Host specificity of Enterocytozoon bieneusi genotypes in Bactrian camels (Camelus bactrianus) in China

Background Enterocytozoon bieneusi is an obligate, intracellular fungus and is commonly reported in humans and animals. To date, there have been no reports of E. bieneusi infections in Bactrian camels (Camelus bactrianus). The present study was conducted to understand the occurrence and molecular characteristics of E. bieneusi in Bactrian camels in China. Results Of 407 individual Bactrian camel fecal specimens, 30.0% (122) were E. bieneusi-positive by nested polymerase chain reaction (PCR) based on internal transcriber spacer (ITS) sequence analysis. A total of 14 distinct E. bieneusi ITS genotypes were obtained: eight known genotypes (genotype EbpC, EbpA, Henan-IV, BEB6, CM8, CHG16, O and WL17), and six novel genotypes (named CAM1 to CAM6). Genotype CAM1 (59.0%, 72/122) was the most predominant genotype in Bactrian camels in Xinjiang, and genotype EbpC (18.9%, 23/122) was the second-most predominant genotype. Phylogenetic analysis revealed that six known genotypes (EbpC, EbpA, WL17, Henan-IV, CM8 and O) and three novel genotypes (CAM3, CAM5 and CAM6) fell into the human-pathogenic group 1. Two known genotypes (CHG16 and BEB6) fell into the cattle host-specific group 2. The novel genotypes CAM1, CAM 2 and CAM4 cluster into group 8. Conclusions To our knowledge, this is the first report of E. bieneusi in Bactrian camels. The host-specific genotype CAM1 was the predominant genotype, which plays a negligible role in the zoonotic transmission of E. bieneusi. However, the second-most predominant genotype, EbpC, has greater zoonotic potential.


Background
Microsporidia are a diverse group of emerging obligate intracellular eukaryotic fungi and there are approximately 1300 microsporidian species in 160 genera [1]. To date, there are at least 14 microsporidian species reported to be infectious to humans [2]. Enterocytozoon bieneusi is the most frequently detected species in humans [3], as well as in domestic animals and wildlife [4], and even in environmental water samples [5].
More than 200 E. bieneusi genotypes have been identified in humans and animals by polymerase chain reaction (PCR) based on ribosomal internal transcribed spacer (ITS) gene sequence analysis [2,6]. Molecular phylogenetic analysis has shown that all E. bieneusi ITS genotypes are clustered into nine large groups, including the potentially zoonotic group 1, and some host-specific groups (Group 2 to Group 9) [7].
The Bactrian camel (Camelus bactrianus) was the major means of transportation on the ancient Silk Road. Today, the population of Bactrian camels in China has been estimated at 242,000, most of which are domesticated in desert and semi-desert areas of northwestern China and play an important role in the livelihood of pastoralists through providing milk and meat [8]. There are some reports of intestinal pathogen infections in camels and Bactrian camels in the Middle East countries and China, such as Eimeria spp. and Cryptosporidium spp. [9,10]. However, E. bieneusi infection has not been previously reported in Bactrian camels.
This study was undertaken to better understand the prevalence of E. bieneusi in Bactrian camels and assess the host specificity of E. bieneusi infections in Bactrian camels in China.

Specimen collection
A total of 407 individual fresh fecal specimens from Bactrian camels were collected from 18 different grazing Bactrian camel groups in 11 collection sites of Xinjiang Uygur Autonomous Region (hereinafter referred to as Xinjiang) of northwestern China. Only one specimen was collected per animal. These specimens were collected during August and September of 2013 and from July 2016 to July 2017 ( Table 1). The grazing Bactrian camel groups were kept outdoors and shared pastures with cattle, sheep, goats and wild animals, and each group had approximately 30-300 animals.
After animal defecation, about 50-100 g of each fresh specimen was collected immediately from the ground using sterile gloves. Each specimen was collected in a plastic container and marked with the specimen number and site. The specimens were transported to the laboratory and stored in 2.5% (w/v) potassium dichromate solution at 4°C before DNA extraction.

DNA extraction and PCR amplification
Approximately 200 mg of each fecal specimen was washed at least three times with distilled water by centrifugation at 5000× g for 5 min to remove the potassium dichromate. DNA was extracted using the E.Z.N.A.R® Stool DNA Kit (Omega Biotek Inc., Norcross, GA, USA) according to the manufacturer's instructions. For E. bieneusi screening, nested PCR assays were used to amplify an rRNA gene fragment containing the entire internal transcriber spacer (ITS) [6]. Each specimen was analyzed in duplicate using positive and negative controls. The secondary PCR products were examined by electrophoresis in a 1.5% agarose gel and visualized after staining with GelRed™ (Biotium Inc., Hayward, CA, USA).

Sequencing and phylogenetic analysis
The positive secondary PCR amplicons were sent to a commercial company (GENEWIZ, Suzhou, China) for sequencing. The sequence accuracy was confirmed with bidirectional sequencing, and the sequences obtained were aligned with reference sequences downloaded from Table 1 The infection status of E. bieneusi and genotypes in Bactrian camels in Xinjiang, China GenBank to determine the genotypes, using the program ClustalX 2.0 (http://www.clustal.org/). The genotypes of E. bieneusi isolated in this study were compared with known E. bieneusi ITS genotypes with a neighbor-joining analysis in the Mega 5 program [6]. A bootstrap analysis was used to assess the robustness of the clusters using 1000 replicates. The established nomenclature system was used in naming the E. bieneusi ITS genotypes [11].

Nucleotide sequence accession numbers
The nucleotide sequences reported in this paper have been submitted to the GenBank database at the National Center for Biotechnology Information under the accession numbers: MG602791-MG602796.

Statistical analysis
Chi-square test was used to compare the prevalence of E. bieneusi infections and predominant genotypes distributions. Differences were considered significant at P < 0.05.
To the best of our knowledge, this is the first report of E. bieneusi in Bactrian camels, and the pathogen is widespread in Xinjiang, northwestern China. In China, the average prevalence of E. bieneusi in animals ranges from 0.9% (4/426) in rabbits [12] to 45.6% (426/934) in pigs [13]. However, E. bieneusi infection has only been reported in some animals in northwestern China (Table  2), the average prevalence ranging from 1.1% (4/353) in white yaks [14] to 47.8% (22/46) in sheep [2]. In Xinjiang, only dairy calves [15] and grazing horses [16] have been previously reported to have E. bieneusi infections, with a prevalence of 16.5% (85/514) and 30.9% (81/262), respectively. The high prevalence in Bactrian camels found in this study may be the result of free feeding and drinking water, and mixed feeding with cattle, sheep, goats and other animals in the same pastures, and with the poor veterinary service.
The phylogenetic analysis of the ITS genotypes revealed the following clusters: Group 1, Group 2 and Group 8. The six known genotypes (EbpC, EbpA, WL17, Henan-IV, CM8 and O) and three novel genotypes (CAM3, CAM5 and CAM6) identified in this study fell into the human-pathogenic Group 1 (Fig. 1), which is the genotype of major zoonotic potential suggesting that the Bactrian camels play a potential role in E. bieneusi transmission to humans [11]. In contrast, the two known genotypes CHG16 and BEB6 fell into the cattle hostspecific Group 2. The three novel genotypes CAM1, CAM 2 and CAM4 clustered into Group 8 (69.7%, 85/ Fig. 1 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 > 50% from 1000 replicates are shown at the nodes. The genotypes detected in this study are shown with triangles; previously known genotypes observed in this study are marked with open triangles and the new genotypes are indicated by filled triangles 122) (Fig. 1), suggesting that the host-specific genotype CAM1 in Bactrian camels exhibits less zoonotic potential compared to the genotypes clustered into the humanpathogenic group.
In previous studies, E. bieneusi genotype EbpC and EbpA were reported in humans and various animals and were also the predominant genotypes in the reports of humans and pigs in China [13,17]. Genotypes EbpC and EbpA were the most common E. bieneusi genotypes in grazing horses in Xinjiang [16], and genotype EbpC was also identified in dairy calves in Xinjiang [15]. Similarly, the E. bieneusi zoonotic genotypes EbpC and EbpA were identified in Bactrian camels in the present study. However, there were no published reports of genotypes EbpC and EbpA in animals in Gansu, Ningxia, Qinghai and Shaanxi, northwestern China (Table 2). Further investigations of the epidemiology and host specificity of E. bieneusi in humans and other animals in Xinjiang might be informative.

Conclusions
The present study demonstrated a widespread occurrence of E. bieneusi in Bactrian camels in Xinjiang, China. The host-specific genotype, CAM1, was the most predominant genotype, which plays a negligible role in the zoonotic transmission of E. bieneusi. The secondmost predominant genotype, EbpC, in addition to other genotypes of zoonotic potential, was also commonly identified in Bactrian camels in this study. Bactrian camels could serve as a vector for E. bieneusi transmission to humans and other animals, and vice versa.