A Dangerous Prey Fish? Brachyspira‐Rich Gastrointestinal Microbiota in Pompano Dolphinfish From Neritic and Deep Waters of the South China Sea

TL;DR

Pompano dolphinfish in the South China Sea have gut microbes dominated by Brachyspira, a potential pathogen that may spread from coastal to deep-sea ecosystems.

Contribution

The study identifies Brachyspira as a dominant gut microbe in dolphinfish and suggests a novel ecological route for pathogen dispersal in marine ecosystems.

Findings

Gut microbiota of pompano dolphinfish is simplified and dominated by Brachyspira in both neritic and deep-sea habitats.

Brachyspira was detected in nearshore seawater eDNA, suggesting coastal acquisition and offshore transport by fish.

Gut microbiota functions suggest adaptation to anthropogenic stressors and pathogenic processes.

Abstract

The gastrointestinal microbiota of marine fishes plays important roles in host physiology and ecosystem processes yet remains poorly characterized. Here, we used 16S rRNA metabarcoding to examine the gastrointestinal microbiota of pompano dolphinfish (Coryphaena equiselis
) collected from a neritic site (< 200 m) and a deep‐sea site (> 3800 m) in the South China Sea. Across sites, the microbial community was unexpectedly simplified and strongly dominated by Spirochaetes, particularly the genus Brachyspira. Because Brachyspira includes well‐documented pathogens of terrestrial animals, its high prevalence raises concern that pompano dolphinfish may act as reservoirs of potentially pathogenic bacteria with the capacity for trophic transfer. Coastal exposure may represent a critical source: Brachyspira was also detected at low levels in seawater eDNA from nearshore habitats, suggesting that dolphinfish could acquire these microbes during neritic stages and subsequently transport them into deep‐sea ecosystems. Functional predictions further revealed that gastrointestinal microbiota from both sites were closely associated with pathogenic processes, while neritic individuals additionally exhibited enrichment of functions linked to adaptation to anthropogenic stressors. Our findings underscore the ecological importance of dolphinfish microbiota as both a reservoir of potentially pathogenic lineages and a sensitive indicator of environmental stress in marine ecosystems.

This study reveals an unexpectedly simplified, Brachyspira‐dominated gastrointestinal microbiota in pompano dolphinfish from both neritic and deep‐sea habitats of the South China Sea. Environmental DNA and functional analyses suggest these fish may acquire potential pathogens in anthropogenically influenced coastal waters and transport them offshore, highlighting a novel ecological route for pathogen dispersal in marine ecosystems.

Figures

Pompano dolphinfish and seawater sampling in the South China Sea. (A) Sampling sites as indicated by the green dots, and (B) the picture of one collected Pompano dolphinfish.

Microbiota composition and diversity of pompano dolphinfish gastrointestinal contents and surrounding seawater from neritic (N) and deep‐water (D) sites. The figure compares host gut versus seawater communities and evaluates habitat‐associated differences in gut microbiota. (A) Stacked bar plots showing the relative abundance (%) of dominant bacterial classes in each sample. (B) UpSet plots illustrating the numbers of shared and unique bacteria zOTUs between the gastrointestinal content of Pompano dolphinfish and seawater at two sites. (C) PCoA chart of gut microbiota (zOTU level) based on Bray–Curtis dissimilarity; group differences between sites assessed by ANOSIM (R and p shown). (D) Alpha diversity between the gastrointestinal microbiota of Pompano dolphinfishes at two sites (ZOTU level), quantified using Chao1 and Observed zOTUs (richness) and Shannon and Simpson indices (diversity). Boxplots show the median and interquartile range (IQR), with individual points representing samples. Statistical differences were evaluated using a Wilcoxon rank‐sum test (ns, not significant; *p < 0.05; p < 0.01; *p < 0.001).

PICRUSt2‐predicted functional profiles of dolphinfish gut microbiota at neritic (N) and deep‐water (D) sites. (A) Spearman's rank correlation heatmap showing associations between the gastrointestinal microbiota at the class level and the top 30 predicted KEGG pathways across samples. (B) Error‐bar plot of the top 20 predicted KEGG pathways in gastrointestinal samples from the two sites. Differentially abundant predicted pathways between neritic and deep‐water groups were identified using edgeR (negative binomial generalized linear model), with multiple‐testing correction performed using the Benjamini‐Hochberg (BH) procedure.

PCA based on the Bray Curtis distance of KEGG pathways of the gastrointestinal microbiota of Pompano dolphinfishes at two sites, the peaks represent marginal density curves of the sample ordination coordinates on the X‐axis (PC1; top) and Y‐axis (PC2; right).

Full text

Introduction

1

The structure of marine ecosystems is influenced by biological interactions, environmental conditions, and human activities (Levin and Lubchenco 2008). In marine ecosystems, fish play a pivotal role—not only as consumers and prey, but also as hosts of diverse microbial communities that regulate essential ecological processes (Pikitch et al. 2014; Egerton et al. 2018). The gastrointestinal microbiota of fish is now recognized as a critical determinant of host physiology, influencing digestion, nutrient absorption, immune modulation, and disease resistance (Egerton et al. 2018). The gastrointestinal microbiota of marine fish may act as reservoirs or vectors of potential pathogens, linking microbial ecology to ecosystem functioning and public health (Senderovich et al. 2010). Despite these recognized roles, the gastrointestinal microbiota of many pelagic fishes remains poorly studied, particularly in relation to potential pathogenic taxa.

Community composition of the gastrointestinal microbiota is influenced by host phylogeny, physiology, and environment (Sullam et al. 2012). Carnivorous fishes often exhibit relatively simplified gastrointestinal microbiota compared to omnivorous or herbivorous fishes, reflecting the strong influence of host trophic level and phylogeny on gastrointestinal microbiota structure and diversity (Liu et al. 2016). In many fishes, gut microbiota are typically dominated by a limited number of bacterial phyla, most commonly Proteobacteria, Firmicutes, Bacteroidota, Actinobacteriota, and Fusobacteriota (Sullam et al. 2012; Wang et al. 2018). At finer taxonomic resolution, representative genera frequently reported include Vibrio, and Photobacterium, (Proteobacteria), Lactobacillus and Clostridium (Firmicutes), Bacteroides (Bacteroidota), as well as Cetobacterium (Fusobacteriota), many of which are involved in nutrient metabolism, fermentation, and host immune modulation (Nayak 2010; Egerton et al. 2018; Wang et al. 2018). These broadly conserved patterns provide an ecological baseline for interpreting deviations from typical gut community structures, particularly when unusual dominance by specific bacterial lineages is observed.

The ecological implications of unusual enrichment of potential pathogens extend beyond individual fish health (Riiser et al. 2020; Spilsbury et al. 2022). Marine fish are central prey for larger predators, including tunas, cetaceans, and seabirds, meaning that gastrointestinal microbes may be transferred through the food web. Recent studies have shown that fish‐associated gut microbiota integrate signals from environmental exposure, diet, and anthropogenic stress, thereby linking host‐associated microbial communities to ecosystem‐level disease dynamics rather than isolated host outcomes (Riiser et al. 2020; Suzzi et al. 2022; Aeby et al. 2024). Opportunistic taxa with pathogenic potential, once enriched in fish gastrointestinal tracts, may propagate through predator–prey interactions or be released into the environment via defecation, influencing microbial circulation across broader spatial and trophic scales. Such processes are increasingly viewed through a One Health framework, in which wildlife microbiomes intersect with ecosystem stability and human health concerns, particularly in coastal systems exposed to intense anthropogenic pressure (Zinsstag et al. 2011; Aeby et al. 2024). One possibility is that neritic habitats—where human activity, seabird inputs, and coastal exposure are concentrated—serve as important sources of potential pathogens, which are subsequently acquired by fish and transported offshore during ontogenetic habitat shifts. If verified, this “coastal exposure–deep‐sea transport” pathway would represent a mechanism linking neritic and deep‐sea microbial dynamics.

The South China Sea is one of the most biologically productive and diverse marine regions globally but is also subject to intense anthropogenic pressures (Jiao et al. 2015; Guo et al. 2019; Tao et al. 2021; Liu et al. 2024). Coastal zones experience nutrient loading, aquaculture effluents, and chemical pollutants, while offshore regions remain comparatively less disturbed and stable (Hughes et al. 2013; Tao et al. 2021; Liu et al. 2024). Such environmental gradients provide an opportunity to investigate how human impacts may be reflected in host‐associated microbial communities. For instance, microbes capable of xenobiotic degradation—such as those involved in breaking down hydrocarbons, pesticides, or industrial chemicals—are often enriched in organisms inhabiting polluted environments (Kimes et al. 2014; Spilsbury et al. 2022). If detected in fish gastrointestinal microbiota, these functions could serve as sensitive bioindicators of anthropogenic influence.

The pompano dolphinfish ( Coryphaena equiselis ), a smaller species conformis of the widely studied common dolphinfish ( Coryphaena hippurus ), inhabits tropical and subtropical waters worldwide (Xu et al. 2018). Although pompano dolphinfish is less commercially targeted, it is an ecologically important predator and a potential prey fish for larger fishes, seabirds, and marine mammals (Pitman and Stinchcomb 2002; Mancini and Bugoni 2014; Gardieff 2025). Its wide distribution across coastal and offshore habitats provides a valuable model for examining how host‐associated microbial communities are structured under differing environmental contexts (Xu et al. 2018). Despite this ecological relevance, pompano dolphinfish has received limited scientific attention, and in particular, its gastrointestinal microbiota remains largely unexplored. To date, only a few studies have investigated gut microbial communities in dolphinfishes, with most available data focusing on the common dolphinfish rather than pompano dolphinfish. Existing studies on dolphinfishes indicate that gut microbial community composition can vary substantially across different oceanic regions, sampling periods, and methodological approaches, suggesting that dolphinfish‐associated microbiota are context dependent rather than taxonomically fixed (Givens et al. 2015; Varela et al. 2025). This variability highlights the paucity of data for than pompano dolphinfish and constrains our understanding of species‐specific microbial assemblages and their potential roles in trophic‐mediated microbial transmission within marine food webs.

In this study, we investigated the gastrointestinal microbiota of pompano dolphinfish collected from two distinct sites in the South China Sea—a neritic site and a deep‐sea site. Using 16S rRNA metabarcoding, we assessed microbial diversity, taxonomic composition, and predicted functional profiles and compared them with surrounding seawater microbiota. Our results can provide new insights into the gastrointestinal microbial ecology of pompano dolphinfish and its possible role as a vector of microorganisms with ecological and health relevance.

Materials and Methods

2

Sampling Site and Specimen Collection

2.1

Specimens of pompano dolphinfish (n = 10; body length 17.5–20.4 cm) were collected on 22 and 26 August 2024 from two sites in the South China Sea (Table S1). The deep‐sea site (Site D; 18°25.818′ N, 117°25.980′ E) had a water depth > 3800 m, whereas the neritic site (Site N; 19°30.848′ N, 112°25.423′ E) was < 200 m deep (Figure 1A). Five individuals were captured at each site using a hand‐operated dip net (scoop net) from a licensed fishing vessel (Figure 1B). Fish were immediately placed in pre‐chilled biological sample bags, flash‐frozen in −20°C ultra‐low temperature freezers within 15 min of capture, and maintained at −20°C during transit using dry ice‐cooled containers. Transfer to −80°C laboratory storage was completed within 8 h post‐landing.

Pompano dolphinfish and seawater sampling in the South China Sea. (A) Sampling sites as indicated by the green dots, and (B) the picture of one collected Pompano dolphinfish.

Before dissection, each fish was thoroughly rinsed with sterilized deionized water to remove surface contaminants. The gastrointestinal tract was then carefully excised using sterilized forceps and scalpels. A new sterile blade was used for each dissection. The gastrointestinal tract contents were removed and processed under aseptic conditions, gently homogenized, transferred into sterile cryogenic tubes, and immediately stored at −80°C until DNA extraction. This procedure followed established best‐practice protocols for gut microbiome studies, which emphasize sterile handling and rapid low‐temperature preservation to minimize compositional bias during sample processing (Knight et al. 2018; Song et al. 2016; Marotz et al. 2021). Strict aseptic techniques were employed throughout the procedure: the dissection tray was lined with new aluminum foil for each fish and disinfected with 75% ethanol between dissections; forceps were sterilized with 75% ethanol between samples, gloves and knives were changed after handling each individual to prevent cross‐contamination. During transportation, gastrointestinal content samples were not suspended in bacterial saline or other preservation media. Instead, samples were flash‐frozen shortly after collection and maintained at −20°C during transport using dry ice‐cooled containers, followed by transfer to −80°C laboratory storage within 8 h post‐landing (Song et al. 2016; Marotz et al. 2021).

At each site, 30 L of surface seawater was collected for environmental DNA (eDNA) analysis. The water was divided into two 15 L subsamples, each filtered and enriched with a separate eDNA filter kit (E‐genomics Technology, Nanjing, China), getting two independent seawater samples per site.

DNA Extraction and PCR Amplification

2.2

For bacterial community profiling, the overall workflow (DNA extraction, 16S rRNA gene amplification, library preparation, sequencing, and downstream bioinformatic analyses) was based on established 16S rRNA amplicon sequencing practices and validated Ion Torrent/Ion Proton manufacturer protocols, with minor adjustments to accommodate sample characteristics and platform‐specific requirements (Pollock et al. 2018).

DNA was extracted from gastrointestinal contents and seawater samples using different protocols because these sample types represent distinct matrices with different biomass levels and potential co‐extracted inhibitors (gut contents vs. low‐biomass eDNA on membrane filters). For the 10 fish gastrointestinal samples, DNA was extracted with the E.Z.N.A. Tissue DNA Kit (Omega Bio‐tek, Norcross, GA, USA). For the 4 seawater eDNA samples, DNA was extracted using the Environmental DNA Extraction Kit (MT058; E‐genomics Technology, Nanjing, China). Extracted DNA was quantified with a Qubit 4 Fluorometer (Q33226; Invitrogen, USA), and purity was assessed by measuring A260/280 nm and A260/230 nm ratios with a NanoDrop ND‐5000 spectrophotometer (Thermo Fisher Scientific, USA).

For characterization of bacterial communities, the V3 region of 16S rRNA gene was amplified using the EGN P3 primer set (a modified primer pair targeting the V3 region; forward: ACCTACGGGRSGCWGCAG, reverse: TTACCGCGGCKGCTGG), generating a targeted amplicon of approximately ~150 bp (Lv et al. 2023). PCR amplification was performed in 30 μL reactions containing 15 μL of 2× Taq Mix, 1.5 μL of each primer (10 μM), 1 μL of template DNA, and 11 μL of nuclease‐free water. Amplification was conducted on a thermal cycler (TC‐96/G/H (b) B; TaKaRa, Japan) with a heated lid at 105°C, initial denaturation at 95°C for 3 min, followed by 28 cycles of 95°C for 15 s, annealing at 64°C for 30 s, extension at 72°Cfor 15 s, and a final extension at 72°C for 5 min, then held at 4°C.

To minimize stochastic PCR errors, each sample was amplified in triplicate. PCR products were verified on 2% (w/v) agarose gels, visualized with a Gel Doc EZ system (735BR00194; Bio‐Rad, USA), and then pooled equimolarly using an automated liquid‐handling workstation. Pooled amplicons were purified using VAHTS DNA Clean Beads (N411‐02; Vazyme, China) and re‐quantified with a Qubit 4 Fluorometer.

DNA Sequencing

2.3

Sequencing libraries were prepared with the VAHTS Universal DNA Library Prep Kit for Ion Torrent V2 (ND702; Vazyme, China) according to the manufacturer's protocol. Final barcoded libraries were normalized to 100 pM, and template‐positive Ion Sphere Particles were generated and enriched using the Ion OneTouch 2 System and Ion OneTouch ES (Thermo Fisher Scientific, USA). Sequencing was performed on a Proton semiconductor sequencer using the Ion PI Chip Kit V3, following the corresponding Ion Torrent sequencing protocols.

Bioinformatics Analysis

2.4

High‐throughput sequencing data were processed on an Ubuntu 14.04 platform using EcoView v3.0 analyze system (E‐genomics Technology, Nanjing, China), which implements established marker‐gene amplicon analysis steps commonly used in QIIME2‐style workflows (Bolyen et al. 2019; Shuai et al. 2025). Raw reads were quality‐filtered in VSEARCH (v2.21.1) with a minimum base quality threshold of Q30, following commonly adopted quality‐filtering practices for amplicon sequencing (Bokulich et al. 2013; Rognes et al. 2016). Reads were then demultiplexed using unique 12‐bp sample tags, allowing up to two mismatches, consistent with standard barcode demultiplexing settings used in widely adopted microbiome pipelines (Caporaso et al. 2010; Bolyen et al. 2019). Denoising was performed using the UNOISE3 algorithm implemented in VSEARCH (v2.21.1) with the “‐‐cluster_unoise” command to zero‐radius operational taxonomic units (zOTUs) (Edgar 2016; Rognes et al. 2016). Chimeric sequences were identified and removed using VSEARCH (v2.21.1) with the “‐‐uchime3_denovo” command (Rognes et al. 2016). Taxonomic annotation was initially conducted against the SILVA (v 138.2), and subsequently refined and supplemented using BLAST searches in NCBI to improve annotation accuracy. Contaminant sequences based on negative control and singleton were removed prior to downstream analysis and minimize sequencing noise and contamination (Bokulich et al. 2013; Davis et al. 2018). Sequence counts were normalized according to the minimum sum count across the given zOTU table across samples prior to downstream diversity analyses (Weiss et al. 2017). Unless otherwise stated, default parameters of the respective software tools were used.

Data Analyses

2.5

Microbiome alpha diversity and class‐level relative abundances were computed and visualized using the microeco package in R (v4.3.1), with statistical significance in alpha diversity assessed by Wilcoxon rank‐sum tests (Liu et al. 2020). Principal coordinates analysis (PCoA) and analysis of similarities (ANOSIM) were conducted with the vegan package and visualized using the ggplot2 package (Ginestet 2011). UpSet plot was performed using the UpSetR package (Conway et al. 2017). Prokaryotic KEGG functions were predicted from 16S rRNA gene taxonomies using PICRUSt2 (Douglas et al. 2020), generating abundance profiles of KEGG Orthologs (KOs). Differential pathway enrichment among pompano dolphinfish groups was assessed with the ggpicrust2 package using the edgeR method and visualized with the ggplot2 package (Yang et al. 2023; Ginestet 2011). Spearman's rank correlation heatmap was analyzed and visualized using the OECloud tools at https://cloud.oebiotech.com.

Results

3

Composition and Differences Between Gastrointestinal and Seawater Microbiota

3.1

In order to compare the microbial communities of pompano dolphinfish with their surrounding environment, gastrointestinal content DNA from 10 individuals and eDNA from seawater were subjected to metabarcoding sequencing. A total of 1,279,331 reads were obtained from the gastrointestinal contents of 10 pompano dolphinfishes. The gastrointestinal microbiota was dominated by three bacterial classes: Brachyspirae, Brevinematae, and Gammaproteobacteria. Brachyspirae represented over half of the community (average 53.84%), with higher abundance at deep‐sea site (60.84%) than neritic site (46.84%). Brevinematae contributed 23.68% on average, increasing from 15.23% at neritic site to 32.12% at deep‐sea site. Gammaproteobacteria were more abundant in the community of neritic site (27.03%), dominated by Shewanella (14.99%) and Photobacterium (10.09%), whereas Gammaproteobacteria represented only 2.23% of the community at the deep site (Figure 2A). These patterns indicate a simplified community with site‐specific differences in dominant bacterial classes.

*Microbiota composition and diversity of pompano dolphinfish gastrointestinal contents and surrounding seawater from neritic (N) and deep‐water (D) sites. The figure compares host gut versus seawater communities and evaluates habitat‐associated differences in gut microbiota. (A) Stacked bar plots showing the relative abundance (%) of dominant bacterial classes in each sample. (B) UpSet plots illustrating the numbers of shared and unique bacteria zOTUs between the gastrointestinal content of Pompano dolphinfish and seawater at two sites. (C) PCoA chart of gut microbiota (zOTU level) based on Bray–Curtis dissimilarity; group differences between sites assessed by ANOSIM (R and p shown). (D) Alpha diversity between the gastrointestinal microbiota of Pompano dolphinfishes at two sites (ZOTU level), quantified using Chao1 and Observed zOTUs (richness) and Shannon and Simpson indices (diversity). Boxplots show the median and interquartile range (IQR), with individual points representing samples. Statistical differences were evaluated using a Wilcoxon rank‐sum test (ns, not significant; *p < 0.05; **p < 0.01; **p < 0.001).

However, seawater microbiota (512,228 reads from 4 samples) was more diverse and evenly distributed, being composed of 6 major bacterial classes: Gammaproteobacteria, Alphaproteobacteria, Flavobacteriia, Synechococcophycideae, Betaproteobacteria and Deltaproteobacteria. Gammaproteobacteria averaged 28.78% of the community, but their taxonomic composition differed from the dolphinfish gastrointestinal contents, being dominated by Moraxellaceae at neritic site (43.00%) and Candidatus Portiera at deep‐sea site (25.06%). Alphaproteobacteria and Flavobacteriia contributed 18.34% and 16.72% on average, respectively, with Weeksellaceae, Flavobacteriaceae, and Prochlorococcus representing dominant taxa across sites. Betaproteobacteria and Deltaproteobacteria were minor components (< 5%) (Figure 2A).

Only 170 zOTUs were detected in gastrointestinal contents (159 at the neritic site; 106 at the deep‐sea site), whereas seawater microbiota comprised 500 zOTUs (461 at the neritic site; 404 at the deep‐sea site) (Figure 2B). A total of 115 zOTUs were shared between seawater and gastrointestinal microbiota. However, none of these shared zOTUs were dominant in both habitats: the gastrointestinal tracts were overwhelmingly dominated by Spirochaetes, particularly Brachyspira sp. (43.19% in gastrointestinal tracts vs. 0.30% in seawater) and Brevinemataceae (22.95% in gastrointestinal tracts vs. 0.09% in seawater), whereas seawater was dominated by cyanobacteria such as Prochlorococcus sp. (0.05% in gastrointestinal tracts vs. 10.61% in seawater) and Synechococcus sp. (0.09% vs. 0.71%). Notably, the dominant Spirochaetes in the gastrointestinal tracts (Brachyspirae and Brevinematae) were detected only in seawater eDNA from the neritic site (N), but not from the deep‐sea site (D). Overall, these data highlight the selective enrichment of specific bacteria within the dolphinfish gastrointestinal tracts.

Differences in Gastrointestinal Microbiota Between Sampling Sites

3.2

In order to examine how gastrointestinal microbiota varied between neritic and deep‐sea sites, we compared community composition and diversity across sampling sites. At both sites, Brachyspirae was the most abundant class, followed by Gammaproteobacteria and Brevinematae. Of the 106 zOTUs detected in the gastrointestinal contents at the deep‐sea site, 95 overlapped with those at the neritic site, indicating partial similarity in community composition (Figure 2D). Principal coordinates analysis (PCoA) based on Bray–Curtis distances revealed clear clustering by site, with samples at the deep‐sea site being more tightly grouped than those at the neritic site. PCo1 and PCo2 explained 43.77% and 35.65% of the variation, respectively (Figure 2C). ANOSIM analysis further confirmed the significant differences between the two sites (p = 0.005, r = 0.268).

Alpha‐diversity indices indicated adequate sequencing depth, with coverage values close to 1.000 for all samples. Richness indices (Chao1, Observed) were higher in the gastrointestinal microbiota at the neritic site compared to the deep‐sea site, and diversity indices (Shannon, Simpson) also trended higher at the neritic site (Figure 2D). These trends suggest potential biological differences in community structure between the two sites. However, Wilcoxon rank‐sum tests indicated significantly higher richness at the neritic site, as reflected by Chao1 and Observed zOTUs (p < 0.05), whereas Shannon and Simpson indices showed no significant differences (p > 0.05), likely due to high within‐group variability in community evenness, particularly among neritic‐site individuals. Overall, the two groups exhibited significant yet convergent patterns in gastrointestinal microbial composition.

Functional Predictions of the Gastrointestinal Microbiota

3.3

In order to explore the potential ecological functions of the gastrointestinal microbiota, we conducted functional inference using PICRUSt2. At the class level, Brachyspirae was strongly associated with infectious disease pathways (both bacterial and viral), while Brevinematae showed stronger correlations with digestive system‐related pathways (Figure 3A). PCA of predicted KEGG pathway profiles demonstrated clear clustering by site, with samples at deep‐sea site being tightly grouped but those at neritic site being more variable (Figure 4).

PICRUSt2‐predicted functional profiles of dolphinfish gut microbiota at neritic (N) and deep‐water (D) sites. (A) Spearman's rank correlation heatmap showing associations between the gastrointestinal microbiota at the class level and the top 30 predicted KEGG pathways across samples. (B) Error‐bar plot of the top 20 predicted KEGG pathways in gastrointestinal samples from the two sites. Differentially abundant predicted pathways between neritic and deep‐water groups were identified using edgeR (negative binomial generalized linear model), with multiple‐testing correction performed using the Benjamini‐Hochberg (BH) procedure.

PCA based on the Bray Curtis distance of KEGG pathways of the gastrointestinal microbiota of Pompano dolphinfishes at two sites, the peaks represent marginal density curves of the sample ordination coordinates on the X‐axis (PC1; top) and Y‐axis (PC2; right).

The most notable distinctions between the gastrointestinal microbiota of neritic and deep‐sea dolphinfish involved xenobiotic biodegradation and infectious disease‐related pathways. Among the top 20 significantly different pathways, six were related to xenobiotic degradation, including polycyclic aromatic hydrocarbon degradation, xylene degradation, bisphenol degradation, fluorobenzoate degradation, ethylbenzene degradation, and atrazine degradation, all of which were significantly more enriched in the gastrointestinal microbiota at the neritic site. In addition, two host‐interaction/opportunistic virulence–associated annotated pathways were enriched in the neritic group: bacterial invasion of epithelial cells and Chagas disease (Figure 3B). These results suggest that neritic‐water fish harbor microbiota with enhanced functional capacities for pollutant degradation and potential pathogen interactions, reflecting stronger human disturbances in neritic habitats.

Discussion

4

Host‐Driven Assembly of the Gastrointestinal Microbiota

4.1

The limited congruence between the gastrointestinal microbiota of pompano dolphinfish and the surrounding seawater communities underscores the importance of host‐related drivers in structuring intestinal microbial ecosystems, with minimal zOTU overlap. This pattern is consistent with recent cross‐species and habitat‐scale studies demonstrating that fish gut microbiota differ markedly from ambient water and are primarily shaped by host habitat, diet, and physiological filtering rather than passive environmental acquisition (Kim et al. 2021; Sadeghi et al. 2023; Varela et al. 2025). Several host‐specific mechanisms may contribute. First, the dolphinfish gastrointestinal tract provides a low‐oxygen, nutrient‐rich, and immune‐modulated environment, favoring anaerobic and facultative fermenters while excluding photoautotrophic and oligotrophic taxa typical of surface waters (Pérez et al. 2010; Wu et al. 2025; Tolas et al. 2025). Second, host genetics have been shown to influence mucus glycan composition and the immune components in fish, and affect pathogen binding (Thomsson et al. 2022; de Bruijn et al. 2017). Third, trophic interactions introduce prey‐ and environment‐derived microbes into the gastrointestinal tract, but only a subset can persist. Rather than a wholesale transfer, the host gut environment imposes strong filtering effects, favoring lineages capable of adapting to digestive and immune conditions while excluding transient allochthonous taxa (Romero et al. 2014; Vatsos 2017). Similar patterns have been observed where feed‐borne microbial signals were detectable but largely transient, with persistence depending on host growth rate and gut selection processes (Jin et al. 2022; Llewellyn et al. 2014).

This host‐dominated assembly process highlights the limited utility of seawater microbiome surveys in predicting gastrointestinal microbial states in pelagic predators. Therefore, the dolphinfish gastrointestinal tract should be generally viewed as a distinct ecological niche, influenced by but not determined by the environmental microbial pool.

Simplification and Dominance of Spirochaetes and Their Potential Origin

4.2

A notable and somewhat unexpected feature of the pompano dolphinfish microbiota was its extreme simplification, with members of the phylum Spirochaetota dominating both neritic‐water and deep‐water individuals. In contrast to the typically diverse assemblages seen in marine fishes—often comprising mixed Proteobacteria, Firmicutes, and Bacteroidota (Ray et al. 2012; Llewellyn et al. 2014)– the dolphinfish gastrointestinal tract was heavily skewed toward Spirochaetes, particularly the genus Brachyspira. This pattern is consistent with recent syntheses and comparative surveys showing that fish gut microbiota is highly context dependent and can become markedly simplified under particular ecological conditions or stressors (Luan et al. 2023; Xavier et al. 2024; Domingo‐Bretón et al. 2025).

Previous studies have reported the unusual prevalence of Spirochaetes in the gastrointestinal tracts of certain marine fishes, such as common dolphinfish ( Coryphaena hippurus ) and Atlantic cod ( Gadus morhua ) (Givens et al. 2015; Riiser et al. 2020). Recent microbiome datasets for pelagic fishes, including common dolphinfish, indicate substantial variability among host species and sampling contexts, suggesting that strong dominance by any single lineage may not be universal across closely related pelagic hosts or regions (Varela et al. 2025). Of particular concern is the genus Brachyspira, a lineage well documented as a pathogen in terrestrial animals, including pigs and poultry, where it can cause colitis and growth depression (Hampson et al. 2019; Rubin and Rohde 2022; Abd El‐Ghany 2025). Its high relative abundance in some marine fish microbiota raises concerns about its ecological role, impact on host health, and potential transmission to higher predators. In our dataset, the dominant Brachyspira‐associated zOTU (zOTU_1) showed the closest match to the reference sequence Brachyspira sp. AN3617:2/1/02 (GenBank JF430758), originally isolated from a free‐living mallard in Sweden. Notably, this isolate has a close phylogenetic affinity to a pathogen— B. pilosicoli (Jansson et al. 2011).

The ecological plausibility of this dominance warrants careful interpretation. Spirochaetes, particularly Brachyspira, are well‐known anaerobic, motile bacteria that can bind to intestinal mucus and migrate toward mucin gradients, and possess enzymes capable of degrading mucosal glycans as carbon sources (Naresh and Hampson 2010; Nakamura et al. 2006; Hampson and Ahmed 2009). In carnivorous hosts with short intestines such as dolphinfish, gastrointestinal tract conditions often favor fast‐growing, mucosa‐associated taxa, and spirochaetes may exploit these niches effectively (Li et al. 2021). Moreover, recent evidence shows that spirochaete enrichment can occur under environmental perturbations (e.g., extreme warming), supporting a scenario in which spirochaete dominance reflects context‐dependent microbiome restructuring (Domingo‐Bretón et al. 2025).

Furthermore, the potential origin of these Brachyspira warrants consideration. Other results provide evidence for ecological sources and trophic transfer of Brachyspira. Among 11 flat needlefish ( Ablennes hians ) collected around Hainan Island, one individual harbored a gastrointestinal microbiota with 11.03% relative abundance of Brachyspirae, whereas the remaining specimens showed no detectable presence of this group. DNA metabarcoding of this same individual's stomach contents revealed consumption of pompano dolphinfish, implying the possibility that Brachyspirae can be transmitted via predation (Table S3). This indicates that pompano dolphinfish may act as potential vectors, introducing Brachyspira into higher trophic levels of marine food webs (Korry and Belenky 2023).

Our unpublished eDNA data showed that low‐abundance Spirochaetota were consistently detected in seawater samples from the coastal waters of northern South China Sea, and specifically Brachyspira was detected in an eDNA sample collected near a coastal wind farm. Likewise, Brachyspira was detected in the seawater eDNA from the neritic site (N) at a relative abundance of 0.67%, whereas it was not detected at the deep‐sea site (D) (Table S2). These observations suggest that pompano dolphinfish may encounter Brachyspira during their neritic stages in coastal environments influenced by human activities, potentially enriching and transporting them offshore during ontogenetic shifts in habitat use (Song et al. 2025).

At the same time, seabirds may serve as a plausible origin, as they can act as mobile reservoirs and dispersal vectors for enteric bacteria through frequent fecal deposition into coastal and offshore waters. Recent studies show that seabird guano inputs can measurably alter microbial growth and community composition, and that wild waterbirds contribute detectable fecal pollution signals in coastal environments (Justel‐Díez et al. 2023; Signa et al. 2021; Boukerb et al. 2021; Fulke et al. 2025). Consistent with this pathway, Brachyspira have been documented in wild seabirds (Jansson et al. 2015; Abd El‐Ghany 2025). While strain‐level pathogenicity cannot be inferred from amplicon data alone, environmental persistence and waterborne exposure routes are plausible for enteric microbes, and Brachyspira survival in water has been demonstrated experimentally (Sosah et al. 2025; Oxberry et al. 1998). Accordingly, contaminated waters could serve as transmission hubs, and analogous bird‐mediated spillover dynamics have been documented in other marine contexts (Webby and Uyeki 2024).

Functional Specialization, Anthropogenic Signatures and Bioindicator Potential

4.3

Functional prediction of gastrointestinal microbiota has become an increasingly valuable tool to link community composition with ecological processes. Tools such as PICRUSt2 allow inference of metabolic potential from 16S rRNA data, highlighting pathways related to nutrient metabolism, host immunity, infectious disease, and pollutant degradation (Douglas et al. 2020).

Although taxonomic profiles differed only modestly between neritic and deep‐water dolphinfish microbiota, functional predictions revealed striking divergence. Neritic‐water individuals harbored microbiota enriched for xenobiotic degradation pathways and infectious disease‐associated functions. Enrichment of pathways for polycyclic aromatic hydrocarbons (PAHs), bisphenols, and chlorinated aromatics strongly suggests the adaptation to pollutant exposure in the nearshore areas heavily affected by industrial runoff, shipping, and aquaculture.

This functional signature is particularly notable because the neritic‐water sampling site (N) was located < 1.5 nautical miles from the Wenchang offshore drilling platform WC15‐1A. While the taxonomic composition of dolphinfish gut microbiota appears strongly host‐filtered and largely distinct from surrounding seawater, the enrichment of pollutant‐degrading pathways is broadly consistent with microbial responses reported from hydrocarbon‐impacted marine environments, where hydrocarbonoclastic taxa and catabolic functions can become enriched (Achberger et al. 2021; Wang et al. 2025). In this way, the gastrointestinal microbiota may encode a functional record of anthropogenic pressure, even when its taxonomic structure remains host‐dominated. Similar enrichment patterns have been documented in other coastal fish species, where gut microbes harbor catabolic repertoires enabling degradation of anthropogenic compounds (Nikouli et al. 2018; Walter et al. 2019; Magnuson et al. 2023). Whether these functions confer protective benefits to the host by reducing toxin burden, or simply indicate opportunistic colonization by pollutant‐degrading bacteria, remains unanswered. Nevertheless, their consistent presence near areas of human activity underscores the potential of fish‐associated microbiota as sensitive bioindicators of special anthropogenic stressors (Suzzi et al. 2022; Spilsbury et al. 2022). Although site‐specific contaminant measurements were not available for the present sampling, the predicted functional signal—together with the anthropogenic setting—supports the hypothesis that local exposure to petroleum‐related inputs may contribute to the observed enrichment. Future work combining microbiome profiling with targeted chemical measurements would help further evaluate this linkage.

Infectious disease pathways enriched in neritic‐water fish further support this interpretation, as coastal ecosystems are hotspots of pathogen transmission due to high host density, eutrophication, and pollutant‐driven stress (Aeby et al. 2024). Specifically, the neritic group was enriched for the pathway bacterial invasion of epithelial cells (Figure 3B). This enrichment may be partly driven by the higher relative abundance of Vibrionaceae‐associated zOTUs in neritic fish, as members of this family are often linked to host association and opportunistic pathogenicity‐related traits. Thus, the microbiota of pompano dolphinfish may integrate signals of both industrial contamination and infection‐associated functional potential, providing a microbial fingerprint of human influence on pelagic predators.

Beyond xenobiotic biodegradation and infection‐associated functions, we also identified taxa and predicted functions consistent with fermentative potential in the dolphinfish gut microbiota. From the taxonomic profiles, we detected 26 zOTUs affiliated with genera that include well‐known fermentative (often anaerobic) or facultatively fermentative members, including Brachyspira (dominant zOTU_1), Streptococcus (zOTU_445, zOTU_673), Leptotrichia (zOTU_413, zOTU_655), Actinomyces (zOTU_735), and facultative fermenters within Enterobacteriaceae (e.g., Escherichia–Shigella zOTU_607; Klebsiella zOTU_544) and Vibrionaceae (multiple Vibrio and Photobacterium zOTUs). The predicted fermentation‐marker signal was mainly driven by a small number of dominant contributors, particularly zOTU_1 (Brachyspira) and Spirochaetes‐associated zOTUs (e.g., Brevinemataceae‐related zOTU_2 and zOTU_4), with additional contributions from Photobacterium‐associated zOTUs (e.g., zOTU_12 and zOTU_79). Functionally, these marker KOs map primarily to central fermentative routes within pyruvate metabolism and short‐chain fatty acid formation, consistent with potential production of acetate/lactate and butyrate‐related intermediates.

Taken together, these findings highlight not only the functional specialization of dolphinfish microbiota in response to anthropogenic impacts but also their potential as scalable bioindicators of marine ecosystem health across gradients of human activity.

Implications for Trophic Ecology and Pathogen Transfer

4.4

The dolphinfish is a fast‐growing, highly mobile predator and prey fish linking small preys to larger predators, including marlin, billfishes, seabirds, and dolphins (Mancini and Bugoni 2014; Palko et al. 1982). During our field surveys, we repeatedly observed pantropical spotted dolphins ( Stenella attenuata ) preying on pompano dolphinfish (Figure S1), extending the known predator spectrum of this species to odontocetes. Comparable interactions have been reported for rough‐toothed dolphins ( Steno bredanensis ) feeding on common dolphinfish in Hawaiian waters (Pitman and Stinchcomb 2002). Such observations highlight the ecological position of pompano dolphinfish as a conduit of energy—and potentially microbes—through marine food webs.

Direct evidence of microbial transfer was detected in our unpublished data: one flat needlefish ( Ablennes hians ) that had consumed pompano dolphinfish harbored a gastrointestinal microbiota with 11.03% Brachyspirae (Table S3). This finding underscores the potential for spirochaetes to move upward through trophic pathways. Larger predators, including dolphins, billfishes, and seabirds, may similarly be exposed to Brachyspira through ingestion of dolphinfish.

Additional transmission routes may operate through defecation, which releases gastrointestinal microbes into the pelagic zone. Although most strict anaerobes likely decline rapidly after release, macroorganisms can act as repositories and dispersal agents of marine microbial diversity, particularly by sustaining and redistributing members of the rare biosphere (“sustaining the rares”) (Troussellier et al. 2017). In this framework, fecal material and sloughed mucus may create transient microhabitats and short‐lived exposure hotspots, increasing encounter rates between gut‐associated taxa and other organisms in the water column. Such localized enrichment around fish schools could expose planktivores to gut‐associated microbes and facilitate cross‐trophic transfer (Egerton et al. 2018; Rempel et al. 2022), allowing some lineages to re‐enter the food web and potentially circulate back to higher predators (Troussellier et al. 2017).

These pathways resonate with the “One Health” framework, where animal microbiomes intersect with ecosystem and human health (Zinsstag et al. 2011; Destoumieux‐Garzón et al. 2018; Mauad et al. 2025). This poses risks not only to aquatic organisms but also potential zoonotic threats to livestock or humans exposed to contaminated waters, for example, during recreational activities (Sosah et al. 2025). The possibility of environmental transmission therefore suggests that pompano dolphinfish may harbor microbial lineages of potential concern and may act as a potential carrier or reservoir within marine food webs, with possible implications for both local ecosystem health and long‐distance transfer via migration. However, pathogenicity cannot be inferred solely from 16S rRNA signatures, and further work is needed to establish the virulence potential of these strains.

Conclusion

5

This study provides the first comparative analysis of the gastrointestinal microbiota of pompano dolphinfish from two distinct sites in the South China Sea. We found a highly simplified, Spirochaete‐dominated community strongly shaped by host filtering yet responsive to anthropogenic influence. The consistent dominance of Brachyspira highlights the potential of dolphinfish to act as reservoirs of opportunistic pathogens, with possible implications for food‐web dynamics and marine predator health. Direct trophic evidence from needlefish and eDNA surveys further suggest that pompano dolphinfish may acquire Brachyspira in coastal habitats and subsequently transport them offshore, linking human‐impacted neritic zones with pelagic ecosystems. Functional enrichment of pollutant‐degradation and infectious disease pathways in neritic‐water fish reinforces the value of microbiota as sensitive bioindicators of anthropogenic pressure. Together, these findings underscore the ecological and health relevance of fish microbiomes and call for broader, multi‐omics studies to clarify their roles in marine ecosystem resilience.

Author Contributions

Wentao Lu: data curation (lead), formal analysis (equal), investigation (lead), methodology (equal), resources (lead), visualization (lead), writing – original draft (lead). Xinrui Long: formal analysis (supporting), resources (equal), visualization (supporting), writing – review and editing (equal). Liang Fang: funding acquisition (equal), investigation (equal), resources (equal), writing – review and editing (equal). Hancheng Zhao: investigation (supporting), writing – review and editing (supporting). Yuezhong Wang: investigation (supporting), writing – review and editing (supporting). Xunyu Yang: investigation (supporting), writing – review and editing (supporting). Zhao Zheng: investigation (supporting), writing – review and editing (supporting). Yijie He: writing – review and editing (supporting). Bo Liang: writing – review and editing (supporting). Zonghang Zhang: writing – review and editing (supporting). Tao Chen: funding acquisition (equal), writing – review and editing (equal). Jianqing Lin: conceptualization (lead), funding acquisition (lead), investigation (lead), project administration (lead), supervision (lead), writing – review and editing (lead). Wenhua Liu: conceptualization (lead), funding acquisition (lead), investigation (lead), project administration (lead), writing – review and editing (lead).

Funding

This work was supported by Basic and Applied Basic Research Foundation of Guangdong Province (2024A1515011082), Key Programme of National Natural Science Foundation of China (42230413); the Hainan Provincial Science and Technology Plan in the Science and Technology Innovation Joint Project of the Sanya Yazhou Bay Science and Technology City (2021CXLH0004) and National Natural Science Foundation of China (32202937).

Ethics Statement

All animal procedures were reviewed and approved by the Committee on Ethics and Use of Experimental Animals of Shantou University (ethical approval No. STU20240120001), and were conducted in accordance with relevant institutional guidelines for the care and use of experimental animals.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Observation of Pantropical Spotted Dolphins (Stenella attenuata) Preying on the Pompano dolphinfish (Coryphaena equiselis).

Table S1: Pampano dolphinfish samples information.

Table S2: Relative abundance of Bacteria taxa in the gastrointestinal content of Pampano dolphinfish and seawater.

Table S3: Gastrointestinal microbiol of flat noodlefish by 16S DNA metabarcoding.