Introduction

Cryptosporidium spp. and Enterocytozoon bieneusi, causative agents of cryptosporidiosis and microsporidiosis, are two important opportunistic intestinal pathogens that can infect vertebrate and invertebrate, posing a significant threat to public health1. Humans are infected with these pathogens mainly via the fecal-oral route with anthroponotic and zoonotic transmission or via food-borne and water-borne transmission2,3. Clinical manifestations of infection with these pathogens are often inconsistent due to the variabilities in the health condition of infected hosts4,5. In healthy individuals, these pathogens usually cause asymptomatic infection or self-limiting diarrhea6. However, infection may also result in chronic or life-threatening diarrhea in immunocompromised individuals, such as patients with acquired immunodeficiency syndrome and patients who had undergone organ transplantation7. In addition to humans, there are a variety of animals that can act as hosts for these two pathogens, including various mammals, reptiles, birds, amphibians and insects8. Therefore, Cryptosporidium spp. and E. bieneusi have been recognized as category B pathogens by the National Institutes of Health due to their ease of transmission in spite of low mortality9.

To detect and evaluate potential zoonotic transmissions, it is necessary to accurately distinguish Cryptosporidium spp. and E. bieneusi on the molecular level10. To date, at least 37 species and over 70 genotypes of Cryptosporidium spp. have been described11. Among them, 11 Cryptosporidium species have been identified in rodents, Cryptosporidium parvum and C. muris are the most common12. For E. bieneusi, more than 474 genotypes have been identified based on the internal transcribed spacer (ITS) region of the rRNA gene13, and more than 35 genotypes have been determined in rodents14. These genotypes can be classified into eleven groups (groups 1–11) by phylogenetic analyses13. Group 1 comprises the majority of zoonotic potential genotypes, whereas the remaining (groups 2–11) are considered as the host-adapted groups, which are mostly found in specific hosts or water13.

In China, Cryptosporidium spp. and E. bieneusi have been detected in a wide range of hosts, including carnivores, lagomorphs, primates, birds, and rodents14,15. Pet rodents, in particular (e.g. chinchillas, red-bellied tree squirrels, guinea pigs, and chipmunks), are considered potential sources of Cryptosporidium spp. and E. bieneusi infections in humans16,17,18,19. The red squirrel (Sciurus vulgaris) is a popular pet in China, which is widely bred in pet shops and homes for its appearance and mild-mannered nature. However, there is no published data regarding the prevalence of Cryptosporidium spp. and E. bieneusi in pet red squirrels, and the role of the red squirrels in the transmission of the two pathogens remains poorly investigated. Thus, we examined the occurrence of Cryptosporidium spp. and E. bieneusi in red squirrels, and evaluated their potential role in the zoonotic transmission of human cryptosporidiosis and microsporidiosis.

Results

Occurrence of Cryptosporidium spp. and E. bieneusi

The overall prevalence of Cryptosporidium spp. in pet red squirrels was 8.6% (27/314, 95% CI: 5.5–11.7%). All pet shops were positive for Cryptosporidium, and the prevalence ranged from 2.2% to 16.2%; significant differences were observed (χ2 = 0.028, df = 4, P < 0.05; Table 1). The prevalence of Cryptosporidium spp. among males and females were 8.1% and 9%, respectively, but the difference was not statistically significant (χ2 = 0.086, df = 1, P > 0.05). The differences in prevalence of Cryptosporidium spp. among squirrels of different ages were not statistically significant (χ2 = 0.093, df = 1, P > 0.05) (Table 2). Moreover, a significant correlation between the different sources and Cryptosporidium spp. infection (P = 0.01) was observed by logistic regression analysis.

Table 1 Occurrence of Cryptosporidium spp. and Enterocytozoon bieneusi in pet red squirrels from different sources in Southwestern China.
Table 2 Occurrence of Cryptosporidium spp. and Enterocytozoon bieneusi in pet red squirrels according to sex and age.

The overall prevalence of E. bieneusi in pet red squirrels was 19.4% (61/314, 95% CI: 15–23.8%). E. bieneusi was found in all four pet shops investigated, with infection rates ranging between 8.6% and 27.6%. The difference was statistically significant (χ2 = 0.036, df = 4, P < 0.05; Table 1). The prevalence of E. bieneusi in female red squirrels (20.5%) was higher than male (18.2%), but the difference was not statistically significant (χ2 = 0.251, df = 1, P > 0.05). The differences in prevalence of E. bieneusi among squirrels of different ages were not statistically significant (χ2 = 0.353, df = 1, P > 0.05) (Table 2). No mixed infections of the two pathogens were found in red squirrels in our study. Similarly, a significant correlation between the different sources and E. bieneusi infection (P = 0.01) was observed by logistic regression analysis.

Cryptosporidium species/genotypes

Twenty-seven Cryptosporidium-positive samples were genotyped by sequence analysis of the SSU rRNA gene, and five Cryptosporidium species/genotypes were identified: Cryptosporidium rat genotype II (8/27, 30%), Cryptosporidium ferret genotype (8/27, 30%), Cryptosporidium chipmunk genotype III (5/27, 18.5%), Cryptosporidium rat genotype I (4/27, 14.8%), and C. parvum (2/27, 7.4%). Phylogenetic relationship analysis confirmed the identity of Cryptosporidium species/genotypes (Fig. 1). Cryptosporidium rat genotype II and Cryptosporidium ferret genotype were the two most predominant genotypes (Table 1).

Figure 1
figure 1

Phylogenetic relationships between the partial Cryptosporidium SSU rRNA gene from red squirrels and the Cryptosporidium spp. or genotypes deposited in GenBank. The GenBank accession number of each Cryptosporidium species or genotype is shown in parentheses. Bootstrap values above 50% from 1,000 replicates are shown at the nodes. The newly generated sequences are indicated in bold.

At the SSU rRNA locus, eight Cryptosporidium rat genotype II isolates had 100% homology between each other and were identical to that (GQ121025) from Rattus tanezumi in China and those from R. rattus in Australia (JX294365). The eight Cryptosporidium ferret genotype sequences were identical to reference sequences MF411071 (from Sciurus vulgaris in Italy) and AF112572 (from ferrets in the USA). Cryptosporidium chipmunk genotype III was identical to the reference sequence GQ121021 (from Siberian chipmunks in China); Cryptosporidium rat genotype I was identical to the reference sequences KP883292 (from R. rattus in Iran) and JN172971 (from R. norvegicus in Sweden); and C. parvum was identical to reference sequences MF671870 (from dairy cattle in China) and KT151548 (from coturnix in Iraq).

E. bieneusi genotypes

DNA sequencing and subsequent analysis of the ITS-PCR products from the 61 E. bieneusi-positive specimens revealed the existence of three known E. bieneusi genotypes (D, SCC-2, SCC-4) and two novel genotypes, which were named RS01 and RS02. Genotype D was the most prevalent (44.3%, 27/61) and showed 100% homology with the sequences JF927954 (from humans in China) and AY371284 (from humans in Peru). Genotypes SCC-2 and SCC-4 had 100% homology with the two sequences MF410401 and MF410403, respectively.

With regard to the two novel genotypes, RS01 displayed two single nucleotide polymorphisms (SNPs) within the 243 bp of the ITS gene sequence of E. bieneusi (G/A at positions 178 and 324), when compared with the genotype SCC-2 (MF410401), which showed 99% homology. RS02 had three SNPs (G/A at positions 101 and 178, A/T at position 108) in comparison with genotype SCC-2, with 99% homology.

Phylogenetic relationship of E. bieneusi

A phylogenetic analysis of the ITS gene sequences of all the genotypes of E. bieneusi obtained here and reference genotypes published previously revealed that genotype D clustered in group 1 and further clustered into 1a, whereas genotypes SCC-2, SCC-4, and two novel genotypes (RS01 and RS02) clustered in group 10 (Fig. 2).

Figure 2
figure 2

Phylogenetic relationships of the E. bieneusi genotypes identified in this study and other reported genotypes. The phylogeny was inferred with a neighbor-joining analysis of the internal transcribed spacer (ITS) sequences based on distances calculated with the Kimura two-parameter model. Bootstrap values greater than 50% from 1,000 replicates are shown at the nodes. Genotypes with open circles and solid circles are known and novel genotypes identified in this study, respectively.

Discussion

In 314 fecal samples of red squirrel, we first demonstrated the presence of Cryptosporidium spp. and E. bieneusi. The occurrence of Cryptosporidium spp. (8.6%, 27/314) was lower than the average prevalence previously reported in rodents (15.2%, 290/1911) (Table 3), but higher than those reported in laboratory rats in Nigeria (1.5%)20, brown rats in Iran (6.6%)21, and brown rats (6.2%), laboratory mice (1.7%), laboratory rats (4%), hamsters (7.8%), squirrel monkeys (4.2%), and Bamboo rats (3.3%) in China22,23,24. Cryptosporidium spp. has been reported in multiple rodent species, with an infection rate of 1.5% in laboratory rats to 85.0% in guinea pigs16,25,26,27. The relatively low occurrence of Cryptosporidium spp. in this study may be explained by the fact that the pet squirrels lived in clean environments and were kept in separate cages. The prevalence of E. bieneusi was 19.4%, which was similar to that in two recent studies in Sichuan province for E. bieneusi infection rates in red-bellied tree squirrels (16.7%, 24/144)18 and chipmunks (17.6%, 49/279)17. The prevalence of E. bieneusi in rodents ranged from 1.1% to 100% (Table 4)12,19,28,29,30. As proposed in other studies, factors contributing to the prevalence of these pathogens may include the examination method, age, sex, season, host health status, feeding density, sample size, geo-ecological conditions, and living conditions31,32,33.

Table 3 Occurrence of Cryptosporidium species/genotypes in rodents from different countries.
Table 4 Occurrence and genotypes of Enterocytozoon bieneusi in rodents from different countries.

Previous studies have indicated that five Cryptosporidium species and nine Cryptosporidium genotypes exist in various rodents in China (Table 3)12,22,23. In this study, five different Cryptosporidium species/genotypes were identified, including Cryptosporidium rat genotypes I and II, Cryptosporidium ferret genotype, Cryptosporidium chipmunk genotype III, and C. parvum. Cryptosporidium rat genotypes I and II have been found in brown rats in the Philippines34, Nigeria20, Australia35, and China22, even in South Nation River watershed, raw wastewater, and environmental samples in the United Kingdom, Canada, and China36,37,38. Cryptosporidium ferret genotype has been found in ferrets and red squirrels in Italy39. The Cryptosporidium chipmunk genotype III was previously reported in red squirrels, eastern squirrels, eastern chipmunks, and deer mice in the USA40. To date, little is known regarding the disease-causing potential of the four genotypes in humans and livestock; thus bringing attention to the need for epidemiological molecular surveillance of Cryptosporidium spp. for the assessment of infectivity across different hosts.

C. parvum is one of the two predominant Cryptosporidium species in humans41. C. parvum has been identified in humans in Henan province, China42. Moreover, C. parvum infections have been observed in brown rats in Japan, mice and red-backed voles in the USA, brown rats in Iran, striped field mice in Slovak Republic, and hamsters, Siberian chipmunks, chinchillas, and Bamboo rats in China, highlighting the prevalence of C. parvum in rodents (Table 3)12,25,27. In addition, C. parvum has also been found in other animals, such as cattle, sheep, goats, deer, alpacas, horses, dogs, gray wolves, raccoon dogs, cats, and pigs43,44. In this study, only two C. parvum isolates were identified in investigated red squirrels; however, these isolates may result in emerging zoonotic infections through the oral-fecal route.

Three known genotypes (D, SCC-2, and SCC-3) and two novel genotypes (RS01 and RS02) were identified in this study. Genotype D was the predominant genotype (44.3%, 27/61). This finding was similar to previous reports in mice in Czech Republic and Germany45 (32.3%, 10/31), mice in Poland46 (33.3%, 10/30), and brown rats (89.5%, 17/19) and red-bellied tree squirrels (75%, 18/24) in China12,18. In China, genotype D has been identified in humans and various animals, such as nonhuman primates, cattle, sheep, horses, pigs, dogs, cats, and in wastewater14,47,48,49. This study demonstrated the presence of genotype D in red squirrels for the first time, suggesting that red squirrels could play a potential role in the disease dissemination of E. bieneusi to humans.

Conclusions

This is the first report on the incidence of Cryptosporidium spp. and E. bieneusi in pet red squirrels in China. The infection rates of Cryptosporidium spp. and E. bieneusi were 8.6% and 19.4%, respectively. The detection of zoonotic C. parvum and genotype D of E. bieneusi suggests that red squirrels are a potential source of cryptosporidiosis and microsporidiosis in humans. However, the infection sources and transmission dynamics between red squirrels and humans remain unknown, thus emphasizing on the importance of further follow-up studies of the transmission dynamics of these pathogens.

Materials and Methods

Ethics statement

The present study protocol was reviewed and approved by the Research Ethics Committee and the Animal Ethical Committee of Sichuan Agricultural University, and all methods were performed in accordance with the relevant guidelines and regulations. Permission was obtained from the owners or shop managers before the fecal specimens were collected.

Collection of specimens

A total of 314 fecal specimens were collected from red squirrels from four pet shops (n = 269) and owners (n = 45) in the Sichuan province, southwestern China between September 2016 and December 2017 (Table 1). All tested pet shops only raised red squirrels and served as suppliers of red squirrels to other pet shops. Sample size was approximately 20% of the squirrels from each shop, and small-scale shops (population less than 50) were not included. The four pet shops are distributed in Jianyang (104°32′E, 30°24′N), Pengzhou (103°57′E, 30°59′N), Wenjiang (103°51′E, 30°40′N), and Jintang (104°24′E, 30°51′N). Pet squirrels from owners were primarily distributed around the urban areas of Chengdu city (104°03′E–104°08′E, 30°36′N–30°52N′). In both pet shops and homes, red squirrels were housed in separate cages. Approximately 30–50 g fresh fecal samples were collected from the bottom of each cage after defecation using a sterile disposal latex glove and then immediately placed into individual disposable plastic bags. No obvious clinical signs were observed at the time of sampling, and the age, sex, and source were recorded at the same time. All fecal specimens were stored in 2.5% potassium dichromate solution at 4 °C until processing.

DNA extraction

The fecal specimens were washed three times in distilled water with centrifugation at 3,000 × g for 10 min to remove the potassium dichromate. Genomic DNA was extracted from approximately 200 mg of each processed fecal specimen using an E.Z. N. A. R Stool DNA kit (Omega Biotek Inc., Norcross, GA, USA) according to the manufacturer’s recommended instructions. The extracted DNA was stored at −20 °C until molecular analysis.

Genotyping of Cryptosporidium spp. and E. bieneusi

Cryptosporidium spp. were identified by nested polymerase chain reaction (PCR) amplification of an SSU rRNA gene fragment of ~830 bp designed by Xiao et al.50. E. bieneusi genotypes were determined by nested PCR amplification of a 392-bp fragment containing the entire ITS (243 bp) and portions of the flanking large and small subunits of the rRNA gene31 (Supplementary Table 1). TaKaRa Taq DNA Polymerase (TaKaRa Bio, Otsu, Japan) was used for PCR amplification. Positive controls (camel-derived C. andersoni DNA for Cryptosporidium spp. and horse-derived genotype D DNA for E. bieneusi) and negative control with no DNA added were included in all PCR assays. The secondary PCR products were examined by agarose gel electrophoresis and visualized after ethidium bromide staining.

Sequence analysis

All nested PCR positive-products were sequenced using the same PCR primers as those used for the secondary PCRs on an ABI 3730 instrument (Applied Biosystems, Foster City, CA, USA) at the BioSune Biotechnology Company (Shanghai, China). The nucleotide sequences of each obtained gene were aligned and analyzed using the Basic Local Alignment Search Tool and Clustal X (http://www.clustal.org/) with reference sequences retrieved from GenBank to identify Cryptosporidium spp. and E. bieneusi genotypes.

Phylogenetic analyses

To support the Cryptosporidium species/genotypes and assess the genetic relationships between the E. bieneusi genotypes in the present study and reference sequences previously published in GenBank, phylogenetic analysis was performed using Phylip version 3.69 package and by constructing a neighboring-joining tree using Mega 6 software (http://www.megasoftware.net/), which is based on evolutionary distances calculated using a Kimura 2-parameter model. The MegAlign program in the DNA Star software package (version 5.0) was used to determine the degree of sequence identity. The reliability of these trees was assessed using bootstrap analysis with 1,000 replicates.

Statistical analysis

Variations in the occurrence of Cryptosporidium spp. (y1) and E. bieneusi (y2) in red squirrels according to age (x1), sex (x2), and geographical location (x3) were analyzed by χ2 test using SPSS V20.0 (IBM, Chicago, IL, USA). Each of these variables was included in the binary logit model as an independent variable by multivariable regression analysis. When the P value was less than 0.05, the results were considered statistically significant. The adjusted odds ratio (OR) and 95% confidence interval (CI) for each variable were calculated with binary logistic regression, and all risk factors were entered simultaneously.

GenBank accession numbers

Representative nucleotide sequences were deposited in GenBank with the following accession numbers: MH940281-MH940290.