INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Cryptosporidium parvum is now recognized as an important cause of diarrheal infections in animals and humans (4). Wild rodents are thought to provide an important reservoir of infection of C. parvum for farm animals because the oocysts are environmentally resistant and C. parvum has been detected in wild brown rats (Rattus norvegicus), wild house mice (Mus domesticus), wild wood mice (Apodemus sylvaticus), and wild bank voles (Clethrionomys glareolus) (2, 20). Recent research that has genetically characterized isolates of C. parvum from wild Australian mice (Mus domesticus) has revealed that mice carry a distinct genotype (10, 11). This "mouse" genotype has smaller oocysts than C. parvum (4.5 by 4.0 µm) and is genetically different from genotypes carried by cattle and humans (10, 11). Until recently, it was assumed that C. parvum was a uniform species, but there is now strong evidence that C. parvum is composed of numerous distinct genotypes: a "human" genotype found only in humans, a "cattle" genotype found in many domestic animals and also humans, and a number of other genotypes, some of which appear to be host specific (1, 7-17, 19). The aim of this study was to genetically characterize Cryptosporidium isolates from mice from diverse locations in order to determine if the "mouse" genotype is conserved in mouse-derived Cryptosporidium isolates from different geographical areas.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Sources of parasite isolates, DNA purification, and PCR. The sources of the parasite isolates are listed in Table 1, and DNA was purified as described previously (8). Primers and PCR conditions were as described previously (8).

Sequencing. PCR products were sequenced with an ABI Prism Dye Terminator Cycle Sequencing kit (Applied Biosystems, Foster City, Calif.) according to the manufacturer's instructions, except that the annealing temperature was raised to 60°C. Sequences were analyzed with SeqEd, version 1.0.3. (Applied Biosystems), and were aligned with the Clustal V (6) sequence alignment program.

Phylogenetic analysis of sequence information. A phylogenetic analysis based on the nucleotide sequences of 18S rRNA gene (rDNA) and acetyl coenzyme A (acetyl-CoA) synthetase gene regions from different isolates was conducted by using PHYLIP 3.5p (5). A similarity index among Cryptosporidium isolates was created by using the formula for Kimura's distance. Phylograms were constructed from genetic distance matrices by the unweighted pair group method of analysis and with the DRAWGRAM programs available in PHYLIP 3.5p (5).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Sequence analysis of 18S rDNA. Sequence analysis of the 298-bp 18S rDNA product (data not shown) revealed the "mouse" genotype (10, 11) to be highly conserved between three mouse-derived Cryptosporidium isolates from Australia sequenced previously (10) (GenBank accession no. AF099667), six mouse-derived Cryptosporidium isolates from the United Kingdom, two mouse-derived Cryptosporidium isolates from Spain, and a bat isolate from New South Wales in Australia. Five mouse isolates from Australia (isolates M4c, M6c, M8c, M23c, and M27c) (GenBank accession no. AF099668) exhibited the "cattle" genotype and one mouse isolate from Spain (isolate SM4) was identical to Cryptosporidium muris (GenBank accession no. L19069). Cattle isolates from the United Kingdom and Australia were all identical and displayed the "cattle" genotype. Sequence analysis of a wildebeest-derived Cryptosporidium isolate from Spain (isolate WB1) revealed that it was of the "cattle" genotype.

Sequence analysis of the acetyl-CoA synthethase gene. As with the rDNA sequencing results, sequence analysis of the acetyl-CoA synthetase gene revealed distinct differences between cattle and mouse isolates. All cattle isolates analyzed exhibited a common genotype. Mouse isolates from the United Kingdom were identical to five mouse isolates from Australia sequenced previously (10) and to two mouse isolates from Spain (isolates SM1 and SM2) (GenBank accession no. AF102768). Isolate SM4 was not amplified with the acetyl-CoA primers. Australian mouse isolates M4c, M6c, M8c, M23c, and M27c exhibited the "cattle" genotype, as did the wildebeest isolate (isolate WB1) from Spain (GenBank accession no. AF102767). It was not possible to amplify the bat isolate with the acetyl-CoA synthetase gene primers due to the low amount of DNA present.

Phylogenetic analysis of rDNA sequencing results. Additional isolates previously sequenced were also analyzed and included a human isolate (H1) and a cattle isolate (C1) (8), and additional Cryptosporidium isolates retrieved from the rRNA WWW Server on the World Wide Web (18). This extended phylogenetic analysis resulted in three distinct groups within C. parvum (Fig. 1): "human" group; a "cattle" group, which contained the cattle isolates, five Australian mouse isolates, and a wildebeest isolate (isolate WB1); and a "mouse" group, which contained all the Australian mouse isolates, all the mouse isolates from the United Kingdom, two Spanish mouse isolates, and the bat isolate from New South Wales. The remaining Spanish mouse isolate (isolate SM4) grouped with C. muris.

View larger version (16K): [in this window] [in a new window]   FIG. 1.   Phylogram of Kimura's distance generated from 18S rDNA sequence information among isolates of Cryptosporidium clustered by the unweighted pair group method of analysis.

Phylogenetic analysis of acetyl-CoA synthetase gene sequencing results. Phylogenetic analysis of the acetyl-CoA synthetase gene sequence information also produced three main groups (Fig. 2): A "human" group; a "cattle" group, which contained the cattle isolates, five Australian mouse isolates, and a wildebeest isolate (isolate WB1); and a "mouse" group, which contained the Australian mice isolates, all the mouse isolates from the United Kingdom, and two Spanish mouse isolates (isolates SM1 and SM2).

View larger version (11K): [in this window] [in a new window]   FIG. 2.   Phylogram of Kimura's distance generated from acetyl-CoA synthethase gene sequence information among isolates of Cryptosporidium clustered by the unweighted pair group method of analysis.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Mice (M. musculus syn. domesticus) from different geographical areas were shown to carry a distinct genotype of C. parvum (referred to as the "mouse" genotype) by both rDNA and acetyl-CoA synthetase gene sequence analyses, indicating that this genotype is conserved across widely separated geographical areas.

The "mouse" genotype was also identified in a fecal sample from a large-footed mouse-eared bat (Myotus adversus), extending the host range of this genotype. Cryptosporidium has recently been reported in a big brown bat (Eptesicus fuscus) (3). In this report, cryptosporidial bodies (<5 µm) were found to be attached to the microvillar border of enterocytes of paraffin-embedded sections of the small intestine. This bat isolate was not genotyped, but the small sizes of the oocysts indicate that a C. parvum-like isolate was present. Future genotyping studies should examine bat isolates from different geographical locations in order to determine how common this genotype is among bats. The present report of Cryptosporidium in a large-footed bat is the only other report of a cryptosporidial infection in bats.

Cattle isolates from the United Kingdom all exhibited the "cattle" genotype by both rDNA and acetyl-CoA synthetase gene sequence analyses and were identical to an Australian cattle isolate (isolate C1), confirming the conserved and widespread nature of this genotype. The "cattle" genotype was also identified in an adult male wildebeest (Connochaetes taurinus taurinus). This is the first time that this genotype has been identified in this host.

Interestingly, five of the mouse isolates analyzed exhibited the "cattle" genotype, which is known to infect humans. These mice were trapped on farms in Victoria, Australia, where large numbers of sheep were grazing. Under the circumstances, sheep are the most likely source of infection for mice since sheep are known to carry isolates of the "cattle" genotype (10). A recent 2-year survey of wild mice and voles on a farm in Warwickshire, United Kingdom, reported prevalence rates of 22, 21, and 13% for C. parvum in M. musculus, A. sylvaticus, and C. glareolus, respectively (2). The apparent autumnal peak for C. parvum in all three rodent species coincided with the calving period at that farm, and it was concluded that "rodents may represent a significant reservoir of Cryptosporidium with a high potential for infection of man and livestock due to cohabitation" (2). The finding of the "cattle" genotype in Australian mice indicates that sheep and cattle may transmit the "cattle" genotype to mice, which may in turn transmit Cryptosporidium to other domestic animals and also to humans. However, the "mouse" genotype appears to be more common in mice, and as small rodent populations and Cryptosporidium prevalence are highest at the end of the summer, independent of the presence of cattle and sheep (2), it may be that mice are only occasionally infected with the "cattle" genotype during periods of heavy environmental contamination. Recent studies with pigs have shown that they are also capable of carrying two distinct genotypes: a "pig" genotype common to pigs from different geographical areas and the "cattle" genotype (10, 11, 13).

More extensive characterization of rodent isolates of Cryptosporidium from a wider geographical distribution and from both urban and rural habitats is necessary before their role as reservoirs of infection in humans and domestic animals can be more fully determined.