Introduction

Cryptosporidium spp. are among the most important coccidian parasites of humans, as well as of other animals, birds, reptiles, and fish. The ingestion of low numbers of oocysts can cause severe diarrheal disease (cryptosporidiosis). In humans, healthy individuals may clear the infection in less than a month, but those whose immune systems are compromised, particularly AIDS patients, transplant patients, and cancer patients, suffer prolonged and potentially fatal episodes of diarrhea. Cryptosporidiosis may occur sporadically or as outbreaks following zoonotic transmission from farm animals, person-to-person spread, or the contamination of water supplies (Casemore 1991).

New species and genotypes of Cryptosporidium have been identified in recent years and there is evidence that more than one Cryptosporidium species is involved in human disease (Hunter and Thompson 2005; Xiao et al. 2004). Several species of Cryptosporidium have been described including C. parvum and C. hominis which are the most common species causing human disease (Hunter and Thompson 2005). Other species have occasionally been reported to cause human illness including C. felis, C. meleagridis, C. canis, C. muris, C. andersoni, and C. suis (Read et al. 2004; Leoni et al. 2006). Currently, a total of 16 Cryptosporidium species are considered to be valid, but C. hominis and C. parvum appear to be most widely distributed (Slapeta 2006). An important difference between these two genotypes is that C. hominis appears largely responsible for human to human transmission, while C. parvum is typically responsible for animal to human transmission. Cattle have been described as major hosts for C. parvum, C. bovis, Cryptosporidium deer-like genotype, and C. andersoni, and several studies have shown host age-related susceptibility: C. parvum predominates in pre-weaned calves, C. bovis and the deer-like genotype in post-weaned calves, and C. andersoni in older calves and adult cattle (Santin et al. 2004; Slapeta 2006).

The presence of Cryptosporidium spp. oocysts in water is increasingly recognized as a problem throughout the world (Karanis et al. 2007a). Drinking water sources can become contaminated through introduction of infected feces from wild and/or domestic animals or from sewage discharged to surface waters. Outbreaks have been reported worldwide attributed to a combination of disinfection resistance of oocysts and water treatment deficiencies allowing passage of sufficient numbers of protozoa to cause illness (Karanis et al. 2007a).

In Thailand, the most prevalence information on Cryptosporidium in human and cattle has been based on microscopy (Uga et al. 1998; Saksirisampant et al. 2002; Tiangtip and Jongwutiwes 2002; Gatei et al. 2002). Accordingly, information on Cryptosporidium species or genotypes is rare. The aim of this work was to determine the genotypes of Cryptosporidium species among AIDS patients and dairy cattle in Thailand.

Materials and methods

Specimen collection, Cryptosporidium oocyst purification, and microscopic examination

Fecal samples were randomly collected from 46 HIV patients (ranged in age from 24 to 62 years) from Prabat Nampu Temple, Lop Buri Province in central Thailand. Further 200 samples from young dairy cattle (ranged in age from 10 days to 4 months) were collected from five farms in Chon Buri Province, in eastern Thailand. Samples were collected between January and August, 2007. Oocysts from each sample were concentrated by Sheather’s sucrose flotation technique and discontinuous sucrose gradient concentration. Harvested concentrates were washed in distilled water and sedimented by centrifuging. The pellets were stained by the acid fast stain (AFS). Microscopic examination was performed in all samples (Ongerth and Stibbs 1987). The fecal samples were preserved in 2.5% potassium dichromate and stored at 4°C. An aliquot of 200 μl of the sample suspension in 2.5% potassium dichromate was used for genotypic analysis.

Extraction of Cryptosporidium genomic DNA

Altogether, 129 samples (46 human and 83 cattle) were subjected to DNA extraction and nested polymerase chain reaction (PCR) analysis of the SSUrRNA gene according to Nichols et al. (2003), Plutzer and Karanis (2007), and Karanis et al. (2007b). DNA was extracted from fecal samples using the QIAmp DNA Stool Mini Kit (Qiagen GmbH, Hilden, Germany), according to the manufacturer’s descriptions, with the addition of three 10-min freeze–thaw cycles after resuspension in lysis solution in order to rupture the Cryptosporidium oocysts. Liquid nitrogen was used for freezing, and thawing was carried out at 70°C in a Dry Thermo unit (DTU-2B, Taitec). DNA was eluted in 100 μl buffer (Qiagen) and stored at −20°C until use.

Amplification of Cryptosporidium DNA by nested PCR

The 18S SSUrRNA nested PCR was performed as described by Nichols et al. (2003) and reported in previous investigations in our laboratory (Plutzer and Karanis 2007; Karanis et al. 2007b). Briefly, first the PCR product of a 655–667-bp segment, and second, the PCR product of a 435-bp segment were amplified in standard mixtures of 25 μl containing 400 nmol (1 μl) of each SSUrRNA specific primer, 200 μM dNTP (2.5 μl), 1.5 mM MgCl2 (3 μl) and 2.5 U (0.25 μl) HotStarTaq DNA polymerase (Qiagen), and 5 μl of DNA template. The templates were subjected to 35 amplification cycles (94°C for 30 s, 68°C at primary PCR and 60°C at secondary PCR for 60 s, 72°C for 45 s) followed by 1 cycle of 7 min at 72°C and held at 4°C. The PCR products were analyzed by 1.5% horizontal agarose gel electrophoresis with TAE buffer, then stained with 0.5 μg/ml ethidium bromide. Finally, the gels were visualized and photographed under ultraviolet light.

Sequencing and Cryptosporidium genotyping

Direct sequencing of PCR products was performed for genotyping of Cryptosporidium species. Out of ten positive samples by nested PCR, eight were successfully sequenced. The 18S PCR products were purified using the QIAquick PCR Purification Kit (Qiagen GmbH, Germany) and applied as templates for sequencing using the forward and reverse primers of the nested (secondary) PCR. Sequencing was conducted using the Big Dye Terminator V3.1 Cycle Sequencing Kit (Applied Biosystems, USA) according to the manufacturer’s protocol. An ABI Prism 3100 Genetic Analyzer automated sequencer (Applied Biosystems, USA) was used to analyze the sequencing reactions. The obtained sequences were compared with published sequences of Cryptosporidium species and genotypes on the NCBI server (http://www.ncbi.nlm.nih.gov/BLAST/). Data on the nucleotide sequences reported in this paper are available in the GenBank database under accession numbers EU606194–EU606201.

Results

Cryptosporidium identification in HIV patient specimens

Cryptosporidium oocysts in positive samples from HIV patients when acid-fast-stained appeared bright red, round to ovoid bodies against a pale green background, often showing elongated naked sporozoites. Cryptosporidium oocysts were found in 13 of 46 samples examined from HIV patients (eight females and five males; Table 1). The overall prevalence was 28.7%. Out of 46 HIV patients, 13 (28.26%) were exhibiting diarrhea and 33 (71.74%) were not. Five of the patients with diarrhea (38.46%) and eight without diarrhea (61.53%) were found positive for Cryptosporidium. The rate of Cryptosporidium finding in patients with diarrhea (38.46%) was not significantly different (p > 0.05) than in patients without diarrhea (61.53%). No significant association was found between Cryptosporidium positivity and age or sex in HIV patients (p > 0.05). No further information is available from the patients investigated. By nested PCR, Cryptosporidium species infection was diagnosed in two of 46 (4.35%) HIV patients (Table 1). Sequencing data indicated that the species found in each case was C. parvum.

Table 1 Detection and genotyping of Cryptosporidium in the investigated HIV patients by acid fast staining (AFS) and by nested PCR
Full size table

Cryptosporidium identification in cattle samples

Cryptosporidium oocysts were found in 26 out of 200 (13%) cattle samples examined by AFS (Table 2). Using the nested PCR method in 83 samples, Cryptosporidium species infections were identified in eight out of 83 (9.63%) cattle samples (Table 2).

Table 2 Detection and genotyping of Cryptosporidium in the investigated cattle in Thailand by acid fast staining (AFS) and by nested PCR
Full size table

Genotyping of Cryptosporidium species from HIV patient and cattle samples

From two of the HIV patient samples, the DNA sequence of the 18S rRNA gene from the nested PCR positive samples matched the published sequence of C. parvum (Table 1). Table 2 shows five of eight successfully sequenced samples out of the eight nested PCR positive cattle samples and all matched the published sequence for C. parvum.

Discussion

In the present study, the prevalence of Cryptosporidium species in HIV positive patients from Prabat Nampu Temple by using acid fast stain was 28.7%. No significant differences were observed between the infection rates observed in both males and females. According to our AFS results, the prevalence of cryptosporidiosis was higher (3.7%) than reported as for example in 1,500 children with acute diarrhea in Thailand (Thamlikitkul et al. 1987). Cryptosporidium was reported in 17 children (8%) from 205 institutionalized orphans 1–61 months of age in Bangkok using the same method (Janoff et al. 1990). Cryptosporidium oocysts were detected in fecal specimens from orphanage children in Nonthaburi Province by a modified cold Kinyoun acid-fast technique (Jongwutiwes et al. 1990; Jirapinyo et al. 2002). Other researchers have found Cryptosporidium oocysts in 10% of Cryptosporidium spp. in HIV seropositive and 2% of those in HIV seronegative subjects in Thailand (Uga et al. 1998; Waywa et al. 2001; Moolasart et al. 1995; Saksirisampant et al. 2002).

In previous studies, in HIV seropositive patients from other countries, ZN staining indicated Cryptosporidium in 37.3% in France (Cotte et al. 1993), 14% in Zambia (Chintu et al. 1995), 21.5% in South Italy (Brandonisio et al. 1999), 50% in Malaysia (Oguntibeju 2006), and 2.2% in Australia (Stark et al. 2007). A recent study from Malaysia (Zaidah et al. 2008) which identified C. parvum infections in 16% stool specimens collected from HIV-infected hospitalized patients by nested PCR highlights the importance of C. parvum in HIV patients. Other reports from India using ZN staining indicated Cryptosporidium in 8.5% from Manipal (Ballal et al. 1999), 4.6% from Mumbai (Cotte et al. 1993), 10% from Chandigarh (Mohandas et al. 2002), and 17% in acute and 11.86% chronic diarrhea from Chennai (Kumar et al. 2002a, b). Ramakrishnan et al. (2007) reported prevalence of C. parvum infection 28.7% in HIV/AIDS patients with diarrhea in Madurai, South India and Guk et al. (2005) reported C. parvum infections of 10.5% in HIV-infected patients who visited Seoul National University Hospital, Korea.

In our study, the number of Cryptosporidium positive patients by AFS technique was 28.7% and 4.35% by PCR in HIV patients with or without diarrhea, respectively. The number of Cryptosporidium positive patients with diarrhea was not significantly different from the number of Cryptosporidium positive patients without diarrhea (p > 0.05). This result corresponds to reports of other authors according to which asymptomatic cryptosporidiosis is increasing (Houpt et al. 2005; Kaushik et al. 2008). This is illustrated by the report from Tanzania indicating cryptosporidiosis in 16.7% in asymptomatic as compared to 18% in symptomatic HIV-infected patients (Houpt et al. 2005). There, 14.9% asymptomatic HIV seropositive patients without diarrhea were positive for Cryptosporidium infection (Kaushik et al. 2008), and from Bolivia (Esteban et al. 1998) and Korea (Yu et al. 2004), cryptosporidiosis was found in 8–32% of healthy individuals. Cryptosporidium was reported in significantly higher number (22.6%) of HIV seropositive patients with diarrhea as compared to 0.5% in patients without diarrhea by ZN staining from Vellore, India (Mukhopadhya et al. 1999). Cryptosporidium positivity was 9.8% and 6% in HIV seropositive patients with and without diarrhea, respectively, by using safranine methylene blue staining, whereas in a recent study from Vellore (Muthusamy et al. 2006), there were 4.6% in diarrheic cases and 1.2% in non-diarrhea cases in children from Kolkata, India (Das et al. 2006). Thus, the studies in general indicate higher number of Cryptosporidium positive findings in patients with diarrhea as compared to those without diarrhea.

In the samples examined here, Cryptosporidium sp. infections in cattle were 13% by AFS. This is higher than the prevalence (0.6%) of Cryptosporidium found among dairy cows in the Nong Pho region of central Thailand by AFS reported by Jittapalapong et al. (2006) and lower than the prevalence of Cryptosporidium in Tanzania and Zambia (Swai et al. 2007; Geurden et al. 2006). However, Cryptosporidium infection correlated with host age, which is an important factor characterizing the distribution of Cryptosporidium infections and suggesting a degree of resistance conferred to individuals infected early in life (Arslan and Vicki 1999; Castro-Hermida et al. 2002; Tanriverdi et al. 2006; Kvac et al. 2006; Lefay et al. 2000; Geurden et al. 2006; Azami 2007; Uga et al. 2000; Nguyen et al. 2007).

In recent years, molecular techniques for species genotyping identification have been developed and evaluated (Abe et al. 2006; Nichols et al. 2003, 2006). It is advisable to use both methods (microscopy and PCR) in combination. According to other authors, C. parvum is the predominant species infecting humans in Thailand (Tiangtip and Jongwutiwes 2002; Gatei et al. 2002). Xiao et al. (2001) reported five types of Cryptosporidium in Peruvian children. However, in some countries, C. parvum is slightly more commonly identified than C. hominis in both immunocompetent and immunocompromised individuals (Caccio et al. 2000; Ong et al. 2002; Meamar et al. 2007; Cama et al. 2007; Llorente et al. 2007), while other authors reporting Cryptosporidium isolates obtained from HIV-infected patients have identified at least five species: C. hominis, C. parvum, C. felis, C. meleagridis, and C. muris (Xiao et al. 2004).

By sequencing DNA of Cryptosporidium from cattle samples, we identified the genotype of the Cryptosporidium from five samples, all of which were C. parvum. Other Cryptosporidium species have been reported to infect cattle including C. bovis, C. andersoni, and the Cryptosporidium deer-like genotype (Santín et al. 2004; Fayer et al. 2006; Abdul Halim et al. 2008), but none of these were found here in our few PCR positive samples. The most common zoonotic species is C. parvum and it is also the most common species infecting cattle. It has been reported elsewhere as the most prevalent in pre-weaned calves, infecting 35% of all pre-weaned calves examined but diminishing in post-weaned calves, heifers, and mature cows to 0.2%, 0.7%, and 0.4%, respectively, of all animals examined (Fayer et al. 2006). In Vietnam, the prevalence of C. parvum type (bovine genotype) infections was 33.5%, those of the C. andersoni type were 5.6%, and mixed infection of both types was 3.4% (Nguyen et al. 2007). Both C. bovis and the deer-like genotype found in cattle are host specific for these animals and of no known health risk to humans or other animals. One heifer infected with C. suis, a rare infection in humans, has been reported (Fayer et al. 2006). C. andersoni has been found in post-weaned calves, heifers, and cows. No C. andersoni oocysts have been reported in cattle less than 3 months old, supporting other reports that chronic C. andersoni infection usually occurs in adult cattle with no clinical symptoms.

The results described here suggest that C. parvum is the most common species of Cryptosporidium in humans and animals in Thailand. The HIV positive patients attending Prabat Nampu Temple come from several almost exclusively rural and agricultural provinces. Cattle in Chon Buri Province appear typical of those in similar rural agricultural provinces. They may serve as the reservoir of C. parvum for transmission to humans through food and water contamination or from livestock directly. HIV positive patients may have been infected by Cryptosporidium oocysts from direct contact with cattle or other environmental sources.

The results represent the first genetic identification of Cryptosporidium species in cattle in Thailand and in cattle in Chon Buri Province. Finding Cryptosporidium spp. present in domestic animal populations is not surprising based on its virtually ubiquitous distribution according to the literature. The presence of Cryptosporidium infections in the human population in this rural area of Thailand is also not surprising and easily accounted for by direct contact with other infected hosts, either human or animal, or through other environmental sources including food and water. Further, the results described here suggest that dairy farms should be considered as a potential source of water contamination and C. parvum is an important zoonotic species in dairy calf population in Thailand.