Re-examining the relationships among families of Pectinoidea and molecular machinery of the visual cycle in molluscs

Abstract

<p>Biologists have long used eyes as a model to study the evolution of complex traits, and the growing number of molecular datasets have allowed new insights into the conservation and function of the molecular machinery underlying photosensitivity across animals. However, molluscs, the second largest animal phylum, have been largely ignored in these studies despite possessing morphologically diverse eyes. Here, I investigated three levels of biological organization in molluscan eyes: organs, pathways, and proteins. I began by generating a robust phylogenetic hypothesis of the bivalve superfamily Pectinoidea to determine when the mirror-type eye of scallops (Pectinidae) originated. From this study, I propose a novel topology in which Propeamussiidae is non-monophyletic, where a subset of species resolve as sister to the Pectinidae and second species group, including the type species Propeamussium dalli, are sister to the Spondylidae. This relationship suggests a single origin of eyes prior to Pectinoidea with multiple instances of loss throughout the superfamily. Next, I assembled and searched available molluscan transcriptomes using known visual cycle proteins of vertebrates and insects to expand on the current hypothesis of the retinoid visual cycle in molluscs. The search results were then divided by protein family and used to develop protein phylogenies with vertebrate and insect anchor sequences to suggest putative homology of function of molluscan blast results. From this study, I was able to propose a new, light independent retinoid visual cycle for molluscs that includes proteins homologous to both vertebrates and insects. Interestingly, I was unable to identify a homologous isomerohydrolase or retinyl storage pathway in molluscs, thus future studies will require experimental work to determine a possible lineage-specific pathway. Finally, I investigated the relationship between genotype and phenotype in opsins by mutating targeted amino acids of interest and examining how these alternations affected opsin function when expressed in vitro. Comparing two closely related scallop retinochromes, I identified and mutated sites lining the binding pocket of the retinochrome that may interact with the chromophore, but were not conserved between the two retinochrome samples. These experiments showed that sites lining the binding pocket may be responsible for fine tuning the spectra of opsin proteins and that size and class of amino acid side chains may be responsible for changes in the λmax. This study highlights the potential of retinochrome as a model in mapping the relationship of genotype and phenotype in opsins which can be used to build more in depth and accurate prediction models of photopigments.</p>