Cryptosporidium rubeyi, Li & Pereira & Larsen & Xiao & Phillips & Striby & McCowan & Atwill, 2015

Li, Xunde, Pereira, Maria das Graças Cabral, Larsen, Royce, Xiao, Chengling, Phillips, Ralph, Striby, Karl, McCowan, Brenda & Atwill, Edward R., 2015, Cryptosporidium rubeyi n. sp. (Apicomplexa: Cryptosporidiidae) in multiple Spermophilus ground squirrel species, International Journal for Parasitology: Parasites and Wildlife 4 (3), pp. 343-350 : 346-349

3.2. Oocyst infectivity of Cryptosporidium sp. c-genotype

Using C. parvum from dairy calves as a positive control, 83% (5/6) of neonatal BALB/c mice were infected after inoculation of 100 oocysts and 100% (19/19 and 17/17) after inoculation of 5000 and 10,000 oocysts, respectively ( Table 2). In contrast, oocysts of all isolates of Cryptosporidium sp. c-genotype from the three ground squirrel species failed to produce detectable levels of infection in mice. Cryptosporidium oocyst infectivity in mice varies with species and genotype, inoculum size, mouse species and strain, age, and susceptibility ( Finch et al., 1993; Neumann et al., 2000; Hou et al., 2004). It is well established that neonatal BALB/c mice are susceptible to C. parvum infection ( Fayer, 1995; Slifko et al., 2002; Jenkins et al., 2003; Guk et al., 2004). Tarazona et al. (1998) reported that inoculation of 10 4 or more C. parvum oocysts results in 100% infection in BALB/c mice. We determined previously that the 50% infective dose (ID 50) for C. parvum in neonatal BALB/c mice was 70.6 oocysts ( Li et al., 2005) and mice inoculated with 1000 oocysts resulted in 100% infection ( Li et al., 2010). Previously we have shown that inoculation up to 10 4 Sbey03c oocysts failed to infect neonatal BALB/c mice ( Atwill et al., 2004) and the current results confirm this, indicating that Cryptosporidium sp. c-genotype oocysts from Spermophilus ground squirrels are not infectious to neonatal BALB/c mice-and also exhibit some degree of host specificity. In a similar study, inoculation of 10 3 Cryptosporidium oocysts from red squirrels ( Sciurus vulgaris ) failed to generate detectable infection in neonatal and adult CD-1 and BABL/c mice (Kv́ac et al., 2008).

Intestinal homogenates coupled with fluorescent microscopy for determining C. parvum infection in neonatal mice has been shown to be significantly more sensitive than histopathology ( Hou et al., 2004). In the present work, no oocysts were detected from mice inoculated with heat inactivated C. parvum oocysts, which confirmed that oocysts detected in positive control mice were not from direct inoculation and subsequent pass through but instead from patent intestinal infections. No clinical signs of cryptosporidiosis were observed in C. parvum infected mice which is not unusual given that asymptomatic cryptosporidial infections in mice have been documented previously by other investigators ( Tarazona et al., 1998; Kv́ac et al., 2008). Prepatent periods of Cryptosporidium infection in mice vary with species and doses of oocysts, species, age, and susceptibility of mice, with younger mice generally more susceptible ( Youssef et al., 1992; Tarazona et al., 1998; Matsui et al., 1999; Rhee et al., 1999; Yang et al., 2000). In the present work most mice were euthanized at day 7 PI for detection of oocysts, which was appropriate for the detection of C. parvum infection in mice in the present and previous work ( Hou et al., 2004). To explore the possibility of a longer prepatent period for Cryptosporidium infection in mice from inoculation of Spermophilus ground squirrel oocysts, we postponed euthanasia to day 10 PI in some mice inoculated with oocysts from S. lateralis (isolates 113, 128, 155, and 230). Despite this longer period, no oocysts were detected in this cohort of mice. This suggests that the failure to detect Cryptosporidium infection in neonatal BALB/c mice was due to host specificity of Spermophilus -derived Cryptosporidium rather than the length of the prepatent period.

Although only two calves were inoculated for each of the 8 isolates, we did not find evidence of infection in calves from inoculation with up to 5000 oocysts of the c-genotype from eight S. lateralis ground squirrels; rather, calves in all groups including the negative control group (without oocyst inoculation) eventually became infected with Cryptosporidium oocysts that were confirmed to be 100% identical to the C. parvum via sequencing the 18S rRNA gene. This genotype of oocyst was the same as found in our positive control calf whereby the oocysts were collected from a local dairy in the same region where the calves were purchased (data not shown). Given that Cryptosporidium remain genetically stable after passing through mammalian species ( Akiyoshi et al., 2002), these calfhood infections with C. parvum might be due to natural infection before inoculation or cross contamination from, for example, filth flies from nearby commercial dairies and/or from our positive control calves. Our calf pens were in an outdoor open facility which can allow filth flies to circulate between positive control and other calves. Our results of BALB/c mice and calf infectivity studies suggest there exists host specificity for this specific c-genotype Cryptosporidium shed by Spermophilus ground squirrels.

3.3. Multiple gene analysis of Cryptosporidium sp. c-genotype isolates from S. beecheyi

We previously reported DNA fingerprinting of Cryptosporidium isolates from Spermophilus ground squirrels collected throughout California, USA (longitude of 114 ǫ 8 Ɩ W to 124 ǫ 24 Ɩ W and latitude of 32 ǫ 30 Ɩ N to 42 ǫ N) ( Pereira et al., 2010). In this present work additional fingerprinting using 18S rRNA, actin, and HSP70 genes was conducted on new Cryptosporidium isolates from S. beecheyi collected in 2011 from the Central Coastal region of California (e.g., latitude of 35 ǫ 16 ƖN and longitude: 120 ǫ 39 ƖW) to confirm our earlier findings of a new species of Cryptosporidium in this host species. Among the 100 S. beecheyi squirrel fecal samples (each from a different squirrel), 18, 14, and 3 fecal samples with oocysts were successfully sequenced for the c-genotype by using the 18S rRNA, actin, and HSP70 gene, respectively. Using our previous nomenclature based on host species, year of isolation, and genotype, in this manuscript we describe the c-genotype collected in 2011 as Sbey11c (host S. beecheyi , 2011 isolation, genotype-c). According to electrophoresis and DNA sequencing results, no positive squirrels were found to be shedding more than one genotype at a time. The GenBank accession numbers of representative c-genotype sequences are KM010224 of the 18S rRNA gene, KM010227 of the actin gene, and KM010229 of the HSP70 gene, respectively. Given the small amount of fecal sample obtained using trap and release procedures for squirrels, it can be difficult to have sufficient oocysts to successfully complete PCR and multiple gene sequencing from a single isolate from this host species, but one isolate of c-genotype was successfully sequenced for all three genes. Phylogenetic trees based on DNA sequences representing the c-genotype of the three genes were constructed and juxtaposed against reference sequences of Cryptosporidium species /genotypes selected as mentioned above ( Figs. 2 ‾ 4).

BLAST results (as of March 12, 2015) of DNA sequences of the three genes are shown in Table 3. With respect to the actin gene, the Sbey11c (KM010227) was not 100% identical to any Cryptosporidium sequence in the GenBank, with maximal similarity of only ~93% to a Cryptosporidium sp. chipmunk genotype I (JX978270). Phylogenetic analysis of the actin gene sequences revealed similar results as the BLAST analysis in that Sbey11c did not form a distinct clade with any existing Cryptosporidium sequence in GenBank ( Fig. 3 ). For the HSP70 gene, maximal similarity of Sbey11c (KM010229) to currently available sequences was at best only ~92% similar to two isolates of Cryptosporidium sp. chipmunk genotype I (JX978275, JX978276). Similarly, the phylogenetic analysis of the HSP70 gene shows that Sbey11c did not form a distinct clade with any existing Cryptosporidium sequences ( Fig. 4 ).

For the 18S rRNA gene, BLAST results show that the Sbey11c (KM010224) was 100% identical to Sbey05c (DQ295012) and Sbey03c (AY462233), 99.64% similar to Sltl05c (DQ295014), and 98.67% similar to Sbld05c (DQ295013) ( Table 3). Sbey05c and Sbey03c represent for the most common Cryptosporidium c-genotype from S. beecheyi squirrels collected in 2005 and 2003; Sltl05c represents the typical Cryptosporidium sp. c-genotype from S. lateralis squirrels collected in 2005; Sbld05c represents the typical Cryptosporidium sp. c-genotype from S. beldingi squirrels collected in 2005, as previously reported ( Pereira et al., 2010). It is interesting that additional fingerprinting of new isolates collected in 2011consistently confirm the presence of Sbey-c genotype Cryptosporidium in S. beecheyi . Phylogenetic analysis of the 18S rRNA gene sequences revealed similar results as the BLAST analysis. The Sbey11c formed a distinct clade with Cryptosporidium isolated from all three host species ( S. beecheyi , S. lateralis , S. beldingi ) (Sbey05c, Sbey03c, Sltl05c, Sbld05c) compared to existing Cryptosporidium sequences ( Fig. 2 ). In particular, BLAST and phylogenetic analyses of 18S rRNA, actin, and HSP70 genes sequences demonstrated that S. beecheyi are a mammalian host of the Sbey11c genotype, and that this unique genotype is also present in other species of the genus Spermophilus from throughout California, USA.

Focusing on the 18S rRNA gene that is commonly used for Cryptosporidium speciation ( Checkley et al., 2015) and has the most sequences of described species and genotypes in the GenBank, our work over a decade ( Atwill et al., 2001, 2004; Pereira et al., 2010 current work) has demonstrated that Cryptosporidium sp. c-genotype is the most prevalent genotype in Spermophilus ground squirrels. Combining the evidence of the presence of novel Cryptosporidium species in Spermophilus ground squirrels from our previous ( Atwill et al., 2004; Pereira et al., 2010) and current work, we propose to name the Cryptosporidium sp. c-genotype in Spermophilus ground squirrels as Cryptosporidium rubeyi n. sp. GenBank accession numbers of DNA sequences for C. rubeyi n. sp. are DQ295012, AY462233, and KM010224 for the 18S rRNA gene, KM010227 for the actin gene, and KM010229 for the HSP70 gene.

Figures

Fig. 2. Phylogenetic relationships of partial 18S rRNA gene sequences of Cryptosporidium sp. Sbey11c from California ground squirrels (S. beecheyi) and other Cryptosporidium spp. inferred by neighbor-joining analysis with 1000 bootstrapping replicates.

Fig. 3. Phylogenetic relationships of partial actin gene sequences of Cryptosporidium sp. Sbey11c from California ground squirrels (S. beecheyi) and other Cryptosporidium spp. inferred by neighbor-joining analysis with 1000 bootstrapping replicates.

Fig. 4. Phylogenetic relationships of partial HSP70 gene sequences of Cryptosporidium sp. Sbey11c from California ground squirrels (S. beecheyi) and other Cryptosporidium spp. inferred by neighbor-joining analysis with 1000 bootstrapping replicates.