Host matters: coral reef fish species show distinct skin microbiome responses to abrupt environmental change

Host matters: coral reef fish species show distinct skin microbiome responses to abrupt environmental change | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Host matters: coral reef fish species show distinct skin microbiome responses to abrupt environmental change Laura Luna Lardinois, Natasha A. Hinojosa, Helio Quintero Arrieta, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8158492/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 21 Apr, 2026 Read the published version in BMC Microbiology → Version 1 posted 13 You are reading this latest preprint version Abstract Background Disentangling the drivers structuring microbiomes can help predict organisms’ responses to rapid environmental change. However, despite microbial communities being important for both host and environmental health, large gaps remain in our understanding of how host-associated microbiomes are structured and respond to different stimuli, especially in marine environments. Here, we leverage seasonal upwelling in Panama’s Tropical Eastern Pacific to test how abrupt environmental changes linked to seasonal upwelling influence the diversity and composition of coral reef fish skin microbiomes in ten species spanning four trophic groups. Results Fish skin microbiomes varied greatly within and among host species and were distinct from the microbiomes of the surrounding seawater. All species had diverse skin microbiomes, with a dominance of Proteobacteria (65%), Bacteroidota (13%), and Cyanobacteria (6%). Host species and trophic group played a greater role in determining fish skin microbiome structure than seasonal and regional environmental variation, despite water microbiomes responding strongly to both season and region. Nevertheless, three out of five trophic groups: the herbivores, carnivores, and planktivore, also displayed significant changes in their microbiomes during upwelling, albeit to a lesser extent than water samples. We performed differential abundance (DA) analyses on these fish and compared microbial taxa that changed between seasons and regions in fish versus water samples. While water communities had thousands of significant DA taxa, fish had around 40 times fewer (n = 17 to 73) and only shared 30 DA taxa with the water samples. Differences between these microbial communities likely arise from both host selection via fishes’ immune system and the skin mucus serving as an environmental filter. However, neither host-associated nor environmental predictors fully explained the variation in microbiome composition, highlighting its complexity. Conclusions Our results show how ecological differences between host species may elicit distinct microbiome responses to environmental changes, with potential cascading effects on ecosystem dynamics under global climate change. Further characterization of marine microbial communities, as well as additional physicochemical and host-related parameters, will be key to monitoring and predicting how these communities will respond to the increasingly rapid and widespread environmental changes our oceans are facing. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Background Microbes are the unseen majority on Earth. Trillions of microbes; including bacteria, archaea, viruses, and fungi, grow on and inside living organisms, forming their unique microbiome [ 1 , 2 ]. These microbial communities establish a dynamic relationship with their animal hosts, responding to factors like the host’s biology, diet, and the surrounding environment [ 3 ]. Some members of the microbiome fulfil essential functions such as providing nutrients, regulating metabolism, and protecting against pathogens, whereas others can be pathogenic or parasitic [ 4 , 5 ]. Furthermore, microbiomes not only influence host traits and the surrounding environment, but also contribute an “extended genetic repertoire” and can rapidly respond to their environment - via changes in community composition, gene expression, and rapid evolution, with important implications for their hosts [ 6 – 8 ]. At the ecosystem scale, microbes play crucial roles in nutrient cycling, animal and plant health, agriculture, and aquaculture [ 9 ]. The cumulative impacts of climate change, including ocean warming and decreasing oxygen levels, are threatening the biosphere, especially the marine ecosystems that make up over 70% of the planet [ 10 , 11 ]. These stressors not only impact animal hosts directly, but also disrupt the balance between hosts and their microbiomes, yet the full extent of their impacts on microbiomes are unknown [ 9 , 12 ]. Emerging research from coral reefs, which are among the most biodiverse and threatened ecosystems, shows cause for concern: for instance, butterflyfish gut microbiomes were altered on degraded reefs, and reef fish gut microbiomes changed when exposed to nutrient pollution in the laboratory [ 13 , 14 ]. Untangling the complex factors that influence the myriads of resident and transient microbial associates is key to understanding how organisms will respond in an era of rapid environmental change. Microbes that colonize the surface of hosts’ skin live at the interface between hosts and the external environment, providing an ideal system to study the interactions between hosts, microbes, and responses to environmental change. Furthermore, skin microbiomes play a crucial role in host health, serving as the first line of defence against pathogens, which is especially important for marine organisms in constant contact with microbes in the surrounding water [ 5 , 15 ]. Despite their importance for host health, much less is known about the skin microbiomes of wild organisms, including coral reef fish, compared to the gut microbiome (but see [ 16 – 22 ]). While concerns have been raised about the potential vulnerability of these microbiomes to environmental change and biodiversity loss, reef fish skin microbiomes’ responses to changing environments have yet to be studied in nature [ 17 ]. We aimed to fill this knowledge gap and expand our understanding of host-microbe interactions by characterizing the skin microbiomes of a community of coral reef fish exposed to natural environmental variation created by seasonal upwelling in Panama’s Tropical Eastern Pacific (Fig. 1 ). Within the Tropical Eastern Pacific (TEP), the Gulf of Panama experiences upwelling during the dry season ( Jan-Apr ), when trade winds displace surface water, causing cold, low-oxygen, nutrient-rich water to rise from the depths [ 23 , 24 ]. To the west, the Cordillera Central mountains block these trade winds, weakening upwelling in the nearby Gulf of Chiriquí (Fig. 2 ) [ 24 , 25 ]. During the wet season ( Apr-Dec ), conditions in the two gulfs return to being similar [ 25 ]. This phenomenon is an ideal natural experiment to explore how environmental changes impact skin microbiomes and whether these microbiomes are most strongly influenced by environmental factors or host-associated traits [ 26 ]. Both regions harbour similar communities of coral reef fishes with many ecologically and economically important species [ 27 ]. In this study, we used 16S rRNA marker gene sequencing to quantify how the skin microbiome responds to changing environmental conditions across regions and seasons in 10 species of reef fish spanning a range of trophic groups. We compared host-associated microbiomes to microbial communities in the surrounding water, and whether these communities respond in similar ways to upwelling conditions. We hypothesized that, if hosts are the principal determinants of microbial communities, the skin microbiomes on fish belonging to the same species would be most similar in composition and relative abundance, with host species and host-associated factors such as trophic group playing a greater role than region or season. Alternatively, if skin microbiomes are primarily influenced by environmental factors, microbiomes from the same region and season would be most similar across species. Furthermore, we would expect greater similarity between skin and water microbiomes, as well as parallel responses to upwelling conditions. Understanding the factors that shape these complex microbial communities in a natural system is essential to better predict the effects of stressors under global climate change. Methods Study area and sample collection We collected fish samples from two island archipelagos in the Tropical Eastern Pacific: Coiba in the Gulf of Chiriquí, which does not experience strong upwelling, and Las Perlas in the Gulf of Panama, which experiences strong upwelling during the dry season (Fig. 1 ). Within each archipelago, we sampled from several sites to ensure that we captured a representative sample of fish host-associated microbiomes ( Table S1 ). We conducted two rounds of sampling in each region across 2021-22: near the end of the wet season (non-upwelling; Oct. 15 – Dec. 3, 2021), and at the end of the dry season (upwelling; Mar. 11 – Apr. 29, 2022). Environmental data, including water temperature and dissolved oxygen (DO), were collected with HOBO and miniDOT data loggers, respectively, in both gulfs (Fig. 2 ; Fig. S1 ). We selected 10 abundant, ecologically important reef-dwelling fish species in five trophic groups: four herbivores (roving and territorial), two invertivores, two carnivores, a planktivore, and an omnivore, representing four families: Acanthuridae (surgeonfishes), Chaetodontidae (butterflyfishes), Pomacentridae (damselfishes), and Serranidae (groupers) (Fig. 1 ; Table S2 ). We replicated fish species within each trophic group wherever possible to differentiate between species-specific responses and those occurring within or across trophic groups. Given that prior work has shown high inter- and intra-specific microbiome variability [ 16 ], we aimed to collect 10 adult fish per species from each region and season. We collected morphological data, including sex, length, mass, and gonad mass, as sex, size, body condition, and reproductive status could all have effects on the microbiome [ 28 – 30 ]. We excluded fish that expressed juvenile characteristics (size and/or coloration), and focused on individuals of the same size classes from each region, to control for microbiome shifts that are known to occur across ontogeny [ 31 , 32 ]. We used freediving and spearfishing (1–15 m depth) to collect the fish following protocols approved by the Smithsonian Tropical Research Institute’s IACUC (SI-22047). Using gloved hands, we rubbed a sterile swab across the upper dorsal side of each fish 3–5 times, circumventing bodily fluids and avoiding contact with the fisher’s hand. After swabbing, fish were euthanized using the rapid chilling method, then kept on ice until we returned from the field. We placed swabs in cryotubes, immediately flash-froze them in liquid nitrogen, then stored them at -80ºC until DNA extraction. Control swabs (exposed to air in field) were collected at each site and season. Additionally, we collected 2-liter water samples above the reef at each site with Whirl-Pak bags. Water samples were kept on ice for transport, run through Millipore™ 0.22 um MCE membranes the same day, and membranes stored at -80ºC until DNA extraction. DNA Extraction & Sequencing We extracted fish skin microbial DNA using the ZymoBIOMICS 96 well DNA kit (four 96-well plates in total), following the manufacturer’s instructions. Briefly, we snipped swabs containing fish skin mucus with sterile scissors and placed the cotton tip in 750 uL Zymo DNA/RNA Shield (instead of lysis solution) in bead-beating tubes, before continuing with the standard protocol. Environmental (water) DNA was extracted from membranes using the Qiagen DNeasy PowerSoil Kit. We amplified the V4 region of the 16S ribosomal RNA gene (16S rRNA) using primers 515F and 806R adapted for Illumina sequencing, following a modified version of the Earth Microbiome Project 16S protocol [ 33 ] (PCR conditions: Table S3 ). Phased primers - one set per plate – were used to increase sample library complexity and augment sequencing quality. Negative PCR controls and extraction controls were included in each plate. Index PCR was performed to attach unique barcodes, then all four plates: 384 samples including controls, were pooled for sequencing on the Illumina MiSeq sequencing platform of the Smithsonian Tropical Research Institute's (STRI) Naos facilities in Panama. Water samples were PCR-amplified independently and run on a separate sequencing run to avoid potential contamination. Sequence Data Processing Sequenced libraries were demultiplexed using the MiSeq Reporter Software. We trimmed these libraries with Cutadapt (v 4.1) to remove primers and adaptors before reading them into R version 4.2.2 for downstream analyses following the DADA2 pipeline (tutorial v 1.16) [ 34 ]. Trimmed sequence reads are available at EMBL-EBI under accession number [PRJEB104461]. We filtered and trimmed (truncation at 220 (Fwd) and 180 (Rev) base pairs), dereplicated, inferred amplicon sequence variants (using the pseudo-pool method to capture rare ASVs) and merged forward and reverse reads. Next, we removed chimeras and assigned taxonomy using the SILVA reference taxonomy (v. 138.1). Contaminants, along with ASVs assigned to chloroplasts, mitochondria, eukaryotes, or unassigned at the phylum level were removed, as were control samples. We then filtered out potentially spurious ASVs found in less than two samples. Additionally, samples with fewer that 1,000 reads were removed, bringing us down to 341 fish samples in the cleaned dataset (Table S4). To test for the influence of uneven sequencing depths on patterns of community composition, we conducted all downstream analyses with both the unrarefied and rarefied datasets. The rarefied dataset was assembled by computing rarefaction curves and normalizing the reads to equal library sizes based on these curves (first using mirl(); repeatedly rarefying to 2500 reads (10 times)), following recent publications on best practices for normalizing microbiome data (42–44). Given that results were consistent between unrarefied and rarefied datasets, we present the unrarefied data in the main text, with the rarefied results available in the supplementary materials. To calculate phylogenetic diversity, we created a maximum likelihood phylogenetic tree (GTR + G + I) for all ASVs, using DECIPHER (v 2.26.0) and phangorn (v 2.11.1) packages for multiple alignment and tree construction, respectively [ 34 ]. Microbiome analyses Assessing alpha diversity (diversity within hosts) We leveraged Hill numbers, a trio of complementary metrics: observed (species richness, which provides higher leverage to rare taxa), Shannon exponential (uses a logarithmic scale, balancing rare and common taxa), and Simpson’s multiplicative inverse (emphasizing common taxa), to measure alpha diversity in our samples [ 34 , 35 ] (Figs. 4 , S2). Together, these metrics allowed us to visualize within-sample diversity in the fish species, across sampling gulfs and seasons. We then ran non-parametric Kruskal-Wallis tests with post hoc Dunn tests to compare alpha diversity among our fish species, and across sampling regions and seasons [ 13 ]. Additional diversity, evenness, and dominance metrics are compiled in supplementary materials ( Table S5) . Assessing beta diversity (diversity between hosts) We used a range of metrics to tease apart the influence of rare versus common ASVs (Jaccard and Bray Curtis), and the phylogenetic relatedness of ASVs (UniFrac and weighted UniFrac) on patterns of community dissimilarity. We ran individual permutational multivariate analyses of variance (PERMANOVA) to test for the effect of host species, host trophic group, sampling region, and sampling season. We then ran additional PERMANOVAs testing for interactions between these key variables, constraining permutations to within a trophic group, given the nestedness of species within trophic groups. Results using Bray-Curtis dissimilarity are presented in the main text. All other metrics are in the supplementary materials ( Table S7 ). Ordinations (PCoA and NMDS) were used to compare microbial community (dis)similarity (Figs. 5 , S3, S4). We visualized microbial relative abundance across host species, regions, and seasons with stacked bar plots (Figs. 3 , S5). All analysis and visualization scripts are available on GitHub ( https://github.com/lardinois21/RRR_Fish_Microbiome_16S.git ) and are archived on Zenodo ( https://doi.org/10.5281/zenodo.17739195 ). Differentially abundant taxa between seasons and regions We assessed which microbial taxa were differentially abundant between seasons and regions using differential abundance analyses. DESeq2 (differential gene expression analysis based on the negative binomial distribution) and MaAsLin2 (microbiome multivariable association with linear models) were among the best performing DA tests, particularly for low sample sizes, in a recent comparison [ 36 – 38 ]. Initial tests on our water dataset showed that the DESeq2 and MaAsLin2 results were comparable ( supp. mat.: MaAsLin2 ), but we ran DESeq2 on the fish dataset given the flexibility provided by the “contrasts” argument for testing pairwise comparisons. Differential abundance tests were only run for fish species with significant differences in microbiome composition based on the PERMANOVA results. Un-normalized counts were used, as DESeq2 corrects for differences in library size. Prior to running the tests, we filtered out taxa present in less than 10% of samples in each dataset, as rare taxa can impact the model assumptions and false discovery rate (FDR) penalty applied [ 39 ]. We specified contrasts to test four pairwise comparisons: (1) upwelling-independent seasonal changes in the Gulf of Chiriquí (Gulf of Chiriquí wet season vs. dry season), (2) upwelling-associated seasonal changes in the Gulf of Panama (Gulf of Panama dry season vs. wet season), (3) “baseline” inter-region differences (Gulf of Panama wet season vs. Gulf of Chiriquí wet season, when environmental conditions are similar between the two gulfs), and (4) inter-region differences during upwelling (Gulf of Panama dry season vs. Gulf of Chiriquí dry season) ( Table S8, Figs. S6-S14 ). Finally, we ran each of the significant DA taxa in the fish datasets against the DA taxa from the water dataset to distinguish between changes in taxa that may be occurring in the surrounding environment and carrying over to the fish skin, versus taxa that are changing independently in the fish, irrespective of the surrounding water microbial community ( Table S9 ). Results Strong seasonal shifts in temperature and dissolved oxygen in the Gulf of Panama As expected for upwelling, during the dry season water temperatures dropped sharply in the Gulf of Panama beginning in late February (min: 17.5ºC) whereas they remained stable throughout the year (~ 29ºC) in the Gulf of Chiriquí (Fig. 2 ). This was accompanied by a drop in dissolved oxygen concentrations in the Gulf of Panama, from an average of 6.55 mg/l (96% DO saturation), down to as low as 0.86 mg/l (11% sat.; Fig. S1 ) . In contrast, DO concentrations varied widely daily in the Gulf of Chiriquí but exhibited no obvious seasonal patterns (range: 0.9-19.92 mg/l (14–249% sat.), avg.: 6.45 mg/l (99% sat.); Fig. S1 ). Microbiome library sizes and microbial taxa pre- and post-rarefaction We sequenced 359 fish skin microbiome samples and 19 water samples. The raw fish microbiome dataset had 4,418,389 total reads corresponding to 23,440 ASVs (range: 1–48,663 sequences/sample). After filtering samples with few reads and removing rare ASVs, 6551 ASVs in 341 samples remained (range: 407 − 39,934). The normalized dataset included 6,493 ASVs across 279 samples (62 samples below the 2,500 read threshold were dropped; Table S4 ). The water sample dataset had 1,507,994 total reads, corresponding to 2,712 ASVs across 19 samples (range: 42,156 − 99,992 sequences/sample). The water dataset was normalized to two different sequencing depths - first to the lowest sequencing depth across all water samples (40,000) to compare water microbiomes between regions and seasons, then to the smaller library size across the whole dataset (fish and water: 2,500) for direct comparisons between the fish and water microbiomes. The fish skin microbiome is highly variable, dominated by key phyla, and distinct from seawater Fish skin microbiomes were highly diverse, varied greatly within and among host species, yet remained distinct from microbial communities in the surrounding water (Figs. 3 , 5 ). At the phylum level, the dominant members of the fish skin microbiome were Proteobacteria (65%), Bacteroidota (13%), Cyanobacteria (6%), Actinobacteriota (4%), and Verrucomicrobiota (3%, Fig. 3 ). Water samples also contained these phyla, however, their relative abundances differed, with less Proteobacteria (54%) and more Cyanobacteria (18%, Fig. 3 K). Dominant phyla were shared between fish skin and water samples, yet certain less abundant phyla were only found in fish: Latescibacteria, MBNT15, Deferrisomatota, Deferribacterota, LCP-89, and Sumerlaeota. Only four phyla were unique to the water samples: AncK6, Nitrospinota, Aenigmarchaeota, and Poribacteria. Some microbial phyla were highly represented in multiple samples of a given fish species, for instance, fusobacteriota in Abudefduf concolor (Fig. 3 A), fibrobacterota in Prionurus laticlavius (Fig. 3 D), and patescibacteria in Chaetodon humeralis (Fig. 3 E). Firmicutes appeared in many fish samples, irrespective of species, with its abundance varying considerably to make up over 25% of some samples while being almost completely absent from others. Note that although prokaryotic nomenclature has recently been revised [ 40 ], we refer to the phylum names in the SILVA database (v. 138.1) used for taxonomic assignment. Microbial richness increases in dry season for herbivores and planktivores in the Gulf of Panama When analysing the full dataset, there were no significant differences in fish skin microbiome alpha diversity across regions and seasons for any of the three metrics (observed, Shannon exponential, and Simpson’s multiplicative inverse; Kruskal-Wallis with BH correction; P > 0.05), although the Gulf of Panama tended to have higher observed richness during the dry season (upwelling) compared to wet season samples from both the Gulf of Panama and the Gulf of Chiriquí ( P = 0.12 and 0.17, respectively). However, pairwise comparisons showed significant differences in alpha diversity between many fish species, with the most pronounced differences in carnivores, which had higher diversity, and invertivores, which had lower diversity (Figs. 4 , S2; Tables S5, S6 ). Splitting the dataset by species, we see a notable increase in alpha diversity during the dry season (upwelling) in the Gulf of Panama across all four herbivores and the planktivore, whereas no clear regional or seasonal pattern is visible in the other species and trophic groups (Fig. 4 ). Similar patterns were seen across all Hill numbers, and differences tended to be more pronounced in observed richness, suggesting these differences are driven in part by rare ASVs. In comparison, for the water samples, Simpson’s multiplicative inverse was significant ( P = 0.019), suggesting that shifts in common taxa in the Gulf of Panama during the dry season cause these microbial communities to become significantly different from those in both the dry and wet season in the Gulf of Chiriquí. Fish skin microbiome beta diversity is primarily driven by host species and trophic group Host species and trophic group had a greater influence on fish skin microbiome structure than environmental factors such as sampling region or season. Although each of these factors are significant, across the full dataset host species explains 7.9% of the variation in microbiome composition, trophic group explains 4.5%, whereas region and season only explain 1.4% and 1.3%, respectively (PERMANOVA – Bray-Curtis, species : R 2 = 0.079, P < 0.001; trophic group : R 2 = 0.045, P < 0.001; region : R 2 = 0.014, P < 0.001; season : R 2 = 0.013, P = 0.001, Table S7 ). We also found significant interactions between many combinations of predictors, including region and season (R 2 = 0.008, P < 0.001, Bray-Curtis); diet and region (R 2 = 0.017, P = 0.002), and diet, species, region, and site (R 2 = 0.108, P = 0.051). Combinations of predictors better explained microbiome composition than a single predictor. Other potential explanatory factors such as fish sex and size either had no significant effect on the skin microbiome (sex), or minimal explanatory power (length and mass). Importantly, even when combining the best predictors we measured, over 66% of the variation in fish skin microbiome remained unexplained. Unconstrained ordinations (PCoA and NMDS; Bray-Curtis dissimilarity) did not reveal consistent clustering in the fish microbiome: samples from different regions and seasons generally overlapped in two-dimensional space, likely due to the low level of variance explained by these factors (Figs. 5 , S3, S4). Given the importance of host species for skin microbiome community structure, we tested the effect of environmental factors on each species separately. We found consistent patterns for fish belonging to the same trophic group. Across trophic groups, hosts generally fell into two categories: ones for which neither region nor season were significant (the invertivores and the omnivore), and ones where both region and season had a significant impact on the skin microbiome (the herbivores and carnivores, the planktivore, and the water) ( Table S7 ). Furthermore, fish in the latter category differed in whether there was a significant interaction between region and season: the territorial herbivores and one carnivore ( E. labriformis ) had no interaction whereas the roving herbivores, the other carnivore ( C. panamensis ), the planktivore, and the water had significant interactions between season and region. When comparing the microbial beta diversity results obtained using Bray-Curtis dissimilarity to those from Jaccard and UniFrac, all three indices generally showed similar trends (i.e., significant differences for the same host species, regions, and seasons) in our microbiome data, but differed in the species-by-species analyses. For instance, we found more significant differences between regions and seasons across host species for Bray-Curtis and Jaccard compared to UniFrac, signalling that there are changes in relative abundance and presence/absence of taxa, but that at the phylogenetic level, the communities do not change as much ( Table S7 ). In other words, while individual microbial taxa inhabiting the skin may change between host species, and vary across both season and region, they are likely being replaced by similar taxa, such that from a phylogenetic perspective, the microbial communities are not that different. Compared to the fish, region and season influenced the water microbial communities to a much greater extent, as reflected by the higher percent variance (PERMANOVA; Bray-Curtis, season : R 2 = 0.313, P < 0.001; region : R 2 = 0.190, P = 0.005, Table S7 ). Furthermore, as would be expected given that only one region experiences strong seasonal upwelling, we found significant interactions between sampling season and region (R 2 = 0.15, P < 0.001). These effects were equally apparent in our ordinations, where samples from the Gulf of Panama and the Gulf of Chiriquí clustered tightly together during the wet season and broke apart into two separate, distinct clusters in the dry (upwelling) season (Fig. 5 K). More differentially abundant taxa between regions and seasons in water than in fish After analysing the differences in microbial community composition across seasons, regions, and hosts, we leveraged differential abundance analyses to better understand which microbial taxa might be driving these differences. Among our water samples, 524 taxa (of 1,298 filtered taxa) differed significantly in abundance between the Gulf of Panama wet and dry season samples (DESeq2; Table S8, Fig. S7 ). Of these, 318 (24%) were enriched (log-fold change > 0) and 206 (16%) decreased in the dry season. There were fewer DA taxa across gulfs during the wet season (116), suggesting that seasonal differences outweigh regional ones in the water microbiomes ( Table S8, Fig. S6 ). Compared to the water, fish had fewer significant DA taxa ( Table S8 ). Given the fish skin microbiome’s diversity, the same 10% filter prior to running DESeq2 resulted in only 16–24% of each fish host’s taxa being retained for the analysis (mean: 658 taxa). While certain fish, such as M. dorsalis and P. laticlavius (two herbivores) followed expected patterns, with the greatest number of DA taxa between the wet and dry seasons in the Gulf of Panama ( Table S8 ; Figs. S9, S11 ), the majority did not follow this pattern. Instead, DA taxa occurred randomly across the four pairwise region-season comparisons, with many DA taxa in certain host species while others had few (range: 17–73; Table S8 ). Limited overlap in DA taxa between fish and water microbiomes Very few ASVs identified as DA in fish were also differentially abundant in the water ( Tables S8, S9 ). Only 30 unique microbial taxa were significantly differentially abundant in both water and at least one host fishes’ skin ( Tables S8, S9 ). Twelve of these appeared multiple times (DA across various host species and/or in several region-season comparisons; Table S9 ). Most of the shared taxa were Proteobacteria. The three fish that shared the most DA taxa with water were all herbivores ( Table S8 ). The direction of change tended to match between the fish and water samples, such that if taxa were enriched in water in the Gulf of Panama during the dry season, for instance, they would also be enriched in the fish ( Table S9 ). Furthermore, most shared DA taxa were found within one of two comparisons: Gulf of Panama wet vs. dry (n = 22), and Gulf of Panama dry vs. Gulf of Chiriquí dry (n = 27). These two comparisons capture upwelling-driven seasonal differences in the Gulf of Panama ( Table S8 ). Discussion Host and Trophic Group Effects This study sought to quantify how the skin microbiomes of reef fish species inhabiting two gulfs of the TEP are structured and respond to contrasting seasonal change across two regions. Host species identity, in conjunction with trophic group, was the strongest predictor of both alpha and beta diversity in reef fishes’ skin microbiomes. This aligns with other work that has also found unique host-associated skin microbiomes relative to the surrounding water, suggesting that fishes’ skin selects for a distinct microbial community, rather than taking up microbes indiscriminately from the surrounding water [ 16 , 17 , 41 , 42 ]. While trophic group and/or diet are known to strongly influence the gut microbiome, our work is one of the first to show that these factors also play an important role for fish skin microbiome composition [ 43 , 44 ]. Our unique study design helps parse out phylogenetic effects from trophic group, as we have representatives from the same family (or genus) with different diets (Pomacentridae: two herbivores and an omnivore, Serranidae: two carnivores and a planktivore; Table S2 ) and fish with similar diets from different families (herbivores: Acanthuridae & Pomacentridae). Importantly, host species remains significant when including diet as a predictor, which precludes the observed host species effect from simply being an artefact of a given species’ dietary preferences [ 42 ]. Additional support for this host effect comes from paired microbiome and host genotyping work in four sympatric Serrasalmidae (piranha) species, which showed that skin microbiomes and host genotypes covaried significantly both within and across species [ 45 ]. Conserved relationships between hosts and their skin microbiomes could, in turn, lead to phylosymbiosis, such that microbial communities mirror their hosts’ phylogenies [ 46 ]. However, given the high degree of intra-specific variability in these fishes’ microbiomes, we caution against recent attempts to detect phylosymbiosis in teleosts using datasets with low species-level replication [ 47 , 48 ]. Understanding how hosts may shape their skin microbiomes requires a closer look at the skin’s surface. From a microbial perspective, the skin mucosa provides a unique niche space of gelatinous mucin [ 49 ]. However, for microbes to establish communities in the fish skin requires that they evade – or coexist with – enzymes, immune proteins, antimicrobial peptides, and other components of the hosts’ innate immune system, in addition to other competing microbial taxa [ 50 ]. Hosts share similar physical traits with their conspecifics, including the skin, scales, and mucosal layer, forming a similar environment that selects for certain microbes [ 51 ]. Additionally, certain immune defences in teleost skin are highly-conserved, whereas others appear to be clade or species-specific, or develop throughout an individual host’s lifetime, thus imposing distinct constraints on potential colonizers [ 52 , 53 ]. Shared host traits and innate immunity would help account for the significant explanatory power of host species in determining the fish skin microbiome. Environmental Effects Upwelling systems are distributed globally and have wide-ranging impacts on the productivity of marine ecosystems [ 54 ]. How microbial communities in the water column respond to upwelling has received considerable attention in other upwelling areas, including the California Current, the Benguela Current, and Tongoy Bay [ 55 – 57 ]. However, our study is the first to assess these phenomena in the Tropical Eastern Pacific of Panama, in which we observe large shifts in water microbial communities during upwelling, particularly in the Gulf of Panama, where upwelling is strongest (Figs. 5 , S3, S4, S6). Water microbiomes were more similar between gulfs during the wet season (non-upwelling) than they were within a gulf across seasons, indicating that seasonal turnover of microbial taxa outweighs fixed differences between gulfs (Fig. 5 , Table S8 ). Interestingly, while the greatest number of differentially abundant taxa in the water communities were, as expected, seen in the Gulf of Panama during the dry season, when there is strong upwelling, we also saw a clear signal of seasonal change in the Gulf of Chiriquí during the dry season, despite no strong changes in temperature nor dissolved oxygen during this time. This may be due to other seasonal changes in coastal hydrology, such as surface runoff, altering water microbiomes during the wet season, or be linked to a localized upwelling hotspot in the Gulf of Chiriquí [ 25 , 58 ]. Furthermore, this is the first study to look at the effects of upwelling-driven environmental changes on both water and vertebrate host-associated microbiomes. Compared to the surrounding seawater, significant seasonal changes were only detected in a subset of our host species’ microbiomes and were of lesser magnitude. Host-associated microbiomes had around 40 times fewer significant differentially abundant (DA) taxa across region-season comparisons than water ( Table S8 ). Among these, hosts shared 0–34% of their DA taxa with water samples, meaning that although a minority of the microbes that changed seasonally on hosts may be picked up from – and thus directly match changes in - the surrounding environment, most host-associated microbes respond to environmental changes independently. Seasonal changes in temperature and dissolved oxygen (Figs. 2 , S1) associated with upwelling likely played a role in the differences seen in the fish skin microbiomes, as both variables (alongside salinity and pH) correlated with shifts in microbiome diversity in a recent review of the few studies to date that have reported DO and temperature data alongside microbiome results for saltwater fishes’ skin [ 59 ]. Hosts whose skin microbiomes were altered by seasonal upwelling – namely, the herbivores, carnivores, and planktivore – may be more susceptible to future environmental changes. However, without further functional information, we cannot determine whether the community-level changes we observe are harmful or helpful to the host. Microbes demonstrate high levels of phenotypic plasticity and can thrive under diverse conditions, allowing them to respond to changing environmental conditions [ 60 , 61 ]. Thus, rapid shifts in microbial community composition and metabolic activity could help the host, if the microbiome is altered in such a way that it continues to provide necessary services [ 62 ]. On the other hand, disruptions in the existing microbiome could favour the establishment of pathogenic taxa with detrimental effects. The differences in responses to seasonal environmental changes between seawater and host-associated microbial communities highlight the need for further study of the two in tandem, to better understand how host-microbe relationships may be impacted by environmental changes in different host species, populations, and habitats at risk. Microbiome Complexity and Residual Variation Individual fishes’ traits, life histories, and adaptive immune responses, in combination with microbe-microbe interactions and stochastic changes in the community due to drift, may help explain the 66% variation that was neither captured by host species nor season or region [ 60 ]. When considering this residual variation, it is worth noting that, while we focus here on host species as a predictor, we also assessed the effects of factors such as sex and size, which have previously been shown to impact fish gut microbiomes [ 28 – 30 ]. However, we found no significant effect of sex and only limited effects of size, suggesting that these traits are not major determinants of microbiome structure in these fishes. Genetic diversity within host species may be another factor contributing to host-associated microbiome variation, as reef fishes often display both high genetic diversity and low genetic differentiation, even across distant reef sites [ 63 , 64 ]. Thus, future work within pairing host genomics and microbiome analyses could serve the dual purpose of furthering our understanding of these reef fishes’ population genetic structure and how it may affect their associated microbial communities. Furthermore, once microbes successfully reach the fish skin, potential colonizers must compete to establish themselves in this limited niche space. Priority effects, including niche pre-emption (the first taxa to arrive take over), niche modifications (microbial taxa altering their environment to favour their own growth), and microbial antagonism (competition for limited resources and antimicrobial defences), play an important role in shaping host-associated microbial communities [ 65 ]. While such microbial assembly mechanisms are difficult to measure, especially in wild populations, repeated sampling of individuals prior to, during, and following upwelling may aid to further disentangle individual, species, regional, and seasonal effects in microbial community assembly. Conclusions Our findings suggest that host-associated microbiome responses to future environmental changes will likely be host-specific, with certain species or taxonomic groups responding more strongly. Being associated with a host imposes constraints on the microbial communities that can form, via host immunity and environmental filtering, which acts as a buffer against the fluctuations seen in water microbial communities. These results raise additional questions: (1) Why do certain hosts’ microbiomes respond more strongly to environmental changes and is this an indicator of vulnerability under future climate change? (2) Which microbiome functional shifts are occurring, and what are their repercussions on host and ecosystem health? (3) Are there additional deterministic factors that may address the unexplained variation in these host-associated microbial communities? Further sampling of marine organisms’ microbiomes with greater replication within and across species, families, and trophic groups would provide crucial insights into this first question. Techniques such as metabolomics and metatranscriptomics can be leveraged to determine functional shifts in microbial communities. Finally, monitoring additional physicochemical and host-related parameters, including sub-species level population structure via genomics research, may help further disentangle microbial community dynamics. Given the global scale of the environmental changes our oceans are facing, it is imperative that we build on our nascent understanding of the role microbes play in host and ecosystem health. Declarations Ethics approval and consent to participate All fish were handled and euthanized following protocols approved by the Smithsonian Tropical Research Institute’s Institutional Animal Care and Use Committee (IACUC; SI-22047). We used rapid chilling (hypothermic shock) to euthanize fish. This was done by rapidly transferring fish after capture into individual labelled bags - allowing direct contact with the chilled water but not the ice - and placing them in an ice slurry in a large cooler on the boat. This method was deemed appropriate as we are working with relatively small, tropical fish that are acclimated to temperatures around 29–31ºC. We also wanted to avoid the use of chemical agents, as these could potentially impact our downstream microbiome analyses. Fish were kept in the cooler for a minimum of 20 minutes, then assessed to confirm lack of movement, and kept on ice until we returned to the lab and began dissections. Consent for publication Not applicable. Competing interests We declare we have no competing interests. Funding This project was made possible through funding from the Mark and Rachel Rohr Foundation, the Smithsonian Tropical Research Institute (STRI) – including a Short-Term Fellowship, a Graduate Mobility Award from the McGill Biology Department, a Delise Alison Graduate Student Development Award from the Redpath Museum, a FRQNT doctoral scholarship [DOI: 10.69777/330691 ], a NSERC Discovery Grant [grant number 2019–04549], a Canada Research Chair, and a Quebec Centre for Biodiversity Science (QCBS) Excellence Award. Author Contribution L.L.L.: conceptualization, data curation, formal analysis, investigation, methodology, visualization, writing – original draft, writing – review and editingM.L.: conceptualization, funding acquisition, methodology, project administration, supervision, writing – review & editingR.D.H.B.: conceptualization, funding acquisition, methodology, supervision, writing – review & editingN.A.H., H.Q.A., A.J.S.: investigation, writing – review & editing Acknowledgement We would like to thank the Rohr Reef Resilience team, including project manager Anabell J. Cornejo and intern Javiera Mora for their assistance with sample collection. Special thanks to intern Haesung Jee for help with sample processing, Marta Vargas (STRI molecular lab manager) for sequencing and lab support, and Sean Connolly for providing additional lab space. Thank you to Steve Paton and the STRI Physical Monitoring Program for providing TEP environmental data. Additional thanks to the Coiba National Park rangers and Naos boat drivers for facilitating access to sites, sharing knowledge of local fish communities, and aiding with sample collection. Data Availability Additional methods, metadata, figures, and tables are available as supplementary materials. Raw sequence reads have been deposited to the European Nucleotide Archive (ENA) at EMBL-EBI under accession number [PRJEB104461]. Code for analyses and figures are available as maintained versions on GitHub (https://github.com/lardinois21/RRR_Fish_Microbiome_16S.git) and have been archived on Zenodo https://doi.org/10.5281/zenodo.17739195). References Wilkins LGE, Leray M, O'Dea A, Yuen B, Peixoto RS, Pereira TJ, Bik HM, Coil DA, Duffy JE, Herre EA et al. 2019 Host-associated microbiomes drive structure and function of marine ecosystems. Plos Biology . 17, (10.1371/journal.pbio.3000533). Lennon JT, Locey KJ. 2020 More support for Earth's massive microbiome. Biology Direct . 15, (10.1186/s13062-020-00261-8). Huttenhower C, Gevers D, Knight R, Abubucker S, Badger JH, Chinwalla AT, Creasy HH, Earl AM, Fitzgerald MG, Fulton RS et al. 2012 Structure, function and diversity of the healthy human microbiome. Nature . 486, 207–214. (10.1038/nature11234). McFall-Ngai M, Hadfield MG, Bosch TCG, Carey HV, Domazet-Loso T, Douglas AE, Dubilier N, Eberl G, Fukami T, Gilbert SF et al. 2013 Animals in a bacterial world, a new imperative for the life sciences. Proceedings of the National Academy of Sciences of the United States of America . 110, 3229–3236. (10.1073/pnas.1218525110). Sehnal L, Brammer-Robbins E, Wormington AM, Blaha L, Bisesi J, Larkin I, Martyniuk CJ, Simonin M, Adamovsky O. 2021 Microbiome Composition and Function in Aquatic Vertebrates: Small Organisms Making Big Impacts on Aquatic Animal Health. Frontiers in Microbiology . 12, (10.3389/fmicb.2021.567408). Henry LP, Bruijning M, Forsberg SKG, Ayroles JF. 2021 The microbiome extends host evolutionary potential. Nature Communications . 12, (10.1038/s41467-021-25315-x). Lau JA, Lennon JT. 2012 Rapid responses of soil microorganisms improve plant fitness in novel environments. Proceedings of the National Academy of Sciences of the United States of America . 109, 14058–14062. (10.1073/pnas.1202319109). Soen Y. 2014 Environmental disruption of host-microbe co-adaptation as a potential driving force in evolution. Frontiers in Genetics . 5, (10.3389/fgene.2014.00168). Cavicchioli R, Ripple WJ, Timmis KN, Azam F, Bakken LR, Baylis M, Behrenfeld MJ, Boetius A, Boyd PW, Classen AT et al. 2019 Scientists' warning to humanity: microorganisms and climate change. Nature Reviews Microbiology . 17, 569–586. (10.1038/s41579-019-0222-5). Webster NS, Negri AP, Botte ES, Laffy PW, Flores F, Noonan S, Schmidt C, Uthicke S. 2016 Host-associated coral reef microbes respond to the cumulative pressures of ocean warming and ocean acidification. Scientific Reports . 6, (10.1038/srep19324). Halpern BS, Walbridge S, Selkoe KA, Kappel CV, Micheli F, D'Agrosa C, Bruno JF, Casey KS, Ebert C, Fox HE et al. 2008 A global map of human impact on marine ecosystems. Science . 319, 948–952. (10.1126/science.1149345). Apprill A. 2017 Marine Animal Microbiomes: Toward Understanding Host-Microbiome Interactions in a Changing Ocean. Frontiers in Marine Science . 4, (10.3389/fmars.2017.00222). Clever F, Sourisse JM, Preziosi RF, Eisen JA, Guerra ECR, Scott JJ, Wilkins LGE, Altieri AH, McMillan WO, Leray M. 2022 The gut microbiome variability of a butterflyfish increases on severely degraded Caribbean reefs. Communications Biology . 5, (10.1038/s42003-022-03679-0). Degregori S, Casey JM, Barber PH. 2021 Nutrient pollution alters the gut microbiome of a territorial reef fish. Marine Pollution Bulletin . 169, (10.1016/j.marpolbul.2021.112522). Gomez JA, Primm TP. 2021 A Slimy Business: the Future of Fish Skin Microbiome Studies. Microbial Ecology . 82, 275–287. (10.1007/s00248-020-01648-w). Chiarello M, Auguet J-C, Bettarel Y, Bouvier C, Claverie T, Graham NAJ, Rieuvilleneuve F, Sucre E, Bouvier T, Villeger S. 2018 Skin microbiome of coral reef fish is highly variable and driven by host phylogeny and diet. Microbiome . 6, (10.1186/s40168-018-0530-4). Chiarello M, Auguet J-C, Graham NAJ, Claverie T, Sucre E, Bouvier C, Rieuvilleneuve F, Restrepo-Ortiz CX, Bettarel Y, Villeger S et al. , . 2020 Exceptional but vulnerable microbial diversity in coral reef animal surface microbiomes. Proceedings of the Royal Society B-Biological Sciences. 287, (10.1098/rspb.2020.0642) Pereira A, Soares MC, Santos T, Pocas A, Perez-Losada M, Apprill A, Sikkel PC, Xavier R. 2022 Reef Location and Client Diversity Influence the Skin Microbiome of the Caribbean Cleaner Goby Elacatinus evelynae. Microbial ecology . (10.1007/s00248-022-01984-z). Xavier R, Mazzei R, Perez-Losada M, Rosado D, Santos JL, Verissimo A, Soares MC. 2019 A Risky Business? Habitat and Social Behavior Impact Skin and Gut Microbiomes in Caribbean Cleaning Gobies. Frontiers in Microbiology . 10, (10.3389/fmicb.2019.00716). Xavier R, Pereira A, Pagan A, Hendrick GC, Nicholson MD, Rosado D, Soares MC, Perez-Losada M, Sikkel PC. 2020 The effects of environment and ontogeny on the skin microbiome of two Stegastes damselfishes (Pomacentridae) from the eastern Caribbean Sea. Marine Biology . 167, (10.1007/s00227-020-03717-7). Galand PE, Ruscheweyh HJ, Salazar G, Hochart C, Henry N, Hume BCC, Oliveira PH, Perdereau A, Labadie K, Belser C et al. 2023 Diversity of the Pacific Ocean coral reef microbiome. Nature Communications . 14, (10.1038/s41467-023-38500-x). Berggren H, Yıldırım Y, Nordahl O, Larsson P, Dopson M, Tibblin P, Lundin D, Pinhassi J, Forsman A. 2024 Ecological filtering drives rapid spatiotemporal dynamics in fish skin microbiomes. Mol Ecol . 33, e17496. (10.1111/mec.17496). Fiedler PC, Lavín MF. Oceanographic Conditions of the Eastern Tropical Pacific. Coral Reefs of the Eastern Tropical Pacific. Dordrecht: Springer; 2017. pp. 59–83. D., E. I. Randall CJ, Toth LT, Leichter JJ, Maté JL, Aronson RB. 2020 Upwelling buffers climate change impacts on coral reefs of the eastern tropical Pacific. Ecology . 101, (10.1002/ecy.2918). Crawford G, Mepstead M, D’íaz-Ferguson E. Characterizing oceanographic conditions near Coiba Island and Pacific Panama using 20 years of satellite-based wind stress, SST and chlorophyll-a measurements. Mar Fish Sci. 2024;37. https://doi.org/10.47193/mafis.37X2024010112) . Leray M, Wilkins LGE, Apprill A, Bik HM, Clever F, Connolly SR, De Leon ME, Duffy JE, Ezzat L, Gignoux-Wolfsohn S et al. 2021 Natural experiments and long-term monitoring are critical to understand and predict marine host-microbe ecology and evolution. Plos Biology . 19, (10.1371/journal.pbio.3001322). Robertson DR, Allen GR. Shorefishes of the Tropical Eastern Pacific: online information system. Balboa, Panamá: Smithsonian Tropical Research Institute; 2015. Li H, Lu L, Chen R, Li S, Xu D. 2022 Exploring Sexual Dimorphism in the Intestinal Microbiota of the Yellow Drum (Nibea albiflora, Sciaenidae). Frontiers in Microbiology . 12, (10.3389/fmicb.2021.808285). Messina M, Iacumin L, Pascon G, Tulli F, Tibaldi E, Cardinaletti G. 2023 Effect of feed restriction and refeeding on body condition, digestive functionality and intestinal microbiota in rainbow trout (Oncorhynchus mykiss). Fish Physiology and Biochemistry . 49, 169–189. (10.1007/s10695-023-01170-z). Bates KA, Higgins C, Neiman M, King KC. 2023 Turning the tide on sex and the microbiota in aquatic animals. Hydrobiologia . 850, 3823–3835. (10.1007/s10750-022-04862-4). Bledsoe JW, Peterson BC, Swanson KS, Small BC. 2016 Ontogenetic characterization of the intestinal microbiota of channel catfish through 16S rRNA gene sequencing reveals insights on temporal shifts and the influence of environmental microbes. PLoS ONE . 11, (10.1371/journal.pone.0166379). Rosado D, Pérez-Losada M, Pereira A, Severino R, Xavier R. 2021 Effects of aging on the skin and gill microbiota of farmed seabass and seabream. Animal Microbiome . 3, (10.1186/s42523-020-00072-2). Caporaso GJ, Ackermann G, Apprill A, Bauer M, Berg-Lyons D, Betley J, Fierer N, Fraser L, Fuhrman JA, Gilbert JA et al. , . EMP 16S Illumina Amplicon Protocol V.2. 2023 [cited 2024; Available from : Callahan BJ, Sankaran K, Fukuyama JA, McMurdie PJ, Holmes SP. 2016 Bioconductor Workflow for Microbiome Data Analysis: from raw reads to community analyses. F1000Research . 5, 1492. Roswell M, Dushoff J, Winfree R. 2021 A conceptual guide to measuring species diversity. Oikos . 130, 321–338. (10.1111/oik.07202). Cappellato M, Baruzzo G, Di Camillo B. 2022 Investigating differential abundance methods in microbiome data: A benchmark study. PLoS Comput Biol . 18, e1010467. (10.1371/journal.pcbi.1010467). Mallick H, Rahnavard A, McIver L, Ma S, Zhang Y, Nguyen L, et al. Multivariable association discovery in population-scale meta-omics studies. PLoS Comput Biol. 2021;17. https://doi.org/10.1371/journal.pcbi.1009442) . Love M, Huber W, Anders S. 2014 Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15. Ollberding N. Identifying Differentially Abundant Features in Microbiome Data. 2021. Oren A, Garrity GM. 2021 Valid publication of the names of forty- two phyla of prokaryotes. International Journal of Systematic and Evolutionary Microbiology . 71, (10.1099/ijsem.0.005056). Berggren H, Tibblin P, Yıldırım Y, Broman E, Larsson P, Lundin D, Forsman A. 2022 Fish Skin Microbiomes Are Highly Variable Among Individuals and Populations but Not Within Individuals. Frontiers in Microbiology . 12, (10.3389/fmicb.2021.767770). Sadeghi J, Chaganti SR, Johnson TB, Heath DD. 2023 Host species and habitat shape fish-associated bacterial communities: phylosymbiosis between fish and their microbiome. Microbiome . 11, 258. (10.1186/s40168-023-01697-6). Degregori S, Schiettekatte NMD, Casey JM, Brandl SJ, Mercière A, Amato KR, Mazel F, Parravicini V, Barber PH. 2024 Host diet drives gut microbiome convergence between coral reef fishes and mammals. Molecular Ecology . 33, (10.1111/mec.17520). Youngblut ND, Reischer GH, Walters W, Schuster N, Walzer C, Stalder G, Ley RE, Farnleitner AH. 2019 Host diet and evolutionary history explain different aspects of gut microbiome diversity among vertebrate clades. Nature Communications . 10, (10.1038/s41467-019-10191-3). Sylvain FE, Normandeau E, Holland A, Val AL, Derome N. 2022 Genomics of Serrasalmidae teleosts through the lens of microbiome fingerprinting. Molecular Ecology . 31, 4656–4671. (10.1111/mec.16574). Lim SJ, Bordenstein SR. 2020 An introduction to phylosymbiosis. Proceedings of the Royal Society B-Biological Sciences . 287, (10.1098/rspb.2019.2900). Liu Y, Li X, Kormas KA, Li Y, Li H, Li J. 2025 Variable phylosymbiosis and cophylogeny patterns in wild fish gut microbiota of a large subtropical river. mSphere . e0098224-e0098224. (10.1128/msphere.00982 – 24). Han GH, Yu J, Kang MJ, Park MJ, Noh CH, Kim YJ, Kwon KK. 2024 Phylosymbiosis in Seven Wild Fish Species Collected Off the Southern Coast of Korea: Skin Microbiome Most Strongly Reflects Evolutionary Pressures (vol 87,153,2024). Microbial Ecology . 87, (10.1007/s00248-024-02478-w). Dash S, Das SK, Samal J, Thatoi HN. 2018 Epidermal mucus, a major determinant in fish health: a review. Iran J Veterinary Res. 19, 72–81. Wang LC, Chen LH, Chiu YC, Liou CY, Chen HC, Lu CY, Chen JL. 2023 Teleost skin microbiome: An intimate interplay between the environment and the host immunity. Fish & Shellfish Immunology . 139, (10.1016/j.fsi.2023.108869). Guardiola FA, Cuartero M, Collado-González MD, Arizcún M, Baños FGD, Meseguer J, Cuesta A, Esteban MA. 2015 Description and comparative study of physico-chemical parameters of the teleost fish skin mucus. Biorheology . 52, 247–256. (10.3233/bir-15052). Guardiola FA, Cuesta A, Abellán E, Meseguer J, Esteban MA. 2014 Comparative analysis of the humoral immunity of skin mucus from several marine teleost fish. Fish & Shellfish Immunology . 40, 24–31. (10.1016/j.fsi.2014.06.018). Buchmann K, Karami AM, Duan YJ. 2024 The early ontogenetic development of immune cells and organs in teleosts. Fish & Shellfish Immunology . 146, (10.1016/j.fsi.2024.109371). Wang DW, Gouhier TC, Menge BA, Ganguly AR. 2015 Intensification and spatial homogenization of coastal upwelling under climate change. Nature . 518, 390–394. (10.1038/nature14235). Rocke E, Cheung S, Gebe Z, Dames NR, Liu HB, Moloney CL. 2020 Marine Microbial Community Composition During the Upwelling Season in the Southern Benguela. Frontiers in Marine Science . 7, (10.3389/fmars.2020.00255). Reji L, Tolar BB, Chavez FP, Francis CA. 2020 Depth-Differentiation and Seasonality of Planktonic Microbial Assemblages in the Monterey Bay Upwelling System. Frontiers in Microbiology . 11, (10.3389/fmicb.2020.01075). Molina V, Belmar L, Levipan HA, Ramírez-Flandes S, Anguita C, Galán A, Montes I, Ulloa O. 2020 Spatiotemporal Distribution of Key Pelagic Microbes in a Seasonal Oxygen-Deficient Coastal Upwelling System of the Eastern South Pacific Ocean. Frontiers in Marine Science . 7, (10.3389/fmars.2020.561597). D'Croz L, O'Dea A. 2007 Variability in upwelling along the Pacific shelf of Panama and implications for the distribution of nutrients and chlorophyll. Estuarine Coastal and Shelf Science . 73, 325–340. (10.1016/j.ecss.2007.01.013). Bell AG, McMurtrie J, Bolaños LM, Cable J, Temperton B, Tyler CR. 2024 Influence of host phylogeny and water physicochemistry on microbial assemblages of the fish skin microbiome. Fems Microbiology Ecology . 100, (10.1093/femsec/fiae021). Nemergut DR, Schmidt SK, Fukami T, O'Neill SP, Bilinski TM, Stanish LF, Knelman JE, Darcy JL, Lynch RC, Wickey P et al. 2013 Patterns and Processes of Microbial Community Assembly. Microbiology and Molecular Biology Reviews . 77, 342–356. (10.1128/mmbr.00051 – 12). Webster TUM, Rodriguez-Barreto D, Castaldo G, Gough P, Consuegra S, de Leaniz CG. 2020 Environmental plasticity and colonisation history in the Atlantic salmon microbiome: A translocation experiment. Molecular Ecology . 29, 886–898. (10.1111/mec.15369). Petersen C, Hamerich IK, Adair KL, Griem-Krey H, Oliva MT, Hoeppner MP, Bohannan BJM, Schulenburg H. 2023 Host and microbiome jointly contribute to environmental adaptation. Isme Journal . 17, 1953–1965. (10.1038/s41396-023-01507-9). Pineros VJ, Gutierrez-Rodriguez C. 2017 Population genetic structure and connectivity in the widespread coral-reef fish Abudefduf saxatilis: the role of historic and contemporary factors. Coral Reefs . 36, 877–890. (10.1007/s00338-017-1579-4). Klanten OS, Choat JH, van Herwerden L. 2007 Extreme genetic diversity and temporal rather than spatial partitioning in a widely distributed coral reef fish. Marine Biology . 150, 659–670. (10.1007/s00227-006-0372-7). Chavez-Arroyo A, Radlinski LC, Baumler AJ. 2025 Principles of gut microbiota assembly. Trends in microbiology . (10.1016/j.tim.2025.02.014). Additional Declarations No competing interests reported. Supplementary Files FishSkinMicrobiomePaperSupplementaryMaterialsBMCMicrobiology.docx Cite Share Download PDF Status: Published Journal Publication published 21 Apr, 2026 Read the published version in BMC Microbiology → Version 1 posted Editorial decision: Revision requested 28 Jan, 2026 Reviews received at journal 20 Jan, 2026 Reviews received at journal 06 Jan, 2026 Reviewers agreed at journal 06 Jan, 2026 Reviewers agreed at journal 05 Jan, 2026 Reviewers agreed at journal 05 Jan, 2026 Reviews received at journal 29 Dec, 2025 Reviewers agreed at journal 23 Dec, 2025 Reviewers agreed at journal 08 Dec, 2025 Reviewers invited by journal 05 Dec, 2025 Editor assigned by journal 02 Dec, 2025 Submission checks completed at journal 01 Dec, 2025 First submitted to journal 30 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8158492","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":553326325,"identity":"57822be8-2e02-49bd-a7fd-3450ed3e7618","order_by":0,"name":"Laura Luna Lardinois","email":"data:image/png;base64,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","orcid":"","institution":"McGill University","correspondingAuthor":true,"prefix":"","firstName":"Laura","middleName":"Luna","lastName":"Lardinois","suffix":""},{"id":553326326,"identity":"c4833889-16e2-46cc-8f17-f3f6bd7c93a2","order_by":1,"name":"Natasha A. Hinojosa","email":"","orcid":"","institution":"University of California, Santa Cruz","correspondingAuthor":false,"prefix":"","firstName":"Natasha","middleName":"A.","lastName":"Hinojosa","suffix":""},{"id":553326327,"identity":"f080ed69-5a9c-4c66-8bb3-29e54964192b","order_by":2,"name":"Helio Quintero Arrieta","email":"","orcid":"","institution":"Smithsonian Tropical Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Helio","middleName":"Quintero","lastName":"Arrieta","suffix":""},{"id":553326328,"identity":"09600ea5-4ab2-4c18-a5cc-0079adb76813","order_by":3,"name":"Andrew J. Sellers","email":"","orcid":"","institution":"Smithsonian Tropical Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Andrew","middleName":"J.","lastName":"Sellers","suffix":""},{"id":553326329,"identity":"2bbdb495-340b-4695-9aad-01990150e598","order_by":4,"name":"Matthieu Leray","email":"","orcid":"","institution":"Smithsonian Tropical Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Matthieu","middleName":"","lastName":"Leray","suffix":""},{"id":553326330,"identity":"06a0b435-bd9f-4c7b-b726-3d4d4199a513","order_by":5,"name":"Rowan D. H. Barrett","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Rowan","middleName":"D. H.","lastName":"Barrett","suffix":""}],"badges":[],"createdAt":"2025-11-19 20:23:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8158492/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8158492/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12866-026-05036-1","type":"published","date":"2026-04-21T15:59:26+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":97250340,"identity":"57ee109c-55ea-486c-b5d4-8cc2f4dc1d99","added_by":"auto","created_at":"2025-12-02 13:14:18","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3407061,"visible":true,"origin":"","legend":"","description":"","filename":"FishSkinMicrobiomePaperBMCMicrobiologyRevised20251127.docx","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/cde823d97e6f8a0e15db5195.docx"},{"id":97231418,"identity":"65757c60-35b1-4fd5-9644-6c1fb2fb42c6","added_by":"auto","created_at":"2025-12-02 09:25:41","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9547,"visible":true,"origin":"","legend":"","description":"","filename":"3ab6dc3354bf479f87938be5e5718176.json","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/153db7a8fb0c4495c66f6841.json"},{"id":97231423,"identity":"26996380-a04a-4251-b500-84ab0eb2be4b","added_by":"auto","created_at":"2025-12-02 09:25:41","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5656982,"visible":true,"origin":"","legend":"","description":"","filename":"FishSkinMicrobiomePaperSupplementaryMaterialsBMCMicrobiology.docx","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/817f7593420a37c11cb92be3.docx"},{"id":97231437,"identity":"cd3a3745-b16e-47f2-8e40-dd0ea1eff35c","added_by":"auto","created_at":"2025-12-02 09:25:42","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":149911,"visible":true,"origin":"","legend":"","description":"","filename":"3ab6dc3354bf479f87938be5e57181761enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/52246a135f0da74d4228e176.xml"},{"id":97250589,"identity":"94bf11c1-7eb5-42cd-9fa1-932f90f94f26","added_by":"auto","created_at":"2025-12-02 13:14:47","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":462116,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/b3c6ba2daeb03ce6337e1716.png"},{"id":97251276,"identity":"39732b8c-9282-48d8-b43f-a085c6aa3027","added_by":"auto","created_at":"2025-12-02 13:16:39","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":159338,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/e3b21927be45de739b1c9724.jpeg"},{"id":97231425,"identity":"452cf0b3-effd-4ffe-80e5-1c13f57104c8","added_by":"auto","created_at":"2025-12-02 09:25:41","extension":"emf","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2885824,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.emf","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/a1642236bb1154ecd0b89a9f.emf"},{"id":97251109,"identity":"ed358f48-590a-459f-963e-f0d4bb3ba4d8","added_by":"auto","created_at":"2025-12-02 13:16:06","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1759,"visible":true,"origin":"","legend":"","description":"","filename":"groupimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/0b6a6853dc4fc02a2bd55417.jpeg"},{"id":97250327,"identity":"34839e6a-e282-4796-b68f-9859e3047e3f","added_by":"auto","created_at":"2025-12-02 13:14:16","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":76581,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/707b2b1f8eb4323d374250b1.png"},{"id":97250955,"identity":"6c7a42d9-d502-4ed6-aabb-1d58bf7d5017","added_by":"auto","created_at":"2025-12-02 13:15:40","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35668,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/36cd58813ca0672450751b77.png"},{"id":97250291,"identity":"58a8df05-4c6d-48fd-95f7-4f9ed59facad","added_by":"auto","created_at":"2025-12-02 13:14:13","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":235464,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/2042c53abfd86517871d319c.png"},{"id":97231420,"identity":"a5f704fc-dd86-4ec4-8df1-a635944db364","added_by":"auto","created_at":"2025-12-02 09:25:41","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":83250,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/dc044cb3a614153c88f82d36.png"},{"id":97231432,"identity":"1de102c6-b097-4ac8-9cad-207c28b90ff6","added_by":"auto","created_at":"2025-12-02 09:25:42","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79651,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/6d00b6d9b5315b66e16b1999.png"},{"id":97250231,"identity":"f198535a-797d-488a-b9bb-ad55ed91ff9b","added_by":"auto","created_at":"2025-12-02 13:14:09","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1183,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinegroupimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/ed518f38a9b9e2bacf8d4053.png"},{"id":97231435,"identity":"7f7a6b9f-7912-4419-9d27-2c51a8d53d5c","added_by":"auto","created_at":"2025-12-02 09:25:42","extension":"xml","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":147959,"visible":true,"origin":"","legend":"","description":"","filename":"3ab6dc3354bf479f87938be5e57181761structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/07f419bb3466cb7c0266c7e1.xml"},{"id":97231429,"identity":"50381d4f-a36c-4836-9fec-a57a37b5aac7","added_by":"auto","created_at":"2025-12-02 09:25:42","extension":"html","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":163504,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/639fbc4ddc885adc28f33366.html"},{"id":97231415,"identity":"db92dcb0-bf53-48a9-acbb-fb8253ebe5e3","added_by":"auto","created_at":"2025-12-02 09:25:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":268847,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eStudy design and sampling sites in Panama’s Tropical Eastern Pacific. \u003c/strong\u003e\u003c/em\u003eSamples collected from 10 fish per species (\u003cstrong\u003eA\u003c/strong\u003e) from two regions: (\u003cstrong\u003eB\u003c/strong\u003e) Gulf of Chiriquí, \u003cem\u003eorange\u003c/em\u003e; (\u003cstrong\u003eC\u003c/strong\u003e) Gulf of Panama; \u003cem\u003eturquoise\u003c/em\u003e, during the wet (Oct-Nov) and dry (Mar-Apr) seasons. Map source file: M. Solano (2022).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/3c8349cbce8acd914539e319.png"},{"id":97231427,"identity":"1fa14caa-38ff-4b99-a23f-41197027b26c","added_by":"auto","created_at":"2025-12-02 09:25:42","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":770860,"visible":true,"origin":"","legend":"\u003cp\u003eWater temperature in the TEP\u003cem\u003e\u003cstrong\u003e. \u003c/strong\u003e\u003c/em\u003eTemperatures (ºC) in the Gulf of Chiriquí (\u003cem\u003eorange\u003c/em\u003e) and Gulf of Panama (\u003cem\u003eturquoise\u003c/em\u003e). Shaded bars: sampling in wet (\u003cem\u003epurple\u003c/em\u003e) and dry (\u003cem\u003egreen\u003c/em\u003e) seasons.\u003cem\u003e \u003c/em\u003eDry season sampling began after water temperatures dropped in the Gulf of Panama, indicative of upwelling. Simple moving average (48hr avg. temp) overlaid over raw temperature values.\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/7cdfc53890cfa843052a41d7.jpeg"},{"id":97231433,"identity":"73179f0f-e7a0-435d-9e81-49670dc3c0fe","added_by":"auto","created_at":"2025-12-02 09:25:42","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":567384,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eMicrobial relative abundances.\u003c/strong\u003e\u003c/em\u003e Bar plots of phylum-level microbiome composition, split by gulf (\u003cem\u003eleft:\u003c/em\u003e Coiba - Gulf of Chiriquí, \u003cem\u003eright:\u003c/em\u003e Las Perlas - Gulf of Panama). \u003cstrong\u003e(A - J) \u003c/strong\u003eEach bar represents a single fishes’ microbiome. Panels arranged by trophic group (coloured boxes), fish icons denote species (see \u003cem\u003eFig. 1\u003c/em\u003e), and coloured lines along the bottom of each panel mark seasons (\u003cem\u003epurple\u003c/em\u003e = wet, \u003cem\u003egreen\u003c/em\u003e = dry). \u003cstrong\u003e(H)\u003c/strong\u003e Water microbiome community.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/439307683dd753dad1e8ac16.png"},{"id":97231416,"identity":"bc4a77cd-3f2e-4360-8ce5-fb9a413ac72e","added_by":"auto","created_at":"2025-12-02 09:25:41","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":502806,"visible":true,"origin":"","legend":"\u003cp\u003eAlpha diversity - observed richness by gulf. Boxplots of observed richness (q0) for each host species, split by gulf: (A)Gulf of Chiriquí (orange box) and (B) Gulf of Panama (turquoise box). In each panel, samples are split by season; left (dark shade) = wet season, right (light shade) = dry season. Bar colours and fish icons indicate trophic groups and study species, respectively (see \u003cem\u003eFig. 1\u003c/em\u003e). Right-most panels: water samples.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/ba76610abafe1eabcfbae322.jpeg"},{"id":97250404,"identity":"165686c4-a61d-4510-9f5b-003873e45acd","added_by":"auto","created_at":"2025-12-02 13:14:26","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":37960,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eDissimilarity between microbial communities. \u003c/strong\u003e\u003c/em\u003ePCoA plots based on Bray-Curtis dissimilarities between communities. \u003cstrong\u003e(A – J) \u003c/strong\u003eFish skin microbiomes across regions (\u003cem\u003ecircle\u003c/em\u003e: Gulf of Chiriquí; \u003cem\u003etriangle\u003c/em\u003e: Gulf of Panama) and seasons (\u003cem\u003epurple\u003c/em\u003e: wet, \u003cem\u003egreen\u003c/em\u003e: dry), split by host species. Axis colours and icons indicate trophic groups and study species, respectively (see \u003cem\u003eFig. 1\u003c/em\u003e). \u003cstrong\u003e(K)\u003c/strong\u003eWater microbiome.\u003c/p\u003e","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/619a6b04adf9a265781cf125.png"},{"id":107928702,"identity":"df74528e-2bde-438c-b28e-b4816e049c1e","added_by":"auto","created_at":"2026-04-27 16:12:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2495147,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/26710e04-90fd-4491-9f66-5c439277fb9c.pdf"},{"id":97251283,"identity":"e9e03905-7e0a-4b66-a5fd-db7ab99166af","added_by":"auto","created_at":"2025-12-02 13:16:40","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":5656982,"visible":true,"origin":"","legend":"","description":"","filename":"FishSkinMicrobiomePaperSupplementaryMaterialsBMCMicrobiology.docx","url":"https://assets-eu.researchsquare.com/files/rs-8158492/v1/701c4b8e5d26f288e72688cc.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Host matters: coral reef fish species show distinct skin microbiome responses to abrupt environmental change","fulltext":[{"header":"Background","content":"\u003cp\u003eMicrobes are the unseen majority on Earth. Trillions of microbes; including bacteria, archaea, viruses, and fungi, grow on and inside living organisms, forming their unique microbiome [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. These microbial communities establish a dynamic relationship with their animal hosts, responding to factors like the host\u0026rsquo;s biology, diet, and the surrounding environment [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Some members of the microbiome fulfil essential functions such as providing nutrients, regulating metabolism, and protecting against pathogens, whereas others can be pathogenic or parasitic [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Furthermore, microbiomes not only influence host traits and the surrounding environment, but also contribute an \u0026ldquo;extended genetic repertoire\u0026rdquo; and can rapidly respond to their environment - via changes in community composition, gene expression, and rapid evolution, with important implications for their hosts [\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAt the ecosystem scale, microbes play crucial roles in nutrient cycling, animal and plant health, agriculture, and aquaculture [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The cumulative impacts of climate change, including ocean warming and decreasing oxygen levels, are threatening the biosphere, especially the marine ecosystems that make up over 70% of the planet [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. These stressors not only impact animal hosts directly, but also disrupt the balance between hosts and their microbiomes, yet the full extent of their impacts on microbiomes are unknown [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Emerging research from coral reefs, which are among the most biodiverse and threatened ecosystems, shows cause for concern: for instance, butterflyfish gut microbiomes were altered on degraded reefs, and reef fish gut microbiomes changed when exposed to nutrient pollution in the laboratory [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Untangling the complex factors that influence the myriads of resident and transient microbial associates is key to understanding how organisms will respond in an era of rapid environmental change.\u003c/p\u003e\u003cp\u003eMicrobes that colonize the surface of hosts\u0026rsquo; skin live at the interface between hosts and the external environment, providing an ideal system to study the interactions between hosts, microbes, and responses to environmental change. Furthermore, skin microbiomes play a crucial role in host health, serving as the first line of defence against pathogens, which is especially important for marine organisms in constant contact with microbes in the surrounding water [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Despite their importance for host health, much less is known about the skin microbiomes of wild organisms, including coral reef fish, compared to the gut microbiome (but see [\u003cspan additionalcitationids=\"CR17 CR18 CR19 CR20 CR21\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]). While concerns have been raised about the potential vulnerability of these microbiomes to environmental change and biodiversity loss, reef fish skin microbiomes\u0026rsquo; responses to changing environments have yet to be studied in nature [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. We aimed to fill this knowledge gap and expand our understanding of host-microbe interactions by characterizing the skin microbiomes of a community of coral reef fish exposed to natural environmental variation created by seasonal upwelling in Panama\u0026rsquo;s Tropical Eastern Pacific (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eWithin the Tropical Eastern Pacific (TEP), the Gulf of Panama experiences upwelling during the dry season (\u003cem\u003eJan-Apr\u003c/em\u003e), when trade winds displace surface water, causing cold, low-oxygen, nutrient-rich water to rise from the depths [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. To the west, the Cordillera Central mountains block these trade winds, weakening upwelling in the nearby Gulf of Chiriqu\u0026iacute; (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. During the wet season (\u003cem\u003eApr-Dec\u003c/em\u003e), conditions in the two gulfs return to being similar [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. This phenomenon is an ideal natural experiment to explore how environmental changes impact skin microbiomes and whether these microbiomes are most strongly influenced by environmental factors or host-associated traits [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Both regions harbour similar communities of coral reef fishes with many ecologically and economically important species [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIn this study, we used 16S rRNA marker gene sequencing to quantify how the skin microbiome responds to changing environmental conditions across regions and seasons in 10 species of reef fish spanning a range of trophic groups. We compared host-associated microbiomes to microbial communities in the surrounding water, and whether these communities respond in similar ways to upwelling conditions. We hypothesized that, if hosts are the principal determinants of microbial communities, the skin microbiomes on fish belonging to the same species would be most similar in composition and relative abundance, with host species and host-associated factors such as trophic group playing a greater role than region or season. Alternatively, if skin microbiomes are primarily influenced by environmental factors, microbiomes from the same region and season would be most similar across species. Furthermore, we would expect greater similarity between skin and water microbiomes, as well as parallel responses to upwelling conditions. Understanding the factors that shape these complex microbial communities in a natural system is essential to better predict the effects of stressors under global climate change.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eStudy area and sample collection\u003c/h2\u003e\u003cp\u003eWe collected fish samples from two island archipelagos in the Tropical Eastern Pacific: Coiba in the Gulf of Chiriqu\u0026iacute;, which does not experience strong upwelling, and Las Perlas in the Gulf of Panama, which experiences strong upwelling during the dry season (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Within each archipelago, we sampled from several sites to ensure that we captured a representative sample of fish host-associated microbiomes (\u003cem\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/em\u003e). We conducted two rounds of sampling in each region across 2021-22: near the end of the wet season (non-upwelling; Oct. 15 \u0026ndash; Dec. 3, 2021), and at the end of the dry season (upwelling; Mar. 11 \u0026ndash; Apr. 29, 2022). Environmental data, including water temperature and dissolved oxygen (DO), were collected with HOBO and miniDOT data loggers, respectively, in both gulfs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; \u003cem\u003eFig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/em\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe selected 10 abundant, ecologically important reef-dwelling fish species in five trophic groups: four herbivores (roving and territorial), two invertivores, two carnivores, a planktivore, and an omnivore, representing four families: Acanthuridae (surgeonfishes), Chaetodontidae (butterflyfishes), Pomacentridae (damselfishes), and Serranidae (groupers) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e; \u003cem\u003eTable S2\u003c/em\u003e). We replicated fish species within each trophic group wherever possible to differentiate between species-specific responses and those occurring within or across trophic groups. Given that prior work has shown high inter- and intra-specific microbiome variability [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], we aimed to collect 10 adult fish per species from each region and season. We collected morphological data, including sex, length, mass, and gonad mass, as sex, size, body condition, and reproductive status could all have effects on the microbiome [\u003cspan additionalcitationids=\"CR29\" citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. We excluded fish that expressed juvenile characteristics (size and/or coloration), and focused on individuals of the same size classes from each region, to control for microbiome shifts that are known to occur across ontogeny [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eWe used freediving and spearfishing (1\u0026ndash;15 m depth) to collect the fish following protocols approved by the Smithsonian Tropical Research Institute\u0026rsquo;s IACUC (SI-22047). Using gloved hands, we rubbed a sterile swab across the upper dorsal side of each fish 3\u0026ndash;5 times, circumventing bodily fluids and avoiding contact with the fisher\u0026rsquo;s hand. After swabbing, fish were euthanized using the rapid chilling method, then kept on ice until we returned from the field. We placed swabs in cryotubes, immediately flash-froze them in liquid nitrogen, then stored them at -80\u0026ordm;C until DNA extraction. Control swabs (exposed to air in field) were collected at each site and season. Additionally, we collected 2-liter water samples above the reef at each site with Whirl-Pak bags. Water samples were kept on ice for transport, run through Millipore\u0026trade; 0.22 um MCE membranes the same day, and membranes stored at -80\u0026ordm;C until DNA extraction.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eDNA Extraction \u0026 Sequencing\u003c/h3\u003e\n\u003cp\u003eWe extracted fish skin microbial DNA using the ZymoBIOMICS 96 well DNA kit (four 96-well plates in total), following the manufacturer\u0026rsquo;s instructions. Briefly, we snipped swabs containing fish skin mucus with sterile scissors and placed the cotton tip in 750 uL Zymo DNA/RNA Shield (instead of lysis solution) in bead-beating tubes, before continuing with the standard protocol. Environmental (water) DNA was extracted from membranes using the Qiagen DNeasy PowerSoil Kit. We amplified the V4 region of the 16S ribosomal RNA gene (16S rRNA) using primers 515F and 806R adapted for Illumina sequencing, following a modified version of the Earth Microbiome Project 16S protocol [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] (PCR conditions: \u003cem\u003eTable S3\u003c/em\u003e). Phased primers - one set per plate \u0026ndash; were used to increase sample library complexity and augment sequencing quality. Negative PCR controls and extraction controls were included in each plate. Index PCR was performed to attach unique barcodes, then all four plates: 384 samples including controls, were pooled for sequencing on the Illumina MiSeq sequencing platform of the Smithsonian Tropical Research Institute's (STRI) Naos facilities in Panama. Water samples were PCR-amplified independently and run on a separate sequencing run to avoid potential contamination.\u003c/p\u003e\n\u003ch3\u003eSequence Data Processing\u003c/h3\u003e\n\u003cp\u003eSequenced libraries were demultiplexed using the MiSeq Reporter Software. We trimmed these libraries with Cutadapt (v 4.1) to remove primers and adaptors before reading them into R version 4.2.2 for downstream analyses following the DADA2 pipeline (tutorial v 1.16) [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Trimmed sequence reads are available at EMBL-EBI under accession number [PRJEB104461]. We filtered and trimmed (truncation at 220 (Fwd) and 180 (Rev) base pairs), dereplicated, inferred amplicon sequence variants (using the pseudo-pool method to capture rare ASVs) and merged forward and reverse reads. Next, we removed chimeras and assigned taxonomy using the SILVA reference taxonomy (v. 138.1). Contaminants, along with ASVs assigned to chloroplasts, mitochondria, eukaryotes, or unassigned at the phylum level were removed, as were control samples. We then filtered out potentially spurious ASVs found in less than two samples. Additionally, samples with fewer that 1,000 reads were removed, bringing us down to 341 fish samples in the cleaned dataset (Table S4). To test for the influence of uneven sequencing depths on patterns of community composition, we conducted all downstream analyses with both the unrarefied and rarefied datasets. The rarefied dataset was assembled by computing rarefaction curves and normalizing the reads to equal library sizes based on these curves (first using mirl(); repeatedly rarefying to 2500 reads (10 times)), following recent publications on best practices for normalizing microbiome data (42\u0026ndash;44). Given that results were consistent between unrarefied and rarefied datasets, we present the unrarefied data in the main text, with the rarefied results available in the supplementary materials. To calculate phylogenetic diversity, we created a maximum likelihood phylogenetic tree (GTR\u0026thinsp;+\u0026thinsp;G\u0026thinsp;+\u0026thinsp;I) for all ASVs, using DECIPHER (v 2.26.0) and phangorn (v 2.11.1) packages for multiple alignment and tree construction, respectively [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e\n\u003ch3\u003eMicrobiome analyses\u003c/h3\u003e\n\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003eAssessing alpha diversity (diversity within hosts)\u003c/h2\u003e\u003cp\u003eWe leveraged Hill numbers, a trio of complementary metrics: observed (species richness, which provides higher leverage to rare taxa), Shannon exponential (uses a logarithmic scale, balancing rare and common taxa), and Simpson\u0026rsquo;s multiplicative inverse (emphasizing common taxa), to measure alpha diversity in our samples [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, S2). Together, these metrics allowed us to visualize within-sample diversity in the fish species, across sampling gulfs and seasons. We then ran non-parametric Kruskal-Wallis tests with post hoc Dunn tests to compare alpha diversity among our fish species, and across sampling regions and seasons [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Additional diversity, evenness, and dominance metrics are compiled in supplementary materials (\u003cem\u003eTable S5)\u003c/em\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eAssessing beta diversity (diversity between hosts)\u003c/h2\u003e\u003cp\u003eWe used a range of metrics to tease apart the influence of rare versus common ASVs (Jaccard and Bray Curtis), and the phylogenetic relatedness of ASVs (UniFrac and weighted UniFrac) on patterns of community dissimilarity. We ran individual permutational multivariate analyses of variance (PERMANOVA) to test for the effect of host species, host trophic group, sampling region, and sampling season. We then ran additional PERMANOVAs testing for interactions between these key variables, constraining permutations to within a trophic group, given the nestedness of species within trophic groups. Results using Bray-Curtis dissimilarity are presented in the main text. All other metrics are in the supplementary materials (\u003cem\u003eTable S7\u003c/em\u003e). Ordinations (PCoA and NMDS) were used to compare microbial community (dis)similarity (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, S3, S4). We visualized microbial relative abundance across host species, regions, and seasons with stacked bar plots (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, S5). All analysis and visualization scripts are available on GitHub (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/lardinois21/RRR_Fish_Microbiome_16S.git\u003c/span\u003e\u003cspan address=\"https://github.com/lardinois21/RRR_Fish_Microbiome_16S.git\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and are archived on Zenodo (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/zenodo.17739195\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.17739195\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eDifferentially abundant taxa between seasons and regions\u003c/h3\u003e\n\u003cp\u003eWe assessed which microbial taxa were differentially abundant between seasons and regions using differential abundance analyses. DESeq2 (differential gene expression analysis based on the negative binomial distribution) and MaAsLin2 (microbiome multivariable association with linear models) were among the best performing DA tests, particularly for low sample sizes, in a recent comparison [\u003cspan additionalcitationids=\"CR37\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Initial tests on our water dataset showed that the DESeq2 and MaAsLin2 results were comparable (\u003cem\u003esupp. mat.: MaAsLin2\u003c/em\u003e), but we ran DESeq2 on the fish dataset given the flexibility provided by the \u0026ldquo;contrasts\u0026rdquo; argument for testing pairwise comparisons. Differential abundance tests were only run for fish species with significant differences in microbiome composition based on the PERMANOVA results. Un-normalized counts were used, as DESeq2 corrects for differences in library size. Prior to running the tests, we filtered out taxa present in less than 10% of samples in each dataset, as rare taxa can impact the model assumptions and false discovery rate (FDR) penalty applied [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. We specified contrasts to test four pairwise comparisons: (1) upwelling-independent seasonal changes in the Gulf of Chiriqu\u0026iacute; (Gulf of Chiriqu\u0026iacute; wet season vs. dry season), (2) upwelling-associated seasonal changes in the Gulf of Panama (Gulf of Panama dry season vs. wet season), (3) \u0026ldquo;baseline\u0026rdquo; inter-region differences (Gulf of Panama wet season vs. Gulf of Chiriqu\u0026iacute; wet season, when environmental conditions are similar between the two gulfs), and (4) inter-region differences during upwelling (Gulf of Panama dry season vs. Gulf of Chiriqu\u0026iacute; dry season) (\u003cem\u003eTable S8, Figs. S6-S14\u003c/em\u003e). Finally, we ran each of the significant DA taxa in the fish datasets against the DA taxa from the water dataset to distinguish between changes in taxa that may be occurring in the surrounding environment and carrying over to the fish skin, versus taxa that are changing independently in the fish, irrespective of the surrounding water microbial community (\u003cem\u003eTable S9\u003c/em\u003e).\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eStrong seasonal shifts in temperature and dissolved oxygen in the Gulf of Panama\u003c/h2\u003e\u003cp\u003eAs expected for upwelling, during the dry season water temperatures dropped sharply in the Gulf of Panama beginning in late February (min: 17.5\u0026ordm;C) whereas they remained stable throughout the year (~\u0026thinsp;29\u0026ordm;C) in the Gulf of Chiriqu\u0026iacute; (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). This was accompanied by a drop in dissolved oxygen concentrations in the Gulf of Panama, from an average of 6.55 mg/l (96% DO saturation), down to as low as 0.86 mg/l (11% sat.; \u003cem\u003eFig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e)\u003c/em\u003e. In contrast, DO concentrations varied widely daily in the Gulf of Chiriqu\u0026iacute; but exhibited no obvious seasonal patterns (range: 0.9-19.92 mg/l (14\u0026ndash;249% sat.), avg.: 6.45 mg/l (99% sat.); \u003cem\u003eFig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/em\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eMicrobiome library sizes and microbial taxa pre- and post-rarefaction\u003c/h2\u003e\u003cp\u003eWe sequenced 359 fish skin microbiome samples and 19 water samples. The raw fish microbiome dataset had 4,418,389 total reads corresponding to 23,440 ASVs (range: 1\u0026ndash;48,663 sequences/sample). After filtering samples with few reads and removing rare ASVs, 6551 ASVs in 341 samples remained (range: 407\u0026thinsp;\u0026minus;\u0026thinsp;39,934). The normalized dataset included 6,493 ASVs across 279 samples (62 samples below the 2,500 read threshold were dropped; \u003cem\u003eTable S4\u003c/em\u003e). The water sample dataset had 1,507,994 total reads, corresponding to 2,712 ASVs across 19 samples (range: 42,156\u0026thinsp;\u0026minus;\u0026thinsp;99,992 sequences/sample). The water dataset was normalized to two different sequencing depths - first to the lowest sequencing depth across all water samples (40,000) to compare water microbiomes between regions and seasons, then to the smaller library size across the whole dataset (fish and water: 2,500) for direct comparisons between the fish and water microbiomes.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eThe fish skin microbiome is highly variable, dominated by key phyla, and distinct from seawater\u003c/h2\u003e\u003cp\u003eFish skin microbiomes were highly diverse, varied greatly within and among host species, yet remained distinct from microbial communities in the surrounding water (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). At the phylum level, the dominant members of the fish skin microbiome were Proteobacteria (65%), Bacteroidota (13%), Cyanobacteria (6%), Actinobacteriota (4%), and Verrucomicrobiota (3%, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Water samples also contained these phyla, however, their relative abundances differed, with less Proteobacteria (54%) and more Cyanobacteria (18%, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eK). Dominant phyla were shared between fish skin and water samples, yet certain less abundant phyla were only found in fish: Latescibacteria, MBNT15, Deferrisomatota, Deferribacterota, LCP-89, and Sumerlaeota. Only four phyla were unique to the water samples: AncK6, Nitrospinota, Aenigmarchaeota, and Poribacteria. Some microbial phyla were highly represented in multiple samples of a given fish species, for instance, fusobacteriota in \u003cem\u003eAbudefduf concolor\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA), fibrobacterota in \u003cem\u003ePrionurus laticlavius\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD), and patescibacteria in \u003cem\u003eChaetodon humeralis\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). Firmicutes appeared in many fish samples, irrespective of species, with its abundance varying considerably to make up over 25% of some samples while being almost completely absent from others. Note that although prokaryotic nomenclature has recently been revised [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e], we refer to the phylum names in the SILVA database (v. 138.1) used for taxonomic assignment.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eMicrobial richness increases in dry season for herbivores and planktivores in the Gulf of Panama\u003c/h2\u003e\u003cp\u003eWhen analysing the full dataset, there were no significant differences in fish skin microbiome alpha diversity across regions and seasons for any of the three metrics (observed, Shannon exponential, and Simpson\u0026rsquo;s multiplicative inverse; Kruskal-Wallis with BH correction; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05), although the Gulf of Panama tended to have higher observed richness during the dry season (upwelling) compared to wet season samples from both the Gulf of Panama and the Gulf of Chiriqu\u0026iacute; (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.12 and 0.17, respectively). However, pairwise comparisons showed significant differences in alpha diversity between many fish species, with the most pronounced differences in carnivores, which had higher diversity, and invertivores, which had lower diversity (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, S2; \u003cem\u003eTables S5, S6\u003c/em\u003e). Splitting the dataset by species, we see a notable increase in alpha diversity during the dry season (upwelling) in the Gulf of Panama across all four herbivores and the planktivore, whereas no clear regional or seasonal pattern is visible in the other species and trophic groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Similar patterns were seen across all Hill numbers, and differences tended to be more pronounced in observed richness, suggesting these differences are driven in part by rare ASVs. In comparison, for the water samples, Simpson\u0026rsquo;s multiplicative inverse was significant (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.019), suggesting that shifts in common taxa in the Gulf of Panama during the dry season cause these microbial communities to become significantly different from those in both the dry and wet season in the Gulf of Chiriqu\u0026iacute;.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eFish skin microbiome beta diversity is primarily driven by host species and trophic group\u003c/h2\u003e\u003cp\u003eHost species and trophic group had a greater influence on fish skin microbiome structure than environmental factors such as sampling region or season. Although each of these factors are significant, across the full dataset host species explains 7.9% of the variation in microbiome composition, trophic group explains 4.5%, whereas region and season only explain 1.4% and 1.3%, respectively (PERMANOVA \u0026ndash; Bray-Curtis, \u003cb\u003especies\u003c/b\u003e: R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.079, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; \u003cb\u003etrophic group\u003c/b\u003e: R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.045, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; \u003cb\u003eregion\u003c/b\u003e: R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.014, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; \u003cb\u003eseason\u003c/b\u003e: R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.013, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001, \u003cem\u003eTable S7\u003c/em\u003e). We also found significant interactions between many combinations of predictors, including region and season (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.008, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, Bray-Curtis); diet and region (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.017, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002), and diet, species, region, and site (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.108, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.051). Combinations of predictors better explained microbiome composition than a single predictor. Other potential explanatory factors such as fish sex and size either had no significant effect on the skin microbiome (sex), or minimal explanatory power (length and mass). Importantly, even when combining the best predictors we measured, over 66% of the variation in fish skin microbiome remained unexplained.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eUnconstrained ordinations (PCoA and NMDS; Bray-Curtis dissimilarity) did not reveal consistent clustering in the fish microbiome: samples from different regions and seasons generally overlapped in two-dimensional space, likely due to the low level of variance explained by these factors (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, S3, S4). Given the importance of host species for skin microbiome community structure, we tested the effect of environmental factors on each species separately. We found consistent patterns for fish belonging to the same trophic group. Across trophic groups, hosts generally fell into two categories: ones for which neither region nor season were significant (the invertivores and the omnivore), and ones where both region and season had a significant impact on the skin microbiome (the herbivores and carnivores, the planktivore, and the water) (\u003cem\u003eTable S7\u003c/em\u003e). Furthermore, fish in the latter category differed in whether there was a significant interaction between region and season: the territorial herbivores and one carnivore (\u003cem\u003eE. labriformis\u003c/em\u003e) had no interaction whereas the roving herbivores, the other carnivore (\u003cem\u003eC. panamensis\u003c/em\u003e), the planktivore, and the water had significant interactions between season and region.\u003c/p\u003e\u003cp\u003eWhen comparing the microbial beta diversity results obtained using Bray-Curtis dissimilarity to those from Jaccard and UniFrac, all three indices generally showed similar trends (i.e., significant differences for the same host species, regions, and seasons) in our microbiome data, but differed in the species-by-species analyses. For instance, we found more significant differences between regions and seasons across host species for Bray-Curtis and Jaccard compared to UniFrac, signalling that there are changes in relative abundance and presence/absence of taxa, but that at the phylogenetic level, the communities do not change as much (\u003cem\u003eTable S7\u003c/em\u003e). In other words, while individual microbial taxa inhabiting the skin may change between host species, and vary across both season and region, they are likely being replaced by similar taxa, such that from a phylogenetic perspective, the microbial communities are not that different.\u003c/p\u003e\u003cp\u003eCompared to the fish, region and season influenced the water microbial communities to a much greater extent, as reflected by the higher percent variance (PERMANOVA; Bray-Curtis, \u003cb\u003eseason\u003c/b\u003e: R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.313, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; \u003cb\u003eregion\u003c/b\u003e: R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.190, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.005, \u003cem\u003eTable S7\u003c/em\u003e). Furthermore, as would be expected given that only one region experiences strong seasonal upwelling, we found significant interactions between sampling season and region (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.15, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). These effects were equally apparent in our ordinations, where samples from the Gulf of Panama and the Gulf of Chiriqu\u0026iacute; clustered tightly together during the wet season and broke apart into two separate, distinct clusters in the dry (upwelling) season (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eK).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eMore differentially abundant taxa between regions and seasons in water than in fish\u003c/h2\u003e\u003cp\u003eAfter analysing the differences in microbial community composition across seasons, regions, and hosts, we leveraged differential abundance analyses to better understand which microbial taxa might be driving these differences. Among our water samples, 524 taxa (of 1,298 filtered taxa) differed significantly in abundance between the Gulf of Panama wet and dry season samples (DESeq2; \u003cem\u003eTable S8, Fig. S7\u003c/em\u003e). Of these, 318 (24%) were enriched (log-fold change\u0026thinsp;\u0026gt;\u0026thinsp;0) and 206 (16%) decreased in the dry season. There were fewer DA taxa across gulfs during the wet season (116), suggesting that seasonal differences outweigh regional ones in the water microbiomes (\u003cem\u003eTable S8, Fig. S6\u003c/em\u003e). Compared to the water, fish had fewer significant DA taxa (\u003cem\u003eTable S8\u003c/em\u003e). Given the fish skin microbiome\u0026rsquo;s diversity, the same 10% filter prior to running DESeq2 resulted in only 16\u0026ndash;24% of each fish host\u0026rsquo;s taxa being retained for the analysis (mean: 658 taxa). While certain fish, such as \u003cem\u003eM. dorsalis\u003c/em\u003e and \u003cem\u003eP. laticlavius\u003c/em\u003e (two herbivores) followed expected patterns, with the greatest number of DA taxa between the wet and dry seasons in the Gulf of Panama (\u003cem\u003eTable S8\u003c/em\u003e; \u003cem\u003eFigs. S9, S11\u003c/em\u003e), the majority did not follow this pattern. Instead, DA taxa occurred randomly across the four pairwise region-season comparisons, with many DA taxa in certain host species while others had few (range: 17\u0026ndash;73; \u003cem\u003eTable S8\u003c/em\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003eLimited overlap in DA taxa between fish and water microbiomes\u003c/h2\u003e\u003cp\u003eVery few ASVs identified as DA in fish were also differentially abundant in the water (\u003cem\u003eTables S8, S9\u003c/em\u003e). Only 30 unique microbial taxa were significantly differentially abundant in both water and at least one host fishes\u0026rsquo; skin (\u003cem\u003eTables S8, S9\u003c/em\u003e). Twelve of these appeared multiple times (DA across various host species and/or in several region-season comparisons; \u003cem\u003eTable S9\u003c/em\u003e). Most of the shared taxa were Proteobacteria. The three fish that shared the most DA taxa with water were all herbivores (\u003cem\u003eTable S8\u003c/em\u003e). The direction of change tended to match between the fish and water samples, such that if taxa were enriched in water in the Gulf of Panama during the dry season, for instance, they would also be enriched in the fish (\u003cem\u003eTable S9\u003c/em\u003e). Furthermore, most shared DA taxa were found within one of two comparisons: Gulf of Panama wet vs. dry (n\u0026thinsp;=\u0026thinsp;22), and Gulf of Panama dry vs. Gulf of Chiriqu\u0026iacute; dry (n\u0026thinsp;=\u0026thinsp;27). These two comparisons capture upwelling-driven seasonal differences in the Gulf of Panama (\u003cem\u003eTable S8\u003c/em\u003e).\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003eHost and Trophic Group Effects\u003c/h2\u003e\u003cp\u003eThis study sought to quantify how the skin microbiomes of reef fish species inhabiting two gulfs of the TEP are structured and respond to contrasting seasonal change across two regions. Host species identity, in conjunction with trophic group, was the strongest predictor of both alpha and beta diversity in reef fishes\u0026rsquo; skin microbiomes. This aligns with other work that has also found unique host-associated skin microbiomes relative to the surrounding water, suggesting that fishes\u0026rsquo; skin selects for a distinct microbial community, rather than taking up microbes indiscriminately from the surrounding water [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. While trophic group and/or diet are known to strongly influence the gut microbiome, our work is one of the first to show that these factors also play an important role for fish skin microbiome composition [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. Our unique study design helps parse out phylogenetic effects from trophic group, as we have representatives from the same family (or genus) with different diets (Pomacentridae: two herbivores and an omnivore, Serranidae: two carnivores and a planktivore; \u003cem\u003eTable S2\u003c/em\u003e) and fish with similar diets from different families (herbivores: Acanthuridae \u0026amp; Pomacentridae). Importantly, host species remains significant when including diet as a predictor, which precludes the observed host species effect from simply being an artefact of a given species\u0026rsquo; dietary preferences [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Additional support for this host effect comes from paired microbiome and host genotyping work in four sympatric Serrasalmidae (piranha) species, which showed that skin microbiomes and host genotypes covaried significantly both within and across species [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Conserved relationships between hosts and their skin microbiomes could, in turn, lead to phylosymbiosis, such that microbial communities mirror their hosts\u0026rsquo; phylogenies [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. However, given the high degree of intra-specific variability in these fishes\u0026rsquo; microbiomes, we caution against recent attempts to detect phylosymbiosis in teleosts using datasets with low species-level replication [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eUnderstanding how hosts may shape their skin microbiomes requires a closer look at the skin\u0026rsquo;s surface. From a microbial perspective, the skin mucosa provides a unique niche space of gelatinous mucin [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. However, for microbes to establish communities in the fish skin requires that they evade \u0026ndash; or coexist with \u0026ndash; enzymes, immune proteins, antimicrobial peptides, and other components of the hosts\u0026rsquo; innate immune system, in addition to other competing microbial taxa [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. Hosts share similar physical traits with their conspecifics, including the skin, scales, and mucosal layer, forming a similar environment that selects for certain microbes [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. Additionally, certain immune defences in teleost skin are highly-conserved, whereas others appear to be clade or species-specific, or develop throughout an individual host\u0026rsquo;s lifetime, thus imposing distinct constraints on potential colonizers [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. Shared host traits and innate immunity would help account for the significant explanatory power of host species in determining the fish skin microbiome.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003eEnvironmental Effects\u003c/h2\u003e\u003cp\u003eUpwelling systems are distributed globally and have wide-ranging impacts on the productivity of marine ecosystems [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. How microbial communities in the water column respond to upwelling has received considerable attention in other upwelling areas, including the California Current, the Benguela Current, and Tongoy Bay [\u003cspan additionalcitationids=\"CR56\" citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e]. However, our study is the first to assess these phenomena in the Tropical Eastern Pacific of Panama, in which we observe large shifts in water microbial communities during upwelling, particularly in the Gulf of Panama, where upwelling is strongest (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, S3, S4, S6). Water microbiomes were more similar between gulfs during the wet season (non-upwelling) than they were within a gulf across seasons, indicating that seasonal turnover of microbial taxa outweighs fixed differences between gulfs (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, \u003cem\u003eTable S8\u003c/em\u003e). Interestingly, while the greatest number of differentially abundant taxa in the water communities were, as expected, seen in the Gulf of Panama during the dry season, when there is strong upwelling, we also saw a clear signal of seasonal change in the Gulf of Chiriqu\u0026iacute; during the dry season, despite no strong changes in temperature nor dissolved oxygen during this time. This may be due to other seasonal changes in coastal hydrology, such as surface runoff, altering water microbiomes during the wet season, or be linked to a localized upwelling hotspot in the Gulf of Chiriqu\u0026iacute; [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eFurthermore, this is the first study to look at the effects of upwelling-driven environmental changes on both water and vertebrate host-associated microbiomes. Compared to the surrounding seawater, significant seasonal changes were only detected in a subset of our host species\u0026rsquo; microbiomes and were of lesser magnitude. Host-associated microbiomes had around 40 times fewer significant differentially abundant (DA) taxa across region-season comparisons than water (\u003cem\u003eTable S8\u003c/em\u003e). Among these, hosts shared 0\u0026ndash;34% of their DA taxa with water samples, meaning that although a minority of the microbes that changed seasonally on hosts may be picked up from \u0026ndash; and thus directly match changes in - the surrounding environment, most host-associated microbes respond to environmental changes independently. Seasonal changes in temperature and dissolved oxygen (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, S1) associated with upwelling likely played a role in the differences seen in the fish skin microbiomes, as both variables (alongside salinity and pH) correlated with shifts in microbiome diversity in a recent review of the few studies to date that have reported DO and temperature data alongside microbiome results for saltwater fishes\u0026rsquo; skin [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e]. Hosts whose skin microbiomes were altered by seasonal upwelling \u0026ndash; namely, the herbivores, carnivores, and planktivore \u0026ndash; may be more susceptible to future environmental changes. However, without further functional information, we cannot determine whether the community-level changes we observe are harmful or helpful to the host. Microbes demonstrate high levels of phenotypic plasticity and can thrive under diverse conditions, allowing them to respond to changing environmental conditions [\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e]. Thus, rapid shifts in microbial community composition and metabolic activity could help the host, if the microbiome is altered in such a way that it continues to provide necessary services [\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e]. On the other hand, disruptions in the existing microbiome could favour the establishment of pathogenic taxa with detrimental effects. The differences in responses to seasonal environmental changes between seawater and host-associated microbial communities highlight the need for further study of the two in tandem, to better understand how host-microbe relationships may be impacted by environmental changes in different host species, populations, and habitats at risk.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003eMicrobiome Complexity and Residual Variation\u003c/h2\u003e\u003cp\u003eIndividual fishes\u0026rsquo; traits, life histories, and adaptive immune responses, in combination with microbe-microbe interactions and stochastic changes in the community due to drift, may help explain the 66% variation that was neither captured by host species nor season or region [\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e]. When considering this residual variation, it is worth noting that, while we focus here on host species as a predictor, we also assessed the effects of factors such as sex and size, which have previously been shown to impact fish gut microbiomes [\u003cspan additionalcitationids=\"CR29\" citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. However, we found no significant effect of sex and only limited effects of size, suggesting that these traits are not major determinants of microbiome structure in these fishes. Genetic diversity within host species may be another factor contributing to host-associated microbiome variation, as reef fishes often display both high genetic diversity and low genetic differentiation, even across distant reef sites [\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e, \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e]. Thus, future work within pairing host genomics and microbiome analyses could serve the dual purpose of furthering our understanding of these reef fishes\u0026rsquo; population genetic structure and how it may affect their associated microbial communities. Furthermore, once microbes successfully reach the fish skin, potential colonizers must compete to establish themselves in this limited niche space. Priority effects, including niche pre-emption (the first taxa to arrive take over), niche modifications (microbial taxa altering their environment to favour their own growth), and microbial antagonism (competition for limited resources and antimicrobial defences), play an important role in shaping host-associated microbial communities [\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e]. While such microbial assembly mechanisms are difficult to measure, especially in wild populations, repeated sampling of individuals prior to, during, and following upwelling may aid to further disentangle individual, species, regional, and seasonal effects in microbial community assembly.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eOur findings suggest that host-associated microbiome responses to future environmental changes will likely be host-specific, with certain species or taxonomic groups responding more strongly. Being associated with a host imposes constraints on the microbial communities that can form, via host immunity and environmental filtering, which acts as a buffer against the fluctuations seen in water microbial communities. These results raise additional questions: \u003cem\u003e(1)\u003c/em\u003e Why do certain hosts\u0026rsquo; microbiomes respond more strongly to environmental changes and is this an indicator of vulnerability under future climate change? \u003cem\u003e(2)\u003c/em\u003e Which microbiome functional shifts are occurring, and what are their repercussions on host and ecosystem health? \u003cem\u003e(3)\u003c/em\u003e Are there additional deterministic factors that may address the unexplained variation in these host-associated microbial communities? Further sampling of marine organisms\u0026rsquo; microbiomes with greater replication within and across species, families, and trophic groups would provide crucial insights into this first question. Techniques such as metabolomics and metatranscriptomics can be leveraged to determine functional shifts in microbial communities. Finally, monitoring additional physicochemical and host-related parameters, including sub-species level population structure via genomics research, may help further disentangle microbial community dynamics. Given the global scale of the environmental changes our oceans are facing, it is imperative that we build on our nascent understanding of the role microbes play in host and ecosystem health.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003cp\u003eAll fish were handled and euthanized following protocols approved by the Smithsonian Tropical Research Institute\u0026rsquo;s Institutional Animal Care and Use Committee (IACUC; SI-22047). We used rapid chilling (hypothermic shock) to euthanize fish. This was done by rapidly transferring fish after capture into individual labelled bags - allowing direct contact with the chilled water but not the ice - and placing them in an ice slurry in a large cooler on the boat. This method was deemed appropriate as we are working with relatively small, tropical fish that are acclimated to temperatures around 29\u0026ndash;31\u0026ordm;C. We also wanted to avoid the use of chemical agents, as these could potentially impact our downstream microbiome analyses. Fish were kept in the cooler for a minimum of 20 minutes, then assessed to confirm lack of movement, and kept on ice until we returned to the lab and began dissections.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003cp\u003eNot applicable.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003eCompeting interests\u003c/h2\u003e\u003cp\u003eWe declare we have no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis project was made possible through funding from the Mark and Rachel Rohr Foundation, the Smithsonian Tropical Research Institute (STRI) \u0026ndash; including a Short-Term Fellowship, a Graduate Mobility Award from the McGill Biology Department, a Delise Alison Graduate Student Development Award from the Redpath Museum, a FRQNT doctoral scholarship [DOI: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.69777/330691\u003c/span\u003e\u003cspan address=\"10.69777/330691\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e], a NSERC Discovery Grant [grant number 2019\u0026ndash;04549], a Canada Research Chair, and a Quebec Centre for Biodiversity Science (QCBS) Excellence Award.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eL.L.L.: conceptualization, data curation, formal analysis, investigation, methodology, visualization, writing \u0026ndash; original draft, writing \u0026ndash; review and editingM.L.: conceptualization, funding acquisition, methodology, project administration, supervision, writing \u0026ndash; review \u0026amp; editingR.D.H.B.: conceptualization, funding acquisition, methodology, supervision, writing \u0026ndash; review \u0026amp; editingN.A.H., H.Q.A., A.J.S.: investigation, writing \u0026ndash; review \u0026amp; editing\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe would like to thank the Rohr Reef Resilience team, including project manager Anabell J. Cornejo and intern Javiera Mora for their assistance with sample collection. Special thanks to intern Haesung Jee for help with sample processing, Marta Vargas (STRI molecular lab manager) for sequencing and lab support, and Sean Connolly for providing additional lab space. Thank you to Steve Paton and the STRI Physical Monitoring Program for providing TEP environmental data. Additional thanks to the Coiba National Park rangers and Naos boat drivers for facilitating access to sites, sharing knowledge of local fish communities, and aiding with sample collection.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eAdditional methods, metadata, figures, and tables are available as supplementary materials. Raw sequence reads have been deposited to the European Nucleotide Archive (ENA) at EMBL-EBI under accession number [PRJEB104461]. Code for analyses and figures are available as maintained versions on GitHub (https://github.com/lardinois21/RRR_Fish_Microbiome_16S.git) and have been archived on Zenodo https://doi.org/10.5281/zenodo.17739195).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eWilkins LGE, Leray M, O'Dea A, Yuen B, Peixoto RS, Pereira TJ, Bik HM, Coil DA, Duffy JE, Herre EA et al. 2019 Host-associated microbiomes drive structure and function of marine ecosystems. \u003cem\u003ePlos Biology\u003c/em\u003e. 17, (10.1371/journal.pbio.3000533).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLennon JT, Locey KJ. 2020 More support for Earth's massive microbiome. \u003cem\u003eBiology Direct\u003c/em\u003e. 15, (10.1186/s13062-020-00261-8).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuttenhower C, Gevers D, Knight R, Abubucker S, Badger JH, Chinwalla AT, Creasy HH, Earl AM, Fitzgerald MG, Fulton RS et al. 2012 Structure, function and diversity of the healthy human microbiome. \u003cem\u003eNature\u003c/em\u003e. 486, 207\u0026ndash;214. (10.1038/nature11234).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMcFall-Ngai M, Hadfield MG, Bosch TCG, Carey HV, Domazet-Loso T, Douglas AE, Dubilier N, Eberl G, Fukami T, Gilbert SF et al. 2013 Animals in a bacterial world, a new imperative for the life sciences. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e. 110, 3229\u0026ndash;3236. (10.1073/pnas.1218525110).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSehnal L, Brammer-Robbins E, Wormington AM, Blaha L, Bisesi J, Larkin I, Martyniuk CJ, Simonin M, Adamovsky O. 2021 Microbiome Composition and Function in Aquatic Vertebrates: Small Organisms Making Big Impacts on Aquatic Animal Health. \u003cem\u003eFrontiers in Microbiology\u003c/em\u003e. 12, (10.3389/fmicb.2021.567408).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHenry LP, Bruijning M, Forsberg SKG, Ayroles JF. 2021 The microbiome extends host evolutionary potential. \u003cem\u003eNature Communications\u003c/em\u003e. 12, (10.1038/s41467-021-25315-x).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLau JA, Lennon JT. 2012 Rapid responses of soil microorganisms improve plant fitness in novel environments. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e. 109, 14058\u0026ndash;14062. (10.1073/pnas.1202319109).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSoen Y. 2014 Environmental disruption of host-microbe co-adaptation as a potential driving force in evolution. \u003cem\u003eFrontiers in Genetics\u003c/em\u003e. 5, (10.3389/fgene.2014.00168).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCavicchioli R, Ripple WJ, Timmis KN, Azam F, Bakken LR, Baylis M, Behrenfeld MJ, Boetius A, Boyd PW, Classen AT et al. 2019 Scientists' warning to humanity: microorganisms and climate change. \u003cem\u003eNature Reviews Microbiology\u003c/em\u003e. 17, 569\u0026ndash;586. (10.1038/s41579-019-0222-5).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWebster NS, Negri AP, Botte ES, Laffy PW, Flores F, Noonan S, Schmidt C, Uthicke S. 2016 Host-associated coral reef microbes respond to the cumulative pressures of ocean warming and ocean acidification. \u003cem\u003eScientific Reports\u003c/em\u003e. 6, (10.1038/srep19324).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHalpern BS, Walbridge S, Selkoe KA, Kappel CV, Micheli F, D'Agrosa C, Bruno JF, Casey KS, Ebert C, Fox HE et al. 2008 A global map of human impact on marine ecosystems. \u003cem\u003eScience\u003c/em\u003e. 319, 948\u0026ndash;952. (10.1126/science.1149345).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eApprill A. 2017 Marine Animal Microbiomes: Toward Understanding Host-Microbiome Interactions in a Changing Ocean. \u003cem\u003eFrontiers in Marine Science\u003c/em\u003e. 4, (10.3389/fmars.2017.00222).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eClever F, Sourisse JM, Preziosi RF, Eisen JA, Guerra ECR, Scott JJ, Wilkins LGE, Altieri AH, McMillan WO, Leray M. 2022 The gut microbiome variability of a butterflyfish increases on severely degraded Caribbean reefs. \u003cem\u003eCommunications Biology\u003c/em\u003e. 5, (10.1038/s42003-022-03679-0).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDegregori S, Casey JM, Barber PH. 2021 Nutrient pollution alters the gut microbiome of a territorial reef fish. \u003cem\u003eMarine Pollution Bulletin\u003c/em\u003e. 169, (10.1016/j.marpolbul.2021.112522).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGomez JA, Primm TP. 2021 A Slimy Business: the Future of Fish Skin Microbiome Studies. \u003cem\u003eMicrobial Ecology\u003c/em\u003e. 82, 275\u0026ndash;287. (10.1007/s00248-020-01648-w).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChiarello M, Auguet J-C, Bettarel Y, Bouvier C, Claverie T, Graham NAJ, Rieuvilleneuve F, Sucre E, Bouvier T, Villeger S. 2018 Skin microbiome of coral reef fish is highly variable and driven by host phylogeny and diet. \u003cem\u003eMicrobiome\u003c/em\u003e. 6, (10.1186/s40168-018-0530-4).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChiarello M, Auguet J-C, Graham NAJ, Claverie T, Sucre E, Bouvier C, Rieuvilleneuve F, Restrepo-Ortiz CX, Bettarel Y, Villeger S et al. ,\u003cem\u003e. 2020 Exceptional but vulnerable microbial diversity in coral reef animal surface microbiomes. Proceedings of the Royal Society B-Biological Sciences. 287, (10.1098/rspb.2020.0642)\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePereira A, Soares MC, Santos T, Pocas A, Perez-Losada M, Apprill A, Sikkel PC, Xavier R. 2022 Reef Location and Client Diversity Influence the Skin Microbiome of the Caribbean Cleaner Goby Elacatinus evelynae. \u003cem\u003eMicrobial ecology\u003c/em\u003e. (10.1007/s00248-022-01984-z).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXavier R, Mazzei R, Perez-Losada M, Rosado D, Santos JL, Verissimo A, Soares MC. 2019 A Risky Business? Habitat and Social Behavior Impact Skin and Gut Microbiomes in Caribbean Cleaning Gobies. \u003cem\u003eFrontiers in Microbiology\u003c/em\u003e. 10, (10.3389/fmicb.2019.00716).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXavier R, Pereira A, Pagan A, Hendrick GC, Nicholson MD, Rosado D, Soares MC, Perez-Losada M, Sikkel PC. 2020 The effects of environment and ontogeny on the skin microbiome of two Stegastes damselfishes (Pomacentridae) from the eastern Caribbean Sea. \u003cem\u003eMarine Biology\u003c/em\u003e. 167, (10.1007/s00227-020-03717-7).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGaland PE, Ruscheweyh HJ, Salazar G, Hochart C, Henry N, Hume BCC, Oliveira PH, Perdereau A, Labadie K, Belser C et al. 2023 Diversity of the Pacific Ocean coral reef microbiome. \u003cem\u003eNature Communications\u003c/em\u003e. 14, (10.1038/s41467-023-38500-x).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBerggren H, Yıldırım Y, Nordahl O, Larsson P, Dopson M, Tibblin P, Lundin D, Pinhassi J, Forsman A. 2024 Ecological filtering drives rapid spatiotemporal dynamics in fish skin microbiomes. \u003cem\u003eMol Ecol\u003c/em\u003e. 33, e17496. (10.1111/mec.17496).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFiedler PC, Lav\u0026iacute;n MF. Oceanographic Conditions of the Eastern Tropical Pacific. Coral Reefs of the Eastern Tropical Pacific. Dordrecht: Springer; 2017. pp. 59\u0026ndash;83. D., E. I.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRandall CJ, Toth LT, Leichter JJ, Mat\u0026eacute; JL, Aronson RB. 2020 Upwelling buffers climate change impacts on coral reefs of the eastern tropical Pacific. \u003cem\u003eEcology\u003c/em\u003e. 101, (10.1002/ecy.2918).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCrawford G, Mepstead M, D\u0026rsquo;\u0026iacute;az-Ferguson E. Characterizing oceanographic conditions near Coiba Island and Pacific Panama using 20 years of satellite-based wind stress, SST and chlorophyll-a measurements. Mar Fish Sci. 2024;37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.47193/mafis.37X2024010112)\u003c/span\u003e\u003cspan address=\"10.47193/mafis.37X2024010112)\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeray M, Wilkins LGE, Apprill A, Bik HM, Clever F, Connolly SR, De Leon ME, Duffy JE, Ezzat L, Gignoux-Wolfsohn S et al. 2021 Natural experiments and long-term monitoring are critical to understand and predict marine host-microbe ecology and evolution. \u003cem\u003ePlos Biology\u003c/em\u003e. 19, (10.1371/journal.pbio.3001322).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRobertson DR, Allen GR. Shorefishes of the Tropical Eastern Pacific: online information system. Balboa, Panam\u0026aacute;: Smithsonian Tropical Research Institute; 2015.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi H, Lu L, Chen R, Li S, Xu D. 2022 Exploring Sexual Dimorphism in the Intestinal Microbiota of the Yellow Drum (Nibea albiflora, Sciaenidae). \u003cem\u003eFrontiers in Microbiology\u003c/em\u003e. 12, (10.3389/fmicb.2021.808285).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMessina M, Iacumin L, Pascon G, Tulli F, Tibaldi E, Cardinaletti G. 2023 Effect of feed restriction and refeeding on body condition, digestive functionality and intestinal microbiota in rainbow trout (Oncorhynchus mykiss). \u003cem\u003eFish Physiology and Biochemistry\u003c/em\u003e. 49, 169\u0026ndash;189. (10.1007/s10695-023-01170-z).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBates KA, Higgins C, Neiman M, King KC. 2023 Turning the tide on sex and the microbiota in aquatic animals. \u003cem\u003eHydrobiologia\u003c/em\u003e. 850, 3823\u0026ndash;3835. (10.1007/s10750-022-04862-4).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBledsoe JW, Peterson BC, Swanson KS, Small BC. 2016 Ontogenetic characterization of the intestinal microbiota of channel catfish through 16S rRNA gene sequencing reveals insights on temporal shifts and the influence of environmental microbes. \u003cem\u003ePLoS ONE\u003c/em\u003e. 11, (10.1371/journal.pone.0166379).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRosado D, P\u0026eacute;rez-Losada M, Pereira A, Severino R, Xavier R. 2021 Effects of aging on the skin and gill microbiota of farmed seabass and seabream. \u003cem\u003eAnimal Microbiome\u003c/em\u003e. 3, (10.1186/s42523-020-00072-2).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCaporaso GJ, Ackermann G, Apprill A, Bauer M, Berg-Lyons D, Betley J, Fierer N, Fraser L, Fuhrman JA, Gilbert JA et al. ,\u003cem\u003e. EMP 16S Illumina Amplicon Protocol V.2. 2023 [cited 2024; Available from\u003c/em\u003e:\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCallahan BJ, Sankaran K, Fukuyama JA, McMurdie PJ, Holmes SP. 2016 Bioconductor Workflow for Microbiome Data Analysis: from raw reads to community analyses. \u003cem\u003eF1000Research\u003c/em\u003e. 5, 1492.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRoswell M, Dushoff J, Winfree R. 2021 A conceptual guide to measuring species diversity. \u003cem\u003eOikos\u003c/em\u003e. 130, 321\u0026ndash;338. (10.1111/oik.07202).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCappellato M, Baruzzo G, Di Camillo B. 2022 Investigating differential abundance methods in microbiome data: A benchmark study. \u003cem\u003ePLoS Comput Biol\u003c/em\u003e. 18, e1010467. (10.1371/journal.pcbi.1010467).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMallick H, Rahnavard A, McIver L, Ma S, Zhang Y, Nguyen L, et al. Multivariable association discovery in population-scale meta-omics studies. PLoS Comput Biol. 2021;17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pcbi.1009442)\u003c/span\u003e\u003cspan address=\"10.1371/journal.pcbi.1009442)\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLove M, Huber W, Anders S. 2014 Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOllberding N. Identifying Differentially Abundant Features in Microbiome Data. 2021.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOren A, Garrity GM. 2021 Valid publication of the names of forty- two phyla of prokaryotes. \u003cem\u003eInternational Journal of Systematic and Evolutionary Microbiology\u003c/em\u003e. 71, (10.1099/ijsem.0.005056).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBerggren H, Tibblin P, Yıldırım Y, Broman E, Larsson P, Lundin D, Forsman A. 2022 Fish Skin Microbiomes Are Highly Variable Among Individuals and Populations but Not Within Individuals. \u003cem\u003eFrontiers in Microbiology\u003c/em\u003e. 12, (10.3389/fmicb.2021.767770).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSadeghi J, Chaganti SR, Johnson TB, Heath DD. 2023 Host species and habitat shape fish-associated bacterial communities: phylosymbiosis between fish and their microbiome. \u003cem\u003eMicrobiome\u003c/em\u003e. 11, 258. (10.1186/s40168-023-01697-6).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDegregori S, Schiettekatte NMD, Casey JM, Brandl SJ, Merci\u0026egrave;re A, Amato KR, Mazel F, Parravicini V, Barber PH. 2024 Host diet drives gut microbiome convergence between coral reef fishes and mammals. \u003cem\u003eMolecular Ecology\u003c/em\u003e. 33, (10.1111/mec.17520).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYoungblut ND, Reischer GH, Walters W, Schuster N, Walzer C, Stalder G, Ley RE, Farnleitner AH. 2019 Host diet and evolutionary history explain different aspects of gut microbiome diversity among vertebrate clades. \u003cem\u003eNature Communications\u003c/em\u003e. 10, (10.1038/s41467-019-10191-3).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSylvain FE, Normandeau E, Holland A, Val AL, Derome N. 2022 Genomics of Serrasalmidae teleosts through the lens of microbiome fingerprinting. \u003cem\u003eMolecular Ecology\u003c/em\u003e. 31, 4656\u0026ndash;4671. (10.1111/mec.16574).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLim SJ, Bordenstein SR. 2020 An introduction to phylosymbiosis. \u003cem\u003eProceedings of the Royal Society B-Biological Sciences\u003c/em\u003e. 287, (10.1098/rspb.2019.2900).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu Y, Li X, Kormas KA, Li Y, Li H, Li J. 2025 Variable phylosymbiosis and cophylogeny patterns in wild fish gut microbiota of a large subtropical river. \u003cem\u003emSphere\u003c/em\u003e. e0098224-e0098224. (10.1128/msphere.00982\u0026thinsp;\u0026ndash;\u0026thinsp;24).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHan GH, Yu J, Kang MJ, Park MJ, Noh CH, Kim YJ, Kwon KK. 2024 Phylosymbiosis in Seven Wild Fish Species Collected Off the Southern Coast of Korea: Skin Microbiome Most Strongly Reflects Evolutionary Pressures (vol 87,153,2024). \u003cem\u003eMicrobial Ecology\u003c/em\u003e. 87, (10.1007/s00248-024-02478-w).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDash S, Das SK, Samal J, Thatoi HN. 2018 Epidermal mucus, a major determinant in fish health: a review. Iran J Veterinary Res. 19, 72\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang LC, Chen LH, Chiu YC, Liou CY, Chen HC, Lu CY, Chen JL. 2023 Teleost skin microbiome: An intimate interplay between the environment and the host immunity. \u003cem\u003eFish \u0026amp; Shellfish Immunology\u003c/em\u003e. 139, (10.1016/j.fsi.2023.108869).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuardiola FA, Cuartero M, Collado-Gonz\u0026aacute;lez MD, Arizc\u0026uacute;n M, Ba\u0026ntilde;os FGD, Meseguer J, Cuesta A, Esteban MA. 2015 Description and comparative study of physico-chemical parameters of the teleost fish skin mucus. \u003cem\u003eBiorheology\u003c/em\u003e. 52, 247\u0026ndash;256. (10.3233/bir-15052).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuardiola FA, Cuesta A, Abell\u0026aacute;n E, Meseguer J, Esteban MA. 2014 Comparative analysis of the humoral immunity of skin mucus from several marine teleost fish. \u003cem\u003eFish \u0026amp; Shellfish Immunology\u003c/em\u003e. 40, 24\u0026ndash;31. (10.1016/j.fsi.2014.06.018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBuchmann K, Karami AM, Duan YJ. 2024 The early ontogenetic development of immune cells and organs in teleosts. \u003cem\u003eFish \u0026amp; Shellfish Immunology\u003c/em\u003e. 146, (10.1016/j.fsi.2024.109371).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang DW, Gouhier TC, Menge BA, Ganguly AR. 2015 Intensification and spatial homogenization of coastal upwelling under climate change. \u003cem\u003eNature\u003c/em\u003e. 518, 390\u0026ndash;394. (10.1038/nature14235).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRocke E, Cheung S, Gebe Z, Dames NR, Liu HB, Moloney CL. 2020 Marine Microbial Community Composition During the Upwelling Season in the Southern Benguela. \u003cem\u003eFrontiers in Marine Science\u003c/em\u003e. 7, (10.3389/fmars.2020.00255).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eReji L, Tolar BB, Chavez FP, Francis CA. 2020 Depth-Differentiation and Seasonality of Planktonic Microbial Assemblages in the Monterey Bay Upwelling System. \u003cem\u003eFrontiers in Microbiology\u003c/em\u003e. 11, (10.3389/fmicb.2020.01075).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMolina V, Belmar L, Levipan HA, Ram\u0026iacute;rez-Flandes S, Anguita C, Gal\u0026aacute;n A, Montes I, Ulloa O. 2020 Spatiotemporal Distribution of Key Pelagic Microbes in a Seasonal Oxygen-Deficient Coastal Upwelling System of the Eastern South Pacific Ocean. \u003cem\u003eFrontiers in Marine Science\u003c/em\u003e. 7, (10.3389/fmars.2020.561597).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eD'Croz L, O'Dea A. 2007 Variability in upwelling along the Pacific shelf of Panama and implications for the distribution of nutrients and chlorophyll. \u003cem\u003eEstuarine Coastal and Shelf Science\u003c/em\u003e. 73, 325\u0026ndash;340. (10.1016/j.ecss.2007.01.013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBell AG, McMurtrie J, Bola\u0026ntilde;os LM, Cable J, Temperton B, Tyler CR. 2024 Influence of host phylogeny and water physicochemistry on microbial assemblages of the fish skin microbiome. \u003cem\u003eFems Microbiology Ecology\u003c/em\u003e. 100, (10.1093/femsec/fiae021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNemergut DR, Schmidt SK, Fukami T, O'Neill SP, Bilinski TM, Stanish LF, Knelman JE, Darcy JL, Lynch RC, Wickey P et al. 2013 Patterns and Processes of Microbial Community Assembly. \u003cem\u003eMicrobiology and Molecular Biology Reviews\u003c/em\u003e. 77, 342\u0026ndash;356. (10.1128/mmbr.00051\u0026thinsp;\u0026ndash;\u0026thinsp;12).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWebster TUM, Rodriguez-Barreto D, Castaldo G, Gough P, Consuegra S, de Leaniz CG. 2020 Environmental plasticity and colonisation history in the Atlantic salmon microbiome: A translocation experiment. \u003cem\u003eMolecular Ecology\u003c/em\u003e. 29, 886\u0026ndash;898. (10.1111/mec.15369).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePetersen C, Hamerich IK, Adair KL, Griem-Krey H, Oliva MT, Hoeppner MP, Bohannan BJM, Schulenburg H. 2023 Host and microbiome jointly contribute to environmental adaptation. \u003cem\u003eIsme Journal\u003c/em\u003e. 17, 1953\u0026ndash;1965. (10.1038/s41396-023-01507-9).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePineros VJ, Gutierrez-Rodriguez C. 2017 Population genetic structure and connectivity in the widespread coral-reef fish Abudefduf saxatilis: the role of historic and contemporary factors. \u003cem\u003eCoral Reefs\u003c/em\u003e. 36, 877\u0026ndash;890. (10.1007/s00338-017-1579-4).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKlanten OS, Choat JH, van Herwerden L. 2007 Extreme genetic diversity and temporal rather than spatial partitioning in a widely distributed coral reef fish. \u003cem\u003eMarine Biology\u003c/em\u003e. 150, 659\u0026ndash;670. (10.1007/s00227-006-0372-7).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChavez-Arroyo A, Radlinski LC, Baumler AJ. 2025 Principles of gut microbiota assembly. \u003cem\u003eTrends in microbiology\u003c/em\u003e. (10.1016/j.tim.2025.02.014).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-microbiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mcro","sideBox":"Learn more about [BMC Microbiology](http://bmcmicrobiol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/mcro","title":"BMC Microbiology","twitterHandle":"#bmcmicrobiology","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8158492/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8158492/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eDisentangling the drivers structuring microbiomes can help predict organisms\u0026rsquo; responses to rapid environmental change. However, despite microbial communities being important for both host and environmental health, large gaps remain in our understanding of how host-associated microbiomes are structured and respond to different stimuli, especially in marine environments. Here, we leverage seasonal upwelling in Panama\u0026rsquo;s Tropical Eastern Pacific to test how abrupt environmental changes linked to seasonal upwelling influence the diversity and composition of coral reef fish skin microbiomes in ten species spanning four trophic groups.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eFish skin microbiomes varied greatly within and among host species and were distinct from the microbiomes of the surrounding seawater. All species had diverse skin microbiomes, with a dominance of Proteobacteria (65%), Bacteroidota (13%), and Cyanobacteria (6%). Host species and trophic group played a greater role in determining fish skin microbiome structure than seasonal and regional environmental variation, despite water microbiomes responding strongly to both season and region. Nevertheless, three out of five trophic groups: the herbivores, carnivores, and planktivore, also displayed significant changes in their microbiomes during upwelling, albeit to a lesser extent than water samples. We performed differential abundance (DA) analyses on these fish and compared microbial taxa that changed between seasons and regions in fish versus water samples. While water communities had thousands of significant DA taxa, fish had around 40 times fewer (n\u0026thinsp;=\u0026thinsp;17 to 73) and only shared 30 DA taxa with the water samples. Differences between these microbial communities likely arise from both host selection via fishes\u0026rsquo; immune system and the skin mucus serving as an environmental filter. However, neither host-associated nor environmental predictors fully explained the variation in microbiome composition, highlighting its complexity.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eOur results show how ecological differences between host species may elicit distinct microbiome responses to environmental changes, with potential cascading effects on ecosystem dynamics under global climate change. Further characterization of marine microbial communities, as well as additional physicochemical and host-related parameters, will be key to monitoring and predicting how these communities will respond to the increasingly rapid and widespread environmental changes our oceans are facing.\u003c/p\u003e","manuscriptTitle":"Host matters: coral reef fish species show distinct skin microbiome responses to abrupt environmental change","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-02 09:25:37","doi":"10.21203/rs.3.rs-8158492/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-28T16:36:47+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-20T13:00:33+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-06T23:09:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"38321038059743066786121050111458886384","date":"2026-01-06T09:04:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"91185735253171276211035321111994083308","date":"2026-01-05T18:54:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"62459991298908325709029296577423758084","date":"2026-01-05T14:23:49+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-29T10:38:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"326880421960050877082045213762570578142","date":"2025-12-23T15:55:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"250296408290465339034007030434289886696","date":"2025-12-08T16:38:14+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-05T08:49:08+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-02T07:22:31+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-01T10:11:30+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Microbiology","date":"2025-12-01T02:42:06+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-microbiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mcro","sideBox":"Learn more about [BMC Microbiology](http://bmcmicrobiol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/mcro","title":"BMC Microbiology","twitterHandle":"#bmcmicrobiology","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"d6ea83b2-e119-4961-9880-551e70c6e1a8","owner":[],"postedDate":"December 2nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-27T16:07:22+00:00","versionOfRecord":{"articleIdentity":"rs-8158492","link":"https://doi.org/10.1186/s12866-026-05036-1","journal":{"identity":"bmc-microbiology","isVorOnly":false,"title":"BMC Microbiology"},"publishedOn":"2026-04-21 15:59:26","publishedOnDateReadable":"April 21st, 2026"},"versionCreatedAt":"2025-12-02 09:25:37","video":"","vorDoi":"10.1186/s12866-026-05036-1","vorDoiUrl":"https://doi.org/10.1186/s12866-026-05036-1","workflowStages":[]},"version":"v1","identity":"rs-8158492","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8158492","identity":"rs-8158492","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}