Molecular epidemiology of Cryptosporidium in humans and cattle in The Netherlands

Available online at www.sciencedirect.com

International Journal for Parasitology 38 (2008) 809–817 www.elsevier.com/locate/ijpara

Molecular epidemiology of Cryptosporidium in humans and cattle in The Netherlands q Peter R. Wielinga a,*, Ankje de Vries a, Tjeerd H. van der Goot a, Theo Mank b, Maria Henriette Mars c, Laetitia M. Kortbeek d, Joke W.B. van der Giessen a a

National Institute for Public Health and the Environment (RIVM), Centre for Infectious Disease Control (Cib), Laboratory for Zoonoses and Environmental Microbiology (LZO), Antonie van Leeuwenhoeklaan 9, P.O. Box 1, Bilthoven, The Netherlands b Laboratory of Public Health, Department of Parasitology, Boerhaavelaan 26, 2035 RC Haarlem, The Netherlands c Animal Health Service (GD), Deventer, The Netherlands d RIVM, Laboratory for Infectious Diseases Surveillance and Screening, The Netherlands Received 18 July 2007; received in revised form 12 October 2007; accepted 16 October 2007

Abstract The protozoan parasite Cryptosporidium is found world-wide and can cause disease in both humans and animals. To study the zoonotic potential of Cryptosporidium in The Netherlands we isolated this parasite from the faeces of infected humans and cattle and genotyped those isolates for several different markers. The overall genotyping results showed: for humans isolates, 70% Cryptosporidium hominis, 19% Cryptosporidium parvum, 10% a combination of C. hominis and C. parvum, and 1% Cryptosporidium felis; and for cattle isolates 100% C. parvum. Analysis of the genetic variants detected for the HSP70, ML1 and GP60 markers showed: for human isolates, one C. hominis and two C. parvum variants (C. parvum and C. parvum NL) for HSP70, one C. hominis and five C. parvum variants (C1, C2, C3, and C2 NL1 and C2 NL2) for ML1, four C. hominis (mainly IbA10G2) and four C. parvum variants (mainly IIaA15G2R1) for GP60; and the cattle isolates only C. parvum (not C. parvum NL1) for HSP70, C1 and C2 for ML1, and 17 different IIa sub-types (mainly IIaA15G2R1) for GP60. Molecular epidemiological analysis of the human data showed a C. hominis peak in autumn. The majority (80%) of the human cases were children aged between 0 and 9 years and >70% of these were caused by C. hominis. Patients >25 years of age were infected mainly with C. parvum. We conclude that C. hominis IbA10G2 is found at high frequencies in autumn in humans and not in cattle. The high prevalence of C. parvum IIaA15G2R1 in both humans and cattle indicates that cattle may be a reservoir for this sub-type in The Netherlands. Ó 2007 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Cryptosporidium; Molecular epidemiology; Genotyping; Zoonose; Human; Ruminants

1. Introduction Cryptosporidium is a globally distributed protozoan parasite which has been found in both vertebrates (e.g., humans, cattle, and birds) and invertebrates (e.g., clams) (Xiao and Ryan, 2004; Caccio et al., 2005). In most cases, q

Note. Nucleotide sequence data reported in this paper are available in GenBankä under the Accession Nos. DQ388376–DQ388391 and EF576952–EF576987. * Corresponding author. Tel.: +31 30 2743666; fax: +31 30 2744434. E-mail address: [email protected] (P.R. Wielinga).

including those in humans, Cryptosporidium causes an enteric infection leading to gastrointestinal (GI) problems such as severe diarrhoea. In birds, Cryptosporidium may also infect the respiratory tract causing respiratory problems (Morgan et al., 2001). Due to the relatively uniform appearance and parasitic life-cycle of different Cryptosporidium strains, classical morphological, and phenotypical classification are limited in distinguishing the various species and genotypes found in humans and animals. Molecular genetic techniques are able to distinguish the different species and genotypes and have shown that some are very host-specific while others have a broad host range. For

0020-7519/$34.00 Ó 2007 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijpara.2007.10.014

810

P.R. Wielinga et al. / International Journal for Parasitology 38 (2008) 809–817

instance, Cryptosporidium baileyi and Cryptosporidium meleagridis are mainly associated with birds; Cryptosporidium canis with dogs; Cryptosporidium felis with cats and Cryptosporidium molnari with fish; however Cryptosporidium parvum has a much broader host range and is found in cattle, sheep, goats, deer, raccoon dogs, horses and humans (for more details see reviews Xiao and Ryan, 2004; Caccio et al., 2005). In humans, the species mainly found to cause disease are Cryptosporidium hominis (also found in monkeys) and C. parvum. Other species (e.g., C. meleagridis, Cryptosporidium andersoni, C. canis, and C. felis) are only found in humans sporadically (Caccio et al., 2002). The incidence of gastroenteritis and the role of a broad range of pathogens in The Netherlands has been studied before (Mank, T.G., 1997. Intestinal protozoa and diarhoea in general practice. Academic thesis, Vrije Universiteit Amsterdam, The Netherlands. ISBN9 056690856; de Wit et al., 2001a,b,c,d). In these studies Cryptosporidium was found in about 2–3% of gastroenteritis cases, compared with a prevalence of about 0.2% in the general population. In these studies the Cryptosporidium isolates were not further genetically analysed. Several studies have reported molecular detection of Cryptosporidium in The Netherlands in humans (Verweij et al., 2004), water (Medema and Schijven, 2001) and animals (Homan et al., 1999b; Huetink et al., 2001; Schets et al., 2007). C. hominis (or genotype 1) was found in humans and C. parvum (or genotype 2) in animals and sporadically in humans (Homan et al., 1999b). Microsatellite 1 analysis showed the presence of C. parvum C1 and C3 in dairy cows (Huetink et al., 2001). Detailed molecular epidemiological information on zoonotic potential and transmission routes in The Netherlands is, however, limited. We sought to study the zoonotic potential of Cryptosporidium in The Netherlands by studying the prevalence and genetic diversity of Cryptosporidium in humans and farm animals. Faecal samples from humans and farm animals with cryptosporidiosis were genotyped for several loci and linked to epidemiological information. Several different methods and genetic loci have been used to genotype Cryptosporidium (Xiao and Ryan, 2004; Caccio et al., 2005; Chalmers et al., 2005; Thompson et al., 2005). In the present study we chose to use PCR and DNA sequencing of a broad spectrum of markers to try to include both conserved and polymorphic markers. The following markers were tested: the 18S ribosomal RNA gene (18S) (Morgan et al., 2001), the Cryptosporidium outer wall protein gene (COWP) (Pedraza-Diaz et al., 2000), the heat shock protein 70 gene (HSP70) (Morgan et al., 2001), the 60 kDa glycoprotein gene (GP60) (Strong et al., 2000; O’Connor et al., 2002), and the two microsatellite markers ML1 and ML2 (Caccio et al., 2000, 2001). The sub-types found for these markers were compared with each other and correlated with the available epidemiological information such as age, recent travel history, gender, region, and seasonality.

2. Materials and methods 2.1. Origin of the samples and DNA isolation Between 2003 and 2005 patients consulting their general practitioner with GI complaints (mostly diarrhoea and abdominal pain) indicating a parasitic infection were tested in collaboration with five clinical diagnostic laboratories for the presence of Cryptosporidium in their stool. Ninety-one of these Cryptosporidium-positive (modified Ziehl–Neelsen stain; Henriksen and Pohlenz, 1981) faecal samples were anonymized and send to the National Institute for Public Health and the Environment (RIVM) for genotyping of the Cryptosporidium infection. The samples originated from five different regions: Goes (n = 11), Haarlem (n = 55), Veldhoven (n = 22), Utrecht (n = 9), Enschede (n = 1). Together with information on the region, each patient was asked for information on gender, occurrence of the disease, age and recent travel abroad. Following the Dutch law on privacy protection this information was decoupled from each patient’s name and address to protect their privacy. In the period 2005–2006, in collaboration with the Dutch Animal Health Service, 162 Cryptosporidium positive faecal animal samples (160 cattle, one goat, and one sheep) were collected by microscopically screening faecal samples of clinical animal cases from different farms throughout The Netherlands. Information on the region, the age and species of the animals was collected and for reasons of privacy protection this information was decoupled from the specific address and name of the farm. The positive samples were sent anonymously to the RIVM for genotyping of the Cryptosporidium infection. For DNA isolation, two methods were used with equal success. In the first method a crude Cryptosporidium oocyst suspension was prepared from 1 g of non-preserved faeces as previously described (Homan et al., 1999a). Oocyst were then isolated using magnetic anti-Cryptosporidium beads (Dynabeads GC-combo from Dynal Biotech GmbH, Hamburg, Germany) and after heating the sample in lysis buffer for 10 min at 100 °C the DNA was isolated by a total DNA purification system (Gentra systems, Minneapolis, Minnesota, USA). In the second method, 1 g of non-preserved faeces was mixed with 2.5 mL GITC-lysis buffer (bioMe´rieux, Boxtel, The Netherlands) and heated for 10 min at 100 °C, after which the insoluble material was removed by centrifugation. DNA was isolated from the liquid phase using a Nucleosens kit (bioMe´rieux, Boxtel, The Netherlands) according to the manufacturer’s instructions. 2.2. Genotyping Cryptosporidium DNA genotyping was performed by PCR and DNA sequence analysis for 18S, COWP, HSP70, GP60, ML1, and ML2 as described (Caccio et al., 2000, 2001; Strong et al., 2000; Pedraza-Diaz et al., 2000; Morgan et al., 2001), using the primers listed in Table 1. PCR reactions were run on a Px2 thermal cycler (Thermo Electron

P.R. Wielinga et al. / International Journal for Parasitology 38 (2008) 809–817

811

Table 1 DNA sequence and location on the Cryptosporidium genome of the primers used in this study Gene locus

Primer

Primer DNA sequence (5 0 –3 0 )

18S

18S-F

AGTCATATGCTTGTCTCAAAGATT

HSP70

18S-R HSPR4-F

TTAACAAATCTAAGAATTTCACC GGTGGTGGTACTTTTGATGTATC

COWP

HSPR4-R BCOWPF

GCCTGAACCTTTGGAATACG ACCGCTTCTCAACAACCATCT TGTCCTC CGCACCTGTTCCCACTCAAT GTAAACCC CTAAAAATGGTGGAGAATATTC

COWPR ML1

F5

ML2

R5 M15-F

GP60

M16-R ATGFOR

CAACAAAATCTATATCCTC CAATGTAAGTTTACTTATG ATTAT CGACTATAAAGATGAGAGAAG ATGAGATTGTCGCTCATTATC

AL3533REV

AGATATATCTTGGTGCG

Positiona

Repeat sequence b

Reference

2449–3309 Chr 7 (GI 46229367)

Morgan et al. (2001)

3772–4217 Chr 2 (GI 46226355)

Morgan et al. (2001)

409492–410260 Chr 6 (GI 46228893)

Pedraza-Diaz et al. (2000)

824371–824607 Chr 3 (GI 46228038)

GAG

Caccio et al. (2000)

61488–61720 Chr 6 (GI 32398642)

AG

Caccio et al. (2001)

178999–178345 Chr 6 (GI 46228893)

TC-A/G/T & ACATCA

Strong et al. (2000)

a

Position of the PCR product on the chromosome (Chr) of Cryptosporidium parvum (Abrahamsen et al., 2004) for the indicated sequence referred to by its GenBank Index number. b Note that the 18S locus is found on chromosomes 2, 7, and 8 of C. parvum (Abrahamsen et al., 2004) and on chromosomes 2 and 7 of Cryptosporidium hominis (Xu et al., 2004). Only the chromosome 7 locus is used for comparison here.

Corporation, Breda, The Netherlands). In short, PCR reactions were conducted in 50 ll reactions mixtures containing: 10 mM Tris–HCl, pH 8.3, 50 mM KCl, 0.2 mM deoxyribonucleotide triphosphates, 1 U of AmpliTaq polymerase (Applied Biosystems, Nieuwerkerk a/d IJssel, The Netherlands), 0.25–0.5 lM of each primer, 2–5 ll template DNA and the following MgCl2 concentrations: for 18S 3.5 mM; for HSP70 1.75 mM; for COWP, ML1, ML2 and GP60 1.5 mM. The PCR was carried out under the following conditions: 5 min of denaturation at 94 °C followed by different amplifications schemes and ending with 10 min elongation at 72 °C. The following amplification schemes were used: for 18S, 52 cycles of 30 s 95 °C, 30 s 55 °C, 45 s 72 °C; for COWP, 30 cycles of 30 s 95 °C, 30 s 59 °C, 30 s 72 °C; for ML1, 30 cycles of 30 s 95 °C, 30 s 50 °C, 60 s 72 °C; for ML2, 40 cycles of 30 s 95 °C, 30 s 50 °C, 60 s 72 °C; for HSP70, 40 cycles of 30 s 95 °C, 30 s 54 °C, 30 s 72 °C; for GP60, 45 cycles of 30 s 95 °C, 30 s 46 °C, 30 s 72 °C. For sequencing, PCR products were purified using a PCR cleaning system (Qiagen/Westburg, Leusden, The Netherlands) and in some cases, when direct sequencing of the PCR product failed, the PCR product was isolated from agarose gel using a Qiaquick gel extraction kit (Qiagen/ Westburg, Leusden, The Netherlands). DNA sequencing reactions were performed in both directions using BigDyeTerminator v3.1 (Applied Biosystems, Nieuwerkerk a/d IJssel, The Netherlands). 2.3. DNA sequence analysis DNA sequences were stored and analysed using BioNumerics software version 4.5 (Applied Maths, Ghent,

Belgium). For species determination DNA sequences were compared with two reference strains for which the complete genomes have been sequenced: C. hominis (Xu et al., 2004) isolated from a child in Uganda (Akiyoshi et al., 2002; Tumwine et al., 2003) and C. parvum (Abrahamsen et al., 2004) isolated from a calf from Iowa, USA (Okhuysen et al., 1999). When differences with these were found, sequences were compared with those found in GenBank using a BLAST search (http://www.ncbi.nlm.nih.gov/). Representative DNA sequences, determined by the consensus sequence of each homogenous cluster of sequences, were submitted to GenBank, Accession Nos: DQ388376–DQ388391 and EF576952–EF576987. 3. Results 3.1. Genotyping of human Cryptosporidium isolates The 91 human isolates were genotyped for all six markers used in this study. Table 2 shows the genotypes identified for the human Cryptosporidium isolates. Sixty-four isolates were typed as C. hominis, 17 as C. parvum, nine as mixed C. hominis/C. parvum and one as C. felis. For COWP, all human C. parvum sequences analysed were identical to the C. parvum reference strain (Abrahamsen et al., 2004). The C. hominis isolates were identical to C. hominis AF266265 from GenBank and showed a single A to G transition (649 bp from the start of the forward primer) compared with the C. hominis reference strain (Xu et al., 2004). Genotyping for 18S was complicated by the finding that the 18S locus is present at multiple sites in the

812

P.R. Wielinga et al. / International Journal for Parasitology 38 (2008) 809–817

Table 2 Genotyping results for the human cryptosporidiosis cases analysed in this study Sample

Combineda

COWP

18S

ML1b

ML2b

HSP70

GP60b

1–63 64 65 66 67 68 69 70 71 72 73

Cryptosporidium hominis C. hominis Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed

C. hominis ndc C. parvum C. parvum C. parvum C. parvum C. parvum nd nd C. hominis nd

C. hominis NL1/NL2d nd C. parvum C. parvum C. parvum (KSU) C. parvum C. parvum nd nd C. hominis NL1 C. hominis NL1

C. hominis H1 nd C. parvum C3 C. parvum C2 NL1 C. parvum C2 NL2 C. parvum C1 C. parvum C2 C. hominis H1 C. hominis H1 C. hominis H1 C. hominis H1

C. hominis nd nd ML2-197 ML2-176 C. hominis C. hominis C. parvum C. hominis nd ML2-197

C. C. C. C. C. C. C. C. C. C. C.

hominis hominis parvum parvum NL parvum NL hominis hominis hominis parvum parvum parvum NL

IbA10G2 IdA17 IbA10G2 IcA5G3R2 IdA14 IIaA15G2R1 nd nd nd nd nd

74–80 81 82 83 84 85–87 88–89 90 91

Cryptosporidium parvum C. parvum C. parvum C. parvum C. parvum C. parvum C. parvum C. parvum C. felis

C. parvum C. parvum C. parvum nd C. parvum C. parvum C. parvum C. parvum nd

C. parvum C. parvum C. parvum nd C. parvum C. parvum C. parvum C. parvum nd

C. parvum C. parvum C. parvum nd C. parvum C. parvum C. parvum C. parvum nd

ML2-229 C. parvum nd nd ML2-191 ML2-227 ML2-197 ML2-191 nd

C. C. C. C. C. C. C. C. C.

parvum parvum parvum parvum parvum parvum parvum NL parvum felis

IIaA15G2R1 IIaA17G1R1 IIdA15G1 IIdA16G1 IIdA18G1 nd nd nd IbA10G2

C1 C2 C3 C3 C1 C2 NL1 C3

a

Combined results for the different loci studied; ‘‘mixed’’ indicates that different markers resulted in a different strain assignment. ML1/2 genotypes are given according to Caccio et al. (2000, 2001) and GP60 genotypes according to Strong et al. (2000), O’Connor et al. (2002), and Sulaiman et al. (2005). c nd, no data available. d For the C. hominis isolates the 18S sequences analysis showed in 33% of the cases the C. hominis NL1 variant and in the other cases the C. hominis NL2 variant (see text for details). COWP, Cryptosporidium outer wall protein gene. b

Cryptosporidium genome (Table 1). With the exception of one isolate identical to the KSU-1 strain (AF308600), all C. parvum isolates were identical to the 18S sequence on chromosome 7 of the reference strain (Table 2). For C. hominis two variants were found. One was named C. hominis NL1 (39% of cases) and contained a G to A transition (282 bp from the start of the forward primer) compared with the sequence on chromosome 2 of the reference strain. The second variant was named C. hominis NL2 (61% of cases) and was identical to the sequence on chromosome 2 of the reference strain. For ML1 all C. hominis sequences were of the H1 type (Caccio et al., 2000) and identical to the reference sequence (Xu et al., 2004) (Table 2). The human C. parvum isolates grouped into C1 (55%), C2 (25%), and C3 (20%) (Caccio et al., 2000). The C1 and C3 sub-types matched the published sequences (Caccio et al., 2000), with the C1 identical to the C. parvum reference sequence (Abrahamsen et al., 2004). For C2 we found three subtypes, one identical to the C2 published before (Caccio et al., 2000) and the other two, C2 NL1 and C2 NL2, containing a G to A transition at position 129 (bp from the start of the forward primer) and C2 NL2 in addition to a G to A transition at position 99 (bp from the start of the forward primer). For ML2 all C. hominis samples were identical to the C. hominis reference strain. For the C. parvum isolates sequencing was complicated due to the long GA-repeat

and for those that could be genotyped we found ML2176/191/197/227/299 variants (Caccio et al., 2001) (Table 2). In the samples for which the length of the GA-repeat could not be determined the flanking sequences were used to type the sample as either C. parvum or C. hominis (Table 2). For HSP70 all C. hominis isolates were identical to the C. hominis reference sequence (Xu et al., 2004). For C. parvum two genotypes were found: C. parvum (76%) and C. parvum NL (24%), the latter containing a C to T transition compared with the C. parvum reference sequence (Abrahamsen et al., 2004) (204 bp from the start of the forward primer) and not matching a sequence in GenBank. One of the isolates (1%) from a 2 year old was identical to C. felis AF221538. For GP60 four C. hominis genotypes were found belonging to the three sub-type families Ib/c/d: IbA10G2 (92.5%), IdA17 (2.5%), IdA14 (2.5%), and IcA5G3R2 (2.5%). IbA10G2 was the predominant genotype and was not identical to the C. hominis IaA19R3 reference strain. For C. parvum five sub-types were found belonging to the two sub-type families IIa/b: IIaA15G2R1 (14% of all GP60 positive cases), IIaA17G1R1 (2%), IIdA15G1 (2%), IIdA16G1 (2%), and IIdA18G1 (2%). IIaA15G2R1 was the dominant sub-type and was identical to the C. parvum reference sequence (Abrahamsen et al., 2004). The isolate identified by HSP70 as C. felis showed a IbA10G2 subtype.

P.R. Wielinga et al. / International Journal for Parasitology 38 (2008) 809–817 30%

Based on the success rate and the degree of differentiation found for the six markers in the human genotyping study, the farm animal isolates, which were isolated in the latter part of this study, were typed for HSP70, ML1, and GP60. Only C. parvum was found in the farm animals isolates (Table 3). For HSP70 a single C. parvum sub-type was found. For ML1 two sub-types were found: C. parvum C1 (92%) and C. parvum C2 (8%). For GP60 18 different sub-types were found belonging to sub-type families IIa and the recently published IIj (Misic and Abe, 2007). In 69% of the GP60-positive cases IIaA15G2R1 was found followed by 11% IIaA17G1R1, 5% IIaA16G3R1, 1.5% IIaA13G2R1, IIaA14G2R1, IIaA17G2R1, IIaA18G4R1, IIaA18R1, or IIaA19G2R1, and 0.75% IIaA11G2R1, IIaA16G1R1, IIaA19G1R1, IIaA18G1R1, IIaA16G2R1, IIaA18G3R1, IIaA21G3R1, or IIaA12G2R1. One isolate was typed as IIjA24R2 (0.75%).

25%

3.3. Correlation of human Cryptosporidium genotypes and patient information Only minor differences with the overall distribution of C. hominis (70%) and C. parvum (19%) were found between the different regions from which the samples originated (not shown). Fig. 1 shows the relationship between cryptosporidiosis cases and the time of year. The number of C. hominis cryptosporidiosis cases showed a peak in the period September–November, coinciding with the autumn

Percentage of total

3.2. Genotyping of cattle Cryptosporidium isolates

813

Cryptosporidiosis

C. hominis C. parvum

*

20%

*

15% 10% 5% 0%

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Month

Fig. 1. Seasonality of human cryptosporidiosis in The Netherlands. The number (squares) of human Cryptosporidium cases given as the percentage of the total number of samples, throughout the year. The distribution of the number Cryptosporidium hominis (triangle) and Cryptosporidium parvum (rhombus) cases is given as the percentage of the total number of Cryptosporidium positive cases. v2 testing shows a significant increased prevalence of C. hominis in the months of August and October (asterisk) with P-values <0.0001.

Cryptosporidium peak (Fig. 1). The C. parvum cryptosporidiosis cases were found more or less evenly distributed over the year. Analyses of the age distribution of the patients showed two patients groups, one consisting of children aged between 0 and 9 years and a second consisting of adults over 25 years old (Fig. 2A). The majority of cryptosporidiosis cases occurred in the children’s group (Fig. 2A). Genotypic analyses showed that in the group of children, C. hominis infections were found significantly more frequently than C. parvum (Fig. 2B). C. parvum infections

Table 3 Genotyping results for the cryptosporidiosis cases found in animals in this study Sample

Host

ML1a

HSP70

GP60a

1 2 3–4 5–6 7–87 88–96 97 98 99–104 105–116 117–118 119–120 121 122 123–124 125–126 127 128–129 130 131 132–159 160–161

Cow Cow Cow Cow Cow Cow and sheep Cow Cow Cow Cow Cow Cow Goat Cow Cow Cow Cow Cow Cow Cow Cow Cow

Cryptosporidium parvum C1 C. parvum C1 C. parvum C1 C. parvum C1 C. parvum C1 C. parvum C2 C. parvum C1 C. parvum C1 C. parvum C1 C. parvum C1 C. parvum C2 C. parvum C1 C. parvum C2 C. parvum C1 C. parvum C1 C. parvum C1 C. parvum C1 C. parvum C1 C. parvum C1 C. parvum C2 C. parvum C1 C. parvum C2

C. parvum ndb C. parvum C. parvum nd C. parvum C. parvum C. parvum C. parvum C. parvum C. parvum nd C. parvum C. parvum nd C. parvum C. parvum C. parvum C. parvum C. parvum nd C. parvum

IIaA11G2R1 IIaA12G2R1 IIaA13G2R1 IIaA14G2R1 IIaA15G2R1 IIaA15G2R1 IIaA16G1R1 IIaA16G2R1 IIaA16G3R1 IIaA17G1R1 IIaA17G1R1 IIaA17G2R1 IIaA18G1R1 IIaA18G3R1 IIaA18G4R1 IIaA18R1 IIaA19G1R1 IIaA19G2R1 IIaA21G3R1 IIjA24R2 nd nd

a ML1/2 genotypes are given according to Caccio et al. (2000, 2001) and GP60 genotypes according to O’Connor et al. (2002), Strong et al. (2000), and Sulaiman et al. (2005). b nd, no data available.

814

P.R. Wielinga et al. / International Journal for Parasitology 38 (2008) 809–817

showed a similar ML1 and GP60 genotype distribution over the entire sample period. Of the cattle samples analysed, 87% came from dairy farms and 13% from meat producing farms. There was no significant difference in ML1 (dairy 91% C1 and 9% C2; meat 89% C1 and 11% C2) or GP60 (IIaA15G2R1, 68% dairy and 63% meat) genotype distribution between the two types of animals. Comparison of the country of origin and the GP60 showed that all 161 farm animal samples belonged to the IIa sub-type family, while a single IIjA24R2 isolate originated from a calf from Lithuania.

40%

30%

(%)

Human cryptosporidiosis cases

A

20%

10%

0% 60 - 65

55 - 60

50 - 55

45 - 50

40 - 45

35 - 40

30 - 35

25 - 30

20 - 25

15 - 20

10 - 15

5 - 10

0-5

B

Percentage of total number of cases

Age (years)

100%

*** **

0-9 years (n = 48) >25 years (n = 20)

80%

* 60%

40% 20% 0% C. hominis

C. parvum

C. hominis/ parvum

C. felis

Fig. 2. Age-dependent distribution of human cryptosporidiosis in The Netherlands. (A) Relationship between the number of cryptosporidiosis cases and patient age, given as the percentage of the total number of cases (n = 80) for which the patient’s age was known. For 11 samples the ages of the patients were not disclosed. (B) Age-dependent distribution of Cryptosporidium hominis and Cryptosporidium parvum for the two age groups, those between 0 and 9 years (black bars) and those over 25 years (white bars), expressed relative to the total number of cases found in the different age groups. v2 testing shows significant differences between the children and adults with P-values of *0.009, **0.1, ***0.002.

were found more frequently in the adult group (Fig. 2B). There was no significant correlation between gender and C. parvum or C. hominis prevalence: 69% and 21% for males, and 59% and 26% for females, respectively. 3.4. Correlation of animal Cryptosporidium genotypes with epidemiology data The mean age of the animals studied and suffering from Cryptosporidiosis was 13 days (range 2–125 days) with only 7% of the animals older than 3 weeks. Differentiating the samples between those originating from animals older and younger than 21 days showed no difference in ML1 and GP60 genotype distribution and C. parvum C1IIaA15G2R1 was the predominant genotype (older group 75%; younger group 69%). The farm animal samples (36– 40 per month) were collected in the period September–February, before, during and after the human cryptosporidiosis peak (see Fig. 2). Comparison of the different months

3.5. Comparison of human and animal genotypes To study the zoonotic potential of Cryptosporidium, the human and animal genotypes were compared. Table 4 compares the human and farm animal HSP70, ML1 and GP60 results for those samples for which all three markers were available. IbA10G2 was only found in humans. In addition, isolates genotyped as HSP70 C. parvum NL, ML1 C. parvum C2 NL1/2 and C. parvum C3, and GP60 IId genotypes were only detected in humans. Multilocus type C. parvum C1IIaA15G2R1 and C. parvum C1-IIaA17G1R1 were found most frequently in both humans and cattle. The goat isolate showed the novel sub-type IIaA18G1R1. Other than sporadic cases of IIa and IIj, the three multilocus types C. parvum C1-IIaA17G1R1, C. parvum C1-IIaA16G3R1 and C. parvum C2-IIaA15G2R1 were only detected in the sheep and cattle isolates. 4. Discussion Ninety-one Cryptosporidium-positive human faecal samples were genotyped for 18S, COWP, HSP70, GP60, ML1, and ML2. With the exception of one C. felis isolate (1%), which is normally found in cats and sporadically in humans (Pieniazek et al., 1999; Leoni et al., 2006), all human infections were caused by C. hominis and/or C. parvum, similar to findings in other countries (Enemark et al., 2002; Hunter et al., 2004; Caccio et al., 2005; Chalmers et al., 2005). There was a good correlation between the six genetic markers that we used: in 90% of the human cases that we analysed the isolates assigned to the same species by all markers: 70% C. hominis and 19% C. parvum. The remaining 10% showed a mixed genotype of both C. hominis and C. parvum depending on the marker considered. This discrepancy between the markers might be due to the presence of recombined strains, which might have been established through genetic recombination (e.g., sexually) of C. hominis and C. parvum (Feng et al., 2002) or it might be caused by a mixed C. hominis/C. parvum infection. For the farm animals in this study we found only C. parvum and no C. andersoni or Cryptosporidium bovis. This correlates with what has been found in cattle in the USA, where C. parvum is only found in young calves (younger than 2 months) and C. andersoni and C. bovis only in older cows (Fayer et al., 2007). For HSP70 a single C. parvum

P.R. Wielinga et al. / International Journal for Parasitology 38 (2008) 809–817

815

Table 4 Comparison of HSP70, ML1 and GP60 genotypes found in humans and farm animals HSP70

ML1

GP60

Cryptosporidium Hominis

C. parvum C1 C. hominis H1

IIaA15G2R1 IbA10G2 IdA17

Cryptosporidium parvum

C. parvum C1

IIaA11G2R1 IIaA13G2R1 IIaA14G2R1 IIaA15G2R1 IIaA16G1R1 IIaA16G2R1 IIaA16G3R1 IIaA17G1R1 IIaA17G2R1 IIaA18G3R1 IIaA18G4R1 IIaA18R1 IIaA19G1R1 IIaA19G2R1 IIaA21G3R1

1 2 2 62 1 1 5 10 1 1 1 2 1 2 1

IIaA15G2R1 IIaA17G1R1 IIaA18G1R1 IIjA24R2

7 2

C. parvum C2

C. parvum NL a

Cow (%)

Goat (%)

Sheep (%)

Human (%) 2 74 2

10

100a 2 100a

1

C. parvum C3

IbA10G2 IIdA15G1 IIdA18G1

2 2 2

C. parvum C2 NL1 C. parvum C2 NL2

IcA5G3R2 IdA14

2 2

Note that this value is base on (n = 1) one sample.

variant was found and for ML1 two variants were found, C. parvum C1 (8%) and C. parvum C2 (92%). For GP60, with one exception, 18 variants were identified belonging to the IIa family, with IIaA15G2R1 (69%), IIaA17G1R1 (11%), and IIaA16G3R1 (5%) as the variants found most frequently. Besides the IIa isolates, one IIjA24R2 isolate was identified, which came from a calf from Lithuania. Comparison with GenBank showed that this genotype has only previously been found in Serbia and Montenegro (Misic and Abe, 2007). In the present study, we found only C. parvum C1 and C2, whereas C. parvum C3 and C1 were previously reported in Dutch dairy cattle (Huetink et al., 2001). This discrepancy may be caused by the difference in sample selection. We only included clinical cases of animals with a mean age of 13 days originating from farms throughout The Netherlands, while in the earlier study healthy animals with a mean age of 1–48 months coming from a single farm were studied. Comparison of the different genotypes found in humans and farm animals was performed for the HSP70, ML1 and GP60 markers. COWP, 18S, and ML2 were not considered for the animal study as the human study revealed that the COWP and 18S loci were highly conserved and ML2 genotyping had a low success rate. In addition, the finding that multiple and different copies of the 18S locus exist in a single strain, made this a less favourable marker for genotyping.

Of the two HSP70 variants found, only the C. parvum NL variant was found in humans, suggesting that this strain is most probably transmitted from human to human and may be part of an anthroponotic cycle, as is the case for C. hominis. This assumption is strengthened by the finding that the HSP70 C. parvum NL sequence is linked to the ML1 C. parvum C2 NL1/2 variants which were also restricted to the human isolates. Of the five human ML1 variants (C1, C2, C3, and C2 NL1 and C2 NL2) only C1 and C2 were found in the farm animals. This suggests that in The Netherlands C1 and C2 strains are potentially zoonotic, while the C2 NL1/2 strains may be part of an anthroponotic cycle. Besides single cases of IdA14, IdA17, and IcA5G3R2, the majority of the human GP60 C. hominis cases were identified as IbA10G2; none of these strains were found in farm animals. The multilocus genotype C. parvum C1-IIaA15G2R1 was predominant in both humans and farm animals and thus humans may have acquired this strain either by zoonotic or anthroponotic transmission. This may also be the case for the C. parvum C2-IIaA17G1R1 strain which was found in both the human and cattle isolates, though at a much lower frequency. The IId strains were only found in humans, suggesting these are anthroponotic or acquired from another reservoir. However, as the prevalence of these strains is very low, we cannot exclude that they are present in the

816

P.R. Wielinga et al. / International Journal for Parasitology 38 (2008) 809–817

farm animal population at a low prevalence (<1%) and can only be detected by testing more isolates. Comparing the seasonality of human C. hominis and C. parvum infections shows a peak for C. hominis in the period September–November, which is also the period during which most cryptosporidiosis cases are diagnosed (Mank, T.G., 1997. Intestinal protozoa and diarhoea in general practice. Academic thesis, Vrije Universiteit Amsterdam, The Netherlands. ISBN9 056690856). This September– November peak was not found for C. parvum cases, which were found at a lower frequency throughout the year. A similar seasonality for human cryptosporidiosis cases was found by others in Europe and also for livestock and small wild animals (mainly C. parvum) in the UK (Sturdee et al., 2003). Our data, however, clearly show that the peak in human infections is not caused by zoonotic, but by anthroponotic transmission of C. hominis. Correlation of age and genotypes showed that there is a tendency to find more C. parvum in human patients over 25 years old, and more C. hominis in 0–9 years old patients. It is unclear why older patients show a higher frequency of C. parvum than C. hominis infections. A possible explanation may be age-related resistance, which has been found for C. parvum infection in lambs (Ortega-Mora and Wright, 1994) which may be related to increased intestinal flora (Harp, 2003) and IL-12- or MyD88-status (Smith et al., 2001; Rogers et al., 2006). The finding that C. hominis IbA10G2 is the dominant genotype in humans in The Netherlands and is not found in farm animals indicates that most of the human Cryptosporidium infections occur via an anthroponotic cycle. The opposite holds true for IIaA15G2R1, which forms the most dominant C. parvum strain in both humans and farm animals, suggesting that this strain might be exchanged between humans and farm animals. Similar findings have been reported for other countries, for instance India, Kenya, Malawi, Peru, South Africa, Thailand, Uganda, the UK and USA, which is indicative of the success of these strains in infecting humans and animals (Peng et al., 1997; Sulaiman et al., 1998; Xiao et al., 2001; Leav et al., 2002; Gatei et al., 2003; Peng et al., 2003; Gatei et al., 2006; Leoni et al., 2006). There are several exceptions to this distribution, e.g., children in Kuwait were mainly infected with IIaA15G2R1 and IIdA20G1 (Sulaiman et al., 2005) and a study of sporadic cryptosporidiosis cases in Wisconsin (USA) showed mainly infections with IIaA15G2R1 (Feltus et al., 2006). The human isolates typed as IIdA15G1, IIdA18G1, IIaA16G2R1, IIaA16G3R1, IcA5G3R2, IdA14, and IdA17, were found with a low frequency and these were not found in the farm animals. This finding raises various questions about the origin of these sub-types. From which other (animal) reservoir and via which transmission routes were they acquired? Are they human-specific and do they form a specific anthroponotic transmission cycle or are they acquired during travel? Currently, we are trying to collect such data by collaborating in the Cryptosporidium and Giarda workgroup ZOOPNET (http://hypocrates.rivm.nl/).

In ZOOPNET we have constructed a web-based database to collects DNA sequence and epidemiological data. Ultimately, we anticipate that via this collaborative database we will obtain a better picture of the molecular epidemiology, the zoonotic potential and the major and minor transmission routes of Cryptosporidium. Acknowledgements We thank the diagnostic laboratories of Enschede, Goes, Utrecht and Veldhoven for providing us with human patient samples. This study was financially supported by the Dutch Food and Consumer Product Safety Authority (VWA). References Abrahamsen, M.S., Templeton, T.J., Enomoto, S., Abrahante, J.E., Zhu, G., Lancto, C.A., Deng, M., Liu, C., Widmer, G., Tzipori, S., Buck, G.A., Xu, P., Bankier, A.T., Dear, P.H., Konfortov, B.A., Spriggs, H.F., Iyer, L., Anantharaman, V., Aravind, L., Kapur, V., 2004. Complete genome sequence of the apicomplexan, Cryptosporidium parvum. Science 304, 441–445. Akiyoshi, D.E., Feng, X., Buckholt, M.A., Widmer, G., Tzipori, S., 2002. Genetic analysis of a Cryptosporidium parvum human genotype 1 isolate passaged through different host species. Infect. Immun. 70, 5670–5675. Caccio, S., Homan, W., Camilli, R., Traldi, G., Kortbeek, T., Pozio, E., 2000. A microsatellite marker reveals population heterogeneity within human and animal genotypes of Cryptosporidium parvum. Parasitology 120, 237–244. Caccio, S., Pinter, E., Fantini, R., Mezzaroma, I., Pozio, E., 2002. Human infection with Cryptosporidium felis: case report and literature review. Emerg. Infect. Dis. 8, 85–86. Caccio, S., Spano, F., Pozio, E., 2001. Large sequence variation at two microsatellite loci among zoonotic (genotype C) isolates of Cryptosporidium parvum. Int. J. Parasitol. 31, 1082–1086. Caccio, S.M., Thompson, R.C., McLauchlin, J., Smith, H.V., 2005. Unravelling Cryptosporidium and Giardia epidemiology. Trends Parasitol. 21, 430–437. Chalmers, R.M., Ferguson, C., Caccio, S., Gasser, R.B., Abs EL-Osta, Y.G., Heijnen, L., Xiao, L., Elwin, K., Hadfield, S., Sinclair, M., Stevens, M., 2005. Direct comparison of selected methods for genetic categorisation of Cryptosporidium parvum and Cryptosporidium hominis species. Int. J. Parasitol. 35, 397–410. de Wit, M.A., Koopmans, M.P., Kortbeek, L.M., van Leeuwen, N.J., Bartelds, A.I., van Duynhoven, Y.T., 2001a. Gastroenteritis in sentinel general practices, The Netherlands. Emerg. Infect. Dis. 7, 82–91. de Wit, M.A., Koopmans, M.P., Kortbeek, L.M., van Leeuwen, N.J., Vinje, J., van Duynhoven, Y.T., 2001b. Etiology of gastroenteritis in sentinel general practices in The Netherlands. Clin. Infect. Dis. 33, 280–288. de Wit, M.A., Koopmans, M.P., Kortbeek, L.M., Wannet, W.J., Vinje, J., van Leusden, F., Bartelds, A.I., van Duynhoven, Y.T., 2001c. Sensor, a population-based cohort study on gastroenteritis in The Netherlands: incidence and etiology. Am. J. Epidemiol. 154, 666–674. de Wit, M.A., Kortbeek, L.M., Koopmans, M.P., de Jager, C.J., Wannet, W.J., Bartelds, A.I., van Duynhoven, Y.T., 2001d. A comparison of gastroenteritis in a general practice-based study and a communitybased study. Epidemiol. Infect. 127, 389–397. Enemark, H.L., Ahrens, P., Juel, C.D., Petersen, E., Petersen, R.F., Andersen, J.S., Lind, P., Thamsborg, S.M., 2002. Molecular characterization of Danish Cryptosporidium parvum isolates. Parasitology 125, 331–341. Fayer, R., Santin, M., Trout, J.M., 2007. Prevalence of Cryptosporidium species and genotypes in mature dairy cattle on farms in eastern United

P.R. Wielinga et al. / International Journal for Parasitology 38 (2008) 809–817 States compared with younger cattle from the same locations. Vet. Parasitol. 145, 260–266. Feltus, D.C., Giddings, C.W., Schneck, B.L., Monson, T., Warshauer, D., McEvoy, J.M., 2006. Evidence supporting zoonotic transmission of Cryptosporidium spp. in Wisconsin. J. Clin. Microbiol. 44, 4303–4308. Feng, X., Rich, S.M., Tzipori, S., Widmer, G., 2002. Experimental evidence for genetic recombination in the opportunistic pathogen Cryptosporidium parvum. Mol. Biochem. Parasitol. 119, 55–62. Gatei, W., Das, P., Dutta, P., Sen, A., Cama, V., Lal, A.A., Xiao, L., 2006. Multilocus sequence typing and genetic structure of Cryptosporidium hominis from children in Kolkata, India. Infect. Genet. Evol. 7, 197–205. Gatei, W., Greensill, J., Ashford, R.W., Cuevas, L.E., Parry, C.M., Cunliffe, N.A., Beeching, N.J., Hart, C.A., 2003. Molecular analysis of the 18S rRNA gene of Cryptosporidium parasites from patients with or without human immunodeficiency virus infections living in Kenya, Malawi, Brazil, the United Kingdom, and Vietnam. J. Clin. Microbiol. 41, 1458– 1462. Harp, J.A., 2003. Cryptosporidium and host resistance: historical perspective and some novel approaches. Anim. Health Res. Rev. 4, 53–62. Henriksen, S.A., Pohlenz, J.F., 1981. Staining of cryptosporidia by a modified Ziehl–Neelsen technique. Acta Vet. Scand. 22, 594–596. Homan, W., van Gorkom, T., Kan, Y.Y., Hepener, J., 1999a. Characterization of Cryptosporidium parvum in human and animal feces by single-tube nested polymerase chain reaction and restriction analysis. Parasitol. Res. 85, 707–712. Homan, W., van Gorkom, T., Kan, Y.Y., Hepener, J., 1999b. Characterization of Cryptosporidium parvum in human and animal feces by single-tube nested polymerase chain reaction and restriction analysis. Parasitol. Res. 85, 707–712. Huetink, R.E., van der Giessen, J.W., Noordhuizen, J.P., Ploeger, H.W., 2001. Epidemiology of Cryptosporidium spp. and Giardia duodenalis on a dairy farm. Vet. Parasitol. 102, 53–67. Hunter, P.R., Hughes, S., Woodhouse, S., Syed, Q., Verlander, N.Q., Chalmers, R.M., Morgan, K., Nichols, G., Beeching, N., Osborn, K., 2004. Sporadic cryptosporidiosis case-control study with genotyping. Emerg. Infect. Dis. 10, 1241–1249. Leav, B.A., Mackay, M.R., Anyanwu, A., O’ Connor, R.M., Cevallos, A.M., Kindra, G., Rollins, N.C., Bennish, M.L., Nelson, R.G., Ward, H.D., 2002. Analysis of sequence diversity at the highly polymorphic Cpgp40/15 locus among Cryptosporidium isolates from human immunodeficiency virus-infected children in South Africa. Infect. Immun. 70, 3881–3890. Leoni, F., Amar, C., Nichols, G., Pedraza-Diaz, S., McLauchlin, J., 2006. Genetic analysis of Cryptosporidium from 2414 humans with diarrhoea in England between 1985 and 2000. J. Med. Microbiol. 55, 703–707. Medema, G.J., Schijven, J.F., 2001. Modelling the sewage discharge and dispersion of Cryptosporidium and Giardia in surface water. Water Res. 35, 4307–4316. Misic, Z., Abe, N., 2007. Subtype analysis of Cryptosporidium parvum isolates from calves on farms around Belgrade, Serbia and Montenegro, using the 60 kDa glycoprotein gene sequences. Parasitology 134, 351–358. Morgan, U.M., Monis, P.T., Xiao, L., Limor, J., Sulaiman, I., Raidal, S., O’Donoghue, P., Gasser, R., Murray, A., Fayer, R., Blagburn, B.L., Lal, A.A., Thompson, R.C., 2001. Molecular and phylogenetic characterisation of Cryptosporidium from birds. Int. J. Parasitol. 31, 289–296. O’Connor, R.M., Thorpe, C.M., Cevallos, A.M., Ward, H.D., 2002. Expression of the highly polymorphic Cryptosporidium parvum Cpgp40/15 gene in genotype I and II isolates. Mol. Biochem. Parasitol. 119, 203–215. Okhuysen, P.C., Chappell, C.L., Crabb, J.H., Sterling, C.R., DuPont, H.L., 1999. Virulence of three distinct Cryptosporidium parvum isolates for healthy adults. J. Infect. Dis. 180, 1275–1281. Ortega-Mora, L.M., Wright, S.E., 1994. Age-related resistance in ovine cryptosporidiosis: patterns of infection and humoral immune response. Infect. Immun. 62, 5003–5009.

817

Pedraza-Diaz, S., Amar, C., McLauchlin, J., 2000. The identification and characterisation of an unusual genotype of Cryptosporidium from human faeces as Cryptosporidium meleagridis. FEMS Microbiol. Lett. 189, 189–194. Peng, M.M., Meshnick, S.R., Cunliffe, N.A., Thindwa, B.D., Hart, C.A., Broadhead, R.L., Xiao, L., 2003. Molecular epidemiology of cryptosporidiosis in children in Malawi. J. Eukaryot. Microbiol. 50, 557–559. Peng, M.M., Xiao, L., Freeman, A.R., Arrowood, M.J., Escalante, A.A., Weltman, A.C., Ong, C.S., Mac Kenzie, W.R., Lal, A.A., Beard, C.B., 1997. Genetic polymorphism among Cryptosporidium parvum isolates: evidence of two distinct human transmission cycles. Emerg. Infect. Dis. 3, 567–573. Pieniazek, N.J., Bornay-Llinares, F.J., Slemenda, S.B., da Silva, A.J., Moura, I.N., Arrowood, M.J., Ditrich, O., Addiss, D.G., 1999. New Cryptosporidium genotypes in HIV-infected persons. Emerg. Infect. Dis. 5, 444–449. Rogers, K.A., Rogers, A.B., Leav, B.A., Sanchez, A., Vannier, E., Uematsu, S., Akira, S., Golenbock, D., Ward, H.D., 2006. MyD88dependent pathways mediate resistance to Cryptosporidium parvum infection in mice. Infect. Immun. 74, 549–556. Schets, F.M., van den Berg, H.H., Engels, G.B., Lodder, W.J., de Roda Husman, A.M., 2007. Cryptosporidium and Giardia in commercial and non-commercial oysters (Crassostrea gigas) and water from the Oosterschelde, The Netherlands. Int. J. Food Microbiol. 113, 189–194. Smith, L.M., Bonafonte, M.T., Campbell, L.D., Mead, J.R., 2001. Exogenous interleukin-12 (IL-12) exacerbates Cryptosporidium parvum infection in gamma interferon knockout mice. Exp. Parasitol. 98, 123–133. Strong, W.B., Gut, J., Nelson, R.G., 2000. Cloning and sequence analysis of a highly polymorphic Cryptosporidium parvum gene encoding a 60kilodalton glycoprotein and characterization of its 15- and 45kilodalton zoite surface antigen products. Infect. Immun. 68, 4117– 4134. Sturdee, A.P., Bodley-Tickell, A.T., Archer, A., Chalmers, R.M., 2003. Long-term study of Cryptosporidium prevalence on a lowland farm in the United Kingdom. Vet. Parasitol. 116, 97–113. Sulaiman, I.M., Hira, P.R., Zhou, L., Al-Ali, F.M., Al-Shelahi, F.A., Shweiki, H.M., Iqbal, J., Khalid, N., Xiao, L., 2005. Unique endemicity of cryptosporidiosis in children in Kuwait. J. Clin. Microbiol. 43, 2805–2809. Sulaiman, I.M., Xiao, L., Yang, C., Escalante, L., Moore, A., Beard, C.B., Arrowood, M.J., Lal, A.A., 1998. Differentiating human from animal isolates of Cryptosporidium parvum. Emerg. Infect. Dis. 4, 681–685. Thompson, R.C., Olson, M.E., Zhu, G., Enomoto, S., Abrahamsen, M.S., Hijjawi, N.S., 2005. Cryptosporidium and cryptosporidiosis. Adv. Parasitol. 59, 77–158. Tumwine, J.K., Kekitiinwa, A., Nabukeera, N., Akiyoshi, D.E., Rich, S.M., Widmer, G., Feng, X., Tzipori, S., 2003. Cryptosporidium parvum in children with diarrhea in Mulago Hospital, Kampala, Uganda. Am. J. Trop. Med. Hyg. 68, 710–715. Verweij, J.J., Blange, R.A., Templeton, K., Schinkel, J., Brienen, E.A., van Rooyen, M.A., van Lieshout, L., Polderman, A.M., 2004. Simultaneous detection of Entamoeba histolytica, Giardia lamblia, and Cryptosporidium parvum in fecal samples by using multiplex realtime PCR. J. Clin. Microbiol. 42, 1220–1223. Xiao, L., Bern, C., Limor, J., Sulaiman, I., Roberts, J., Checkley, W., Cabrera, L., Gilman, R.H., Lal, A.A., 2001. Identification of 5 types of Cryptosporidium parasites in children in Lima, Peru. J. Infect. Dis. 183, 492–497. Xiao, L., Ryan, U.M., 2004. Cryptosporidiosis: an update in molecular epidemiology. Curr. Opin. Infect. Dis. 17, 483–490. Xu, P., Widmer, G., Wang, Y., Ozaki, L.S., Alves, J.M., Serrano, M.G., Puiu, D., Manque, P., Akiyoshi, D., Mackey, A.J., Pearson, W.R., Dear, P.H., Bankier, A.T., Peterson, D.L., Abrahamsen, M.S., Kapur, V., Tzipori, S., Buck, G.A., 2004. The genome of Cryptosporidium hominis. Nature 431, 1107–1112.