Genetic diversity of the Chinese goat in the littoral zone of the Yangtze River as assessed by microsatellite and mtDNA

Abstract The objective of this study was to assess the genetic diversity and population structure of goats in the Yangtze River region using microsatellite and mtDNA to better understand the current status of those goat genetic diversity and the effects of natural landscape in fashion of domestic animal genetic diversity. The genetic variability of 16 goat populations in the littoral zone of the Yangtze River was estimated using 21 autosomal microsatellites, which revealed high diversity and genetic population clustering with a dispersed geographical distribution. A phylogenetic analysis of the mitochondrial D‐loop region (482 bp) was conducted in 494 goats from the Yangtze River region. In total, 117 SNPs were reconstructed, and 173 haplotypes were identified, 94.5% of which belonged to lineages A and B. Lineages C, D, and G had lower frequencies (5.2%), and lineage F haplotypes were undetected. Several high‐frequency haplotypes were shared by different ecogeographically distributed populations, and the close phylogenetic relationships among certain low‐frequency haplotypes indicated the historical exchange of genetic material among these populations. In particular, the lineage G haplotype suggests that some west Asian goat genetic material may have been transferred to China via Muslim migration.


| INTRODUC TI ON
The goat (Capra hircus) is a culturally and economically important domestic animal worldwide. The goat is also one of the major landmarks in the evolution of human modes of subsistence (Doro et al., 2014;Zeder, 2008).
The genetic diversity of several types of Chinese goats has also been reported, including the Chinese dairy goat (Wang et al., 2015), goats indigenous to southwest China (Wei et al., 2014;Zhao et al., 2014), the black goat (Miao et al., 2015;Zhong et al., 2013), and the cashmere goat (Di et al., 2011;Liu et al., 2013).
Most studies primarily consider a relatively small group of breeds within an administrative area; however, due to geographical isolation and adaptation to different nutrient supplies and climates, the population structure always follows the geographical distribution and management history (Ling et al., 2012;Yadav et al., 2015). Human social activities play an equally important role in shaping genetic diversity and the evolution of the population genetic structure of domestic animals.
The Yangtze River, which is the longest river in Asia and the third longest river in the world, plays an important role in the archeology, civilization, and economy of China, particularly in the upstream region of the Yangtze River (Jin et al., 2012). For thousands of years, the Yangtze River has been used for transportation and boundary-marking. Moreover, due to a multitier transport network comprising water, railways, and roads, a new economic belt was created alongside the river. This transport network has driven a new round of human mass migration and cultural exchange. In addition, the upstream region of the Yangtze River is the origin of indigenous domesticated goats and human civilization in China (China National Commission of Animal Genetic Resources, 2011).
In this study, we estimated the molecular genetic diversity and population phylogenetic structure of domestic goat populations in the upstream region of the Yangtze River using microsatellites and D-loop DNA to better understand that Yangtze River, as natural landscape and carrier of human civilization development, what is its influence in shaping the genetic diversity of domestic animal. In addition, the aim of this research was also to identify the diversity among breeds and populations, and provide basis for genetic resources conservation strategy in native goats in this area.

| Animals and microsatellite marker methods
The experimental conditions in this study were approved by the Committee on the Ethics of Animal Experiments of Southwest University (No. [2007] 3) and the Animal Protection Law of China.
In total, 426 individual blood samples from goats in 16 populations in the littoral zone of the Yangtze River (Table 1) were collected in EDTA tubes, and a standard phenol:chloroform protocol (Sambrook & Russell, 2001) was used to extract genomic DNA; a 0.8% agarose gel was used to assess the DNA quality, and a DTX microplate reader (Beckman Coulter, US) was used to quantify the extracted DNA.
All goats were genotyped using twenty-one microsatellite markers (Table 2) as recommended by FAO (FAO, 2011). The genotyping was conducted using PCR protocols and a 3130xl Genetic Analyzer (Applied Bio Systems) as previously described by E et al. (2016).
The mean number of alleles (N A ), observed (H o ) heterozygosity, expected heterozygosity (H E ), polymorphism information content (PIC), and inbreeding coefficient (F IS ) were estimated using FSTAT 2.9.3.2 (Goudet, 1995) and Microsatellite Toolkit software (Park, 2001). To identify possible deviations from Hardy-Weinberg equilibrium (HWE), we used Fisher's exact test with Bonferroni correction using GENEPOP 3.4 software (Raymond & Rousset, 1995). Pairwise differences in populations (F ST ) (Slatkin, 1995) were identified using R-lequin software version 3.5.1.3 (Excoffier & Lischer, 2010). Phylogenetic neighbor-joining trees were derived from the Reynolds genetic distance using the PHYLIP software package (Felsenstein, 2005). The Reynolds genetic distance among populations was visualized using a Neighbor-Net generated using SPLITSTREE4 software (Huson & Bryant, 2006). The Bayesian clustering algorithm was implemented in STRUCTURE 2.3.3 (Falush, Stephens, & Pritchard, 2003) to determine the population structure and explore the assignment of individuals and populations to specific gene clusters using a burn-in of 50 000, followed by 100,000 Markov Chain Monte Carlo (MCMC) iterations from K = 1 to K = 16 in 100 runs, and a merger of these runs within populations. In particular, the lineage G haplotype suggests that some west Asian goat genetic material may have been transferred to China via Muslim migration. each of the K values obtained from CLUMPP software (Jakobsson & Rosenberg, 2007) was visualized using DISTRUCT 1.1 software (Rosenberg, 2004). A graphical display of the simulated results and the optimal K value were generated using STRUCTURE_Harvester (Earl & vonHoldt, 2012).

| Animals and experimental methods using mtDNA D-loop variants
The mtDNA D-loop variants of 326 related goats are shown in Table 6.
The high-variability region in the mtDNA control region was amplified using the primers CAP-F (5′-CGTGTATGCAATGACATAC-3′) and CAP-R (5′-CTGATTAGTCATTAGTCCATC-3′) . The PCR amplification was conducted using a Bio-RAD T100 ™ Thermal Cycler (Bio-Rad Laboratories Pty. Ltd, New South Wales, Australia) with a total reaction volume of 50 μl, containing 150 ng of DNA, 25 μl of 2× PCR-Mix (BioMed, Beijing, China), 20 pmol/L each of the forward and reverse primer, and ddH 2 O to bring the volume up to 50 μl. The PCR protocol was as follows: an initial denaturation step at 95°C for 3 min, followed by 28 cycles of 95°C for 30 s, 58°C for 30 s, and 72°C for 1 min. The final cycle was followed by a 72°C extension for 10 min. Tanyibiotech (Wuhan, China) performed the sequencing using CAP-F.
The sequence alignments of the D-loop region in all goats were constructed using Clustal X software (1.83) (Cummings, Neel, & Shaw, 2008) with GenBank No. AF533441 as the reference sequence. DnaSP5.10 (Librado & Rozas, 2009) was conducted to screen for haplotypes and polymorphisms. Bayesian inference (BI) and maximum likelihood (ML) frameworks were used to examine the phylogenetic relationships. The best fitting model of DNA substitution for BI (the Hasegawa-Kishino-Yano model (G + I)) was obtained using jModelTest V. 0.1.1. software (Posada, 2008). The neighborjoining phylogenetic network was rebuilt using MEGA (6.0) software (Tamura, Stecher, Peterson, Filipski, & Kumar, 2013), and the bootstrap values to support the nodes of the tree were based on 1,000 iterations of the heuristic search. Twelve reference sequences (lineage A (AY155721, EF618134, and EF617965), lineage B (EF617965 and DQ121578), lineage C (AY155708 and DQ188892), lineage D (AY155952 and DQ188893), lineage F (DQ241349 and DQ241351), and lineage G (EF618084)) from a previous study (Naderi et al., 2007) were used in this study. All sequences generated in this study were deposited in the GenBank database (KX660779-1003, KU891395-494).

| Diversity estimation and population structure of Chinese goat in the littoral zone of the Yangtze River using microsatellite analysis
Across all 16 goat populations in the littoral zone of the Yangtze River (see Table 1), 435 alleles were found by analyzing 21 microsatellite markers (  (Table 4).
On average, each locus deviated from HWE in 7.90 populations.
The most extreme markers, that is, ILSTS005 and SRCRSP7, deviated from HWE in 14 populations (Table 3). In addition, the number TA B L E 2 Primer information for twenty-one microsatellites as recommended by the FAO  (Table 4 and Appendix S1). Among all populations, the mean number of private alleles was 6.57 per marker, and the range was from 1 (MAF209) to 19 (INRA023).
In the F ST analysis, the greatest divergence was observed between HW and EB (0.549), and the smallest divergence was observed between WG and BJ (0.025, Table 5). Thus, the distribution of F ST did not reveal a significant divergence among different ecogeographically distributed populations. However, the populations in closer distribution locations, particularly those in southwest China (i.e., Chongqing, Sichuan, and Yunnan), had a significantly lower divergence compared with that in populations from geographically distant populations (Table 5, Figure 1a). Hubei and separated into Cluster VII, (Figure 1b).
The clustering of individuals into 1 ≤ K ≤ 16 was estimated using the STRUCTURE software by setting the optimal K value to 2 and 8 (Appendice S2). At K = 2, the populations were separated into two different groups; the first group comprised YW, EB, MG, CN, and JY, and the other group comprised the remaining populations. At K = 8, the population structure was separated into seven group, which was similar to the groups separated using the neighbor-joining tree.
The population structure analysis using both STRUCTURE and the TA B L E 3 Genetic diversity of 21 microsatellite locus across sixteen goat populations in the littoral zone of the Yangtze River phylogenetic neighbor-joining network revealed a population structure without a unanimous ecogeographical distribution (Figure 1c).

| Diversity estimation and population structure of Chinese goat in the littoral zone of the Yangtze River using mtDNA D-loop
To analyze the mtDNA D-loops of domestic goats in the littoral zone of the Yangtze river, samples from 326 goats from 13 populations were obtained, and the sequences were 582 bp in length as published in the GenBank database (Table 6). A meta-analysis of 168 mtDNA D-loop sequences from nine other populations was conducted using previous reports and data obtained in this study to generate a larger sample size (   with systematic nonrandom mating. In addition, it is well known that inbreeding occurs within populations when some individuals in large populations separate into smaller flocks or are improperly managed by humans. Thus, nonrandom mating should be the major reason for the deviation from HWE and high F IS .

| D ISCUSS I ON
In addition, the sample size, as one of reason above, in this study could not completely represent the real diversity of each population or other types of population genetic events that occurred within the goat populations. Moreover, certain high-frequency private alleles within populations indicated that these populations could have independently undergone long-term domestication history (Granevitze et al., 2007). Last but not least important, the null alleles could also contribute to the deviation due to incorrect genotyping, particular like 1 bp difference between the large fragments in traditional gel conditions may not sufficiently accurate to separated, which cause some different allele ignored. However, the classification platform used in this study was sufficient to identify the genotype of 1 bp differences as well known.
According to the pairwise differences (Slatkins linearized F ST ) in this study, the distribution of F ST showed low genetic divergence (not significant) among populations in general. The EB population, which is an indigenous breed of goats that live on the Enshi Grand flow and migration among populations occurred with a high frequency in the Chongqing, Sichuan, and Yunnan goat populations.
The STRUCTURE analysis showed a clear clustering of these 16 goat populations, which was similar to the neighbor-joining network pattern described above using the Reynolds genetics distance. Based on a microsatellite marker analysis, the results elucidated strong gene flow and genetic material exchange among different regions in the littoral zone of the Yangtze River, which was likely caused by human migration, commercial trade, and extensive transport during the long-term domestication of goats.
To further discover possible evidence of gene flow and genetic material exchange among the goat populations in the upstream re-  (Lin et al., 2013;Zhao et al., 2014). Four haplotypes (H_50, H_52, H_34, and H_44) from four goats were lineage C, and four haplotypes (H_38, H_45, H_153, and H_151) from nine animals were lineage D. In a previous study, lineage C and lineage D were observed to have a low frequency in Tibetan goats (Liu, Lei, Liu, & Yang, 2007); these data are consistent with our data, which showed that nearly all goats from the QH and CM populations and one goat from the DZ population belonged to lineages C and D. Previous studies have indicated that the lineage C haplotype is not found in the DZ population Zhong et al., 2013), but the lineage C haplotype has been found in the Yangtze River delta white goat (YDW, Jiangsu, China) in the littoral zone of the Yangtze River (Liu, Cao, Chen, Yao, & Liu, 2009). In addition, the lineage D haplotype, which was only observed in the QH and CM populations in this study, has been detected in goats from YDW, Inner Mongolia, and Taihang (Liu et al., 2009;Zhong et al., 2013 (Muhammad, 2016). In addition, Qinghai is located in the center of China's western region and is at the intersection of three major arteries, including the Silk Road, Tang-Fan Road, and Ancient Tea Road. Due to its vast natural pasture and cultivated land, Qinghai is not only the birthplace of many ethnic groups but is also a breeding and blood convergence ground for the development of many religions, such as Islam, which is one of the earliest religions spread in our country (Li, 2016;Yang, 2016). In particular, during the Song and Yuan Dynasties (AD 960-1368), a large number of Muslim businessmen moved to Qinghai, and until the Ming Dynasty, Islam was one of the core religions in Qinghai (Lv & Wang, 2006). Therefore, the infiltration of Muslim culture and the migration of Muslim groups not only resulted in Qinghai indigenous cultural exchange but also, unsurprisingly, in changes in agriculture and animal husbandry.
In summary, we estimated the diversity of large-scale Chinese goat populations in the littoral zone of the Yangtze River using microsatellite and partial mtDNA D-loop variants. These showed a high diversity level within related goat populations. According to their population structure (based on the mtDNA and microsatellite markers), there was a dispersed geographical clustering, indicating that a high frequency of genetic material exchange and gene flow among these populations of goats occurred. In particular, we discovered lineage C mtDNA in the southwestern goat populations and lineage G mtDNA in the Qinghai goat population, which was inconsistent with previous studies. Based on our results, we infer that high anthropic movement and migration related to the acclimation process of domestic animals occurred. Therefore, the results of this study are helpful for the understanding of the domestication process of domestic animals in South China and can be used to support new, planned strategies for their conservation.

CO N FLI C T O F I NTE R E S T
None declared.