Portal de Periódicos da CAPES

MORE THAN BINDIN DIVERGENCE: REPRODUCTIVE ISOLATION BETWEEN SYMPATRIC SUBSPECIES OF A SEA URCHIN BY ASYNCHRONOUS SPAWNING

2012; Oxford University Press; Volume: 66; Issue: 11 Linguagem: Inglês

10.1111/j.1558-5646.2012.01700.x

1558-5646

Rachel M. Binks, Jane Prince, Jonathan P. Evans, W. Jason Kennington,

Coral and Marine Ecosystems Studies

EvolutionVolume 66, Issue 11 p. 3545-3557 Free Access MORE THAN BINDIN DIVERGENCE: REPRODUCTIVE ISOLATION BETWEEN SYMPATRIC SUBSPECIES OF A SEA URCHIN BY ASYNCHRONOUS SPAWNING Rachel M. Binks, Rachel M. Binks School of Animal Biology, The University of Western Australia, Crawley, Western Australia 6009, Australia E-mail: rachel.binks@grs.uwa.edu.auSearch for more papers by this authorJane Prince, Jane Prince School of Animal Biology, The University of Western Australia, Crawley, Western Australia 6009, AustraliaSearch for more papers by this authorJonathan P. Evans, Jonathan P. Evans School of Animal Biology, The University of Western Australia, Crawley, Western Australia 6009, AustraliaSearch for more papers by this authorW. Jason Kennington, W. Jason Kennington School of Animal Biology, The University of Western Australia, Crawley, Western Australia 6009, AustraliaSearch for more papers by this author Rachel M. Binks, Rachel M. Binks School of Animal Biology, The University of Western Australia, Crawley, Western Australia 6009, Australia E-mail: rachel.binks@grs.uwa.edu.auSearch for more papers by this authorJane Prince, Jane Prince School of Animal Biology, The University of Western Australia, Crawley, Western Australia 6009, AustraliaSearch for more papers by this authorJonathan P. Evans, Jonathan P. Evans School of Animal Biology, The University of Western Australia, Crawley, Western Australia 6009, AustraliaSearch for more papers by this authorW. Jason Kennington, W. Jason Kennington School of Animal Biology, The University of Western Australia, Crawley, Western Australia 6009, AustraliaSearch for more papers by this author First published: 20 May 2012 https://doi.org/10.1111/j.1558-5646.2012.01700.xCitations: 12AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract The evolution of reproductive barriers is crucial to the process of speciation. In the Echinoidea, studies have focused on divergence in the gamete recognition protein, bindin, as the primary isolating mechanism among species. As such, the capacity of alternate mechanisms to be effective reproductive barriers and the phylogenetic context in which they arise is unclear. Here, we examine the evolutionary histories and factors limiting gene exchange between two subspecies of Heliocidaris erythrogramma that occur sympatrically in Western Australia. We found low, but significant differentiation between the subspecies in two mitochondrial genes. Further, coalescent analyses suggest that they diverged in isolation on the east and west coasts of Australia, with a subsequent range expansion of H. e. erythrogramma into Western Australia. Differentiation in bindin was minimal, indicating gamete incompatibility is an unlikely reproductive barrier. We did, however, detect strong asynchrony in spawning seasons; H. e. erythrogramma spawned over summer whereas H. e. armigera spawned in autumn. Taken together, we provide compelling evidence for a recent divergence of these subspecies and their reproductive isolation without gamete incompatibility. Western Australian H. erythrogramma may therefore present an intriguing case of incipient speciation, which depends on long-term persistence of the factors underlying this spawning asynchrony. The process of speciation is initiated by barriers to gene exchange, which over time lead to the accumulation of genetic differences and reproductive isolation (Mayr 1942, 1963). Initial barriers to gene flow can arise by geographic displacement (i.e., allopatry), through biological, topographic, or oceanographic limitations to dispersal (e.g., Lessios et al. 2001; Johnson and Black 2006; Sherman et al. 2008), or in the same geographic location (i.e., sympatry) through ecological shifts and habitat specialization (e.g., Feder et al. 1988; Barluenga et al. 2006). Following extended periods of isolation and subsequent genetic differentiation, a variety of reproductive barriers can arise, which can be split into two categories: prezygotic barriers, which reduce the opportunity or ability for gametes to successfully fertilize, and postzygotic barriers, which impact the fitness of hybrid offspring, either through impaired survival or sterility (Mayr 1963; Dobzhansky 1970). Studying these barriers in recently diverging groups can provide considerable insight into the evolutionary processes leading to speciation. Marine species, particularly broadcast spawners, have been especially well studied in the context of understanding the evolution of prezygotic reproductive barriers owing to the ease of gamete access and manipulation (Palumbi 1994; Lessios 2007). In such species, reproductive isolation can arise in the form of temporal barriers by asynchronies in the timing of spawning (e.g., Lessios 1984; Harriot 1985; Toro et al. 2002; Kruse et al. 2004; Levitan et al. 2004), whereas others may have spatial barriers to reproduction by aggregating in discrete habitats (e.g., Billet and Hausen 1982; Pernet 1999), or chemical barriers, including gamete attraction by species-specific chemo-attractants that direct sperm toward conspecific eggs (e.g., Miller 1979, 1997; Riffell et al. 2004). Finally, where gametes do interact directly, proteins expressed on the surface of gametes can exhibit species-specific structural differences that generate functional failures during heterospecific fertilization attempts (reviews in Swanson and Vacquier 2002; Zigler et al. 2005; Palumbi 2009). The Echinoidea have recently gained considerable attention in marine speciation studies owing to the characterization of the gamete recognition protein, bindin. The extent of divergence in the bindin gene between species is directly correlated with gamete incompatibility (Zigler and Lessios 2003b; Zigler et al. 2005) and these proteins are therefore thought to play a major role in the reproductive isolation and ultimately speciation of echinoids. Divergence can arise rapidly by positive selection and as a result strong incompatibility can be found between both distantly (e.g., Strathman 1987) and closely related congeneric species (e.g., Palumbi and Metz 1991). Moreover, only genera consisting of sympatric species show bindin evolving under selection (Echinometra: Metz and Palumbi 1996; Strongylocentrotus: Biermann 1998; Heliocidaris: Zigler et al. 2003), whereas genera of allopatric species typically show neutral bindin evolution (Arbacia: Metz et al. 1998; Tripneustes: Zigler and Lessios 2003b; Lytechinus: Zigler and Lessios 2004). Although the resulting functional incompatibility is rarely 100% efficient or symmetric (e.g., Lessios and Cunningham 1990; Vacquier et al. 1995; Rahman et al. 2001; McClary and Sewell 2003; Rawson et al. 2003; Zigler et al. 2003), bindin divergence and gamete incompatibility is generally considered to be a major factor underlying speciation processes in echinoids. As a result of the recent interest in bindin evolution, relatively little attention has been paid to alternate reproductive barriers among echinoid species. Asynchronous spawning is a well-known isolating mechanism among mass-spawning species, such as corals (e.g., Wilson and Harrison 2003; Levitan et al. 2004; Mangubhai and Harrison 2009), but has also been reported across a wide range of marine invertebrates (e.g., Toro et al. 2002; Kruse et al. 2004), fish (e.g., Danilowicz 1995), and algae (e.g., Pearson and Serrão 2006). Such spawning events are typically triggered by species-specific environmental cues that may allow for reproductive isolation in heterogeneous environments. However, although lunar cycles have been shown to allow temporal gamete isolation among some Diadema species (Lessios 1984), the temperature and photoperiod cues that tend to trigger annual spawning cycles in echinoids are considered too spatially or temporally variable for spawning asynchrony to be an effective reproductive barrier (see Lessios 2007 for a review). The sea urchin Heliocidaris erythrogramma (VALENCIENNES 1846) is a well-known species for studies of developmental biology and evolution due to its unusual shift in reproductive strategy from the typical planktotrophic larvae to lecithotrophic larvae (e.g., Laegdsgaard et al. 1991; Raff et al. 1999; Wilson et al. 2005; Smith et al. 2009). As a consequence of this shift, H. erythrogramma larvae develop rapidly and have a relatively short planktonic phase, such that genetic subdivision occurs at much lower spatial scales in this species compared to its planktotrophic congener H. tuberculata (McMillan et al. 1992). In addition to this reproductive shift is strong bindin divergence and gamete incompatibility between these two species where they occur sympatrically in Eastern Australia (Zigler et al. 2003). More recently, questions have arisen as to whether similar reproductive barriers may be present within H. erythrogramma. Two subspecies, H. e. erythrogramma and H. e. armigera, are broadly distributed between east and west coasts of Australia, respectively (Mortensen 1943), but recent work has revealed that they both occur within Western Australia, albeit with a strong dominance of H. e. armigera (Binks et al. 2011a). In this region of sympatry, these subspecies can be distinguished by spine morphology and coloration, where the thicker spined H. e. armigera consistently occurs with red-violet and red-violet-green coloration whereas the thinner spined H. e. erythrogramma typically exhibits white-green, red-green, and white-violet morphs (Binks et al. 2011a). Further examination with microsatellite markers revealed genetic differentiation four times greater between subspecies than within either subspecies. This trend remained consistent even across very fine spatial scales ( tB); (2) geographic barriers formed within Western Australia to isolate and differentiate subspecies in allopatry within Western Australia (i.e., local vicariance: tA < tB); or (3) ecological barriers formed within Western Australia to isolate and differentiate each subspecies in sympatry (also tA < tB). Further assessment of consistency in the levels of divergence across loci could then be used to differentiate between the latter two hypotheses (i.e., concordant patterns across all loci with local vicariance due to random drift acting equally across the genome, or discordant patterns in a subset of loci acting in sympatry). We were also interested in estimations of m1 and m2 to determine the extent and direction of gene flow between H. e. erythrogramma and H. e. armigera since their divergence. To strengthen these two analyses, we combined the mitochondrial data from this study with microsatellite data; all Western Australian microsatellite data were acquired from a previous study (Binks et al. 2011b) whereas Eastern Australian data were obtained by amplifying individuals collected in the current study with the methods described in Binks et al. (2011b). Thus, both analyses were run with a total of five loci; CO1 and 16S (nA= 32, nB= 44 for each locus) were run under the Hasegawa-Kishino-Yano (HKY) mutational model and all three microsatellite loci (C101, C115, and D105) were run under the stepwise mutational model using a subsample of randomly selected individuals (nA= 378, nB= 268 for each locus). For each analysis, several preliminary runs were performed with wide prior parameter distributions to optimize the final run which consisted of 150 chains for a total of 10 million steps (sampling every 100), following a burnin period of 1,000,000 steps. We included a geometric heating scheme to increase mixing (g1= 0.9, g2= 0.8). During preliminary trials, the posterior distributions for t contained flat tails that did not reach zero. To set a logical upper bound for this prior, we followed the methods of Peters et al. (2007); we assumed that the time since divergence could not be older than the time of the most recent common ancestor (TMRCA). We then averaged the posterior distribution of TMRCA across loci and used the upper 95% highest posterior density (HPD) of TMRCA as the upper bound for t. For each analysis, three identical runs with different random seed numbers were performed to ensure consistency, evolutionarily stable strategy values were all above 1000 and parameter plots showed no trends, which together, indicated convergence had been reached. To convert the IMa parameter estimates to biologically meaningful demographic values, we followed the methods described and implemented in Won and Hey (2005). In doing so, we applied a mutation rate of 3.5% per million years for CO1 evolution in sea urchins (Zigler et al. 2003) and a generation turnover of two years. BINDIN SEQUENCING AND ANALYSIS Pure genomic DNA, extracted as detailed above, was utilized from 10 individuals of each subspecies (i.e., five from each Western Australian site). Approximately 1537 bp of the mature gene and its intron were amplified using primers HeF1 and HeR1 and sequenced using primers HeF1, HeR1, Heout5, Heout51, and MB1130+ (Zigler et al. 2003). PCR conditions involved a preliminary 3-min denaturation stage at 95°C, followed by 35 amplification cycles of 45 s denaturation at 94°C, 45 s at the annealing temperature (56°C) and 150 s extension at 72°C with a final extension cycle for 10 min at 72°C. All individuals were sequenced twice to ensure data quality against sequencing errors. Once the intron was excluded, no more than one heterozygous site was detected throughout the mature gene within a single individual and therefore cloning was not necessary to obtain haplotype data. Sequences were aligned and edited using Sequencher version 4.5 (Gene Codes Corp.). Haplotype sequences for H. e. armigera have been deposited in GenBank under accession numbers JQ364936–38. Following Zigler et al. (2003), we divided the bindin gene into three regions: 31 amino acids of the hotspot of rapid evolution (Metz and Palumbi 1996; Biermann 1998); 55 amino acids of the highly conserved core region (Vacquier et al. 1995; Zigler and Lessios 2003a); and 117 amino acids that flank either side of these two regions making up the rest of the gene. However, the low levels of variation detected among the Western Australian sequences across all regions negated the typical application of statistical analyses to compare the ratio of synonymous to nonsynonymous substitutions or tests for selection within or between subspecies. Patterns of bindin divergence between these subspecies were therefore described qualitatively for each region. We added all Western Australian haplotypes to the Eastern Australian Heliocidaris bindin phylogeny generated by Zigler et al. (2003) to determine where these Western Australian subspecies fit in the Heliocidaris genus, particularly for comparison with our CO1 phylogeny. We used PAUP* version 4.0b10 (Swofford 2003) to regenerate the phylogeny using the substitution model of Tamura and Nei (1993) with gamma correction and 9999 iterations for bootstrap support by neighbor joining methods. In doing so, we included bindin haplotypes from Eastern Australian H. e. erythrogramma (accession numbers AF529574–80; AF530401–13) and H. tuberculata (accession numbers AF529581–84; AF530414–43) (Zigler et al. 2003), as well as two E. mathaei haplotypes (Metz and Palumbi 1996; accession numbers U39511–12) as an outgroup, all available on GenBank. SPAWNING SEASONALITY Between 2008 and 2010, regular collections of approximately 15 individuals of each subspecies were taken from each respective site, starting in October each year and continuing fortnightly until all spawning activity had ceased (June). October was chosen as an early starting point because eastern populations of H. e. erythrogramma are known to spawn annually between the summer months of December and February (Williams and Anderson 1975; Dix 1977; Laegdsgaard et al. 1991). Only individuals greater than 50 mm in test diameter were collected to ensure sexual maturity (Dix 1977) and the size range was kept to a minimum to avoid any influence of body size on gonadal measures (Gonor 1972). Each individual was weighed to the nearest 0.001 g and then we induced spawning with a 3 mL injection of 0.5 M KCl through the peristomial membrane. For both subspecies, the proportion of each sex spawning at a given sampling time was assessed within a 30-min period following the injection. Only individuals that spawned copiously (>0.2 mL) were considered successful, that is, those most likely to be ready to spawn naturally. Individuals with limited response were classified as not spawning given that they may be in early stages of the gametogenic cycle or have already spawned (Lessios 1984). Following spawning trials, each individual was dissected for a quantitative measure of reproductive activity. The gonadal somatic index (GSI) was calculated as the total gonadal wet weight (g) divided by the total body wet weight (g), multiplied by 100 (%). Previously spawned gametes were collected dry and recombined with gonadal tissue to be included in the gonadal wet weight measure. Gonadal indices were calculated during the second spawning season only (October 2009 to June 2010). We tested each bimonthly sample for deviations from a 1:1 sex ratio using χ2 tests and then calculated the overall significance for each subspecies by summing the independent χ2 values and associated degrees of freedom. Proportion data for both spawning responses and gonadal indices were arcsine transformed (Zar 1984) and used in two-way analyses of variance (ANOVA) using JMP® version 8.0.1. We tested for variation in spawning response between sexes (fixed) and years (fixed) for each subspecies, and found no significant difference in the proportion of males and females spawning responses (HEA: F1, 40= 0.15, P= 0.70 and HEE: F1, 28= 4.23, P= 0.05), or spawning between years (HEA: F1, 40= 0.33, P= 0.57 and HEE: F1, 28= 3.47, P= 0.07). We also tested for variation in gonadal indices between sexes (fixed) and bimonthly sample (fixed) for each subspecies and found significant differences between sampling times (HEA: F17, 170= 6.16, P < 0.0001 and HEE: F16, 187= 15.25, P < 0.0001), but no difference between sexes (HEA: F1, 170= 0.41, P= 0.53 and HEE: F1, 187= 0.36, P= 0.55) or their interaction (HEA: F17, 170= 0.22, P= 0.99 and HEE: F16, 187= 0.23, P= 0.99). This suggests that the changes in GSI over time were synchronous between the sexes. Data for both spawning receptivity and gonadal indices, therefore, were pooled across the sexes and data for spawning were also pooled across years. These pooled data were then used in two separate ANOVAs to test for differences in both spawning receptivity and gonadal indices between bimonthly samples (fixed) and subspecies (fixed). Sea surface temperature (SST) and photoperiod data for all sites over the study period were obtained for comparison with spawning and GSI data. Monthly averages of SST data were provided by GIovanni Ocean Color Radiometry: Online Visualization and Analysis (http://gdata1.sci.gsfc.nasa.gov/daac-bin/G3/gui.cgi?instance_id=ocean_month) and data for photoperiod were obtained from Geoscience Australia (http://www.ga.gov.au/bin/astro/sunrisenset). There was very little spatial or temporal variation in either SST or photoperiod, so data were averaged across sites and years for each parameter. Results MITOCHONDRIAL DIVERGENCE AND PHYLOGENY Haplotype networks of H. erythrogramma for each of CO1 and 16S are shown in Fig. 2A and B. A total of 14 haplotypes were found for CO1, of which five were singletons. Two of these haplotypes were found within Western Australia and the remaining 12 in Eastern Australia, with no shared haplotypes between east and west coasts. Within Western Australia, the two haplotypes were differentiated by four substitutions and each haplotype was consistently associated with either the RV H. e. armigera or the WG H. e. erythrogramma. In contrast, there was no haplotype differentiation between RV and WG H. e. erythrogramma in Eastern Australia, and while each state tended to be dominated by one or two haplotypes, there were three haplotypes shared across the entire eastern coastline. As expected in a more conserved gene, less variation was observed for 16S, but the results support those from CO1. A total of nine haplotypes were found, five of which were singletons. Most of the variation was found in Western Australia (five haplotypes). Again, each was unique to the west coast and no haplotypes were shared between H. e. armigera