Alterations in children’s sub-dominant gut microbiota by HIV infection and anti-retroviral therapy

Objective We investigated the impact of human immunodeficiency virus (HIV) infection and anti-retroviral therapy (ART) on the gut microbiota of children. Design This cross-sectional study investigated the gut microbiota of children with and without HIV. Methods We collected fecal samples from 59 children with HIV (29 treated with ART [ART(+)] and 30 without ART [HIV(+)]) and 20 children without HIV [HIV(–)] in Vietnam. We performed quantitative RT-PCR to detect 14 representative intestinal bacteria targeting 16S/23S rRNA molecules. We also collected the blood samples for immunological analyses. Results In spearman’s correlation analyses, no significant correlation between the number of dominant bacteria and age was found among children in the HIV(−) group. However, the number of sub-dominant bacteria, including Streptococcus, Enterococcus, and Enterobacteriaceae, positively correlated with age in the HIV(−) group, but not in the HIV(+) group. In the HIV(+) group, Clostridium coccoides group positively associated with the CD4+ cell count and its subsets. In the ART(+) group, Staphylococcus and C. perfringens positively correlated with CD4+ cells and their subsets and negatively with activated CD8+ cells. C. coccoides group and Bacteroides fragilis group were associated with regulatory T-cell counts. In multiple linear regression analyses, ART duration was independently associated with the number of C. perfringens, and Th17 cell count with the number of Staphylococcus in the ART(+) group. Conclusions HIV infection and ART may influence sub-dominant gut bacteria, directly or indirectly, in association with immune status in children with HIV.


Introduction
The gut microbiota comprises approximately 100 trillion microbes from more than 1000 bacterial species [1,2]. The gut microbiota plays a major role in nutrient absorption, food metabolism, intestinal barrier protection from pathogens, and the modulation of gut immune function [3][4][5]. Although the composition of the gut microbiota may be influenced by age, diet, genetics, and geography, four phyla (i.e., Firmicutes, Bacteroidetes, Actinobacteria, and Proteobacteria) are dominant and stable in healthy individuals [4,6,7]. CD4 + T cells and their subsets, such as type 1 helper T cells (Th1), Th2, Th17, and regulatory T (Treg) cells, have been associated with the gut microbiota, and their interactions are associated with various diseases, such as inflammatory bowel disease, rheumatoid arthritis, and cancer [8,9].
Gut-associated lymphoid tissue (GALT) is the largest replication site, and it serves as a reservoir of human immunodeficiency virus type 1 (HIV) [10][11][12]. Progressive HIV infections are characterized by a depletion of CD4 + T cells in the GALT, followed by microbial translocation, gut microbiota dysbiosis, and chronic immune activation [11,[13][14][15][16][17]. Despite the sustained viral suppression and immune recovery provided by anti-retroviral therapy (ART), the imbalance in gut microbiota is, at best, only partially restored for a long time after initiating ART in adults [15,16,18].
In the gut microbiota of healthy children, the dominant phyla are Bacteroidetes and Actinobacteria, particularly the Bifidobacterium genus of Actinobacteria. These bacteria have a functional composition similar to that of healthy adults [7,[19][20][21][22]. A few studies from Africa and India have shown reduced bacterial diversity in the gut microbiota of children with HIV and children treated with ART compared to the microbiota of children without HIV [23][24][25]. However, no consensus exists on whether ART in children with HIV may restore the gut microbiota to the state observed in children without HIV [23][24][25]. The impacts of HIV and ART on the gut microbiota in children remain poorly understood.
In Vietnam, no study has focused on understanding the gut microbiota in children with HIV. Therefore, the current study aimed to investigate the impact of HIV infection and ART on the gut microbiota among children in Vietnam.

Study population
This non-randomized, cross-sectional study was conducted at the Vietnam National Children's Hospital (VNCH) in Hanoi, Vietnam, in 2012 [26,27]. Children with HIV who did not start ART [HIV(+) group], children with HIV who received ART [ART(+) group], and children without HIV infection [HIV(−) group, control] were recruited.
The inclusion criteria for children with HIV were followed at the VNCH and �2 years old. Exclusion criteria were progression of HIV to acquired immunodeficiency syndrome (AIDS), treatment with anything that may influence the immune system, any antibiotics except cotrimoxazole, hospitalization in the prior 8 weeks, or symptoms of gastrointestinal infection, such as diarrhea, nausea, and vomiting, with fever, at the time of recruitment. The children in the ART(+) group resided at an orphanage near Hanoi. The children in the HIV(+) group resided at their own homes. The children in the HIV(−) group resided at a different orphanage in Hanoi [26,27].

Collection and preparation of fecal samples
Immediately after defecation, fecal samples were collected in a plastic container (Sarstedt AG & Co., Nümbrecht, Germany), kept at 4˚C, and transferred to the laboratory using containers maintained at 4˚C. At the laboratory, the fecal samples were weighed, suspended in RNA-stabilizing solution (RNAlater; Ambion, Inc., Austin, TX, USA), and homogenized (20 mg of feces/mL). The fecal homogenate (200 μL) was added to 1 mL of Dulbecco's Phospahte Buffered Saline (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan). After centrifuging the mixture at 12,000 × g for 5 min, the pellet was stored at −80˚C until used for RNA extraction. The whole process was completed within 24 hours after defecation [28].

Quantification of bacteria in human feces by RT-qPCR
Total RNA extraction and subsequent reverse-transcription and quantitative polymerase chain reaction (RT-qPCR) assays were performed using the methods described by Matsuda et al. [29,30]. Briefly, 4 mg of feces were subjected to total RNA extraction, and each extracted RNA sample was serially diluted 10-fold. Three serial dilutions of the extracted RNA samples (corresponding to 1/400, 1/4,000, and 1/40,000 of the extracted RNA) were subjected to RT-qPCR with specific primer sets that targeted the 16S or 23S rRNA of the 14 representative intestinal bacteria in four main phyla, including: Firmicutes, such as Clostridium coccoides group, C. leptum subgroup, C. difficile, C. perfringens, Lactobacillus spp., Streptococcus, Enterococcus, and Staphylococcus; Actinobacteria, such as Bifidobacterium and Atopobium cluster; Bacteroidetes, such as Bacteroides fragilis group and Prevotella; and Proteobacteria, such as Enterobacteriaceae and Pseudomonas [29][30][31][32]. The counts of Lactobacillus spp. obtained with RT-qPCR were expressed as the sum of six Lactobacillus subgroups and two species. In the same experiment, a standard curve was generated with the RT-qPCR data (by threshold cycle: C T value) and the cell counts (by DAPI staining) of the dilution series of total RNA from the standard strain for each bacterial target. The C T values from fecal samples were normalized to the standard curve to obtain the bacterial count per gram wet weight of feces.
In addition, the individual bacteria in the fecal microbiota are present at different microbial cell counts. Previous reports revealed that the average total bacterial count is approximately 10 11 cells/g of feces [29,31]. We regarded the threshold for dominance in abundance at 1.0% of the total bacterial count, and the threshold in counts was set at 10 9 cells/g [6,[33][34][35].

Laboratory methods
White blood cell (WBC) counts, WBC differentiation, hemoglobin level (Beckman Coulter, Lh 780, USA), total cholesterol level, and fasting blood sugar (Olympus AU640, Japan) were measured at the clinical laboratory of VNCH. Plasma HIV viral loads were measured as described previously [26].

Statistical analysis
Statistical analyses were performed using SPSS version 25 (IBM SPSS Statistics, USA) and R version 3.6.2 [39]. The chi-squared test or Fisher's exact test was performed to assess the differences in bacterial detection rates. The gut microbiota patterns were presented as biplots with the principal component analysis (PCA) using the prcomp function from the ggbiplot package in R. The number of bacteria was compared between the groups using the Mann-Whitney U test. Spearman's rank test was used to analyze the pairwise correlations between bacteria and possibly related factors, such as age, ART duration, CD4 + cells and their subsets, CD8 + cells, the proportion of activated CD8 + cells, and the use of cotrimoxazole. The correlations were visualized as a heatmap using the corrplot function in R. Simple linear regression was used to assess the linear relationship of the significantly correlated pairs. The significant relationship was confirmed in multiple linear stepwise regression analysis. CD4 + cells were not included in the multiple linear regression analysis due to the multicollinearity with their subsets. P < 0.05 was considered significant.

Study approval
This study was carried out according to the World Medical Association's Declaration of Helsinki, the Japanese Ethics Guidelines for Human Genome/Gene Analysis Research, and the Vietnamese Ethics Guidelines. The protocol was approved by the Ethics Committee of Kanazawa University [2011-080 (5775)] and the Ethics Committee of the VNCH (09/2012/ BVNTWW-HD3), Hanoi, Vietnam. Each child's parents or guardians were informed, and written consent was obtained for all participants. This study is registered at UMIN-CTR: UMIN000015044.

Recruitment and characteristics of the study population
Approximately 500 children with HIV were followed at the VNCH in Hanoi, Vietnam, in 2012, and 40 of them did not start ART according to the Guidelines of the Ministry of Health in Vietnam (No. 3003/QÐ-BYT dated 19/08/2009) [40]. We invited all 40 of the children who did not receive ART, 30 of whom consented to participate in this study [13 females and 17 males; median age 5.9 years, range 2.0-8.8 years; HIV(+) group]. We tried to recruit age-and gender-matched children with HIV who were on ART [n = 29: 12 females and 17 males; median age 6.1 years, range 3.6-8.6 years; median duration of ART: 3.5 years, range 0.8-5.8 years; ART(+) group] and children without HIV as a control [n = 20, 8 females and 12 males; median age 4.1, range 2.0-8.3 years; HIV(-) group], though we could only recruit a smaller number of children without HIV who were 2 years younger than the HIV(+) and ART(+) groups (P = 0.048 and P = 0.009, respectively). Their detailed demographic characteristics and immune status are provided in Table 1 and elsewhere [26,27].
PCA revealed that the HIV(−) and HIV(+) groups had similar gut microbiota structures. The gut microbiota structure of the ART(+) group was different from the other groups and characterized by the abundance of Atopobium cluster, Bifidobacterium, Prevotella, and Lactobacillus spp. (Fig 1).  [25].
In the ART(+) group, the number of Staphylococcus significantly correlated with the CD4 + cell count and its subsets (CD4 + cells: Rho = 0. 46 Table).  Table].

Independent predictors of the gut microbiota in children with HIV
The multiple linear regression analyses including age, ART duration, immune status, and use of cotrimoxazole showed that the ART duration was independently associated with the number of C. perfringens (Beta coefficient = 0.726, P < 0.001), and Th17 count with the number of Staphylococcus (Beta coefficient = 0.428, P = 0.02) in the ART(+) group (Tables 2 and 3). The linear regression analysis for C. coccoides group in the HIV(+) group showed no significant association (S5 Table).

Discussion
In the present study, we investigated the impact of HIV infection and ART on the gut microbiota of Vietnamese children using RT-qPCR. We found that several sub-dominant gut bacteria were positively associated with age in children without HIV, but this was not observed in the children with HIV. In addition, Staphylococcus negatively correlated with age, i.e. the duration of HIV infection, in the children vertically infected with HIV, and ART duration had an independent positive association with C. perfringens and Th17 count with Staphylococcus in the HIV-infected children on ART. These findings indicate an impact of HIV infection and ART on the sub-dominant gut microbiota, such as C. perfringens and Staphylococcus, in children. Our findings highlight the importance of investigating the role of the sub-dominant gut microbiota in the pathogenesis of HIV infection.
To the best of our knowledge, this study is the first to apply RT-qPCR techniques to evaluate the gut microbiota, particularly sub-dominant bacteria, in children with and without HIV. The sum of the six dominant bacterial groups in fecal samples detected by RT-qPCR was previously reported to cover 71.3% of the total intestinal bacterial count estimated by hybridization with a universal probe [29,41]. This finding ensures the validity of using RT-qPCR methods to identify the main gut microbiota profile in this study, though our results may not be comparable directly to the results of the other studies using NGS, since the RT-qPCR method is not appropriate to calculate the diversity indices and the relative abundance of the selected bacteria. In addition, the RT-qPCR approach can detect and enumerate the gut bacteria at the population level between 10 2 and 10 11 cells/g of feces, whereas the lower detection limit of next generation sequencing (NGS) methods is 10 7 to 10 8 cells/g [33]. The counts of the sub-dominant bacteria, including Lactobacilli and potential pathogens, such as C. perfringens, were near the detection limit of NGS or lower [33]; thus, we took advantage of RT-qPCR to estimate the counts of these less abundant, but clinically significant, targets.
In this study, the numbers of dominant gut bacteria, including C. coccoides group, C. leptum subgroup, Bifidobacterium, and Atopobium cluster, did not correlate with age in the HIV(−) group, i.e., children aged 2 to 8 years. This finding is consistent with previous findings that the gut microbiota of healthy children stabilizes and becomes similar to that of adults at around 2 or 3 years of age [7,21,22,33]. In contrast, several sub-dominant gut bacteria, including Streptococcus, Enterococcus, and Enterobacteriaceae, positively correlated with age in the HIV(−) group. This finding is also consistent with previous findings in Japanese children evaluated using the same RT-qPCR methods [33]. These results suggest that the dominant gut microbiota may reach stable levels by 2 or 3 years of age, whereas the sub-dominant bacteria may still be in transition in children aged 2 to 8 years.
We found that C. coccoides group positively correlated with the CD4 + cell count and its subsets in the HIV(+) group, as well as the Treg cell count in the ART(+) group. The C. coccoides spp. are known to promote the expansion and differentiation of Treg cells, which play a central role in regulating gut inflammation through the production of butyrate, a short-chain fatty acid (SCFA), in mice [42,43]. In the ART(+) group, the B. fragilis group positively correlated with the Treg cell count. This finding is consistent with a previous study that found that B. fragilis promotes Treg cell function by producing polysaccharide A [44]. Thus, HIV infection and ART may also influence the immune status by changing the levels of gut bacterial metabolites, such as SCFAs [45,46]. Further studies on bacterial metabolites and their networks in the gut microbiota of children with HIV treated with and without ART may elucidate the underlying mechanisms of immune modulation in HIV infection and ART interventions.
Multiple regression analysis showed a positive association between gut Staphylococcus and Th17 count in the ART(+) group, which was shown for the first time. Th17 cells produce interleukin-17, which is important for promoting neutrophil recruitment to clear bacteria and has a specific role in the host defense against Staphylococcus aureus skin infection [47]. Thus, it would be interesting to investigate the interaction between Th17 and gut Staphylococcus in order to understand the pathophysiology of HIV infection in children who are on ART.
The use of cotrimoxazole reportedly influences some gut bacteria and reduces gut inflammation in children with HIV [48][49][50]. In the current study, the use of cotrimoxazole was associated only with C. perfringens in the ART(+) group. However, in multiple regression analysis, we found that ART duration, but not the use of cotrimoxazole, was independently associated with C. perfringens, which is a potentially harmful bacterium [51]. These findings suggest that a novel therapeutic approach, such as ingesting probiotics and/or prebiotics, may be necessary to restore gut microbiota homeostasis in children with HIV who are on ART [52].
In this study, all of the children in the ART(+) group were treated with NRTIs as backbone drugs and only four also received a PI. Therefore, the positive correlation between the number of C. perfringens and ART duration may be due to the use of NRTIs, but not PIs, even though PIs are known to lower the diversity of the gut microbiota [53,54]. Our study is consistent with the previous study that ART, especially NRTI-including regimen, had more suppressive impacts on the composition and the variability of the gut microbiota [55]. Further study is needed to investigate whether NRTIs influence the gut microbiota, directly or indirectly, through the restoration of immune status in children with HIV who are on ART.
This study has some limitations. First, the children in the HIV(−) group were 2 years younger than the other groups. The diets were not controlled among the groups, though the children in the HIV(−) and ART(+) groups who resided at orphanages were provided the same diets. The children in the HIV(+) group who resided in their own homes appeared to have poorer nutritional status than the children in other groups, which could be due to the uncontrolled diet and/or HIV infection [26]. Considering the influence of age and diet factors on the gut microbiota [7,21,22,33,[56][57][58], we did not focus on comparing the gut microbiota between the groups, but highlighted the factors associated with the gut microbiota in each group. Second, the number of patients recruited in the present study was relatively small, which may limit the significance of our findings. Third, we have not mentioned the HIV-exposure history of the children in the HIV(−) group because we could not confirm the history via documents, though they were reportedly born to mothers with HIV. Machiavelli et al. reported that the gut microbiota of HIV-exposed but uninfected children is similar to that of HIV-unexposed and uninfected children at the age of 18 months except in several taxa [59]. Therefore, the impact of HIV-exposure history on the gut microbiota in the children without HIV over 2 years of age would be limited. These findings require confirmation in longitudinal studies that compare the gut microbiome between age-matched children with and without HIV, with and without HIV-exposure history, and/or before and after initiating ART to assess the effect of ART on the composition of the gut microbiota in children with HIV.
This study provided new insights into the alterations in the gut microbiota, particularly the sub-dominant groups of bacteria, among children with HIV. Our results suggest that HIV infection and ART influence the sub-dominant gut microbiota, directly or indirectly, in association with the immune status of children with HIV.  Table. Linear regression analysis of Clostridium coccoides group with immune status in the HIV(+) group. ART: anti-retroviral therapy; Treg: regulatory T cells; Beta: regression coefficient; SE: standard error. Stepwise multiple linear regression analysis was not done, since no factor with P < 0.05 was found in the simple linear regression analysis. (DOCX)