Study on Genetic Diversity of Six Duck Populations with Microsatellite DNA

In this study, we investigated the genetic diversity and phylogenetic relationship of six duck populations by employing the genetic polymorphisms of 20 microsatellites. The parameters used in this study included number of alleles, average effective numbers of alleles (E) and average rates of heterozygosity of each population. The results showed that all the microsatellite loci were highly polymorphic except that the locus AJ515896 in Muscovy duck was 0. The average PIC (0.762), average h (0.7843) and average E (5.261) of the six duck populations were all high, indicating that the gene polymorphisms and genetic diversity were high. The test of Hardy-Weinberg equilibrium showed that the six populations in this study were all in Hardy-Weinberg disequilibrium. The F-statistic analysis results showed the range of FST was from 0.0205 (AJ515895) to 0.2558 (AJ515896). The mean FST was 0.0936. Phylogenetic study revealed that Peking duck (Z1 and Z4), Shaoxing duck, Cherry Valley duck and Aobaixing duck were clustered in one group, while the Muscovy duck was clustered in one group alone. The phylogenetic relationships among different populations were in accordance with their breeding history and distribution. Our data suggested that the 20 microsatellite loci were effective markers for analysis of genetic relationships among duck populations. (


INTRODUCTION
China was one of the earliest nations in the world to domesticate ducks (Anas platyrhynchos) and currently raises the largest populations of these birds.However, in past decades, China has imported foreign ducks and used them to improve native duck performance.This resulted in the decrease of native duck populations and, even worse, caused the disappearance of some native populations.Since the native duck populations are invaluable genetic resources, it remains urgent to systematically investigate their genetic diversity and genetic characteristics.There have been many reports using microsatellites to study the genetic diversity of chickens (Romanov and Weigend, 2001;Chen et al., 2003;Wu et al., 2004;Chen et al., 2004;Osman et al, 2005;Kong et al., 2006;Osman et al., 2006;Liu et al, 2006;Chang et al, 2007;Ahlawat et al, 2008).In ducks, Chen et al. (2001), Zuo et al. (2004) and Yan et al. (2005) analyzed the genetic diversity of some populations using random amplification polymorphic DNA (RAPD) and amplification fragment length polymorphism (AFLP).Stai et al. (2003), Huang et al. (2005), Li et al. (2006) and Tang et al. (2007) separately analyzed wild and domestic Muscovy ducks and some native ducks in China by microsatellites.Microsatellites are characterized by a core sequence that consists of a number of tandemly repeated units with a length of 1-6 base pairs.They have high diversity and are easily examined.With the availability of high-throughout DNA sequencing, sizing of microsatellite alleles has become efficient.Therefore, microsatellites have been widely used in many applications.Microsatellites are deemed to be one of the most valuable genetic markers (Aranguren-Méndez et al., 2002;Wang et al., 2004).In this paper, we investigated the genetic diversity of six duck populations (Peking duck (Z1 and Z4), Muscovy duck, Shaoxing duck, Cherry Valley duck and Aobaixing duck) with 20 microsatellite loci.

Population samples and microsatellite primers
Three hundred ducks were randomly chosen from six duck populations in China (50 ducks of each population): Peking duck Z1, Z4, Cherry Valley duck and Aobaixing duck from Beijing, Shaoxing duck and Muscovy duck from Zhejiang.Blood (5 ml) was collected from each duck into a tube containing 1 ml anticoagulant and was then preserved in a -20°C freezer.
Twenty microsatellite primers were designed based on the report of Maak et al. (2000Maak et al. ( , 2003)).The sequences of primers are listed in Table 1.

Genomic DNA extraction and PCR condition
The hydroxybenzene-chloroform method was used to extract duck genomic DNA as described in the Molecular Cloning-A Laboratory Manual (2nd edition) (Sambrook et al., 1999).The PCR amplification was carried out in 20 μl of a mixture containing 1 μl DNA template (40 ng/μl), 10×PCR Buffer 2 μl, 0.4 μl dNTP(10 mmol/L), 2.0 μl MgCL 2 (20 mmol/L), 0.8 μl of each primer (10 pmol/μl) and 0.4 μl Taq DNA polymerase (2 U/μl) from Dingguo Biotech (Beijing).Double-distilled water was added to a final volume of 20 μl.After a denaturing step of 5 min at 94°C, samples were processed through 30 cycles of 45 s at 94°C, 35 s at an optimal annealing temperature and 40 s at 72°C.The last elongation step was lengthened to 5 min at 72°C.

Electrophoresis
Both agarose gel (1.5-2%) and polyacrylamide gel (8%) were used to analyze PCR products.Finally, the Alphalmager software was used to analyze the segmental length of the genes.

Data analysis
Allele frequencies were computed by Pi = (2 (ii)+(ij1)+ (ij2)+ …+(ijn))/2N, where Pi was the frequency of the ith allele for which i represented the allele, while j1, j2, … jn were the co-dominant alleles to i, and n was the number of the allele.
Polymorphic information content (PIC) was calculated by the following formula (Botstein et al., 1980): The formulas of Heterozygosity (h) and effective number of alleles (E) (Xiao, 2002) are listed below: Where m was the number of the allele, and Pi and Pj were the frequencies of the ith and the jth allele, respectively.
The Hardy-Weinberg equilibrium was tested by X 2 .The formula is listed below: Where O i showed observed times of genotypes; E i showed expected times of genotypes; k showed the number of genotypes.
The F-statistic was calculated by the followed formula: Where F IT was the fixation indices of individuals relative to its subpopulations; F IS was the fixation indices of individuals relative to the total population; F ST was the fixation indices of subpopulation relative to the total population.
The genetic distance (Nei et al., 1983) and the dendrogram between different populations were estimated by the neighbor-joining method (NJ) (Saitou and Nei, 1987) in the DISPAN software (Ota, 1993).

Gene diversity and allele frequencies
Figure 1 shows one of the polyacrylamide gels (PAGE) of PCR products.The total number of alleles in 20 microsatellite loci of six populations was 281.There were

Polymorphism information content (PIC)
As shown in Table 2, the PIC of the microsatellite AJ515883 in the Peking duck Z1 was the highest (PIC = 0.9515), but PIC of the microsatellite AJ515896 in Muscovy duck was 0. Among microsatellite loci, the locus AJ515895 had the highest PIC (PIC = 0.8723) while the locus AJ515896 had the lowest PIC (0.5968).The average PIC of all sites and populations was 0.762.PIC of most microsatellite sites was more than 0.5.This indicated that the selected microsatellite loci had high diversity and can reflect the genetic relationship of different populations on a molecular level.

Heterozygosity (h)
The heterozygosity is summarized in Table 3; heterozygosity was the highest (0.8079) in Muscovy duck and varied from 0 to 0.9118; followed by the Peking duck Z1 (0.7954) which varied from 0.6289 to 0.9004; whereas the Aobaixing duck was lowest (0.7654) and varied from 0.5886 to 0.8881.The average h of all populations was 0.7843.

Effective number of alleles (E)
The effective number of alleles is also an index used to reveal the genetic diversity of the populations.Table 4 shows the results for the effective number of alleles.The E value of 20 microsatellite loci was the highest (5.7351) in Muscovy duck and, varied from 0 to 11.3443; while it was the lowest in Charry Valley duck (4.7289) and varied from 2.4534 to 7.9819.The average E of all populations was 5.261.

The test of Hardy-Weinberg equilibrium
The Hardy-Weinberg equilibrium was used for testing whether the genotypes were maintained in balance or deviated from balance.The test results of Hardy-Weinberg equilibrium are listed in Table 5.The six duck populations were all in Hardy-Weinberg disequilibrium.

F-statistic analysis
The F-statistic was used in testing the genetic differentiation among subpopulations.The F-statistics were calculated for 20 microsatellite loci.The results are shown in Table 6.In this study, the range of F ST was from 0.0205 (AJ515895) to 0.2558 (AJ515896).The mean of F ST was 0.0936.

Genetic distances and clustering of ducks
The Nei's genetic distance (D A ) was calculated by the allele's frequencies and the results are summarized in Table 7. Genetic distance between Shaoxing duck and Cherry Valley duck was the shortest (0.1839), followed by Shaoxing duck and Peking duck Z1 (0.1947); the genetic distance between Muscovy duck and Peking duck Z4 was the longest (0.4558).The dendrogram was constructed using the NJ method of DISPAN software (Figure 2).The six populations were clustered into two groups.Peking duck Z4, Aobaixing duck, Peking duck Z1, Shaoxing duck and Cherry Valley duck were clustered together, while Muscovy duck was not included in the group.

Genetic diversity among loci and populations
In this study, we investigated the genetic diversity and phylogenetic relationships of six duck populations in China.The polymorphisms revealed differences among 20 microsatellite loci in the six populations.Average effective number of alleles of all microsatellite loci was 5.261, but it was different in each population and locus and varied from 0 to 11.407.
The PIC was a good index for genetic diversity evaluation.Botstein et al. (1980) first pointed out that PIC index can be used to evaluate the level of gene variation: when PIC>0.5, the locus has high diversity; when PIC< 0.25, the locus has low diversity; and the locus has intermediate diversity when PIC between 0.25 and 0.5.In this study, the PIC of 20 microsatellites in each population all showed high diversity except the AJ515896 locus in Muscovy duck.This suggested that the 20 microsatellite loci are all high diversity loci in the duck.
Heterozygosity (h) is one of the indices used to assay the genetic variation of each population.The values of h indicate the diversity level of the molecular marker.When the value is high, the molecular marker's diversity is high too.Among all loci studied, the h of the locus AJ515895 had the highest value (0.876), while the locus AJ515896 had the lowest h value (0.6212).In all populations, the h of Muscovy duck was the highest (0.8079), followed by Peking duck Z1 (0.7954), whereas Aobaixing duck was the lowest (0.7654).Our data indicated that genetic diversity of each population was high and there were enough gene resources in duck populations.This may be related to the breeding history and environment of each population.
Effective number of alleles (E) is used to assay the effect of alleles in each population.The result of this study revealed that the E of Muscovy duck was the highest (5.7351) and Cherry Valley duck was the lowest (4.7289) among all duck populations.Average E of all populations was 5.261.It was also shown that the E of 20 microsatellite loci in all duck populations was high.This result suggested that the populations have better abilities to keep the effective alleles when selection, mutation or genetic drift has occurred.
Taken together, each duck population used in this study had high average effective number of alleles (E), heterozygosity (h) and polymorphism information content (PIC), suggesting that genetic diversity of the 20 microsatellite loci in the six populations was high.

The structure of six duck populations
The Hardy-Weinberg equilibrium was used in testing whether the genotypes were maintained in balance or deviated from balance.In this study, the six duck populations were all in Hardy-Weinberg disequilibrium.These results showed that the structures of six duck populations were all being destroyed.Selection, nonrandom mating and inbreeding were the main reasons which induced the disequilibrium.In addition, in this study there were other reasons causing the disequilibrium, such as the excursion caused by mutation, genetic drift, selection, etc.
The F-statistic was used in testing the genetic differentiation among subpopulations.The F IS and F IT may be positive values or negative values, however, the F ST values were always positive.When there is no differentiation, the value of F ST is 0; when alleles among populations are quite different, the value of F ST equals 1.In  this study, the range of F ST was from 0.0205 (AJ515895) to 0.2558 (AJ515896).The mean of F ST was 0.0936.

Phylogenesis of duck populations
There are many methods to measure the genetic distance.Nei showed that average codon margin of each locus can be estimated by the gene frequency of abundant loci.In this study, genetic distance and the dendrogram (Figure 2) between different populations were estimated by the neighbor-joining method (NJ) in the DIASPAN software.
The dendrogram of populations in Figure 2 reflected the genetic relationships of populations.The six populations clustered into two groups: one group included Peking duck Z4, Aobaixing duck, peking duck Z1, Cherry Valley duck and Shaoxing duck, while Muscovy duck was clustered into another group.In group one, Peking duck Z4 had the shortest genetic distance with Aobaixing duck, followed by Peking duck Z1, Cherry Valley duck and Shaoxing duck.The result of this cluster was consistent with breeding history and environment of the five populations.According to evolutional history, Peking duck appeared two hundred years ago.Peking duck Z1 and Z4 were new lines, which were cultivated in the 1980s in China.Female line Peking duck Z1 was mainly used to lay eggs, while Peking duck Z4 was the male line and used to breed with the Aobaixing duck.Cherry Valley duck was imported into China from England and Aobaixing duck from France (Zhou, 2002).They were crossed with Peking duck.Shaoxing duck was an egg-duck breed whose feather color, production performance, body size were different from Peking duck.Muscovy duck was another genera (Cairina moschata), and the form and biological characters were different from other duck populations.Therefore, it was clustered alone.The result was consistent with that of Yan et al. (2005).
In conclusion, our data suggested that 20 microsatellite loci could be used to investigate the genetic diversity and phylogenetic relationships among different duck populations.

Table 3 .
The heterozygosity of 20 microsatellite sites in different duck populations

Table 4 .
The effective number of alleles of 20 microsatellite loci in different duck populations

Table 5 .
The test of HDW with χ 2 based on genotype frequency of 20 microsatellites

Table 7 .
D A genetic distances among populations