SCRS/2001/111

A GENETIC PERSPECTIVE ON ATLANTIC STOCK STRUCTURE

Jan R. McDowell1 and John E. Graves

SUMMARY

The genetic basis of stock structure of Atlantic sailfish (Istiophorus platypterus) was investigated using analyses of the mitochondrial control region and three nuclear microsatellite loci. Relatively robust samples (n= 293) from 6 locations throughout the Atlantic were screened for variation at these four hypervariable gene regions. While considerable variation was revealed, almost all of the variation was present in individual samples. The null hypothesis that samples were drawn from a common gene pool could not be rejected when each sample location was considered separately, or when collections were combined into eastern and western Atlantic samples. The apparent genetic homogeneity reinforces recent reports on the distribution of sailfish across the Atlantic and calls to question the current application of eastern and western Atlantic stocks.

RÉSUMÉ

La base génétique de la structure du stock de voilier de l=Atlantique (Istiophorus platypterus) a été examinée au moyen d=analyses de la région de contrôle mitochondriale et de trois loci microsatellitaires. Des échantillons relativement robustes (n=293) de six locations de l=Atlantique ont été étudiés à la recherche de variations de ces quatre régions à gènes hypervariables. Bien qu=une variation considérable ait été observée, elle était presque entièrement présente dans les échantillons individuels. L=hypothèse nulle que les échantillons proviennent d=une source génétique commune n=a pas pu être rejetée lorsque chaque location d=échantillonnage était examinée séparément, ou lorsque les collections étaient combinées en échantillons de l=Atlantique est et ouest. L=homogénéité génétique apparente renforce les rapports récents sur la distribution du voilier dans l=Atlantique et remet en question l=application actuelle de stocks est et ouest dans l=Atlantique.

RESUMEN

Se ha investigado la base genética de la estructura de stock del pez vela del Atlántico (Istiophorus platypterus) utilizando análisis de la región de control mitocondrial y tres loci microsatélite nucleares. Se han examinado muestras relativamente robustas (n=293) de 6 lugares de todo el Atlántico en busca de alguna variación en estas cuatro regiones genéticas hipervariables. Aunque se descubrió una variación considerable, prácticamente toda la variación estaba presente en muestras individuales. La hipótesis nula de que las muestras fueron extraídas de una masa genética común no podía rechazarse al considerar por separado cada lugar de la muestra, o al combinar las recopilaciones en muestras del Atlántico oriental y occidental. La aparente homogeneidad genética refuerza los informes recientes sobre la distribución de pez vela a través del Atlántico y cuestiona la actual aplicación de los stocks del Atlántico este y oeste.

KEYWORDS

Atlantic sailfish, DNA, genetic, population genetics, population structure, geographical distribution, stock identification.

1 Virginia Institute of Marine Science, School of Marine Science, College of William and Mary. Rt. 1208 Greate Rd. Gloucester Point, VA, 23062. [email protected]

INTRODUCTION

The sailfish, Istiophorus platypterus, is globally distributed in tropical and sub-tropical marine waters and ranges from 40 N to 40 S in the western Atlantic Ocean and from 50 N to 32 S in the eastern Atlantic. Sailfish are largely coastal spawners and spawning is thought to occur off the East Coast of Florida from May through October, off Senegal in July and August, and off Brazil from November to February (Nakamura, 1985).

Historically, the International Commission for the Conservation of Atlantic Tunas (ICCAT) has managed Atlantic sailfish as separate eastern and western stocks with an arbitrary line drawn at 30o W between the two management units. This model is based on both morphological and tag and recapture data. Specifically, eastern Atlantic sailfish tend to be larger than western Atlantic sailfish, and spots that occur on the inter-radial area of the dorsal fin of sailfish from Brazil, West Africa and the Indian Ocean are absent on western north Atlantic sailfish (ICAAT, 1994). It is important to note that these types of morphological characters can be significantly influenced by environmental conditions, and differences in morphology may represent environmental differences rather than different genetic stocks. In addition, although the Cooperative Tagging Center (CTC) and The Foundation (TBF) have tagged 94,299 Atlantic sailfish in the years 1959-1999, there have been no trans-Atlantic tag recaptures of sailfish, suggesting a lack of mixing between east and west. However the tag recovery rate for sailfish is extremely low, (1.78% Prince et al; 2001) and while the vast majority of sailfish are recaptured in the vicinity of their release, there have been several movements in excess of 1,000 nautical miles. These long distance returns indicate that sailfish have the capacity to undertake extensive movements.

Seasonal and spatial distribution data can also be used to infer stock structure. Unfortunately, commercial longline catch data has been problematic because sailfish and spearfish have historically been reported together in ICCAT landing statistics, making it impossible to estimate the distribution of sailfish across the Atlantic. However beginning in 1994, Japan began reporting catches of these two separately. Although he Japanese effort during these years was predominantly concentrated in the eastern Atlantic, preliminary analysis based on the 1994-1996 Japanese database shows that while the catch rate of sailfish is higher in coastal waters, sailfish comprised 10-25% of the total spearfish- sailfish catch past the 30W line (Uozumi, 1997). These results, in contrast with what is known from tagging and morphology studies, call into question the validity of the two-stock hypothesis.

Recently molecular techniques have been useful in delineating the stock structure of other billfish species including striped (Graves and McDowell, 1994), (Graves and McDowell, 1998), (Reeb et al., 2000, Alvarado-Bremer et al., 1996, Rosel and Block, 1996, Alvarrado Bremer et al., 1995), and blue marlin (Graves and McDowell 1998, Buonacccrosi et al., 2001). In the current study, both mtDNA and nuclear DNA markers were used to examine a total of 294 samples of sailfish taken from Brazil, Venezuela and Florida in the western Atlantic and from Senegal in Ghana in the eastern Atlantic.

METHODS

A total of 163 samples was collected from Brazil, Venezuela, and Florida in the western Atlantic and 131 samples from Senegal and Ghana in the eastern Atlantic. DNA was isolated according to the methods of Sambrook et al. (1989) and amplified using primers specific for the D-loop region of mtDNA (Palumbi, 1996). Amplified fragments were subsequently cut with five variable restriction enzymes, Dde I, Hinc II, Hinf I, Nci I, and Sty I, and composite haplotypes constructed. In addition, three microsatellite loci were amplified using the primers MN01, MN08, and MN10 as described in Buonaccorsi et al., 2001. Microsatellite alleles were visualized using a LiCor 4000 automated DNA sequencer and number of repeats scored using the RFLPScan software (Scanalytics, CSPI).

Data were analyzed with samples collected in multiple years at a location held separately and, since there was no evidence of temporal heterogeneity between years at a location, subsequently combined. For mtDNA, both haplotypic diversity (H), nucleotide sequence diversity (p), and a matrix of nucleotide sequence divergences between haplotypes were calculated using the program REAP (McElroy et al., 1991). For both mtDNA and microsatellite data, pairwise genetic distances were calculated and significance was assessed via randomization in the program ARLEQUIN (Schneider et al., 2000). Exact tests of population differentiation were carried out using the methods of Raymond and Rousset (1995) and the level of population differentiation was estimated using the AMOVA algorithm (Excoffier et al., 1992) as implemented in ARLEQUIN ver. 2.0 (Schneider et al., 2000). Conformance of microsatellites to Hardy-Weinburg equillibrium was assessed using the exact test of Guo and Thompson as implemented in ARLEQUIN 2.0.

RESULTS

As with other istiophorid , Atlantic sailfish harbor considerable genetic variation. MtDNA analysis resulted in a total of 42 composite haplotypes (Table 1). Haplotype diversities ranged from 0.84 in Brazil and Ghana to 0.87 in Senegal and nucleotide diversities ranged from 0.020 in Brazil and Ghana to 0.024 in Senegal (Table 1). An analysis of molecular variance (AMOVA) which considered the entire Atlantic as a single group found that 99.95% of the variance was attributable to variation within samples and 0.05% (p=0.400) was attributable to variation between samples. When samples were divided into two groups, eastern and western Atlantic, 100% of the variance was due to variation between populations; no population structure was apparent in the mtDNA data using an AMOVA. Likewise, no significant differences were found between any two samples in pairwise comparisons, indicating that the distribution of genetic variation is homogeneously distributed among samples (Table 2). Exact tests of population differentiation between all between all pairs of populations and between east and west were also non-significant.

Analysis of three hypervariable microsatellite loci gave results concordant with the mtDNA analysis. MN01 had a total of 16 alleles in the Atlantic ranging from 6-31 repeats, MN08 had 13 alleles ranging from 5-28 repeats and MN10 had 17 alleles ranging from 12-35 repeats. Each locus had several alleles present at high frequency as well as several low frequency alleles (Table 3). All populations were found to be in Hardy Weinburg equilibrium at both the locus and haplotype level after correction for multiple tests. An AMOVA considering the Atlantic as a single group found that 100% of the variance could be attributed to variation within populations using both number of different alleles (FST) and the sum of squared size difference (RST) as estimators. Dividing samples into eastern and western groups produced similar results; no significant partitioning of genetic variation was attributable to variance among groups or among populations within groups (Table 4). As with the mtDNA analysis, pairwise comparisons failed to find significant differences between samples.

DISCUSSION

Neither mtDNA nor microsatellite data revealed the presence of stock structure among sailfish samples within the Atlantic Ocean. Failure to find stock structure using genetic techniques does not necessarily mean that there is not structure, just that the method used failed to detect structure. However, similar results have been found in population genetic studies of other Atlantic istiophorid billfishes, most notably the white marlin (Graves and McDowell, 1998). RFLP analysis of the mtDNA region of white marlin showed that although there was sufficient variation to detect stock structure, the distribution of genetic variation was homogeneously distributed among geographic locations. These results contrast with the results of studies involving Indo-Pacific istiophorids. Population genetic studies of both the (Graves and McDowell, 1994) and the Indo-Pacific sailfish (McDowell and Graves unpublished data) were found to have levels of genetic variation similar to their Atlantic counterparts, however both were found to comprise distinct genetic stocks. Since, there is no obvious biological reason why stock structure should exist in the Pacific but not in the Atlantic for these species pairs, the most likely explanation appears to be the relative size of the Atlantic Ocean as compared to the Indo-Pacific Ocean. These results combined with the 1994-1996 Japanese landing statistics demonstrating a continuous distribution of sailfish across the Atlantic Ocean suggest that the assumption that sailfish are divided into eastern and western stocks should be reevaluated.

REFERENCES

ALVARADO BREMER, J.R., J. Mejuto, T.W. Greig, B. Ely 1996. Global population structure of the swordfish (Xiphias gladius L.) as revealed by analysis of the mitochondrial DNA control region. J. Exp. Mar. Biol. Ecol. 197(2): 295-310.

ALVARADO BREMER, J.R, A.J. Baker, J. Mejuto 1995. Mitochondrial DNA control region sequences indicate extensive mixing of swordfish (Xiphias gladius) populations in the Atlantic Ocean. Can. J. . Aquat. Sci. 52(8) 1720-1732.

BUONACCORSI, V.P., J.R. McDowell, and J.E. Graves 2001. Reconciling patterns of inter-ocean molecular variance from four classes of molecular markers in blue marlin ( nigricans). Mol. Ecol. 10: 1179-1196.

EXCOFFIER, L., P.E. Smouse and J.M. Quattro 1992. Analysis of molecular variance inferred from metric distances among mtDNA haplotypes-Application to human mitochondrial DNA restriction data. Genetics 131: 479-491.

GRAVES, J.E. and J.R. McDowell 1994. Genetic analysis of striped marlin ( audax) population structure in the Pacific Ocean. Can. J. Fish. Aquatic Sci. 51: 1762-1768

GRAVES, J.E. and J.R. McDowell 1998. Population Genetic Structure of Atlantic Istiophorid Billfishes. Report of the third ICCAT billfish workshop. Inter. Comm. Cons. Atl. Tunas. Coll. Vol. Sci. Pap. XLVII: 329-341.

NAKAMURA, I. 1985 5: Billfishes of the world: An annotated and illustrated catalogue of , , spearfishes, and known to date. FAO Species Catalogue, 5, 1-68.

PALUMBI, S.R. 1996. Nucleic acids II: The polymerase chain reaction. IN: D.M. Hillis, C. Moritz, and B.K. Mable, eds., Molecular systematics, 2nd edition. Sinauer Press, Sunderland, MA

PRINCE, E., M.A. Ortiz, D. Rosenthal, A. Venizelos, and K. Davy 2001. An update of the tag release and recapture files for Atlantic Istiophoridae. Inter. Comm. Cons. Atl. Tunas. Coll. Vol. Sci. Pap. 53: 198-204.

REEB, CA, L. Arcangeli, B.A. Block 2000. Structure and migration corridors in Pacific populations of the Swordfish Xiphius gladius, as inferred through analyses of mitochondrial DNA. Mar. Biol., 136(6): 1123-1131.

ROSEL, P.E. B.A. Block 1996. Mitochondrial control region variability and global population structure in the swordfish, Xiphias gladius. Mar. Biol., 125(1): 11-22.

SAMBROOK, J., E.F. Fritsch, and T. Maniatis 1989. Molecular cloning: a laboratory manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

SCHNEIDER, S., D. Rosseli, L. Excoffier 2000. ARLEQUIN A software for population genetics data analysis, Ver. 2.0 Geneva, Switzerland.

UOZUMI, Y. 1997. Distribution of sailfish and in the Atlantic Ocean during 1994- 1996 based on the logbook database of the Japanese longline fishery. Inter. Comm. Cons. Atl. Tunas. Coll. Vol. Sci. Pap. 52: 1-6. Table 1. Sailfish composite mtDNA haplotypes.

WA WA WA WA EA EA ID Hap CAR FLA BRA VEN SEN GHA Count SAIL01 AAABA 4 7 14 2 10 23 60 SAIL02 AABAA 5 14 32 3 9 26 89 SAIL03 AAAAA 1 4 9 2 3 12 31 SAIL05 AAABC 0 0 0 1 0 2 3 SAIL06 CBDAC 0 2 7 2 0 6 17 SAIL07 AABBA 0 1 4 1 1 2 9 SAIL08 BABAA 0 0 5 0 2 6 13 SAIL09 CBAAC 0 2 1 0 3 2 8 SAIL10 AABAE 2 0 3 1 0 0 6 SAIL12 ACBAA 0 2 1 0 0 2 5 SAIL13 ADBAA 0 0 2 1 1 1 5 SAIL15 EABAA 0 0 2 0 1 0 3 SAIL16 AABAC 0 1 2 0 0 1 4 SAIL17 ABBAA 0 0 0 0 1 1 2 SAIL19 ABAAC 0 1 0 0 0 0 1 SAIL22 ABAAA 0 2 0 0 0 0 2 SAIL23 ABDAB 1 0 1 0 0 0 2 SAIL24 AAEBA 0 0 2 0 0 0 2 SAIL25 AACAA 0 2 1 0 0 0 3 SAIL28 ABCAA 0 0 1 1 0 0 2 SAIL29 CABAA 0 1 1 0 0 0 2 SAIL30 CACAC 0 0 0 0 2 0 2 SAIL31 CBCAA 0 1 1 0 0 0 2 SAIL32 CBCAC 0 0 1 0 1 0 2 SAIL33 CBDAA 0 0 0 0 0 1 1 SAIL36 AACAC 0 0 0 0 1 0 1 SAIL37 AABAB 0 0 0 0 0 1 1 SAIL38 AADAA 0 0 1 0 0 0 1 SAIL40 CAAAA 0 0 1 0 0 0 1 SAIL41 GABAA 0 0 0 0 0 1 1 SAIL42 FABAA 0 0 0 0 1 0 1 SAIL43 EBCAC 0 0 0 0 1 0 1 SAIL44 CEDAC 0 0 0 0 0 1 1 SAIL45 CCDAC 0 0 0 0 0 1 1 SAIL46 CCAAA 0 0 0 0 0 1 1 SAIL49 ADABA 0 0 0 0 0 1 1 SAIL50 ADAAB 0 0 0 1 0 0 1 SAIL51 ABDBA 0 0 1 0 0 0 1 SAIL52 IAEBA 0 1 0 0 0 0 1 SAIL53 ABCAC 0 1 0 0 0 0 1 SAIL54 CBAAA 1 0 0 0 0 0 1 SAIL56 AAFAA 0 0 0 0 0 1 1 Total 14 42 93 15 37 92 293

Haplotype Diversity 0.85 0.85 0.84 0.94 0.87 0.84 Nucleotide Sequence 0.017 0.021 0.020 0.022 0.024 0.020 Diversity

Table 2. Hierarchical analysis of molecular variance (AMOVA) of sailfish mtDNA RFLP data. Distance method used was pairwise differences with 10,000 permutations.

Structure Tested Variance Components Percent Variation p-value Atlantic Ocean Among populations 0.00054 Va FST: 0.00049 0.05 0.40050 Within populations 1.10973 Vb 99.95

East/West Atlantic Among groups 0.0274 Va FCT: 0.00246 0.25 0.33724 Among pops within groups -0.00557 Vb FSC: -0.00501 -0.50 0.66764 Within populations 1.11648 Vc FST: -0.00246 100.25 0.67840

Table 3. Descriptive statistics for microsatellite data.

GATA01 POP Gene Copies No Alleles Repeat Range Gene Diversity FLA 66 10 6-15 0.8406 +/- 0.0189 BRA 232 11 7-17 0.8652 +/- 0.0077 VEN 34 10 7-18 0.8431 +/- 0.0314 SEN 74 9 7-15 0.8193 +/- 0.0210 GHA 182 11 6-16 0.8245 +/- 0.0127 CAR 30 10 7-30 0.8782 +/- 0.0353

GATA08 POP Gene Copies No Alleles Repeat Range Gene Diversity FLA 70 4 5-10 0.6791 +/- 0.0309 BRA 232 6 5-11 0.5832 +/- 0.0283 VEN 28 4 5-10 0.6402 +/- 0.0526 SEN 68 6 5-10 0.6765 +/- 0.0429 GHA 166 10 5-19 0.7352 +/- 0.0242 CAR 30 7 5-28 0.7494 +/- 0.0525

GATA10 POP Gene Copies No Alleles Repeat Range Gene Diversity FLA 72 11 15-26 0.8103 +/- 0.0258 BRA 234 15 15-35 0.8259 +/- 0.0127 VEN 36 7 15-25 0.7810 +/- 0.0455 SEN 56 8 15-24 0.8104 +/- 0.0308 GHA 174 12 15-26 0.7853 +/- 0.0171 CAR 28 11 13- 28 0.9048 +/- 0.0281

Table 4. Hierarchical analysis of molecular variance (AMOVA) of sailfish microsatellite data. Distance method used was number of different alleles (FST) with 10,000 permutations. Sum of squared size difference (RST) gave similar results.

Structure Tested Variance Components Percent Variation p-value Atlantic Ocean Among populations -0.00455 Va FST: -0.00429 -0.43 0.97099 Within populations 1.06441 Vb 100.43

East/West Atlantic Among groups 0.00615 Va FCT: -0.00829 0.58 0.10119 Among pops within groups -0.00875 Vb FSC: 0.00579 -0.82 0.99891 Within populations 1.06441 Vc FST: -0.00245 100.24 0.97356