|
|
||||||||
ANIMAL GENETICS |
,2
* Animal Genomics and Bioinformatics Division, National Livestock Research Institute, RDA, Suwon, Gyeonggi 441-706, Republic of Korea and
and
USDA-ARS, Corn Insects and Crop Genetics Research Unit, Genetics Laboratory, Iowa State University, Ames 50011
| Abstract |
|---|
|
|
|---|
Key Words: Genetic Relationship Genetic Structure Korean Native Pig Microsatellite
| Introduction |
|---|
|
|
|---|
The genetic relationship between the Asian and the European pig breeds has been evaluated using both mitochondrial DNA and nuclear DNA analyses (Kim et al., 2002a
,b
; Kim and Choi, 2002
). Phylogenetic analysis of mitochondrial D-loop DNA (Kim et al., 2002a
) revealed that Asian native pigs are closely related but differ from European pigs, except for the Berkshire and Large White breeds. Amplified fragment-length polymorphism (Kim et al., 2002b
) and microsatellite markers (Kim and Choi, 2002
) together showed that the Korean native pig has a low level of genetic diversity and is distinct from the Western pig breeds. Nonetheless, population genetics information, especially in a population management context, was limited because of small sample size as well as a limited number of microsatellite markers.
In this study, we used 16 microsatellite markers to evaluate the genetic structure of Korean and Chinese breeds, to explain the genetic relationships between Asian pig and European pig breeds, and to determine the assignment accuracy of individuals into their correct breeds of origin.
| Materials and Methods |
|---|
|
|
|---|
Microsatellite Genotyping
Sixteen microsatellites were selected based on their genomic location, allele number, and ease of scoring (Table 1
). Microsatellite markers that labeled with fluorescence were amplified by PCR using 10 ng of pig genomic DNA as a template. The polymerase chain reaction was performed in a 10-µL final volume with 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 200 uM of each nucleotide (dNTP), 3 pmol of each primer, and 0.5 units of Taq DNA polymerase (TaKaRa Shuzo Co., Shiga, Japan). Thermal cycling conditions in Gene-Amp PCR System 9600 (Applied Biosystems, Foster City, CA) included an initial denaturation for 5 min at 94°C, followed by 35 cycles of 30 s at 94°C, 30 s at annealing temperature (Table 1
), 1 min at 72°C, and a final extension step of 72°C for 10 min.
|
Data Analyses
Allele frequencies (available from the authors upon request), the mean number of alleles per locus (MNA), observed heterozygosity (HO), and heterozygosity expected from Hardy-Weinberg (HW) assumptions (HE) for each locus were computed using the GENETIX software package (Belkhir et al., 1996
). The two measures of heterozygosity are highly correlated, but in this study, we focused on HE because it is considered a better estimator of the genetic variability present in a population (Nei and Kumar, 2000
). To compare the number of alleles between different sample sizes, allelic richness (AR), which measures the number of alleles independent of sample size, was calculated using the program FSTAT v. 2.9.3 software package (Goudet, 2001
).
The program FSTAT was used to calculate two different measures of the genetic differentiation over subpopulations, FST (Weir and Cockerham, 1984
) and RST (Rousset, 1996
), where measures of FST and RST are based on the infinite allele model (Kimura and Crow, 1964
) and the stepwise mutation model (Kimura and Ohta, 1978
), respectively. Pairwise FST and inbreeding coefficients (FIS and FIT) were calculated using the program FSTAT (Goudet, 2001
). The sequential Bonferroni correction was applied to derive significance levels for the analysis involving multiple comparisons (Rice, 1989
).
The probability test approach described by Guo and Tomson (1992) and implemented in the GENEPOP software (Raymond and Rousset, 1995
) was employed to test for HW equilibrium. The HW test for each locus in each population and global tests for all populations were performed to investigate whether there was a heterozygote excess or deficit.
The genetic divergence between the populations based on allele frequencies was calculated according to DA genetic distance (Nei et al., 1983
) using the DISPAN computer program (Ota, 1993
). Phylogenetic trees were constructed by using the neighbor-joining (NJ) clustering (Saitou and Nei, 1987
) and the unweighted pair group method with the arithmetic mean (UPGMA; Sneath and Sokal, 1973
) from DA distance. Bootstrap re-sampling (n = 1,000) was performed to test the robustness of the dendrogram topologies.
To represent geometric relationships among the pig breeds, a principal component analysis (PCA) was applied using gene frequencies of all variable loci. The frequencies of all alleles at a single locus were considered to be independent variables, even though they were not independent from each other, as their sum is unity. Accordingly, correlation matrices were computed from the gene frequencies of all loci. In addition, the eigenvalues of all principal components, the proportions of individual eigenvalues to the total variance (contribution rates of components), and the factor scores of every pig for each of the principal components were computed. A scattergram of the score data was examined to visualize the geometric relationship among pig breeds. The PCA was performed using the XLSTAT program (Agresti, 1990
).
The statistical certainty of assignment or exclusion for individuals into their reference populations was evaluated using the program GeneClass v.2.0 (Piry et al., 2004
). The exclusion method was carried out using the Bayesian approach developed by Rannala and Mountain (1997)
because the Bayesian method showed better accuracy than the frequency- and distance-based methods (Cornuet et al., 1999
; Koskinen, 2003
).
Assignment of each individual was tested using the "leave one out" procedure (Efron, 1983
), which means each individual was excluded from the data set when performing its assignment.
The principle of the exclusion method has been well described elsewhere (Cornuet et al., 1999
; Koskinen, 2003
). Frequency probabilities of multilocus genotypes in each reference population were performed using Monte Carlo simulations of 10,000 independent individuals for the population. The assignment criterion estimate of an individual in question was then compared with the frequency distribution of simulated genotypes of each reference population, allowing exclusion of individuals from populations with a specified degree of confidence (e.g., 0.001).
| Results |
|---|
|
|
|---|
|
|
Across multiple loci, the Korean wild pig, Berkshire, and Landrace showed a significant value of inbreeding coefficient (FIS) after correction for multiple tests. Most pig populations, except for two Chinese native pigs (Min pig and Wuzhishan pig), showed a deviation (P < 0.05) from HW equilibrium. Across populations and loci, 6 of 144 deviated from the HW equilibrium (P < 0.05) after correction for multiple tests (Table 2
). All of these deviated cases are related to the positive FIS, indicating HW equilibrium deviation in the direction of heterozygote deficit.
The overall FIS values per locus ranged from 0.0435 (SW2409) to 0.1393 (SW2), showing an overall FIS of 0.067 (Table 3
). The FST and RST estimates of genetic differentiation were similar for all 16 microsatellites loci (Table 3
). The FST values ranged from 0.2189 (SW2612) to 0.3308 (SW510). The mean FST value of 0.261 from all loci indicated that 73.9% of the genetic variation was caused by the differences among individuals and 26.1% was due to the differentiation among breeds.
Genetic Distances
Neis DA genetic distance and mean FST estimates between each pair of nine porcine populations are shown in Table 4
. The genetic distance ranged from 0.139 (between Landrace and Yorkshire) to 0.684 (between Duroc and Wuzhishan pig). Pairwise FST estimates ranged from 0.092 (between Landrace and Yorkshire) to 0.438 (between Korean native pig and Wuzhishan pig). Genetic divergence of Asian native pig breeds from European breeds was pronounced, whereas genetic divergence among European breeds was relatively small (0.092 to 0.279). Phylogenetic trees of the nine porcine breeds were reconstructed based on Neis DA genetic distances (Figure 1
). Trees from both NJ and UPGMA methods showed a similar topology, but the bootstrap values were slightly different. The Korean native pig and Min pig were grouped into the same branches with commercial western pig breeds with high bootstrap support values (93% in NJ and 98% in UP-GMA). Conversely, the Xiang pig, Wuzhishan pig, and Korean wild pig were clustered into different branches.
|
|
|
|
| Discussion |
|---|
|
|
|---|
The low genetic diversity in the Korean native pig can be attributed to its breeding history for improved traits. Because the Korean native pigs original characteristics were not economically favorable (black coat, small BW, slow growth, and small litter size), since 1910, animal breeders tried to improve its productivity by crossing it with commercial breeds such as Berkshire.
Recently, the Korean native pig has been recognized as an important genetic resource because of its indigenous adaptation and specific traits. When the Korean native pig faced the threat of extinction during the 1980s, because of crossbreeding with commercial breeds and the introduction of many commercial breeds, several animal breeders and geneticists tried to restore the genetic characteristics of the Korean native pig. A small number of pigs were used as founder animals for the genetic restoration of the Korean native pig; however, the founder effects and closed breeding resulted in the loss of genetic variation. These facts further explain why the Korean native pig had the lowest heterozygosity and was clustered with the commercial pig breeds owing to its close genetic distance with Berkshire.
Both phylogenetic trees and PCA revealed that Korean native pig and Min pig (North Chinese pig breed) were grouped together with commercial pig breeds, whereas the Xiang pig and Wuzhishan pig were clustered distantly from other pig breeds.
The current findings confirm previous results, indicating the close genetic distance between the Korean native pig and the North Chinese breed (Kim et al., 2002a
; Kim and Choi, 2002
). Korean native pigs also were placed on the same branch of the phylogenetic tree as the Berkshire and Yorkshire based on the analysis of mitochondrial DNA, even though the Korean native pig was distinct from the European pig breeds (Swedish, Landrace, Duroc, Welsh, and Yucatan; Kim et al., 2002a
). However, our findings largely differ from the previous reports in terms of genetic relationships. Fang et al. (2005)
reported that all Chinese pig breeds, including Min pig, are distinct from Western pig breeds based on a comprehensive microsatellite loci survey for 32 Chinese local pig breeds as well as three Western pig breeds. Moreover, Kim and Choi (2002)
reported that the Korean native pig was placed in a different branch from commercial breeds such as Duroc, Landrace, and Yorkshire (no analysis was made for Berkshire). The discrepancy in the relationship studies could be more likely due to the use of different pig samples, although the use of different kinds and numbers of microsatellites cannot be ruled out. Indeed, the sample size of Min pig in this study was relatively small, so we acknowledge that our sample might not represent the real population of Min pig.
Despite these discrepancies, the close genetic relationship between the Korean native pig, and Berkshire, followed by Min pig, provided evidence supporting the historical hypothesis that the Korean native pig originated from North China, spread to South Korea via North Korea, and then experienced introgression of European alleles through progressive interbreeding with Western pig breeds. Chinese breeds were classified into six types according to their geographic origin, distribution, body conformation, and coat color (Li et al., 2004
). According to their classification, the Min pig belongs to Type I (North China), and the Xiang pig and Wuzhishan pig are classified into Type IV (South China). Therefore, the results of the present study support the historical hypothesis with respect to the North China origin of the Korean native pig. To address this issue fully, however, more rigorous investigation with a greater number of pig samples will be needed.
According to some researchers, domestication of the Asian and European pigs was done independently from wild boar subspecies in Asia and Europe (Giuffra et al., 2000
; Okumura et al., 2001
). Conversely, more recent research revealed that domestication of pigs was undertaken at multiple centers across Eurasia (Larson et al., 2005
). Jones (1998)
insisted that Chinese pig breeds were used to develop European pig breeds such as Berkshire, Small White, and Middle White. This might explain why the Min pig was clustered with commercial breeds. Although the Min pig did not contribute directly to the development of commercial breeds, it is possible that a Chinese pig breed that is close to the Min pig was used for the development of commercial breeds. Once again, however, detailed scrutiny from a larger number of pig samples will be required to answer this issue.
The close relationship between the Korean pig breed and Western pig breeds could be explained by two possible hypotheses: 1) the Korean breed has been introgressed with European breeds or 2) a significant number of European pigs used in our study have Asian haplotypes; thus, they cluster with Min and Korean. Introgression of European alleles into Korean native pigs, as well as the relationship between Min pig and Korean native pig, could benefit from mitochondrial DNA sequence comparison between pig breeds sampled in this study.
All pig breeds, with the exception of two Chinese pigs, deviated significantly from HW equilibrium (P < 0.05). Random mating without artificial selection could cause the HW equilibrium in the two Chinese native pigs, or genetic drift might not have a detectable effect on the genotypic frequencies of these populations. Deviation from HW equilibrium in other pig breeds could be the result of the sampling of a single population with different allele frequencies in the subpopulations (Wahlund effect), nonrandom sampling, and/or inbreeding. In the present study, at least three populations, the Korean wild pig, Berkshire, and Landrace, showed evidence of significant inbreeding (Table 2
). Deficiency of heterozygotes caused by the Wahlund effect has been proposed in other domestic animals, particularly in the Mexican hairless pig population (Lemus-Flores et al., 2001
) and Iberian cattle breeds (Martin-Burriel et al., 1999
).
Our estimation (overall FST = 0.261) of genetic differentiation indicates significant population subdivision over pig breeds. This value belongs in the high category of genetic differentiation between pig breeds examined thus far. The greatest differentiation is observed in European pig breeds (FST = 0.27; Laval et al., 2000
) and the least in Chinese breeds (FST = 0.077; Yang et al., 2003
). Nonetheless, these differences are most likely to be explained by the different types of markers analyzed as well as the current genetic status of source population sampled.
The high level of genetic differentiation among pig breeds in this study increases success rate for individual assignment. This study was able to assign 91.3% of 242 animals into their correct reference breeds. This result was very similar to a previous finding of 92.14% of 420 animals (Yang et al., 2003
). All individuals from Korean and Chinese native pigs showed 100% exclusion success from all alternative reference pig breeds with a high level (P < 0.001) of confidence; however, some portion (9.7%) of Korean native pigs also was excluded from their source population, indicating the need for more efforts in genetic refinement of Korean native pigs. In contrast, Western pig breeds, except for Duroc, could not be excluded successfully from their alternative reference breeds. In particular, Landrace and York-shire could not successfully exclude each other by either direct or simulation approaches, suggesting the possibility of an admixture of gene pools between these two samples studied.
In conclusion, the Korean native pig had low heterozygosity and clustered with the Min pig and Western pig breeds; however, the Korean native pig was distinct from the Korean wild pig and South China pig breeds (Xiang pig and Wuzhishan pig). The findings suggested that the Korean native pig originated from a North China pig breed with a black coat color similar to the Min pig, but which possessed the traits of Western breeds such as Berkshire and Landrace because of crossbreeding. The results of this study provide evidence supporting the fact that commercial breeds were used mainly as sires for improving Korean native pigs. These findings can be used as genetic information for the preservation and further genetic improvement of the Korean native pig.
| Footnotes |
|---|
2 These authors contributed equally to this work. ![]()
3 Correspondence: 560 Omockchun-dong, Suwon, Gyeonggi 441-706 (phone: 82-31-290-1603; fax: 82-31-290-1602; e-mail: kth6160{at}rda.go.kr).
Received for publication February 7, 2005. Accepted for publication June 14, 2005.
| Literature Cited |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. A. Vicente, M. I. Carolino, M. C. O. Sousa, C. Ginja, F. S. Silva, A. M. Martinez, J. L. Vega-Pla, N. Carolino, and L. T. Gama Genetic diversity in native and commercial breeds of pigs in Portugal assessed by microsatellites J Anim Sci, October 1, 2008; 86(10): 2496 - 2507. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Tadano, M. Sekino, M. Nishibori, and M. Tsudzuki Microsatellite Marker Analysis for the Genetic Relationships Among Japanese Long-Tailed Chicken Breeds Poult. Sci., March 1, 2007; 86(3): 460 - 469. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |