MATERIALS AND METHODS

Sample collection: A total of 469 purebred chicken samples from 20 breeds were acquired in Korea from Hanhyup and the NIAS. Eight commercial breeds of chickens (HH, HF, HG, HV, HS, HW, HY and HA) were collected from Hanhyup. Samples from 12 purebred breeds were obtained from the NIAS as follows: Red KNC (NR), Yellow KNC (NY), Gray KNC (NG), Black KNC (NL), White KNC (NW), Ogye (NO), Leghorn F (NF), Leghon K (NK), Black Cornish (NH), Brown Cornish (NS), Rhode Island Red C (NC) and Rhode Island Red D (ND) (Table 1).

Table 1:	Description of the 20 chicken breeds used in this study with 27 microsatellite markers

Table 2:	Description of the 7 chicken breeds used in the case study with multiplex-PCR marker sets

Multiplex PCR was used on 339 commercial chickens (Table 2). Genomic DNA (g DNA) was extracted from blood samples collected from the wing veins. DNA from the blood was extracted according to the standard protocol provided with the PrimePrep Genomic DNA isolation kit for blood (GeNetbio, Korea).

Marker selection and multiplexing: For the analysis of genetic diversity, 27 microsatellite markers from the FAO/ISAG recommended diversity panel were used. (ISAG/FAO Standing Committee, 2004; Table 3). From these results, 16 highly polymorphic MS markers were selected based on the number of alleles, expected heterozygosity (H_exp) and polymorphic information content (PIC). The selected markers were classified into two sets of 16 markers each (SET1 and SET2). Size fractionation of all 16 markers in each group was made possible by capillary electrophoresis using a combination of fluorescent-labeled primers (Table 6). The possibility of multiplexing the size fractionation by capillary electrophoresis was also considered in the selection of suitable markers.

PCR: All 469 DNA samples were amplified using a GeneAmp PCR system 9700 (Applied Biosystems, Foster City, CA, USA). The amplifications were carried out using 15 μL reaction mixtures containing genomic DNA (5-20 ng μL^–1), primer (10 pM) and 1.5 U Thermo Scientific Maxima Hot Start Taq DNA polymerase (Thermo Scientific, Waltham, MA, USA), which were then subjected to 30 cycles of 30 sec at 95°C, 30 sec at 55°C and 30 sec at 72°C. The multiplex PCR program was the same as the program described above.

Genotyping and statistical analysis: Genotyping of the amplified DNA was performed using an automated Genetic Analyzer 3130xl (Applied Biosystems, USA). The results were obtained using GeneMapper V 4.1 (Applied Biosystems, USA). The genotyped data were analyzed using the Cervus V 3.0 program⁸ and Excel MS toolkit version 3.1⁹ to calculate the allele frequencies at each locus for each population, the H_exp, the observed heterozygosity (H_obs) and the PIC values. The levels of inbreeding-like effects within subpopulations (genetic distance [F_st]), among subpopulations (within inbreeding [F_is]) and within the entire population (total inbreeding [F_it]) were analyzed by F-statistics¹⁰. The STRUCTURE program¹¹ was used to analyze the proportion of the membership within the studied breeds. The expected probability of identity values among genotypes of random individuals (PI), random half sibs (PI_half-sibs) and random sibs (PI_sibs) were calculated using API-CALC (ver 1.0)¹². The neighbor-joining method¹³ was used to construct a phylogenetic tree. Principal coordinates analysis (PCoA) was performed on 7 breeds using the GenAlEx 6.4 program. A correspondence analysis, which is a weighted principal component analysis (PCA), was performed using the allele frequency data for the individuals of all 7 breeds (Table 2) and 16 MS markers with GENETIX software¹⁴.

This study was approved by the research Ethics Committee of Hankyung National University 2018-2.

RESULTS AND DISCUSSION

Microsatellite polymorphisms, within and between populations: The microsatellite polymorphism, evaluated by the number of alleles (NA) per locus, mean heterozygosity, PIC and F-statistics values for each breed are described in Table 4. A total of 316 alleles were observed at the 27 microsatellite loci distributed in 469 Korean chickens representing the 20 Korean chicken populations. All the microsatellite loci typed were polymorphic. The NA per locus ranged from 5 (ADL0268) to 20 (MCW0127), with a mean of 11.7 alleles. The mean H_Exp across loci was 0.797, with estimates per locus ranging from 0.688 (MCW0288) to 0.881 (MCW0104). For H_Obs, the mean for all loci was 0.547, with a range of 0.359 (GCT0016) to 0.677 (MCW0145, MCW0029). In this study, all the loci had high PIC values (PIC>0.5). Estimation of the genotypic diversity of heterozygosity and PIC value data of the MS markers were previously used for the determination of animal breed selection and animal traceability. Accordingly, markers with PIC>0.5 and H_Exp>0.6 are the most reasonably informative loci for applications in genetics^15-17. F-statistics were estimated in a fixation index as genetic differentiation (F_st), the global heterozygote deficit among the 20 chicken breeds (F_it) and the heterozygote deficit within the breed (F_is) among the 27 MS markers (Table 4). Among these markers, estimation of fixation index has been discovered for F_st, F_it and F_is, with values ranging from 0.119 to 0.382, 0.199 to 0.565 and -0.037-0.337, respectively.

Table 3:	Primer information for 27 selected microsatellite markers in the study

The estimated mean values of F_st, F_it and F_is were 0.233, 0.321 and 0.115, respectively. The breed statistics generated by the 27 microsatellite markers in nine chicken breeds are shown in Table 5. The mean number of alleles (MNA) in each breed ranged from 3.59 (NF) to 6.63 (NY). The two most diverse breeds were the HF and NL breeds, which had the highest mean H_Exp (0.745 and 0.734), H_Obs (0.705 and 0.675) and PIC (0.575 and 0.686).

Table 4:	Statistical analysis of heterozygosity (HObs and HExp), polymorphic information content (PIC) and F-statistics value using 27 selected microsatellite markers among the 20 chicken breeds
HEXP: Expected heterozygosity, Hobs: Observed heterozygosity, PIC: Polymorphic information content, Fit: Total inbreeding, Fst: Genetic distance, Fis: Within inbreeding

Table 5:	Statistical analysis of the mean number of alleles (MNA), heterozygosity (HObs and HExp), polymorphic information content (PIC) and within inbreeding (Fis) observed across 27 microsatellite loci for each population
MNA: Mean number of alleles, HEXP: Expected heterozygosity, Hobs: Observed heterozygosity, PIC: Polymorphic information content, Fis: Within inbreeding, HH: Hanhyup H, HF: Hanhyup F, HG: Hanhyup G, HV: Hanhyup V, HS: Hanhyup S, HW: Hanhyup W, HY: Hanhyup Y, HA: Hanhyup A, NR: Red Korea native chicken, NY: Yellow Korea native chicken, NG: Gray Korea native chicken, NL: Black Korea native chicken, NW: White Korea native chicken, NO: Ogye, NF: Leghorn F, NK: Leghorn K, NH: Black cornish, NS: Brown cornish, NC: Rhode Island red C, ND: Rhode Island red D

Table 6:	Characteristics of microsatellite markers flanking each sarcomere gene and their primer sequences and modifications

Table 7:	Expected probability values among genotypes of random individuals (PI), random half-sib (PIhalf-sibs), random sibs (PIsibs) and total expected probability (PE) for the discrimination of chicken breeds using multiplex-PCR marker SET1 and SET2

Construction of multiplex PCR microsatellite marker sets and case study: Genotyping of all 27 markers was performed for 469 Korean chickens. Table 4 shows the characteristics of different alleles for each marker including the repeat number, frequency, mean heterozygosity (MH) and PIC. Among the 27 loci that were polymorphic in the 20 Korean chicken breeds sampled, 16 highly polymorphic MS markers were selected based on the number of alleles, H_exp and PIC. They were segregated into two sets for multiplex-PCR based on fragment size, genetic variation and peak interpretation (Table 6). For these chicken breeds, a satisfactory expected probability of identity among the genotypes of PI, PI_half-sibs and PI_sibs was suggested to require approximately 16 markers. Thus, the expected probability of the identity values from the 16 MS markers in PI, PI_half-sibs and PI_sibs was estimated as 1.79×10^–21, 6.76×10^–16 and 5.70×10^–8 respectively. Overall, the total expected probability (PE) of the identity values was 98.89% for the discrimination of Korean chickens (Table 7). Our study identified multiplex PCR marker sets in Korean chicken breeds.

Figure 1 illustrates the population relationships based on the PCoA using the individual multilocus genotypes of the 16 MS markers, with 2 types of multiplex PCR marker sets from the 7 Korean commercial chicken breeds. With the first type, the two components contributed 29.46 and 12.05%, respectively. By the first component, the HF and HH breeds were confirmed to be clearly separated from the other groups and by the second component, the HY breed was confirmed to be clearly separated from the HS breed.

Fig. 1(a-b):	Principal components analysis of allele frequencies from (a) SET1 and (b) SET2 in 7 populations using GENETIX
	H: Hanhyup H, F: Hanhyup F, S: Hanhyup S, W: Hanhyup W, Y: Hanhyup Y, HCC: Hanhyup 3, WM: Woorimatdag

With the second type, the two components contributed 24.52 and 8.93% respectively and the intergroup relationship showed similarity with the analysis using the first type marker set. From the multidimensional scatter (MDS) plot using 600K SNP genotypes, it can be inferred that HF and HH were classified and maintained together until relatively recently, as their clusters overlap and are spread widely¹. Upon comparison, the two results in this study are similar.

Our results identified multiplex PCR marker sets in Korean chicken breeds could be applied in future breeding plans and utilized as discrimination markers for these breeds. This finding suggested that, considering the scale of the domestic chicken market, the 16 identified markers are sufficient and the maintenance of the strains of Korean chickens with appropriate discrimination markers is essential for the conservation of these breeds. Our results indicated that these multiplex PCR marker sets will have considerable applications in small- and large-scale population genetic structure analyses. In addition, they can be used to aid the conservation, traceability and future improvement of these native Korean chicken breeds.

SIGNIFICANCE STATEMENT

This study identified multiplex PCR microsatellite marker sets for chicken breeds that can be beneficial for investigating the genetic diversity of Korean chicken breeds and classifying the breeds. This study will help Korean chicken breeds to be isolated and characterized. Thus, these specific markers could aid the selection process in the development of native chickens that are more suitable for the chicken industry.

ACKNOWLEDGMENT

This study was supported by Golden Seed Project, funded by Korea Institute of Planning and Evaluation for Technology in Food, Agriculture and Forestry (IPET) [PJ012820052019 (213010053WT251)].