Gut Microbiome Analysis Reveals Core Microbiota Variation Among Allopatric Populations of the Commercially Important Euryhaline Cichlid Etroplus suratensis

Abstract

The gut microbiome plays a critical role in host physiology and adaptation, shaped by both intrinsic host factors and extrinsic environmental conditions. In this study, we investigated the influence of habitat type and geographical isolation on gut microbial communities in habitat-isolated populations of the euryhaline cichlid Etroplus suratensis, which inhabit freshwater and brackish water environments. Using 16S rRNA gene amplicon sequencing, we compared microbial assemblages in fish guts and their corresponding habitats to assess patterns of community divergence. Alpha and beta diversity analyses revealed significant differences in microbial composition between gut and water samples, with limited overlap, particularly in brackish water, indicating strong host-mediated filtering of environmental microbiota. Notably, brackish and freshwater habitats harbored 2244 and 3136 unique water-associated taxa, respectively, while only 36 and 426 taxa were shared between water and gut in each habitat. Despite habitat divergence, 59 microbial taxa were consistently shared across gut samples from both populations, indicating the existence of a conserved core microbiome that likely fulfills essential functional roles. These findings support the notion that the fish gut serves as a selective ecological niche, enabling the persistence of functionally relevant microbes while restricting the entry of environmental transients. Moreover, the observed divergence in gut microbiota across habitats, coupled with a shared core, highlights the interplay between local adaptation and conserved host–microbe associations, with potential implications for understanding microbial contributions to vertebrate ecological diversification and allopatric speciation.

1. Introduction

The microbiome, a pivotal element in host biology, is gaining increasing recognition for its influence on nutrition, immunity, development, and adaptation in a wide range of organisms [1]. The dynamic nature of host-associated microbial communities, shaped by both internal and external factors, offers valuable insights into how organisms navigate environmental challenges [1,2,3,4]. In fish, the gut microbiota is a key player in metabolism, disease resistance, and overall fitness [1,5].

For fish, a multitude of factors, including diet, water quality, trophic ecology, and salinity, have been demonstrated to significantly influence gut microbial composition [4,5,6]. Cross-environment comparisons, spanning freshwater, estuarine, and marine habitats, reveal robust associations between microbial diversity and salinity gradients [7]. Furthermore, studies suggest that local habitat conditions may exert as much, if not more, influence on gut microbiota than host phylogeny or taxonomy [8]. These findings underscore the pivotal role of the microbiome in shaping host adaptation to diverse habitats, thereby enriching our understanding of the ecological significance of the microbiome.

Euryhaline fishes, with their ability to thrive in a wide range of salinity levels, serve as excellent models for exploring the relationships between salinity variations and gut microbiome composition. Several euryhaline species exhibit distinct gut microbiota depending on whether they inhabit freshwater or brackish water habitats, and in catadromous fishes, their gut microbiota also alters [8]. Studies on three-spined sticklebacks have revealed substantial differences in gut microbiota, even among populations in close geographical proximity, with ecological factors being one of the key drivers of this differentiation [9]. In Mummichog, habitat and environmental conditions have a stronger influence on gut microbiome composition than population or genetic factors [10]. Findings from studies [9,10] highlight the crucial role of environmental drivers in shaping microbial communities and influencing host responses across diverse habitats.

The South Indian cichlid Etroplus suratensis (pearlspot), a commercially important euryhaline fish, is primarily found in the brackish backwaters of Kerala and also maintains sustainable populations in the freshwater systems of the Western Ghats [11,12]. Despite its ecological and aquaculture relevance, our understanding of its gut microbiota is limited. Establishing baseline information on its microbial composition is crucial for both ecological understanding and future applications in aquaculture management. High-throughput sequencing of 16S rRNA amplicons is a widely employed approach for profiling microbial communities, providing insights into their composition, abundance, and phylogenetic relationships [13,14]. This method is particularly useful for identifying uncultivable or low-abundance taxa, thereby enabling a more comprehensive understanding of microbiome structure [15,16].

Here, we investigate whether there is divergence in gut microbiomes among allopatric populations of a euryhaline fish species. We specifically try to understand: (a) the ecological influence on the gut microbiomes of allopatric populations; (b) the core microbiota variation among populations signaling the role of the microbiome in local adaptation. In addition, this study serves as a pilot investigation to characterize the gut microbiota of this commercially valuable species and to establish a baseline dataset, which can support future studies aiming to systematically examine microbiome variations across aquaculture systems, including both freshwater and brackish water environments.

2. Materials and Methods

2.1. Sample Collection

Fishes were sampled from a brackish water habitat (pH: 7.65, salinity: 23.00 ppt) in Thalassery (11.766727, 75.473069) and a freshwater habitat (pH: 7.36, salinity: 0.006 ppt) in Kacherikadavu (12.074194, 75.741028), Kerala, India. Three individuals of E. suratensis were collected from each location, strictly adhering to the protocols of the Institutional Animal Ethics Committee (IAEC). Water samples were simultaneously collected from the same locations in triplicates. These samples were used for further DNA isolation and sequencing.

2.2. DNA Extraction and Sequencing

Genomic DNA from the whole fish guts was extracted using the Promega Wizard^® Genomic DNA Purification Kit (Promega, Madison, WI, USA) following the manufacturer’s instructions. Prior to extraction, the gut was aseptically removed, rinsed with sterile water to eliminate loosely associated microbes, cut into small pieces, and macerated using a sterile pestle and mortar in the presence of sterile 1× PBS to facilitate homogenization. The homogenate was centrifuged at 10,000× g for 10 min at 4 °C, and the resulting pellet was used as input for DNA extraction.

DNA from water samples (1 L each) was sequentially vacuum-filtered using 0.2-micron Mixed cellulose esters (MCEs) and 47 mm diameter filter paper (Merck Millipore Ltd., Tullagreen, Ireland). The filter paper was then cut into small pieces and carefully transferred to 2 mL centrifuge tubes. Lysis buffer (525 µL; 1.5 M NaCl; 100 mM EDTA; 100 mM Na₂₂HPO₄₄; CTAB-1%; pH 8.0) was added to the tubes, and they were vortexed for one minute after adding two to three sterile glass beads. Five microliters of lysozyme were added and incubated at 37 °C for 1 h. After incubation, 60 µL of 10% SDS was added, and the contents were gently mixed. This was followed by the addition of proteinase K, which was then mixed by gentle inversion. Tubes were incubated at 55 °C for 2.5 h in a water bath. After this, one volume of Chloroform-Isoamyl Alcohol (24:1) was added and mixed by gentle inversions. Tubes were centrifuged at 10,000× g for 10 min at room temperature. The aqueous phase was then transferred carefully into freshly labeled tubes. The chloroform extraction step was repeated. The extraction process was further carried out by following the Macherey-Nagel (MACHERY-NAGEL GmbH & Co. KG. Duren, Germany) Soil Kit protocol. The quality and quantity of DNA extracted from the fish gut (six samples) and water (six samples) were determined using a Nanodrop UV-VIS Spectrophotometer (Synergy MX, BioTek, Winooski, VT, USA). The microbial 16S rRNA V3-V4 region was amplified by 341F (5′CCTACGGGNGGCWGCAG 3′) and 805R (5′GACTACHVGGGTATCTAATCC 3′) primers targeting the hypervariable ~460 bp region [17]. Paired-end read libraries (2 × 300 nt) were sequenced on Illumina’s (Illumina, San Diego, CA, USA) MiSeq platform using the Herculase II Fusion DNA Polymerase Nextera XT Index V2 Kit (Illumina, San Diego, CA, USA).

2.3. Bioinformatic and Statistical Analysis

The paired-end reads obtained after next-generation sequencing were processed and analyzed using Quantitative Insights Into Microbial Ecology 2 (QIIME 2 v2024.5) software [18]. The Cutadapt plugin in QIIME 2 was used to remove primer contaminations and Nextera adapters. Reads of the quality score above 30 were included for further analysis. Denoising, paired-end read merging, chimera removal of reads, and determination of amplicon sequence variants (ASVs) were performed using DADA2 [19] algorithm (--p-trim-left-f 17 --p-trim-left-r 21 --p-trunc-len-f 261 --p-trunc-len-r 220 --p-max-ee-f 5 --p-max-ee-r 5) as implemented in QIIME2. As mandated by QIIME 2, a metadata file was prepared reflecting the habitat, location, and type of samples (Table S1). Alpha rarefaction (--p-max-depth 70,800) was performed to normalize differences in library sizes among samples, thereby improving the accuracy of alpha diversity comparisons. Goods coverage value was calculated to determine if the sequencing depth was sufficient to capture all the microbial diversity. For core metric phylogenetic analysis --p-sampling-depth = 9900 was used based on the DADA2 feature table summary. It is the minimum sequence count across samples, ensuring that all included samples met this threshold for downstream analysis. The taxonomic composition of the samples was explored using a QIIME 2-compatible SSU SILVA reference database based on the curated NR99 (version 138.1) database [20]. RESCRIPt (REference Sequence annotation and CuRatIon Pipeline) was used to retrieve full-length sequences from the NCBI RefSeq database (version 138.1). These sequences were then reverse-transcribed and cleaned to remove low-quality or partial entries. The feature classifier plugin was used to extract specific sequences corresponding to the primer used in this study. The resulting sequences were further dereplicated using RESCRIPt, retaining unique sequences for downstream analysis [21]. In the further steps, the sequences were classified using a Naive Bayes classifier, followed by filtering to remove chloroplast and mitochondrial sequences, as removing cryptic organellar DNA is crucial for improving the reliability of the data [22].

The sample metadata, table, taxonomy, and tree were exported from the QIIME 2 environment and imported into R (v4.3.2) [23,24] for further statistical analyses after building a Phyloseq (v1.46.0) project [25]. Data analysis and visualization were carried out using the packages tidyverse (v2.0.0) [26] and vegan (v2.6-8) [27] in association with the phyloseq project.

Overall taxon diversity was examined by estimating the microbial richness and composition (alpha diversity) across host species and habitats, as well as the microbial differentiation between groups (beta diversity). These analyses helped to (1) identify the shared and unique microbial communities between freshwater and brackish water samples (both gut and water samples) and to (2) assess the species-specific and habitat-specific effects on microbiota composition.

Variations in alpha diversity of microbial communities in gut and water samples from brackish water and freshwater habitats were determined using different ecological diversity indices (Shannon’s diversity index, Observed Features, Faith’s Phylogenetic Diversity). After computing diversity metrics, we examined the microbial composition of the samples in relation to the sample metadata. Testing for associations between categorical metadata columns and alpha diversity data was performed for Faith Phylogenetic Diversity (a measure of community richness) and evenness metrics. These steps were carried out in R, using appropriate statistical and visualization packages.

Beta diversity was evaluated through weighted and unweighted UniFrac dissimilarity matrices, implemented in R software. Bray–Curtis dissimilarity (non-phylogenetic) matrices were used for comparison of the communities. Sample composition in the context of categorical metadata was analyzed using PERMANOVA, while the homogeneity of multivariate dispersions (PERMDISP) was also tested to evaluate differences in within-group variability. Both PERMANOVA and PERMDISP were calculated using weighted UniFrac distance matrices in QIIME 2.

Principal Coordinate Analysis (PCoA) was performed to visualize the patterns of microbial community structure and their relatedness based on distance metrices. Double Principal Coordinate Analysis (DPCoA) was used to incorporate phylogenetic relationships and species composition of the microbial community. DPCoA provides an integrated representation of samples and taxa, highlighting both species composition and their evolutionary aspects.

UniFrac quantifies the distance between microbial communities using the phylogenetic data of OTUs. Weighted UniFrac is a quantitative metric for β diversity that assesses variations in both the number of sequences from each lineage and the presence of different taxa.

The Microbiome package in R was used to examine the core microbiota of samples from freshwater and brackish water habitats. The UpSetR [28] package was used to prepare a plot highlighting the core microbiota. Differential abundance was calculated using the DESeq2 (v1.42.1) [26] and ANCOMBC (v2.4.0) [27] packages in R. Heatmap highlighting the 20 most divergent taxa was prepared using the R package Complex Heatmap (v2.17.0) [29].

3. Results

3.1. Taxon Composition

A comprehensive analysis of microbial communities across gut and water samples from freshwater and brackish water habitats revealed the 12 most dominant bacterial phyla (Figure 1b) and the 25 most dominant genera (Figure 1a) based on relative abundance. There is an increasing proportion of taxonomically unassigned ASV sequences as the host organisms were classified at more specific taxonomic levels, such as genus or species, compared to broader levels like phylum. Proteobacteria was the most abundant phylum based on the highest proportion of ASVs assigned to it and was consistently present across all samples from the two study sites, with particularly elevated abundance in brackish water. In contrast, Firmicutes was the second most abundant phylum observed in all samples, with considerable variation in relative abundance. Freshwater gut samples contained a markedly high proportion of Firmicutes, exceeding 45% whereas this phylum was present at substantially lower levels in the corresponding water samples. ASVs assigned to the phylum Fusobacteria were consistently detected across all gut samples, with a pronounced abundance in brackish water gut samples. Meanwhile, the phylum Actinobacteria was widely distributed in both freshwater and brackish water samples with comparable levels of relative abundance.

ASVs of Verrucomicrobiota dominated freshwater samples (>25%) but remained <5% in all other groups. Bacteroidota exhibited similar relative abundance in water samples from both habitats and were minimally represented in gut samples (<3%). Planctomycetota and Chloroflexi were more prevalent in freshwater habitats, the latter particularly enriched in freshwater gut samples. Armatimonadota was exclusively found in freshwater water samples, while Cyanobacteria were present in water samples from both habitats and at low levels in freshwater gut samples.

The relative abundance profiles of the most represented genera (Figure 1a) reveal that Cetobacterium, Romboutsia, Turicibacter, Clostridium_sensu_stricto_1, and Terrisporobacter were highly prevalent in the gut microbiota of fish from both freshwater and brackish water habitats. Freshwater gut samples were uniquely characterized by the presence of Bacillus, Paraclostridium, and Anoxybacillus. Genera such as Enterovibrio, Photobacterium, Vibrio were exclusively detected in the gut microbiota of the brackish water fish. Additionally, water samples from the brackish water habitat contained the Genera Candidatus Aquiluna and AEGEAN-169 marine group, IS–44, CL500-29 marine group, Candidatus Actinomarina as well as the NS5 marine group which are either absent or minimally represented in other samples. The genus hgcl clade was detected in water samples from both locations.

The total abundance of microbial phyla (Figure 2) indicates the relative abundance of each phylum across the samples. The highly abundant phyla included Proteobacteria, Verrucomicrobiota, Bacteroidota, Firmicutes, Fusobacteria, Actinobacteriota, and Chloroflexi. Among these major bacterial phyla observed, Proteobacteria exhibited the highest abundance and prevalence. Their relative abundance varied noticeably across samples ranging from as low as 10¹ to as high as 10⁵, and prevalence between 0.2 and 1 (fraction of samples). Other phyla in this group also displayed moderate to high abundance and prevalence. While taxa are prevalent in many samples, their abundance varies significantly across samples. In our samples, Abditibacteriota, Aenigmarchaeota, Ascomycota, FW 113, LCP-89, WOR-1, S2, Cloacimonadota, FCPU 426 are characterized by a total read count of less than 10 across all samples and prevalence found in only 10% of samples. Therefore, these phyla can be considered as rare and infrequent taxa. Other microbial phyla exhibited abundance and prevalence values that were intermediate between those of the highly abundant and rare phyla.

3.2. Alpha Diversity

A statistical comparison of alpha diversity metrics reveals a significant difference between the gut and water samples. Water samples exhibited higher observed richness (p = 0.041) and Shannon diversity (p = 0.026) compared to gut samples (Figure 3) reflecting the typically greater microbial diversity in environmental water. Water samples from freshwater habitat (FW) showed the highest richness, indicated by a narrow interquartile range, followed by those from brackish water habitat (BW). Both exhibited greater richness than their corresponding gut samples. The wider distribution of species richness in freshwater gut (FW gut) indicates greater variability among these communities.

In contrast, the brackish water gut (BW gut) exhibited the lowest richness, indicating the presence of a more specialized microbiota. FW gut and BW showed comparable Shannon diversity. Overall, the water samples (FW and BW) were more diverse than the gut samples (Figure 3). Among the water samples, those from freshwater habitats supported greater richness and diversity.

3.3. Beta Diversity

The Principal Coordinate Analysis (PCoA) plot (Figure 4) reveals clear clustering of gut and water samples from two habitats. The tight clustering of brackish water gut samples indicates a similar microbial composition among them. In contrast, the more dispersed distribution of remaining samples indicates greater variability in their microbial composition. Additionally, water samples from both habitats are separated from gut samples, highlighting that the microbial communities in the water are distinct from those in the fish gut. From the Double Principal Coordinate Analysis (DPCoA) plot (Figure S1a) [30,31], a phylogenetic ordination method that provides a biplot representation of both samples and taxonomic categories, the water samples from both habitats cluster tightly and distinctly from gut samples. This pattern suggests a lower within-group variability and a more conserved microbial community structure in water samples, whereas the gut samples exhibit a more dispersed distribution, reflecting higher intra-group variability. A few key microbial taxa are likely responsible for driving the observed community divergence between the gut and water samples from the two habitats.

PERMANOVA was performed to test for differences in microbial community composition among the four groups. Analysis revealed significant differences in community structure across groups (pseudo-F = 10.38, p = 0.001; Supplementary Table S2). Brackish water gut clusters more closely with brackish water suggesting a partial overlap in community structure, whereas freshwater gut microbiota are separated from other groups, showing a distinct microbial assemblage (Figure S2b). To evaluate whether PERMANOVA results were influenced by differences in within group variability PERMDISP was conducted. Box plot in Figure S2a displays the results of PERMDISP test, which evaluates the homogeneity of group dispersions. The results reveal differences in beta diversity dispersion between groups (F = 1.44, p = 0.101; Supplementary Table S2). Water samples exhibit tight clustering with low variability. Brackish water gut samples show moderate variability whereas freshwater gut samples exhibited greatest within group variability. The PERMDISP analysis showed no statistically significant homogeneity of group dispersion.

The upset plot (Figure 5) reveals a notable degree of uniqueness across different samples. The highest number of unique taxa was found in freshwater samples (FW = 3139), followed by brackish water (BW) with 2244 taxa and the freshwater fish gut sample (FW gut = 2176), while the brackish water fish gut sample (BW gut) contained fewer unique taxa (638), indicating a more restricted microbial assemblage. Freshwater gut and water samples share 426 taxa, whereas the shared taxa between brackish water and brackish water fish gut samples are limited to 36. Fifty-nine taxa are shared among gut samples from both habitats. On the other hand, brackish water and freshwater share only 12 taxa. There is minimal overlap between samples, with no more than nine taxa shared across any three-group combination. Strikingly, only one taxon was found to be present in all four sample types, indicating a highly limited core microbiome.

The heatmap (Figure 6) indicates the distinct composition of microbial taxa in gut and water samples. The gut-associated microbes include Escherichia-Shigella (Proteobacteria), Clostridium sensu stricto 13 (Firmicutes), Bacteroides vadinHA17 (Bacteroidota), Rhodobacter (Proteobacteria), Romboutsia (Firmicutes), Brevibacillus (Firmicutes), and Vibrio (Proteobacteria). Among these microbes, Clostridium sensu stricto 13, Vibrio, and Escherichia-Shigella were exclusively found in the gut of brackish water fish, whereas, water-associated microbes comprise the CL500-29 Marine Group (Actinobacteria), Rhodopirellula (Planctomycetota), Bacteroides vadinHA17 (Bacteroidota), env.OPS_17 (Bacteroidota), Luteolibacter (Verrucomicrobiota), hgcl_clade (Actinobacteria), Comamonadaceae (Proteobacteria), SH3-11 (Verrucomicrobiota), Fluviicola (Bacteroidota), Bryobacter (Acidobacteroita), Oscillatoria PCC-10802 (Cyanobacteria), and some uncultured bacteria belonging to the phylum Proteobacteria. Among the water-associated microbes, only Rhodopirellula and CL500-29 Marine Group dominated the brackish water habitats. Hierarchical clustering in the heatmap also shows grouping of samples with similar microbial profiles. Here, FFGUT03 (freshwater fish gut) is clustered with brackish water fish gut samples, while BFGUT003 (brackish water fish gut) is grouped with freshwater fish gut samples. Based on the heatmap, freshwater samples appear to have contributions from a wider range of phyla, such as Firmicutes, Bacteroidota, Verrucomicrobiota, Actinobacteriota, Acidobacteriota, and Proteobacteria, compared to brackish water samples, which are dominated mainly by two phyla: Actinobacteria and Planctomycetota.

All of the goods coverage values estimated were close to one, indicating most of the diversity has been captured. The alpha rarefaction curve (Figure S3) represents the sufficient sequencing depth to capture the diversity within samples. Some of the samples have a higher Shannon index, indicating greater richness and evenness, whereas samples with a lower Shannon index reveal a community dominated by fewer taxa.

4. Discussion

4.1. Habitat-Driven Divergence and Gut Microbiome Differentiation in Allopatric Populations of Etroplus suratensis

4.1.1. Distinct Microbial Profiles in Gut and Water Samples

The present study aimed to investigate the gut microbial composition and diversity of Etroplus suratensis in allopatry and compare them with the microbial communities of their corresponding habitat. We examined the gut microbiota of E. Suratensis from allopatric populations using bacterial 16S rRNA (V3-V4) gene sequencing.

Etroplus suratensis, a highly sought-after fish species, is extensively cultivated in the region and commands premium prices for farmers. However, a common perception exists that the freshwater populations of E. suratensis are less palatable than their brackish water counterparts. This study reveals significant differences in the core microbiota between freshwater and brackish water populations. Developing probiotic feeds for freshwater aquaculture of this species, an outcome of this research, could significantly increase its value.

In line with earlier research on fishes, the most abundant phyla observed in water samples were Proteobacteria, Actinobacteria, Verrucomicrobiota, Bacteroidota, Planctomycetota, Chloroflexi, Cyanobacteria, and Armatimonadota [1,32,33]. The dominant phyla in the gut samples observed were Fusobacteria and Firmicutes. The total abundance of microbial phyla (Figure 3) exhibits variation in distribution patterns, indicating niche-specific structuring of microbial communities. Differences in abundance patterns indicate the influence of habitat in shaping microbiota composition and dynamics.

4.1.2. Microbial Community Structure Across Habitats

Previous studies on fishes have shown that the composition of the gut microbiome is shaped primarily by the environment rather than genetic factors [6,32]. Alpha diversity analysis revealed variations in richness and evenness, with some samples exhibiting higher Shannon diversity indices, indicating differences in environmental exposure and dietary preferences. Thus, significant differences in the diversity metrics between gut and water samples suggest the selective nature of the fish gut environment. As shown in Figure 3, water samples from both habitats exhibited a higher observed richness (p = 0.041) and Shannon diversity (p = 0.026) compared to their corresponding gut samples. This pattern aligns with earlier studies on fishes showing that the environmental microbial communities exhibit greater diversity than that of the gut [1,33], possibly due to the absence of a host-driven microbial selection. The freshwater habitat supported the highest richness, indicated by both diversity indices and a narrow interquartile range, suggesting a more stable and diverse microbial community. The wider distribution in the richness of freshwater gut samples suggests inter-individual variability that may be influenced by factors such as a diverse diet, differential environmental exposure, or host physiology [34]. In contrast, brackish water gut samples exhibited the lowest richness and narrowest distribution, suggesting a more specialized microbial community. A similar pattern has been observed in euryhaline fish species, where gut microbes tend to be less diverse and more restricted in saline conditions compared to their freshwater counterparts [8].

Further calculation of the goods coverage value, which was close to one, indicates that the sequencing depth was sufficient to capture the microbial diversity in all samples. The alpha rarefaction curve (Figure S3) further confirms this conclusion.

Beta diversity analysis exhibited a profound difference in microbial composition between gut and water samples across two habitats. From PCoA and DPCoA plots (Figures 4 and Figure S1), the distinct clustering pattern suggests that both external and internal factors may influence the gut microbiota. The PCoA plot (Figure 4) demonstrates lower inter-individual variability and a relatively homogeneous microbiota in brackish water gut samples. In contrast, brackish water samples, as well as gut and water samples from freshwater habitats, exhibit dispersed clustering, indicating greater compositional variability within these groups. There is a clear separation between the gut and water samples, emphasizing the divergence between the host and environmental microbiomes. The results of PERMANOVA and PERMDISP (Figure S2a,b; Supplementary Table S2) support this finding. PERMANOVA exhibited significant differences in community composition indicating microbial communities are compositionally different across habitats and sample types, while PERMDISP indicated no significant differences in dispersion suggesting those differences are not due to unequal variability. Although the results are statistically significant, the small sample size limits the strength of this conclusion. The results of beta diversity analyses reinforces the idea that fish gut microbiota is not merely a reflection of the ambient microbial pool but rather represents a selectively filtered and distinct microbial community. Phylogenetic and compositional differentiation among microbial communities is evident in the DPCoA plot (Figure S1b). The dispersion of taxa associated with the gut indicates greater within-group variability and suggests functional heterogeneity in the gut microbiome [35,36]. The observed distinct clustering pattern suggests that external factors (environmental) or/and internal factors (diet, physiology) may have shaped the microbiota composition [37].

4.1.3. Selective Filtering and Taxonomic Differentiation of Gut and Water Microbiota

The upset plot (Figure 5) illustrates a detailed visualization of shared and unique taxa across different sample types. The fewer shared taxa between gut and water samples suggest a higher degree of selectivity in the gut microbiota. Diversity is higher among water samples compared to gut samples. Analysis of the core microbiome reveals that some microbes are habitat-specific, while others remain stable across diverse environments. The highest number of unique taxa in freshwater samples (3139) reflects the presence of a rich and diverse environmental microbial reservoir. A substantial number of unique taxa were also observed in freshwater gut samples (2176); however, the shared taxa between gut and water samples from freshwater habitats were limited to 426 taxa. This limited overlap indicates that the habitat-associated microbiota plays a major role in defining the gut community particularly in freshwater which is characterized by lower salinity and higher microbial diversity [38,39]. The unique taxa associated with brackish water samples were 2244, while the brackish water fish gut harbored only 638 unique taxa, which is notably lower than the unique taxa observed in all other samples.

Furthermore, only 36 taxa were shared between the brackish water and its corresponding gut samples. This suggests that gut microbial communities in brackish water habitats are more selective and act as a strong ecological filter, supporting only those environmental microbes capable of persisting under host-associated conditions such as increased salinity or other challenging physicochemical conditions [40,41,42,43]. The presence of 59 shared taxa between gut samples from both freshwater and brackish water habitats points to the existence of a conserved core gut microbiota in Etroplus suratensis. Despite the salinity difference and environmental microbial composition, these taxa are capable of surviving and persisting within the host gut. Such microbes may play essential functional roles—including nutrient metabolism, digestion, immunity, growth, and overall host health—in fishes [37,44,45]. Only a single taxon was present in all four sample types, highlighting the strong niche differentiation and specialization of microbial communities between habitats and gut samples. The sole shared taxon may represent a functionally critical or highly adaptable microorganism capable of surviving in all conditions [46].

The heatmap (Figure 6), with hierarchical clustering of both taxa and samples, demonstrates distinct microbial assemblages between gut and water samples, reflecting clear segregation between environmental and host-associated microbial communities. Gut samples were enriched with Escherichia-Shigella (Proteobacteria), Clostridium sensu stricto 13 (Firmicutes), Bacteroides vadinHA17 (Bacteroidota), Rhodobacter (Proteobacteria), Romboutsia (Firmicutes), Brevibacillus (Firmicutes), and Vibrio (Proteobacteria), which are consistent with typical gut-associated bacteria known to perform functions like nutrient digestion, fermentation of dietary substances, immune responses, and maintenance of tissue homeostasis [47,48,49,50]. In contrast, the water samples were dominated by bacterial taxa from different phyla which are all well-known members of free-living aquatic microbial communities [8,51].

The dominance of Cetobacterium in the gut samples from both habitats is consistent with previous studies highlighting its key metabolic roles in fish, reflecting its involvement in vitamin B₁₂ and acetate production, which contribute to host energy metabolism and glucose homeostasis [52]. Romboutsia, Turicibacter, Clostridium_sensu_stricto_1, and Terrisporobacter are generally associated with carbohydrate fermentation, amino acid metabolism, and the production of short-chain fatty acids [5,53,54]; additionally, Turicibacter is involved in the regulation of immune and inflammatory responses in fish [8]. The presence of Vibrio, Photobacterium, and Enterovibrio in brackish water fish guts reflects a community shift toward taxa adapted to marine environments. These genera produce enzymes, such as amylase, lipase, and chitinase, which support the breakdown of protein- and chitin-rich diets, although some strains are also opportunistic pathogens [5,55,56].

Epulopiscium and Paraclostridium are known to ferment carbohydrates and proteins, generating short-chain fatty acids (SCFAs) that act as an energy source for the host and help to maintain the gut health [57]. Lutibacter and Flavobacterium are involved in breaking down of complex polysaccharides and proteins, facilitating nutrient recycling and potentially suppressing pathogenic microbes through competitive exclusion [58,59]. Bacillus strains are not only enzyme-producing bacteria but also promising candidates for enhancing fish health and combating vibriosis in aquaculture systems [60,61].

Water samples harbored distinct environmental clades, including Candidatus Aquiluna, AEGEAN-169 marine group, CL500-29 marine group, and the NS5 marine group, which are known for dissolved organic matter degradation and niche partitioning in marine systems but are generally considered transient in fish guts [5,55]. The detection of the hgcl clade in both habitats further supports the role of surrounding water as a reservoir of microbial inputs to the gut.

Interestingly, the cross-habitat clustering of the freshwater sample ‘FFGUT03′ and the brackish water sample ‘BFGUT003′ indicates individual-level variation in microbiota composition, possibly due to differences in environmental exposure, diet, or physiological stress. This unexpected clustering pattern also suggests the plasticity of the microbiome in Etroplus suratensis. Additionally, the broader taxonomic diversity in freshwater samples reflects the greater microbial richness and complexity of the freshwater ecosystem. In contrast, brackish water samples are predominantly composed of Actinobacteria and Planctomycetota, indicating a more selective or constrained microbial habitat. These results align with other findings in the present study.

The primary limitation of this study is the small sample size (n = 3 per group), which constrains the statistical power of our analyses and the extent to which we can generalize these findings to entire populations. Therefore, these results should be interpreted as a strong preliminary characterization that establishes a baseline for future, larger-scale investigations. The ecological and functional significance of the gut microbiota in ecological success of populations sustaining in vastly different habitats also need to be explored further. Such efforts would ideally enable the integration of functional roles of gut microbiota into aquaculture management, thereby enhancing sustainability and productivity.

5. Conclusions

In summary, our study highlights the fundamental differences in gut microbiome divergence among allopatric populations of Etroplus suratensis, specifically in relation to their respective habitats. The freshwater habitat nurtures more diverse microbial communities, in contrast to the brackish water habitat, which exhibits lower diversity and higher homogeneity. Future research, including metagenomic/metatranscriptomic studies, could further elucidate the adaptive roles of specific gut microbes in relation to habitat conditions. Also, expanding sample sizes in future work will help confirm our findings and better capture individual-level variations more effectively.