Depth-structured lineages in the coral Stylophora pistillata of the Northern Red Sea

Abstract

Coral reefs are biodiversity hotspots, where new species continue to be discovered. Stylophora pistillata, a depth-generalist coral, is widely distributed throughout the Indo-Pacific and has long been considered the poster child for phenotypic plasticity. It occupies a wide range of reef habitats and exhibits a myriad of gross morphologies. Here, we used reduced representation genome sequencing (nextRAD) to assess the genetic structure of adults and recruits of S. pistillata across shallow and mesophotic populations in the northern Red Sea (Gulf of Aqaba). Across analytical approaches, we observed a complex genetic structure with at least four genetically divergent lineages occurring sympatrically with little to no admixture and structured by depth. Morphological and physiological differences previously documented suggest that the long-considered ecological opportunism of S. pistillata in the Red Sea may, in fact, have a genetic basis. Assessment of both adult colonies and recruits within each of the lineages also revealed the prevalence of local recruitment and genetic structuring across the eight-kilometer section of the Israeli Red Sea coastline. Overall, the observed patterns confirm the presence of undescribed diversity within this model organism for coral physiology and corroborate a broader pattern of extensive undescribed diversity within scleractinian corals.

Fig. 1. Sampling map.

Sampling sites of adult fragments and juvenile spats Stylophora pistillata along the shoreline of Eilat, Red Sea. Sites are indicated by yellow dots, with name and depth of collection. Red line contours indicate isobaths of 20 m depth intervals. For further images, Japanese gardens_45m is referred as Natural reserve_45 (NAT45).

Fig. 2. Overall genetic structure of Stylophora pistillata.

Neighbor joining tree (based on 281 individuals and 38,836 SNPs) shows at least four major lineages and structure analysis (based on five replicate datasets with random SNPs separated by at least 2500 bp, each containing 281 individuals and 7185 SNPs) sowing K from three to 10. Colors indicate distinct genetic clusters, with each individual represented by a vertical bar with different colors indicating the relative proportion of each genetic cluster. Location shows original sites where samples were collected, with different colors and shapes indicating sampling depth and life stage, respectively.

Fig. 3. Principal component analysis for red and yellow clusters.

Principal Component Analysis (PCA) based on 24,151 SNPs for the Red cluster and 29,617 SNPs for the Yellow cluster of all sampled individuals (left; 94 and 100 individuals for the Red and Yellow clusters, respectively) and only adults (right; 56 and 50 individuals for the Red and Yellow clusters, respectively) for the two major genetic lineages, Red and Yellow. Populations (pop) indicates site and depth of sampling. KIS Kissosky, KAT Katza, KPO Katza polluted, NAT Natural Reserve, IUI IUI, PRI Princess. Numbers 10 and 45 indicates samples collected from 7 to 10 and at 45 m depth, respectively. Life stage indicates adults (AD) and recruits (SP). Percentages of variance explained by the illustrated PCs are shown.

Fig. 4. Genetic structure for Red and Yellow clusters.

Genetic structure performed with the software ADMIXTURE of the Red (93 individuals and 24,151 SNPs) and Yellow (100 individuals and 29,617 SNPs) clusters for K equal two and three. Each individual is represented by a vertical bar with different colors indicating the relative proportion of each genetic cluster. Populations (pop) indicates site, depth of sampling and life stage. KIS Kissosky, KAT Katza, KPO Katza polluted, NAT Natural Reserve, IUI IUI, PRI Princess. Numbers 10 and at 45 indicates samples collected from 7 to 10 and 45 m depth, respectively. AD adults and SP recruits.

Fig. 5. Spat count per m² at different locations and periods.

Spat counts expressed as mean ± SEM Standard Error of the Mean) (n = 4–12) per location and sampling period. a Comparison between locations. Different lowercase letters indicate significantly different mean values between locations for each period (p < 0.05). b Comparison between periods. Different lowercase/ uppercase letters indicate significantly different mean values between months in 2017 and 2018 respectively (p < 0.05). Asterisks indicate significantly different mean values between measurements in 2017 and 2018 for each month (*p < 0.05, **p < 0.01, ***p < 0.001). Statistical significance was determined by a three-way ANOVA, followed by pairwise comparisons with FDR correction.