MATERIALS AND METHODS

Samples Collection
Totally 149 samples of adult Sander lucioperca were caught from 3 regions including 50 samples from the Coasts of Talesh, 50 samples from Anzali Wetland and 49 samples from the Coasts of Chaboksar in 2007-2008 from Southwest Caspian Sea in Iran (Fig. 1). For each sample, 2-3 g dorsal fin tissues was collected and conserved in absolute alcohol for subsequent DNA extraction and amplification.

DNA Extraction
Total genomic DNA was extracted from fin tissue using proteinase-K digestion, phenol: chloroform; isoamylalchol precipitation as described by Pourkazemi et al. (1999). The quality and concentration of DNA were assessed using agarose gel electrophoresis and spectrophotometry (model Nano Drop, ND-1000). Finally, DNA was stored at -20°C until use.

PCR Profiles and Primer Sequences
Nuclear DNA was amplified using 13 microsatellites primers (YP13, YP17, YP41, YP60, YP68, YP78, YP110, YP111, YP113, Pfla L2, Pfla L3, Pfla L8, Pfla L9) designed for Perca flavescens (Li et al., 2007; Leclerc et al., 2000) (Table 1).

Fig. 1:	Map showing sampling regions of three populations of Sander lucioperca: Talesh Coasts (▲), Anzali Wetland (■) and Chaboksar Coasts (●)

Table 1:	Loci, repeat motif, primers sequence, gene bank number and primer source used at present study

The Polymerase Chain Reaction (PCR) conditions, especially annealing temperatures, were optimized for the microsatellite loci as necessary to produce scorable amplification products. Polymerase chain reaction was performed in a 20 μL reaction volume containing 100-150 ng of template DNA, 10 pmol of each primer, 200 μM each of the dNTPs, 1U of Taq DNA polymerase (Cinnagen, Tehran, Iran), 1.5 mM MgCl₂ and 1x PCR buffer. The temperature profile consisted of 2 min initial denaturation at 94°C, followed by 35 cycles of 30 sec denaturation at 94°C, 30 sec annealing at a locus-specific temperature (Table 2), 30 sec extention at 72°C and a final 5 min extension at 72°C for the first 9 microsatellite primers (YP13, YP17, YP41, YP60, YP68, YP78, YP110, YP111 and YP113). For the rest microsatellite primers (Pfla L2, Pfla L3, Pfla L8 and Pfla L9) the temperature profile consisted of 3 min initial denaturation at 96°C, 30 sec annealing at a locus-specific temperature (Table 2), 1 min extention at 72°C and a final 5 min extension at 72°C. Polymerase chain reaction products were separated on 6% polyacrylamide gels and stained by silver nitrate. Alleles were sized using BioCapt software and each gel contained an allelic ladder (100 bp) to assist in consistent scoring of alleles.

Table 2:	Polymorphic amplified locus, allele size (bp) and annealing temperature on Sander lucioperca

DNA Analysis
Allelic frequency, observed and expected hetrozigosities, genetic distance (Nei, 1972) and genetic identity (Nei, 1972) were computed in GenAlex 6.0 software (Peakall and Smouse, 2005). This package was also used to calculate F_ST and R_ST, Nm, Hardy-Weinberg (HW) tests of equilibrium and AMOVA (Analysis or Molecular Variance).

RESULTS

Amplification and Banding Patterns
Out of 13 sets of microsatellite primers, two sets (YP78 and YP113) have not shown any flanking sites on pikeperch genome. Eleven sets of primers were successfully amplified where 5 sets (YP17, YP41, YP68, YP111 and Pfla L2) showed monomorphic pattern in all samples. Therefore, totally 6 sets of primers produce polymorphic bands. Loci YP13, YP60 and Pfla L8 had the highest numbers of alleles (4), while the locus YP110 had the lowest (2).

Genetic Variation Within Sampling Regions
The average number of alleles found per locus in Talesh Coasts, Anzali Wetland and Chaboksar Coasts samples were 3.5, 2.7 and 3, respectively. The average observed hetrozigosities in these 3 regions were 0.540, 0.570 and 0.517, respectively (Table 3). The observed heterozigosity of all 3 regions at YP13 and Pfla L9 were significantly higher than the corresponding expected hetrozygosity (p<0.01). The differences between all three sampling area were not statistically significant (p>0.05), neither for the average number of alleles nor for the observed heterozygosities.

Significant deviations from Hardy-Weinberg equilibrium at the locus level are shown in Table 3. Talesh Coasts, Anzali Wetland and Chaboksar Coasts regions deviated at 4, 3 and 3 loci, respectively, mostly due to the deficiency of heterozygosities.

Genetic Variation Among Sampling Regions
Genetic distance calculated between each pair of collections ranged from 0.035 (between Talesh and Chaboksar Coasts) to 0.068 (between Anzali Wetland and Chaboksar Coasts, Table 4). The highest range of genetic difference were observed between Chaboksar Coasts and Anzali Wetland (F_ST = 0.051, R_ST = 0.088, p≤0.01) and the lowest between Talesh and Chaboksar Coasts (F_ST = 0.021, R_ST = 0.036, p≤0.01). However, F_ST and R_ST estimates often differ in a pronounced manner (Balloux and Lugon- Moulin, 2002). Values of pairwise R_ST among samples were consistently much higher (as much as an order of magnitude) than equivalent F_ST values (Table 5).

Table 3:	Variability of six microsatellite loci in three populations of Sander lucioperca from Caspian Sea
*Significant at p<0.05, A: Number of alleles; Ho: Observed heterozigosity; He: Expected heterozigosity; P: p-value of χ2 tests for Hardy-Weinberg equilibrium; ns: Not signifocent

Table 4:	Pairwise population of genetic distance (below diagonal) a nd genetic identity (above diagonal) (Nei, 1972) detected at 6 loci in pikeperch samples

Table 5:	Pairwise estimates of genetic differentiation detected at 6 loci in pikeperch samples, using unbiased RST (below diagonal) and FST values (above diagonal)
Probabilities of RST or FST Determined by AMOVA tests at p≤ 0.01

DISCUSSION

The long-term persistence of an endangered fish species can be investigated by allelic diversity, gene diversity, effective population size and population structure (Yue et al., 2004). Despite the importance of pikeperch as a highly economic fish in Iran, natural populations of this fish are declined. Information about these populations is pivotal for their conservation and sustainable use. Unfortunately, the knowledge on the molecular population genetic structure of this species has not conducted yet.

In this study we have employed 6 polymorphic microsatellite loci to assess the genetic relationship among populations of pikeperch from three regions in Iran along the Southwest Coasts of Caspian Sea. According to the results, all of three sampling regions had low number of alleles but Anzali Wetland samples showed the lowest number of alleles (2.7 in average). From the 6 primers used in this study, the first 3 primers have not been used for other pikeperch populations yet. But these primers have shown high number of alleles in the studied species (Li et al., 2007). So, these primers were used in this study because of their polymorphic character. The rest of primers were used in a few individuals (6-8) of Sander lucioperca (Leclerc et al., 2000). The results showed that the average number of alleles per locus similar to present study was low, but the number of individuals was insufficient for population genetic statistical analysis and the results were not comparable with present study. In another study of genetic structure of pikeperch in Rhone delta (Poulet et al., 2009), 5 alleles were produced in Pfla L3 locus, which is higher than present produced alleles (3 in average) for this locus. As a conclusion, the low number of alleles per locus in the present study in comparison to earlier studies for used primers indicates the occurrence of a bottleneck effect in the progeny of selective breeding stocks in Aras dam that are transferred to Caspian Sea and Anzali Wetland. This may also occure due to the founder effect in the Caspian Sea and Anzali Wetland pikeperch populations.

The value of F_ST is a useful measure of genetic differentiation among populations (Peakall and Smouse, 2005). At present study the F_ST value in all 3 sampling regions was low but significant (p<0.01), suggesting that the 3 populations are genetically differentiated and don’t represent a single panmictic population. The lowest of F_ST value on AMOVA was between Talesh and Chaboksar Coasts populations with the high gene flow. However, there is no information about the main origin of the Talesh and Chaboksar Coasts populations. Perhaps these populations are related to other rivers (e.g., Ural, Volga, Kura, and Sefid Roud) or neighboring Coasts.

The genetic distance between these 3 populations were in the range of 0.035-0.068. Shaklee et al. (1982) and Thorpe and Sol-Cava (1994) showed that genetic distance values (Nei, 1972) for conspecific populations averaged 0.05 (range: 0.002-0.07) and for congeneric species averaged 0.30 (range: 0.03-0.61). The distance value obtained in the present study falls within the average value of conspecifics, which indicate that the genetic difference among the studied populations is not pronounced. Regarding the fishery returning, it is possible that the caught samples are a combination of the various generation and different birth places which have formed a gathering to feed.

Although significant deviations from Hardy-Weinberg equilibrium were found at more loci in the Talesh Coast population than in the Anzali Wetland and Chaboksar Coast populations, there were no significant differences in the average expected and observed heterozygosities among all three populations (p>0.05). The significant deviations from Hardy-Weinberg equilibrium could be explained either by sample bias or the present of null alleles. In the presence of null alleles heterozygotes possessing a null allele could be erroneously recorded as homozygotes for the variant allele leading to a deficiency of heterozygotes in the respective population.

The data generated in this study showed that there are 3 different populations of pikeperch in the studied regions of Southwest Caspian Sea. This information can be applied for future genetic improvement by selective breeding and to design suitable management guidelines for the genetic materials. However, in order to have better conservational policy and restocking programs, further studies are recommended on determining the other different populations of this important species in other regions of Caspian Sea.

ACKNOWLEDGMENTS

This research was performed in the Molecular Genetic Laboratory of the Dr. Dadman International Sturgeon Research Institute, Rasht, Iran. The authors express their sincere gratitude to all staffmembers especially, Mr. Nuruzfashkhami, head of Genetic Department, Mr. Hassanzadeh, Mr. Chakmedouz, Mr. Bagheri and Mrs. Haghighi for their great co-operation and support.