Background

Cryptosporidium spp., Giardia duodenalis and Enterocytozoon bieneusi are three important intestinal protozoa that can cause gastrointestinal discomfort and diarrhea in various hosts [1, 2]. The infections cause by these pathogens are self-limiting in healthy individuals, but in immunocompromised individuals, the infection period can be protracted, and even turn out to be life-threatening [3, 4]. To date, at least 42 Cryptosporidium spp. and 60 genotypes have been identified [5], with most of these species and genotypes being host-specific. Giardia duodenalis is currently classified into eight genetic assemblages (A–H) of which assemblages A and B are zoonotic [6]. For E. Bieneusi, > 500 distinct genotypes have been reported, and phylogenetic analysis has divided these into 11 distinct groups (groups 1–11), with > 90% of the genotypes belonging to groups 1 or 2 [7]. Some genotypes are found in a variety of animals, including humans, thus indicating their zoonotic potential.

The prevalence and genotype distribution of these three pathogens in sheep and goats has been widely reported [8,9,10], but most previous studies have involved an intensive farming environment. Tibetan sheep, which are highly adapted to the high altitudes of Qinghai Province and economically important to local herders, are generally raised using a combination of supplementary feeding and semi-stocking. During the growing season (June–October), when natural pasture can provide enough herbage, Tibetan sheep are always raised in free pastures [11]. Recent studies on Cryptosporidium spp., G. duodenalis and E. bieneusi infections in Tibetan sheep have been reported conflicting results [12,13,14,15]. To date, there has been no systematic study on the seasonal distribution of these pathogens in Tibetan sheep.

The aim of the present study was to examine the prevalence, genotype characterization and seasonal distribution of Cryptosporidium spp., G. duodenalis and E. bieneusi in Tibetan sheep in Qinghai, China, and to assess the zoonotic transmission potential of these pathogens and their impact on public health.

Methods

Sample collection

From May 2016 to August 2017, 761 fecal samples were collected from Tibetan sheep in seven counties in Qinghai Province, China. All samples were collected from grazing sheep with no adverse clinical symptoms. The age difference between sheep was relatively small. As the sheep were raised in a natural pasture, we collected the top layers of the fecal material immediately after defecation, thus avoiding the part in contact with the ground. The sheep were numbered before sampling, and only one fecal sample was collected per animal. These samples were transported to the laboratory under cool conditions and preserved in 2.5% potassium dichromate at 4 °C until DNA extraction.

DNA extraction

Each fecal sample (0.5 mg) was washed 3 times with distilled water to remove the potassium dichromate. DNA was extracted using the Stool DNA Kit (OMEGA, China) according to the manufacturer’s instructions and then stored at −20 °C until PCR amplification.

Detection, genotyping and subtyping of Cryptosporidium spp.

Cryptosporidium spp. were examined by PCR analysis of an approximately 830-bp fragment of the small subunit ribosomal RNA gene (18S rRNA) [16]. The Cryptosporidium spp. present in the samples were identified to the species level by sequence analysis of the secondary PCR products. Cryptosporidium ubiquitum was then subtyped using a PCR assay and sequence analysis of an approximately 850-bp fragment of the 60-kDa glycoprotein gen (gp60), as described previously [17].

Detection, genotyping and subtyping of G. duodenalis

Genotyping of G. duodenalis was performed by PCR analysis of an approximately 532-bp fragment of triosephosphate isomerase genetic locus (TPI) [18]. Assemblages of G. duodenalis were determined using sequence analysis of the secondary PCR products.

Detection, genotyping and subtyping of E. bieneusi

Enterocytozoon bieneusi was detected by PCR analysis of an approximately 390-bp fragment of the internal transcribed spacer gene (ITS) [19]. Genotypes of E. bieneusi were determined by sequence analysis of the PCR products.

DNA sequence analysis

All DNA samples which tested for the pathogens were sent to Sangon Biotech Co., Ltd. (Shanghai (China) for bidirectional DNA sequence analysis. Raw sequences were assembled using DNAStar 5.0 [20] and aligned using Clustal X 1.83 [21], following which the sequences were used to construct a phylogenetic analysis tree using the maximum likelihood (ML) method, with MEGA 7.0.26 software [22]. The Hasegawa-Kishino-Yano (HKY) model and gamma distribution were used to calculate the substitution rates to identify the genotypes of E. bieneusi. The reliability of each phylogenetic tree was assessed using a bootstrap analysis with 1000 replicates.

Statistical analysis

The Chi-square test (χ2 test) was used to determine the relationships between positivity rates and locations of Cryptosporidium spp., G. duodenalis and E. bieneusi, as well as the relationships between the positivity rates and seasons. Statistical analysis was implemented in SPSS software version 20.0 (SPSS IBM, Armonk, NY, USA) for Windows. Differences were considered significant at the 0.05 level.

Results

Mixed infection of Cryptosporidium spp., G. duodenalis and E. bieneusi in Tibetan sheep

Five fecal samples were identified having mixed infections. One was positive for both Cryptosporidium spp. and G. duodenalis; one was positive for both Cryptosporidium spp. and E. bieneusi; and the remaining three fecal samples contained a mixture of G. duodenalis and E. bieneusi. However, in none of the samples were all three pathogens detected concurrently.

Prevalence and seasonal distribution of Cryptosporidium spp. in Tibetan sheep

PCR analysis confirmed that 28 (3.68%) of fecal samples collected from Tibetan sheep were positive for Cryptosporidium spp. Across the seven counties in Qinghai Province where samples were collected from sheep, Cryptosporidium spp. was only found in four counties, where the positivity rates ranged from 2.80% (Huangnan County, 3/107) to 6.13% (Haibei County, 13/212) (Table 1); however, the differences were not statistically significant (χ2 = 10.18, df = 6, P ˃ 0.05).

Table 1 Prevalence and species/assemblage/genotype distribution of Cryptosporidium spp., Giardia duodenalis and Enterocytozoon bieneusi in Tibetan sheep in Qinghai Province
Full size table

Samples positive for Cryptosporidium spp. were found in all seasons, with the highest rate, 7.56% (16/212), in the summer (Table 1). Across different seasons, the positivity rate of Cryptosporidium spp. showed significant differences (χ2 = 13.36, df = 3, P < 0.01).

The results of the DNA sequence analysis of the 18S rRNA gene products showed that the sequences were highly similar (> 99%) to those of known Cryptosporidium spp. Subsequent phylogenetic analysis of these sequences identified four species among the 28 isolates of Cryptosporidium spp.: C. xiaoi (n = 13 samples), C. ubiquitum (n = 8), C. bovis (n = 6) and C. ryanae (n = 1), with C. xiaoi being the predominant species (13/28, 46.43%) in Tibetan sheep in Qinghai Province. For C. ubiquitum, only three of the eight positive samples were successfully subtyped, yielding subtype XIIa. Cryptosporidium ryanae was only detected in one sample, and the sequence showed 100% homology to subtype KT922234 derived from a calf in Ethiopia.

Prevalence and seasonal distribution of G. duodenalis species in Tibetan sheep

Of the 761 fecal samples collected from Tibetan sheep in Qinghai Province, 12 (1.58%) tested positive for G. duodenalis. These positive samples came from three counties: Xining (5/164, 3.05%), Haibei (5/212, 2.36%) and Hainan (2/124, 1.61%) (Table 1); however, the differences in positivity rate were not statistically significant (χ2 = 7.31, df = 6, P < 0.01).

Positive specimens of G. duodenalis were found in three seasons, but not in winter. The positivity rate was higher in spring (4.92%, 9/183) than in summer and autumn, and the differences were statistically significant (χ2 = 12.60, df = 3, P < 0.01).

DNA sequence analysis led to the identification of two genotypes, and comparison of the similarity with those from from GenBank data (Additional file 1: Dataset 1) showed > 99% similarity. Two samples showed a similarity of 99.81% to zoonotic assemblage A, and the remaining ten sequences were identical to assemblage E, with similarity to GenBank sequences ranging from 99.43% to 100% after BLAST (Basic Local Alignment Search Tool) analysis.

Prevalence and seasonal distribution of E. bieneusi genotypes in Tibetan sheep

The PCR results on the ITS locus showed that 49 (6.44%) samples from Tibetan sheep were positive for E. bieneusi. Enterocytozoon bieneusi was detected in samples from all counties except Golog, with positivity rates ranging from 1.92% to 13.41%. The highest positivity rate was detected in Xining (22/164, 13.41%) (Table 1). Analysis showed that the differences in positivity rate were statistically significant (χ2 = 19.39, df = 6, P < 0.01).

Positive samples of E. bieneusi were found across all seasons, with the highest rate in summer (13.21%, 28/212) (Table 1). The results also showed that the differences in positivity rates of E. bieneusi in different seasons were significant (χ2 = 24.25, df = 3, P < 0.01).

Comparison of the sequences with those in the GenBank database using BLAST analysis revealed five genotypes: BEB6 (n = 21 samples), COS-I (n = 14), CHS3 (n = 11), CGS1 (n = 2) and PIGEBITS5 (n = 1). Phylogeny analysis indicated that, with the exception of genotype PIGEBITS5, which belongs to group 1, the remaining genotypes all belonged to group 2.

Discussion

In this study, we found that the prevalence of Cryptosporidium spp., G. duodenalis and E. bieneusi in Tibetan sheep was 3.68, 1.58 and 6.44%, respectively. The results of this study showed that the prevalence of these pathogens differed significantly across seasons (Fig. 1). Prior to this study, prevalence data on the seasonal distribution of these pathogens were limited for sheep in China, with the few previous studies reporting on the prevalence of these pathogens in livestock in Ireland, India and Jordan [23,24,25]. Other related studies mainly focused on humans. The reasons for the seasonal differences observed in the present study are unclear. Many factors, including levels of sunlight and germicidal ultraviolet radiation, environmental temperatures, humidity, breeding density and precipitation, can contribute to such results [26,27,28].

Fig. 1
figure 1

Prevalence of Cryptosporidium spp., Giardia duodenalis and Enterocytozoon bieneusi in Tibetan sheep across the different seasons

Full size image

Cryptosporidium spp. are important protozoan parasites that target the gastrointestinal tract of various hosts, including humans, domestic animals and wildlife [29]. In the present study, the overall infection rate of Cryptosporidium spp. in Tibetan sheep was 3.68%. In comparison, previous studies reported that the infection rates of Cryptosporidium spp. in sheep and goats were between 2.75% and 45.5% in different provinces and cites in China [12, 14, 30,31,32,33]. The Cryptosporidium spp. infection rate found in the present study is higher than that reported in Papua New Guinea (2.2%) [34] and Egypt (2.5%) [35], but lower than that reported in other countries, such as Greece, Spain, Algeria, Tunisia, Jordan, Poland, Norway and Mexico, where studies reported a wide range, from 5.1% to 67.5% [26, 36,37,38,39,40,41,42]. The differences in infection rates between these studies can be attributed to a variety of reasons, such as sample sizes, climate, animal age and animal management methods.

To date, > 10 species of Cryptosporidium have been identified in sheep, including C. xiaoi, C. ubiquitum, C. parvum, C. andersoni, C. fayeri, C. ryanae, C. scrofarum, C. hominis, C. suis and C. bovis [30]. In the present study, four Cryptosporidium spp. were isolated from Tibetan sheep in Qinghai: C. xiaoi (46.43%, 13/28), C. ubiquitum (28.57%, 8/28), C. bovis (21.43%, 6/28) and C. ryanae (3.57%, 1/28). Cryptosporidium xiaoi was the dominant species, consistent with previous reports on Tibetan sheep in Qinghai and Inner Mongolia in China [12, 33]. For C. ubiquitum, only three isolates were successfully subtyped, among which subtype XIIa has been found in humans and ruminants worldwide. This subtype has also been detected in Tibetan sheep, reflecting its zoonotic potential [43, 44]. Previous studies reported the C. ryanae was common in bovines, barking deer, Cervus uincolor, buffalo and deer [45]. Our study is the first to detect this species in Tibetan sheep. Mirhashemi et al. detected C. ryanae in sheep in Ireland and reported that it was the dominant Cryptosporidium species in cattle [25]; it has also been reported in yaks in Qinghai [46]. During summer, which is the growing season, yaks generally share the same pasture with Tibetan sheep; therefore, C. ryanae has the potential to spread between yaks and Tibetan sheep, and the animals can infect each other by contaminating the pasture.

Similar to the Cryptosporidium spp. infection rates, the infection rates of G. duodenalis reported in the present study are drastically different from those reported in previous studies. We found an infection rate in Tibetan sheep of 1.58% which, when compared with rates previously reported in China, are similar to those documented for Tibetan sheep in Gansu (1.7%) [14] and Qinghai (1.3%) [47], but higher than those obtained for Tibetan sheep (0.6%) and goats (0%) in Tibet [48] and sheep in Qinghai (0%) [49]. However, the infection rate is lower than those reported in previous studies on sheep in Heilongjiang (4.3%) [50] and Inner Mongolia (4.3%) [51], and especially the Tibetan sheep in Qinghai (13.1%) [13]. Globally, many researchers have conducted extensive investigations on sheep and goats infected with G. duodenalis, and the reported infection rates vary from 1.5% to 55.6% [52, 53]. In addition, in our study there was no significant difference between G. duodenalis infections at different altitudes (the altitude variation among the seven sampling counties was 1980 m), which is consistent with the results of a study in the Qinghai-Tibetan Plateau Area (QTPA) (which includes Qinghai, Yunnan and Tibet) [47].

Three assemblages (A, B, E) have been isolated from sheep to date. Assemblage E is the predominant genotype and has a significantly higher prevalence than assemblages A and B [8, 50, 54, 55]. In the present study, sequence comparison showed that two assemblages, E and A, were present in Tibetan sheep. In the past, livestock-specific assemblage E was not considered to be zoonotic as it was mostly detected in sheep, goats, pigs, among others [52]. However, there are emerging reports about this assemblage being detected in three human fecal samples in Egypt [56], and it was subsequently found in persons living in rural settings in Egypt [57], Rio de Janeiro, Brazil [58] and Queensland, Australia [59] and in primates (red colobus) of western Uganda [60]; these results show that assemblage E has zoonotic potential. Therefore, Tibetan sheep herders should be alert to this risk of infection.

In the present study, the infection rate of E. bieneusi in Tibetan sheep was 6.44%. Worldwide, several studies have been conducted to identify and assess the prevalence of E. bieneusi in sheep and goats. These data are mostly from China [8, 9, 14, 15, 51, 61,62,63], with a few other reports from Iran [64], Brazil [61] and Sweden [65]. The prevalence of E. bieneusi infection in sheep reported in these studies ranges from 4.4% to 69.3%, whereas in goats, it ranges from 7.5% to 32.9%. Three studies reported the infection rates of E. bieneusi in Tibetan sheep from Qinghai, Gansu and Tibet in China to be 23.4, 34.5 and 15%, respectively [14, 15, 48]. Compared with the results of the majority of these earlier studies, in our study the infection rate of E. bieneusi in Tibetan sheep in Qinghai was relatively lower.

Many genotypes of E. bieneusi have been found in ovines globally through phylogenetic analysis [66]. Most cluster with host-specific groups 1 and 2, which are zoonotic; only the CM4 and CHG21 genotypes belong to group 9 (Table 2). However, many new genotypes are isolated from ovines every year, constantly supplementing the genotype distribution in these animals. In the present study, five genotypes were identified from 49 E. bieneusi-positive samples using phylogenetic analysis: BEB6, COS-I, CGS1 and CHS3 belonging to group 2, and PIGEBITS5 belonging to group 1 (Fig. 2). BEB6 (42.9%, 21/49) was the dominant genotype in Tibetan sheep in the present study, which is consistent with the results of previous studies in Qinghai, Henan and Inner Mongolia [9, 15, 51]. CGS1 is a novel genotype that was first identified in Tibetan sheep in Gansu [14]; to date, it has not been isolated from other animals. This new genotype may be a result of host–parasite interactions. Recently, the PIGEBITS5 genotype was found in three Tibetan sheep fecal samples in Tibet [48]. Worldwide, the PIGEBITS5 genotype was first identified in swine in the USA [19]. A subsequent study by Abe and Kimata on pigs in Japan provided strong evidence that the PigEBITS5 genotypes are pig-specific [67], a finding which has been confirmed by many subsequent studies [67–69]. However, this genotype has also been detected in dairy calves [70, 71], Macaca nemestrina [72], dogs in China [73] and humans in Czech Republic [74], implying that it may infect a wide range of hosts and is of zoonotic potential.

Table 2 Distribution of Enterocytozoon bieneusi genotypes in ovines in previous studies
Full size table
Fig. 2
figure 2

Phylogenetic tree of Enterocytozoon bieneusi internal transcribed spacer sequences based on the maximum likelihood method

Full size image

Conclusions

The findings if the present study demonstrate the prevalence, genotype characterization and seasonal distribution of Cryptosporidium spp., G. duodenalis and E. bieneusi in Tibetan sheep in Qinghai Province, China. Four species of Cryptosporidium spp. were detected, with C. xiaoi being the dominant species, and Cryptosporidium ryanae in Tibetan sheep is reported for the first time. The frequency of G. duodenalis assemblages E and A showed that the risk of this pathogen to public health in this region may not be high. Furthermore, based on the ITS region, five genotypes of E. bieneusi were detected, which clustered into zoonotic phylogenetic groups 1 and 2. This result indicates that Tibetan sheep may be a potential source of zoonotic E. bieneusi infection. Systematic analysis was used to detect the seasonal differences for these three protozoan pathogens. More detailed studies are required to assess the zoonotic transmission ability of Cryptosporidium spp., G. duodenalis and E. bieneusi from sheep, and the impact of these pathogens on public health.