- Top of page
Twenty-nine faecal specimens from Slovenian patients in which Cryptosporidium oocysts had been identified were studied. A fragment of the Cryptosporidium 18S rRNA gene and a fragment of the Cryptosporidium COWP gene were amplified by PCR and sequenced. Cryptosporidium parvum was identified in 26 of the 29 specimens, Cryptosporidium hominis in two, and Cryptosporidium cervine genotype in one. The fact that C. parvum, which is associated traditionally with animals, was identified in the majority of human faecal specimens suggests that cryptosporidiosis may have primarily a zoonotic origin in Slovenia.
The use of molecular methods in determining the taxonomy of Cryptosporidium spp. has led to increased recognition of the diversity of species infecting humans . Human cryptosporidiosis is caused mainly by Cryptosporidium hominis, which is found almost exclusively in humans, and Cryptosporidium parvum, which is found in most livestock, some wild animals and humans [2,3]. The occurrence of both of these species in humans indicates that anthroponotic and zoonotic transmission cycles can occur in human infections . In addition to C. hominis and C. parvum, humans are also known to be infected by Cryptosporidium meleagridis, Cryptosporidium muris, Cryptosporidium felis, Cryptosporidium canis, Cryptosporidium suis and the cervine genotype, which are associated traditionally with animals . The prevalence and significance of the different species and genotypes in humans are not yet clear. Moreover, potential reservoir hosts and transmission pathways for novel species infecting humans have not yet been elucidated. Genotyping of isolates from different parts of the world is therefore essential for a more precise understanding of the epidemiology of Cryptosporidium spp. [1,5].
In Slovenia, only five Cryptosporidium isolates from human patients have been typed to date, all five of which were C. parvum[6,7]. In the present study, isolates from 29 faecal specimens obtained from sporadic cases of cryptosporidiosis, collected at the Institute of Microbiology and Immunology, Ljubljana, Slovenia, between 2000 and 2003 were genotyped. The specimens were obtained from 29 immunocompetent patients who attended health centres and hospitals in various parts of Slovenia because of clinical symptoms consistent with cryptosporidiosis. Five of these patients were hospitalised because of cryptosporidiosis (Table 1); none of the infections was hospital-acquired.
Table 1. Isolate genotypes and clinical and epidemiological data for patients with Cryptosporidium infection in Slovenia
|Isolate code||Year of collection||Patient age (years)||Gender||Type of region||Symptoms||Hospitalisation||Species, genotype (18S)||Species, genotype (COWP)|
|SI 1||2002||32||F||Urban||Diarrhoea||No||C. parvum (A)||C. parvum|
|SI 2||2002||12||F||Rural||NA||Yes||C. parvum (A)||C. parvum|
|SI 3||2003|| 8||F||Rural||NA||No||C. parvum (A)||C. parvum|
|SI 4||2002||31||M||Rural||NA||No||C. parvum (A)||C. parvum|
|SI 5||2003|| 3||M||Rural||Diarrhoea||No||C. parvum (A)||C. parvum|
|SI 6||2002||23||F||Rural||Enterocolitis||No||C. parvum (A)||–a|
|SI 7||2002||23||F||Rural||Diarrhoea||No||C. parvum (A)||C. parvum|
|SI 8||2002|| 1||M||Rural||NA||Yes||C. parvum (A)||C. parvum|
|SI 9||2002||23||M||Urban||NA||No||C. parvum (A)||C. parvum|
|SI 10||2002||18||F||Urban||NA||Yes||C. parvum (A)||C. parvum|
|SI 11||2002||28||F||Urban||Gastroenterocolitis||No||C. parvum (A)||C. parvum|
|SI 12||2002|| 6||F||Urban||NA||Yes||C. parvum (A)||C. parvum|
|SI 13||2000||18||F||Rural||Enterocolitis||Yes||C. parvum (A)||C. parvum|
|SI 14||2000||NA, child||M||Rural||NA||No||C. hominis||C. hominis|
|SI 15||2000|| 9||F||Urban||Enterocolitis||No||C. parvum (A)||C. parvum|
|SI 16||2001|| 8||F||Rural||NA||No||C. parvum (A)||C. parvum|
|SI 17||2001|| 8||F||Rural||NA||No||C. parvum (A)||C. parvum|
|SI 18||2001||11||F||Urban||Gastroenterocolitis||No||C. parvum (B)||– a|
|SI 19||2001||29||F||Urban||NA||No||C. hominis||C. hominis|
|SI 20||2001|| 1||M||Rural||NA||No||C. parvum (A)||C. parvum|
|SI 21||2002||11||M||Urban||Abdominal pain||No||C. parvum (A)||C. parvum|
|SI 22||2002|| 1||M||Rural||NA||No||C. parvum (A)||C. parvum|
|SI 23||2002|| 1||F||Rural||Diarrhoea||No||Cervine||Cervine|
|SI 24||2002|| 3||M||Rural||Enterocolitis||No||C. parvum (A)||C. parvum|
|SI 25||2002|| 6||M||Rural||NA||No||C. parvum (A)||C. parvum|
|SI 26||2002|| 6||M||Rural||Enterocolitis||No||C. parvum (A)||C. parvum|
|SI 27||2002|| 4||M||Rural||Gastroenterocolitis||No||C. parvum (A)||C. parvum|
|SI 28||2003|| 2||M||Rural||Enterocolitis||No||C. parvum (A)||C. parvum|
|SI 29||2003||52||F||Urban||NA||No||C. parvum (A)||C. parvum|
Cryptosporidium oocysts were identified microscopically in faecal smears after staining with modified Ziehl–Neelsen stain, and by use of a direct immunofluorescence test (MeriFluor; Meridian Bioscience, Cincinnati, OH, USA). DNA was extracted from faecal specimens with the QIAamp DNA stool mini kit (Qiagen, Hilden, Germany). A c. 830-bp fragment of the Cryptosporidium 18S rRNA gene that spanned the hyper-variable region and a 553-bp fragment of the Cryptosporidium COWP gene were amplified by nested PCR and PCR, respectively, as described previously [8,9]. PCR products were sequenced in both directions on an ABI Prism 310 Genetic Analyser (Applied Biosystems, Foster City, CA, USA). Overlapping bidirectional sequences were assembled using SeqMan sequence analysis software (DNASTAR Inc., Madison, WI, USA) and were subjected to a BLAST search to determine their identities and to assess their similarities to sequences in GenBank. The sequences were aligned using the ClustalV program. A neighbour-joining tree was constructed from the 18S rRNA gene fragment information by using the TreeconW program, and evolutionary distances were calculated by Kimura two-parameter analysis. The 18S rRNA and COWP gene sequences of five representative patient isolates (Fig. 1) have been deposited in the European Molecular Biology Laboratory (EMBL) database, under accession numbers AJ849457–AJ849465.
Figure 1. Phylogenetic relationship of Cryptosporidium sequences inferred from neighbour-joining analysis of the partial 18S rRNA gene. The Cryptosporidium muris (AF093498), Cryptosporidium andersoni (L19069) and Cryptosporidium serpentis (AF093502) sequences were used as an outgroup, and the tree was rooted with this outgroup. Values on branches are percentage bootstrap values using 1000 replicates. Bootstrap values > 50% are shown. Bar = 0.02 substitutions per site. Numbers in parentheses following species or genotypes are GenBank accession numbers. Sequences obtained from Slovenian isolates are marked in bold type.
Download figure to PowerPoint
All 29 specimens gave the expected c. 830-bp amplicon for the 18S rRNA gene. The comparison of the 18S rRNA gene sequences with published reference sequences by multiple sequence alignment and phylogenetic analysis (Fig. 1) showed that the 29 isolates fell into three main groups. The first group comprised 26 C. parvum isolates, SI 1–13, 15–18, 20–22 and 24–29 (Fig. 1). Sequence analysis showed that these isolates were of two different subunit types; 25 isolates (SI 1–13, 15–17, 20–22 and 24–29), which were identical to one another, had C. parvum type A subunit sequences, while the remaining isolate (SI 18) had a C. parvum type B subunit sequence. The second group comprised C. hominis isolates SI 14 and SI 19 (Fig. 1). The third group was represented by a single isolate (SI 23) whose sequence was identical to the published sequence of the Cryptosporidium cervine genotype identified in lemurs (AF442484)  (Fig. 1), which has only been reported once previously in humans ; however, this organism could emerge as an important human pathogen following increasing contact between humans and wildlife .
PCR amplification of the COWP gene fragment was successful for 27 of the 29 isolates, yielding amplicons of 553 bp. Following sequencing, 24 amplicons (SI 1–5, 7–13, 15–17, 20–22 and 24–29) proved to be C. parvum sequences, two (SI 14, SI 19) were C. hominis sequences, and one (SI 23) was a Cryptosporidium cervine genotype sequence. This sequence was identical to that of the isolate from lemurs described by da Silva et al. .
In developed countries, most cases of cryptosporidiosis occur in children aged 1–4 years, perhaps because of increased exposure as they explore their environment . However, in the present study, there was a slightly greater proportion of cases in the group aged 5–14 years than in the group aged 1–4 years. Moreover, there were more cases involving children aged ≤ 14 years than there were adult cases. However, in Slovenia, while cryptosporidiosis is mainly a disease of pre-school and school-aged children, it is also a disease of adults.
Genetic characterisation of the Cryptosporidium isolates revealed that in Slovenia, as in other European countries, C. parvum is generally found in humans more frequently than C. hominis (90% of cases vs. 7% of cases in the present study). The fact that Cryptosporidium spp. associated traditionally with animals were identified in most human faecal specimens (26 cases involving C. parvum and one case involving Cryptosporidium cervine genotype, compared with two cases involving C. hominis) suggests that cryptosporidiosis may have primarily a zoonotic origin in Slovenia. This is not surprising, as most parts of Slovenia are predominantly rural, and even large cities have rural suburbs where individuals can come into contact with animals. However, the significance of finding parasites associated traditionally with animals in humans should be interpreted with caution. For example, many human cases of C. parvum infection may originate from humans . Indeed, results of subtyping studies have shown the presence of human-adapted C. parvum subtypes, even in areas with extensive transmission of C. parvum between humans and farm animals [14–16]. Thus, anthroponotic transmission of C. parvum and also other Cryptosporidium spp. associated traditionally with animals is probably not unusual .