Genetic diversity and connectivity of chemosynthetic cold seep mussels from the U.S. Atlantic margin

Abstract

Background: Deep-sea mussels in the subfamily Bathymodiolinae have unique adaptations to colonize hydrothermal-vent and cold-seep environments throughout the world ocean. These invertebrates function as important ecosystem engineers, creating heterogeneous habitat and promoting biodiversity in the deep sea. Despite their ecological significance, efforts to assess the diversity and connectivity of this group are extremely limited. Here, we present the first genomic-scale diversity assessments of the recently discovered bathymodioline cold-seep communities along the U.S. Atlantic margin, dominated by Gigantidas childressi and Bathymodiolus heckerae.

Results: A Restriction-site Associated DNA Sequencing (RADSeq) approach was used on 177 bathymodiolines to examine genetic diversity and population structure within and between seep sites. Assessments of genetic differentiation using single-nucleotide polymorphism (SNP) data revealed high gene flow among sites, with the shallower and more northern sites serving as source populations for deeper occurring G. childressi. No evidence was found for genetic diversification across depth in G. childressi, likely due to their high dispersal capabilities. Kinship analyses indicated a high degree of relatedness among individuals, and at least 10-20% of local recruits within a particular site. We also discovered candidate adaptive loci in G. childressi and B. heckerae that suggest differences in developmental processes and depth-related and metabolic adaptations to chemosynthetic environments.

Conclusions: These results highlight putative source communities for an important ecosystem engineer in the deep sea that may be considered in future conservation efforts. Our results also provide clues into species-specific adaptations that enable survival and potential speciation within chemosynthetic ecosystems.

Keywords: Adaptation; Bathymodioline; Bathymodiolus heckerae; Chemosynthetic; Genomics; Gigantidas childressi; RADseq; Recruitment; SNP.

Fig. 1

Map of sampling locations along the U.S. Atlantic Margin. Norfolk Canyon Seep= NCS, Baltimore Canyon Seep = BCS, Chincoteague Seep = CTS. Photo credit: DeepSearch

Fig. 2

Principal components analysis (PCA) plot representing genetic differentiation among G. childressi samples collected from Norfolk Canyon Seep (NCS, red circles), Chincoteague Seep (CTS, orange triangles) and Baltimore Canyon Seep (BCS, blue squares)

Fig. 3

Average probability of membership graph for G. childressi (n = 81) collected from seeps at Norfolk canyon (NCS), Chincoteague (CTS) and Baltimore Canyon (BCS). K = 2 clusters (or ancestral populations) as identified by STRUCTURE

Fig. 4

Principal Components Analysis (PCA) plot representing genetic differentiation among B. heckerae samples from six discrete collections along Blake Ridge Seep (BRS) (1, red circle: mussel pot B6 [n = 33], 2, orange triangle: remotely operated vehicle grab- ROV_G [n = 1], 3, filled blue square: slurp- SBlue_02 [n = 1], 4, teal cross: mussel pot B1 [n = 17], 5, blue square: mussel pot B2 [n = 10], 6, pink asterisk: mussel pot B4 [n = 25])

Fig. 5

A: Manhattan plot revealing outlier SNPs (adjusted p < 0.05, above red line) with a minimum allele frequency (mAF) > 0.05 (default). B: A score plot displaying the projection of each sample onto the principal components (PC) of the PCA conducted in pcadapt. Samples are color coded by site for G. childressi (Baltimore canyon seep (BCS), pink); Norfolk Canyon seep (NCS), blue; and Chincoteague seep (CTS), green) and/or species B. heckerae (purple)

Fig. 6

REViGO treemap summary of Gene Ontology information for biological processes associated with the outlier SNPs in the two bathymodioline species. Color blocks represent grouped terms based on common overlapping ontology