Microsatellite Analysis of Six Populations of Chrysichthys nigrodigitatus from Nigeria

Aim: To investigate the patterns and levels of genetic polymorphism and population structures of wild C. nigrodigitatus using the microsatellite DNA in Nigeria. Place and Duration of Study: in Nigeria between 2008 and 2009, and Laboratory of Mariculture, Ocean University of China, 2009-2010. Methodology: A total of 93 individuals of Chryscichthys nigrodigitatus obtained from 6 sites in Nigeria were used for the study. DNA extracted from alcohol preserved muscle tissue was amplified by PCR. Amplified products were detected using the silver staining technique to visualize bands. Results: The four microsatellite loci indicate high genetic variation in all 6 populations of the species with the number of alleles and H O varying from 2-10 and 0.800-1.00 per locus, respectively. The NJ tree revealed a marginal genetic differentiation between two clades, which were not well supported. Significant genetic differences were detected between most samples. Freshwater or brackishwater habitat, limited long-distance dispersal of the adult and juveniles may account for breakdown of gene flow. Conclusion: C. nigrodigitatus intra population genetic variability moderate inter


INTRODUCTION
Chrysichthys nigrodigitatus is a euryhaline fish of tropical Africa which supports thriving commercial fisheries with great potentials for aquaculture in West Africa. It occurs in a variety of fresh and brackish waters habitats such as rivers, mangrove swamps, lakes, and estuaries and low salinity coastal areas [1][2][3]. The species is characterized by spawning migration undertaken in the rainy season from more saline brackishwaters to freshwaters where spawning occurs. In the reverse, the juveniles follow the flood water back to the saline environment to feed and grow [1]. However, the wild population of C. nigrodigitaus is observed to be declining due to destructive fishing methods, environmental pollution and overfishing. Most water bodies in the Niger Delta are polluted from petroleum extraction and allied activities, urbanization and agriculture. Fishermen target migrating gravid females thereby disrupting the reproductive and recruitment cycle. Furthermore, culture of the species still relies on the capture of fry from the wild for stocking. All these may be interacting to decrease the genetic diversity of the populations. As an important commercial and aquaculture species understanding the level of genetic diversity and patterns of population genetic structure are of paramount significance.
There are several molecular markers for assessing genetic diversity. Among all the types of molecular markers, the microsatellites DNA are widely relied on for the analysis of genetic diversity and population structure in fish. Microsatellites DNA are characterized by a core sequence that consists of a number of tandemly repeated units with a length of 1-6 base pairs. Microsatellites would potentially produce higher values of polymorphism than mtDNA sequences, which mutate at a higher speed than mtDNA sequences [4].
The population genetic structure and genetic diversity of C. nigrodigitatus obtained from the Niger Delta was analyzed using four microsatellite markers, in order to address the following questions: What is the current level of genetic diversity and how is the population structured? Is the genetic diversity higher or lower in the Niger Delta populations when compared with populations from other regions in Nigeria?
The present study aims to investigate the patterns and levels of genetic polymorphism and population structures of wild C. nigrodigitatus using the microsatellite DNA to [1] estimate the level of genetic diversity in the C. nigrodigitatus populations in Nigeria, [2] calculate the distribution of variability within a population and among populations and [3] examine the genetic relationship between the populations.

Sample Collection
Chrysichthys nigrodigitatus specimen for this study were collected between December 2008 and July, 2009) from 8 sites in Nigeria (Fig. 1). These sites were 3 Niger Delta locations, namely; Benin River at Koko (KO), Warri River (WA) and Buguma Creek (BU). The remaining populations were obtained from Lagos lagoon (LA) and the River Niger at the downstream (LN) and upstream (UN), respectively. Muscle tissue was excised from each individual fish and preserved in 95% alcohol until DNA extraction. Total genomic DNA was extracted using standard phenol-chloroform method [5]. Genomic DNA was suspended in 100µl distilled water.

Microsatellite Genotyping
Four microsatellite loci CN13, CN25, CN45 and CN67 listed in Table 1  Amplified products were checked for yield on an agarose minigel before loading, along with size standards internal to every lane, onto a 6% denaturing polyacrylamide gel ran for 2h at 50°C using Sequi-Gen GT Sequencing cell (Bio-Rad, USA) and finally detected using the silver staining technique to visualize bands. A 50 bp DNA ladder was used as a reference marker to enable the determination of allele sizes.

Scoring Microsatellite Markers
Data collation was done manually. All microsatellite loci were scored by visually. Microsatellite loci amplified by PCR are characterized by a distinct banding pattern consisting of the main alleles together with fainter stutter bands. Genotypes from scored loci were entered into the Excel spreadsheets.

Statistical Analyses
Descriptive statistics of microsatellite loci used to quantify genetic diversity included observed

Scoring Microsatellite Markers
Data collation was done manually. All microsatellite loci were scored by visually. Microsatellite loci amplified by PCR are characterized by a distinct banding pattern consisting of the main alleles together with fainter tutter bands. Genotypes from scored loci were entered into the Excel spreadsheets.
Descriptive statistics of microsatellite loci used to quantify genetic diversity included observed heterozygosity (H O ), expected heterozygosity (H E ), number of alleles, polymorphic information content (PIC) and Wright's F-statistics. H E for each population across the loci and those for each locus across populations were calculated using the GENEPOP programme Version 3.3 [7], according to We Cockerham's F IS [8]. Expected heterozygosity is the heterozygosity that would be obtained given the allele frequencies at a particular locus under conditions of HWE. The discrepancies between  ), expected heterozygosity ), number of alleles, polymorphic information statistics. H O and for each population across the loci and those for each locus across populations were calculated using the GENEPOP programme 3.3 [7], according to Weir and [8]. Expected heterozygosity is the heterozygosity that would be obtained given the allele frequencies at a particular locus under The discrepancies between Allelic richness (AR) was also calculated using FSTAT v.2.9.3 [9]. Allelic richness was computed to allow for a comparison among samples of different sizes [10].

C. nigrodigitatus
Genic differentiation between populations was estimated using unbiased estimate of P-value of the probability test (or Fisher's exact test) as described by Raymond and Rousset [7]. The genetic differentiation among populations was also measured by F ST and analyzed by a hierarchical analysis of genetic diversity (AMOVA) implemented in ARLEQUIN version 3.11 [11]. We also conducted the AMOVA analysis with two groups representing the freshwater group and brackishwater group. The division of groups is shown in Table 4.
Genetic distances of Cavalli-Sforza and Edwards [12] between populations were calculated using the programme POPULATIONS 1.2.30 [13]. A hybrid neighbour-joining tree was constructed with the software MEGA [14]. The tree topology was based on Da genetic distance [15] with statistical support from 1000 bootstrapped data sets for phylogenetic relationships [16].
Multilocus estimates of the effective number of migrants (Nm) between populations were calculated using the private allele method of Slatkin [17] and were corrected for sample size as given in Barton and Slatkin [18].

RESULTS
Genetic diversity at four microsatellite loci was determined using the primers CN67, CN45, CN25 and CN13 reported as useful for amplification studies in C. nigrodigitatus [6]. The four microsatellite loci used in this study exhibited high polymorphism in the 6 populations under investigation (Table 2). Across the populations, the number of alleles ranged from 2 to 10 per locus. The locus CN67 had the highest number of alleles [10] and the locus CN25 had the least number of alleles [2]. The PIC values were higher than 0.5 except for population LN at CN25 locus.
The highest average number of alleles was recorded in KO and LN populations (7.5), followed by BU (7.25), LA (7.0), UN (6.75) and WA (6.25 The index of inbreeding F IS was negative for all populations across loci except for KO at locus CN13 (Table 2). Negative F IS imply there is an excess of heterozygotes relative to Hardy-Weinberg expectations. The exact test for genic differentiation revealed highly significant differentiation between the studied populations at three loci at 0.05% level (CN25, P<0.001; CN67, P<0.001 and CN13, P<0.001). AMOVA shows that population specific F ST indices (Table 2) varied from 0.0281 in BU to 0.0357 in LA, indicating that 2.81 to 3.57% of total genetic variation comes from intrapopulational variation and 97.19-96.43% of the variations are due to variation within populations. The mean F ST was estimated at 0.0309. Thus, about 97% of total genetic variation resides within each population. The AMOVA analysis based on freshwater and brackishwater group (Table 3) showed that 7.37% of the genetic diversity was found between freshwater and brackishwater groups (P<0.05). A small (3.55%) but significant (P<0.01) amount of genetic diversity was found among populations within groups. The AMOVA analysis also showed that a large and significant genetic differentiation (89.08%, P<0.01) accounts for individuals within populations.  The neighbour-joining tree clearly showed a marginal genetic differentiation between two groups (Fig. 2). This clade is differentiated by a bootstrap support value of 69.

DISCUSSION
A total of eight populations were screened for microsatellite variation. However, two populations, Cross River Basin (CR) and Shiroro Lake (SH) were excluded from analyses because most individuals were monomorphic for all the loci and in others the bands were poor due to degraded DNA. Earlier mtDNA and AFLP analysis of the two populations showed they possess very low diversity [19]. The Niger Delta region of Nigeria is one of the ten most important wetlands in the world. Oil spill and discharges from many other industries is common in the area resulting in environmental contamination. Environmental contamination result in loss of biodiversity and destruction of habitats. In line with the aim of this study, that is, to determine whether genetic diversity in the Niger Delta populations of C. nigrodigitatus have been reduced as a result of long-term environmental pollution and degradation, microsatellite analysis was performed on the same set of individuals as in mtDNA and AFLP [19].
Among the populations, mtDNA and AFLP polymorphisms was high except in SH and CR whose populations experienced a bottleneck (19). Only two mtDNA haplotypes were found in SH due to the isolating effect of the Shiroro Dam coupled with overfishing [20]. Three mtDNA haplotypes were also found in CR attributable to overfishing and possible environmental degradation. The mtDNA and AFLP analysis did not show any less genetic diversity in the Niger Delta populations (BU, KO, and WA) when compared with LA, UN and LN populations. Mulvey et al. [21] did not also find any evidence of decreased diversity in the mtDNA control region of Fundulus heteroclitus populations in a highly contaminated environment. Microsatellite analysis of five populations of C. nigrodigitatus excluding any Nigerian population revealed high levels genetic variability comparable to those of marine species [6]. The microsatellite data in the present study revealed a similar scenario. However, it is surprising that while mtDNA sequencing was able to detect some variation in SH and CR; microsatellite analysis could not detect any. The probable reason for this lack of variation in SH and CR could be genotyping or scoring errors and the few mtDNA haplotypes have been excluded because of poor genomic DNA.
The range of heterozygosity values of 80-100% and the number of alleles per locus (2-10 per population) in this study was reported for some catfish e.g., [22][23][24]. All the loci had high values of heterozygosity in all the populations of C. nigrodigitatus. However, Kotoulas et al. [6] in a study of four natural populations of C. nigrodigitatus detected 29-30 alleles at three microsatellite loci. The small sample sizes may have limited the detection of more number of alleles in the present study. The microsatellite PIC values higher than 0.5 in this study indicate that more genetic information can be provided by SSR loci [25].
The level of genetic differentiation detected between pairs of population was moderate. This may suggest limited level of gene flow and limited dispersal between collection sites. The same conclusion was reached by Song et al. [19]. Indeed, fish which distribute in freshwater and brackishwater environment may show genetic differentiation somewhat in between those that inhabit barrier free marine environment and barrier isolated freshwater [26][27][28][29].
Significant (P< 0.05) population differentiation was found in more of the pairs of populations F ST 's, showing obviously that the populations were genetically structured. Although the pairwise F ST between populations LA/KO, BU/KO, LA/WA, and KO/WA were large, they were insignificant due to the low bootstrap values. The low support values may be due to the small number of individuals and few loci examined. More number of individuals per population and more number of microsatellite loci are expected to give a better definition of genetic diversity and population structure.
The inbreeding coefficient (F IS ; Table 2) defined as the probability that two homologous alleles present in the same individual are identical by descent showed non-significant negative values for all populations, indicating the presence of more heterozygous individuals in these populations.
Overall, C. nigrodigitatus populations were found to be generally out crossing with little or no inbreeding. The AMOVA analysis showed that 7.37% of the genetic diversity was found between freshwater and brackishwater groups (P=0.047). A small (3.55%) but significant (P=0.000) amount of genetic diversity was found among populations within groups. The AMOVA analysis also showed that a large and significant genetic differentiation (89.08%, P=0.000) accounts for individuals within populations. Unrooted NJ trees based on microsatellites revealed populations UN and LN were differentiated from other populations. The result was consistent with hierarchical analysis of molecular variance above. Freshwater or brackishwater habitat, limited long-distance dispersal of the adult and juveniles may account for breakdown of gene flow.
On the whole, the obtained data show that C. nigrodigitatus maintains sufficient intrapopulation genetic variability and moderate interpopulation differentiation in the studied area. Similar conclusions were reached with mtDNA and AFLP analyses [19]. The potential loss of genetic diversity in the Niger Delta due to disturbances, such as environmental pollution from oil extraction, destruction of breeding sites, capture of fry for aquaculture and overexploitation, are reasons for concern and the conservation of genetic diversity may be one of the most important issues facing the future of C. nigrodigitatus. However, because of limited sampling size, and lack of representative populations from most of the distribution area, further investigation need to be carried out.

CONCLUSION
Freshwater or brackishwater habitat, limited longdistance dispersal of C. nigrodigitatus adults and juveniles may account for breakdown of gene flow and consequently responsible for the current genetic structure. Furthermore, genetic diversity of the species in the Niger Delta was comparable with populations from outside the polluted Niger Delta aquatic environment, maintaining sufficient intrapopulation polymorphism and moderate interpopulation diffentiation.
However, an extensive studies comprising more populations and increased population size are needed to make far reaching conclusion.