32 



Fishery Bulletin 94(1), 1996 



of anchovies after an allozyme study that covered a 

 large area of the western Mediterranean. In contrast, 

 Spanakis et al. (1989) were able to distinguish stocks 

 from the Ionian and Aegean seas by morphometries 

 and allozyme electrophoresis, noting that a degree of 

 migration between these areas was probable. 



The application of DNA technology to stock dis- 

 crimination in E. encrasicolus is an obvious step, 

 given the limited success of allozyme analysis, and 

 mitochondrial DNA (mtDNA) has several distinctive 

 properties which commend it to such investigations. 

 Its inheritance is essentially clonal, reducing the ef- 

 fective population size to 1/4 of that for nuclear genes; 

 mtDNA variation is hence more susceptible to the 

 effects of stochastic processes (genetic drift and par- 

 tial extinctions), and interpopulation variation ac- 

 cumulates more rapidly. This process is perhaps ac- 

 celerated by the higher mutation rates often reported 

 for mtDNA (Brown et al., 1979). Several studies have 

 demonstrated stock differentiation in teleosts using 

 mtDNA where none was detected at the allozyme 

 level (see Ward and Grewe, 1994), although Ward 

 and Grewe ( 1994 ) pointed out that the reverse is also 

 sometimes true. It should be noted, however, that, 

 since mtDNA acts as a single locus, the ability to 

 screen a number of independent loci is an advantage 

 of allozyme and nuclear DNA (nDNA) methods. 



Despite their economic importance, relatively few 

 studies of mtDNA variation in clupeoid fish have been 

 published. Tringali and Wilson (1993) reported no 

 significant spatial variation in the distribution of 

 Sardinella aurita Val. mtDNA haplotypes in the east- 

 ern Gulf of Mexico but found that these fish were 

 distinct from a sample taken in waters off southern 

 Brazil. Kornfield and Bogdanowicz ( 1987) described 

 geographic heterogeneity in Atlantic herring, Clupea 

 harengus, although this variation has subsequently 

 been shown to be nonsignificant (Roff and Bentzen, 

 1989). Other clupeoid species studied include Ameri- 

 can shad, Alosa sapidissima (Bentzen et al., 1988; 

 Nolan et al., 1991; Chapman et al., 1994) and men- 

 haden, Brevoortia tyrannus and Brevoortia patronus 

 (Bowen and Avise, 1990), three species whose life 

 history strategies include estuarine or freshwater 

 components. The only published study on E. 

 encrasicolus mtDNA variation concentrated on the 

 occurrence of heteroplasmy (multiple forms of 

 mtDNA within an individual), attributed to "pater- 

 nal leakage" (Magoulas and Zouros, 1993). 



Until recently, analysis of mtDNA involved labori- 

 ous extraction protocols (e.g. Lansman et al., 1981) 

 or hybridization of mtDNA probes with Southern- 

 blotted digests of total DNA, but the revolution in 

 molecular biology due to the discovery of polymerase 

 chain reaction (PCR) amplification of DNA (Saiki et 



al., 1988) has largely removed these obstacles. Re- 

 searchers may now apply the technique to concen- 

 trate efforts on mtDNA regions of particular inter- 

 est, whether for species identification or for the de- 

 tection of intraspecific variation and stock markers 

 (Chow et al., 1993; Cronin et al., 1993; Chapman et 

 al., 1994). 



The genes of the NADH dehydrogenase complex 

 (ND genes) have been the subject of a number of re- 

 cent investigations, usually exhibiting sufficient 

 variation to provide useful genetic markers (Cronin 

 et al, 1993; Hall, 1993; Park et al., 1993). In this study 

 we have investigated the utility of restriction frag- 

 ment length polymorphism (RFLP) analysis of PCR- 

 amplified mtDNA ND genes in the study of popula- 

 tion structure in anchovies from the Adriatic Sea and 

 surrounding waters. Our sampling strategy has en- 

 abled the comparison of results from this rapid, state- 

 of-the-art approach with those previously obtained 

 by allozyme analysis. 



Materials and methods 



The anchovy samples used for analysis were from 

 the ports of Trieste, Ancona, and Vieste in the Adri- 

 atic Sea, from the Sicilian Channel, and from the 

 Ionian, Tyrrhenian, and Aegean seas (Fig. 1). Each 

 sample comprised 20 individuals. Fish were captured 

 by commercial vessels with technology typical of the 

 fishing grounds, i.e. light seine (lampara) and pair 

 trawl (volante), and immediately placed on solid CO., 

 (dry ice). Fish were later stored at -80°C until labo- 

 ratory analysis. 



DNA extraction 



Further details on the preparation of solutions may 

 be found in Sambrook et al. (1989). Apiece of muscle 

 tissue approximately 5x3x3 mm was added to 300 

 pL extraction solution (0.1 M Tris, 0.01 M EDTA, 0.1 

 M NaCl, 29c SDS, proteinase K 0.8 rngmL" 1 , pH 8.0), 

 mixed, and incubated at 55°C for 1 hour. After diges- 

 tion and phenol extraction, 2.5 volumes of ice-cold 

 99*# ethanol were added, the tube was inverted sev- 

 eral times, and spun for 5 minutes in a micro- 

 centrifuge to pellet the precipitated DNA. The pellet 

 was washed with 709c ethanol, air-dried for 30 min- 

 utes, and dissolved in 100 pL TE buffer ( 0.01 M Tris, 

 0.02 M EDTA, pH 7.5). 



PCR amplification of mtDNA 



The main set of PCR primers employed were "uni- 

 versal" vertebrate sequences which amplified a 2.5 



