Prevalence and genetic characteristics of Blastocystis hominis and Cystoisospora belli in HIV/AIDS patients in Guangxi Zhuang Autonomous Region, China

Blastocystis hominis and Cystoisospora belli are considered to be common opportunistic intestinal protozoa in HIV/AIDS patients. In order to investigate the prevalence and genetic characteristics of B. hominis and C. belli in HIV/AIDS patients, a total of 285 faecal samples were individually collected from HIV/AIDS patients in Guangxi, China. B. hominis and C. belli were investigated by amplifying the barcode region of the SSU rRNA gene and the internal transcribed spacer 1 (ITS-1) region of the rRNA gene, respectively. Chi-square test or Fisher’s exact test were conducted to assess the risk factors related to B. hominis and C. belli infection. The prevalence of B. hominis and C. belli was 6.0% (17/285) and 1.1% (3/285) respectively. Four genotypes of B. hominis were detected, with ST3 (n = 8) and ST1 (n = 6) being predominant, followed by ST6 (n = 2) and ST7 (n = 1). Females had a statistically higher prevalence of B. hominis (11.6%) than males (4.2%). The statistical analysis also showed that the prevalence of B. hominis was significantly associated with age group and educational level. Our study provides convincing evidence for the genetic diversity of B. hominis, which indicates its potential zoonotic transmission and is the first report on the molecular characteristics of C. belli in HIV/AIDS patients in China.

www.nature.com/scientificreports/ least 26 genotypes, of which 10 genotypes, namely ST1-ST9 and ST12, have been found in humans [12][13][14] . ST3 is reportedly the most common genotype that causes gastrointestinal symptoms 13 . Although the prevalence of B. hominis in HIV/AIDS patients has been reported worldwide, the genetic diversity of B. hominis in HIV/AIDS patients is rarely documented by empirical data. So far, just three studies of B. hominis in HIV/AIDS patients have been carried out in Anhui, Hunan, and Yunnan provinces, China, for which the prevalence of B. hominis was 16.2%, 9.9%, and 3.7%, respectively 14 . Cystoisospora belli (formerly Isospora belli), originally described by Zaman in 1968 15 , also has a global distribution but it is more common in tropical and subtropical countries, such as India, Nigeria, and Ethiopia 16 . Importantly, C. belli has been frequently detected in patients with HIV, Human T-lymphotropic Virus Type I (HLTLV1), Alzheimer's disease, and colorectal cancer, as well as those having undergone liver and renal transplantation [17][18][19][20] . In those HIV/AIDS patients with diarrhea symptoms, the prevalence of C. belli could reach up to 20% 21 . Although it occurs worldwide, C. belli is a generally overlooked parasite, and one not well studied in China. Hence, further study of this parasite is imperative.
Several cluster studies of B. hominis and C. belli in HIV/AIDS patients have been conducted, especially in low-and middle-income countries 16,22,23 . But, surprisingly, comparable reports are quite limited for China, especially for C. belli. To our best knowledge, the molecular study presented here is the first on this parasite in China. Guangxi Zhuang Autonomous Region (hereon Guangxi), a provincial-level autonomous region, is located in southern China and borders Vietnam in the southwest. It has a subtropical monsoon climate with abundant rainfall and light. In 2020, the total population of the registered residence in Guangxi was ~ 50 million. It has been reported that the prevalence of HIV/AIDS in Guangxi was 6.6% in 2013, 8.4% in 2014, and 11.2% in 2015, all of which were higher than the corresponding numbers for the general population in China (0.05%) 24,25 . Furthermore, by the end of 2017, Guangxi had reported 124,282 cases of HIV/AIDS, indicating an increase of 78.7% since June 2011 (69,548 cases) and ranking it as second in terms of HIV seropositive cases among the 31 provinces of China 26,27 . Since the Guangxi leads this country in the number of HIV/AIDS patients, investigating their intestinal parasites is of great public health significance 28 . In our previous research, Microsporidia, Cryptosporidium, and Giardia were respectively detected with prevalence of 11.6%, 0.7%, and 2.8% in those participants 5,29,30 . Both B. hominis and C. belli are recognized as the pathogens that most often cause opportunistic infections in HIV/AIDS patients, posing a prominent threat to the public health 13,16 . Accordingly, the present study aimed to assess the prevalence in addition to genetic characteristics and risk factors of B. hominis and C. belli in HIV/AIDS patients of Guangxi, China.

Methods
Ethics approval and consent to participate. The present study obtained the approval from the Ethics Committee of the National Institute of Parasitic Diseases, Chinese Centre for Disease Control and Prevention (reference no. 2012-12) and all methods were conducted in accordance with the relevant guidelines and regulations as provided in the Declaration of Helsinki. All the patients enrolled in the study were given an oral explanation of the objectives, procedures and potential risks for collection of their faecal samples. Adult participants signed the written informed consent personally. All participants were over the age of 20, so no parental consent was required.
Sample size calculation. Due to the lack of prevalence of C. belli in China, the sample size was calculated based on the prevalence of B. hominis. The required sample sized was determined using the formula 31 , where α = 0.05, δ = 0.05, and p = the estimated prevalence of B. hominis among HIV/AIDS patients, which was taken as 16.2% 32 . The resulting sample size was 209 patients. Considering that an estimated 10% of the patients might fail to participate in the study, the final sample size was increased to 230 patients. Finally, a total of 285 HIV/AIDS patients registered in nine hospitals in Guangxi were enrolled into the present study.
Questionnaire, faecal sample collection, and processing. From July 2013 to July 2014, fresh faecal samples (one per patient) were collected, with an in-person, structured questionnaire survey conducted faceto-face to collect the patients' demographic information (gender, age, education level, occupation), behavioral information (whether drinking boiled water or not, whether receiving HAART treatment or not), HIV transmission route, clinical symptoms (diarrhea), and CD4 + cell count.
The collected fresh faecal samples (≥ 200 mg) were first stored at 4 °C with 2.5% potassium dichromate, and then sent to the laboratory of National Institute of Parasitic Diseases, Chinese Centre for Disease Control and Prevention. There, all samples were washed three times with deionized water at 14 000 rpm for 10 min, to remove the potassium dichromate. Then, DNA was extracted from 180 to 220 mg of each faecal sample, by using QIAamp DNA Stool Mini Kit (Qiagen, Hilden, Germany) and following the manufacturer's instructions. For a higher DNA yield, the lysis temperature was adjusted to 95 °C according to the manufacturer's recommendation. The final amount of DNA was 200 μl per sample, stored at -30 °C until the PCR analysis.
Detection of B. hominis and C. belli. PCR was used to amplify the barcode region of the SSU rRNA gene of B. hominis, using primer sets as described by Scicluna 33 , while C. belli was detected with the primer sets based on the ITS-1 region of the rRNA gene designed by Reza et al. 17 . All DNA samples were analyzed at least three times. Either B. hominis-positive DNA or C. belli-positive DNA and nuclease-free water served as the positive and negative controls, respectively. The PCR products (5 μl) were checked by 2% gel electrophoresis, and the products of an expected size (approximately 600 bp for B. hominis and 450 bp for C. belli) were analyzed using an ABI 3730 DNA Analyzer and Big Dye Terminator v3. Phylogenetic and statistical analysis. Phylogenetic trees for B. hominis and C. belli were constructed in MEGA 6.0 (http:// mega. softw are. net), based on 1000 bootstrap replicates, using the sequences obtained in this study and representative sequences downloaded from NCBI. SPSS V20.0 software (https:// spss. en. softo nic. com/) was used to statistically analyze the data. Chi-square test or Fisher's exact test was implemented to compare the prevalence between groups classified by gender, age, education level, occupation, whether drinking boiled water or not, whether receiving HAART treatment or not, HIV transmission route, diarrhea, and CD4 + cell count. Differences were regarded as statistically significant at P < 0.05.
The prevalence of C. belli was 1.1% (3/285), and all the three infected individuals were males. Moreover, coinfection of C. belli with B. hominis was observed in a farmer aged 50 who had bacterial pneumonia.
Analysis of risk factors of B. hominis and C. belli. All 285 participants properly completed the survey questionnaire. They were divided into different groups by gender, age, education level, occupation, whether drinking boiled water or not, whether receiving HAART treatment or not, HIV transmission route, diarrhea, Univariate analysis showed a statistically significant correlation between the prevalence of C. belli and different education levels (P = 0.002). No significant differences, however, were found between groups formed by other risk factors (all P-values > 0.05; Table 1).
The sequences of the three C. belli isolates showed 100% homology with isolates (HM630352) previously identified from an HIV/AIDS patient in Argentina (Fig. 2).

Discussion
Blastocystis hominis and C. belli are common, opportunistic intestinal protozoa in immunocompromised individuals, especially in HIV/AIDS patients . We determined that the overall prevalence of both species was 6.7% (19/285), with B. hominis at least fivefold more prevalent (6.0%, 17/285) than C. belli (1.1%, 3/285). As far as we know, this is the first molecular investigation of B. hominis in HIV/AIDS patients in Guangxi, China, and of C. belli in China.
Blastocystis hominis is found globally, but its prevalence in HIV/AIDS patients among countries varies tremendously, from 0.9 to 72.4% 4 . A higher prevalence that reached 88.7% was observed in indigenous children from the Colombian Amazon Basin 34 . Due to the controversial pathogenicity of B. hominis, less attention has been paid to it than to Cryptosporidium, Giardia and Microsporidia in China. Nevertheless, a waterborne outbreak of B. hominis in China that afflicted 1122 people highlighted the public health significance of this intestinal parasite 35 . Therefore, detection of B. hominis is imperative in China. Generally, few studies of B. hominis relying on molecular approaches have been conducted in China. Genotypes of B. hominis were analyzed in seven provinces, cities or autonomous regions-Heilongjiang, Guangxi, Yunnan, Zhejiang, Shanghai, Xinjiang and Chongqing-among different population groups, for which its prevalence ranged from 3.7% to 36.6% 14,[36][37][38][39][40][41][42][43][44][45][46][47] (Table 3). We found an occurrence of B. hominis in HIV/AIDS patients of 6.0%, lower than that in Fuyang, Anhui Province (16.2%) 48 , yet slightly higher than that in Tengchong, Yunnan Province (3.7%) 14 . In other countries, a higher prevalence of B. hominis in HIV/AIDS patients was reported for Peru (24.6%) 49 , Rome (25.0%) 50 , Bogota (25.2%) 51 and Indonesia (72.4%) 4 , whereas lower values characterized Nepal (0.9%) 52 and India (3.3%) 53 . Such disparate infection rates may arise from multiple factors, including the research scope, the selected subjects, their living environment as well as local social, economic, and cultural conditions. For example, low socio-economic conditions coupled to a humid and high temperature climate can increase the transmission of B. hominis 54 . In addition, diagnostic methods differ in their sensitivity and specificity in detecting B. hominis. Molecular approaches based on PCR were shown to be superior to traditional microscopic examination of faecal samples 55 . In southwestern Iran, B. hominis occurred in 12.3% and 19.0% of humans according to the Formol-Ether technique and PCR, respectively 56 . In another study of newly arrived immigrants in Qatar, the prevalence of B. hominis was 7.6% when assessed by coproscopic methods but much higher (65.2%) by RT-PCR 57 . In other work, in HIV-infected patients, microscopic examination suggested a B. hominis prevalence of 21.8% but according to PCR it was 25.0% 50 .  48 , we found that women were more likely to get infected with B. hominis than men. Differences in nutritional status and access to medical facilities between men and women may be responsible for the increased risk of parasite infection among women 58 . Individuals with an education level of primary schools and below have the highest infection rate of B. hominis in Guangxi, hence a limited education and poor awareness of hygiene may contribute to the high infection rate in specific groups. Prevalence in patients grouped by occupation, CD4 + cell count, HIV transmission route, and whether they were receiving HAART treatment or not, all showed no statistical differences. Previous studies suggested that people with a CD4 + cell count less than 200 are more susceptible to B. hominis 56 , and that drinking raw (untreated) water can increase its infection risk 45 . Moreover, three waterborne outbreaks of B. hominis have been reported 22,35 , but in the present study the prevalence was similar between boiled water drinkers and raw water drinkers. This result may have two explanations. First, there is no B. hominis in the water of this area; second, the ratio of boiled water drinkers was high, accounting for 93.3%.   61 . Although ST3 was the predominant genotype identified in our study, no statistical association was observed between B. hominis infection and diarrhea.
Presently, B. hominis is known to harbour at least 26 genotypes 11,12 . Ten genotypes (ST1-ST9 and ST12) were found in humans, among which ST3 was the most common 37 . The ST1-ST8 have been identified both in humans and various animals, highlighting their capability for zoonotic transmission [12][13][14] . We know of nine genotypes of B. hominis detected so far in HIV/AIDS patients worldwide: ST1-ST7, ST9 and ST12 14,36,45,50,56,[62][63][64] (Table 4). In our study, four genotypes of B. hominis were identified-ST1, ST3, ST6 and ST7-with ST3 the dominant one (47.1%; 8/17), a result consistent with that of studies in HIV/AIDS patients in southwestern Iran and cancer patients in China 37,56 . All the genotypes have zoonotic potential according to the latest research: both ST1 and ST3 are considered the most abundant genotypes 13,56 and are widely associated with animals, such as non-human primates, dogs and pigs 65,66 . ST6, detected in indigenous children in the Colombian Amazon Basin 34 , and ST7, detected in HIV/AIDS patients in China's Yunnan Province and in Iran 14,64 , have been identified from cattle, rodents, birds, chickens, and pigs 65,66 . Additionally, B. hominis is viewed as a waterborne pathogen, given its identification in different water bodies including river, wastewater, drinking water, school ponds, and canal water and the three waterborne outbreaks of B. hominis infection documented in Italy, China, and Nepal that drew garnered attention to this parasite 22 . Hence, future research should focus on investigating the prevalence and genetic diversity of B. hominis in both animals and water in our study area to better understand its transmission dynamics.
The prevalence of C. belli in HIV/AIDS patients varies from country to country, even so within the different areas of the same region 16 . Infection by C. belli happens worldwide, especially in tropical and subtropical regions such as India, Iran, Brazil and Ethiopia, but mostly these were sporadic cases or small aggregate outbreaks 16,17,21 . Here, in Guangxi, C. belli was detected in three participants, at a prevalence of 1.1%, which is close to that reported for HIV/AIDS patients in Ethiopia (1.3%) 67 , but lower than that of HIV/AIDS patients in India (16.5%) 68 and higher than of patients in Burkina Faso (0.7%) 69 . Severe diarrhea due to C. belli infection has been reported in immunocompromised patients 17,70 . Contrary to our expectation, the three C. belli-infected individuals all reported no diarrhea symptoms. In this study, non-diarrheal individuals accounted for 89.8% (256/285) of the sampled population, which may induce the low prevalence, given that patients with diarrhea often show a high prevalence of C. belli (25.0%) 71 .
Here we found that people with a primary school and below education level were more prone to infection by this protist. Yet no statistical link was discernable between its prevalence and the many other risk factors examined. Similarly, a study in Nigeria also could not link isosporiasis to either the gender or age of hosts 72 .

Conclusions
The present study revealed the prevalence, genetic characteristics, and risk factors of B. hominis and C. belli in HIV/AIDS patients in Guangxi, China. Our data indicated the zoonotic transmission of B. hominis in this area and relationships between B. hominis infection and gender, age, and educational level. Furthermore, our study provided new insight into the molecular characteristics of C. belli, providing the first molecular data for this parasite in China. Future studies should be focus on investigating B. hominis and C. belli in animals and water in the Guangxi area to figure out the sources and routes of these protozoans' transmission.

Data availability
The datasets generated and/or analysed during the current study are not publicly available in order to protect participant confidentiality. The gene sequences identified in this study were submitted to GenBank with accession numbers MK922974, MK930346, MK930362, MK934329-MK934342, and MK935345-MK935347. Table 4. Genotype distribution of B. hominis in HIV/AIDS patients worldwide.

Country
No. examined Infection rate(%) Genotype References