Nuclear DNA restriction site polymorphisms and the phylogeny and population structure of an intertidal snail species complex (Littorina)

Proteomic similarity of the Littorinid snails in the evolutionary context

BACKGROUND

The introduction of DNA-based molecular markers made a revolution in biological systematics. However, in cases of very recent divergence events, the neutral divergence may be too slow, and the analysis of adaptive part of the genome is more informative to reconstruct the recent evolutionary history of young species. The advantage of proteomics is its ability to reflect the biochemical machinery of life. It may help both to identify rapidly evolving genes and to interpret their functions.

METHODS

Here we applied a comparative gel-based proteomic analysis to several species from the gastropod family Littorinidae. Proteomes were clustered to assess differences related to species, geographic location, sex and body part, using data on presence/absence of proteins in samples and data on protein occurrence frequency in samples of different species. Cluster support was assessed using multiscale bootstrap resampling and the stability of clustering-using cluster-wise index of cluster stability. Taxon-specific protein markers were derived using IndVal method. Proteomic trees were compared to consensus phylogenetic tree (based on neutral genetic markers) using estimates of the Robinson-Foulds distance, the Fowlkes-Mallows index and cophenetic correlation.

RESULTS

Overall, the DNA-based phylogenetic tree and the proteomic similarity tree had consistent topologies. Further, we observed some interesting deviations of the proteomic littorinid tree from the neutral expectations. (1) There were signs of molecular parallelism in two Littoraria species that phylogenetically are quite distant, but live in similar habitats. (2) Proteome divergence was unexpectedly high between very closely related Littorina fabalis and L. obtusata, possibly reflecting their ecology-driven divergence. (3) Conservative house-keeping proteins were usually identified as markers for cryptic species groups ("saxatilis" and "obtusata" groups in the Littorina genus) and for genera (Littoraria and Echinolittorina species pairs), while metabolic enzymes and stress-related proteins (both potentially adaptively important) were often identified as markers supporting species branches. (4) In all five Littorina species British populations were separated from the European mainland populations, possibly reflecting their recent phylogeographic history. Altogether our study shows that proteomic data, when interpreted in the context of DNA-based phylogeny, can bring additional information on the evolutionary history of species.

Species and gene divergence in Littorina snails detected by array comparative genomic hybridization

Background

Array comparative genomic hybridization (aCGH) is commonly used to screen different types of genetic variation in humans and model species. Here, we performed aCGH using an oligonucleotide gene-expression array for a non-model species, the intertidal snail Littorina saxatilis. First, we tested what types of genetic variation can be detected by this method using direct re-sequencing and comparison to the Littorina genome draft. Secondly, we performed a genome-wide comparison of four closely related Littorina species: L. fabalis, L. compressa, L. arcana and L. saxatilis and of populations of L. saxatilis found in Spain, Britain and Sweden. Finally, we tested whether we could identify genetic variation underlying “Crab” and “Wave” ecotypes of L. saxatilis.

Results

We could reliably detect copy number variations, deletions and high sequence divergence (i.e. above 3%), but not single nucleotide polymorphisms. The overall hybridization pattern and number of significantly diverged genes were in close agreement with earlier phylogenetic reconstructions based on single genes. The trichotomy of L. arcana, L. compressa and L. saxatilis could not be resolved and we argue that these divergence events have occurred recently and very close in time. We found evidence for high levels of segmental duplication in the Littorina genome (10% of the transcripts represented on the array and up to 23% of the analyzed genomic fragments); duplicated genes and regions were mostly the same in all analyzed species. Finally, this method discriminated geographically distant populations of L. saxatilis, but we did not detect any significant genome divergence associated with ecotypes of L. saxatilis.

Conclusions

The present study provides new information on the sensitivity and the potential use of oligonucleotide arrays for genotyping of non-model organisms. Applying this method to Littorina species yields insights into genome evolution following the recent species radiation and supports earlier single-gene based phylogenies. Genetic differentiation of L. saxatilis ecotypes was not detected in this study, despite pronounced innate phenotypic differences. The reason may be that these differences are due to single-nucleotide polymorphisms.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-687) contains supplementary material, which is available to authorized users.

Micro-spatial distribution of two sibling periwinkle species across the intertidal indicates hybrdization

Populations of periwinkles Littorina saxatilis (Olivi 1792) and L. arcana Hannaford Ellis, 1978 are well suited for microevolutionary studies, being at the same time closely related and intraspecifically diverse. The divergence between these two sibling species, sympatric over large parts of their distribution areas, is small, the only morphological difference being the pallial gland complex structure in females. Molecular identification is possible with the use of a RAPD nuclear marker (cloned A2.8 DNA fragment) typical for L. arcana. However, in some individuals from sympatric populations molecular and morphological criteria suggest conflicting species affiliation, which may be explained either by hybridization or by shared ancestral polymorphism. We tested the hybridization hypotheses examining the micro-spatial distribution of these two species across the intertidal zone in two distant sites at the Barents Sea. We found that (a) the frequency of putative hybrids in sympatric populations was proportional to the frequency of L. arcana; (b) L. saxatilis bearing A2.8 DNA fragment were almost absent in the lower part of the intertidal zone, where L. arcana was absent too; (c) there was a close positive correlation between the distribution of potential parent molluscs and putative hybrids. Moreover, logistic regression models showed a good agreement between the distribution of putative hybrid frequencies and that of parental species frequencies. All our observations taken together support the hypothesis of hybridization between L. saxatilis and L. arcana. Elucidating the mechanisms that support the species status of these sympatric populations is necessary.

Phylogeographic analysis reveals a deep lineage split within North Atlantic Littorina saxatilis

Phylogeographic studies provide critical insight into the evolutionary histories of model organisms; yet, to date, range-wide data are lacking for the rough periwinkle Littorina saxatilis, a classic example of marine sympatric speciation. Here, we use mitochondrial DNA (mtDNA) sequence data to demonstrate that L. saxatilis is not monophyletic for this marker, but is composed of two distinct mtDNA lineages (I and II) that are shared with sister species Littorina arcana and Littorina compressa. Bayesian coalescent dating and phylogeographic patterns indicate that both L. saxatilis lineages originated in the eastern North Atlantic, around the British Isles, at approximately 0.64 Ma. Both lineages are now distributed broadly across the eastern, central and western North Atlantic, and show strong phylogeographic structure among regions. The Iberian Peninsula is genetically distinct, suggesting prolonged isolation from northeastern North Atlantic populations. Western North Atlantic populations of L. saxatilis lineages I and II predate the last glacial maximum and have been isolated from eastern North Atlantic populations since that time. This identification of two distinct, broadly distributed mtDNA lineages further complicates observed patterns of repeated incipient ecological speciation in L. saxatilis, because the sympatric origins of distinct ecotype pairs on eastern North Atlantic shores may be confounded by admixture of divergent lineages.