Integrative and Comparative Biology Integrative and Comparative Biology, volume 52, number 1, pp. 173–180 doi:10.1093/icb/ics083 Society for Integrative and Comparative Biology SYMPOSIUM Patterns of Nuclear Genetic Variation in the Poecilogonous Polychaete Streblospio benedicti Downloaded from https://academic.oup.com/icb/article-abstract/52/1/173/742368 by New York University user on 29 August 2019 Matthew V. Rockman1 Department of Biology and Center for Genomics & Systems Biology, New York University, 12 Waverly Place, New York, NY 10003, USA From the symposium ‘‘Poecilogony as a Window on Larval Evolution: Polymorphism of Developmental Mode within Marine Invertebrate Species’’ presented at the annual meeting of the Society for Integrative and Comparative Biology, January 3–7, 2012 at Charleston, South Carolina. 1E-mail: [email protected] Synopsis The evolution of marine larvae is replete with transitions in trophic mode, but little is known about how, at the genetic level, these transitions are achieved. Basic parameters, including the number of underlying loci, their molecular characteristics, and the population-genetic processes that drive transitions remain unknown. Streblospio benedicti, an abundant benthic polychaete with heritable poecilogony, provides a unique genetically tractable system for addressing these issues. Individuals of S. benedicti vary in diverse aspects of development. Some females produce small, planktotrophic larvae, whereas others produce large, yolky larvae capable of settling without feeding. Here, I present estimates of basic features of nuclear genetic variation in S. benedicti to lay the foundations for subsequent efforts to understand the genetic basis of poecilogony. Sequence of 20 kb of random nuclear DNA indicates that the nucleotide composition, at 62.1% A þ T, is typical of lophotrochozoan genomes. Population-genetic data, acquired by sequencing two loci (2500 bp) in multiple animals of each developmental morph, indicate that the morphs exhibit very little differentiation at random loci. Nucleotide heterozygosity (^)is0.5–1% per site, and linkage disequilibrium decays within a few kilobases (^ 3 Â 10À3 per site). These data suggest that genetic mapping by association will require a high density of markers, but linkage mapping and identification of regions of elevated inter-morph differentiation hold great promise. Introduction have necessarily passed through a stage of develop- The evolution of marine larvae is characterized by mental polymorphism, with heritable variation in de- multiple independent transitions from obligate velopmental mode due to segregating alleles with planktotrophy, with a long phase of larval dispersal, effects on life history, only one species is known to to obligate lecithotrophy, with dispersal abbreviated exhibit such polymorphism in contemporary popu- or abandoned (Strathmann 1978; Wray and Raff lations: the spionid polychaete Streblospio benedicti. 1991; Smith et al. 2007). Many species exhibit inter- Since the discovery of developmental variation mediate phenotypic states, including facultative within and among S. benedicti populations (Levin planktotrophy (Allen and Pernet 2007) and environ- 1984), this species has become a paragon of poeci- mentally plastic development (Rice and Rice 2009). logony and the subject of an extensive literature on Evolutionary transitions from one fixed mode of development and ecology. Because trophic mode in development to another require, in addition, S. benedicti is now understood to include facultative intermediate population-genetic states, as alleles planktotrophy (Pernet and McArthur 2006), I refer that contribute to lecithotrophic development accu- to the two ends of the developmental spectrum as mulate in genetically variable populations. Although the ‘‘small-egg’’ and ‘‘large-egg’’ morphs, with the all populations with derived lecithotrophic larvae former encompassing the obligate planktotrophs Advanced Access publication June 1, 2012 ß The Author 2012. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved. For permissions please email: [email protected]. 174 M. V. Rockman and the latter including both facultative plankto- The experimental tractability of a species with trophs and obligate lecithotrophs. heritable poecilogony holds out promise for the The two developmental morphs differ in a large identification and molecular characterization of the and diverse array of features, starting with the ova, causal alleles that underlie life-history transitions. which differ in volume by a factor of six (Levin and The distribution of these causal alleles in natural Bridges 1994). In the earliest stages of embryogenesis, populations and their temporal and spatial fluctua- the proportion of the zygote segregated into the tions could, in principle, be directly measured. polar lobe differs between morphs, as do the relative Such studies would provide exceptional insight into Downloaded from https://academic.oup.com/icb/article-abstract/52/1/173/742368 by New York University user on 29 August 2019 sizes of the blastomeres and micromeres (McCain the intersection of development, evolution and 2008). Later in development, the morphs exhibit ecology. differences in the timing and pattern of development Here, I provide a preliminary account of several of the gut and coelem (Gibson et al. 2010; Pernet aspects of the nuclear genome and its patterns of and McHugh 2010). Several traits exhibit discrete variation in S. benedicti as a first step toward creating differences: larvae from small eggs produce long, tools to discover the causal alleles underlying the conspicuous larval bristles that are absent from diverse differences between planktotrophic and larvae from large eggs (Levin 1984), and they also lecithotrophic modes of larval development. I have several cell types, including bacillary cells and describe four attributes—nucleotide composition, some mucous cells, that large-egg larvae fail to heterozygosity, linkage disequilibrium, and popula- generate (Gibson et al. 2010). Phenotypic differences tion differentiation—that will be valuable for future between larval morphs are reflected in differences in efforts to map genes by linkage and association. transcriptome complexity (Marsh and Fielman 2005). Animals of the small-egg morph are released from Materials and methods the maternal brood pouches with fewer setigers than Adult S. benedicti were collected from Newark Bay, is true of the large-egg morphs and they spend New Jersey, and Long Beach, California. I collected longer in the water prior to settlement (Levin and animals from intertidal mud at the north end of Bridges 1994). Later still, animals of the two morphs Bayonne Park, Bayonne, New Jersey (4084101100N, differ in age and size at maturity and adults differ in 7480604800W). Animals from Long Beach were a brood size and in the investments of carbon and generous gift from Bruno Pernet. From each locality, nitrogen in each brood (reviewed by Levin and broods from at least six females were observed and in Bridges 1994). every case the Bayonne females produced small feed- The heritable nature of peocilogony in S. benedicti ing larvae with long larval bristles and the Long was established through a series of meticulous and Beach females produced large larvae, lacking larval creative studies (reviewed by Levin and Bridges bristles, and capable of settling without feeding. 1994). Experimental crosses confirmed the substan- These animals are facultative planktotrophs and tially genetic basis for the alternative larval types larvae fed Dunaliella salina ate it. The Bayonne (Levin et al. 1991). Importantly, intermediate population has been observed for 3.5 years and the forms, putative hybrids between the discrete larval phenotype has been consistent. Californian morphs described earlier, occur in natural popula- populations consist exclusively of the large-egg devel- tions (Levin and Huggett 1990), and their attributes, opmental phenotype (Levin 1984). Streblospio along with those of experimental intercross progeny, benedicti on the West Coast of North America is demonstrate that the genetic basis of developmental likely a recent introduction from East-Coast mode in S. benedicti is complex and multifarious, populations (Schulze et al. 2000). with different genetic bases for different aspects Individual worms were maintained in artificial of the larval syndromes. Moreover, molecular seawater (Instant OceanÕ). After worms were iso- population-genetic data from the mitochondrial lated from all mud for at least 24 h to purge their genome indicate that the two developmental modes guts and were washed several times with artificial share alleles and are not genetically differentiated, at seawater, they were frozen at À808C. DNA was iso- least at this single locus (Schulze et al. 2000; Mahon lated from each worm either by phenol–chloroform et al. 2009). These findings suggest that the molecu- extraction or by Qiagen’s DNeasy kit. lar genetic basis for larval evolution in S. benedicti To generate genomic clones for sequencing, I must involve multiple unlinked loci that are differ- digested 400 ng of genomic DNA isolated from a entiated between the larval morphs on a genetic pool of eight Bayonne individuals with the restriction background homogenized by gene flow. enzymes EcoRI and HindIII (NEB) for 8 h at 378C Genetic variation in S. benedicti 175 and recovered digested DNA in several size ranges by phase