Microbiome differences in disease-resistant vs. susceptible Acropora corals subjected to disease challenge assays

Abstract

In recent decades coral gardening has become increasingly popular to restore degraded reef ecosystems. However, the growth and survivorship of nursery-reared outplanted corals are highly variable. Scientists are trying to identify genotypes that show signs of disease resistance and leverage these genotypes in restoring more resilient populations. In a previous study, a field disease grafting assay was conducted on nursery-reared Acropora cervicornis and Acropora palmata to quantify relative disease susceptibility. In this study, we further evaluate this field assay by investigating putative disease-causing agents and the microbiome of corals with disease-resistant phenotypes. We conducted 16S rRNA gene high-throughput sequencing on A. cervicornis and A. palmata that were grafted (inoculated) with a diseased A. cervicornis fragment. We found that independent of health state, A. cervicornis and A. palmata had distinct alpha and beta diversity patterns from one another and distinct dominant bacteria. In addition, despite different microbiome patterns between both inoculated coral species, the genus Sphingomonadaceae was significantly found in both diseased coral species. Additionally, a core bacteria member from the order Myxococcales was found at relatively higher abundances in corals with lower rates of disease development following grafting. In all, we identified Sphingomonadaceae as a putative coral pathogen and a bacterium from the order Myxococcales associated with corals that showed disease resistant phenotypes.

Figure 1

A. cervicornis and A. palmata show distinct microbial communities. There were significant differences between A. cervicornis and A. palmata in microbial (A) Shannon diversity (rarefied to 8,721) (B) evenness (rarefied to 8,721) and (C) beta-diversity (values were centered log ratio (CLR) transformed and plotted with a Euclidean distance on a principal component analysis (PCA) and only amplicon sequence variants (ASVs) present in >4 samples were used.). (D) The cumulative relative abundances of the most abundant microbial genera (>0.05%, not rarefied) per genotype. Each stacked color bar represents a different genus. The figure is grouped by coral species and treatment. For figures (A–C), circle = control samples, triangles = inoculated samples, blue = Acropora cervicornis, and tan = Acropora palmata.

Figure 2

A. cervicornis and A. palmata show distinct beta-diversity patterns to disease exposure. (A) Principal component analysis (PCA) with a Euclidean distance of A. cervicornis colored by experiment outcome. (B) Dispersion of beta-diversity from the experiment outcome for A. cervicornis. (C) PCA with a Euclidean distance of A. palmata colored by experiment outcome (D) Dispersion of beta-diversity from the experiment outcome for A. palmata.

Figure 3

ASVs from four bacterial families were significantly associated with disease exposure. (A) Box plots show the relative abundance of a differential abundance analysis of the experiment outcome that resulted in two significantly abundant ASVs in A. cervicornis. (B) Differential abundance analysis of the experiment outcome resulted in three significantly abundant ASVs in A. palmata. (C) The average relative abundance of significantly differentiated ASVs per sample and grouped by outcome. For figures A–C bacterial families are represented by different colors.