Multilocus genotypes and broad host-range of Enterocytozoon bieneusi in captive wildlife at zoological gardens in China

Background Enterocytozoon bieneusi is a common opportunistic pathogen that is widely detected in humans, domestic animals and wildlife, and poses a challenge to public health. The present study was performed to evaluate the prevalence, genotypic diversity and zoonotic potential of E. bieneusi among wildlife at Chengdu and Bifengxia zoological gardens in Sichuan Province, China. Results Of the 272 fresh fecal samples harvested from 70 captive wildlife species at Chengdu Zoo (n = 198) and Bifengxia Zoo (n = 74), 21 (10.6 %) and 22 (29.7 %) tested positive for E. bieneusi by internal transcribed spacer (ITS) sequencing analysis, respectively. Specifically, genotypes D, Peru 6, CHB1, BEB6, CHS9, SC02 and SC03, and genotypes D, CHB1, SC01 and SC02 were detected in the Chengdu and Bifengxia Zoo samples, respectively. Five known genotypes (D, Peru 6, BEB6, CHS9 and CHB1) and three novel genotypes (SC01, SC02 and SC03) were clustered into the zoonotic group (group 1) and host-adapted group (group 2). Multilocus sequence typing (MLST) analysis targeting three microsatellites (MS1, MS3 and MS7) and one minisatellite (MS4) were successfully sequenced for 37, 33, 35 and 37 specimens, generating 8, 3, 11 and 15 distinct locus types, respectively. Altogether, we identified 27 multilocus genotypes (MLGs) among the E. bieneusi isolates by MLST. These data highlight the high genetic diversity of E. bieneusi among zoo wildlife. Conclusions To our knowledge, this is the first report on the prevalence and genotypic diversity of E. bieneusi infections among captive wildlife in zoos in southwest China. Notably, we identified three novel E. bieneusi genotypes, as well as six new mammalian hosts (Asian golden cats, Tibetian blue bears, blackbucks, hog deer, Malayan sun bears and brown bears) for this organism. Moreover, the occurrence of zoonotic genotypes suggests that wildlife may act as reservoirs of E. bieneusi that can serve as a source of human microsporidiosis. The findings presented here should contribute to the control of zoonotic disease in China. Electronic supplementary material The online version of this article (doi:10.1186/s13071-016-1668-1) contains supplementary material, which is available to authorized users.


Background
Microsporidia, classified as fungi, are the causative agents of microsporidiosis, an important emerging infectious disease [1,2]. Among the approximately 1,300 microsporidian species identified, Enterocytozoon bieneusi is the most frequent cause of microsporidial infections in humans [3]. Enterocytozoon bieneusi is an obligate intracellular pathogen that is widely distributed in a variety of animals, including domestic animals and wildlife, and can also be found in water and contaminated food [4][5][6][7][8]. Enterocytozoon bieneusi colonizes the epithelium of the small intestine, localizing predominantly within the apical portion of the villus [9]. While microsporidiosis is typically associated with self-limiting diarrhea among healthy individuals, immunocompromised patients, particularly those suffering from AIDS, can develop life-threatening chronic diarrhea [10][11][12][13].
Due to the small size of its spores and the uncharacteristic staining properties of this organism, it is difficult to detect E. bieneusi by light microscopy [5]. As a result, molecular methods, particularly PCR-based amplification of E. bieneusi-specific sequences, are primarily utilized to detect and confirm E. bieneusi infections [6]. Currently, due to the high degree of diversity observed among E. bieneusi isolates, amplification and sequencing of the ribosomal internal transcribed spacer (ITS) is widely used to identify and genotype these strains [14]. To date, over 200 E. bieneusi genotypes, clustered into eight groups (Group 1-8), have been defined [15,16]. While the strains comprising Group 1 have been isolated from both animals and humans and are generally associated with a major zoonotic potential, those of the other groups are considered host-adapted, as they exhibit a narrow host-range and possess little to no zoonotic potential; however, these organisms remain a potential public health concern [15,17].
A wide variety of wildlife species are housed at Bifengxia Zoo and Chengdu Zoo. Indeed, Chengdu Zoo is one of the largest zoos in southwest China. Zoo animals are considered domesticated in that they have been separated from their natural habitat. Furthermore, they live under unnatural conditions and in higher densities than those observed in nature [18]. Previous studies have found Cryptosporidium andersoni in Bactrian camels and zoonotic Cryptosporidium at Bifengxia Zoo [19,20]. To protect the health of wildlife and to avoid potential public health risks, it is necessary to investigate the occurrence of E. bieneusi in captive wild animals. The aim of this study was to examine the prevalence of E. bieneusi in various wild animal species in Bifengxia Zoo and Chengdu Zoo, and to genotype the resulting E. bieneusi isolates via ITS sequencing and multilocus sequence typing (MLST) analyses. Furthermore, we assessed the zoonotic potential of each E. bieneusi strain isolated.

Sample collection and DNA extraction
A total of 272 fecal samples were obtained from wildlife in Chengdu Zoo (n = 198) and Bifengxia Zoo (n = 74), which are located in Chengdu and Ya'an, respectively, in Sichuan Provence, China, between June 2014 and September 2015. All samples were placed on ice in separate containers, and transported to the laboratory immediately. Prior to use, specimens were stored in 2.5 % potassium dichromate at 4°C in a refrigerator.
Fecal samples were washed with distilled water and centrifuge at 3,000× g for three min. This process was repeated in triplicate. Genomic DNA was then extracted from approximately 200 mg of each semi-purified product using an E.Z.N.A.® Tool DNA Kit (D4015-02; Omega Bio-Tek Inc., Norcross, GA, USA) following the manufacturer's instructions. DNA samples were stored in 200 μl of the kit Solution Buffer at -20°C until use.

Phylogenetic analyses
All nucleotide sequences obtained in this study were aligned with E. bieneusi reference sequences downloaded from the GenBank database using Blast [23] and Clus-talX software [24]. Phylogenetic analysis of ITS sequences was performed using Mega software [25], and Maximum Likelihood analysis of the aligned E. bieneusi sequences was utilized to support genotype classifications. A total of 1,000 replicates were used for bootstrap analysis.
Phylogenetic analyses based on ITS sequencing indicated that all representative isolates detected in this work belong to Group 1 or Group 2 ( Fig. 1). Specifically, isolates with the known genotypes (D and Peru 6) and the three new genotypes (SC01, SC02 and SC03) fell into Group 1, while strains with genotypes CHB1, BEB6 or CHS9 were categorized as Group 2. Moreover, the genotype D strains identified in this study clustered into Subgroup 1a, while the Peru 6, SC01 and SC02 strains and the SCO3 strains were clustered into subgroups 1b and 1 day, respectively (Fig. 1).
ITS analysis of the three novel genotypes showed genetic variability. There was a three-nucleotide difference between the ITS of SC01 strains and that of genotype CM3 (KF305589). The ITS sequences of SC02 strains isolated from Tibetian blue bears, Asiatic black bears, sun bears and northern raccoons differed from that of the CHN-DC1 genotype (KJ710333) by two SNPs. Finally, the ITS sequences harbored by strains of the newly-identified genotype SC03 contained four SNPs relative to that of genotype EbpC (KP262381).  (TTA TTT TTT CCA TTT TTC TTC TTC TAT  TTC CTT TA) (Table 4).

Discussion
The    animals) [30] and of 28.2 % and 12.3 % among free-ranging rhesus monkeys and newly captured baboons in Kenya, respectively [13,31]. The prevalence of E. bieneusi among animals of the order Artiodactyla was 4 % in Chengdu Zoo, which was similar to that observed in golden takins (4.7 %) in a prior study [1], but markedly lower than the average infection rate detected in reindeers (16.8 %) and in dairy cattle (24.3 %) [32,33].
Notably, no such animals tested at Bifengxia Zoo were infected with E. bieneusi. The results of this investigation indicate that the occurrence of E. bieneusi varies between zoos and animal species. ITS sequencing and phylogenetic analyses detected two known (D and Peru 6) and three novel E. bieneusi genotypes (SC01, SC02 and SC03) among Group 1 strains, which exhibit zoonotic potential, whereas Group 2 was comprised of the CHB1, BEB6 and CHS9 genotypes. In Chengdu Zoo, seven genotypes were detected, with the zoonotic D genotype being the most prevalent followed by Peru 6. Notably, genotype D was found in four animal species, suggesting cross-transmission between these animals. Indeed, previous reports have detected genotype D in humans and wildlife in various different countries [34][35][36][37]. Therefore, these findings indicate that zoonotic transmission to humans and between wildlife species may occur in Chengdu Zoo. In Bifengxia Zoo, four E. bieneusi genotypes were identified, with CHB1 being the predominant genotype; this genotype was especially common among black bears. Our results provide the first evidence of CHB1 infection among Malayan sun bears, red pandas, brown bears, Tibetian blue bears and ring-tailed lemurs. Additionally, a recently published study reported CHB1 infection in black bears [26]. The common existence of the CHB1 genotype among animals of the family Ursidae indicates that these animals may be more susceptible to infection by E. bieneusi than other species housed at this zoo.
In this study, E. bieneusi infection was detected in a total of 18 wildlife species. Of these, six had previously never been found to be infected with this organism, including Asiatic golden cats, Tibetian blue bears, Malayan sun bears, brown bears, blackbucks and hog deer. As such, our findings extend the known host-range for this parasite. These newly identified hosts belong to the order Artiodactyla or Carnivora, indicating that animals in these orders may be more susceptible to infection by E. bieneusi. Two Malayan sun bears and an alpaca were observed to be infected with CHB1 and BEB6, respectively. In contrast, genotype J was identified in Malayan sun bears and genotypes CHALT1 and J were detected in alpacas in Zhengzhou Zoo [26]. Similar to the results of a previous study, we detected genotype BEB6 in sika deer [38]; we also detected the newly identified genotype SC03 in these animals. Interestingly, the isolation of two novel genotypes from these deer, as well as the five novel genotypes identified by the study of Zhao et al. [38] suggest that genetic variability between deer-derived E. bieneusi may be common. While previous studies reported infections with genotypes D, Ebpc, WL1, WL2, WL3 and WL15 in raccoons [26,39], we detected only genotypes D and SC02 in these animals. However, the presence of the novel SC02 genotype indicates that raccoons likely harbor strains of E. bieneusi that have yet to be characterized. Tian et al. [27] detected the I-like and EbpC genotypes in giant pandas and red pandas, respectively. In contrast, we detected the Peru 6 genotype in giant pandas and the CHB1 genotype in red pandas. Recently, several studies have examined the prevalence and types of E. bieneusi infections among NHP species. These studies demonstrated that NHP can be infected by a wide range of genotypes, including Type IV, D, Henan V, Peru8, PigEBITS7, EbpC, WL15, LW1d, Peru11, Peru7, BEB6, I, O, EbpA, Henan-IV, BEB4, PigEBITS5, EbpD, CS-1, CM1-CM18, Macaque 1, Macaque 2 and KB1-KB6 [29][30][31]40]. Here, we further these findings by providing the first evidence that NHP can be also infected by genotype CHB1, and by demonstrating that northern white-cheeked gibbons can harbor genotype D.
MLST analyses involving the amplification and sequencing of housekeeping genes are widely used to study genetic profiles of pathogens with high resolution, sensitivity and specificity. Indeed, this method plays an important role in parasite research, including in studies of Cryptosporidium and E. bieneusi, and has been applied to the evaluation of E. bieneusi strains isolated from humans, pandas, golden takins, baboons and other NHP [1,22,27,[40][41][42][43][44][45][46]. We therefore utilized this approach to analyze 43 ITS-positive E. bieneusi wildlife-derived isolates. Our analyses indicated that these 43 strains were comprised of 27 distinct MLGs. Several Asiatic black bears at Bifengxia Zoo were infected with the same three MLGs (MLG14, MLG19 and MLG23), indicating likely transmission of E. bieneusi between these animals. Interestingly, despite belonging to distinct orders, both a ring-tailed lemur (Primates) and an Asiatic black bear (Carnivora) were infected with MLG16; however, it is unclear which animal was the source of the infection. To (See figure on previous page.) Fig. 1 Phylogenetic relationships of ITS nucleotide sequences of the Enterocytozoon bieneusi genotypes identified in this study and other reported genotypes. The phylogeny was inferred by a maximum likelihood analysis. Bootstrap values were obtained using 1,000 pseudoreplicates and greater than > 50 % was shown on nodes. The genotypes in this study are marked by empty triangles and the novel genotypes are marked by filled triangles

Conclusions
The results of our study describe the prevalence of E. bieneusi infections among captive wildlife in zoos in southwest China. Furthermore, they provide the first evidence of E. bieneusi infections in Asian golden cats, Tibetian blue bears, blackbucks, hog deer, Malayan sun bears and brown bears, thereby expanding the recognized host-range of this organism. The detection of zoonotic genotypes among various animals highlights the potential for zoonotic transmission to humans. Thus, methods for controlling this transmission are needed. Our novel E. bieneusi sequencing data will facilitate future molecular epidemiology research. However, further multi-locus genotyping analyses, involving a larger number of isolates from humans and wildlife, are needed to better assess the zoonotic potential and transmission dynamics of E. bieneusi.

Additional files
Additional file 1: Table S1. List of mammals in Chengdu Zoo examined in the present study.  Animal species infected with Enterocytozoon bieneusi reported for the first time Abbreviation: ns, not successfully sequenced or unsuccessful PCR amplification