Comparative metagenomics of tropical reef fishes show conserved core gut functions across hosts and diets with diet-related functional gene enrichments

Abstract

Fish gut microbial communities are important for the breakdown and energy harvesting of the host diet. Microbes within the fish gut are selected by environmental and evolutionary factors. To understand how fish gut microbial communities are shaped by diet, three tropical fish species (hawkfish, Paracirrhites arcatus; yellow tang, Zebrasoma flavescens; and triggerfish, Rhinecanthus aculeatus) were fed piscivorous (fish meal pellets), herbivorous (seaweed), and invertivorous (shrimp) diets, respectively. From fecal samples, a total of 43 metagenome assembled genomes (MAGs) were recovered from all fish diet treatments. Each host-diet treatment harbored distinct microbial communities based on taxonomy, with Proteobacteria, Bacteroidota, and Firmicutes being the most represented. Based on their metagenomes, MAGs from all three host-diet treatments demonstrated a baseline ability to degrade proteinaceous, fatty acid, and simple carbohydrate inputs and carry out central carbon metabolism, lactate and formate fermentation, acetogenesis, nitrate respiration, and B vitamin synthesis. The herbivorous yellow tang harbored more functionally diverse MAGs with some complex polysaccharide degradation specialists, while the piscivorous hawkfish's MAGs were more specialized for the degradation of proteins. The invertivorous triggerfish's gut MAGs lacked many carbohydrate-degrading capabilities, resulting in them being more specialized and functionally uniform. Across all treatments, several MAGs were able to participate in only individual steps of the degradation of complex polysaccharides, suggestive of microbial community networks that degrade complex inputs.

Importance: The benefits of healthy microbiomes for vertebrate hosts include the breakdown of food into more readily usable forms and production of essential vitamins from their host's diet. Compositions of microbial communities in the guts of fish in response to diet have been studied, but there is a lack of a comprehensive understanding of the genome-based metabolic capabilities of specific microbes and how they support their hosts. Therefore, we assembled genomes of several gut microbes collected from the feces of three fish species that were being fed different diets to illustrate how individual microbes can carry out specific steps in the degradation and energy utilization of various food inputs and support their host. We found evidence that fish gut microbial communities share several core functions despite differences in microbial taxonomy. Herbivorous fish harbored a functionally diverse microbial community with plant matter degraders, while the piscivorous and invertivorous fish had microbiomes more specialized in protein degradation.

Keywords: fish gut; herbivore; invertivore; metagenome; piscivore.

Fig 1

Coverage of 30S ribosomal protein annotations from the assembled metagenomes compared to coverage within each MAG. Samples shown are (a) piscivorous hawkfish, (b) herbivorous yellow tang, and (c) invertivorous triggerfish. The comparison highlights which taxa are represented by MAGs. Presumably, those not in MAGs have a high degree of species diversity and could not be accurately binned. The taxonomy of the 30S ribosomal protein gene was determined by the best hit in UniProt, and the taxonomy of the MAG was determined by GTDB.

Fig 2

Principal component analysis of normalized enzyme gene copy numbers across all MAGs recovered from the three diet treatments. The total numbers of enzymes of each degradation pathway (proteins, cellulose, hemicellulose, starch, and chitin) normalized to the MAG’s genome completeness were used to generate components. Fatty acids were included as a binary where 1 designated that the MAG could complete all four steps of beta oxidation, while 0 designated that it could not. Ellipses represent t distributions around the centroids. Ellipses are labeled for clarity. Individual points represent individual MAGs. The larger shapes in the ellipses are group centroids.

Fig 3

Polysaccharide degradation pathway to monosaccharide products. Shown are (a) cellulose, (b) xylan, (c) mannan, (d) xyloglucan, and (e) chitin. Enzymes and their CAZyme IDs are noted above arrows connecting substrates and products. MAGs that possess genes for each enzyme are denoted beneath the arrow, where dark brown (left), green (center), and pink (right) font represent the piscivorous, herbivorous, and invertivorous treatments, respectively.

Fig 4

Number of substrate degradation genes per MAG normalized to genome completeness. Shown are (a) cellulose, (b) hemicellulose, (c) chitin, (d) starch, and (e) proteins in each diet treatment. Kruskal–Wallis values above bars represent group comparisons. Pairwise comparison values denoted by brackets are the result of post-hoc Wilcoxon tests between diets.

Fig 5

Heatmaps of metabolic functions that can be performed by recovered MAGs. Samples shown are (a) piscivorous hawkfish, (b) herbivorous yellow tang, and (c) invertivorous triggerfish. Red indicates the ability to perform a metabolic pathway, while blue indicates inability.