Infection by and genotype characteristics of Enterocytozoon bieneusi in HIV/AIDS patients from Guangxi Zhuang autonomous region, China

Background Enterocytozoon bieneusi has been increasingly reported to infect humans and various mammals. Microsporidia cause diarrhea in HIV-infected patients worldwide. PCR amplification and sequencing based on the internal transcribed spacer region have been used to describe the genotypes of E. bieneusi and transmission of microsporidiosis. Methods In this study, we examined E. bieneusi infection and genotypes in HIV-positive patients in Guangxi, China. Stool specimens were collected from 285 HIV-positive patients and 303 HIV-negative individuals. E. bieneusi genotypes were characterized using nested PCR and sequencing. Results Thirty-three (11.58%) HIV-positive patients were infected with microsporidia, and no infection was found in the 303 healthy controls. Three new genotypes were identified and named as GX25, GX456, and GX458; four known genotypes, PigEBITS7, Type IV/K, D, and Ebpc, were also identified. Our data showed that the positive rate for microsporidia was significantly higher in the rural patients than in the other occupation groups. In addition, the positive rate for microsporidia was significantly higher in the patients who drink unboiled water than in those with other drinking water sources. Conclusions Our results will provide baseline data for preventing and controlling E. bieneusi infection in HIV/AIDS patients. Further studies are required to clarify the epidemiology and potential sources of microsporidia. Our study showed that microsporidium infection occurs in the HIV/AIDS patients in Guangxi, China. Electronic supplementary material The online version of this article (10.1186/s12879-017-2787-9) contains supplementary material, which is available to authorized users.


Background
Gastrointestinal infection is a major opportunistic infection in HIV/AIDS patients, and many studies have reported HIV/AIDS patients co-infected with microsporidia. Microsporidia are obligate intracellular parasites that infect a broad range of vertebrates and invertebrates [1][2][3]. They have been increasingly recognized as human pathogens in AIDS patients, and they are mainly associated with life-threatening chronic diarrhea and systemic disease [4,5]. In 1959, the first human case of microsporidiosis was detected, and reports of immunocompromised patients infected by microsporidia have increased [1,6,7]. Among the microsporidial species, Enterocytozoon bieneusi is the most prevalent human pathogenic species [8]. The infection rate of E. bieneusi among HIV patients has been reported to reach up to 50% [9]. Transmission of E. bieneusi may involve personto-person as well as environmental sources, such as ditch water, especially in developing countries with poor sanitation [10,11]. In addition, zoonotic transmission of E. bieneusi has been reported worldwide in various mammal hosts, such as livestock, companion animals, birds, and wildlife. Other routes including waterborn, respiratory or sexual infection have also been reported [12][13][14][15][16].
Considerable genetic variation and genotypes exist within E. bieneusi isolates of human and animal origin, and different pathogenic characteristics and host specificity have been found for E. bieneusi [3]. Molecular diagnostic methods, especially methods that genotype and subtype pathogens, have been used to characterize the transmission of E. bieneusi in HIV patients [17][18][19]. The internal transcribed spacer (ITS) region of the rRNA gene has been extensively used to identify and describe the genotype characteristics and transmission routes of E. bieneusi in humans and animals [20,21]. To date, more than 204 ITS genotypes have been reported by genotyping analysis, and all the ITS genotypes have been divided into zoonotic (Group 1) and host-specific groups (Groups 2-8) by phylogenetic analysis [22]. Group 1 infects humans and animals, while the other groups are found mostly in specific hosts and wastewater [13,15,23]. The presence of the same genotypes of E. bieneusi in both humans and animals indicates potential zoonotic transmission [18,24]. The molecular epidemiologic characterization of E. bieneusi has become essential, to predict possible sources of transmission and control the transmission routes.
E. bieneusi infection is responsible for 30%-51% of all cases of diarrhea in patients with AIDS [25]. In fact, E. bieneusi has been detected in 11.4% and 18.5% of nonhuman primates in Guangxi, and various zoonotic genotypes were identified [14,26]. Hence, humans, especially HIV patients in Guangxi, could face the risk of E. bieneusi infection. To date, no studies have been conducted to describe the E. bieneusi infection in HIV or diarrheal patients in Guangxi. In the present study, we aimed to identify the prevalence and genotypes of E. bieneusi in HIV-infected patients and case controls in Guangxi and compare the differences between the two groups by using。PCR and sequence analysis of the ITS locus. In addition, we evaluated the public health significance of E. bieneusi via phylogenetic analysis and analyzed the risk factors for E. bieneusi in the HIV-infected patients on the basis of demographic and clinical data.

Study population
Between July 2013 and July 2014, stool specimens were collected from 285 HIV-positive patients in Guangxi. Among the patients, 216 (75.8%) were males and 69 (24.2%) were females. Most (76.1%) of the participants were farmers and live in rural areas. Demographic data, education level, presence of diarrhea, infective routes, recent CD4 + cell counts, and potential risk factors related to waterborne and person-to-person routes and marital status were collected from the participants by attending physicians by using a structured questionnaire at the time of enrollment. The demographic data of the two groups are listed in Table 1. In addition, 303

Specimen collection and DNA extraction
The fecal specimens were preserved in 2.5% potassium dichromate and stored at 4°C. Aliquots of the stool specimens were shipped to the laboratory. The specimens were collected from patients with fecal excretion heavier than 200 mg and no less than three events of diarrhea per day. Sufficient samples were collected for DNA extraction and purification with the QIAamp DNA Stool Mini Kit (Qiagen, Hilden, Germany). The extracted DNA was stored at −30°C for PCR and was used for E. bieneusi detection and genotyping.

E. bieneusi detection and genotyping
To detect E. bieneusi, a 392-bp fragment of the rRNA gene, including ITS, was amplified using nested PCR [27]. Primers used for PCR amplification of ITS gene were listed in Table 2.The amplified fragments were analyzed using agarose gel electrophoresis, and the positive samples were used for sequencing. Genotypes of E. bieneusi were determined using sequence analysis of the secondary PCR products and named according to the established nomenclature system. The cycling conditions for E. bieneusi were as follows: the primary cycle consisted of 94°C for 1 min, 35 cycles of 94°C for 50 s, 56°C for 30 s and 72°C for 60 s, followed by 72°C for 10 min, and termination at 4°C . A second reaction was carried out similarly. Each specimen was analyzed at least three times by PCR with E. bieneusi-positive sample as positive control and nucleasefree water as negative controls in each run, respectively.

DNA sequencing and data analysis
For accurate analysis, all of the genes were amplified at least three times and all PCR-positive products were sequenced in both directions by using the secondary primers with the ABI 3730 DNA Analyzer (Applied Biosystems, Foster City, USA) and Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems). Conti-gExpress was used to evaluate the wave peak and assemble the sequences. All the nucleotide sequences obtained in the present study were searched using the Basic Local Alignment Search Tool, aligned with E. bieneusi reference sequences downloaded from GenBank, and analyzed using Clustal X 1.83, MEGA 5 (http://www.megasoftware.net; last accessed in November 2012). Bootstrap analysis with 1000 replicates was used to assess the robustness of the clusters. The chi-square test was used for comparisons.

Nucleotide sequence accession numbers
The relationship between the E. bieneusi genotypes identified in this study and other known genotypes deposited in GenBank was inferred using neighbor-joining analysis of the ITS sequences on the basis of genetic distance by the Kimura two-parameter model. The numbers on the branches are percent bootstrapping values from 1000 replicates. Each sequence was identified by its accession number, host origin, and genotype designation.

Infection rates of E. bieneusi in the participants
Of the 285 fecal specimens from the HIV-positive patients, 33 specimens showed positive results for E. bieneusi after PCR amplification of the ITS locus. The infection rate of E. bieneusi was 11.6% (33/285) in the HIV-positive patients, and E. bieneusi was not found in the HIV-negative patients ( Table 1). The differences in the infection rates between the HIV-positive and HIVnegative patients were statistically significant different (χ 2 = 37.17, P < 0.01). No age-or sex-associated differences were found in the patients of our study.

Risk factors for microsporidiosis
In the present study, a number of risk factors related to E. bieneusi infection were analyzed, such as gender, age, occupation, water sources, CD4 + cell count, marital status, transmission route, and other risk factors ( Table 1). The statistical analysis showed that microsporidium infection was significantly associated with the different occupations of the patients. Farmers showed a higher occurrence of microsporidium infection (14.3%, χ 2 = 6.366, P < 0.01) than the other groups with different occupations. In addition, patients who drank unboiled water were more likely to be infected with microsporidia.

Genotypes of E. bieneusi
A total of seven ITS genotypes were obtained from 33 successfully sequenced specimens from the HIV-positive patients. Of them, four genotypes have been previously reported, namely, genotype D (11 cases), type IV/K (seven cases), PigEBITS7 (seven cases), and EbpC (four cases) ( Table 3). Three new genotypes were found and named as GX25 (one case), GX456 (one case), and GX458 (one case).

Phylogenetic analysis
Phylogenetic analysis was performed to understand the genetic relationship among the E. bieneusi genotypes. A neighbor-joining tree was constructed using the published E. bieneusi ITS nucleotide sequences from humans and domestic animals. These new genotypes were phylogenetically related to Group 1, which contains most of the human pathogenic E. bieneusi genotypes (Fig. 1).

Discussion
PCR and sequence analysis of the ribosomal ITS are regarded as the standard diagnostic technique for identifying and genotyping E. bieneusi isolates [24]. To date, the infection rate among HIV-infected patients has been reported to reach up to 50% [28], and E. bieneusi causes chronic diarrhea in the patients. In this study, we investigated the prevalence of E. bieneusi infection in HIVinfected patients and HIV-negative controls in Guangxi. High prevalence (11.6%, 33/285) of E. bieneusi was observed in 285 HIV-positive patients, and E. bieneusi was not found in the HIV-negative controls. A previous study conducted on E. bieneusi infection in HIV-positive patients in Henan Province showed that the infection rate was 5.7% (39/683) [29]. The difference in the infection rates between the two provinces in China might be attributed to the overall sample size, composition, and health status of the patients, as well as geographical location. In addition, all of the patients in Wang's study received highly active antiretroviral therapy (HAART), while just fewer than half the patients in this study are receiving HAART. In fact, HAART has been reported to reduce the prevalence of microsporidiosis in HIV/AIDS patients in industrialized nations [30,31].
In the present study, farmers showed a higher occurrence of microsporidium infection (P < 0.01) than the other groups with different occupations. The possible reasons could be the following risk factors: First, the living environment and health conditions in farms are poor when compared with those of the other populations with different occupations. In the countryside, the water used to flush toilets is usually not treated, and many people in these localities do not wash their hands after using the toilet [32]. Therefore, the patients could be infected through fecal-oral transmission. Farmers also have a variety of drinking water sources, such as tap water and pump water, and they provide transmission routes for microsporidia. In fact, a study conducted on the prevalence of intestinal parasitic infections among 463 HIV patients in Benin City, Nigeria, showed that HIV patients who used streams and rivers as sources of water exhibited a significantly higher prevalence of microsporidial infections (P = 0.011) [32]. In this study, the patients who drink unboiled water showed a higher microsporidium infection rate (χ 2 = 4.282, P < 0.05) than the other patients. Drinking unboiled water was identified as a risk factor for E. bieneusi infection in the present study, which is consistent with the relatively high occurrence of microsporidia in the farmers.
E. bieneusi is a major human pathogen associated with chronic diarrhea in HIV-infected patients [33][34][35]. In a cross-sectional study of zoonotic E. bieneusi genotypes in HIV-positive patients on antiretroviral therapy, E. bieneusi infection was significantly associated with the occurrence of diarrhea [29]. However, there was no correlation between E. bieneusi infection and the clinical symptoms of the HIV-positive patients, which could be mostly attributed to the immune status of the patients and sampling time. In fact, previous studies have found no association between the intensity of microsporidium infection and clinical symptoms [36,37]. In our study, some other risk factors (age, gender, CD4 + level, etc.) and clinical manifestations (diarrhea, white blood cell level, etc.) were also analyzed. However, no correlation was found between these risk factors and E. bieneusi infection. Although E. bieneusi is nowadays considered to be an opportunistic pathogen in HIV-infected patients or organ transplant recipients, E. bieneusi infections have been found in HIV-negative, immunocompetent, and other healthy people [38][39][40][41]. In our previous study, E. bieneusi was detected using nested PCR in 34 (13.49%) fecal samples from patients with clinical diarrhea in Shanghai [42]. Therefore, detection of E. bieneusi is absolutely imperative for HIV-infected patients and individuals with clinical diarrhea.
In this study, three new genotypes and four known E. bieneusi genotypes were identified. The new genotypes, namely, GX25 (one case), GX456 (one case), and GX458 (one case), are phylogenetically related to Group 1, which contains most of the human pathogenic E. bieneusi genotypes. Sequence alignment and phylogenetic analysis of the E. bieneusi isolates on the basis of sequences of the ITS region revealed that the three new genotypes have a high homology with the isolates from pigs (AF135832) [43,44], indicating their public health significance. The prevalent genotypes were D (11 cases), type IV/K (seven cases), PigEBITS7 (seven cases), and EbpC (four cases). The most frequently observed genotype, D (n = 11), has a large variety of hosts and geographic range. It was first detected in humans in Germany then in American, Asian, and African countries [33,[45][46][47][48][49][50][51][52][53][54]. In fact, genotype D has been identified in HIV patients [29], animals [15,26], and wastewater [13] in China. Type IV/K has been detected in HIV patients and non-human primates in Henan Province [29] and cats and dogs in Heilongjiang Province [14]. PigE-BITS7, previously found in only pigs [55], has been found in humans [29,56] and monkeys [15]. EbpC has been detected in HIV-positive and HIV-negative patients [29], pigs [57], and wastewater [13] in China ( Table 4). The occurrence of the above-mentioned ITS genotypes in the HIV-positive patients of our study suggest the possibility of zoonotic transmission. This is also supported by the fact that genotype D has been detected in animals in Guangxi [15], and further molecular studies with a large sample size and extensive epidemiological information on humans, animals, and water sources are required to better explain the zoonotic transmission of microsporidiosis.

Conclusions
In summary, our study showed the occurrence of microsporidium infection in HIV/AIDS patients in Guangxi,

Authors' contributions
Note: a The sample sizes were not mentioned in the study