion of the data i nto 2x4 tables and calcu lation of the Wald's stat istic for homogeneity

of strata ( Rothman et al. 2008). Adjusted M antei-Haenszel relative risks ( Mantel and Haenszel

1 959) were calculated for the comparisons not showing heterogeneity of strata by the Wald's

test (as indicated by p > 0 .05) . All the calculations were performed using Rothman's Episheet

software ( http ://members .aol .com/krothm an/episheet .x ls , downloaded 27 August 2005) . The

modest sample size precluded a more complex mu ltivariate analysis (see below) .

4.4 Results

Response to the req uest for co-operation. Seven students subm itted stool specimens to

IVABS between 1 4- 1 5 October 2006. One of the specimens, of a l iquid consistency, had been

collected by a student on October 8 and s ince then held in a home refrigerator. Th ree formed

specimens col lected on October 9 by the students who contacted the reg ional public health

service (see above) were reclaimed by the same students and resubmitted to IVABS. An

addit ional three formed specimens were d ispatched to I VABS on October 1 5, with no

1 37

accompanying i nformation . A total of 80 {83%) students completed 0 1 , and 64 (67%)

completed 02. 0 1 was fully answered by 75 {93%) students (Appendix 4 .3 ) .

M icrobiological findings. Al l the human specim ens tested negative tor g roup A rotavi rus,

Salmonella spp. , norovirus and Giardia. Four specimens tested positive tor Cryptosporidium

oocysts by immunofluorescent stain . One Cryptosporidium-positive specimen had been

col lected on October 8 , two on October 9, and one October 1 4. One Cryptosporidium-positive

specimen collected on October 9 had previously tested positive also at the diagnostic laboratory

(see above) .

Only three out of tour Cryptosporidium-positive specimens yielded editable 1 8S rRNA gene

sequences. Two of the positive sequences were i ndist ingu ishable from the C. parvum Type A'

1 8S rRNA gene sequence reported in Genbank u nder accession number L 1 6996 (Johnson et

al . 1 995) ; the th i rd sequence was indistingu ishable from the Type B' polymorphic copy of the

1 8S rRNA gene of C. parvum ( Le Blanq et a l . 1 997) .

The GP60 locus could be ampl ified from two Cryptosporidium-positive human specimens and

the HSP70 from all tour specimens. Al l the ed ited GP60 and HSP70 sequences were longer

than 740 and 380 base pairs, respectively, and com prised the polymorphic repeat reg ions

(Appendix 4 .4) . The two GP60 sequences were indisti nguishable, and belonged to the al lel ic

g roup //a A21 G4R 1 (according to the nomenclature proposed by Sulaiman et a l . 2005) . There

were no previous reports of th is G P60 sequence in New Zealand, and a search in Genbank

revealed that it h ad only been reported in some calves in the Un ited States (GenBank

accession number 006305 1 4) . Most interest ing ly, the same sequence was also found in an

isolate from an unpublished outbreak of cryptosporidiosis among veterinary students in the

U nited Kingdom (GenBank accession number EU262602. 1 ) .

There were two different al leles at the HSP70 locus , differ ing by the addit ion of a s ing le thym i ne

downstream of the repeat region . One HSP70 al lele was common to three isolates, and the

second was only present in one isolate. The same polymorph ism was conserved after re-edit ing

of the second sequence in 2009, arg u ing against an in itial edit ing error (Appendix 4.4) .

Al l the bovi ne specimens from the source farm were negative tor Cryptosporidium oocysts, but

one specimen yielded an amplicon of a sequence that was 99% sim i lar to the 1 8S rRNA gene

of C. bovis (Santin et al . 2004) (Appendix 4 .4) . An identical sequence was not retrieved in

Genbank. No Cryptosporidium oocysts were observed in drinking water collected dur ing the

visit to the large anim al teaching faci lity.

Clin ico-epidemiologal features of the outbreak. The demographic characteristics of the

class, as el icited from the responses to 01 , are reported in Table 4 . 1 . Seventy percent of the

1 38

respondents to 01 were female and 28% were non-New Zealand cit izens. These values were

very s imi lar to those of the entire class rol l ( not shown) . The range and mean of the age of the

students were very s im i lar in both genders. There was no statistical ly-s ign ificant d ifference

between the proport ions of m ales and females com ing from a rural or u rban background (two

tai led Fisher's exact P=0 .3 ) . The proport ions of males and females that had never handled

ruminants prior to the calf-handl ing practicum were not s ign i ficantly different (P=0 . 1 9) . In both

genders, the proportion of students that had previously handled ruminants more than once­

twice was greater in students from a rural background (not shown) . A greater proportion of

males had previously hand led ruminants more than once-twice a year than females (P=0.0096) .

There were 25 cases among the 80 respondents to 01 (3 1 %) , and 1 5 among the 64

respondents to 02 (23%) . The difference in the proportion of cases between 01 and 02 were

not statistical ly significant (two-tailed Fisher's exact P=0.35) . N ine out of 25 (36%) cases sought

medical advice during the course of the i l lness . All the respondents to 02 had attended the calf­

handl ing practicum and none reported having d runk tap water at LATU, ru l ing out a water-borne

source of infection there .

The duration of i l l ness and distribution of the c l in ical signs, as elicited by the responses to 0 1 ,

are reported i n Fig ure 4. 1 . N otice that the median duration of the i l lness was 5-6 days (range 2-

26 days) , and that the m ajority of cases suffered from diarrhoea and abdom inal d iscomfort ,

whi lst vomiti ng was only reported in six cases. The epidemic curve is reported in F igure 4.2 .

Only 2 out of 25 (8%) cases reported an i l lness start ing before the date of the calf-handl ing

practicu m ; the median and mode of the date of onset of the i l l ness coincided on Day 5 after the

same practicu m .

Risk factor a nalysis. The resu lts o f the analysis for the presence o f risk factors are

summarised in Table 4 .2 . There were no differences in the proportion of cases between m ales

and females (p=0.8) , New Zealand cit izens and non-New Zealanders , overall (P=0 .3) , and in

each gender separately. I n both genders, the habit of not wash ing hands after practicums was

not significantly associated with an i ncreased r isk of i l l ness (P=0.4 in females and 0.6 in m ales) .

No s ign ificant difference in the risk of i l l ness was observed between students from rural or

u rban background (P=0.3) . However, there was a reduction of the r isk of i l l ness in males from

rural backg round as compared to their urban counterparts ( R R=O.OO; p=0 .008), as wel l as

compared to rural and urban females (the statistical comparisons between rural males and rural

and urban females not shown but the data are reported in Table 4. 1 ) . I n agreement with this

finding , there was a greater risk of i l lness in males that had n ot handled rum i nants (P=0 .0 1 ) , or

handled rumi nants up to twice a year (p=0 .0 1 ) as compared with thei r counterparts, but these

associations were not seen among females. However, as said, a greater proportion of males

from rural background had reported handl ing ruminants more than once-twice a year than

females (Table 4 .2 ) .

1 39

Five students reported us ing chronic med ication , of which three were defined as cases

(Appendix 4 .3 ) . In terestingly, these three cases spontaneously reported the chron ic use of anti­

asthma inhaled steroids (not shown) .

Age range (mean)

Nu mber of students that sought med ica l advice during the i l lness/total

respondents

Number of students that subm itted a faecal

specimen/tota l respondents

Smokers

Students of rural backgrou nd/u rban

background

Nu mber of students that never h and led

ru m i na nts/total respondents

Number of students that hand led ru m i nants more

than once-twice a year/tota l respon dents

Males Females

1 9-34 (22.5) 1 8-37 (22 .5)

1 /24 8/56

1 /24 6/56

0

1 1 /23 1 9/56

7/24 25/56

1 3/24 1 3/56*

Table 4.1 Demographic characteristics of the outbreak of cryptosporidiosis among a class of veterinary

students reported in Section 4. 1 . Asterisks denote a signif icant d ifference between the proportions in both

genders ( Fisher's exact test, p<0. 05}

1 40

2

2

- -Abdominal Diarrhoea

discomfort

4 5 6 8 1 0 1 4 1 5 1 6 1 7 26

Duration of illness (in days)

1 5

6

0 0 0

Vomiting Abdominal Abdom inal Diarrhoea + Abdominal

discomfort + discomfort + vomiting discomfort +

diarrhoea

Sym ptoms

vomiting diarrhoea +

vomiting

Figure 4.1 Duration of i l l ness (upper graph) and distribution of symptoms (lower graph) in the outbreak of

cryptosporidiosis, as el icited from the responses to questionnaire 0 1 . The number of respondents in the

lower g raph is reported above each bar.

1 41

1 0

: I 7 '

6

5

4 '

3

2

I I 1 �1 I -8 -7 -6 -5 -4 -3 -2 - 1 0 2 3 4 5 6 7 8 9 1 1

Figure 4.2 Epidemic curve of the outbreak of cryptosporidiosis as e l icited by the responses to 01

1 42

Putative risk factor Factor Factor RR (Cl ; p- x2 (questionnaire number) present absent value) Wald

test Number of Number of cases/total cases/total

resQondents resQondents ExQosure factors Attended the calf-handling practicum (02) 1 5/64 n/a n/a n/a

Drank tap water during the calf-handling 010 1 5/64 n/a n/a practicum (02)

Do not wash hands after practicums (01 ) all 4/1 3 2 1 /67 0.98 (OA - 2 A ;0.9)

females 2/4 1 6/52 1 .6 (0.5 - 4. 7; OA) p=0.30 males 2/9 5/1 5 0 .66 ( 0 . 1 - 2 . 7 ; 0 . 6)

MHrr 1 .03 (OA - 2 . 5; 0.95)

Constitutional factors Gender = male 7/24 1 8/56 0.9 (OA - 1 .9 ; 0.8) n/a

Non New Zealander (01 ) all 9 /23 1 6 /57 1 A (0. 7 - 2.7; 0.33) females 5 /1 5 1 3 /4 1 1 .0 (OA - 2 A ; 0.9)

P=0.22 males 4 /8 3 /1 6 2.7 ( 0 . 7- 9. 1 ; 0 . 1 2)

MHrr 1 A (0 . 7-2 .7 ; 0.33)

Rural background (01 ) all 1 7/49 7/30 1 .5 (0. 7 - 3 . 1 5; 0.3) females 7/19 1 1 /37 1 .2 (0 .5 - 2.7; 0 . 6) n/c males 0/1 1 6/1 2 0 ; P=0.008

MHrr 0.7 (0.3- 1 A ; 0. 3 1 )

Never handled ruminants (01 ) all 1 3/32 1 2/48 1 .6 (0.9 - 3. 1 ; 0. 1 4) females 8/25 1 0/31 1 (0.5 - 2 . 1 ; 0.98) P=0.025 males 5/7 2/1 7 6 ( 1 . 5 - 24; 0 . 0 1 )

MHrr not calculated

Handled ruminants up to twice a year, or never (01 )

all 20/54 5/26 1 .9 (0.8 - 4.6; 0 . 1 ) females 1 4/43 4/1 3 1 .05 (OA - 2.7; 0.9) P=0.07 males 6/1 1 1 / 1 3 7 ( 1 - 50; 0 . 0 1 )

MHrr 1 .8 (0.8 - 4 ; 0 . 1 1 )

Table 4.2 Risk factor analysis of the outbreak of cryptosporidiosis reported i n Sect ion 4. 1 . R R , relative

r isk; MHrr, Mantei- Haenszel relative ri sk; C l , 95% confidence intervals; n/a, not appl icable;nlc=not

calcu lated due to a value of zero in the data

4.5 D iscussion

The present outbreak provided a un ique opportunity to perform the i nvestigation in the context

of a retrospective cohort study. As in other investigations of outbreaks of cryptosporidiosis ( Reif

et a l . 1 999 ; Anonymous 2001 ; Mathieu et al . 2004; Wheeler et a l . 2007), the avai labi l ity of on ly

four laboratory-confi rmed cases was a l im i tation, which could have had implicat ions o n the

specificity of the case defin it ion. For i nstance, two cases reported an i l lness before the

1 43

practicum (F igure 4 .2) but nevertheless, were i ncluded in the analysis. Their i nclusion was

necessary because the same students may have also suffered from cryptosporidiosis. On the

other hand, the fact that the investigation could be performed with on ly short delay was a

strength , which most l ikely prevented recal l bias i n 01 .

The finding of C. parvum as the sole agent i n multiple specimens corroborated the occurrence

of an outbreak caused by th is parasite species. The outbreak was in it ial ly overlooked by the

health authority, as on ly one out of three faecal speci mens from the students tested positive for

Cryptosporidium at the diagnostic laboratory, which precluded the identification of a cluster of

cases. In this case, the authorities m ay have over-emphasized the value of the laboratory test

results. To the author's knowledge , the diagnostic laboratory used a commercial ant igen­

detection kit for the analysis of cryptosporidiosis, whereas the the same specimens were tested

us ing immunofluorescence at M U . Due to the potential for massive outbreaks of

cryptosporidiosis, it would be prudent to assess the sensitivity of the d iagnostic methods

employed by the d iagnostic laboratories in New Zealand.

The resu lts of the molecular analysis of the C. parvum isolates supported the hypothesis of a

point source oubreak, as in New Zealand the l ikel ihood of f inding such a rare G P60 al lele i n

isolates from indiv iduals acquir ing infections from i ndependent sources wou ld have been very

smal l . The results of the responses to 02 and the water testing ruled out a water-borne

outbreak, leaving the direct contact with newborn calves at the practicum as the on ly l i kely route

of transmission.

Excluding the two cases with an onset on Days ( -4) and ( -8) (wh ich were probably unrelated to

the outbreak), if transmission dur ing the calf-handl ing practicum is assumed, the median and

mode of the incubat ion period coincide at five days post transm ission , and the range is between

0- 1 1 days post-exposure (F igure 4. 1 ) . These values fit the median and range observed in

experimental infect ion trials i n human volunteers for which most current knowledge of the

natural history of cryptosporidiosis is derived (Chappel l et a l . 1 999; Chappel l and Okhuysen

2006; Okhuysen et al . 1 999}, and t hose extrapolated in an early retrospective outbreak

invest igation (M i l lard et al. 1 994) . The self- l imit ing diarrhoea! i l l ness accompan ied by abdominal

discomfort and in a small proport ion of cases also vom it ing (Figure 1 ) were consistent with the

resu lts of previous retrospective studies in immunocompetent populations (Wolfson et al. 1 985 ;

Mac Kenzie e t a l . 1 994 ; Mathieu e t a l . 2004; Anonymous 200 1 ; Causer et a l . 2006} . However,

vom it ing had been reported by 82% of the cases in one outbreak of cryptosporidiosis from

fresh-pressed apple cider (M i l lard et al. 1 994} .

As i n other published cryptosporidiosis outbreak i nvestigations (Hel lard et a l . 200 0 ; Anonymous

2001 ; Turabel idze et al . 2007} , an environmental source of i nfect ion could not be

microbiologically traced. Unfortunately, calves less than one month of age were not avai lable for

1 44

sampl ing , and so C. parvum cou ld not be detected, as i n calves the oocyst shedding period is

usual ly short and concentrated in the fi rst three weeks of l ife (see Section 1 .6 . 1 ) . The f inding of

a nucleotide sequence mostly sim i lar, but n ot identical to the sequence of the 1 8S rRNA gene of

C. bovis in calves was not surpris ing , given the extensive genetic diversity revealed i n genus

Cryptosporidium since the advent of molecular tools for the genetic characterisation of the

isolates . This sequence remains of uncertain taxonomic position ing and pathogenic

sign if icance . Also the s ing le nucleotide polymorph ism observed i n the H SP70 locus was not

surpris ing . I ndeed, genetic differences between Cryptosporidium isolates had been documented

in previous outbreaks in humans (G iaberman et al . 2002; Leoni et a l . 2007) , and in our previous

investigations of enzootic cryptosporidiosis in foals (see Section 3.2) . In th is case the genetic

polymorphism in the HSP70 locus m ay have reflected a PCR or sequencing artefact, or the

presence of genetical ly heterogeneous parasites.

In the p resent outbreak, no sign ificant associations between any exposu re or constitutional

factor and the r isk of i l l ness were observed at the bivariate level, but some associations reached

a statistical sign ificance after stratif icat ion of the data. In particular, males com ing from a rural

background had a lower risk of i l lness than urban males and rural and urban fem ales (Table

4.2) . This risk-reduction should not be viewed as evidence of an effect of gender on the

susceptib i l ity to C. parvum, as this effect was confounded by the presence of a g reater

proport ion of males reporting frequent contact with ruminants than females (Table 4 . 1 ) . A

plausible explanation for the risk-reduction i n males from rural backgrou nd could be that of the

presence of a greater rate of immunity to C. parvum i n males, acqu i red throug h previous

exposures d uring repeated contacts with ruminants . Unfortunately, the modest size of this

cohort p recluded a more exhaustive m u ltivariate analys is to disentangle the relat ionships

between the variables, as in it ial explorat ion of logistic regression models including the

independent variables of gender, backg round and the frequency of handling ruminants, resu lted

in an unbalanced design , perhaps due to the small size of the cohort ( not shown) . l t is ,

however, noteworthy, that the effect of previous exposure on the susceptibi l ity to

cryptosporidiosis is not wel l understood. So far th is effect has only been i nvestigated by m eans

of experimental infections in human volu nteers, with confl ict ing resu lts. As would be expected,

vo lunteers experimental ly pre-exposed to C. parvum showed a partial resistance to re- infection

in one study (Okhuysen et al . 1 999) . However, in another experimental infection study,

volunteers with pre-existing anti- Cryptosporidium antibodies exhibited a more severe disease

than their sero-negative counterparts (Chappel l et al. 1 999) .

I n conclus ion , an explosive outbreak of cryptosporidiosis i n New Zealand caused by a rare

subgenotype of C. parvum is reported. The attack rate was 3 1 %. The infections were most

l ikely transm itted through di rect contact with young calves . The i l lness was self- l im iting i n al l

cases, with a median incubation period of 5 days. Symptoms included diarrhoea and abdomina l

discomfort, and vom iting in a l im ited proport ion of cases. A sig nificant d isease-risk-reduction in

1 45

m ales comi ng from a rural background, attributable to an immunity acqu i red through previous

exposures to C. parvum i n the farm environment, was observed . To the author's knowledge,

th is is the first study in which a constitut ional risk factor for cryptosporidiosis other than

immunosuppression was identified . However, due to the modest size of the cohort, th is study

can be viewed as a hypothesis-generati ng rather than test ing, and so the results should be

corroborated by large-scale epidemiological studies .

4.6 R EFERENCES

Alpert G, Bell LM, Kirkpatrick CE, Budnick LD, Campos JM, Friedman HM, Plotkin SA.

Outbreak of cryptosporidiosis in a day-care center. Pediatrics 77, 1 52-7, 1 986

Anonymous. Protracted outbreaks of cryptosporid iosis associated with swimming pool use -

Ohio and Nebraska, 2000. Centers for Disease Control and Prevention . Morbidity and Mortality

Weekly Report 200 1 ; 50 : 406-4 1 0 .

http ://www.cdc.gov/mmwr/preview/m mwrhtml!m m50 20a3 .htm. Accessed February 2009

Beaudeau P, de Valk H, Vail lant V, Mannschott C , Ti l l ier C, Mouly D, Led rans M . Lessons

learned from ten investigations of waterborne gastroenteritis outbreaks, France, 1 998-2006.

Journal of Water and Health 6, 491 -503, 2008

Causer LM, Handzel T, Welch P, Carr M , C u l p D, Lucht R, M udahar K, Robinson D,

Neavear E, fenton S, Rose C, Craig L, Arrowood M Wahlquist S, Xiao L, Lee YM, M i rel L,

Levy D, Beach MJ, Poquette G, Dworkin MS. An outbreak of Cryptosporid ium hom in is

infection at an I l l ino is recreational waterpark. Epidemiology and Infection 1 34, 14 7-1 56, 2006

C happell CL, Okhuysen PC. Cryptospor id iosis in health adu lt volu nteers. I n :

Cryptosporidiosis, t h e analytical chal lenge. M . S m ith and K C Thompson (eds) . The Royal

Society of Chemistry, Cambridge, UK, Pp 62-72, 200 1

Chappell CL, Okhuysen PC, Sterl ing CR, Wang C, Jakubowski W, DuPont H L . I nfectivity of

Cryptosporidium parvum in healthy adu lts with pre-exist ing anti -C. parvum serum

immunoglobul in G. A merican Journal of Tropical Medicine and Hygiene 60, 1 57-64, 1 999

Combee CL, Col l i n ge ML, B ritt EM. Cryptosporidiosis in a hospital-associated day care

center. Pediatric Infectious Diseases 5, 528-32, 1 986

Crombie IK. The l imitations of case-control stu dies In the detection of environmental

carcinogens, Journal of Epidemiology and Community Health 35, 28 1 -7, 1 98 1

1 46

Ethel berg S. Lisby M , Vestergaard LS, Enemark HL, Olsen KEP, Stensvold CR, Nielsen

HV, Porsbo LJ , Plesner AM, Molbak K. A foodborne outbreak of Cryptosporidium hominis

i nfection . Epidemiology and Infection, 1 37 , 348-56, 2009

Fayer R. Biology. I n : Fayer , R . , Xiao, L. ( Eds.) , Cryptosporidium and Cryptosporidiosis, second

ed. C R C Press and lW A Publ ishing, Boca Ratan, FL, pp. 1 -42, 2008

Gait R, Soutar RH, Hanson M, Fraser C, Chal mers R. Outbreak of cryptosporidiosis among

veterinary students . Veterinary Record 1 62 , 843-5, 2008

Glaberman S, Moore JA, Lowerly CJ, Chalmers R , Sulaiman I , Elwin K , Rooney PJ,

Mil lar BC, Dooley J S G, Lal AA, X i ao L. Three dri nking-water-associated cryptosporidiosis

outbreak, Northern I reland. Emerging Infectious Diseases 8, 631 -3, 2002

Hajdu A, Void L, Ostmo TA, Helleve A, Hel gebostad S R , Krogh T, Robertson L, d e Jong

8, Nygard K. I nvest igatio n of Swedish cases reveals an outbreak of cryptosporidiosis at a

Norweg ian hote l with possible l i nks to in-house water systems. BMC Infectious Diseases 8, 1 52,

2008

Heij bel H, Slaine K, Seigel B, Wall P, McNabb SJ, G i bbons W, lstre G R . Outbreak of

diarrhea in a day care center with spread to household m embers: the role of Cryptosporidium.

Pediatric Infectious Diseases Journal 6 , 532-5, 1 987

Hellard M E , Sinclair Ml, Fairley CK, And rews RM, Bai ley M, Black J , Dharmage SC, Kirk

MD. An outbreak of cryptosporidiosis in an urban swim ming pool : why are such outbreaks

difficult to detect? Australian and New Zealand Journal of Public Health 24, 272-5, 2000

Hoek M R , Ol iver I, Barlow M, Heard L, Chalmers R, Paynter S. Outbreak of Cryptosporidium

parvum among chi ldren after a school excursion to an adventure farm , south west Eng land.

Journal of Water and Health 6, 333-8, 2008

Hunter PR, Nichols G. E pidemiology and cl in ical featu res of Cryptosporidium i nfect ion in

immu nocomprom ised patients. Clinical Microbiology Reviews 1 5 , 1 45-54, 2002

Johnson DW, Pieniazek NJ, Griffi n DW, Misener L, Rose JB. Development of a PCR

protocol for sensit ive detection of Cryptosporidium oocysts in water samples Applied and

Environmental Microbiology 6 1 , 3849-55, 1 995

1 47

Jones M, Boccia D, Kealy M , Salkin 8, Ferrero A, Nichols G, Stuart J M . Cryptosporidium

outbreak l inked to interactive waterfeature, UK : Importance of guidel ines. EuroSurveil/ance 1 1 ,

1 26-8, 2006

LeBiancq SM, Khramtsov NV, Zamani F, Upton SJ, Wu TW. R ibosomal RNA gene

organization in Cryptosporidium parvum. Molecular and Biochemical Parasitology 90, 463-78,

1 997

Leoni F, Mallon ME, Smith HV, Tait A, McLauchl in J. Mult i locus analysis of Cryptosporidium

hominis and Cryptosporidium parvum isolates from sporadic and outbreak-related human cases

and C. parvum iso lates from sporadic l ivestock cases in the U nited Kingdom . Journal of Clinical

Microbiology 45: 3286-94, 2007

Levine JF, Levy MG, Walker RL, Crittenden S. Cryptosporid iosis i n veterinary students.

Journal of the A merican Veterinary Medical Association 1 93, 1 4 1 3-4, 1 988

Mantel, N. , Haen szel, W. Statistical aspects of the analysis of data from retrospective studies

of disease. Journal of the National Cancer Institute 22, 7 1 9-7 48, 1 959

M athieu E, Levy DA, Veverka F, Parrish M, Sarisky J, Shapiro N, Johnston S, Handzel T,

H ightower A, X iao L, Lee V, York S, Arrowood M, Lee R , Jones JL. Epidem iolog ic and

environmental i nvestigation of a recreational water outbreak caused by two genotypes of

Cryptosporidium parvum in Oh io in 2000 . American Journal of Tropical Medicine and Hygiene

7 1 , 582-9, 2004

M i l lard PS, Gensheimer KF, Addiss DG, Sosin DM, Beckett G A, Houc k-Jankoski A,

H udson A. An outbreak of cryptosporidiosis from fresh-pressed apple cider. Journal of the

A merican Medical Association 2 72, 1 592-6, 1 994

M i ron D, Kenes J, Dagan R . Calves as a sou rce of an outbreak of cryptosporidiosis among

young chi ldren in an agricultural closed comm unity. Pediatric Infectious Diseases Journal 1 0,

438-41 , 1 99 1

Neira-M unos N , Okoro C , M cCarthy NO. Outbreak of waterborne cryptosporidiosis associated

with low oocyst concentrations . Epidemiology and Infection 1 35, 1 1 59-64, 2007

Okhuysen PC, Chappell CL, C rabb JH, Sterl i ng CR, Du Pont HL. Virulence of three distinct

Cryptosporidium parvum isolates for healthy adu lts. Journal of Infectious Diseases 1 80 , 1 275-

8 1 , 1 999

1 48

Pohjola S, Oksanen H, Jokipii L, Jokipi i AM . Outbreak of cryptosporidiosis among veterinary

students. Scandinavian Journal of Infectious Diseases 1 8 , 1 73-8, 1 986

Preiser G , Preiser L, Madeo L. An outbreak of cryptosporidiosis among veterinary science

students who work with calves. Journal of the A merican College Health 5 1 , 2 1 3-5, 2003

Reif JS, Wimmer L, Smith JA, Dargatz DA, Cheney J M . Human cryptosporidiosis associated

with an epizootic in calves. A merican Journal of Public Health 79, 1 528-30, 1 989

Rothman KJ, G reen land S, Lash T . I n : Modern Epidemiology. Lippincott Wi l l iams and Wi lk ins ,

Phi ladelph ia, PA, USA, p. 227, 2008

Santin M, Trout JM, Xiao L, Zhou L, G reiner E, Fayer R. Prevalence and age-related

variat ion of Cryptosporidium species and genotypes in dairy calves Veterinary Parasitology

1 22, 1 03- 1 7 ' 2004

Shield J, Baumer JH, Dawson JA, Wilki n son PJ. Cryptosporidiosis - an educatio nal

experience. Journal of Infection, 2 1 , 297-30 1 , 1 990

Sulaiman IM, H i ra PR, Zhou L, AI-Ai i FM, AI-Shelahi F A, Shweiki HM, lqbal J, Kha lid N,

Xiao L. Un ique endemicity of cryptosporidiosis in ch i ldren in Kuwait. Journal of Clinical

Microbiology 43, 2805-09, 2005

Stefanogiannis N, Mclean M , Van M i l H. Outbreak of cryptosporidiosis l i nked with a farm

event. New Zealand Medical Journal 1 1 4, 5 1 9-2 1 , 2001

Turabelidze G, Lin M, Weiser T, Zhu BP. Communitywide outbreak of cryptosporid iosis i n

rural M issouri associated with attendance at ch i ld care centers . Archives of Pediatric and

Adolescent Medicine 1 6 1 , 878-83 , 2007

Valderrama AL, Hlavsa MC, C ronq uist A, Cosgrove S, Joh nston SP, Roberts JM, Stock

ML, Xiao L, Xavier K, Beach MJ. Multiple risk factors associated with a large statewide

i ncrease in cryptosporidiosis . Epidemiology and Infection 27, 1 -8 , 2009

Vonberg RP, Gastmeier P. Quality of outbreak descriptions in medical literature. The Lancet

Infectious Diseases 7, 699-700, 2007

Warren CA, G uerrant R L. Clinical disease and patho logy. In : Cryptosporid ium and

cryptosporidiosis (editors : R. Fayer and L . Xiao ) , 2nd edit ion, Pp. 235-8, CRC press, Boca

r�on, FL, USA, 2008

1 49

Wheeler C, Vugia DJ, Thomas G, Beach MJ, Carnes S, Maier T, Gormam J, X i ao L,

Arrowood MJ G il l i ss D, Werner SB. Outbreak of cryptosporidiosis at a Cal ifornia waterpark:

employee and patron roles and the long road towards prevention. Epidemiology and Infection

1 35, 302-1 0 , 2007

Wolfson JS, R ichter J M , Waldron M A, Weber DJ, McCarthy DM, Hopki n s CC.

Cryptosporidiosis in imm unocompetent patients. New England Journal of Medicine 3 1 2 , 1 278-

82, 1 985

1 50

CHAPTER 5

GENERAL DISCUSSION

The f irst publ ications by Tyzzer defined most of the knowledge about the biology of

Cryptosporidium parasites, whi lst the recent advent of mo lecu lar tools for the genotyping of the

isolates is currently enabl ing a clearer understand ing of the epidemiology of cryptosporidiosis.

The studies presented in th is thesis were performed dur ing a period of rapid expansion i n the

use of molecu lar tools for the study of Cryptosporidium, and applied a variety of such too ls .

Chapter 2 reports two studies that analysed m u!ti !ocus genotype data to explore the popu lation

genetic structure of C. parvum and C. hominis. For the analys is of the data, classic analytical

techniques of population molecular genetics and rarefact ion analysis - a stat istical test

traditional ly used in ecological and paleonto logical stud ies - were combined . Conversely,

Chapter 3 reports straightforward studies that applied simple genotyp ing schemes for

epidem iolog ical source-tracki ng . U lt imately, the epidemiological and molecular expert ise

acqui red by the author over the years was used to investigate an outbreak of cryptosprid iosis i n

humans in N e w Zealand.

The studies presented in Chapter 2 contribute to the discuss ion on the population structure of

C. parvum and C. hominis i n several ways. Fi rstly, the resu lts of the study i n Section 2 .2 provide

statistical support to the emerging idea of the existence of anth roponotic C. parvum cycles that

do not i nvolve local cattle reservoirs in Scotland. Unfortunately, the modest sample s ize

precluded a comparative analys is between human and bovi ne C. parvum in New Zealand us ing

rarefaction analysis. Also the use of different genetic m arkers by the research groups in

Scotland (Wellcome Centre for Molecular Parasitology) and the USA (Tufts University)

precluded a d i rect comparison between the sam ples from Scotland and New Zealand. However,

anthroponotic C. parvum cycles could not be identified in New Zealand. I ndeed, a str ik ing

homogeneity in the genetic repertoire of C. parvum was seen i n this country using multi locus

genotyping, with bovine and human isolates from North and South Island composing a s ing le

star-l ike eBURST network ( Section 2 .2) . Th is homogeneity was later corroborated by

comparative analysis of human , bovine and equine isolates us ing single and bi locus genotyp ing

schemes in the studies presented in Sections 3 .2- 3.5. I n addition, putative anthroponotic C.

parvum parasites carrying G P60 al leles belong ing to the a l lele fami ly 1/c could not be identified

in N ew Zealand.

Secondly, the study in Section 2.2 provides new data on the structuring role of geography in C.

parvum and C. hominis popu lations. On the whole, the resu lts show that gene flow is not

sufficient to erase genetic divergence among geographical ly separated parasite populations,

which basical ly remain al lopatric. This result was somewhat surprising because C. parvum and

C. hominis are cosmopol itan m icroorganisms smal l enoug h to be transported mechanically, or

1 51

through overt o r subcl in ical i nfections, thus enabling m ix ing of oocysts from different sources.

In the same study, the clear identification of five C. hominis isolates introduced to the UK f rom

Pakistan via travel a lso demonstrate the feasibil ity of tracking C. parvum and C. hominis

parasites to thei r country of or ig in using multi locus genotyp ing . The author hypothesises that the

ecology of cryptosporidiosis may have selected for the best-adapted parasites in each

environment, and that imported parasites would be un l ikely to spread if environmental factors

are unfavourable .

Lastly, the results of the study i n Sect ion 2 .2 are inconsistent with a strict categorization of C.

parvum and C. hominis as either clonal or panmictic. I nstead, they support the notion of the eo­

occurrence of both mutational and recombi natorial diversification, with geography playing an

important structuring role. In some regions where sanitary cond itions are good (as in the UK) ,

selfing seems to prevai l , g iv ing rise to what is often referred to as a "clonal" popu lation ,

characterized by the occurrence of a smal l number of m u ltilocus genotypes and a h igh ly

abundant putative founder type. Conversely, i n some developing regions (as i n Uganda) the

genetic d iversification of C. parvum and C. hominis seems to be accelerated by recombination

between genetical ly-heterogeneous gametes. This may be due to suboptimal sanitation and

h igh H IV prevale nce , which en hance i nfections with envi ronmental oocysts or ig inat ing from

m ultiple sou rces . The process generates a so-cal led 'panmictic' population, characterized by

the occurrence of a large number of mu lt i locus genotypes and the absence of an obvious

founder type. These differences in the structu res of C. parvum and C. hominis popu lations

between cou ntries may plausib ly explain - at least i n part - the severity of cryptosporidiosis

observed in some developi ng reg ions, as the accelerated diversification by recombination could

enhance the frequent appearance of novel, h i ghly viru lent genetic variants. However, the

cl in ico-epidem iogical impl icat ions of these differences remain to be establ ished, taking into

account confounding factors related to the host, such as the higher prevalence of H IV-related

immunosuppression in some developing countries.

lt is relevant to h ighlight here the l im itations of the above and previously publ ished studies of

popu lation genet ic structure of C. parvum and C. hominis. In al l these studies, the structures

have been inferred based on the observed genetic variation between isolates, as revealed

fol lowing PCR am plification of mu ltiple polymorphic markers. However, th is approach m ay

introduce bias, as the true genetic variation in Cryptosporidium is between the sporozoites,

with in the isolates . Unfortunately, this variat ion is diff icult to assess using PC R , due to the

problem of preferential ampl ificatio n of the templates, as discussed in Sect ion 1 .3 .

The studies presented i n Sections 3 . 1 -3 .3 defi ned the basic molecu lar and cl in ico­

epidem iolog ical aspects of cryptosporidiosis in foals and, to the author's knowledge, represent

the most comprehensive series of studies of cryptosporidiosis in these hosts reported in the

l i terature. The study in Sect ion 3 . 1 represents the f i rst known descript ion of the d isease in foals

1 52

which i ncludes cl in ico-epidemiological and pathological data, as well as the genetic

identificat ion of the iso lates as C. parvum. At the t ime the study was done, genetic subtyping

tools for Cryptosporidium were sti l l embryon ic and so the characterisat ion of the isolates was

l imi ted to the defin it ion of the taxon . In the fol lowing years, the suggestion that a

Cryptosporidium 'horse genotype' exists and the notion of the occurrence of anthroponotic C.

parvum cycles triggered the study presented in Section 3 .2 . The genetic characterisation of n i ne

Cryptosporidium-positive specimens col lected from foals for that study resu lted in the

identification of C. parvum in al l cases. Therefore , the notion of the occurrence of a

Cryptosporidium ' horse genotype' in horses could not be validated . After identifyi ng C. parvum

as the dominant species parasitis ing young foals in New Zealand, it was necessary to perform a

comparative genetic siudy of the isolates from foals, calves, and humans, in order to i nfer the

transm ission routes and assess the zoonotic potential of cryptosporidiosis in foals. Such a study

was then feasible, as numerous polymorphic loci in the genome of C. parvum had already been

reported . The results of the genetic comparison of diarrhoeagenic C. parvum isolates from foals,

humans and cattle indicated the three host-species were infected with genetically-s imi lar

parasites, and so cryptosporidiosis in foals should be considered potent ia l ly zoonotic. Final ly ,

the study presented in Sect ion 3.3 was m otivated by the lack of knowledge about the incidence

of cryptosporidiosis in foals. The detection of a new outbreak i n a broodm are farm situated near

Massey Un iversity in the course of the study faci l i tated a longitudinal observation of fou r

affected foals . The resu lts i ndicate that the cryptosporidiosis in foals is l i kely to be more

com mon than thought. The observed short patent period and self l im it ing nature of

cryptosporidiosis in foals suggest the condit ion may be underdiagnosed, and m ay account for a

proport ion of cases empirica l ly diagnosed as foal-heat diarrhoea. I n the two documented

outbreaks in foals, the ani mals co-existed with ruminants on the same land. However, newborn

calves are considered the main amplifiers of C. parvum i n nature, but such an imals were not

present on the farms dur ing the foal ing season , and no specimens from o lder ruminants were

analysed. In addition , an environmental source of infection, such stream water, was also l ikely.

Further studies to assess the i mpact of cryptosporidiosis in neonatal d iarrhoea of foals in New

Zealand are warranted.

The study presented in Section 3.4 was motivated by the paucity of data on the dairy-farm

prevalence of C. parvum in New Zealand, which contrasted with the huge i mportance of

dairying in th is country. Despite the short and concentrated calving seaso n , Cryptosporidium

oocysts were detected in 40 percent of the surveyed farms. However , it is acknowledged that

the phenotypic identification of the parasites in th is study did not stand the test of t ime. In fact, a

novel 1 8S rRNA gene sequence, most s imi lar to the sequence of C. bovis, was identif ied i n a

faecal specimen from a calf in the successive study reported i n Section 4, ind icat ing that non­

parvum Cryptosporidium parasites do cycle in young cattle in New Zealand. Nevertheless, a

large amount of data from overseas indicate C. parvum as the most com mon taxon cycl ing in

unweaned calves and so it is l ikely that most parasites isolated i n the study i n Section 3 .4 were

1 53

i ndeed C. parvum. Although not designed to est imate the farm level prevalence, the successive

f inding of C. parvum in a large number of farms in the Waikato in the study presented in Section

3.5 clearly indicates that this potentia l ly zoonotic species is widespread in cattle in New

Zealand.

Many questions related to the incidence of i nfections with C. parvum i n cattle and the immun ity

deve lopi ng as a result of cryptosporidiosis remain unanswered. Likewise, l i ttle is known about

the effect of the age on the susceptibi l ity to C. parvum i nfections in calves which are not

exposed to the i nfection in the perinatal period. In previous studies (which were not included this

thesis} , this and other authors have observed an i ncidence of cryptosporidiosis in 1 00% of the

calves on farms with year round calvin g (see Section 1 .6 . 1 ) . However, the same may not

necessari ly occur in New Zealand, where the short and concentrated cavi ng season m ight

prevent - at least in theory - the accummulatio n of the pathogen in the farm environment.

The absence of oocysts in the samples from farms rear ing Jersey cattle in the study presented

in Section 3 .4 was intr igu ing . To the author's surprise, th is seem ingly m inor f inding has been

later corroborated in a study in the USA (Starkey SR , Zeig ler PE, W ade SE , et a l . Factors

associated with shedd ing of Cryptosporidium parvum versus Cryptosporidium bovis among

dairy cattle i n New York State . JAVMA 229 , pp. 1 623-26, 2006}, which tr iggered a successive

letter to the editor of JAVMA by the author (JAVMA 23, p. 339, 2007} . The f inding of low

infection rates in Jersey cattle in independent studies i n d ifferent countries would suggest an

effect of the breed on the susceptibi l ity to C. parvum, which warrants further i nvestigation .

The study presented in Section 3 .5 represents the fi rst systematic genetic com parison between

human and bovine C. parvum i n New Zealand us ing h igh ly discriminatory sub-typing tools. The

results of the study indicate a significant homogeneity of the genetic repertoi re of bovine and

human C. parvum. Despite the wide spectrum of C. parvum G P60 al leles in circulat ion, two

alleles - belong ing to al lele fami l ies //a A1 8G3 R 1 and //a A1 9G4 R 1 - persistently dominated the

samples, each year. As hypothesised in the study presented in Section 3 .4 us ing microscopy­

only, the author postu lates that the cyclic dom inance of these subgenotypes is due to the

selection of a 'rapacious' phenotype, able to rapidly f i l l the niche during the calvi ng season and

survive between seasons. The pathways of survival of C.parvum i n the New Zealand ecosystem

between the calving seasons remain to be establ ished.

Final ly, Chapter 4 reported the i nvestigat ion of a serendipit ious outbreak of a gastrointestinal

i l l ness among a cohort of young adults. Cryptosporidium parvum was the only enteropathogen

identified from mu ltiple faceal specimens. The explosive nature of the outbreak and the h igh

attack rate observed underscore the potential publ ic health hazard posed by C. parvum i n New

Zealand. As in many other investigations of outbreaks of cryptosporidiosis publ ished in the

scientific literature , the sou rce of i nfection could not be traced, presumably due to the delay i n

1 54

the i nvest igation . I ndeed, the outbreak was init ial ly overlooked b y the local health authorit ies

based on the laboratory test resu lts, which precluded the occurrence of a cluster of cases.

However , a point source i nfection could be determined fol lowing the identification of a rare

subgenotype of C. parvum in two isolates. Also the c i rcumstantial evidence and the epidemic

curve strong ly suggested a poi nt-source zoonotic transm ission through d i rect contact with

calves dur ing a calf-hand l i ng practicu m . it is hoped that the lessons learned dur ing the

investigat ion of this outbreak will en hance preparedness.

In addit ion to the valuable data on the natural h istory of cryptospor id iosis obtained duri ng the

investigation , a h istory of previous i nteractions with ruminants in the farm environment was

signif icantly associated with a reduct ion of the risk of cryptosporidiosis in males. This result

reinforces the intu itive not ion that exposure to C. parvum confers some protective immun ity.

However, in this study the effect of previous i nteractions with rumi nants on the susceptibi l i ty to

cryptosporidiosis may have been confounded, and should therefore be corroborated i n large­

scale epidemiolog ical studies.

Many important questions on cryptosporidiosis remain u nanswered . Future studies of the

popu lation genetic structure of Cryptosporidium should address the question of the intra-isolate

genetic d iversity, which is technically d ifficult to assess us ing PCR. I n the author's opin ion , th is

diversity could be captured using increas ing ly-accessible PCR-free sequencing techno logies.

Large-scale epidemiolg ical studies are needed in o rder to assess the impact of

cryptosporidiosis, and other infections, on the health of newborn foals. A re-assessment of the

farm-prevalence of C. parvum i n New Zealand using molecular tools, as wel l as incidence

studies with long itudinal studies of large cohorts of calves in m ult iple farms would also be

necessary in order to define the farm-level prevalence and animal-level incidence of

cryptosporidiosis caused by C. parvum i n New Zealand, and elsewhere. Such studies, if

adequately powered, could also assess the presence of risk factors for C. parvum i nfect ions at

the farm-leve l . Finally, stud ies assessing the potential zoonotic impact of Cryptosporidium

parasites cycl ing in other dominant domestic and wild species in New Zealand, such as sheep,

deer, possums and birds, are also warranted.

1 55

APPENDICES

APPENDIX 2.1 Mult i locus genotypin g results o f the study reported i n Section 2 .2 .

M LG_I D: identif ier of the mu lti locus genotype; eBU RST code: code used in the s i ngle locus

variant networks using eBU RST software.

Numbers represent the allele s izes in base-pairs. The markers are those described in Section

2.2. N otice that the iso lates from the USA were not included in the eBU RST diagrams .

These data are a lso avai lable a s a n on- l ine supplement t o the publ ished paper at

http ://aem .asm .org/

1 00

1 00

1 36

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

266

274

274

308

1 56

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 68

1 68

1 68

209

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

205

1 75

1 75

1 75

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

250

280

306

1 66

250

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

235

244

244

244

244

244

244

244

244

244

244

244

244

244

244

244

244

244

382 298

382 298

382 298

394 228

382 283

382 283

382 283

382 283

382 3 1 6

382 3 1 6

382 3 1 6

382 3 1 6

382 3 1 6

382 3 1 6

382 283

382 274

382 283

382 283

382 283

382 283

382 283

382 283

382 283

382 283

382 283

382 283

382 283

382 283

382 283

382 283

382 298

382 298

1 56

1 71

1 84

1 84

1 7 1

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 74

1 77

1 74

1 74

1 74

1 74

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 79

1 92

1 92

1 79

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

U K 1

U K 2

U K 3

UK4

UG5

UG6

6

6

UG7

UG7

UG7

UG7

UG7

UG7

UG8

UG9

UG 1 0

U G 1 0

UG 1 0

U G 1 0

UG1 1

UG1 1

UG1 1

U G 1 1

U G 1 1

U G 1 1

U G 1 1

UG1 1

UG1 1

UG1 1

1 2

1 2

39

39

39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

39

39

39

39

39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

1 65

245

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 84

1 84

1 87

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

244

244

244

250

250

250

250

250

250

250

250

250

250

1 95

244

239

244

244

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

382

382

382

382

382

298

3 1 6

3 1 6

283

283

382 283

382 283

382 283

382 283

382 283

382 283

382 283

382 307

484 283

382 334

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

382 298

1 57

1 77

1 77

1 77

1 74

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 94

1 77

1 7 1

1 71

1 7 1

1 71

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 71

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 71

1 7 1

1 7 1

1 7 1

1 71

1 71

1 71

1 71

1 71

1 71

1 71

1 71

1 71

1 71

1 7 1

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 82

1 85

1 92

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 2

1 3

1 3

1 4

1 5

1 5

1 5

1 5

1 5

1 5

1 5

1 6

1 7 1 8

1 9

20

2 1

2 1

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

U G 1 2

UG 1 3

UG 1 3

UG 1 4

UG 1 5

UG 1 5

UG1 5

UG 1 5

UG 1 5

UG 1 5

UG 1 5

UG 1 6

UG 1 7 UG 1 7

UK20

UK21

UK21

UK22

UK22

UK22

UK22

UK22

UK22

UK22

U K22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

39

39

39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

1 39

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

266

274

274

274

274

274

274

274

274

274

274

274

276

276

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 52

1 52

1 52

1 52

1 52

1 52

1 52

1 68

1 68

1 68

1 68

1 68

1 68

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

1 75

1 75

205

205

205

205

205

1 75

205

205

205

1 35

205

250

250

250

250

250

250

250

250

250

250

250

250

250 250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

250

306

244

244

244

244

244

268

244

244

244

244

244

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

460

460

382

382

382

382

382

382

382

382

382

382

382

382

382

1 58

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

298

3 1 6

3 1 6

325

325

298

298

298

298

298

298

298

3 1 6

3 1 6

274

298

298

298

274

274

1 71

1 71

1 71

1 71

1 7 1

1 7 1

1 71

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 71

1 71

1 7 1

1 7 1

1 7 1

1 7 1

1 71

1 7 1

1 7 1

1 7 1

1 71

1 71

1 7 1

1 71

1 84

1 84

1 84

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 7 1

1 71

1 7 1

1 7 1

1 7 1

1 84

1 7 1

1 71

1 71

1 77

1 77

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 92

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 79

1 92

1 92

1 79

1 79

1 79

1 79

1 79

1 92

1 79

1 79

1 79

1 79

1 79

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

22

23

24

24

24

25

25

26

26

27

27

28

29

30

30

30

3 1

3 1

32

33

33

33

34

35

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK22

UK23

U K24

U K24

U K24

U K25

U K25

UK26

U K26

U K27

U K27

U K28

U K29

UK30

UK30

U K30

UK31

UK31

U K32

UK33

U K33

U K33

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

39

285

286

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

291

1 52

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 68

1 82

1 82

1 75

1 75

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 74

1 35

1 74

280

280

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

239

244

244

244

244

244

244

244

244

244

244

244

244

244

244

244

248

248

248

248

248

248

250

250

250

250

250

268

248

382

382

382

382

382

382

382

382

382

382

382

382

382 382

382

382

382

382

382

382

472

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

382

1 59

298

325

286

286

298

307

307

307

3 1 5

3 1 5

3 1 5

3 1 6

3 1 6 3 1 6

3 1 6

3 1 6

3 1 6

3 1 6

3 1 6

3 1 6

3 1 6

298

274

274

274

274

283

283

283

283

283

283

286

298

3 1 5

3 1 6

3 1 6

283

283

286

297

307

3 1 6

274

286

297

298

325

274

297

1 84

1 84

1 77

1 77

1 77

1 74

1 77

1 77

1 74

1 74

1 77

1 74

1 74

1 74

1 74

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 77

1 74

1 74

1 77

1 77

1 77

1 77

1 74

1 77

1 74

1 74

1 77

1 77

1 77

1 77

1 74

1 77

1 77

1 50

1 77

1 74

1 77

1 77

1 74

1 74

1 79

1 92

1 82

1 82

1 82

1 85

1 82

1 85

1 85

1 85

1 85

1 85

1 85

1 85

1 85

1 82

1 82

1 85

1 85

1 85

1 85

1 87

1 82

1 82

1 87

1 92

1 85

1 87

1 82

1 82

1 85

1 85

1 85

1 92

1 85

1 85

1 82

1 82

1 82

1 82

1 85

1 85

1 85

1 79

1 82

1 85

1 82

1 82

1 87

1 85

36

3 7

39

39

40

4 1

42

43

44

44

45

46

46

46

46

47

47

48

48

48

49

50

5 1

5 1

52

53

54

55

56

56

57

57

58

59

60

6 1

62

63

63

64

65

66

67

68

69

70

7 1

72

73

74

UK36

UK37

UG39

UG39

UG40

UG41

UG42

UG43

UG44

UG44

UG45

UG46

UG46

UG46

UG46

UG47

UG47

UG48

UG48

UG48

UG49

UG51

UG5 1

UG54

UG55

UG56

UG56

UG57

UG57

UG58

UG60

UG61

UG62

UG63

UK63

UK64

UK65

UK66

UK67

UG69

UG70

UG71

UG72

UG74

1 88 1 88 1 88

1 88 1 88

291

291

297

300

300

300

300

322

274

276

276

279 279 279 279

279 279

1 88 279 1 88 279 1 88 279

1 88 279

1 88 279 188 279 188 279 1 88 279

1 88 279 1 88 279 1 88 279 1 88 279 1 88 279 188 279 188 279 1 88

1 88

279

279

279 279 279 322

324

333

333 333

333 345 274

322

1 82

1 87

1 52

1 68

1 68

1 68

1 68

209

1 87

209

225

225

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

1 87

225

232

225

209

1 87

1 87

1 87

1 87

1 87

232

209

1 74

1 74

1 75

1 75

1 75

1 75

1 75

205

232

232

1 98

1 98

205

205

205

232

232

232

232

232

232

232

232

232

232

232

232

232

232

232

232

232

232

232

232

232

232

232

1 98

232

232

232

232

232

205

1 98

232

250

244

268

226

244

256

280

250

1 95

1 95

2 1 4

2 1 4

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

1 95

235

1 95

1 95

2 1 4

1 95

1 95

1 95

1 95

1 95

1 66

2 1 4

2 1 4

382 274

382 283

382 298

382 298

382 334

382 274

382 298

460 298

534 229 528 229

454 1 79

460 1 79

504 229 534 229 540 229 382 229

484 229 492 229 504 229 522 1 93 528 229

528 229

528 229 528 229 534 229 534 229

534 229 534 229 534 2 29 534 229 534 229 534 229 534 229 540 229

546 229

528 229 528 229 472 229 460 1 79

472 229

466 1 93

466 1 93 484 229

484 229 492 1 93 474 1 79

528 229

1 60

1 77

1 77

1 84

1 84

1 84

1 94

1 84

1 65

1 94

1 94

1 50

1 50

1 94

1 94

1 94

1 94

1 94

1 94

1 74

1 94

1 94

1 94

1 94

1 94

1 74

1 94

1 94

1 94

1 94

1 94

1 94

1 94

1 94

1 94

1 94

1 94

1 94

1 74

1 50

1 74

1 74

1 74

1 94

1 94

1 74

1 50

1 50

1 87

1 85

1 92

1 92

1 92

1 92

1 92

230

21 3

2 1 3

220

220

2 1 3 2 1 3

2 1 3 1 85

2 1 3 2 1 3 2 1 3 2 1 3 2 1 3

21 3

2 1 3

2 1 3 187 2 1 3

2 1 3

2 1 3

2 1 3 2 1 3 2 1 3 2 1 3 2 1 3 2 1 3

2 1 3

2 1 3 2 1 3

2 1 3

220

1 8 7

1 87

1 8 7 1 87

187 2 1 3 220

220

75

76

77

78

79

80

8 1

82

UG76

UK77

UK78

UK79

UK80

UK81

UK82

UG1 1

UG 1 2

UG 1 3

UG 1 3

UG 1 3

UG 1 3

UG 1 4

UG 1 5

U G 1 5

UG 1 5

UG 1 5

UG 1 5

UG 1 5

324

324

324

324

324

324

324

274

274

276

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

322

324

324

324

324

324

324

324

324

324

324

225

232

232

232

232

225

232

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

205

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

205

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

21 4

2 1 4

21 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

454

474

474

474

474

454

460

460

460

460

447

454

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

460

472

447

447

447

447

447

447

447

447

447

447

1 6 1

1 93

1 75

1 79

1 79

1 85

1 93

1 79

1 79

1 93

1 79

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 79

1 79

1 79

1 79

1 79

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 93

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

223

223

220

220

220

220

220

220

223

220

220

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

223

236

223

220

220

220

220

220

223

223

223

223

223

28

29

30

30

3 1

32

33

34

35

36

37

38

39

39

39

39

39

39

39

40

40

40

40

40

4 1

4 1

4 1

4 1

4 1

4 1

4 1

4 1

4 1

4 1

4 1

4 1

4 1

4 1

42

43

44

44

44

44

44

45

45

45

45

45

IL28

TR29

TR30

TR30

TR3 1

I L32

TR34

NZ35

TR36

IL37

NZ40

NZ40

NZ40

NZ40

NZ40

NZ41

NZ41

NZ41

NZ41

NZ41

NZ41

NZ4 1

NZ4 1

NZ4 1

NZ4 1

NZ41

NZ41

NZ41

NZ4 1

NZ42

NZ43

IL44

IL44

IL44

IL44

I L44

I L45

I L45

I L45

I L45

I L45

324

324

324

324

324

324

324

324

324

340

340

279 322

322

322

322

322

322

322

324

324

324

322

322

322

324

324

324

340

322

322

322

324

324

225

225

225

225

225

225

225

225

225

225

225

1 87

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

225

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

205

205

232

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

1 98

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

1 95

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

2 1 4

447 1 93

454 1 93

454 1 93

454 1 93

454 1 93

460 1 93

460 1 93

474 1 79

474 1 79

385 333

460 1 93

5 3 4 229 447 1 93

447 1 93

447 1 93

447 229

454 1 93

454 1 93

454 206

447 1 93

472 229

472 229

4 1 5 229

447 220

454 1 93

4 1 5 229

447 229

472 206

447 229

447 229

454 229

454 229

447 1 93

447 229

1 62

1 70

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 94

1 50

1 70

1 70

1 70

1 70

1 70

1 50

1 70

1 50

1 70

1 50

1 50

1 70

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

1 50

2 1 3

220

220

220

223

220

223

223

223

223

223

2 1 3

236

236

236

236

236

236

236

236

236

236

236

236

236

236

236

236

236

236

21 3

236

236

236

A P P E N D I X 3 . 5 The G P60 sequences identified in bovine and human C. parvum i n the study i n Sect ion 3 .5

>lla A1 8G3R 1 , bovine a n d hu man C. parvum

ATTTAAAGGATGTTCCTGTTGAGGGCTCATCATCGTCATCGTCATCGTCATCATCATCATCATCATCAT

CATCATCATCATCATCATCAACATCAACCGTCGCACCAGCAAATAAGGCAAGAACTGGAGAAGACGCA

GAAGGCAGTCAAGATTCTAGTGGTACTGAAGCTTCTGGTAGCCAGGGTTCTGAAGAGGAAGGTAGTG

AAGACGATGGCCAAACTAGTGCTGCTTCCCAACCCACTACTCCAGCTCAAAGTGAAGGCGCAACTAC

CGAAACCATAGAAG CTACTCCAAAAGAAGAATGCGGCACTTCATTTGTAATGTGGTTCGGAGAAGGTA

C CCCAGCTGCGACATTGAAGTGTGGTGCCTACACTATCGTCTATGCACCTATAAAAGACCAAACAGAT

CCCGCACCAAGATATATCTCTGGTGAAGTTACATCTGTAACCTTTGAAAAGAGTGATAATACAGTTAAA

ATCAAGGTTAACG GTCAGGATTTCAGCACTCTCTCTGCTAATTCAAGTAGTCCAACTGAAAATGGCGG

ATCTGCGGGTCAGGCTTCATCAAGATCAAGAAGATCACTCTCAGAGGAAACCAGTGAAGCTGCTGCA

ACCGTCGATTTGTTTGCCTTTACCCTTGATGGTGGTAAAAGAATTGAAGTGGCTGTACCAAACGTCGA

AGATGCATCTAAAAGAGACAAGTACAGTTTGGTTGCAGACGATAAACCTTTCTATACCGGCGCAAACA

GCGGCACTACCAATGGTGTCTACAGGTTGAATGAGAACGGAGACTTGGTTGATAAG

> /la A 18G3R 1 , h u man C. parvum

TGTTCCTGTTGAGGGCTCATCATCGTCATCGTCATCGTCATCATCATCATCAICATCATCATCATCATCA

TCATCATCAACATCAACCGTCgCaCCAg eAAATaAGG CAAGAaCTGgAGAAGACGeAGAAGGCAGTCAA

GA TTCTaGTGGTaCTGAaGCTTCTGGT AGCCAGGGTT e TGAAGAGGAAGGTag TGAAGACGATGGCCA

AACTagTGCTGCTTCCCaACCCACTACTCCaGCTCAAAGtGaAGGCGCAACTACCGAAACCATAGAAgC

T ACTCCAAAAGAAGAA TGCGGeaCTT ea TTTGTaATGTGGTT eGgAGAAgGtACCCCaGCTGCGACA TTG

AaGtGTGGtGCCTACACTATCGTCTATGCACCTATAAAAGACCAAAeAGATCCCGCAC CAAgATATATCT

CTGGTGAAGTTACATCTGTAACCTTTGAAAAGAGTGATAATACAGTTAAAATCaAG GTTAACGGTeAGG

ATTTCAgCaCTCTCTCTGCTAATTCAAGTAgTCCAACTGAAaATGGeGGATCTGCGGGTeAgGCTTCATC

AaGATCAAGAAGATeACTCTCAGAgGAAACCagTGAAGCTGCTGCAACCGeCgATTTGTTTGCCTTTACe

CTTGATGGTGGTAAAAGAATTGAAGTGGCTGTACCAAACGTCGAAGATGCATCTAAAAGAGACAAGTA

CAGtTTGGTIGCAGACGATAAACCTTTCTATACCGGCGCAAACAGCGGeACTACeaATGGTGTCTACAG

GTTGAATGAgAACGGAGAC

>lla A1 9G4R 1 , bovine C. parvum

CTTTAAaGGATGTTCCTGTTGAGGGCTCATCATCGTCATCGTCATCGTCATCGTCATCATCATCATCAT

CATCATCATCATCATCATCATCATCAACATCaaCCGTCgCACCAGCaAATaAGGeAAGAaCTGGAGAAGA

CGCAGAAGGCAGTCAAGATTCTaGTGGTACTGAAGCTTCTGGTAGCCAG GGTTCTGAAGAGGAAGGT

aGTGAAGACGATGGCCaAACTagTGCTGCTTCCCaACCCACTACTCCAGCTCAAAGTGaAGGCGCAAC

T ACCGAAACCA T AGAAGCT ACTCCAAAAGAAGAA TGCGGeaCTT ea TTTGTaAtGTGGtTeGgAGAAGGT A

CCCCAGCTGCGACATTGAaGtGTGGTGCCTACACTATCGTCTATGCACCTATAAAAGACCAAAeAGATC

CCGCACCAAgATATATCTCTGGTGAAGTTACATCTGTAACCTTTGAAAAGAGTGATAATACAGTTAAAA

TCaAGGTTAACG G TCAGGATTTCAgCACTCTCTCTGCTAATTCAAGTAGTCCAACTGAAaATGGeGGAT

CTGCGGGTeAGGCTICATCAAGATCAAGAAGATeACTCTCAGAGGAAACCAGTGAAGCTG CTGCAACC

GTC GATTTGTTTGCCTTTACCCTTGATGGTGGTAAAAGAATTGAAGTGGCTGTACCAAACGTCGAAGA

TGCATCTAAAAGAGACAAGTACAGTtTGGTTGeAGACGATAAAeCTTTCTATACCGGCGCAAACAGCGG

eACTACeaATGGTGTCTACAGGTTGAATGAgAACGGAGACTTGG

1 63

APPENDIX 4.1 Questionnaire 01 ( See Section 4 .3) .

1 - D id you experience one of the following i n the last four weeks? (tick the box)

Abdominal discomfort---

Diarrhoea------------------

Vomiting-----------------

None of the above---------

2 - If you answered yes to Q 1 , please circle the date when you first experienced the symptoms and the date w hen you first felt healthy (the dates of Mechanisms of Disease

Case Scenarios are given for your orientation) :

Date Mechanisms of disease Case Scenario Monday 1 8/09/06--------------------- --- A cat h aving difficulty breathing

Tuesday 1 9/09/06

Wednesday 20/09/06 Thursday 2 1109/06

Friday 2 2/09/06 Saturday 23/09/06

S unday 24/06/06 Monday 25/09/06 ------------------------------- A c alf w ith severe diarrhoea

Tuesday 26/09/06 ------------------------------- Visit to LA TU (calf h andling) Wednesday 27/09/06 Thursday 28/09/06 Friday 2 9/09/06 Saturday 30/09/06 Sunday 0 1 1 1 0/06 Monday 02/ 1 0/06 ------------------------------- A thirsty, balding portly poodle

Tuesday 03/ 1 0/06 Wednesday 04/ 1 0/06

Thursday 05/ 1 0/06

Friday 06/1 0/06 Saturday 07/ 1 0/06 Sunday 08/ 1 0/06 Monday 09/ 1 0/06----------------------------------Il l thrift in a group of lambs

Tuesday 1 0/ 1 0/06 Wednesday 1 11 1 0/06 Thursday 1 2/ 1 0/06 Friday 1 3/1 0/06 Saturday 1 4/ 1 0/06

Sunday 1 51 1 0/06 Monday 1 61 1 0/06----------------------- --------------Dog with hindlimb lamenes s

Tuesday 1 7/ 1 0/06 Wednesday 1 8/ 1 0/06/Today

1 64

3 - Did you seek medical advice? Circle either Yes or No

4 - Did you submit faecal samples for analysis? Yes No

3 - Gender: Male D Female D

4 - Age:

5 -Background: NZ Australia American Asian Other

6 - In general , how do you quench your thirst during practical sessions?

Bottled water Tap water

7 Are you a smoker? Yes No

Soft drink

8 - Is it your habit to wash your hands after practical c lasses where animal s

are handled? Yes No

9 - Are you on any medication for acute or chronic disorders (e.g. insulin,

thyroxine) . Yes No. If yes, could you specify?

1 0 Did you grow u p in: Rural Urban

Other

1 1 -Before starting the Veterinary course, how often did physically handled ruminants :

Never Once/twice a year Monthly Weekly /more

Thank you very much for your help,

1 65

APPENDIX 4.2 Questionnaire 02 ( See Section 4 .3)

Please tick the appropriate box ! Did you declare suffering from abdominal d iscomfort

Did you attend last year's practical "A Did you drink tap water at and/or diarrhoea and/or vomiting in the

calf with severe diarrhoea" at LATU? LATU on that d ay? questionnaire?

YES NO YES NO YES NO I 2

3

4

5

6

7

8

9

1 0

1 1

1 2

1 3

1 4

1 5

1 6

1 7

1 8

1 9

20

2 1

22

23

24

25

26

27

28

29

30

3 1

32

33

34

35

36

37

38

39

40

4 1

42

43

44

45

46

47

48

49

50

5 1

1 66

APPENDIX 4.3 Class responses to questionnai re 01 (See also Tables 4 . 1 and 4.2)

Key to appendix:

Abdom: abdominal d iscomfort: yes= 1 ; no= 0

Diarrhoea : yes= 1 ; no= 0

Vomiting: vomiting . Yes= 1 ; no= 0

Date first: the date of the onset of the symptoms

Date healthy: last day of symptoms

Gender: 1 = m ale; 0= female

Nationality: New Zealand= 1 ; Austral ian= 2 ; U SA= 3; Asia=4 ; other= 5

Tap: dr ink ing tap water duri ng practicals. Yes= 1 ; no= 0

Wash hands: washing hands after practical classes where an imals are hand led. Yes= 1 ; no=O

Chronic med ic: taking chronic medicat ion . Yes= 2 ; no= 0

Rural/urba n : rural= 0 ; urban=1 0,

Handled ruminants: no previous physical i nteraction with rum inants= 0; previous physical

interact ion with ruminants= 1

Frequency handled ruminants: up to once-twice/year= 0 ; m ore than once-twice/year= 1

Diseased Abdom diarrhoea vom Date first Date healthy gender nationality Tap wash hands

Frequenl chronic Handled handle;

medic Rural/urban calves ruminan

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0 0

0 0

0 0

0

0 0

0

0 0

0

0

0 0

0

0

0

0 0

0

0

0

0

0

28/09/2006 6 1 1 0/2006

1 1 1 0/2006 5/1 012006

1 11 0/2006 1 71 1 0/2006

211 0/2006

26/09/2006 2810912006

1 / 1 0/2006 4/1 012006

7/1 0/2006 1 3/1 0/2006

3/1 012006 6/1 0/2006

211 0/2006 1 0/1 0/2006

1 67

0

0

3

3

5

5

3

3

4

0

0

0

Q

0

0 0

0

0

o I 0 0

0

0

o I 0 0 0

0 0 0

0

0

0 0 0

Q 0 0

Q 0

0 0

0

0 0

0

0

0

1 68

APPENDIX 4.4 G P60 and HSP70 sequences of C. parvum isolates from affected students and the 1 88 rRNA

sequence found in a calf dur ing the outbreak investigation reported in Section 4 . 1 .

Key: HSP70_1_student: sequence of the C. parvu m HSP70 gene fou nd in 3 students ;

HSP70_2_students : sequence of the C . parvum HSP70 gene found i n one student ; 1 8S_rRN A_

calf : f the 1 88 rRNA gene sequence found in one calf.

> HSP70_1 _student

ATCTGAAATTGATGAGGCTGAGAAGAAGATCAAGGATGCTCTTGACTGGCTCGAGCACAACCAAACTGCTGAAAAGGACGAGTTTGA

ACATCAACAAAAGGAGATTGAAACTCATATGAATCCACTCATGATGAAGATCTACTCTGCTGAGGGTGGTATGCCAGGTGGAATGCC

AGGTGGTATGCCAGGCGGTATGCCAGGTGGAATGCCAGGTGGTATGCCAGGTGGAATGCCAGGCGGTATGCCAGGTGGTATGCCA

GGTGGT ATGCCAGGTGGTATGCCAGGA TCT AATGGTCCAACTGTTGAAGAGGTCG A C T AA TT .A.TTTT!\GTC.t•.CCA/\AAAAACTCACTC

AAAATGGAAAGTTAAGAACTATTTACACACTTTCAATTTCTAGTTATTTTTTACCAAAATAAGAAGAAAAGCACACTCTCCCTTTAGG

>HSP70_2_student

A TGAgGCTGAGAaGAAGA TCAAGGA TGCTCTTGACTGGCTCGAGCACAACcAAACtGctGAAAAGGACGAGTTTGAACA TCAACAAAAG

GAgATTGAAACTCATATGAATCCACTCATGATGAAGATCTACTCTGCTGAGGGTGGTATGCCAGGTGGAATGCCAGGTGGTATGCCA

GGCGGT ATGCCAGGTGGAATGCCAGGtGGT ATGCCAGGTGGAA TGCCAgGCGGT ATGCCAGGtGGT ATGCCAGGtGGT ATGCCAGGt

GGT ATGCCAgGATcT AatGgtCCAaCTGTTGaAgagGtcGACTAA TtA TTTT AgTCACCtAaaAAAACTcaCTCAAAAtggAAAGtT AAgAACT AttT

ACACACtTtCAaTTTCTagTTaTTTTTTACCAAA

>GP6_student_ 1

GGATGTTCCTGTTGAGGGCTCATCATCGTCATCGTCATCGTCATCGTCATCATCATCATCATCATCATCATCATCATCATCATCATCAT

CATCAACATCAACCGTCGCACCAGCAAATAAGGCAAGAACTGGAGAAGACGCAGAAGGCAGTCAAGATTCTAGTGGTACTGAAGCTT

CTGGTAGCCAGGGTTCTGAAGAGGAAGGTAGTGAAGACGATGGCCAAACTAGTGCTGCTTCCCAACCCACTACTCCAGCTCAAAGT

GAAGGCGCAACTACCGAAACCATAGAAGCTACTCCAAAAGAAGAATGCGGCACTTCATTTGTAATGTGGTTCGGAGAAGGTACCCCA

GCTGCGACATTGAAGTGTGGTGCCTACACTATCGTCTATGCACCTATAAAAGACCAAACAGATCCCGCACCAAGATATATCTCTGGT

GAAGTTACATCTGTAACCTTTGAAAAGAGTGATAATACAGTTAAAATCAAGGTTAACGGTCAGGATTTCAGCACTCTCTCTGCTAATTC

AAGTAGTCCAACTGAAAATGGCGGATCTGCGGGTCAGGCTTCATCAAGATCAAGAAGATCACTCTCAGAGGAAACCAGTGAAGCTGC

TGCAACCGTCGATTTGTTTGCCTTTACCCTTGATGGTGGTAAAAGAATTGAAGTGGCTGTACCAAACGTCGAAGATGCATCTAAAAGA

GACAAGTACAGTTTGGTTGCAGACGATAAACCTTTCTATACCGGCGCAAACAGCGGCACTACCAATGGTGTCTACAGGTTGAATGAG

AACGGAGACTTGG

>GP6_student_3

GAaCTTT AAGGatGTTCCTGTTGAGGGCTCATCA TCGTCATCGTCATCGTCA TCGTcATCAtCA TCATCATCATCATCATCATCA TCATCA

TCA TCA T C a TCAaCA TCaaCCGtCGCACCAgcAAA TaAgGcAAGAaCTGgAGAAGACGcAGaAGGcAGTCAaGA TTCTaGTGGTaCTGAaG

CTtCTGGTAgCcAGGGTtcTG AAGAgGaAGGTagTGAAGACGATGGCCaaACTagTGCTGCTTCCCaACCcACTACTCCaGCTCAAaGTGa

AGGCGCaACT ACCGaAACCA T AGAAGCT ACTCCAAAAGAAGaA TGCGgcaCTT ea TTTGTaAtGTGGtTcGgAGAAGGtACCCCaGCTGCG

ACA TTGAaGtGTGGtGCCT ACACT ATCGTCT A TGCACCTAT AAAAGACCAAACAGATCCCGCACCAAGATATATctCTGGTGAAGtT ACA T

CTGTAACCTTTGAAAAgAGTGATAATACAGTTAAAATCaAGGTTAACGGTCAgGATTTCAgcACTCTCTCTGCTAATTCAAGTAgTCCAA

CTGAaaA TGGcGGA TCTGCGGGTcAGGCTTCA TCAaGA TCAAGAAGA TCACTCTCAGAGGAAACCaGTGaAGCTGCTGCaACCGtCgA T

TTGtTTGCCTTT ACcCTTGA TGGTGGtAAAAGAa TTGAAGTGGCTGT ACCAaACGtCGAAGA TGCA TCT AAAAGAGAcAAGT ACAGttTGG

TTGcAGAcGATAaAcCTTTCT AT ACCGGcGCAAACAGCGGcACTACcaATGGTGTCTACAGGTtGAA TGAgAACGGAgACTT

> l BS_rRNA_ calf

TTTGGTGACTCATAATAACTTTACGGATCACATTATGTGACATATCATTCAAGTTTCTGACCTATCAGCTTTAGACGGTAGGGTATTGG

CCTACCGTGGCTATGACG GGTAACGGGGAATTAGGGTTCGATTCCGGAGAGGGAGCCTGAGAAACGGCTACCACATCTAAGGAAG

GCAGCAGGCGCGCAAATTACCCAATCCTAATACAGGGAGGTAGTGACAAGAAATAACAATACAGAGCCTTACGGTTTTGTAATTGGA

ATGAGTTAAGTATAAACCCCTTAACAAGTATCAATTGGAGGGCAAGTCTGGTGCCAGCAGCCACGGTAATTCCAGCTCCAATAGCGT

ATATTAAAGTTGTTGCAGTTAAAAAGCTCGTAGTTAATTTTCTGTTAATTTTTATATACAATGCTACGGTATTTATATAATATTAACATAA

TTCATATTACTTTTTAGTATATGAAACTTTACTTTGAGAAAATTAGAGTGCTTAAAGCAGGCTATTGCCTTGAATACTCCAGCATGGAAT

AATATTAAGGATTTTTATTCTTCTTATTGGTTCTAGAAT AAAAATAATGATTAATAGGGACAGTTGGGGGCATTTGTATTTAACAGTCAG

AGGTGAAATTCTTAGATTTGTTAAAGACAAACTACTGCGAAAGCATTTGCCAAGGATGTTTTCATTAATCAAGAACGAAAGTTAGGGG

ATCGAAAGACGATCAGATACCGTCGTAGTCTTAACCATAAACTATGCCAACTAGA

1 69