Extensive intra- and inter-genetic admixture of Chinese gamecock and other indigenous chicken breeds revealed by genomic data

Abstract

Genomic admixture is a widespread phenomenon among domestic animal breeds, including chickens. However, reports on admixture within Chinese gamecocks or other indigenous chickens are limited. This study focuses on the population genetic structure and admixture of 5 Chinese gamecock breeds and the admixture with 9 other indigenous Chinese chicken breeds. Our results showed that Turpan and Henan gamecocks were grouped into one cluster, whereas Luxi, Zhangzhou, and Xishuangbanna gamecocks were grouped into the other cluster. Gene flow occurred between Xishuangbanna and Turpan and Turpan and Luxi gamecocks. Simultaneously, gene flow was observed between gamecocks and indigenous chickens, such as Xishuangbanna and Wenchang. Ancestral component analysis indicated that modern domestic chickens in southern China played an important role in the history of the domestication of modern Chinese gamecock. Our study will be helpful in better understanding the domestication and evolution of Chinese gamecock.

INTRODUCTION

The complex landform, diverse ecology, and culture of China have led to breeding of abundant poultry resources. There are over 110 local breeds of chickens distributed extensively throughout China. Modern Chinese gamecock is a special derivative of cockfighting culture dating back about 2,800 yr ago (Qu et al., 2009). Gamecock breeds are characterized by extreme aggression and stronger muscles and bones compared to other indigenous chickens. According to the distinct geographical division, the main Chinese gamecock breeds are the Luxi (LX), Henan (HN), Xishuangbanna (XSBN), Zhangzhou (ZZ), and Turpan (Tur) breeds. LX and HN gamecocks are common in central China. The ZZ gamecock is found in the southeastern Fujian province. The XSBN gamecock is found in southwestern Yunnan province, and the Tur gamecock is found in Xinjiang, northwestern China. In addition to being used for entertainment, the selected characteristics of gamecocks make them of interest for behavioral and physiological research (Luo et al., 2020). According to the National Animal Husbandry Services of China, the number of XSBN, LX, HN, ZZ, and Tur gamecocks is 135,156, 50,390, 97,422, 1,920, and 825, respectively. Besides, the cross-breeding of gamecocks in the folk and the lack of good breeding measures may have led to intra- and inter-genetic admixture of Chinese gamecock.

With the method known as genomic selection, a method that involves using dense molecular markers such as single nucleotide polymorphisms (SNP) to predict the genomic value of individuals without information regarding their phenotype (Meuwissen et al., 2001), several high-density (HD) SNP genotyping arrays were developed, such as 600K Affymetrix Axiom HD genotyping array (Kranis et al., 2013). These HD SNP arrays have enabled the implementation of genomic selection in layer and broiler breeding. DNA microarray has been widely used in genetic-related studies (Drobik-Czwarno et al., 2018; Ma et al., 2018; Negro et al., 2019; Manimekalai et al., 2020; Zhang et al., 2020) because of the high SNP density and the large number of individuals that can be genotyped in a given period.

The genetic diversity of Chinese gamecocks has been reported using re-sequencing or mtDNA data (Liu et al., 2006; Luo et al., 2020; Zhang et al., 2020); however, there are few reports on admixture within gamecock breeds or the interaction between gamecocks and other indigenous chicken breeds. This study aimed to analyze the admixture within Chinese gamecock breeds and investigate the patterns of admixture between gamecock and other indigenous chicken breeds.

MATERIALS AND METHODS

Sample Collection and SNP Genotyping

We collected 333 adult chickens, including 54 gamecocks from 5 gamecock breeds and 279 from 9 other indigenous chicken breeds in China. Table 1 and Figure 1 provide detailed sampling information on all 14 populations. Blood samples were collected according to local ethical standards from the subwing vein using a standard procedure for venous blood collection. Genomic DNA of the 5 gamecock breeds, BigBone breed, and Wenchang breed were extracted from the blood using the TIANamp Blood DNA Kit DP348 (Tiangen Biotech Co., Ltd., Beijing, China) and sent to Nyogin Biotechnology Co., Ltd. (Shanghai, China) for sequencing. Microarray data of the other 7 chicken breeds were accumulated by our laboratory several years ago. The 600K Affymetrix Axiom Chicken Genotyping Array (Affymetrix, Inc., Santa Clara, CA) includes 580,961 SNPs across the entire chicken genome. Axiom Analysis Suite v4.0.1 (AxAS) software was used for genotype analysis. After genotyping, PED and MAP format files were produced for downstream analyses. SNP quality control was performed using PLINK software (v 1.90) (Purcell et al., 2007) with the following criteria: the SNP missing rate of an individual was greater than 0.1 (using the"–mind" command), the SNP missing rate of a single locus was greater than 0.1 (using the"–geno" command), and the minor allele frequency of SNP loci was less than 0.05 (using the"–maf" command). After filtering and removing SNPs of the sex chromosome, the final number of animals was 325, including 52 gamecocks, and the SNPs retained for analysis were 466,175.

Table 1.

Detailed information on the chicken samples used in our study.

Population	Geographic origin	Classification	Number
Henan Game (HN)	Henan Province, China	Fight breeds	13
Luxi Game (LX)	Shandong Province, China	Fight breeds	10
Turpan Game (Tur)	Xinjiang Province, China	Fight breeds	11
XishuangbannaGame (XSBN)	Yunnan Province, China	Fight breeds	10
Zhangzhou Game (ZZ)	Fujian Province, China	Fight breeds	10
BigBone (BB)	Liaoning Province, China	Indigenous breeds	10
Beijing You (BJ)	Beijing, China	Indigenous breeds	50
Chahua (CH)	Yunnan Province, China	Indigenous breeds	11
Hongshan (HS)	Hubei Province, China	Indigenous breeds	48
Piaoji (PJ)	Yunnan Province, China	Indigenous breeds	9
Shouguang (SG)	Shandong Province, China	Indigenous breeds	50
Taihe Silky (TS)	Jiangxi Province, China	Indigenous breeds	50
Tibetan (Tibet)	Tibet, China	Indigenous breeds	40
Wen chang (WC)	Hainan Province China	Indigenous breeds	11
Total			333

Figure 1.

Geographical location of chickens. Photos were obtained from animal genetic resources in China.

Population Structure Analysis

Genetic structure was inferred using the Admixture (v1.3.0) software (Alexander et al., 2009), which uses a maximum likelihood-based method to estimate individual ancestries from multilocus SNP genotype datasets. To explore divergence between populations, we set the predefined genetic clusters (K) ranging from 2 to 14 to cover the maximum number of lineages. The results were presented graphically using the online tool (http://pophelper.com/). The best-fit number of K was considered to hold the smallest CV error value. Next, we constructed a phylogenetic tree based on the high-quality SNPs using the SNPhylo (v 20180901) software (Lee et al., 2014), which uses a maximum likelihood method for inference of phylogeny. One thousand bootstraps were used to assess branch reliability. Beautification of the tree was implemented using the iTOL online tool (https://itol.embl.de/itol.cgi). The PLINK (v1.90) software was also used for principal component analysis (PCA) on the samples of all breeds. The online tool (https://hiplot.com.cn/ho me/index.html) was used to visualize PCA.

Test for Gene Flow

Treemix (v1.13) software (Pickrell et al., 2012) was used to test the possible gene flow, and the results were visualized with R. We ran Treemix with migration events given from 2 to 5 and generated their corresponding residual matrix. The Red Junglefowl was specified as an outgroup in the analysis.

RESULTS

The Gamecock Population Structure Described by PCA

PCA was used to evaluate the structure of gamecock populations (Figure 2A) and overall populations (Figure 2B). For the 5 gamecock populations, the first 2 principal components (PCs) captured 9.50% and 3.63% of the variance, respectively. The clustering in the PCA of the genome-wide SNP data illustrated the close relationship between the HN, LX, and Tur gamecock populations. The ZZ and XSBN gamecock populations were outliers. For overall populations in this study, the first 2 PCs captured 25.04% and 21.65% of the variance, respectively. Gamecocks showed close relationships with all indigenous populations except TS, SG, and BJ populations.

Figure 2.

Principal component analysis of (A) gamecock breeds and (B) all breeds.

Population Admixture Analysis Further Verified the Relationship Between All Populations

The results of the admixture analysis for all populations, with a range of 2 to 14 potential clusters, showed that the best-fitting number of populations present in the total sample was K = 12 (Figure 3A, Supplementary Figure 1). When K increased from 5 to 12 (Supplementary Figure 2), all the individuals of the TS and most of BJ populations were separated from the other 11 populations. This was consistent with the population structure indicated by the PCA (Figure 2B). Notably, when K ranged from 6 to 12, CH and about half of Tibet individuals showed similar ancestral components. Although gamecocks showed complicated ancestral components, ZZ, LX, and HN gamecocks had relatively simple ancestral components at K = 3, 4 and 11, 12 compared to Tur and XSBN gamecocks. To better understand the genetic history within gamecock populations, we did an admixture analysis only using the data of gamecocks. At K = 2 (Figure 3B), which showed the minimal CV error value (Supplementary Figure 3), the genetic background of ZZ gamecock was simpler than that of the other 4 populations, which agrees with the results of PCA.

Figure 3.

Admixture results of (A) all breeds at K = 12 and (B) gamecock breeds at K = 2.

Maximum-Likelihood Tree Reflected the Phylogenetic Relationship of All Populations

A phylogenetic tree was constructed using the genome-wide SNP data of all the populations. The tree showed 4 genetic clusters (Figure 4). One cluster included TS and SG populations. Another cluster included all the samples of BB and HS populations and 1 sample of the WC population. The third cluster comprised indigenous and gamecock populations, including all BJ individuals, 10 Tur, 11 HN, and 1 LX individual. Cluster 4 included all Tibet, XSBN, ZZ, PJ, and CH chickens and 1 Tur, 9 LX, and 6 WC chickens.

Figure 4.

Maximum likelihood tree of the clustering pattern among all populations in this study.

Treemix Analysis Revealed Intra- and Inter-Gene Flows of Gamecocks

The population structure and phylogenetic relationship analysis suggested that admixture existed extensively within gamecock populations and between indigenous chicken populations. To better infer the admixture events and understand the phylogenetic relationship, we employed Treemix to construct an ML tree using an RJF as an outgroup. The results of the migration event at 3 are shown in Figure 5. The tree showed a similar phylogenetic relationship to the phylogenetic tree above (Figure 4). When migration events ranged from 2 to 5 (Supplementary Figure 4), 2 gene flows were found between XSBN and Tur populations and Tur and LX populations. Moreover, a gene flow was observed between XSBN and WC populations.

Figure 5.

Maximum likelihood tree inferred from all populations with 3 migration edges to detect gene flow. Migration weight is indicated according to the color of the arrows. The scale bar denotes 10 times the average standard error of the entries in the sample covariance matrix.

DISCUSSION

Before making selection and breeding plans and implementing chicken conservation programs, understanding the population’s genetic diversity and structure is crucial. Admixture event between different chicken populations is a key factor in the formation of genetic diversity. The present study's results suggested that over the breeding course of more than 2,800 yr, admixture occurred within gamecock populations and between gamecock and other indigenous chicken populations, consciously or unconsciously.

During the breeding of Chinese gamecock, cross-breeding with other chicken populations was often avoided to preserve its unique characteristics. As a result, gamecocks are on a relatively independent branch among indigenous chicken populations. In this study, 5 gamecock populations were grouped into 2 clusters: one included Tur and HN populations, whereas the other included LX, ZZ, and XSBN populations. The result was not exactly similar to the results of previous studies based on mtDNA data (Liu et al., 2006; Qu et al., 2009). According to the archaeological and ethnological records in China, with the migration of humans, old HN and LX gamecocks around the Wei and Yellow Rivers moved to Yunnan province, and some of them differentiated into XSBN gamecocks. HN, LX, and XSBN were brought to Turpan and surrounding areas in Xinjiang province, and after a long domestication period, the Tur gamecock was formed. This migration path was also confirmed in our admixture and Treemix results (XSBN and Tur and Tur and LX gamecocks had similar ancestral components); we also observed gene flow between XSBN and Tur and Tur and LX gamecocks. The relatively independent ancestral component of modern ZZ gamecocks can be explained by the breeding of hybridization between indigenous gamecocks and those mainly from Thailand about 200 yr ago.

Similar ancestral components were found in gamecocks and indigenous chicken populations in southern China, including CH, PJ, and WC populations and some of Tibet, XSBN, and LX populations. Gene flow was also observed between WC and XSBN populations. This suggests that modern domestic chickens in southern China played an important role in the history of the domestication of modern Chinese gamecock.

In conclusion, based on the 600K microarray data of 333 chickens, including 54 gamecocks, we found that admixture and gene flow existed within 5 Chinese gamecock breeds and between other indigenous chickens. Tur and HN gamecocks were grouped into one cluster, while LX, ZZ, and XSBN gamecocks were grouped into another cluster. Gene flow occurred between XSBN and Tur and Tur and LX gamecocks. Simultaneously, gene flow was observed between gamecocks and indigenous chickens, such as XSBN and WC populations. Ancestral component analysis indicated that modern domestic chickens in southern China played an important role in the history of the domestication of modern Chinese gamecock. Our results will help to better understand the evolutionary history of modern Chinese gamecock and are of significance to provide some references for designing and implementing conservation strategies of Chinese gamecocks.