538 



Fishery Bulletin 100(3) 



and banding patterns differed by as little as 15 base pairs, 

 making alternate patterns difficult to distinguish. Innes et 

 al. (1998) were able to discriminate among seven species 

 of billfish found in Australian waters with RFLP analysis 

 of a 1400-bp region of the mtDNA control region (D-loop). 

 Their analysis, which employed four restriction enzymes, 

 revealed relatively high levels of intraspecific variation of 

 the diagnostic characters within some species, and there 

 was some overlap of banding patterns between species. 

 In neither study was an independent nuclear marker 

 developed to corroborate specific identifications based on 

 analyses of mtDNA. 



In this article we present a molecular key to the iden- 

 tification of istiophorid and xiphiid billfishes using RFLP 

 analyses of independent mitochondrial and nuclear DNA 

 regions. We demonstrate low intraspecific variation of the 

 characters within large collections of individuals sampled 

 from throughout each species' range and show the utility 

 of the markers for the identification of fillets and early life 

 history stages. 



Materials and methods 



Collections of striped marlin (Tetraptiiriis audax), white 

 marlin (Tetraptu/'us albidus). blue marlin iMakaira nigri- 

 cans), and sailfish ilstiophortis platypterus) were available 

 from previous analyses of stock structure (Graves and 

 McDowell 1994, 1995; Graves, 1998), and individuals from 

 locations throughout each species' range were selected 

 for the present study (Table 1). These DNA samples con- 

 sisted of the nuclear and mitochondrial bands resulting 

 from mtDNA purifications with the equilibrium den- 

 sity gradient centrifugation protocols of Lansman et al. 

 (1981). Samples of black marlin iMakaira indica), longbill 

 spearfish (Tetrapturus pfluergeri), shortbill spearfish (Tet- 

 rapturus angustirostris). and swordfish iXiphias gladius), 

 were obtained from recreational and commercial fisher- 

 men (Table 1) and consisted of either frozen heart tissue 

 or white muscle tissue preserved in DMSO storage buffer 

 (Seutin et al., 1991). DNA was extracted from these tissues 

 following the protocols of Winnepenninckx et al. (1993). 



Evaluation of candidate mitochondrial and nuclear loci 

 involved a two-step process. The first was to ensure consis- 

 tent amplification by the polymerase chain reaction (PCR) 

 of a similar-size product across all taxa. The second step 

 was to screen those loci that successfully amplified across 

 all billfish species with a panel of restriction endonucleas- 

 es to identify enzymes that discriminated among species 

 and revealed limited intraspecific variation. 



Several candidate mitochondrial and nuclear gene re- 

 gions were amplified by PCR (Table 2). The 25 ijL PCR 

 reactions consisted of 0.25 luL template DNA, 2.5 ^L lOX 

 PCR buffer plus magnesium, 0.5 ;iL dNTP mix, 0.25 pL 

 forward primer, 0.25 i.tL reverse primer, 0.125 pL Tag DNA 

 polymerase, and 21.125 pL PCR grade water Primers 

 were ordered from either Life Technologies (Gaithersburg, 

 MD) or Genosys Biotechnologies Inc. (The Woodlands, TX), 

 and PCR reactions were carried out in an MJ Research 

 Corporation PTC-200 Peltier thermal cycler (Watertown, 



MA) by using the Life Technologies PCR reagent system 

 (Gaithersburg, MD). Initial screening demonstrated that 

 the mitochondrial ND4 gene region and the nuclear 

 MN32-2 locus produced the most reliable amplifications 

 across taxa and PCR conditions were optimized for these 

 loci. The cycling parameters for the ND4 gene region were 

 an initial denaturation at 95°C for 5 min., followed by 35 

 cycles of 94°C for 1 min., 47°C for 1 min., 65°C for 3 min., 

 and a final extension at 65°C for 7 min. Amplification of 

 MN32-2 proceeded with an initial denaturation at 95°C 

 for 5 min., 40 cycles of 94°C for 1 min., 57°C for 1 min., 

 65°C for 3 min., and a final extension at 72°C for 7 min. 

 Amplified products were held at 4°C until use. The size of 

 each amplification product was determined on a 1% aga- 

 rose gel run in TBE (45 mM Tris, 45 mM boric acid, I mM 

 EDTA) at 100 volts for 1 hour ND4 amplification resulted 

 in a product of approximately 1.7 kb and MN32-2 amplifi- 

 cation resulted in a product of approximately 1.2 kb. 



Amplified products were screened with a panel of re- 

 striction endonucleases to identify those that discrimi- 

 nated among species, and revealed a minimum level of 

 variation within species. All enzymes were purchased 

 from Gibco/BRL Life Technologies Inc. (Bethesda, MD) 

 with the exception of Banl, which was purchased from 

 Promega (Madison, WI). All were used according to the 

 manufacturers' instructions. 



Restriction fragments were separated on 2.5'/r hori- 

 zontal agarose gels made from 1.25% UltraPure agarose 

 (Life Technologies Inc.. Bethesda, MD) and 1.25% NuSeive 



