Naturally occurring fire coral clones demonstrate a genetic and environmental basis of microbiome composition

Coral microbiomes are critical to holobiont functioning, but much remains to be understood about how prevailing environment and host genotype affect microbial communities in ecosystems. Resembling human identical twin studies, we examined bacterial community differences of naturally occurring fire coral clones within and between contrasting reef habitats to assess the relative contribution of host genotype and environment to microbiome structure. Bacterial community composition of coral clones differed between reef habitats, highlighting the contribution of the environment. Similarly, but to a lesser extent, microbiomes varied across different genotypes in identical habitats, denoting the influence of host genotype. Predictions of genomic function based on taxonomic profiles suggest that environmentally determined taxa supported a functional restructuring of the microbial metabolic network. In contrast, bacteria determined by host genotype seemed to be functionally redundant. Our study suggests microbiome flexibility as a mechanism of environmental adaptation with association of different bacterial taxa partially dependent on host genotype.


Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences
Behavioural & social sciences Ecological, evolutionary & environmental sciences For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Ecological, evolutionary & environmental sciences study design
All studies must disclose on these points even when the disclosure is negative.

Study description
Research sample
Here we examined bacterial community differences of naturally occurring fire coral clones within and between contrasting reef environments (135 clones, 6 genotypes, 3 reef habitats) to assess the relative contribution of host genotype and environment to microbiome structure. Bacterial community composition of fire coral clones differed between reef habitats, highlighting the contribution of the environment. Similarly, but to a lesser extent, microbiomes varied across different genotypes in identical habitats, denoting the influence of host genotype. The effect and interaction of 'Genotype' and 'Habitat' on bacterial ASV composition was tested using a two-way PERMANOVA, while the effect of 'Genotype' and 'Habitat' was investigated separately using one-way PERMANOVAs. Results of the one-way PERMANOVAs were visualized in non-metric multidimensional scaling (NMDS) ordination plots. For the genetic contribution to variation in bacterial communities, each plot shows one of the three habitats (mid slope, upper slope and back reef) in which the different genotypes were found; overall each genotype had between 13 and 38 replicates: G1 = 38, G2 = 16, G3 = 15, G4 = 32, G5 = 13, G6 = 21. For the environmental contribution to variation in bacterial communities, each plot shows one of the six genotypes with clones found in different habitats; overall each habitat had between 19 to 90 replicates: mid slope = 26, upper slope = 90, back reef = 19. We also found bacterial indicator taxa specific to both, host genotype (irrespective of the environment) and reef habitat (irrespective of clonal identity). Indicator bacterial taxa associated with a given genotype and/or habitat was identified using the IndicSpecies package for corrections of unequal sample sizes with the function Indval.g and only ASVs that were highly significantly (P < 0.01) associated with one or several groups were considered: 1-34 genotype-specific ASVs and 3-195 habitat-specific ASVs were identified in this study. At last, our results suggest a functional restructuring of the microbial metabolic network between reef habitats, while the genotype-specific associations seemed to be governed by functional redundancy. Among all functional traits identified using predictive metagenomic analysis, 24 predicted functions distinguished the microbial communities associated with distinct reef habitats (LDA > 2.5), while no discriminant functional traits were identified that differentiated host genotype microbiomes. Taken together, our study argues for a contribution of the microbiome to environmental adaptation, in the form of different taxonomic compositions depending on host genotype.
The work presented here is based on a previous collection of nearly 4,000 geo-referenced colonies of fire corals, for which we could demonstrate that clones of the same genotype display different morphologies across variable environments [1]. This work has provided us a unique opportunity to assess the extent of host genetics and environment on microbiome composition by using genetically identical coral hosts across diverse environmental conditions (temperature and light). Fire corals of the genus Millepora (Cnidaria, Hydrozoa), similar to stony corals (Cnidaria, Scleractinia), are an important component of reef communities worldwide where they shelter symbiotic algae and microbes and build calcareous skeletons, and thus contribute to reef accretion and community dynamics. A recent study of Millepora platyphylla identified several genotypes with clones found across distinct environments on a barrier reef ecosystem in Moorea, French Polynesia [1]. These clones were produced naturally through asexual fragmentation (i.e. likely wave-induced breakage), while dispersed across adjacent habitats (< 210 m apart) via crossreef transport. Specific environmental gradients across spatially adjacent reef habitats, such as light incidence, temperature, nutrients and water flow (among others), have been reported as underlying factors of substantial variation in the occurrence and persistence of bacterial symbionts. Similar to studying human identical twin microbiome structure and function, fire coral clones provide an ideal natural system to study the relative contribution of host genotype and environment to bacterial association. For this study, we selected six genotypes with at least four clonal replicates in at least two of the three surveyed habitats, mid slope, upper slope and back reef (n = 135 samples Methods n/a Involved in the study

ChIP-seq
Flow cytometry MRI-based neuroimaging tissue-covered skeleton (< 2 cm3) were also collected from each colony using a hammer and chisel, and were preserved in 80 % ethanol for further molecular analysis. All colonies were genotyped using microsatellite markers (as described in [1]) to identify clone mates (i.e., genetically identical colonies produced through asexual fragmentation). From these surveys, 135 colonies were retained to assess the relative contribution from host genotype (6 genotypes) and environment (3 reef habitats) to microbiome structure. The temperature and light intensity were monitored over a one-month period (i.e., from August 23 to September 26, 2019) to assess the environmental differences between the three surveyed reef habitats.
The 16S rRNA gene amplicon library was prepared by C.E.D. and sequenced at the KAUST BioScience Core Laboratory on the Illumina HiSeq 2500 platform using the rapid-run mode with 2 x 250 bp overlapping paired-end reads with a 10 % phiX control. Paired-end sequencing reads were processed by C.E.D.  No data were excluded from the analyses.
PCR amplifications of the 16S rRNA gene were run in triplicate per sample, amplification success was verified on a 1 % agarose gel and successful triplicate reactions were pooled. PCR products were subsequently quantified using a Qubit dsDNA HS Kit (Invitrogen, Carlsbad CA, USA) and run on the Bioanalyzer 2100 (Agilent Technologies, Santa Clara CA, USA) to confirm amplicon length and purity.
Randomization was not relevant in our study as we were specifically assessing the bacterial community composition differences in six genotypes of the fire coral Millepora platyphylla with clones occurring naturally in three contrasting reef habitats.
Blinding was not relevant in our study as we were specifically assessing the bacterial community composition differences in six genotypes of the fire coral Millepora platyphylla with clones occurring naturally in three contrasting reef habitats.
The field work was done by boat when swell surge was low to allow coral sampling using SCUBA and snorkeling at low depth (< 1 m to 13 m). The sampling also occurred in the dry season when rainfall is low.
Fire coral colonies (Millepora platyphylla) were collected from three adjacent reef habitats located at Papetoai on the north shore of Moorea Island, French Polynesia (17.5267 S, 149.8348 W): the mid slope (13 m depth), upper slope (6 m depth) and back reef (< 1 m depth).
Samples of coral tissue-covered skeleton were exported from Moorea in French Polynesia to Perpignan in France (CITES -FR1298700028-E). Extracted DNA were brought to KAUST, Saudi Arabia.
There were no disturbances as we only collected a small fragment of the coral colony (< 2cm3).