Genetic diversity of Ukrainian local pig breeds based on microsatellite markers

Federal Science Center for Animal Husbandry named after Academy Member L. K. Ernst, Dubrovitsy, 60, Podolsk District, Moscow Region, 142132, Russia. Tel.: +38-050-991-53-14. E-mail: kssnail0108@gmail.com Kramarenko, S. S., Lugovoy, S. I., Kharzinova, V. R., Lykhach, V. Y., Kramarenko, A. S., & Lykhach, A. V. (2018). Genetic diversity of Ukrainian local pig breeds based on microsatellite markers. Regulatory Mechanisms in Biosystems, 9(2), 177–182. doi:10.15421/021826


Introduction
Preserving the current diversity of the living material on Earth is fundamental for the survival of future generations. In the case of animals, in the recent past, more and more effective breeding programmes have been implemented, and have led to an emphasis on a few specialised stocks. Consequently, breeds that are less suited to current needs tend to see their numbers decline and to be eventually lost. Conservation of variation, however, is necessary to meet future agricultural challenges and particularly food needs, as well as to preserve the rich agricultural heritage of the various regions of the world (SanCristobal et al., 2003).
Livestock populations have been subjected to a variety of evolutionary forces during their histories. The cumulative effect of foundation events, genetic drift and natural or artificial selection has led to the formation of distinct breeds. In the process of evaluating genetic diversity to develop conservation programmes, it is of interest to assess genetic variation between domestic stocks by using powerful tools such as genetic markers (Berthouly et al., 2008).
At present there are more than 730 breeds or lines of pigs throughout the world and more than two thirds are found in China and Europe. Many of these (possibly more than 270 breeds) are now in danger of extinction and others are threatened by inefficient use or loss due to cross breeding. To evaluate genetic uniqueness and breed diversity of pigs and assist in rationalising breed conservation programmes, microsatellites have been and remain efficient markers (Nidup & Moran, 2011).
Europe contains a large proportion of the world population of pigs (circa 30%) as well as of the world genetic diversity of pigs (37% of the breeds included in the FAO inventory). However, the European pig industry relies predominantly on a limited number of breeds, since one single breed, the widely known Yorkshire (Large White in many countries), represents about one third of the slaughter pigs' gene pool of the European Union (Laval et al., 2000). Local breeds can be considered cultural properties in relation to their role as historical witnesses as they often play a central part in the agriculture tenures and in the social life of rural populations. Local breeds can also be likened to cultural properties because they contribute to the preservation of ancient local traditions. The analysis shows that consistent differences can be observed in the cultural values of local breeds, both as historical witness and as custodian, today, of local traditions (Gandini & Villa, 2003). It can be assumed that local breeds contain the genes and alleles per-tinent to their adaptation to particular environments and local breeding goals. Such local breeds are needed to maintain genetic resources permitting adaptation to unforeseen breeding requirements in the future and can serve as a source of research material (Romanov & Weigend 2001).
Microsatellite markers (MS) also referred to as short tandem repeats (STR), short sequence repeats (SSR) or sequence tagged microsatellite sites (STMS) contain repetitive sequences composed of 2-6 nucleotides. The most common motif in the pig genome is (CA) n and its number is estimated to range between 65,000 and 100,000 copies (Winterø et al., 1992). Microsatellites have been proposed as the best markers for evaluating the genetic diversities of domestic animals because of their abundant, even distribution in the genome, high polymorphism and ease of genotyping. The International Society of Animal Genetics (ISAG) and the Food and Agriculture Organization (FAO) have recommended a set of 27 microsatellite loci (www.toulouse.inra.fr/lgc/pig/panel/html) for evaluating the genetic diversities of pigs as part of the global strategy for the management of farm animal genetic resources (Hammond & Leitch, 1998).
The analysis of the MS loci showed that the allelic diversity pattern among breeds was quasi-independent from the diversity pattern based on allele frequencies. Genetic distances showed no particular clustering of local with international breeds, confirming the genetic uniqueness of the European local breeds compared to mainstream international breeds (Ollivier, 2009). Within breeds of the same species (or within one species) it is possible to spot the ancient or the most estranged from the "protogene pool" groups of animals and, thus, to use these data in the breeding strategy aimed at the preservation of animal forms close to the ancestral form. Using various DNA markers, methods of genomic and genetic breeding in nature conservation and genetic selection of domesticcated species provides an opportunity to obtain data on the genetic potential (value, originality) of the breed, which is important for scientific justifycation of its preservation (Stolpovskiy & Zakharov-Gezekhus, 2017).
Thus, the focus in this paper will be on the genetic diversity pigs of five Ukrainian local and one commercial (Duroc) breeds assessed by using genetic markers (microsatellite DNA loci).

Material and methods
This study was carried out to understand the molecular genetic diversity of six swine populations in Ukraine. A total of 350 pigs representing five local pig breeds from Ukraine (Mirgorod, MIR: n = 26; Poltava Meat, PM: n = 13; Ukrainian Meat, UM: n = 128; Ukrainian White Steppe, UWS: n = 67 and Ukrainian Spotted Steppe, USS: n = 25) and one commercial breed were sampled. The commercial breed pigs (Duroc, DUR: n = 91) were included in the investigation in order to complete the picture of the diversity existing in populations of pigs in different origins. The samples, for both sexes (sows and boars), were collected in five different region of Ukraine (Table 1). PCR analysis was carried out on DNA extracted from 350 ethanolfixed small tissue samples (pieces of ear). A DNA extraction using the Nexttec Clean Column kit (Nexttec, Leverkusen, Germany) was performed according to the manufacturer's instructions. Genomic DNA was extracted based on the Zinovieva and Gladyr's (2011) protocol, with minimum adaptations.
Twelve microsatellite loci (SW24, S0155, SW72, SW951, S0386, S0355, SW240, SW857, S0101, SW936, SW911 and S0228) recom-mended by the International Society for Animal Genetics (ISAG) were used to analyze the genetic diversity of six Ukrainian swine breeds. Electrophoresis was conducted using an ABI 3130xl Genetic Analyzer (Applied Biosystems, USA). The size of each allele was visualized and determined by GeneMapper version 4.0 software (Applied Biosystems, USA). The DNA of the samples were stored in the DNA Bank of the Federal Science Center for Animal Husbandry named after Academy Member L. K. Ernst where this experiment was developed.
In the analysis of the genetic variability within and between Ukrainian swine breeds, the GenAIEx version 6.5 software (Peakall & Smouse, 2012) was used to calculate population genetic parameters such as: allelic frequencies observed (Ho) and expected heterozygosity (He) for each locus. The effective allele number (Ae) for each swine breed or locus was calculated using the following formula: Ae = 1 / (1 -He), where He corresponds to the expected heterozygosity for each swine breed or locus, respectively.
We determined levels of genetic differentiation within and among six Ukrainian pig populations using the indices proposed by Weir & Cockerham (1984): F IT (=F), F ST (=Θ) and F IS (=f) in FSTAT version 2.9.3 software (Goudet, 2002). The bootstrap values were obtained to estimate the statistical signifycance for each of the indices by permutation test (999 permutations).
GENEPOP version 4.2 software (Rousset, 2008) was used to find a significant deviations form Hardy-Weinberg equilibrium (HWE) per breed and locus using Markov chain algorithm (Guo, Thompson, 1992) with 10,000 dememorizations, 200 batches and 5,000 interactions per batch (the exact test).
A dendrogram (hierarchical tree diagram) was created using the UPGMA method (Unweighted Pair Group Method with Arithmetic Mean) (Sneath & Sokal, 1973). Bootstrap analyses (with 999 permutations) were used to estimate the internal consistency of the suggested groupings in the PAST version 3.01 software (Hummer et al., 2001). A Principal Coordinate Analysis (PCoA) was done to visualize the geometric relationships among the Ukrainian swine breeds using GenAlEx version 6.5 software (Peakall & Smouse, 2012).
Population genetic structure of the swine breeds was investigated using Pritchard et al. (2000) algorithm implemented in Structure version 2.2 application based on the multilocus microsatellite genotypes. The Structure software is able to determine genetically distinct clusters (populations of origin, K) of the sampled pigs.
Based on the Markov chain Monte Carlo method (MCMC), the Structure Harvester algorithm (Earl, 2012) was used to estimate the natural algorithm of the probability that any particular individual belongs to the assumed K clusters (Evanno et al., 2005). The program provides a fast way to assess the range of possible clusters (K) from 2 to 11, and was run 10 times for each K.

Results
All microsatellite loci exhibit substantial levels of polymorphism across the study samples. A total of 132 alleles were detected at these 12 marker loci in the 350 evaluated samples of Ukrainian pigs. Only the alleles observed once were not included in the analyses. The number of observed alleles (Na) detected per polymorphic microsatellite locus ranged from two (MIR, S0355 and PM, SW951) to 14 (UM, SW24).
Shared alleles were detected in high frequencies among Ukrainian swine breeds in most of the studied loci (Table 2). Eight loci presented from 1 to 2 shared alleles in all Ukrainian local pig breeds: S0155, SW72, SW951, S0386, S0355, SW240, SW857 and S0101. For the UM and UWS pig breeds higher total alleles (101 and 88, respectively) and rare alleles lower than 5% numbers (43 and 41, respectively) were detected across the 12 microsatellite loci analyzed. The Duroc popula-tion had a lower allele number, with 80 alleles, from which 34 had frequency lower than 5%. There exist population-specific alleles found in a single swine breed ('private alleles'). Thirty-four 'private alleles' were identified among all tested pig breeds (Table 3) which were distributed between 11 of the 12 loci. A high number of 'private alleles' was detected in the Duroc population -nine alleles did not occur in other Ukrainian pig breeds. In animals of the UM breed, a total of 16 'private alleles' were detected. The number of 'private alleles' discovered for other Ukrainian local pig breeds was lowest (1-4 only).

Table 3
List of the 'private alleles' (in bp) found among five Ukrainian local and one commercial pig breed The effective number of alleles (Ae) varied from 1.28 to 5.89 and was proportional to the values of expected heterozygosity (He) found in loci SW951 for the USS breed (0.218) and SW24 for the UM breed (0.830), respectively.
The highest heterozygosity (over 80%) was detected for loci S0101 (MIR and PM breeds), SW857 (MIR and PM), SW72 (UM) and SW24 (UWS and DUR). The lowest value (< 20%) was revealed for loci S0355 and SW24 in MIR and for locus SW951 in PM.
The HWE test showed that all of the polymorphic microsatellite loci deviated from HWE (P < 0.05) in at least one population (Table 4). Loci S0355 (5), S0386 (4) and SW24 (4) presented a higher number of populations in the HWE imbalance. As shown in Table 4, two (UM and UWS) of the five studied Ukrainian local swine breeds had deviated from the HWE for most loci. The Ukrainian local pig breeds presented from one (for USS) to seven (for UWS) loci that deviate from HWE, while the Duroc had eight loci that did not fit HWE (Table 4). This underlines a great difference in the number of loci in Hardy-Weinberg disequilibrium among the five Ukrainian local and one commercial pig breed (Table 4).

Table 4
Results of the HWE test for 12 microsatellite loci identified on five Ukrainian local and one commercial pig breed (D -deficit heterozygosity; E -excess heterozygosity; * -P < 0.05; ** -P < 0.01; *** -P < 0.001; ns -P > 0.05) According to the genetic diversity indices (Table 5), the mean allele number (Na) was lowest for PM (2.92 alleles/locus), while the UM and UWS populations presented highest estimates of 8.42 and 7.33, respectively. The total Na was 6.23 alleles/locus (ranging between two to 14), whereas only two (UM and UWS) of the Ukrainian local pig breeds and the Duroc presented Na above the total mean score. The average values of observed heterozygosity (Ho) in the studied swine populations varied from 0.382 ± 0.096 (for PM) to 0.668 ± 0.030 (for UM). For the PM, the effective number of alleles (Ae = 2.08 ± 0.42) and the expected heterozygosity (He = 0.452 ± 0.087) were the lowest values amongst all analyzed pig breeds. On the other hand, UM was the breed that presented the highest estimated values of effective number of alleles (Ae = 3.83 ± 0.29) and expected heterozygosity (He = 0.718 ± 0.027). Comparing the estimates of expected heterozygosity and the effective number of alleles, Duroc possesses a high amount of genetic diversity compared to all other studied pig breeds, with the exception of UM and UWS.
The variability within pig breeds estimated using the inbreeding coefficient (F IS ) showed that the highest value was observed for the PM and MIR breeds (0.184 ± 0.099 and 0.178 ± 0.102, respectively). In the local swine breeds it can be inferred that significant and positive F IS values are a consequence of the inbreeding effect, arising from matings between related sows and boars. The Duroc presented a high significant inbreeding value also (F IS = 0.111 ± 0.070). The high positive values for F IS in five pig breeds also showed a significant deviation from HWE ( Table 4). The fixation coefficient of populations (F ST ) per locus varied from 0.057 (SW72) to 0.206 (S0101), with a mean value of 0.122 ± 0.010. Thus, 12.2% of the total genetic variation was explained by differences between pig populations (Table 6). For the F ST -index, six of the loci (50.0%) presented values outside the 95% confidence interval, with four loci indicating significant values to determine differences between studied pig breeds.
The mean F IT and F IS values for all microsatellite loci were 0.204 and 0.092 with 41.7% and 50.0% of the values outside the 95% confi-dence interval, respectively. The mean fixation index within the populations (F IS ) in each locus represented positive values, except for loci SW72 (-0.086), SW857 (-0.039), S0101 (-0.035) and SW936 (-0.050), which presented negative F IS , indicating an excess of heterozygous (primarily for Duroc, MIR and PM). The positive F IS (0.092 ± 0.039) for all loci of pig breeds reflects the deviations from the HWE for most Ukrainian local pig breeds (Table 4). The genetic distances (D A ) between each pair of pig breeds are shown in Table 7. The highest Nei's D A value (0.824) was found between the MIR and Duroc breeds, while the UM and the USS breeds were quite close to each other (0.159). Nei's D A value indicated that the UM was the Ukraine local pig breed closest to the commercial Duroc breed (0.268), and that the MIR was the most genetically distant from Duroc (0.824). Overall, the Duroc breed revealed the longest distances in relation to all Ukrainian local pig breeds, probably due to the geographic isolation, drift and/or sampling effects. This finding supports the fact that the evolutionary path of Ukrainian local pig breeds is totally different from the evolutionary path of commercial pig breeds. The first axis of the PCoA plot of the six breeds ( Fig. 1) clearly distinguishes the MIR and PM from the other breeds, which indicates that the Nei's D A values among the MIR and PM, on the one hand, and the other Ukrainian local pig breeds, on the other hand, is high. Figure 2 presents a dendrogram built with the UPGMA method from Nei's (1972) D A matrix. Two different groups were observed: the first one (with 55% confidence) formed by Duroc and three local pig breeds (UM, USS and UWS); and the second group (with 67% confidence) formed by the MIR and PM breeds (Fig. 1 also).
An assignment-test was used to investigate relationships between six pig breeds ( Table 8). The percentage of individuals assigned to the breed of origin varied from 69.2 (for PM) to 100% (for USS). Generally, we found that 92.6% of the individuals (324 out of 350) were correctly assigned to the respective breed of origin, which is probably a consequence of the clear intrabreed structure. The individuals of the Duroc breed were assigned to its true population of origin with a frequency of 92.6%. Only few individuals had a high probability of mixed ancestry (e.g., two MIR individuals clustered together with PM and three PM with MIR; five UM were assigned to UWS and four to USS). Thus, the assignment-test results yielded further evidence for genetic originality of the five Ukrainian local pig breeds. The Bayesian approach implemented by Structure Harvester was used to evaluate the most likely number of ancestral populations underlying the observed interbreed genetic diversity. The likelihood of the observed data given the number of putative population of origin [Ln Pr(X|K)] is shown in Figure 3, for numbers of inferred populations (K) ranging from 2 to 11. The mean value of Ln Pr(X|K) increased up to K = 4 and then descended, with a large increase in its variation. Thus, it was assumed that K = 4 is the optimal number of clusters (ancestral populations) for the pig populations studied (Fig. 4).
Probabilities from the allocation test of individuals for the six pig breeds estimated using Structure Harvester (for K = 4) are presented in Table 9. In cluster 1 the highest grouping probabilities were revealed for MIR (0.917) and PM (0.750) breeds. Local breeds UM (0.824) and USS (0.772) were grouped in cluster 2. Cluster 3 was related to the local breed UWS (0.873). Cluster 4 indicated high allocation probabilities for the commercial pig breed Duroc (0.924). It is noteworthy that, all pig populations studied, despite having presented higher proportions in certain of the clusters, had several individuals allocated in other clusters.

Table 8
Assignment analysis of the five Ukrainian local and one commercial pig breed (values are the number of individuals from each breed (rows) assigned to each population (columns); italic indicates correct assignment. Accurate -the proportions of individuals derived from their source population)  Fig. 4. Individual membership of the five Ukrainian local and one commercial pig breed of the K clusters (K = 4) inferred by Structure analysis: codes on the x-axis indicate the putative population of origin; see Table 1 for location abbreviations; each colour denotes a cluster from the structure analysis

Discussion
This is the first attempt to specifically quantify the population genetic diversity of the local Ukrainian pig breeds based on microsatellite markers. The six pig populations considered in our study showed substantial genetic diversity, with an overall mean of 6.23 alleles per locus (from two to 14) and an average expected heterozygosity of 0.572 for the 12 STR loci. At the breed level, the mean number of alleles per locus was 5.71 (2.92-8.42), with an allelic richness corrected for sample size of 2.84 (2.58-4.83) and an expected heterozygosity of about 0.525 (0.382-0.668). These results are in agreement with those observed for certain European pig breeds, but are somewhat lower than those reported for Asian breeds (Nidup & Moran, 2011).  The mean number of alleles per locus of 5.00-8.42 recorded in the present study for the UM, UWS and USS pig breeds is more than twice the mean number of alleles of 2.92-3.92 recorded for the MIR and PM pig breeds, indicating higher genetic diversity in some local pigs of Ukraine but approximately corresponds to the mean 7.00-7.70 reported by Behl et al. (2002) for two Indian pig breeds.
The relatively high number of alleles found in the UM and USS pig breeds is an indication that the effects of isolation and artificial selection of these populations has been moderate. The lower number of alleles in the Duroc population (6.67) reflects a relatively recently established Ukrainian population of limited size. The number of alleles in the Ukrainian Duroc pigs is, however, still higher than values of 2.39-2.80 reported in Belgian and certain Asian swine populations ( Van Zeveren et al., 1995;Fan et al., 2002;Chang et al., 2009).
Thus, the PM breed, based on estimates of effective number of alleles and observed heterozygosity, can be regarded as the Ukrainian local pig breed with lowest genetic diversity. As previously mentioned in Niang Megha pig (Indian local breed), the low effective number of alleles may be due to very low frequency of most of the alleles at each locus and a very few alleles might have contributed the major part of the allelic frequency at each locus (Zaman et al., 2013). Overall, the three Ukrainian local pig breeds presented higher mean values for the intrapopulation genetic diversity parameters (such as mean estimation of Na, total allele and rare allele numbers), than the ones obtained for the commercial pig breed (Duroc). Additionally, the high amount of 'private alleles' found, mainly in the local swine breeds, shows their importance and the necessity of their preservation and conservation.
Such remark shows the high diversity of the local pig populations in comparison to the specialized (commercial) ones. This higher diversity may be explained by the fact that the locally adapted genetic groups are not subject to constant improvement programs for specific characteristics such as specialized breeds (Silva et al., 2011). Factors such as the level of inbreeding, population size, the history or origin of the breeding population, the level of artificial selection pressure and husbandry practices affect the genetic diversity of domestic animal populations (Ayizanga et al., 2016).
There was a significant level of inbreeding recorded at all loci studied. The lack of compliance with the HWE that was observed for at least one locus in most Ukrainian pig breeds is probably associated with the substantial deficit in heterozygosity. This shortage could be a consequence of inbreeding or intrabreed substructure, which are common properties in local breeds of low census population size (Ollivier et al., 2005;SanCristobal et al., 2006).
Moreover, in the analysed pig breeds, the high deviation from HWE loci number is probably due to the fact that the animals were raised by various pig breeding farms in Ukraine. These breeding farms often carry out matings between related individuals, especially due to the small size of the herd. On second hand, studying European commercial pig breeds, Laval et al. (2000) reported that the majority of them remained within HWE.
The mean overall fixation index (F IT ) of 20.4% recorded in the present study shows a great deal of genetic differentiation in individual animals relative to the total population. This index combines the genetic effects of non-random mating within populations together with the effects of genetic drift among populations (Ayizanga et al., 2016).
F ST was the smallest at SW240 and SW24 loci (0.085-0.088) while at SW951 locus it was the greatest (0.145). A mean F ST of 0.122 indicates that 12.2% of total genetic variability occurs among the Ukrainian local pig populations and this is indicative of moderate genetic differentiation. This is comparable with results of Ayizanga et al. (2016) who reported a significant F ST value of 12% among the local pigs of Ghana. AMOVA results of five Brazilian genetic groups (local and Landrace) obtained by Sollero et al. (2009) showed that 14% of all observed diversity came from the difference between the evaluated genetic groups. In another study carried out with Chinese pigs, an F ST value of 7.7% was found (Yang et al., 2003). The highest genetic differentiation values for pig populations ever reported were F ST = 27% (Laval et al., 2000) between European pigs, followed by 26.1% for a differentiation study also carried out with European, Korean and Chinese pigs (Kim et al., 2005).
Results from the assignment test suggested a true genetic structure with significant differentiation among all populations, except the MIR and PM breeds. This result supports the trend from frequency-based D A values, which showed significant differentiation among all population pairs but with the lowest DA observed between the MIR and PM breeds (0.169) and between the UM and USS breeds (0.159). Five populations displayed a pattern of strong distinctiveness, with more than 88% of individuals assigned to a single cluster in each case. By contrast, significant numbers of pigs from the PM breed were distributed over several clusters, with the most prominent cluster containing only 69.2% (9 from 13) of individuals. Overall, the results from the assignment test support the hypothesis of high homogeneity within most Ukrainian local pig breeds sampled (Traspov et al., 2016).

Conclusions
This study was the first based on microsatellite markers for the genetic characterization of the local pig breeds from Ukraine. The results show that levels of genetic diversity in five local Ukrainian pig breeds are moderate to high. Results from an assignment-test confirmed results from F ST which suggested an original genetic structure with significant differentiation between most breeds sampled, but with little differentiation among MIR and PM breeds.
Financial support was received from the Ministry of Education and Science of Ukraine (0117U000485) to A. V. Lykhach and A. S. Kramarenko.