The seashore where the land and sea meet is a very unique environment. It is challenging for marine animals to survive in these habitats because of harsh conditions such as strong waves and exposure to atmosphere. Nonetheless, a variety of organisms are present and they form seashore ecosystems.
Various factors are involved in the evolution of marine life dwelling in these habitats. In case of animal species without the larval stage such as sea slaters (
Representing a large variation between species, cytochrome oxidase subunit I (COI) sequences in the mitochondrial genome are popular as DNA barcoding markers in the identification of animal species (Hebert et al., 2003). COI regions are also suitable for studying intraspecific genetic variation since COI sequences have considerable genetic variation within species (Marosi et al., 2013).
In this study, we compared the population genetic structure of two dominant seashore-dwelling species from Korea, differing in the length of the larval period: periwinkle
Up to forty-six periwinkles and thirty-seven acorn barnacles were collected from seven and four locations in Korea, respectively (Table 1, Fig. 1). Specimens were fixed in 80% ethanol solution at the collection site, and were identified using a stereomicroscope in the laboratory. Parts of the body of each specimen were removed for DNA extraction. DNAs were extracted with DNeasy Tissue Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s protocol.
Summary statistics of genetic variation of Littorina brevicula and Fistulobalanus albicostatus in this study
Part of the COI region was amplified using primers LCO 1490 and HCO2198 (Folmer et al., 1994). Each polymerase chain reaction (PCR) mixture (total volume, 25 μL) was composed of 2.5 μL of PCR buffer (10×), 2.0 μL of dNTP mixture (10 mM), 1.5 μL of MgCl2 solution (25 mM), 1 μL of each primer solution (10 μM), 0.5 μL of
Sequences were saved in FASTA file format. Then, multiple alignments were performed by using CLUSTAL_X (Thompson et al., 1997). The definition of haplotypes was carried out by using DnaSP 5.10.01 (Librado and Rozas, 2009). Genetic polymorphism was investigated with ARLEQUIN 3.11 (Excoffier et al., 2005) by calculating haplotype diversity (gene diversity) and nucleotide diversity per site. Haplotype networks were drawn with NETWORK 22.214.171.124 (Fluxus Technology, http://www.fluxus-engineering.com) by using the median joining algorithm (Bandelt et al., 1999). Recent demographic histories were inferred by performing mismatch distribution (MMD) analysis (Harpending, 1994) and Tajima’s D test (Tajima, 1989) with ARLEQUIN software. Analysis of molecular variance (AMOVA) was performed to examine the hierarchical genetic structure by using ARLEQUIN.
The sequencing results for 46 individuals of periwinkle revealed 8 haplotypes (GenBank accession number: KU977411-KU977418) (658 bp). The two haplotypes with the highest frequencies appeared in almost all locations (Fig. 2A). When all individuals were pooled, haplotype diversity and nucleotide diversity were 0.5961 and 0.0012, respectively. SEOSAN was monomorphic, and therefore, it showed lowest genetic diversity, but AYAJIN showed the highest genetic polymorphism in terms of both haplotype diversity and nucleotide diversity (Table 1).
In case of acorn barnacle, 28 haplotypes were defined in 37 individuals (GeneBank accession number: KU977383-KU977410) (672 bp). The haplotype network of barnacle is more complicated than that of periwinkle (Fig. 2B). Although a few haplotypes with the highest frequencies appeared in most of the locations, 25 out of the 28 haplotypes were private, i.e., they appeared in only one location. It is not likely that the genetic relationship shows an association with geography. Acorn barnacle populations were genetically more diverse than populations of periwinkle. Haplotype diversity and nucleotide diversity of the pooled population of barnacle were 0.9625 and 0.0070, respectively (Table 1).
From the results of MMD analysis, the observed patterns of MMD were not significantly different from the expected patterns under a sudden expansion model in both species (Table 1). This suggests that these two species may have experienced recent population expansion. One remarkable difference between the two species is that acorn barnacle populations showed a bimodal pattern of MMD unlike periwinkle populations which showed a unimodal pattern (Fig. 3). The results of Tajima’s D test indicated a slightly different story. The D value for each local population of both species was not significantly negative. But the pooled population of these two species showed significant negative D values, which suggests they may have experienced expansion of population in view of the whole population. It is, therefore, likely that the results of the two population demographic analyses are consistent on the whole.
[Fig. 3.] Mismatch distribution plots of Littorina brevicula (A) and Fistulobalanus albicostatus (B). Exp and Obs denote an expected pattern of mismatch distribution under a sudden expansion model and an observed pattern of mismatch distribution, respectively.
According to the results of AMOVA, the
Results of analysis of molecular variance (AMOVA) for Littorina brevicula
Results of analysis of molecular variance (AMOVA) for Fistulobalanus albicostatus
On the other hand, these two species were significantly different in terms of the other genetic properties. The most remarkable property was the difference in genetic diversity. Acorn barnacle populations showed much more diversity than periwinkle populations. Moreover, individuals constituting each local population of barnacle were not closely related. The next difference between these two species was detailed histories of populations. In MMD analyses, a bimodal pattern was observed in acorn barnacle populations, but a unimodal pattern was observed in periwinkle populations.
Such differences between these two species could be due to historical as well as recurrent processes. Bimodal patterns in MMD plots usually suggest that past fragmentation was followed by secondary contact or genetic admixture (Strasburg et al., 2007). Therefore, the present acorn barnacle populations may have been formed by secondary contact of genetically diverged populations that had been separated by geological changes probably during the Pleistocene.
The second but a more important factor shaping the different population genetic structure was disparity in gene flow rates between these two species. In general, the longer are the larval periods, the higher is the possibility of recruitment of individuals from the common gene pool to a variety of local seashore habitats. Resultantly, it may weaken local adaptation and lower genetic divergence between local populations (Doherty et al., 1995; Johannesson, 2003). Actually, the larval period of periwinkles is 4-7 weeks, which is longer than the 6-23 day larval period of
This study suggests that the genetic structure of coastal marine animal species could be affected by a variety of factors such as historical events or recurrent processes. Further studies on seashore-dwelling species are expected to be useful in understanding the evolution of the coastal ecosystem around Korean waters.