Mitochondrial DNA evidence for rapid colonisation of ... - Springer Link

Marine Biology (1999) 134: 227±233

Ó Springer-Verlag 1999

D. Gopurenko á J. M. Hughes á C. P. Keenan

Mitochondrial DNA evidence for rapid colonisation of the Indo±West Paci®c by the mudcrab Scylla serrata

Received: 13 October 1998 / Accepted: 15 February 1999

Abstract Scylla serrata (ForskaÊl, 1775) is widely distributed throughout mangrove habitats of the Indo-West Paci®c (IWP) coastal waters. This study investigated the phylogeographic distribution of S. serrata mitochondrial DNA haplotypes sampled throughout the species range. Adults were sampled from three west Indian Ocean locations (N = 21), ®ve west Paci®c sites (N = 28) and three sites from northern and eastern Australia (N = 76). Temperature-gradient gel-electrophoresis and sequencing of 549 base pairs of a mtDNA gene (cytochrome oxidase 1) identi®ed 18 distinct haplotypes. Haplotypes cluster into two clades separated by '2% sequence-divergence. One clade is widespread throughout the IWP, the other is strictly con®ned to northern Australia. Genealogical assessment of sequenced haplotypes suggests that the historical spread of S. serrata throughout the IWP has occurred rapidly and recently ( 20)

Region

p

h

Intra-regional comparison Indian Ocean Paci®c Ocean Northern Australia

0.16 0.21 0.56

0.61 0.62 0.69

Inter-regional comparison Indian vs Paci®c Indian vs Northern Australia Paci®c vs Northern Australia

0.09 1.12 1.08

0.76 0.80 0.72

231

clades consist of distinct haplotype assemblages that di€er by '2% sequence-divergence (Table 4). This level of divergence suggests a coalescent ancestry between the two clades dating back '1 million years before present (Table 4). The demographic histories of these two clades appear to have proceeded along independent trajectories, resulting in markedly di€erent patterns of distribution and population genetic structure. We hypothesised that a vicariant separation of Indian from Paci®c Ocean populations, observed for several IWP marine species, would similarly have a€ected the population genetic structure of Scylla serrata. Results for S. serrata are contrary to this hypothesis, and instead suggest that radiation throughout the IWP occurred as a single rapid wave of expansion, possibly emanating from a west Paci®c origin during the late Pleistocene. For example, the mean corrected percent sequence-di€erence between the Indian and Paci®c regions is similar to uncorrected levels of di€erences seen within each of these two regions (Table 3). The network of haplotype relationships (Fig. 2) indicates that all lineages from Indian and Paci®c localities are nested within Clade 1 and are ultimately derived from a single common ancestor-currently widespread in the West Paci®c. The low level of sequence divergence within this clade indicates a recent origin for the expansion of S. serrata populations throughout the IWP. Assuming that our clock calibrations accurately re¯ect the cumulative mutation rate for mtDNA, then the estimated timing of genealogical coalescence for all Clade 1 haplotypes falls within the late Pleistocene (Table 4). Surprisingly, the presence of ®xed unique haplotypes in each of the populations peripheral to the west Paci®c core suggests that intervening gene ¯ow since the initial radiation has been insucient to overcome the e€ects of local random genetic drift. The

lack of e€ective gene ¯ow among populations appears to have been sucient to allow the development of distinct lineages in discrete populations through mutation and random lineage assortment (Slatkin 1977; Neigel and Avise 1986). Estimated total haplotype diversity for S. serrata is comparable to that reported for other widespread IWP species (Table 5). However, apart from that seen in northern Australia, individual populations of S. serrata are characterised by low haplotype diversity (Table 1). The observed low diversity may be an artifact of the small sample size, as 7 of the 11 locations sampled were represented by fewer than ten individuals (Table 1). However, comparison of haplotype-diversity estimates among the various taxa with comparable sample sizes in Table 5 suggest that average diversity within S. serrata populations is depauperate relative to that of other widespread taxa with planktonic dispersal. Periodic ¯uctuations of local e€ective population size resulting in population bottlenecks may have contributed to the reduction of haplotype diversity at these locations. These ¯uctuations may have allowed the chance ®xation of locally derived haplotypes, especially under conditions where there is a lack of sustained gene ¯ow from neighbouring populations (Neigel and Avise 1986). Contrary to the insular distribution of haplotypes among peripheral populations, the distribution of the Clade 1 most recent common ancestor (Haplotype A) is immense, spanning a distance of '6500 km from east Australia to Okinawa. This north±south distribution testi®es to extensive levels of gene ¯ow along the western margin of the Paci®c. Similar patterns of extensive west Paci®c gene ¯ow have been reported for several marine species, including a number of giant clams (Benzie and Williams 1997) and a species of sea urchin (Palumbi et al. 1997). Remarkably, these gene-¯ow patterns are contrary to present-day ocean currents, which suggests that episodes of population expansion and gene ¯ow along the west Paci®c margin may have been far greater during the Pleistocene than they are at present (Benzie and Williams 1997). We suspect that conditions during the Pleistocene may have facilitated episodes of Scylla serrata population-expansion, resulting in the widespread distribution of the species within the west Paci®c. Levels of gene ¯ow may have been sucient to provide both connectivity among populations via the rapid

Table 5 Scylla serrata. Comparison of mtDNA haplotypediversity estimates for seven species of invertebrates distributed throughout Indo±West Paci®c (IWP) and Paci®c locations, respectively [N(tot) and N(ave) total and average per site sample size,

respectively; h(tot) total-sample haplotype diversity; h(ave) average per-site haplotype diversity; h diversity estimates for all species (except B. latro) calculated using raw data derived directly from cited studies]

Table 4 Scylla serrata. Estimates of maximum coalescent events relative to MRCA 1 and 2 (MRCA most recent common ancestor; % seq maximum percent sequence-di€erence observed between all haplotypes and MRCA within a sample) Coalescent event

% seq

Years to coalescence

MRCA-1:Clade 1 MRCA-2:Clade 2 MRCA-1:MRCA-2

0.546 0.546 2.550

237 000 237 000 1 109 000

Species

Sites

N(tot)

h(tot)

N(ave)

h(ave)

Source

Birgus latro Echinometra oblonga Echinometra mathaei Echinometra sp.nov.A Echinometra sp.nov.C Linckia laevigata Scylla serrata

8 4 8 7 4 17 11

160 37 60 67 39 370 125

0.98 0.79 0.84 0.97 0.92 0.89 0.79

19 9 7.5 9.6 9.8 21.8 11.4

0.97) 0.62 0.57 0.82 0.81 0.74 0.13

Lavery et al. (1996)

IWP Paci®c Paci®c Paci®c Paci®c IWP IWP

Palumbi et al. (1997) Williams and Benzie (1997, 1998) Present study

232

spread of the MRCA haplotype, and to hinder random ®xation of alternative haplotypes among these locations. It may be surmised that fortuitous dispersal e€ected by Clade 1 crabs during the Pleistocene was sucient to found widely distributed IWP populations. Intervening levels of dispersal do not appear to have ensured genetic panmixia throughout the IWP, suggesting that historically, trans-oceanic dispersal and colonisation have occurred in a sporadic context for this species. The distribution and genetic structure of Clade 2 crabs suggest a very di€erent history to that of Clade 1 crabs. Clade 2 crabs are strictly con®ned to populations within northern Australia. The high haplotype-diversity estimates within these populations are markedly greater than in locations outside this region (Table 1), indicating a long-term stability of Clade 2 populations within the region. This second clade di€ers from the IWP clade by '2% sequence-divergence. The estimated coalescence time for these two clades (Table 4) falls within the early Pleistocene, pre-dating that estimated for the most recent IWP expansion by Scylla serrata by '850 000 yr. These results are surprising, given the apparent rapid radiation of Clade 1 crabs throughout the IWP. We propose two scenarios to explain this unusual phylogeographic pattern. The ®rst scenario supports our initial hypothesis; i.e. the geographically restricted Clade 2 crabs represent the remnants of formerly widespread Indian Ocean populations which extended to northern Australia but were separate from the Paci®c. Subsequent retractions of Indian Ocean populations during periods of glacial activity may have led to extinction of the clade in all but the northern Australian locations. These retractions would necessarily have had to occur prior to the current radiation of the Clade 1 Scylla serrata back into the Indian Ocean. Abundant fossil evidence of S. serrata reported from an early Pleistocene formation in South Africa (Cooper and Kensley 1991) clearly predates our estimated timing of the current Clade 1 IWP radiation, but does lend support to the existence of prior Indian Ocean populations of Scylla dating back at least 1 million years before present. An alternative scenario to the Indian Ocean expansion±retraction hypothesis is that Clade 2 crabs have evolved in-situ within or around the northern Australian region. Marine species within this region may have experienced periods of allopatry, isolated from both the Indian and Paci®c Oceans as a consequence of Pleistocene sea-level and tectonic ¯uctuations (Chappell 1976; McManus 1985). The marine environment of northern Australia has been noted as a potential biological province separate from the IWP for a variety of taxa (Briggs 1995). High levels of endemism within this region have been reported for gastropods (Laseron 1956), echinoderms (Endean 1957), sponges (Bergquist 1967), estuarine brachyurans (Davie 1985) and shore®sh (Wilson and Allen 1988). The biological provinciality of this region may have its origins in vicariance, although this issue has never been fully explored. The

presence of Haplotype A as a low-frequency Clade 1 haplotype within northern Australia (Table 1) suggests that gene ¯ow into this region from the IWP has occurred. The absence of any lineage derived from Haplotype A within northern Australia suggests that this gene ¯ow has been recent. Given the high degree of potential isolation of biota within the northern Australian region, it may be pertinent to speculate as to whether Clade 2 crabs represent a remnant race of Scylla serrata, or whether they represent the descendents of an incipient speciation event. If Clade 2 crabs represent a distinct species separate from Clade 1 crabs, then it is expected that a lack of hybridisation between the species will result in a pattern of ®xed nuclear DNA di€erences between the two clades. Comparison of nuclear DNA between Clade 1 and Clade 2 crabs derived from the same location would provide a means of testing this hypothesis. Acknowledgements Our gratitude to M. Gopurenko and B. Kehoe for their assistance in sampling Australian locations. We also appreciate the comments expressed by S.F. Chenoweth concerning some aspects of this study.

References Benzie JAH, Williams ST (1997) Genetic structure of giant clam (Tridacna maxima) populations in the West Paci®c is not consistent with dispersal by present-day ocean currents. Evolution 51: 768±783 Bergquist PR (1967) Australian intertidal sponges from the Darwin area. Micronesica 3: 175±202 Briggs JC (1995) Global biogeography. Elsevier Press, Amsterdam Brower AVZ (1994) Rapid morphological radiation and convergence among races of the butter¯y Heliconius erato inferred from patterns of mitochondrial DNA evolution. Proc natn Acad Sci USA 91: 6491±6495 Campbell NJH, Harriss FC, Elphinstone MS, Baverstock PR (1995) Outgroup heteroduplex analysis using temperature gradient gel electrophoresis: high resolution, large scale, screening of DNA variation in the mitochondrial control region. Molec Ecol 4: 407±418 Chappell JMA (1976) Aspects of late Quaternary paleogeography of the Australian±East Indonesian region. In: Kirk RL, Thorne AG (eds) The origin of the Australians. Australian Institute of Aboriginal Studies, Canberra, pp 11±77 (Human Biology Series No.6) Clary DO, Wolstenholme DR (1985) The mitochondrial DNA molecule of Drosophila yakuba: nucleotide sequence, gene organisation, and genetic code. J Molec Evolut 22: 252±271 Cooper MR, Kensley BF (1991) An early Pleistocene decapod crustacean fauna from Zululand. S Afr J Sci 87: 601±604 Crandall KA, Templeton AR (1993) Empirical tests of some predictions from coalescent theory with applications to intraspeci®c phylogeny reconstruction. Genetics, Austin, Tex 134: 959±969 Davie PJF (1985) The biogeography of littoral crabs (Crustacea: Decapoda: Brachyura) associated with tidal-wetlands in tropical and sub-tropical Australia. In: Bardsley KN, Davie JDS, Woodro€e CD (eds) Coasts and tidal wetlands of the Australian monsoon region. North Australian Research Unit, Casuarina NT, pp 259±275 (Australian National University North Australia Research Unit Mangrove Monogr No. 1) Endean R (1957) The biogeography of Queensland's shallow-water echinoderm fauna (excluding Crinoidea), with a rearrangement of the faunistic provinces of tropical Australia. Aust J mar Freshwat Res 8: 233±273

233 Excoer L, Langaney A (1989) Origin and di€erentiation of human mitochondrial DNA. Am J hum Genet 44: 73±85 Hill BJ (1994) O€shore spawning by the portunid crab Scylla serrata (Crustacea: Decapoda). Mar Biol 120: 379±384 Keenan CP, Davie PJF, Mann DL (1998) A revision of the genus Scylla De Haan, 1833 (Crustacea: Decapoda: Brachyura: Portunidae). Ra‚es Bull Zool 46: 217±245 Kocher TD, Thomas WK, Meyer A, Edwards SV, Paabo S, Villablanca FX, Wilson AC (1989) Dynamics of mitochondrial DNA evolution in animals: ampli®cation and sequencing with conserved primers. Proc natn Acad Sci, USA 86: 6196±6200 Kumar S, Tamura K, Nei G (1994) MEGA: molecular evolutionary genetic analysis software for microcomputers. Cabios 10: 189±191 Laseron CF (1956) The families Rissoidae (Mollusca) from the Solanderian and Damperian provinces. Aust J mar Freshwat Res 7: 384±485 Lavery S, Moritz C, Fielder DR (1996) Indo-Paci®c population structure and evolutionary history of the coconut crab Birgus latro. Molec Ecol 5: 557±570 Maniatis T, Fritsch EF, Sambrook J (1982) Molecular cloning, a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York McManus JW (1985) Marine speciation, tectonics and sea-level changes in southeast Asia. Proc 5th int coral Reef Congr 4: 133±138 [Gabrie C, Harmelin VM (eds) Antenne Museum± EPHE, Moorea, French Polynesia] McMillan WO, Palumbi SR (1995) Concordant evolutionary patterns among Indo±West Paci®c butter¯y®shes. Proc natn Acad Sci USA 260: 229±236 Muralidharan K, Wemmer C (1994) Transporting and storing ®eld collected specimens without refrigeration for subsequent DNA extraction and analysis. BioTech 17: 420±422 Nei M (1987) Molecular evolutionary genetics. Columbia University Press, New York

Neigel JE, Avise JC (1986) Phylogenetic relationships of mitochondrial DNA under various demographic models of speciation. In: Nevo E, Karlin S (eds) Evolutionary process and theory. Academic Press, New York, pp 515±534 Palumbi SR, Grabowsky G, Duda T, Geyer L, Tachino N (1997) Speciation and population genetic structure in tropical Paci®c sea urchins. Evolution 51: 506±1517 Palumbi SR, Martin A, Romano S, McMillan WO, Stice L, Grabowski G (1991) The simple fool's guide to PCR, version 2.0. Department of Zoology, University of Hawaii, Honolulu Roehrdanz RL (1993) An improved primer for PCR ampli®cation of mitochondrial DNA in a variety of insect species. Insect molec Biol 2: 89±91 Schneider S, Kue€er JM, Roessli D, Excoer L (1997) Arlequin version 1.1. A software for population genetic data analysis. Genetics and Biometry Laboratory, University of Geneva, Switzerland Slatkin M (1977) Gene ¯ow and genetic drift in a species subject to frequent local extinctions. Theor Popul Biol 12: 253±262 Tamura K, Aotsuka T (1988) Rapid isolation method of animal mitochondrial DNA by the alkaline lysis procedure. Biochem Genet 26: 815±819 Williams ST, Benzie AH (1997) Indo±West Paci®c patterns of genetic di€erentiation in the high-dispersal star®sh Linckia laevigata. Molec Ecol 6: 559±573 Williams ST, Benzie AH (1998) Evidence of a biogeographic break between populations of a high dispersal star®sh: congruent regions within the Indo±West Paci®c de®ned by color morphs, mtDNA, and allozyme data. Evolution 52: 87±99 Wilson BR, Allen GR (1988) Major components and distribution of the marine fauna. In: Dyne GR, Walton DW (eds) Fauna of Australia. General articles. Vol 1A. Australian Government Public Service, Canberra, pp 43±68