Gene flow between Atlantic and Pacific Ocean basins in three lineages of deep-sea clams (Bivalvia: Vesicomyidae: Pliocardiinae) and subsequent limited gene flow within the Atlantic

Abstract Pliocardiin (vesicomyid) clams rely on microbial symbionts for nutrition and are obligate inhabitants of deep-sea chemosynthetic ecosystems. Unlike many other invertebrate hosts of chemosynthetic microbes, pliocardiin clams are found in every ocean in a variety of reducing habitats, including hydrothermal vents, cold seeps, organic falls and deep-sea fans. The global distribution of pliocardiin clams suggests historical gene flow between ocean basins. We focus on 3 pliocardiin genera— ‘Pliocardia’ I, Calyptogena and Abyssogena —each of which has a pair of sister clades in the Atlantic and Pacific. Our work tests the hypothesis that historical gene flow between the Atlantic and Pacific Oceans within these genera was interrupted by the closure of the Panamanian seaway and tests whether isolation between the ocean basins is the result of vicariance or past colonization. These questions are investigated in the context of fossil evidence, biogeography and phylogenetics. This study revealed a set of substitution rates consistent with other invertebrate studies ( μ =0.8%/My/lineage), and a set consistent with much lower rates often attributed to deep-sea organisms ( μ =0.3%/My/lineage). Among the Pacific/Atlantic sister pairs, ‘ Pliocardia’ I COI divergence per lineage is intermediate (2.5%), Calyptogena is the highest (6.1%) and Abyssogena the lowest (0.8%). The substitution rates suggest that ‘ Pliocardia’ I and Calyptogena have histories of at least 2.8 My in the Atlantic, with Calyptogena likely older. The slower rate, however, is inconsistent with both the maximum age of the family and several well studied fossils: leaving the faster rate preferred. With the faster rate, the Abyssogena southwardae clade diverged from its Pacific sister clade around 1 Mya, which likely post-dates the closure of the Isthmus of Panama and the opening of the Bering Strait. In light of this recent divergence, we test the previously proposed hypothesis that there is a high level of ongoing gene flow between Atlantic populations of A. southwardae. A. southwardae has colonized a broad geographic range of seep sites including the West Florida Escarpment, the Barbados Accretionary Prism, the Lobes of Congo, and the Mid-Atlantic Ridge north and south of the Romanche Transform Fault. Coalescent methods detect gene flow between Barbados and the Mid-Atlantic ridge; and between the West Florida Escarpment and the Lobes of Congo. All other comparisons failed to detect gene flow, contrary to prevailing interpretations of connectivity across the entire Atlantic Basin.