Gut microbial communities of hybridising pygmy angelfishes reflect species boundaries

Abstract

Hybridisation and introgression of eukaryotic genomes can generate new species or subsume existing ones, with direct and indirect consequences for biodiversity. An understudied component of these evolutionary forces is their potentially rapid effect on host gut microbiomes, and whether these pliable microcosms may serve as early biological indicators of speciation. We address this hypothesis in a field study of angelfishes (genus Centropyge), which have one of the highest prevalence of hybridisation within coral reef fish. In our study region of the Eastern Indian Ocean, the parent fish species and their hybrids cohabit and display no differences in their diet, behaviour, and reproduction, often interbreeding in mixed harems. Despite this ecological overlap, we show that microbiomes of the parent species are significantly different from each other in form and function based on total community composition, supporting the division of parents into distinct species, despite the confounding effects of introgression acting to homogenize parent species identity at other molecular markers. The microbiome of hybrid individuals, on the other hand, are not significantly different to each of the parents, instead harbouring an intermediate community composition. These findings suggest that shifts in gut microbiomes may be an early indicator of speciation in hybridising species.

Fig. 1. Overview of factors that may impact gastrointestinal microbial communities of hybridising reef fishes.

Conceptual diagram of the various factors that may influence the microbial community composition of hybridising reef fishes such as pygmy angelfishes. These include factors such as age and sex of individuals, their position within a harem, the specific harem that they belong to, and the makeup of individuals with a harem, use of the surrounding habitat, and the extent of genetic exchange via interbreeding. Photos of fish taken by Tane Sinclair-Taylor (co-author), all other images sourced from the Symbols and Integration and Application Network (ian.umces.edu/media-library).

Fig. 2. Abundance of prokaryotic phyla within the mid and hindguts of the lemonpeel angelfish Centropyge flavissima (n = 7).

Total sequence abundance of phyla in midguts and hindguts of Centropyge flavissima (n = 7). Black horizontal bars indicate individual ASVs within phyla. HG hindgut sample, MG midgut sample.

Fig. 3. Alpha diversity of hindgut microbial communities of pygmy angelfishes and their hybrids (n = 8).

Alpha diversity of hindgut microbial communities based on 16S rDNA gene sequencing. No differences were detected among sample types for any alpha diversity measures (one-way ANOVA, p < 0.05). Observed species number and chao index were calculated using rarefied data; Shannon, Simpson and Fisher’s indices were calculated using unrarefied data. C. eibli = Centropyge eibli, C. flavissima = Centropyge flavissima, Hybrids = hybrids of C. eibli and C. flavissima (n = 8). Boxes denote the interquartile range between the 25th and 75th percentiles of the data, and the horizontal line inside the box reflects the median value. The upper and lower whiskers represent scores outside this middle 50%. Dots outside whiskers represent outliers.

Fig. 4. Non-metric multi-dimensional scaling (nMDs) plot comparing hindgut microbial community composition of pygmy angelfishes and their hybrids (n = 8).

Two-dimensional nMDS ordination plot and 80% confidence ellipses constructed from bootstrap averages for the Bray-Curtis similarity index of square root transformed hindgut microbial community data from the two angelfish species, Centropyge flavissima and C. eibli, and their hybrids (n = 8).

Fig. 5. Abundance of prokaryotic operational taxonomic units (OTUs) that differ within hindguts of the pygmy angelfish and their hybrids (n = 8).

Percent sequence abundance of OTUs that were significantly different within hindguts of the two angelfish species, Centropyge flavissima and C. eibli, and their hybrids (n = 8) based on mvabund analysis (p < 0.01). Taxonomic ranks are listed to genus, or the lowest possible taxonomic rank otherwise.