Analysis of Human Gut Microbiota Composition Associated to the Presence of Commensal and Pathogen Microorganisms in Côte d’Ivoire

Background: The human gut microbiota is a microbial ecosystem contributing to the maintenance of host health with functions related to immune and metabolic aspects. Relations between microbiota and enteric pathogens in sub-Saharan Africa are scarcely investigated. The present study explored gut microbiota composition associated to the presence of common enteric pathogens and commensal microorganisms, e.g., Blastocystis and Entamoeba species, in children and adults from semi-urban and non-urban localities in Côte d’Ivoire. Methods: Seventy-six stool samples were analyzed for microbiota composition by 16S rRDNA sequencing. The presence of adeno-, entero-, parechoviruses, bacterial and protozoal pathogens, Blastocystis, and commensal Entamoeba species, was analyzed by different molecular assays. Results: Twelve individuals resulted negative for any tested microorganisms, 64 subjects were positive for one or more microorganisms. Adenovirus, enterovirus, enterotoxigenic Escherichia coli (ETEC), and Blastocystis were frequently detected. Conclusions: The bacterial composition driven by Prevotellaceae and Ruminococcaceae confirmed the biotype related to the traditional dietary and cooking practices in low-income countries. Clear separation in UniFrac distance in subjects co-harboring Entamoeba hartmanni and Blastocystis was evidenced. Alpha diversity variation in negative control group versus only Blastocystis positive suggested its possible regulatory contribution on intestinal microbiota. Pathogenic bacteria and virus did not affect the positive outcome of co-harbored Blastocystis.


Introduction
The intestinal microbiota consists of a complex microbial community, which exhibits a mutualistic relationship with the host and a crucial impact on human health [1]. About 60% of its bacterial composition persists stably for decades (e.g., the phyla Firmicutes, Bacteroidetes and Actinobacteria), while the remaining proportion is involved in transient modifications induced by different environmental conditions. The high diversity and the relative abundance of beneficial bacterial strains, compared to those potentially harmful, promote the homeostasis of intestinal functions and hostprotective immunity [2], and the combinations of redundant microbial species, differing among the human populations and geographical areas, insure fundamental metabolic functions [3]. However, several factors, e.g., antibiotic therapies, may lead to the loss of microbial diversity and imbalances between commensal versus pathogenic bacterial strains, weakening the recovery of intestinal microbiota [4].
The development of advanced sequencing techniques has allowed a more comprehensive understanding of the relationships between intestinal bacteria and the human host than a culture-based analysis alone [1]. Increasing evidence suggests a distinction between a state of balanced homeostasis, typical of healthy subjects, and different dynamic associations of microbial communities that seem to predispose to non-communicable diseases [5], as supported by studies on animal models [6].
Pioneering microbiota studies were performed in Europe, China, and the United States, evidencing a lower microbial diversity compared to pre-industrial populations [7,8]. Further comparisons have been carried out between different ethnic groups living in America [9,10] and a growing number of data relating to African populations are now available [11][12][13][14].
The characterization of human intestinal microbiota, especially in endemic environments for enteropathy, cannot neglect the potential presence of common intestinal pathogens, including bacteria, viruses, and parasites, and their influence on microbiota balance [15,16]. However, the dynamics of relationships between pathogens and commensals, both prokaryotes and eukaryotes, are often underestimated [17,18] or focused on single microorganisms [19][20][21]. Poor water, sanitation, and hygiene conditions in low and middle-income countries sustain the susceptibility to multiple and asymptomatic infections [22].
The link between natural systems, human health, and infections remains an interesting challenge. In line with these aspects, the present study investigated the gut microbiota in children and adults living in Côte d'Ivoire. A particular, focus was on the changes in microbiota composition related to the presence of Blastocystis, commensal Entamoeba species, and different common enteric pathogens.

Study Area and Sample Collection
Human fecal samples were collected at four sites in the department of Grand-Bassam, in the south of Côte d'Ivoire. Bonoua (5 • 16 17 N, 3 • 35 40 W) was the only semi-urban locality while Assouindé (5 • 09 57.1 N 3 • 28 10.9 W), Kimoukro, and Yaou (5 • 14 25.0 N 3 • 37 37.9 W) were rural/semi-rural villages 6 to 20 km away from Bonoua. Sampling was performed during three distinct seasons characterizing the study area: the heavy rainy season (HRS), the low rainy season (LRS), and the heavy dry season (HDS).

Sample Management and Fecal Nucleic Acids Extraction
From each patient, one fecal sample was collected using an appropriate sterile container. Recent or ongoing antibiotic therapy constituted an exclusion principle. After allocation of an anonymous code, two aliquots (ca. 400 mg each) were prepared from the sample. The first aliquot was preserved for parasitological and NGS analysis by adding 0.5 mL of Qiagen Allprotect Tissue Reagent (QIAGEN, Hilden, Germany) in a sterile Nalgene ® cryogenic vial size 2.0 mL, frozen at −20 • C until transported to the University Microorganisms 2021, 9, 1763 3 of 19 of Rome Tor Vergata (Italy). DNA from this aliquot was extracted by QIAmp Stool Mini Kit (QIAGEN, Hilden, Germany) [16]. The second aliquot for the xTAG GPP assay analysis and real-time PCR was frozen at −20 • C, transported to the University of Rome Tor Vergata, preserved at −80 • C until shipment to the Institute of Virology of the University of Cologne (Germany). In this case, sample preparation was performed using the automated platform VERSANT kPCR Molecular System (Siemens Healthcare Diagnostics, Erlangen, Germany), according to the manufacturer's instructions, as described in [23].

Blastocystis and Commensal Entamoeba spp. Detection and Identification
End-point PCR and sequence analysis of the SSU rDNA region were carried out for the molecular identification of Blastocystis and Entamoeba species. The detection of Blastocystis was performed by amplification of a 600 bp fragment by nested PCR using the primers RD5-BhRDr and Blasto2F-Blasto2R [25] as described in D'Alfonso et al. [26]. The amplification of the specific fragment of Entamoeba spp., including the pathogenic E. histolytica, was obtained using the primers JVC and DSPR2 [27]. The amplicons were from 622 to 667 bp long depending on the species.
All amplicons were purified using a QIAquick Gel Extraction Kit (QIAGEN, Valencia, CA, USA) and sequencing was performed in the Bio-Fab Research Laboratory in Rome (Italy). The identities of the obtained sequences were verified using the Basic Local Alignment Search Tool (BLAST).
Amplification was performed as follows: initial denaturation at 95 • C for 5 min, 25 cycles of denaturation at 95 • C for 40 s, annealing at 53 • C for 40 s, and extension at 72 • C for 60 s, followed by a final extension of 7 min at 72 • C. Before the "library preparation" by index-PCR, amplicons were purified using the Agencourt AMPure ® XP system on a BioMek NX workstation (Beckman Coulter, Germany) following instructions Microorganisms 2021, 9, 1763 4 of 19 provided by the manufacturer. The index PCR attached dual indices using the Nextera XT ® (Illumina Inc., San Diego, CA, USA) Index Kit following instructions provided by the manufacturer. The index PCR reaction was purified as described above followed by a library quantification, normalization, pooling, library denaturing, and MiSeq ® (Illumina Inc., San Diego, CA, USA) sample loading in a final concentration of 10 pM, following the protocol described by the manufacturer. Sequencing was accomplished using MiSeq reagent kit v2 in a 2 × 250 cycle paired-end sequencing run.

16S rRNA Amplicon Data Processing
Sequencing data were processed using the QIIME DADA2 plugin with the denoisepaired option and standard parameters (trunc_q = 2, max_ee = 2, chimera_method = consensus). Taxonomic classification was performed by a Naïve Bayes classifier (sklearn) [29], which was trained on the SILVA database release 128 [30]. Rarefaction curves were determined based on the feature table and analysis of the relative proportion of each bacterial taxon was made after the data were rarefied at a sequencing depth of 4000 sequences per sample.
Statistical analyzes were carried out using R for Statistical Computing (version 3.5.1, R Foundation for Statistical Computing, Vienna, Austria) [31] (Team RC). The QIIME biom data were imported and diversity scores calculated using the phyloseq R package [31,32]. Rarefaction curves were determined based on the feature table and analysis of the relative proportion of each bacterial taxon was made after the data were rarefied at a sequencing depth of 4000 sequences per sample. All continuous data were presented as mean and standard deviation (SD) or median and range, tested with Student's t-test, Mann-Whitney U test, and Kruskal-Wallis-test with Dunn's post-test, as appropriate. In the beta diversity, the UniFrac distances between the samples were visualized using principal coordinate analysis (PCoA), and group effects were tested by a permutational multivariate analysis of variance (PERMANOVA). Differentially abundant taxa were identified using linear discriminant analysis (LDA) effective size (LefSe) [33]. All statistical tests were two-tailed, and a p-value < 0.05 was considered statistically significant.

Study Area and Host Population
Sixty-seven fecal samples were collected from subjects aged between 1 to 59 years old and nine samples from infants (<1 year), representing four different localities: 53 samples from an urban area (Bonoua) and 23 samples from non-urban areas (Assouindé, Kimoukro, Yaou; see Table 1). Fourteen samples were collected during the heavy rainy season, 22 during the low rainy season, and 40 during the heavy dry season.

Viral, Bacterial, and Protozoa Detection
Within the cohort, 12 subjects, including four infants, were tested negative for all analyzed microorganisms; 64 subjects, including five infants, were tested positive for at least one virus, bacterium, or protozoa; 10 subjects, including three infants, were tested positive for a single microorganism, 20 subjects for two microorganisms, and 34 subjects, including two infants, for three or more microorganisms.
Overall, 32 subjects, including five infants, were tested positive for at least one virus, with adenovirus (AdV) and enterovirus (EV) being the most common, as showed in Table 2.  A total of 17 subjects, including two infants, were tested positive for bacterial pathogens with enterotoxigenic E. coli (ETEC), Campylobacter, Shigella, and E. coli O157 being the most frequently identified (Table 2). A total of 54 subjects, including one infant, were positive for at least one protozoa species with Blastocystis being the most common (47 out of 54). Overall, 20 subjects were positive for Entamoeba hartmanni, 11 for Entamoeba coli (En. coli), and two for Entamoeba dispar; Giardia duodenalis was detected in 25 individuals, including one infant (Table 3).

Age
A marked difference in diversity was found between infants compared to the other age groups ( Figure 1A). Infants showed a significant lower alpha diversity, in particular when compared to those aged 6-17 and >18 years (Table 4).
The organisms most frequently present at the genus level are shown in Figure 1B; the related statistical significance in the abundance comparison between infants versus all other groups is reported in Table S1A.

Denovo Clustering
In an attempt to identify different gut microbiome enterotypes in all 67 subjects >1 year of age, we calculated the UniFrac distances between all samples, partitioned (clustered) the data into k clusters around meteroids ( Figure S2A), and calculated the respective gap statistics. Using this approach, we were able to split the subjects into two clusters. The first one, grouping 39 subjects aged 1-59 years and showed higher richness based on Chao1 (p < 0.01), OTUs (p < 0.01), and PD (p < 0.001), and dominance of Ruminococcaceae (Clostridia Class) over Prevotellaceae (Bacteroidia Class). The second cluster, including 28 subjects aged 1-19 years, showed an inverse dominance of Prevotellaceae versus Ruminococcaceae ( Figure S2B-D). At the genus level, both clusters had similar abundances of Bacteroides, Escherichia-Shigella, Faecalibacterium, Roseburia, Eubacterium, Lachnospiraceae, Christensenellaceae R-7 group.
Subsequently, the results relating to the nine infants are not indicated considering their small number and the unexpected variety of isolated pathogens.
The data of alpha and beta diversity concerning five subjects tested positive for different enteric pathogens (adenovirus, enterovirus, parechovirus, norovirus, Campylobacter, Shigella, and G. duodenalis) and four negative subjects were indicated in S3 (Table S3A, Figure S3B,D).
The data of alpha and beta diversity concerning five subjects tested positive for different enteric pathogens (adenovirus, enterovirus, parechovirus, norovirus, Campylobacter, Shigella, and G. duodenalis) and four negative subjects were indicated in S3 (Table S3A, Figure S3B,D).

Localities and Seasonality
The gut microbiota of patients sampled from semi-urban (Bonoua) and non-urban areas (Assouindé, Kimoukro, Yaou) showed no differences in alpha and beta diversity.
The relative abundances of the genus Escherichia-Shigella (0.00125 (0-0.00125)) in samples collected during the low rainy season was lower than in sample collected in the heavy dry season (0.0045 (0-0.59125)) and the heavy rainy season (0.003 (0-0.272)). The genus Prevotella 9 showed a significant lower abundance in the heavy rainy season (0.001375 (0-0.6885)) versus heavy dry season (0.05575 (0-0.7405)) and versus low rainy season (0.52625 (0.022-0.86675)), and in heavy dry season versus low rainy season. The genus Megasphaera showed a significant lower abundance in heavy dry season (0 (0- The relative abundances of the genus Escherichia-Shigella (0.00125 (0-0.00125)) in samples collected during the low rainy season was lower than in sample collected in the heavy dry season (0.0045 (0-0.59125)) and the heavy rainy season (0.003 (0-0.272)). The genus Prevotella 9 showed a significant lower abundance in the heavy rainy season (0.001375 (0-0.6885)) versus heavy dry season (0.05575 (0-0.7405)) and versus low rainy season (0.52625 (0.022-0.86675)), and in heavy dry season versus low rainy season. The genus Megasphaera showed a significant lower abundance in heavy dry season (0 (0-0.121)) versus the heavy rainy season (0 (0-0.6535)). The top 15 bacterial composition at the family and genus levels were shown in Figure 3.

Blastocystis, Commensal Entamoeba spp., and Intestinal Pathogens
Subjects negative for common enteric pathogens (xTAG Luminex panel), EV, PeV, AdV (real-time PCR), Blastocystis, and Entamoeba spp. (PCR) were assumed as the control group ( Figure S3A,C). The other groups contained positive individuals who tested positive for one or more of these microorganisms. Only five individuals aged 5 to 10 years (4F/1M, 7.8%) were found to have multiple infections and exclusively reported frequent abdominal pains (Table S4A). The beta diversity analysis showed a significant lower abundance of the genus Ruminococcaceae UCG-002 in negative (0.0116 [0-0.082]) versus positive individuals (0.065 [0-0.8272]) (p < 0.01). The opposite trend for the genus Faecalibacterium does not reach significance.
No difference was found between positive versus negative subjects for the different age groups (Figure 4).
Microorganisms 2021, 9, x FOR PEER REVIEW 10 of 20 Figure 4. The UniFrac distances between enteric pathogen positive versus enteric pathogen negative subjects is displayed by age groups, using principal coordinate analysis (PCoA). The infant group tends to separate out according to Axis 1 (7.6%) and Axis 2 component (7%). The differences of the infant group from other age groups as shown in Figure 1, are also confirmed for comparisons of enteric pathogen negative versus enteric pathogen positive subjects.

Blastocystis
To shed light on a possible implication of single or co-detection Blastocystis on gut microbiota, the control group (Group 1) was compared with 42 positive Blastocystis patients further divided as follows: 6 subjects positive for Blastocystis only (Group 2); 12 subjects positive for Blastocystis and Entamoeba spp. only (Group 3); 5 subjects positive for Blastocystis, viruses, and bacteria (Group 4a); 6 subjects positive for Blastocystis and Entamoeba spp., viruses and bacteria (Group 4b); 13 positive subjects for Blastocystis and Entamoeba spp. and G. duodenalis and other pathogens (Group 5b). Alpha diversity indices increased when Entamoeba spp. co-infections were present ( Table S4B).
As for bacterial composition, significant differences were observed between Group 1 versus Groups 3, 4a, 4b, and 5b but not versus subjects positive for Blastocystis only (Group 2). In detail, at the genus level, Group 1 showed: (i) a lower median of Alloprevotella (p < 0.01), Rikenellaceae RC9, Succinivibrio (p < 0.05), and Ruminococcaceae UCG-010 (p < 0.01) versus Group 3, (ii) a higher median of Faecalibacterium (p < 0.05) versus Group 4a; (iii) a lower median of Ruminococcaceae UCG-002 and Ruminococcaceae UCG-010 versus Group 4b; (iv) a higher median of Bacteroides (p < 0.05) and Lachnoclostridium (p < 0.05) and a lower median of Ruminococcaceae UCG-010 (p < 0.05) compared with Group 5b. Finally, Group 2 showed an increase of Alloprevotella and Prevotella 2 (p < 0.05) versus Group 3 ( Figure 5). . The UniFrac distances between enteric pathogen positive versus enteric pathogen negative subjects is displayed by age groups, using principal coordinate analysis (PCoA). The infant group tends to separate out according to Axis 1 (7.6%) and Axis 2 component (7%). The differences of the infant group from other age groups as shown in Figure 1, are also confirmed for comparisons of enteric pathogen negative versus enteric pathogen positive subjects.

Blastocystis
To shed light on a possible implication of single or co-detection Blastocystis on gut microbiota, the control group (Group 1) was compared with 42 positive Blastocystis patients further divided as follows: 6 subjects positive for Blastocystis only (Group 2); 12 subjects positive for Blastocystis and Entamoeba spp. only (Group 3); 5 subjects positive for Blastocystis, viruses, and bacteria (Group 4a); 6 subjects positive for Blastocystis and Entamoeba spp., viruses and bacteria (Group 4b); 13 positive subjects for Blastocystis and Entamoeba spp. and G. duodenalis and other pathogens (Group 5b). Alpha diversity indices increased when Entamoeba spp. co-infections were present ( Table S4B).
The presence of E. hartmanni produced a significant increase in alpha diversity indices (Table S4C). The UniFrac distances, using principal coordinate analysis (PCoA), showed the shifting of positive groups for E. hartmanni ( Figure 6B). At the genus level of the Christensenellaceae R-7 group, Ruminococcaceae UCG-002 increased in two groups carrying E. hartmanni; Blastocystis, Faecalibacterium, and Bifidobacterium showed an opposite trend (Table S5C).
The presence of E. hartmanni produced a significant increase in alpha diversity indices (Table S4C). The UniFrac distances, using principal coordinate analysis (PCoA), showed the shifting of positive groups for E. hartmanni ( Figure 6B). At the genus level of the Christensenellaceae R-7 group, Ruminococcaceae UCG-002 increased in two groups carrying E. hartmanni; Blastocystis, Faecalibacterium, and Bifidobacterium showed an opposite trend (Table S5C).  Table S5D.
The next step was to discriminate any differential relationship between the control group (Group 1E) and the subjects infected by pathogens, not including Entamoeba spp. (Group 2E), or infected by different Entamoeba species (Group 3E = positive for En. coli and other pathogens; Group 4E = positive for E. dispar and other pathogens; Group 5E = positive for E. hartmanni and other pathogens). As for alpha diversity, a significant difference was found in Chao1 Index, Observed OTUs (p < 0.05) and Faith's phylogenetic diversity (p < 0.01), comparing Group 1E versus Group 5E; observed OTUs and Faith's phylogenetic diversity (p < 0.05) comparing Group 2E versus Group 5E (Table S4D).
Beta diversity analysis showed: (i) a lower abundance for Bacteroides (p < 0.05) and for Lachnoclostridium (p < 0.01) in Group 3E versus Groups 1E and 2E; (ii) a lower abundance for [Ruminococcus] torques group in Group 3E versus Group 2E; (iii) an increased abundance for Ruminococcaceae UCG-002 and Ruminococcaceae UCG-010 (p < 0.05) in  Table S5D. The next step was to discriminate any differential relationship between the control group (Group 1E) and the subjects infected by pathogens, not including Entamoeba spp. (Group 2E), or infected by different Entamoeba species (Group 3E = positive for En. coli and other pathogens; Group 4E = positive for E. dispar and other pathogens; Group 5E = positive for E. hartmanni and other pathogens). As for alpha diversity, a significant difference was found in Chao1 Index, Observed OTUs (p < 0.05) and Faith's phylogenetic diversity (p < 0.01), comparing Group 1E versus Group 5E; observed OTUs and Faith's phylogenetic diversity (p < 0.05) comparing Group 2E versus Group 5E (Table S4D).

G. duodenalis and Other Intestinal Pathogens
To investigate the influence of G. duodenalis and other enteropathogens, the 67 subjects were divided into seven groups: control group (Group A); 6 subjects positive for Blastocystis only (group C); 11 subjects positive for Blastocystis and E. hartmanni without enteric pathogens (group D); 6 subjects infected by E. hartmanni, Blastocystis and G. duodenalis, and enteric pathogenic bacteria and viruses (group E1); 6 subjects positive with En. coli, Blastocystis and G. duodenalis, and enteric pathogenic bacteria and viruses (group E2); 8 subjects positive with Blastocystis, G. duodenalis, and enteric pathogenic bacteria and viruses (group E3); 5 subjects positive for Blastocystis and enteric pathogenic bacteria and viruses (group M).
Observed OTUs, Faith's PD, and Shannon index were significantly increased in group D compared to the control group and only for the Shannon index compared to group M (Table S4E).
The UniFrac PCoA analysis along axis 1 showed some overlap in the composition of the groups and a non-significant separation between the three groups co-infected with Entamoeba species (D, E1, E2) and the other groups (A, E3, M) ( Figure 7A); negative subjects (group A) were separated from subjects co-infected with Entamoeba spp. (D, E1, E2) as displayed in Figure 6A. Cladograms generated by LEfSe showed differences in taxa between A, C, E1, E2 groups with LDA score (p < 0.01) ( Figure 7B). No significant differences between all groups were observed by the analysis of the relative abundance at the family and genus level ( Figure 7C,D). Group 5E versus Group 2E, and for Ruminococcaceae UCG-010 (p < 0.01) in Group 5E versus Group 1E ( Figure 6C,D).

G. duodenalis and Other Intestinal Pathogens
To investigate the influence of G. duodenalis and other enteropathogens, the 67 subjects were divided into seven groups: control group (Group A); 6 subjects positive for Blastocystis only (group C); 11 subjects positive for Blastocystis and E. hartmanni without enteric pathogens (group D); 6 subjects infected by E. hartmanni, Blastocystis and G. duodenalis, and enteric pathogenic bacteria and viruses (group E1); 6 subjects positive with En. coli, Blastocystis and G. duodenalis, and enteric pathogenic bacteria and viruses (group E2); 8 subjects positive with Blastocystis, G. duodenalis, and enteric pathogenic bacteria and viruses (group E3); 5 subjects positive for Blastocystis and enteric pathogenic bacteria and viruses (group M).
Observed OTUs, Faith's PD, and Shannon index were significantly increased in group D compared to the control group and only for the Shannon index compared to group M (Table S4E).
The UniFrac PCoA analysis along axis 1 showed some overlap in the composition of the groups and a non-significant separation between the three groups co-infected with Entamoeba species (D, E1, E2) and the other groups (A, E3, M) ( Figure 7A); negative subjects (group A) were separated from subjects co-infected with Entamoeba spp. (D, E1, E2) as displayed in Figure 6A. Cladograms generated by LEfSe showed differences in taxa between A, C, E1, E2 groups with LDA score (p < 0.01) ( Figure 7B). No significant differences between all groups were observed by the analysis of the relative abundance at the family and genus level (Figures 7C and D).

Discussion
Age, diet, mode of birth and lactation, environment, circadian rhythm, geographic location, and host genetics, in a complex way, contribute to the individual heterogeneity of the human gut microbiota [1]. The data currently available are mainly from populations living in so-called "Western" contexts such as Europe and North America where Bacteroides dominate the gut microbiota. Limited data are available from sub-Saharan rural or semi-urban areas where a predominance of Prevotella in bacteria composition is evidenced. However, the diffusing industrialization and westernization in these regions could influence these patterns [8,34,35].
The present study describes the composition of the intestinal microbiota in infants, children, and adults living in the Côte d'Ivoire without neglecting the most common endemic, pathogenic, and commensal microorganisms. This investigation aimed to delineate a potential link between Blastocystis, commensal Entamoeba spp., common enteric pathogens, and the gut microbiota [36,37]. Among adults, two clusters, one dominated by Ruminococcaceae and the other one by Prevotellaceae, could be identified. These species could be either termed "biomarkers" as proposed by Gorvitovskaia et al. [38], or "biomes types" as indicated by Huse et al. [39]. These species represent consortia of bacteria mostly influenced by genetic and immune host factors, and, above all, dietary and cooking practices [40]. The reduced abundance of Bacteroidaceae, in both clusters, was in line with previous data from similar eco-geographical areas, from agricultural societies, and countries with a low level of industrialization [10,41]. In our sampling area, weaning occurs earlier than in Western societies and the diet mostly includes local and seasonal vegetables (e.g., okra, cassava and cassava leaf, tubers, and plantain) prevailing on the intake of animal-based product (e.g., eggs, poultry, fish). Among the fats, peanuts, and palm oil dominate. Seasonal fruit consumption is not a widespread habit, due to economic reasons and its rapid deterioration and traditional sugary juices (pineapple, passion fruit, hibiscus, ginger) are preferred, therefore reducing the intake of vitamins, mineral salts and healthy fiber. Overall, the present investigation confirmed the different composition of gut microbiota between infants, children, and the high individual variation among adults [4,42]. Enterobacteriales (Proteobacteria phylum), Bifidobacteriales (Actinobacteria phylum), and Lactobacillales (Firmicutes phylum) showed higher relative abundance in infants, while Firmicutes and Bacteroidetes were the dominant phyla in older individuals, followed by Proteobacteria [43,44]. In all age groups, Faecalibacterium, Lachnospiraceae, Eubacterium rectale, Roseburia, members of Clostridium clusters XIVa and IV, have been detected. These taxa produce crucial metabolites that promote health and maintain gut microbiome colonization resistance [45]. The lack of complete uniformity with previous studies could be attributable to regional and geographical differences [10,[46][47][48].
Regarding sex-related microbiota differences, females showed a higher phylogenetic diversity than males. It is possible to hypothesize that puberty in young women could have an impact on alpha diversity compared to males. Additionally, ethnic and social habits related to daily activities could contribute to an exposition of men and women to different environmental factors [49]. A sex-related divergence of Prevotella 9, Ruminococcaceae UCG-002, and Megasphaera was noted in different age groups. In the present cohort, Megasphaera was almost undetectable in both sexes, except in women >18 years of age, which is different from the study by Takagi et al. [50], who reported high levels in Japanese adult males. The meaning of these differences will require further detailed investigations.
A specific aim of the study was to provide new data on the microbiota composition in subjects harboring Blastocystis, Entamoeba spp., and common enteropathogens. The high detection of protozoa viruses and bacteria confirmed data reported by previous studies conducted in the same area [26,51] and other countries in sub-Saharan Africa [9,19,[52][53][54]. No patients were tested positive for E. histolytica, C. hominis, and C. parvum. Unfortunately, the use of the xTAG GPP molecular test did not allow the isolation of Cryptosporidium meleagridis whose prevalence is relatively high in the same area [55].
The high number of asymptomatic individuals suggests an adaptation to constant and cumulative exposure to enteropathogens due to the common risk factors associated with low-income countries [22,23,[56][57][58].
In children and adults carrying protozoa, bacteria, or viruses, greater diversity and richness in the gut microbiota than in negative subjects was evident, in line with other works [19,20]. Furthermore, the critical taxa involved in the intestinal homeostasis such as Faecalibacterium, Bifidobacterium, Lachnospiraceae, Ruminococcaceae UCG-002, (Eubacterium) coprostanoligenes, (Eubacterium) rectale, and Roseburia group showed variations in richness in the same comparison.
The variable abundance of bacterial genera such as Faecalibacterium, Bifidobacterium, Roseburia group allows for the production of key metabolites as short-chain fatty acids (SCFA), e.g., butyrate and succinate, lactate, or acetate to preserve homeostasis [1,4,59]. This scenario could reflect a bidirectional ability of adaptation between the host's immune response and bacterial composition. Faecalibacterium prausnitzii and E. rectale are responsible for a major fraction of butyrate production and consequently play a crucial role in shaping the local and peripheral immune system. The emerging evidence of diet-induced changes in SCFA-producing bacteria, as reported by Morrison et al. [60], could suggest a possible contribution of Ivorian eating habits in the modulation of adaptive immunity.
Blastocystis was highly prevalent in our cohort and a non-significant variation of alpha diversity between Blastocystis positive-only group and the control group is evidence of the modest effect that Blastocystis carriage has [61,62]. The lower abundance of Bacteroides in subjects infected with Blastocystis seems to confirm the hypothesis of a minor predisposition to harbor this parasite in "Bacteroides-driven enterotype" subjects, as described for Danish and Spanish individuals by O'Brien Andersen et al. [63], and Mexican subjects by Nieves-Ramirez [64]. In all groups carrying Blastocystis, bacterial genera associated with a healthy status, such as E. rectale and coprostanoligenes groups, Roseburia and Succinivibrio showed a constant presence, although with a reduced relative abundance versus control group [46,[65][66][67][68]. These specific OTUs could explain a beneficial influence on normobiosis [61] also in conjunction with a traditional African diet. Considering the genetic diversity within Blastocystis, with at least 17 different subtypes, the possible implications in the pathophysiology of human infections related to the different subtypes should be investigated to better understand the role of this parasite on the human gut microbiota [63,69].
Concerning commensal Entamoeba spp., an increase in alpha diversity in subjects coharboring Entamoeba spp. and Blastocystis, especially in case of E. hartmanni co-occurrence, was evidenced. Clear separation in Unifrac distance was found and the increase of Succinivibrio, Christensenellaceae R-7 group, Ruminococcaeae UCG-002, Rikenellaceae RC9 intestinal group seem associable with Entamoeba perturbations. Bacterial niches, responsible for beneficial or detrimental associations in subjects positive for Blastocystis alone or coharbored with commensal Entamoeba spp., should be further investigated [70,71]. The increase in Succinivibrio in people detected positive for E. hartmanni, Blastocystis, and pathogens seems to confirm the data reported from Cameroon by Morton et al. (19) and in Malawian children by Ordiz et al. [72]. These data suggest that recurrent endemic intestinal infections in rural countries [73], together with eating habits [10] could favor, in some groups of African natives compared to African Americans, the conditions for a stable presence of Succinivibrio in the gut microbiota. The low number of subjects harboring G. duodenalis and pathogenic viruses and bacteria precluded the statistical significance in any comparisons. However, an interesting aspect seems to emerge: some protective taxa (e.g., E. rectale group, Roseburia, Christensenellaceae) showed an expansion despite the reduction of Faecalibacterium, as previously reported [46,[74][75][76][77][78]. The relative abundances of Bifidobacterium and Faecalibacterium were oppositely related. Subjects containing multiple enteric pathogens, without G. duodenalis and Entamoeba spp., showed a decrease in so-called protective taxa (Bifidobacterium and Faecalibacterium) while the prevalence of Megasphaera increased. This result should be better investigated concerning the potential involvement of Megasphaera in the adaptability of the human gut ecosystem [72,79,80].

Conclusions
In conclusion, the co-occurrence of Blastocystis and commensal Entamoeba spp., despite the presence of other enteric pathogens including G. duodenalis, seems to preserve a high diversity, favor different bacterial consortia, and does not compromise the intrinsic ability of intestinal microbiota to restore and/or maintain homeostasis [74,[81][82][83][84][85]. However, the beneficial or potential regulatory effect of Blastocystis is reduced in subjects positive for bacterial and viral pathogens without G. duodenalis and commensal Entamoeba spp., probably reducing the entity of symptoms.
Finally, the complex scenario observed in this cohort from Côte d'Ivoire helps to provide an integrated view of human gut microbiota composition. Simultaneous testing of common intestinal pathogens supplies new data from Africa, still underrepresented in the global picture of the gut microbiota, and helps to better understand the balances of intestinal homeostasis.
Future investigations are needed to verify if the influence of rural diet on gut microbiota can play a role in seasonal predisposition to endemic enteritis. It will be very important to extend the understanding of eukaryotic microorganisms' interactions in the gut microbiota in health status and intestinal infections.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/microorganisms9081763/s1, Figure S1A. The plot of top 15 bacterial relative abundances at the Family level in each age group; Table S1A. The top 15 bacterial relative abundances; Figure S2. Beta diversity of Cluster 1 versus Cluster 2 (Denovo clustering); Table S3A. Alpha diversity indices of positive and negative subjects, grouped in <1 year and >1 year; Figure S3. Positive versus negative subjects comparisons: Beta diversity analysis; Table S4A. All microorganisms detected in the 5 symptomatic subjects; Table S4B. Comparison of Blastocistys groups: Alpha diversity index; Table S4C. Comparison of E. hartmanni groups: Alpha diversity index; Table S4D. Comparison of Entamoeba species groups: Alpha diversity index; Table S4E. Comparison of G. duodenalis and other intestinal pathogens positives: Alpha diversity index; Table S5A. Beta diversity analysis for Entamoeba spp. presence; Table S5B. Beta diversity analysis for Entamoeba spp. presence; Table S5C. Comparison of E. hartmanni groups: Beta diversity at genus level; Table S5D. Beta diversity analysis for Entamoeba spp. detection; Table S5E. Beta diversity analysis for Entamoeba spp. presence.  Institutional Review Board Statement: All procedures performed in this study involving human participants were approved by the Don Orione Center's Medical Committee (Bonoua, Côte d'Ivoire) and followed the 1975 Helsinki Declaration and its later amendments.
Informed Consent Statement: Adults' participation was voluntary after providing a verbal agreement on the methods and purposes of sampling. For minors, informed consent was obtained from their parents and guardians. A local dialect interpreter was used to explain the study's aims, procedures, and significance when necessary. All participants were assured that the human biological sample would provide data exclusively on intestinal microorganisms and not on the human subject. All data were analyzed anonymously.

Data Availability Statement:
The dataset associated to this paper can be accessed on https://www. ncbi.nlm.nih.gov/sra/PRJNA741457 (accessed on 15 August 2021).