Genetic divergence among East Icelandic and ... - Oxford Journals

65

Genetic divergence among East Icelandic and Faroese populations of Atlantic cod provides evidence for historical imprints at neutral and non-neutral markers ¨ . Stefańsson, and Anna K. Danıélsdo´ttir Christophe Pampoulie, Pe´tur Steingrund, Magnus O ¨ ., and Danıélsdo´ttir, A. K. 2008. Genetic divergence among East Icelandic and Faroese populations of Pampoulie, C., Steingrund, P., Stefańsson, M. O Atlantic cod provides evidence for historical imprints at neutral and non-neutral markers. – ICES Journal of Marine Science, 65: 65–71.

During the past decade, genetic markers have been used increasingly to improve stock discrimination and to aid fisheries management. Today, the Icelandic and Faroese Plateau cod (Gadus morhua) are managed as separate units, belonging to ICES Subareas Va and Vb1, respectively. There is little information on the genetic connectivity of the two units, however, except in terms of tagging experiments which revealed limited adult migration between the two areas, and few genetic studies describing genetic differentiation among Faroese and East Icelandic cod. Here, previously published data on the genetic structure of Icelandic cod were combined with new data from the Faroe Plateau to assess the level and the source of genetic variability of Atlantic cod around the Iceland – Faroe Ridge and the potential sources of genetic variation. In all, 771 cod were genotyped at nine microsatellite loci and at the Pantophysin locus (Pan I). The genetic markers employed were congruent and showed that South Icelandic and East Icelandic – Faroese Plateau populations have limited genetic connectivity. Diversifying selection associated with restricted gene flow is likely to explain the observed pattern with the Pan I locus. Further analyses detected historical imprints in the microsatellite data, suggesting that the divergence could be due to isolation of different cod populations during the last glacial maximum. Keywords: Faroe Islands, Gadus morhua, historical signal, Iceland, microsatellite loci, migration, Pantophysin. Received 19 June 2007; accepted 26 October 2007; advance access publication 14 December 2007. ¨ . Stefańsson, and A. K. Danıélsdo´ttir: Marine Research Institute, Sku´lagata 4, 101 Reykjavı´k, Iceland. P. Steingrund: Faroese C. Pampoulie, M. O Fisheries Laboratory, Noátuń, PO Box 3051, FO-110 To´rshavn, Faroe Islands. Correspondence to C. Pampoulie: tel: þ354 575 2038; fax: þ354 575 2001; e-mail: [email protected]

Introduction The Atlantic cod (Gadus morhua) is widely distributed around the North Atlantic, and has been one of the most commercially valuable fish species inhabiting the continental shelves on both sides of the ocean. During the past decade, genetic markers have been used increasingly to assess stock structure of local populations of cod and to improve fisheries management (Nielsen et al., 2003; Sarvas and Fevolden, 2005; Pampoulie et al., 2006). Atlantic cod are genetically structured on small and large geographical scales (Ruzzante et al., 1996; Hutchinson et al., 2001; Pogson et al., 2001; Knutsen et al., 2003; Nielsen et al., 2003). Oceanic fronts (Pampoulie et al., 2006), shoreline topography (Sarvas and Fevolden, 2005), and geographical distance (Mork et al., 1985) promote weak but significant differentiation among cod populations, so these three parameters are likely to affect fisheries management. However, the genetic pattern of present-day organisms is also related to historical events such as estuary formation or postglacial expansion of a species (Beheregaray and Sunnucks, 2001; Turgeon and Bernatchez, 2001; Hardie et al., 2006), so these phenomena too may, to some extent, play a role in the origin of marine population divergence. One of the most challenging tasks for geneticists has been to disentangle historical from contemporary genetic signatures (Hilbish, 1996). To date, the relationship between Icelandic and Faroese cod spawning areas has been investigated using tagging experiments # 2007

(Ta˚ning, 1940; Jońsson, 1996; Steingrund et al., 2005), transferrin gene (de Ligny, 1969), haemoglobin polymorphism (Jamieson and Birley, 1989), and recently the Pan I locus (Case et al., 2005). Whereas tagging experiments revealed limited adult migration between Icelandic and Faroese waters, the haemoglobin study revealed a lower frequency of the HbI 1 allele in Faroese cod (Faroe Plateau) than in cod collected south of Iceland (Jamieson and Birley, 1989), though a higher frequency than in cod caught off north Iceland (de Ligny, 1969). The transferrin gene corroborated this finding and revealed a slightly greater frequency of the predominant allele TfC in Faroese Plateau cod than in Icelandic cod. The Pan I locus data also revealed significant variation in allele frequency among Icelandic and Faroese cod, the latter usually exhibiting a higher frequency for the Pan IA allele (Case et al., 2005). Although managed as a single stock, it has recently been suggested that the Icelandic stock is partitioned into two populations, northeast and southwest (Jońsdo´ttir et al., 2006; Pampoulie et al., 2006). In Icelandic waters, the spawning season lasts from mid-March to May (Marteinsdo´ttir and Bjo¨rnsson, 1999). Pelagic eggs and larvae from the southwest spawning regions passively disperse over vast distances in a clockwise manner around the country, so limiting population divergence (Marteinsdo´ttir et al., 2000; Brickman et al., 2007). In contrast, eggs and larvae from cod spawning off northeast Iceland have

International Council for the Exploration of the Sea. Published by Oxford Journals. All rights reserved. For Permissions, please email: [email protected]

66 been suggested to be retained within fjords and inshore waters (Marteinsdo´ttir et al., 2000). Although spawning site fidelity is stock-dependent and differs widely (Robichaud and Rose, 2004), tagging data in the study area provide evidence that cod exhibit homing behaviour (Ta˚ning, 1940; Jońsson, 1996; Pampoulie et al., 2006), potentially promoting genetic divergence among populations. On the Faroe Plateau, cod spawn in two main spawning areas located north and west of the Faroe Islands (Joensen et al., 2005; Steingrund et al., 2005). Post-spawning cod usually disperse around the Plateau (Ta˚ning, 1940; Joensen et al., 2005). The Faroe Islands are separated from other shallow areas by depths exceeding 500 m (Steingrund and Gaard, 2005), but are nevertheless connected to the Southeast Icelandic shelf through the Iceland– Faroe Ridge (depth ,500 m). Two main currents might play a role in the dispersal of eggs/larvae, the East Icelandic Current, which flows along the east coast of Iceland towards the Faroe Islands, and the Irminger Current, from which a minor branch flows along the south coast of Iceland (Figure 1). Both currents might promote unidirectional dispersal of cod from Icelandic to Faroese populations and thereby homogenize them. Here, the level of genetic variation among cod from the East Icelandic and Faroese spawning areas was assessed using nine microsatellite loci and the Pantophysin locus (Pan I), which is considered to be under positive Darwinian selection (Pogson and Mesa, 2004). Conventional methods and recently developed spatial/geographical (Dupanloup et al., 2002; Manni et al., 2004) approaches were applied to detect areas of genetic discontinuity. Additionally, we attempted to assess which evolutionary forces were responsible for the genetic differentiation we observe. We therefore compared the global variance of allelic identity and allelic size among the studied populations, with the aim of generating information on the relative impact of drift (u) vs. mutation (RST) on the observed divergence (Hardy et al., 2003). These results are then interpreted in the light of results from tagging experiments conducted on cod spawning grounds off the Faroes and Iceland.

Material and methods In all, 771 cod were collected from six spawning areas from the Icelandic south [Area 812 was sampled in the years 2002 and

C. Pampoulie et al. 2003 (812b)] and east coasts and two areas of the Faroe Plateau (Figure 1) during the spawning season. A subset of Icelandic samples analysed in Pampoulie et al. (2006) that was characteristic of the south and north spawning components was chosen to preclude unbalancing sample sizes. The Icelandic samples were identified as described in Pampoulie et al. (2006). Totals of 94 cod per sample were analysed except for samples 612 (n ¼ 84), 621 (77), Far-N (85), and Far-W (55). The depths of the cod samples from Faroe North (Far-N) and West (Far-W) were 103 and 115 m, respectively. Samples were genotyped at nine microsatellite loci: Gmo2, Gmo8, Gmo19, Gmo34, Gmo37, Tch5, Tch11, Tch14, and Tch22 (Brooker et al., 1994; Miller et al., 2000; O’Reilly et al., 2000), and at the Pantophysin locus (Pan I). DNA extraction, PCR, and genotyping were performed as described in Pampoulie et al. (2006). Allele frequencies and observed (Ho) and expected heterozygosity (He) were calculated in GENETIX version 4.03 (Belkhir et al., 1999). Samples were tested for Hardy –Weinberg expectation (HWE) using GENEPOP version 3.1 (Raymond and Rousset, 1995). Estimates of genetic divergence were calculated with FST, using the estimator u (Weir and Cockerham, 1984) in GENETIX. Pairwise FST estimates were tested for significance by permutation (5000) and adjusted with a sequential Bonferroni test (Rice, 1989). A multidimensional scaling (MDS) approach was carried out on the pairwise FST values for both microsatellite loci and the Pan I locus to project the pairwise FST values on a twodimension plane, using the R-package (Ross and Gentleman, 1996). Potential genetic discontinuities were assessed with the Monmonier’s maximum distance algorithm implemented in BARRIER version 2.2 (Manni et al., 2004). The analysis was conducted as described in Pampoulie et al. (2006). The potential number of populations was estimated using spatial analysis of molecular variance (SAMOVA, Dupanloup et al., 2002). SAMOVA employs a simulated procedure and uses allele frequency data along with geographical coordinates of the sample populations to identify groups of populations that exhibit close genetic relationships. The grouping of populations is based on a hierarchical analysis of rST and maximizing the proportion of total genetic variance between groups (rCT). We conducted the analysis for 2 –4 potential populations. A random permutation of different allele sizes at each microsatellite loci among allelic states (2000 replicates) was performed to simulate the FST distribution (rRst), with 95% confidence intervals (CIs) using SPAGEDI version 1.1 (Hardy et al., 2003) and to elucidate the role of mutation in the genetic data. Differentiation is likely to reflect the impact of genetic drift if RST u and if RST is within the 95% CI of rRST. On the other hand, differentiation is likely due to mutation if RST . u, and if it falls above the 95% CI of rRST (Hardy et al., 2003).

Results Microsatellite loci

Figure 1. Location of Atlantic cod samples (dots) and main oceanic currents (modified from Valdimarsson and Malmberg, 1999). Depth contours are 1000 and 2000 m.

All microsatellite loci were highly polymorphic. The number of alleles per locus across samples ranged from 10 (Gmo34) to 49 (Gmo8). Ho averaged over loci and all samples ranged from 0.735 to 0.858, and He ranged from 0.779 to 0.859. Genotypic proportions were significantly out of HWE in 2 of 81 exact tests (this is fewer than expected by chance alone). Genotypic frequencies in all but one sample (811) were in HWE. The genetic variation among

Genetic divergence among East Icelandic and Faroese populations of Atlantic cod

67

Table 1. Estimates of FST (Weir and Cockerham, 1984) among pairs of populations of Gadus morhua. 511 612 621 811 812 812b 822 Far-N Far-W 511 0 20.0011 0.0027 0.0013 0.0023 0.0022 0.0042 0.0010 0.0006 . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . 612 0.1552 0 0.0012 0.0004 0.0013 0.0029 0.0032 20.0009 20.0007 . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . 621 0.2245 0.0146 0 0.0031 0.0063 0.0042 0.0059 20.0009 0.0009 . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . 811 20.0036 0.1740 0.2442 0 0.0016 0.0014 0.0018 0.0004 20.0004 . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . 812 0.0812 0.3727 0.4388 0.0660 0 20.0001 20.0008 0.0043 0.0023 . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . 812b 0.1378 0.4371 0.4971 0.1195 0.0033 0 0.0001 0.0042 0.0032 . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . 822 0.1511 0.4556 0.5160 0.1321 0.0073 20.0044 0 0.0047 0.0035 . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . Far-N 0.1182 20.0062 0.0287 0.1341 0.3145 0.3804 0.3978 0 20.0013 . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . Far-W 0.2324 0.0222 20.0056 0.2520 0.4440 0.5016 0.5204 0.0399 0 The FST estimates for the nine microsatellite loci are above the diagonal and for the Pan I locus below the diagonal. Emboldened values indicate significant values after Bonferroni correction (p , 0.01 for 5% significance level).

and within samples showed a significant FST of 0.002 (CI: 0.0001– 0.0060) and a non-significant FIS of 0.020. After Bonferroni correction, pairwise differentiation between populations yielded 11 significant FST comparisons, of which seven were between the Faroese and Icelandic populations (Table 1). MDS analysis revealed the presence of two distinct groups clustering along the first dimension, namely the South Icelandic group and the East Icelandic –Faroese group (Figure 2a). The Monmonier algorithm analysis detected one strong genetic discontinuity supported by five loci (Gmo2, Gmo8, Gmo34, Tch14, Tch22), between Southeast Icelandic (812, 812b, and 822) and East Icelandic – Faroese samples. The SAMOVA confirmed this finding and revealed that genetic differentiation among groups was highest (rCT ¼ 0.0031) when samples were pooled into two groups, one consisting of the East Icelandic –Faroese samples (plus sample 811), and the other those from South Iceland (Table 2). The RST estimates fell above the 95% CI of rRST at two microsatellite loci (Gmo8 and Tch14; Table 3) and although not significant, the estimates were also elevated at locus Gmo19 and Tch11. Collectively, this resulted in significantly larger RST multi-locus estimates than rRST (RST . u). The results therefore provide evidence that mutation may have been the dominant factor contributing to the genetic divergence observed at these loci.

Pan I locus Genotypic frequencies did not deviate from HWE in any of the samples. The genetic variance among and within samples showed a significant FST of 0.255 (CI: 0.190 –0.310) and a nonsignificant FIS of 20.120. After Bonferroni correction, pairwise differentiation between populations yielded 21 significant FST comparisons (Table 1), among which nine were attributable to the Faroese and South Icelandic samples. MDS analysis revealed two distinct groups clustering along the first dimension, the South Icelandic group and the East Icelandic –Faroese group (Figure 2b). Allelic frequencies varied greatly, and almost no Pan IB alleles were observed in the Faroe Plateau or the East Icelandic samples (612 and 621) (Figure 3). SAMOVA revealed that among-group genetic differentiation was greatest (rCT ¼ 0.3180) when samples were pooled into two groups, one comprising the East Icelandic – Faroese samples (plus sample 811), and the other those from South Iceland (Table 1). As cod samples were collected mainly in shallow areas, no depth pattern could be detected (regression Pan IB frequency and depth; r ¼ 0.1109, p . 0.05).

Figure 2. MDS analysis on eight spawning samples of Atlantic cod collected from off East Iceland and the Faroe Plateau, based on FST values computed for (a) nine microsatellite loci, and (b) the Pan I locus.

68

C. Pampoulie et al.

Table 2. SAMOVA carried out on nine microsatellite loci and the Pan I locus (Dupanloup et al., 2002) Locus Number of groups Group composition r CT r ST Microsatellite loci . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 2 (812, 812b, 822) (511, 612, 621, 811, Far-N, Far-W) 0.0031** 0.0014** . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 3 (811) (812, 812b, 822) (511, 612, 621, Far-N, Far-W) 0.0028* 20.001* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 4 (811) (812, 812b, 822) (511, 612) (621, Far-N, Far-W) 0.0029* 20.0070* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . Pan I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 2 (812, 812b, 822) (511, 612, 621, 811, Far-N, Far-W) 0.3180** 0.0727** . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 3 (812, 812b, 822) (511, 811) (612, 621, Far-N, Far-W) 0.3165** 0.0004** . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 4 (812, 812b, 822) (511) (811) (612, 621, Far-N, Far-W) 0.3062** 0.0015** Significance was assessed by 2000 permutations. Groups are presented in parentheses. *p , 0.01, **p , 0.001.

Table 3. Mean single locus and multi-locus estimates of RST, uST, and rRST (95% CIs are presented in parentheses) between nine samples of Atlantic cod following 2000 permutations (Hardy et al., 2003). Locus uST RST rRST (95% range) Gmo2 20.001 20.002 0.001 (20.005/0.008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . a 20.001 (20.004/0.007) Gmo8 0.002 0.021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gmo19 0.001 0.003 20.003 (20.004/0.007) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gmo34 0.020 20.002 20.001 (20.004/0.008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gmo37 0.001 20.004 20.004 (20.004/0.008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tch5 20.001 20.001 20.001 (20.005/0.008) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tch11 0.001 0.005 20.001 (20.004/0.007) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . a 20.001 (20.005/0.007) Tch14 20.002 0.009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tch22 0.002 0.001 0.001 (20.004/0.007) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.001 (20.002/0.003) Multi-locus 0.002 0.010a a

Locus for which mutation played a significant role in the divergence (RST . rRST and 95% CI).

Figure 3. Allele frequencies at the Pan I locus in the studied area. Black, Pan IB allele; white, Pan IA allele.

Discussion Our study was primarily aimed at assessing the genetic variability of Atlantic cod around the Iceland–Faroe Ridge by comparing local spawning populations collected during the spawning

seasons off East and South Iceland and on the Faroe Plateau. Most striking is the congruence between the results from the microsatellite and the Pan I genotype analyses. Although existing tagging data reveal the quasi-absence of connectivity among Icelandic and Faroese cod (Jońsson, 1996; Joensen et al., 2005; Anon., 2006), both genetic markers revealed (i) the absence of genetic structure in Faroe Plateau waters, (ii) the absence of genetic differentiation among East Icelandic and Faroese spawning populations, (iii) significant genetic differentiation among South Icelandic and Faroese spawning populations, and (iv) historical genetic imprints at the microsatellite loci. Below we discuss each of these findings in more detail.

Pattern of genetic variation Faroe plateau north vs. west The lack of genetic variation among samples collected on the Faroe Plateau observed with both types of genetic marker confirms earlier findings at the haemoglobin locus and transferrin gene, and showing that the frequencies of the HbI 1 and TfC alleles were homogeneous on the Faroe Plateau (de Ligny, 1969; Jamieson and Birley, 1989). In addition, statistical tests based on allele frequencies obtained from Case et al. (2005) revealed that current Far-N and Far-W samples were not significantly different from the sample of Case et al. (2005) collected around the Faroe Plateau (x2 test: Far-N, x2 ¼ 3.60, d.f. ¼ 1, p . 0.05; Far-W, x2 ¼ 0.91, d.f. ¼ 1, p . 0.05). The apparent genetic homogeneity might be due to the strong current flowing in a clockwise manner around the Faroe Plateau promoting the dispersal of eggs and larvae from the two spawning grounds around the Plateau, so limiting the possibility for a build-up of genetic divergence among spawning areas (Steingrund et al., 2005). Juvenile cod growing up in specific nursery areas close to land are expected to originate from the Faroe North and West spawning areas in a random fashion. They frequently select an area deeper and outside the nursery area as they grow (Joensen et al., 2005), and eventually seek one of the two spawning areas when they become sexually mature (Steingrund et al., 2005). Adult cod on the Faroe Plateau are thought to migrate to the same spawning (Ta˚ning, 1940) and feeding area in consecutive years (Steingrund et al., 2005). Such spawning site fidelity, however, does not appear to represent true homing behaviour, because it does not materialize in genetic heterogeneity at microsatellite loci between the two spawning areas.

Genetic divergence among East Icelandic and Faroese populations of Atlantic cod

East Iceland vs. Faroe Plateau Although the haemoglobin locus, the transferrin gene, and the Pan I locus suggested significant variation in the frequency of the HbI 1, TfC, and Pan I alleles among East Iceland and Faroese cod (de Ligny, 1969; Case et al., 2005), we did not detect genetic divergence with either the microsatellite loci or the Pan I locus. The absence of genetic variation among cod from East Iceland and the Faroe Plateau could be due to passive drift of eggs/larvae from the former to the latter. Such unidirectional dispersal could be facilitated by the East Icelandic Current and the east branch of the Irminger Current. A recent particle-tracking model estimating drift probabilities for Icelandic cod suggested possible offshore dispersion of larvae from Iceland to Faroese waters along the Iceland–Faroe Ridge (Brickman et al., 2007), so corroborating the results of a drifter experiment conducted in 1995 and 1996 (Valdimarsson and Malmberg, 1999). Nevertheless, a 0-group survey in Faroese waters conducted annually since 1974 has shown that cod larvae are rarely found near the northern and western spawning areas of the Faroe Islands, and that virtually none are found on the Iceland–Faroe Ridge (J. Reinert, Faroese Fisheries Laboratory, To´rshavn, Faroe Islands, pers. comm.). Additionally, several tagging experiments have shown limited adult migration between Icelandic and Faroese waters. In a 40-year study, researchers caught just five of 10 969 tagged and recaptured Icelandic cod in Faroese waters (Jońsson, 1996). Likewise, just 1 out of 1043 tagged and recaptured cod marked in Faroese waters from 1952 to 1965 were taken in Icelandic waters (Joensen et al., 2005). Recent tagging experiments conducted from 1997 to 2005 in both Icelandic and Faroese waters as well as at the Iceland –Faroe Ridge also confirmed adult cod migration between the two areas to be rare (Anon., 2006).

South Iceland vs. Faroe Plateau Having established genetic similarity between East Iceland and the Faroe Plateau (East Iceland–Faroes group), we turn our attention to South Iceland. Genetic differentiation was significant among South Icelandic and East Icelandic –Faroese “groups” with both types of genetic marker, corroborating the earlier transferrin gene study (de Ligny, 1969). Most of the significant comparisons were between these two groups (Figure 2), and a significant barrier to gene flow was detected with the microsatellite loci (BARRIER). For the microsatellite loci, the level of differentiation among samples was low but significant, and comparable with the differentiation observed on a small geographic scale (Ruzzante et al., 2001), whereas the Pan I locus showed a stronger signal than previous studies carried out at the same geographic scale (Sarvas and Fevolden, 2005). Both adult migration and the passive dispersal of eggs/larvae between South Iceland and the East Iceland– Faroe regions could be hindered by the frontal zone located on the Iceland–Faroe ridge (east –southeast). There, a minor branch of the Irminger Current flowing along the southeastern edge of the Icelandic shelf meets the East Icelandic Current on the Iceland– Faroe Ridge, creating the frontal zone (Figure 1; Pampoulie et al., 2006).

Biological significance of these patterns Although studies of microsatellite loci often determine significant (but weak) divergence among groups of individuals, it can be complicated to assess the biological significance of such divergence (Waples, 1998). One way to resolve this issue would be to combine several genetic markers with biological and/or tagging data

69

(Waples, 1998). Here, such a comparison confirmed that the significant genetic difference between South Iceland and East Iceland–Faroes cod populations does have a biological base. Moreover, Pogson and Mesa (2004) showed that the Pan I locus can be subject to positive Darwinian selection, and it has been suggested that Pan I is under disruptive selection within Icelandic waters (Pampoulie et al., 2006). In the current study, a strong significant positive correlation was found between the levels of genetic divergence at both types of marker employed (r ¼ 0.745, p , 0.001). This suggests that the FST values observed at the Pan I locus are likely to be due to diversifying selection that accentuates differences not readily apparent at neutral loci, so confirming an earlier study performed off Iceland (Pampoulie et al., 2006). Hence, the genotypic pattern observed at both types of locus suggests that the genetic pattern observed could be the result of restricted gene flow among the areas studied.

Historical imprints in the microsatellite loci data Alternatively, the observed genetic pattern might merely reflect similar evolutionary (microsatellite loci) and/or selective (Pan I locus) histories of the Faroese and Icelandic cod. Contemporary levels of differentiation observed at microsatellite loci have commonly been shown to originate from historical restriction of gene flow, rather than a current isolation of populations (Turgeon and Bernatchez, 2001; Hanfling et al., 2002; Hardie et al., 2006). In our study, we observed limited genetic differentiation with microsatellite loci, whereas tagging data reflect limited migration among the populations studied. In the absence of gene flow, genetic differentiation attributable to drift and/or mutation will rise among populations given sufficient time. Low FST estimates in the absence of gene flow can therefore reflect a rapid expansion and recent common origin of populations in historical time, as previously suggested for Atlantic cod by Pogson et al. (1995). Although a significant mutational signal was only detected at two microsatellite loci, a general trend could be observed at four more of the eight studied, indicating the influence of past historical events on microsatellite data (Table 2). During the last glacial maximum (LGM, 25 000 years ago), most of the North Atlantic was covered by an ice sheet, and a reconstruction of temperature then suggested the presence of a sea ice limit (228C isotherm) south of Iceland and west of the British Isles (Siegert and Dowdeswell, 2004). Therefore, present-day North Atlantic waters were inhabitable by Atlantic cod during the LGM, and the low level of genetic differentiation observed at microsatellite loci and the clear gradient in Pan I allele frequencies within the North Atlantic (Case et al., 2005) are likely to reflect historical isolation of cod populations in refugia during the LGM, followed by recolonization of newly ice-free environment. Indeed, the genetic structure of Atlantic cod populations has been suggested to originate from recolonization events of ice-free environment following the last deglaciation (Pogson et al., 1995; Hardie et al., 2006).

Conclusions This is the first study to have investigated genetic variation among Faroe Plateau and Icelandic cod using microsatellite loci and the Pantophysin locus. The pattern observed with both genetic markers was congruent and revealed significant genetic divergence between South Icelandic and Faroese populations, but not between the latter and East Icelandic populations. This genetic pattern is likely to be due to the dispersal of Icelandic cod eggs/larvae

70 to Faroese waters, rather than active migration, as tagging experiments show. Alternatively, the presence of historical imprints in the data (mutational component) suggests that the recent origin of cod populations could be the underlying cause for the limited genetic differentiation observed. Additional research needs to be done to understand more completely the genetic connectivity of the populations.

Acknowledgements The research has been carried out under the METACOD EU project (Q5RS-2001-00 953). We thank the crews of the vessels and researchers for assistance with sampling, Th. D. Jo¨rundsdo´ttir for sample processing, H. Valdimarsson for discussion on oceanic currents, and E. Hjo¨rleifsson for his valuable comments on an early version of the manuscript. Particular thanks are also due to two anonymous referees for their helpful comments on the submitted draft.

References Anon. 2006. Report of the North-Western Working Group, 25 April – 4 May. ICES Document CM2006/ACFM: 26. Beheregaray, L. B., and Sunnucks, P. 2001. Fine-scale structure, estuarine colonization and incipient speciation in the marine silverside fish Odontesthes argentinensis. Molecular Ecology, 10: 2849– 2866. Belkhir, K., Borsa, P., Goudet, J., and Bonhomme, F. 1999. GENETIX: logiciel sous Windows pour la geńe´tique des populations Version 3.0. Laboratoire Geńome and Population, CNRS – UPR, Universite´ de Montpellier II, Montpellier (France). Brickman, D., Marteinsdo´ttir, G., Logemann, K., and Harms, I. H. 2007. Drift probabilities for Icelandic cod larvae. ICES Journal of Marine Science, 64: 49– 59. Brooker, A. L., Cook, D., Bentzen, P., Wright, J. M., and Doyle, R. W. 1994. Organization of microsatellites differs between mammals and cold-water teleost fishes. Canadian Journal of Fisheries and Aquatic Science, 51: 1959– 1966. Case, R. A. J., Hutchinson, W. F., Hauser, L., Van Oosterhout, C., and Carvalho, G. R. 2005. Macro- and micro-geographic variation in pantophysin (Pan I) allele frequencies in NE Atlantic cod Gadus morhua. Marine Ecology Progress Series, 301: 267 – 278. de Ligny, W. 1969. Serological and biochemical studies on fish populations. Oceanography and Marine Biology: An Annual Review, 7: 411– 513. Dupanloup, I., Schneider, S., and Excoffier, L. 2002. A simulated annealing approach to define the genetic structure of populations. Molecular Ecology, 11: 2571 – 2581. Hanfling, B., Hellemans, B., Volckaert, F. A. M., and Carvalho, G. R. 2002. Late glacial history of the cold-adapted freshwater fish Cottus gobio, revealed by microsatellites. Molecular Ecology, 11: 1717– 1729. Hardie, D. C., Gillet, R. M., and Hutchings, J. A. 2006. The effects of isolation and colonization history on the genetic structure of marine-relict populations of Atlantic cod (Gadus morhua) in the Canadian Arctic. Canadian Journal of Fisheries and Aquatic Sciences, 63: 1830– 1839. Hardy, O. J., Charbonnel, N., Freville, H., and Heuertz, M. 2003. Microsatellite allele sizes: a simple test to assess their significance on genetic differentiation. Genetics, 163: 1467– 1482. Hilbish, T. J. 1996. Population genetics of marine species: the interaction of natural selection and historically differentiated populations. Journal of Experimental Marine Biology and Ecology, 200: 67 – 83. Hutchinson, W. F., Carvalho, G. R., and Rogers, S. I. 2001. Marked genetic structuring in localised spawning populations of cod

C. Pampoulie et al. Gadus morhua in the North Sea and adjoining waters, as revealed by microsatellites. Marine Ecology Progress Series, 223: 251– 260. Jamieson, A., and Birley, A. J. 1989. The demography of haemoglobin polymorphism in the Atlantic cod, Gadus morhua L. Journal of Fish Biology, 35: 193– 204. Joensen, J. S., Steingrund, P., Henriksen, A., and Mouritsen, R. 2005. Migration of cod (Gadus morhua): tagging experiments at the Faroes 1952– 1965. Fro´ðskaparrit, 53: 100– 135. Jońsdo´ttir, I. G., Campana, S. E., and Marteinsdo´ttir, G. 2006. Stock structure of Icelandic cod Gadus morhua L. based on otolith chemistry. Journal of Fish Biology, 69: 136– 150. Jońsson, J. 1996. Tagging of cod (Gadus morhua) in Icelandic waters 1948– 1986. Rit Fiskideildar, 14: 1 – 82. Knutsen, H., Jorde, P. E., Andre´, C., and Stenseth, N. C. 2003. Fine-scaled geographical population structuring in a highly mobile marine species: the Atlantic cod. Molecular Ecology, 12: 385– 394. Manni, F., Gue´rard, E., and Heyer, E. 2004. Geographic patterns of (genetic, morphologic, linguistic) variation: how barriers can be detected by “Monmonier’s algorithm”. Human Biology, 76: 173– 190. Marteinsdo´ttir, G., and Bjo¨rnsson, H. 1999. Time and duration of spawning of cod in Icelandic waters. ICES Document CM 1999/ Y: 34. Marteinsdo´ttir, G., Guðmunðsdo´ttir, A., Thorsteinsson, V., and Stefansson, G. 2000. Spatial variation in abundance, size composition and viable egg production of spawning cod (Gadus morhua L.) in Icelandic waters. ICES Journal of Marine Science, 57: 824– 830. Miller, K. M., Le, K. D., and Beacham, T. D. 2000. Development of triand tetranucleotide repeat microsatellite loci in Atlantic cod (Gadus morhua). Molecular Ecology, 9: 238– 239. Mork, J., Ryman, N., Sta˚hl, G., Utter, F., and Sundnes, G. 1985. Genetic variation in Atlantic cod (Gadus morhua) throughout its range. Canadian Journal of Fisheries and Aquatic Sciences, 42: 1580– 1587. Nielsen, E. E., Hansen, M. M., Ruzzante, D. E., Meldrup, D., and Gronkjaer, P. 2003. Evidence of a hybrid-zone in Atlantic cod (Gadus morhua) in the Baltic and the Danish Belt Sea revealed by individual admixture analysis. Molecular Ecology, 12: 1497– 1508. O’Reilly, P. T., Canino, M. F., Bailey, K. M., and Bentzen, P. 2000. Isolation of twenty low stutter di- and tetranucleotide microsatellites for population analyses of walleye pollock and other gadoids. Journal of Fish Biology, 56: 1074– 1086. Pampoulie, C., Ruzzante, D. E., Chosson, V., Jo¨rundsdo´ttir, T. G., Taylor, L., Thorsteinsson, V., Danıélsdo´ttir, A. K., et al. 2006. The genetic structure of Atlantic cod (Gadus morhua) around Iceland: insight from microsatellites, the Pan I locus, and tagging experiments. Canadian Journal of Fisheries and Aquatic Sciences, 63: 2660– 2674. Pogson, G. H., and Mesa, K. A. 2004. Positive Darwinian selection at the pantophysin (Pan I) locus in marine gadid fishes. Molecular Biology and Evolution, 21: 65 – 75. Pogson, G. H., Mesa, K. A., and Boutilier, R. G. 1995. Genetic population structure and gene flow in the Atlantic cod Gadus morhua: a comparison of allozyme and nuclear RFLP loci. Genetics, 139: 375– 385. Pogson, G. H., Taggart, T. C., Mesa, K. A., and Boutilier, R. G. 2001. Isolation by distance in the Atlantic cod Gadus morhua, at large and small geographical scales. Evolution, 55: 131 – 146. Raymond, R., and Rousset, F. 1995. GENEPOP (Version 1.2): population genetic software for exact tests and ecumenicism. Journal of Heredity, 86: 248 – 249. Rice, W. R. 1989. Analyzing table of statistical tests. Evolution, 43: 223– 225.

Genetic divergence among East Icelandic and Faroese populations of Atlantic cod Robichaud, D., and Rose, G. A. 2004. Migratory behaviour and range in Atlantic cod: inference from a century of tagging. Fish and Fisheries, 5: 185– 214. Ross, I., and Gentleman, R. 1996. R: a language for data analysis and graphics. Journal of Computational Graphical and Statistics, 5: 299– 314. Ruzzante, D. E., Taggart, C. T., Cook, D., and Goddard, S. 1996. Genetic differentiation between inshore and offshore Atlantic cod (Gadus morhua) off Newfoundland: microsatellite DNA variation and antifreeze level. Canadian Journal of Fisheries and Aquatic Sciences, 53: 634– 645. Ruzzante, D. E., Taggart, C. T., Doyle, R. W., and Cook, D. 2001. Stability in the historical pattern of genetic structure of Newfoundland cod (Gadus morhua) despite the catastrophic decline in population size from 1964 to 1994. Conservation Genetics, 2: 257 – 269. Sarvas, T. H., and Fevolden, S. E. 2005. Pantophysin (Pan I) locus divergence between inshore v. offshore and northern v. southern populations of Atlantic cod in the north-east Atlantic. Journal of Fish Biology, 67: 444– 469. Siegert, M. J., and Dowdeswell, J. A. 2004. Numerical reconstruction of the Eurasian ice sheet and climate during the late Weichselian. Quaternary Science Reviews, 23: 1273– 1283.

71

Steingrund, P., and Gaard, E. 2005. Relationship between phytoplankton production and cod production on the Faroe shelf. ICES Journal of Marine Science, 62: 163 – 176. Steingrund, P., Hansen, B., and Gaard, E. 2005. Cod in Faroese waters. In Spawning and Life History Information for North Atlantic Cod Stocks. ICES Cooperative Research Report, 274: 50 – 55. Ta˚ning, A. V. 1940. Migration of cod marked on the spawning places off the Faroese. Meddelelser fra Kommissionen for Danmarks fiskeri- og havundersogelser, serie: Fiskeri, 10: 1 – 52. Turgeon, J., and Bernatchez, L. 2001. Clinal variation at microsatellite loci reveals historical secondary intergradation between glacial races of Coregonus artedi (Teleostei: Coregoninae). Evolution, 55: 2274– 2286. Valdimarsson, H., and Malmberg, S. A. 1999. Near-surface circulation in Icelandic waters derived from satellite tracked drifters. Rit Fiskideildar, 16: 23 – 39. Waples, R. S. 1998. Separating the wheat from the chaff: patterns of genetic differentiation in high gene flow species. Journal of Heredity, 89: 438 – 450. Weir, B. S., and Cockerham, C. C. 1984. Estimating F-statistics for the analysis of population structure. Evolution, 38: 1358– 1370. doi:10.1093/icesjms/fsm181