Plio-Pleistocene sea level and temperature fluctuations in the northwestern Pacific promoted speciation in the globally-distributed flathead mullet Mugil cephalus

Abstract

Background: The study of speciation in the marine realm is challenging because of the apparent absence of physical barriers to dispersal, which are one of the main drivers of genetic diversity. Although phylogeographic studies using mitochondrial DNA (mtDNA) information often reveal significant genetic heterogeneity within marine species, the evolutionary significance of such diversity is difficult to interpret with these markers. In the northwestern (NW) Pacific, several studies have emphasised the potential importance of sea-level regression during the most recent glaciations as a driver of genetic diversity in marine species. These studies have failed, however, to determine whether the period of isolation was long enough for divergence to attain speciation. Among these marine species, the cosmopolitan estuarine-dependent fish Mugil cephalus represents an interesting case study. Several divergent allopatric mtDNA lineages have been described in this species worldwide, and three occur in sympatry in the NW Pacific.

Results: Ten nuclear microsatellites were surveyed to estimate the level of genetic isolation of these lineages and determine the role of sea-level fluctuation in the evolution of NW Pacific M. cephalus. Three cryptic species of M. cephalus were identified within this region (NWP1, 2 and 3) using an assignment test on the microsatellite data. Each species corresponds with one of the three mtDNA lineages in the COI phylogenetic tree. NWP3 is the most divergent species, with a distribution range that suggests tropical affinities, while NWP1, with a northward distribution from Taiwan to Russia, is a temperate species. NWP2 is distributed along the warm Kuroshio Current. The divergence of NWP1 from NWP2 dates back to the Pleistocene epoch and probably corresponds to the separation of the Japan and China Seas when sea levels dropped. Despite their subsequent range expansion since this period of glaciation, no gene flow was observed among these three lineages, indicating that speciation has been achieved.

Conclusions: This study successfully identified three cryptic species in M. cephalus inhabiting the NW Pacific, using a combination of microsatellites and mitochondrial genetic markers. The current genetic architecture of the M. cephalus species complex in the NW Pacific is the result of a complex interaction of contemporary processes and historical events. Sea level and temperature fluctuations during Plio-Pleistocene epochs probably played a major role in creating the marine species diversity of the NW Pacific that is found today.

Figure 1

Sampling localitions of Mugil cephalus in the northwestern Pacific. The shaded zone in dark grey is the area of the continental shelves that was exposed during periods of low sea levels. Blue arrows correspond to currents present in the area, C. C. C.: China Coastal Current, S. C. C.: South China Current.

Figure 2

Phylogenetic trees of Mugil cephalus in the northwestern Pacific. Phylogenetic trees reconstructed from (A) 36 COI sequences haplotypes (HT1-HT36) of Mugil cephalus in the northwestern Pacific and its geographic distribution (for location names follow Table 1). The values above the branches are bootstrap values for the NJ, ML and MP analyses and the posterior probabilities for the Bayesian analysis.

Figure 3

Haplotype networks and mismatch distribution for three lineages of Mugil cephalus COI sequences. (A) Haplotype networks for three lineages of Mugil cephalus COI sequences. The lengths of the connecting lines are in relation to the number of mutations between haplotypes. Each circle represents a haplotype, with the diameter of the circle proportional to the number of sequences of that haplotype. The names of the major haplotypes for each cryptic species correspond to those listed in Table S1. (B) Mismatch distributions from the mtDNA COI sequences of M. cephalus from the ten sampling locations. Blue bar: observed distributions; Light blue bar: expected distributions from the sudden expansion model.

Figure 4

Correspondence between mtDNA lineages and nuclear Structure clusters. First row is the distribution of the 3 mtDNA lineages (green: lineage 1, blue: lineage 2, red: lineage 3) in 18 samples of Mugil cephalus in NW pacific. Second to forth row are the results of the assignment test using STRUCTURE [49] for M. cephalus microsatellite data. Each cluster (K) is designated by a different colour with vertical bars representing individuals and the proportion of a bar assigned to a single colour representing the posterior probability that an individual is assigned to that cluster. Assignment results are shown with K = 3, 4, 5 and 6.

Figure 5

The distribution of three cryptic species of Mugil cephalus in the northwestern Pacific. (A) the proportions of the three species in various locations; coloured lines indicate sea surface temperature in the NW Pacific in December, which is the period of reproductive migration of Mugil cephalus; (B) the distribution range of NWP1 and location of the China Coastal Current; (C) the distribution range of NWP2 and location of the Kuroshio Current; (D) the distribution range of NWP3 and location of the South China Sea Current.