Occurrence and genetic characteristics of Cryptosporidium spp. and Enterocytozoon bieneusi in pet red squirrels (Sciurus vulgaris) in China.

Cryptosporidium spp. and Enterocytozoon bieneusi are two well-known protist pathogens which can result in diarrhea in humans and animals. To examine the occurrence and genetic characteristics of Cryptosporidium spp. and E. bieneusi in pet red squirrels (Sciurus vulgaris), 314 fecal specimens were collected from red squirrels from four pet shops and owners in Sichuan province, China. Cryptosporidium spp. and E. bieneusi were examined by nested PCR targeting the partial small subunit rRNA (SSU rRNA) gene and the ribosomal internal transcribed spacer (ITS) gene respectively. The infection rates were 8.6% (27/314) for Cryptosporidium spp. and 19.4% (61/314) for E. bieneusi. Five Cryptosporidium species/genotypes were identified by DNA sequence analysis: Cryptosporidium rat genotype II (n = 8), Cryptosporidium ferret genotype (n = 8), Cryptosporidium chipmunk genotype III (n = 5), Cryptosporidium rat genotype I (n = 4), and Cryptosporidium parvum (n = 2). Additionally, a total of five E. bieneusi genotypes were revealed, including three known genotypes (D, SCC-2, and SCC-3) and two novel genotypes (RS01 and RS02). Phylogenetic analysis revealed that genotype D fell into group 1, whereas the remaining genotypes clustered into group 10. To our knowledge, this is the first study to report Cryptosporidium spp. and E. bieneusi in pet red squirrels in China. Moreover, C. parvum and genotype D of E. bieneusi, previously identified in humans, were also found in red squirrels, suggesting that red squirrels may give rise to cryptosporidiosis and microsporidiosis in humans through zoonotic transmissions. These results provide preliminary reference data for monitoring Cryptosporidium spp. and E. bieneusi infections in pet red squirrels and humans.

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 health[1]. Humans are infected with these pathogens mainly via the fecal-oral route with anthroponotic and zoonotic transmission or via food-borne and water-borne transmission[2,3]. Clinical manifestations of infection with these pathogens are often inconsistent due to the variabilities in the health condition of infected hosts[4,5]. In healthy individuals, these pathogens usually cause asymptomatic infection or self-limiting diarrhea[6]. 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 transplantation[7]. 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 insects[8]. 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 mortality[9].

To detect and evaluate potential zoonotic transmissions, it is necessary to accurately distinguish Cryptosporidium spp. and E. bieneusi on the molecular level[10]. To date, at least 37 species and over 70 genotypes of Cryptosporidium spp. have been described[11]. Among them, 11 Cryptosporidium species have been identified in rodents, Cryptosporidium parvum and C. muris are the most common[12]. For E. bieneusi, more than 474 genotypes have been identified based on the internal transcribed spacer (ITS) region of the rRNA gene[13], and more than 35 genotypes have been determined in rodents[14]. These genotypes can be classified into eleven groups (groups 1–11) by phylogenetic analyses[13]. 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 water[13].

In China, Cryptosporidium spp. and E. bieneusi have been detected in a wide range of hosts, including carnivores, lagomorphs, primates, birds, and rodents[14,15]. 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 humans[16-19]. 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.

Sources	No. of examined	Cryptosporidium spp.				E. bieneusi
		No. of positive	Prevalence (%) (95% CI)	OR (95% CI)	Species/Genotype (n)	No. of positive	Prevalence (%)(95% CI)	OR (95% CI)	Genotype (n)
Pet shop 1	58	2	3.4% (0.014–0.083)	reference	rat genotype I (2)	5	8.6% (0.012–0.161)	reference	D (3), RS01 (2)
Pet shop 2	74	12	16.2% (0.076–0.248)	5.4 (1.2–25.3)	rat genotype II (8), chipmunk genotype III (3), C. parvum (1)	16	21.6% (0.12–0.312)	2.9 (1.0–8.5)	D (6), SCC-2 (8), SCC-3 (2)
Pet shop 3	76	8	10.5% (0.035–0.176)	3.2 (0.7–16.1)	ferret genotype (6), chipmunk genotype III (2)	21	27.6% (0.173–0.379)	4.0 (1.4–11.5)	D (13), SCC-2 (6), RS02 (2)
Pet shop 4	61	4	6.6% (0.002–0.129)	2.0 (0.3–11.2)	rat genotype I (2), ferret genotype (2)	14	23% (0.121–0.338)	3.2 (1.1–9.4)	D (4), SCC-3 (10)
owners	45	1	2.2% (0.023–0.067)	0.6 (0.1–7.2)	C. parvum (1)	5	11.1% (0.016–0.207)	1.3 (0.4–4.9)	D (1), SCC-2 (4)
Total	314	27	8.6% (0.055–0.117)		rat genotype II (8), ferret genotype (8), chipmunk genotype III (5), rat genotype I (4), C. parvum (2)	61	19.4% (0.150–0.238)		D (27), SCC-2 (18), SCC-3 (12), RS01 (2), RS02 (2)

Table 2

Occurrence of Cryptosporidium spp. and Enterocytozoon bieneusi in pet red squirrels according to sex and age.

Factor	Characteristics	No. of examined	Cryptosporidium spp.			E. bieneusi
			No. of positive	Prevalence (%) (95% CI)	OR (95% CI)	No. of positive	Prevalence (%) (95% CI)	OR (95% CI)
Sex	Male	148	12	8.1% (0.037–0.126)	reference	27	18.2% (0.119–0.245)	reference
	Female	166	15	9% (0.046–0.134)	1.1 (0.5–2.5)	34	20.5% (0.143–0.267)	1.2 (0.7–2.0)
Age	≤3 months	154	14	9.1% (0.045–0.137)	reference	32	20.8% (0.143–0.273)	reference
	>3 months	160	13	8.1% (0.038–0.124)	0.9 (0.4–1.9)	29	18.1% (0.121–0.242)	0.8 (0.5–1.5)

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

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

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 China[22-24]. Cryptosporidium spp. has been reported in multiple rodent species, with an infection rate of 1.5% in laboratory rats to 85.0% in guinea pigs[16,25-27]. 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)[12,19,28-30]. 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 conditions[31-33].

Table 3

Occurrence of Cryptosporidium species/genotypes in rodents from different countries.

Country	Host(common name)	Scientific name	No. of samples	No. of positive (%)	Species/Genotype (n)	References
Japan	Brown rats	Rattus norvegicus	50	19 (38)	C. meleagridis (1), C. parvum (1), New genotypes (11)	Kimura et al., 2007
USA	Opossum	Didelphis virginiana	—	2	Marsupial genotype (2)	Feng et al., 2007
	Chipmunk	Tamias striatus	—	1	C. baylei (1)
	Gray squirrel	Sciurus carolinensis	—	1	Skunk genotype (1)
	White-footed mouse	Peromyscus leucopus	—	3	C. parvum (3)
	Deer mouse	Peromyscus maniculatus	—	3	C. parvum (2), Muskrat II genotype (1)
	Red-backed vole	Clethrionomys gapperi	—	2	C. parvum (1), Muskrat II genotype (1)
	Meadow vole	Microtus pennsylvanicus	—	5	Muskrat II genotype (5)
	House mouse	Mus musculus	—	1	C. parvum (1)
Australia	Black rats	Rattus rattus	85	7 (8.2)	rat genotype III (4), rat genotype II (3)	Koehler et al., 2018
	Swamp rats	Rattus lutreolus	21	3 (14.3)	C. viatorum (3)
Philippines	Asian house rat	Rattus tanezumi	83	37 (44.6)	rat genotype III (18), rat genotype IV (5), suis-like genotype (5), C. scrofarum (3), rat genotype I (1), rat genotype II (1), C. muris (1)	Nghublin et al., 2013
	Brown rat	Rattus norvegicus	70	12 (18.6)	rat genotype II (5), rat genotype IV (1), C. muris (2), rat genotype I (2), C. scrofarum (1), rat genotype III (1)
Iran	Brown rat	Rattus norvegicus	91	6 (6.6)	C. parvum + C. muris (6)	Gholipoury et al., 2016
Nigeria	Laboratory rats	Rattus norvegicus	134	2 (1.5)	C. andersoni (1), rat genotype II (1)	Ayinmode et al., 2017
Slovak Republic	Striped field mouse	Apodemus agrarius	103	34 (33)	C. scrofarum (18), C. parvum (9), Muskrat genotype II (3), C. environment isolate (3), C. hominis (1)	Danišová et al., 2017
	Bank vole	Myodes glareolus	72	16 (22.2)	C. scrofarum (4), C. parvum (3), Muskrat genotype I (3), C. environment isolate (6)
	Yellow-necked mouse	Apodemus flavicollis	73	15 (20.5)	C. scrofarum (5), C. suis (4), C. parvum (3), C. environment isolate (3)
Italy	Red squirrels	Sciurus vulgaris	70	17 (24.3)	ferret genotype (15), chipmunk genotype I (2)	Kvác et al., 2008
China	Brown rat	Rattus norvegicus	64	4 (6.2)	C. tyzzeri (3), rat genotype III (1)	Lv et al., 2009
	Asian house rat	Rattus tanezumi	33	5 (15.2)	C. tyzzeri (1), rat genotype II (2), rat genotype III (2)
	Laboratory mouse	Mus musculus	229	4 (1.7)	C. tyzzeri (4)
	Laboratory rat	Rattus norvegicus	25	1 (4)	C. tyzzeri (1)
	Golden hamster	Mesocricetus auratus	50	16 (32)	C. muris (7), C. andersoni (5), C. parvum (4)
	Siberian hamster	Phodopus sungorus	51	4 (7.8)	C. parvum (2), C. muris (1), hamster genotype (1)
	Campbell hamster	Phodopus campbelli	30	3 (10)	C. parvum (2), C. andersoni (1)
	Red squirrel	Sciurus vulgaris	19	5 (26.3)	ferret genotype (5)
	Siberian chipmunk	Tamias sibiricus	20	6 (30)	ferret genotype (4), C. parvum (1), C. muris (1), chipmunk genotype III (1)
	Guinea pig	Cavia porcellus	40	34 (85)	C. wrairi (30)
	Chinchillas	Chinchilla lanigera	140	14 (10)	C. ubiquitum (13), C. parvum (1).	Qi et al., 2015
	Brown rats	Rattus norvegicus	242	22 (9.1)	rat genotype I (14), rat genotype IV (6), C. suis-like genotype (1), C. ubiquitum (1)	Zhao et al., 2018
	Squirrel monkey	Saimiri sciureus	24	1 (4.2)	C. hominis monkey genotype II (1)	Liu et al., 2015a
	Bamboo rats	Rhizomys sinensis	92	3 (3.3)	C. parvum (3)	Liu et al., 2015b

-represents unknown.

Table 4

Occurrence and genotypes of Enterocytozoon bieneusi in rodents from different countries.

Country	Host (common name)	Scientific name	No. of samples	No. of positive (%)	Genotypes (no.)	References
Czech Republic and Germany	East-European house mice	Mus musculus musculus	127	14 (11)	D (6), PigEBITS5 (4), EpbA (2), C (1), H (1)	Sak et al., 2011
	West-European house mice	Mus musculus domesticus	162	17 (10.5)	D (4), Peru 8 (4), CZ3 (4), PigEBITS5 (3), S6 (1), C (1)
United States	Eastern gray squirrel	Sciurus carolinensis	34	11 (32.4)	WL4 (5), Type IV (3), WW6 (2), PtEbV + WL21 (1)	Guo et al., 2014
	Eastern chipmunk	Tamias striatus	7	5 (71.4)	WL4 (3), Type IV (1), WL23 (1)
	Woodchuck	Marmota monax	5	5 (100)	WL4 (2), Type IV + WL20 (1), WL22 (1), WW6 (1)
	Deer mouse	Peromyscus sp.	55	13 (23.6)	WL4 (10), WL23 (2), WL25 (1)
	Boreal red-backed vole	Myodes gapperi	5	1 (20)	WL20 + WL21(1)
	Meadow vole	Microtus pennsylvanicus	10	3(33)	Peru11 (1), Peru11 + type IV (1), WL21 + unknown (1)
	Guinea pigs	Cavia porcellus	60	4 (6.7)	Peru 16 (4)	Cama et al., 2007
	Black-tailed prairie dogs	Cynomys ludovicianus	153	14 (9.2)	Row (14)
Poland	Pallas	Apodemus agrarius	184	79a	D (6), WR8 (2), WR5 (1), WR7 (1), gorilla 1 (1)	Perec-Matysiak et al., 2015
	Yellow-necked mouse	Apodemus flavicollis	60	18a	D (2), WR6 (6), WR4 (1), WR1 (1), WR9 (1)
	Bank vole	Myodes glareolus	46	18a	D (2), WR6 (2), WR10 (2), WR2 (1)
	House mouse	Mus musculus	21	6a	WR3 (1)
Slovakia	House mouse	Mus musculus musculus	280	3 (1.1)	Unknown	Danišová et al., 2015
China	Chinchillas	Chinchilla lanigera	140	5 (3.6)	D (2), BEB6 (3)	Qi et al.,
	Brown rats	Rattus norvegicus	242	19 (7.9)	D (17), Peru6 (2)	Zhao et al., 2018
	Red-bellied tree squirrels	Callosciurus erythraeus	144	24 (16.7)	D (18), EbpC (3), SC02 (1), CE01 (1), CE02 (1)	Deng et al., 2016
	Chipmunks	Eutamias asiaticus	279	49 (17.6)	D (6), SCC-1 (17), SCC-2 (9), SCC-3 (5), CHY1 (5), Nig 7 (4), CHG9 (2), SCC-4 (1)	Deng et al., 2018a

aRepresents positive samples in feces and spleen.

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 Philippines[34], Nigeria[20], Australia[35], and China[22], even in South Nation River watershed, raw wastewater, and environmental samples in the United Kingdom, Canada, and China[36-38]. Cryptosporidium ferret genotype has been found in ferrets and red squirrels in Italy[39]. The Cryptosporidium chipmunk genotype III was previously reported in red squirrels, eastern squirrels, eastern chipmunks, and deer mice in the USA[40]. 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 humans[41]. C. parvum has been identified in humans in Henan province, China[42]. 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 pigs[43,44]. 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 Germany[45] (32.3%, 10/31), mice in Poland[46] (33.3%, 10/30), and brown rats (89.5%, 17/19) and red-bellied tree squirrels (75%, 18/24) in China[12,18]. In China, genotype D has been identified in humans and various animals, such as nonhuman primates, cattle, sheep, horses, pigs, dogs, cats, and in wastewater[14,47-49]. 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 gene[31] (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.

Supplementary information