GENETICS AND GENOMICS

Transcription

1 GENETICS AND GENOMICS Molecular phylogenetic analysis of Chinese indigenous blue-shelled chickens inferred from whole genomic region of the SLCO1B3 gene Seyed Benyamin Dalirsefat, Xianggui Dong, and Xuemei Deng 1 National Engineering Laboratory for Animal Breeding and Key Laboratory of Animal Genetics, Breeding, and Reproduction of the Ministry of Agriculture, China Agricultural University, Beijing , China ABSTRACT In total, 246 individuals from 8 Chinese indigenous blue- and brown-shelled chicken populations (Yimeng Blue, Wulong Blue, Lindian Blue, Dongxiang Blue, Lushi Blue, Jingmen Blue, Dongxiang Brown, and Lushi Brown) were genotyped for 21 SNP markers from the SLCO1B3 gene to evaluate phylogenetic relationships. As a representative of nonblue-shelled breeds, White Leghorn was included in the study for reference. A high proportion of SNP polymorphism was observed in Chinese chicken populations, ranging from 89% in Jingmen Blue to 100% in most populations, with a mean of 95% across all populations. The White Leghorn breed showed the lowest polymorphism, accounting for 43% of total SNPs. The mean expected heterozygosity varied from 0.11 in Dongxiang Blue to 0.46 in Yimeng Blue. Analysis of molecular variation (AMOVA) for 2 groups of Chinese chickens based on eggshell color type revealed 52% within-group and 43% between-group variations of the total genetic variation. As expected, F ST and Reynolds genetic distance were greatest between White Leghorn and Chinese chicken populations, with average values of 0.40 and 0.55, respectively. The first and second principal coordinates explained approximately 92% of the total variation and supported the clustering of the populations according to their eggshell color type and historical origins. STRUC- TURE analysis showed a considerable source of variation among populations for the clustering into blueshelled and nonblue-shelled chicken populations. The low estimation of genetic differentiation (F ST ) between Chinese chicken populations is possibly due to a common historical origin and high gene flow. Remarkably similar population classifications were obtained with all methods used in the study. Aligning endogenous avian retroviral (EAV) HP insertion sequences showed no difference among the blue-shelled chickens. Key words: blue-shelled chicken, SLCO1B3 gene, EAV HP insertion, SNP marker, phylogenetic analysis 2015 Poultry Science 94: INTRODUCTION Eggshell color is both of biological interest and of economic importance, and the nature of the shell pigments in various avian species and biochemical and physiological processes involved in pigment formation and its deposition in and on the shell have been discussed since the 19th century (Wicke, 1858;Sorby,1875; Krukenberg, 1883; Lang and Wells, 1987). Because it is an important aspect of egg quality in many countries, shell color has been emphasized in the sales, promotion, and marketing strategies of egg retailers. Recently, avian egg color has been explained as mainly serving crypsis or mimetism (Underwood and Sealy, 2002). Blue green colors in avian eggs have also been proposed as post-mating signals of female C 2015 Poultry Science Association Inc. Received December 6, Accepted April 14, Corresponding author: deng@cau.edu.cn phenotypic quality to their mates. Morales et al. (2006) suggested that egg color may not only indicate female value, but also the quality of the eggs themselves and of the resulting offspring. Blue green eggshell color in chickens is produced by an autosomal dominant trait called oocyan, and eggs laid by oocyan homozygotes are a darker blue than those from heterozygotes (Punnett, 1933). SLCO1B3 codes a membrane transporter OATP1B3, which is considered a liver specific transporter and is highly expressed in liver, where it transports a wide range of substrates, including bile salts (Popovic et al., 2010; Hagenbuch and Gui, 2008). Wang et al. (2013) suggested that, as blue eggshell is colored mainly by deposition of biliverdin, which is one component of bile salts, expression of SLCO1B3 in the uterus could enhance transportation of biliverdin to the eggshell. They also found that SLCO1B3 is expressed in the shell gland of blue-shelled chickens and not in the glands of brown- or white-shelled chickens, supporting an essential role of the gene in pigmentation of blue eggs. 1776

2 PHYLOGENETIC ANALYSIS OF CHINESE CHICKENS 1777 Figure 1. Geographical locations of the 6 indigenous chicken populations sampled in China Recently, in 2 independent studies, an endogenous avian retroviral (EAV) HP insertion has been discovered in 2 Chinese (Dongxiang and Lushi), an American (Mapuche fowl), and a European (Araucana) blueshelled chicken breeds, associated with the overexpression of SLCO1B3, strongly suggesting this is a causative mutation for the blue eggshell phenotype (Wang et al., 2013; Wragg et al., 2013). In addition, conserved EAV HP integration sites and sequences have been found in South American and European blue-shelled chickens, distinct from those of the Asian chicken, which implies independent integration events in the blue-shelled chickens from the 2 groups of the 3 continents and provides an example of parallel evolution at the molecular level. China has a wide variety of indigenous poultry, with 108 native chicken breeds (Chen et al., 2004), many of which have valuable genetic features. Most of these chickens are local and fancy breeds characterized by medium to low performance and are usually maintained in small populations. The native chicken varieties include some blue-shelled breeds; Dongxiang and Lushi are 2 indigenous blue-shelled chicken breeds reported in recent studies (Zhao et al., 2006; Wang et al., 2013). However, the origin of some local blue-shelled chickens in China remains unclear. Phylogenetic analysis and breed characterization requires basic knowledge of genetic variations that can be effectively measured within and between populations. In recent years, analysis of SNP markers has become the standard approach for diversity analysis and genome-wide studies. Because of their abundance in the genome, genetic stability, and amenity to high-throughput automated analysis, SNP represent one of the more interesting approaches for genotyping (Vignal et al., 2002). The usefulness of SNP in analyses of population diversity and structure has been demonstrated in several studies (McKay et al., 2008; Lin et al., 2010; Edea et al., 2013). Furthermore, with the progress of sequencing technology, whole-genome/gene sequencing has become available for characterizing genetic diversity among farm animals (Yang et al., 2013). This is an improvement on using microsatellites and microarrays because it can detect both SNP and structural variations. The present study was undertaken to analyze the molecular phylogenetic analysis of 8 Chinese indigenous blue- and brown-shelled chickens by using SNP markers inferred from resequencing the whole genomic region of SLCO1B3 and its inserted EAV HP sequence information. MATERIALS AND METHODS Sample Collection and DNA Extraction Eight Chinese indigenous blue- and brown-shelled chicken populations were used in the study. All samples were collected from breed conservation farms in the related regions in China (Figure 1). The breeds were Yimeng Blue from Shandong Province (N = 45), Wulong Blue (N = 45) from Chongqing Municipality, Lindian Blue (N = 40) from Heilongjiang Province, Jingmen Blue (N = 42) from Hubei Province, Dongxiang Blue (N = 28) and Dongxiang Brown (N = 29) from Jiangxi Province, and Lushi Blue (N = 29) and Lushi Brown (N = 30) from Henan Province. As a representative of a nonblue and brown eggshell breed, White Leghorn (N = 30) was also included for reference. Although the

3 1778 DALIRSEFAT ET AL. populations are termed as blue-shelled chicken breeds, the eggshell color trait has not been fixed. Historically, Dongxiang and Lushi chickens have been selected for blue eggshell. The other populations have not been systematically bred until now, so some appearance traits, eggshell color, and feather color do not show homogeneity. Total genomic DNA was extracted from blood using a TIANamp blood DNA kit. The DNA concentration and its purity were examined using spectrophotometric analysis based on absorbance at 260 and 280 nm and agarose gel electrophoresis analysis. Diagnostic Genotyping Test of EAV HP Insertion The retrovirus insertion was genotyped by multiplex PCR using 3 primers (test-nor-up, test-no-down, and test-eav) introduced by Wang et al. (2013). The PCR amplifications were performed in a total volume of 20 μl containing 10 μl 2XTaq PCRMix (TIANGEN Biotech Beijing Co., Ltd.), 0.25 μm each primer, and 50 to 100 ng genomic DNA in the following conditions: 95 C for 5 min, followed by 36 cycles of 95 C for 30 s, 56 C for 30 s, 72 C for 20 s, and a final extension at 72 C for 5 min. The PCR products were separated by 2% agarose gel electrophoresis, and the length of target fragments were 340 bp for test-nor-up and testnor-down, and 425 bp for test-nor-up and test-eav. Resequencing of EAV HP Insertion The EAV HP insertion (approximately 4.2 kb) upstream of SLCO1B3 was amplified using the method described by Wang et al. (2013). With the exception of the Dongxiang Blue, which has been previously sequenced by Wang et al. (2013; GenBank Accession No.: JF837512), we sequenced the EAV HP insertion of all blue-shelled chicken populations (GenBank Accession No.: KP256532, KP276576, KP276577, KP276578, and KP for Yimeng Blue, Wulong Blue, Lindian Blue, Lushi Blue, and Jingmen Blue, respectively). MassARRAY Analysis Twenty-one SNPs found by resequencing of SLCO1B3 (Wang et al., 2013) were used to analyze the genetic variants of SLCO1B3 in the 9 populations. SNP markers were genotyped using an iplex SE- QUENOM MassARRAY platform (Sequenom, CA). This genotyping system uses single-base extension reactions to create allele-specific products that are separated automatically and scored in a matrixassisted laser desorption ionization/time-of-flight mass spectrometer. Primer design was performed using MassARRAY Assay Design software (v3.1) according to Sequenom s instructions. Multiplex PCR amplification of amplicons containing SNP of interest was performed using HotStart Taq Polymerase (Qiagen, CA) with 12 ng genomic DNA. Assay data were analyzed using Sequenom TYPER software (v3.4). Statistical Analysis Genetic diversity Total number of alleles, number of effective alleles, allele frequencies, observed (Ho) and expected (He) heterozygosity (Nei, 1973), Shannon s information index, and fixation index for each population across the loci were estimated using the programs Arlequin (Excoffier et al., 2005) and GenAlEx (Peakall and Smouse, 2012). The fixation index F (also called the inbreeding coefficient) can take values ranging from 1 to +1. Values close to zero are expected under random mating, while substantial positive values indicate inbreeding or undetected null alleles. Negative values indicate excess of heterozygosity due to negative assortative mating or selection for heterozygotes. Deviation from Hardy Weinberg equilibrium (HWE; heterozygote deficiency) was assessed by performing a chi-square test with the PowerMarker program (Liu and Muse, 2005) for each marker and population. Analyses of molecular variance Analyses of molecular variance (AMOVA) were carried out on 3 data sets using the program GenAlEx (Peakall and Smouse, 2012). The first analysis included data only for 8 Chinese chicken populations in one group; the second data set consisted of the Chinese chickens grouped by the 2 eggshell color types, and a third analysis was performed for all 9 populations in three groups (i.e., 6 Chinese blue eggshell, 2 Chinese brown eggshell, and White Leghorn chicken populations). Genetic differentiation The Wright (1978) fixation indices were estimated according to Weir and Cockerham (1984) using the program GenAlEx (Peakall and Smouse, 2012). The significance of fixation indices and pairwise population differentiation values were determined by permutation tests with 1,000 permutations. Relationships among the populations Reynolds genetic distance (Reynolds et al., 1983) between different pairs of chicken populations was calculated using PowerMarker (Liu and Muse, 2005). The unweighted pair-group method with arithmetic mean (UPGMA; Sneath and Sokal, 1973) algorithm was used to construct the dendrogram from Reynolds matrices using the same software. The generated tree was visualized in Mega tree explorer (Tamura et al., 2011). Gene flow between populations, defined as the number of reproductively successful migrants per generation (Nm), was calculated using the program GenAlEx (Peakall and Smouse, 2012). The estimate was based on the relationship F ST = 1/(4Nm +1),whereN is the effective population size, m is the migration rate, and F ST is calculated as the mean over loci. Principal coordinate analysis A principal coordinate analysis (PCoA) was carried out to illustrate the relationships among the populations using the program GenAlEx (Peakall and Smouse, 2012) based on

4 PHYLOGENETIC ANALYSIS OF CHINESE CHICKENS 1779 Table 1. Genetic variability within chicken populations 1. SNP (%) not Polymorphic in HWE Population N Na (SE) Ne (SE) H ob (SE) H ex (SE) I (SE) markers (%) (P 0.05) F (SE) Yimeng Blue (0.00) 1.9 (0.03) 0.75 (0.03) 0.46 (0.01) 0.65 (0.01) (0.04) Wulong Blue (0.00) 1.5 (0.03) 0.39 (0.03) 0.31 (0.02) 0.48 (0.03) (0.02) Lindian Blue (0.00) 1.8 (0.03) 0.72 (0.03) 0.45 (0.01) 0.64 (0.01) (0.04) Dongxiang Blue (0.07) 1.1 (0.01) 0.12 (0.01) 0.11 (0.01) 0.21 (0.02) (0.00) Dongxiang Brown (0.07) 1.3 (0.04) 0.07 (0.02) 0.22 (0.02) 0.37 (0.03) (0.05) Lushi Blue (0.05) 1.3 (0.03) 0.30 (0.02) 0.25 (0.02) 0.40 (0.03) (0.01) Lushi Brown (0.00) 1.4 (0.08) 0.21 (0.04) 0.23 (0.04) 0.36 (0.05) (0.07) Jingmen Blue (0.08) 1.3 (0.03) 0.27 (0.02) 0.22 (0.02) 0.37 (0.03) (0.00) White Leghorn (0.11) 1.4 (0.10) 0.15 (0.04) 0.19 (0.05) 0.27 (0.07) (0.01) Mean 1.9 (0.06) 1.4 (0.08) 0.33 (0.02) 0.27 (0.01) 0.42 (0.02) (0.03) 1 Na = No. different alleles, Ne = No. effective alleles, H ob = Observed heterozygosity, H ex = Expected heterozygosity, I = Shannon s information index, and F = Fixation index. an algorithm published by Orlóci (1978). Two methods of PCoA were carried out to determine population relationships based on the F ST matrix and the covariance matrix, with data standardization by using a multivariate technique that allows one to find and plot the major patterns within a multivariate data set (e.g., multiple loci and multiple samples). Structure analysis For the analysis of population structure, a model-based clustering method using genotype data consisting of unlinked markers was implemented with the program STRUCTURE (Pritchard et al., 2000, 2010). This program can determine the existence of population structure, ascertain distinct genetic populations, assign individuals to populations, and identify migrants and admixed individuals. It is assumed that within populations the loci are at HWE and linkage equilibrium. In the present study, a Monte Carlo Markov chain algorithm was used to estimate allele frequencies in each of the K populations and the degree of admixture for each individual. The true number of clusters (K) wasinferredbythe program STRUCTURE HARVESTER (Earl and von- Holdt, 2012) implementing the Evanno (2005) method, using 10 independent runs with 100,000 iterations following a burn-in period of 20,000, under an admixture ancestry and correlated allele frequency model with K ranging from 2 to 10. The estimated cluster membership coefficient matrices of multiple runs were used as the input file for finding optimal alignments of R replicate cluster analyses of the same data using the program CLUMPP (Jakobsson and Rosenberg, 2007). The most frequent solution for each K was taken as the most probable clustering and visualized using DISTRUCT software (Rosenberg, 2004). EAV HP insertion data analysis Because the EAV HP insertion does not occur in White Leghorn, 2 blue-shelled chicken breeds from South America (Chile, Mapuche fowl) and Europe (France, Araucana) reported by Wragg et al. (2013) were included in this analysis as outgroups. Multiple alignment of the EAV HP insertion sequences of all blue-shelled chickens under study and Mapuche/Araucana (GenBank Accession No.: KC632578) was done using the online Basic Local Alignment Search Tool program of the National Center for Biotechnology Information. RESULTS We used fixation indices (F ST, F IS,andF IT ) as primary metrics for empirically estimating and testing the magnitude of genetic divergence among populations (Table S1). The fixation coefficients of subpopulations within the total population (F ST ) when all populations were analyzed for the 21 loci varied from 0.12 (SNP1B3 4) to 0.54 (SNP1B3 6), with a mean of 0.41 (P < 0.001). All loci contributed significantly to this differentiation. The average overall deficit of heterozygotes across populations (F IT ) was 0.29 (P < 0.001) and the mean F IS was 0.22 (P > 0.05) within populations. All loci showed significant excess of heterozygotes. Numbers of different and effective alleles, ratio of polymorphic markers, Shannon s information index, genetic variability within the different chicken populations, fixation index, and departures from HWE are shown in Table 1. Among Chinese chicken populations, the proportion of polymorphic SNP ranged from 86% in Jingmen Blue to 100% in Yimeng Blue, Wulong Blue, Lindian Blue, and Lushi Brown, with a mean of 95%. The White Leghorn breed showed the lowest proportion of polymorphism (43% polymorphic loci). The mean expected heterozygosity over all loci for each population varied from 0.11 (0.01) in Dongxiang Blue to 0.46 (0.01) in Yimeng Blue. The highest observed heterozygosity was similarly found in the same population (0.75 ± 0.03 in Yimeng Blue), whereas Dongxiang Brown showed the lowest observed heterozygosity (0.07 ± 0.02). The mean values of observed and expected heterozygosities in White Leghorn were 0.15 (0.04) and 0.19 (0.05), respectively. Three populations showed an overall significant deficit of heterozygotes, while 6 populations showed an excess of heterozygous genotypes with respect to the expected value. Effective number of alleles and Shannon s information index ranged from 1.1 (0.01) and 0.21 (0.02) in Dongxiang Blue to 1.9 (0.03) and 0.65 (0.01) in Yimeng Blue, respectively. The highest and the least values of fixation index (F; inbreeding

5 1780 DALIRSEFAT ET AL. Table 2. Analysis of molecular variation (AMOVA) analysis results using different data sets of Chinese chicken populations and White Leghorn. Data set Variance component (%) Among groups Among populations Among population within groups Only Chinese chickens in one group Chinese chickens based on 2 eggshell color groups All 9 populations in 3 groups (Chinese Blue eggshell, Brown eggshell, and White Leghorn group) Within groups Table 3. Pairwise genetic differentiation (F ST ) and gene flow (Nm; in parenthesis) values between the 9 chicken populations (below diagonal) and Reynolds genetic distance (above diagonal). Population Yimeng Wulong Lindian Dongxiang Dongxiang Lushi Lushi Jingmen White Blue Blue Blue Blue Brown Blue Brown Blue Leghorn Yimeng Blue Wulong Blue (5.2) Lindian Blue (64) (6.4) Dongxiang Blue (1.3) (4.8) (1.5) Dongxiang Brown (0.95) (0.38) (0.78) (0.16) Lushi Blue (2.9) (21) (3.4) (9.6) (0.29) Lushi Brown (0.85) (0.33) (0.72) (0.14) (2.7) (0.25) Jingmen Blue (2.5) (15) (2.8) (12) (0.26) (340) (0.23) White Leghorn (0.69) (0.29) (0.58) (0.12) (0.85) (0.22) (1.6) (0.20) < P < 0.01, significance levels were obtained after 1,000 permutations. coefficient) were observed in Dongxiang Brown (0.75 ± 0.05) and Yimeng Blue ( 0.61 ± 0.04), respectively, with an average of 0.12 (0.03) across all populations. All polymorphic SNP were in HWE (P 0.05) in Dongxiang Blue, Lushi Blue, Jingmen Blue, and White Leghorn. However, in the other 5 populations, from 19% (in Lushi Brown) to 95% (in Yimeng Blue) of SNP markers significantly deviated from HWE (P 0.05). For further analysis, we used AMOVA to examine the partitioning of genetic variation (Table 2). When the 8 Chinese chicken populations were analyzed as one group, most of the population variance (71%) could be explained by within-group variability. Analysis of the 8 Chinese chicken populations grouped based on eggshell color types (Chinese blue eggshell and Chinese brown eggshell) showed that within-group genetic differences accounted for 52% of the total variation, with 43% variation among groups. When White Leghorn chickens were included in the analysis, the percentages of variance components were very close to the second data set. To evaluate genetic differentiation and similarity among populations, pairwise population F ST estimates and Reynolds s genetic distance were examined for each pair of chicken populations (Table 3). For most pairs, the F ST values indicated statistically significant differences (0.001 < P < 0.01), revealing significantly differentiated populations. Among Chinese chicken populations, the lowest F ST (0.001) and Reynolds genetic distance (0.002) were both found between Lushi Blue and Jingmen Blue populations, whereas the highest values (0.65 and 0.78, respectively) were both found between Dongxiang Blue and Lushi Brown populations. As expected, average pairwise F ST (0.40) and Reynolds genetic distance (0.55) between White Leghorn and Chinese chicken populations were higher than those between Chinese chickens, with average values of 0.21 and 0.31, respectively. However, average F ST (0.42) and Reynolds genetic distance (0.57) values between all pairs of blue- and brown-shelled chickens were higher than those averages between White Leghorn and Chinese chickens (0.40 and 0.55, respectively). Interestingly, when comparing White Leghorn and Chinese chicken populations, the lowest values of both F ST (0.23 and 0.14) and Reynolds s genetic distance (0.37 and 0.24) were found between White Leghorn and 2 brownshelled types of Chinese chickens. However, some pairs of blue-shelled types did not show significant differentiation (P > 0.05, Table 3). We also estimated gene flow (Nm) between each population pair (Table 3) to examine the possible movement of individuals and genes in space, which affects many important ecological and evolutionary properties of populations. The Nm value across Chinese chickens

6 PHYLOGENETIC ANALYSIS OF CHINESE CHICKENS Lushi Blue Jingmen Blue Wulong Blue Dongxiang Blue Yimeng Blue Lindian Blue White Leghorn Dongxiang Brown Lushi Brown Figure 2. Unweighted pair-group method with arithmetic mean (UPGMA)-based phylogenic tree showing the genetic relationships among the 8 Chinese chicken populations and White Leghorn using Reynolds et al. (1983) genetic distance. ranged from 0.14 (between Dongxiang Blue and Lushi Brown) to 340 (between Lushi Blue and Jingmen Blue). Between Chinese chickens and White Leghorn, Nm varied from 0.12 to 1.6. As expected, the mean number of migrants per generation (Nm) across all Chinese chicken populations (18) was much higher than that between White Leghorn and Chinese chickens (0.56), while the average estimated gene flow across all populations was 14. All Nm values between Chinese chicken populations with the same eggshell color were above 1.0, whereas values between populations with different eggshell colors were below 1.0. In addition, the Nm values between White Leghorn and Chinese populations (except Lushi Brown) were below 1.0. We also used a tree-structured graph to visualize the result of hierarchical clustering and the phylogenetic relationships among the 9 chicken populations, based on Reynolds genetic distance (Reynolds et al., 1983; Figure 2). As expected, the populations clearly separated into 2 major clusters, including all blueshelled chickens in the first major cluster and brownshelled chickens together with White Leghorn in the second major cluster. Within Chinese blue-shelled chicken populations, Lindian Blue and Yimeng Blue formed a closely related subcluster. Lushi Blue and Jingmen Blue also formed a very closely related subcluster. Dongxiang Blue and Wulong Blue were intermediately positioned between these 2 subclusters. In the second major cluster, Dongxiang Brown and Lushi Brown formed a relatively close group separated from White Leghorn. Since matrices such as the pairwise F ST matrix for 9 populations or 318 individual chickens can be difficult to read and interpret, PCoA was used to visualize the patterns of genetic relationship. Analysis of principal coordinates based on the F ST matrix for 21 markers in the 9 chicken populations is illustrated in Figure 3(a). PCoA obviously discriminated blue-shelled chickens from other chicken populations, with the first and second principal coordinates explaining 83 and 8.7%, respectively, of the total variation. Figure 3(b) shows PCoA results based on a genetic distance matrix for 318 individuals and a covariance matrix with data standardization of the 9 chicken populations. This PCoA also apparently separated Chinese blue-shelled chickens from brown-shelled chickens and White Leghorn, with the first and second principal coordinates explaining 75 and 7.2%, respectively, of the total variation. The third principal component accounted for approximately 4.1% of the variation, and slightly separated the White Leghorn population from the brown-shelled chickens, with some overlapping. PCoA analysis confirmed the results of the population structure and dendrogram analysis. The program STRUCTURE was used to investigate population structure as well as inferring the presence of distinct populations, assigning individuals to populations, studying hybrid zones and identifying migrants and admixed individuals. The graphic results of the clustering analysis for K = 2 to 7 are illustrated in Figure 4. Results of the Bayesian clustering approach for K = 2 to 10 are presented in Table S2 and Figure S1, and show that K = 7 was optimal, capturing the major structure proportion present in the data (Figure S2). With K = 2, most blue-shelled chicken populations appeared clearly differentiated from other chickens, and the Yimeng Blue with Lindian Blue populations showed intermediate results (Figure 4 and Table S3). At this K value, 7 populations clustered with more than 76% of membership coefficients, while more than 54% of individuals from Yimeng Blue and Lindian Blue Figure 3. First and second principal coordinates analyses (PCoA) results in the 8 Chinese chicken populations and White Leghorn clusteredbased on populations (a) and individuals (b).

7 1782 DALIRSEFAT ET AL. Figure 4. STRUCTURE clustering of Chinese indigenous blue- and brown-shelled chickens in reference to the White Leghorn. Each individual is represented by a single vertical segments showing estimated membership coefficients of each individual to the inferred K cluster, with K = 2 to 7. Black color line separated the populations indicated below the figure. clustered in the related group (data not shown). This result is in good agreement with results from UPGMA dendrogram and PCoA. With K = 3, a relatively high heterogeneity of membership coefficients was observed within blue-shelled chicken populations, particularly in comparison with nonblue-shelled chickens. Differentiation within Chinese blue-shelled chicken populations was first observed at K = 3, with over 50% of Wulong Blue, Dongxiang Blue, Lushi Blue, and Jingmen Blue assigned to the same cluster, and 62% of the Yimeng Blue and 61% of Lindian Blue sharing the same cluster. However, there was no clear differentiation among blueshelled chicken populations at these inferred clusters and they did not cluster according to their traditional classifications or geographical distribution. Similar to K = 3, 3 clusters were formed at K = 4, but with a lower proportion of individuals in each cluster. At this K value, White Leghorn as an outgroup chicken started to differentiate from brown-shelled chickens. At K = 5 and 6, the same clusters that formed at K = 4 were observed, with approximately close membership coefficient. Finally, at K = 7, Wulong Blue and Dongxiang Blue separated from Lushi Blue and Jingmen Blue to form different clusters. Since EAV HP insertion is the causative mutation for the blue-shelled phenotype (Wang et al., 2013), we sequenced and aligned the EAV HP insertion of the populations to examine potential polymorphism. Multiple sequence alignment from upstream and downstream parts of the EAV HP integration site, corresponding to the 6 blue-shelled chicken populations (GenBank Accession No.: KP256532, KP276576, KP276577, KP276578, KP276579, and JF for Yimeng Blue, Wulong Blue, Lindian Blue, Lushi Blue, Jingmen Blue, and Dongxiang Blue, respectively) and Mapuche/Araucana (GenBank Accession No.: KC632578; Wragg et al., 2013) as an outgroup, compared with the host sequence, Gallus gallus OATP1B3 (SLCO1B3) gene (GenBank Accession No.: JN020139) is illustrated in Figure S3. Complete sequences of the EAV HP insertions revealed 100% identity among Chinese blue-shelled chickens except for the Dongxiang breed (GenBank Accession No.: JF837512; Wang et al., 2013). However, the EAV HP sequences of the all oocyan Chinese chickens were 98% identical to Mapuche/Araucana sequences. Comparing sequences of the complete insertion in the homozygote oocyan chickens illustrated identical host integration sites at Gga1:67,324,647 for all the Chinese breeds, but a difference at Gga1:67,324,624 for the Mapuche/Araucana chickens. In addition, the target-site duplications were identical in all the Chinese breeds (3 - GAGGAG-5 ) but different in the Mapuche/Araucana (3 -CCTTCA-5 ) chickens (Figure S3). DISCUSSION We evaluated the molecular phylogenetic relationships among 8 Chinese indigenous blue- and brownshelled chickens, with White Leghorn as an outgroup, by using SNP markers inferred from resequencing the whole genomic region of SLCO1B3 and its inserted EAV HP sequence. All SNP loci identified in the whole SLCO1B3 gene of Chinese chicken populations exhibited a high degree of polymorphism (average 95%), whereas White Leghorn showed the lowest level (43%; Table 1). Genetic variability was highest in Yimeng Blue chicken (0.46 ± 0.01), while Dongxiang Blue demonstrated the lowest genetic variability (0.11 ± 0.01). The relatively higher genetic diversity observed in the Yimeng Blue population and Lindian could be due to lack of inbreeding (F = 0.61 and 0.58, respectively) and controlled

8 PHYLOGENETIC ANALYSIS OF CHINESE CHICKENS 1783 mating practices. The genetic variability of White Leghorn (0.19 ± 0.05) was lower than for Chinese chickens except Dongxiang Blue (0.11 ± 0.01). The lower genetic variability in White Leghorn is in harmony with the results of Leroy et al. (2012), who observed lower genetic diversity in white egg layers compared with 24 other African local and commercial breeds. This may correspond to the somewhat wider genetic range of founder breeds in local chickens than in white egg layers and to the lower number of breeding generations (Crawford, 1990). The observed heterozygosity values were less than expected in Dongxiang Brown, Lushi Brown, and White Leghorn, causing F > 0, which could indicate an inbreeding system of mating (Templeton and Read, 1994) in these populations. In contrast, higher observed heterozygosity than expected in the remaining populations might indicate an isolate-breaking effect (opposite of the Wahlund effect). Among all chicken populations, the average observed and expected heterozygosity values were lower in this SNP-based study (Table 1) than when estimated using microsatellite markers in several studies, including 52 European chicken breeds (Hillel et al., 2003), 12 Chinese indigenous chicken breeds (Granevitze et al., 2007), 15 Chinese indigenous chicken breeds (Chen et al., 2008), 6 Italian local chicken breeds (Zanetti et al., 2007), 10 Egyptian chicken strains (Eltanany et al., 2011), 5 Japanese native chicken breeds (Tadano et al., 2012), 23 African local breeds (Leroy et al., 2012), and 18 African native chicken breeds (Berima et al., 2013). By contrast, lower values of heterozygosity (0.19 ± 0.02) have been described for Italian native chickens using Amplified Fragment Length Polymorphism (AFLP) markers (De Marchi et al., 2006). The average expected heterozygosity within populations (0.27 ± 0.01) in this study was very close to the value reported for layer lines (0.27), but lower than for broiler lines (0.53) of commercial breeds analyzed by using 17 microsatellites (Crooijmans et al., 1996). However, the observed heterozygosity values of Yimeng Blue (0.75 ± 0.03) and Lindian Blue (0.72 ± 0.03) were substantially higher than those of all chicken breeds reported in the studies noted above. The differences in the results may be explained by differences in locations, sample sizes, experimental chickens, application of different molecular markers (particularly the multi-allelic nature of microsatellite markers), and wider distribution of microsatellite and AFLP markers throughout the genome. Five populations showed significant deviation from HWE (P 0.05) ranging from 19% in Lushi Brown to 95% in Yimeng Blue. This significant violation may be explained by sampling errors (including the Wahlund effect), misclassification of genotypes, measuring 2 or more systems as a single system, population substructure, failure to detect rare alleles and the inclusion of nonexisting alleles, or if inbreeding has occurred in the populations as a whole. As expected, estimates of pairwise population differentiation (F ST ) and Reynolds genetic distance (Table 3) revealed closer relationships between Chinese chicken populations than between White Leghorn and Chinese chickens. This is in good agreement with an AFLP-based fingerprinting study (Gao et al., 2008), which indicated higher genetic similarity among indigenous chicken breeds in China compared with that between Recessive White breed and indigenous Chinese chicken breeds. A low level of differentiation was also observed between each pair of the Chinese populations according to their eggshell color types, which could be attributed to common ancestry, admixture of the population, and lack of selection pressure in these populations. On the other hand, the average F ST among Chinese populations across loci observed in this study (0.35; Table S1) was higher than that reported between 5 closely related lines of Japanese native chickens (F ST = 0.15; Tadano et al., 2012), 15 Chinese indigenous chicken breeds (F ST = 0.16; Chen et al., 2008), and 78 Chinese indigenous chicken breeds (F ST = 0.11, Qu et al., 2006), but lower than that between local Italian breeds of chickens (F ST = 0.44; Zanetti et al., 2007). However, the range of pairwise F ST between all populations in the present study (from to 0.68; Table 3) was more extensive than that between 5 closely related lines of Japanese native chickens, the Nagoya breed (from to 0.25), assessed based on microsatellite polymorphisms (Tadano et al., 2012). The average F IS value over loci ( 0.22 ± 0.02; Table S1), which indicates the degree of departure from random mating, was significantly negative in this study, similar to that of all commercial crosses and 2 Nagoya lines of Japanese native chickens ( 0.14; Tadano et al., 2012), reflecting excess heterozygosity. In contrast, Italian chicken breeds (0.042; Zanetti et al., 2007) and15 Chinese indigenous chicken breeds (0.02; Chen et al., 2008) had positive and higher values of F IS, indicating heterozygosity deficiency and departures from random mating as a result of inbreeding within populations. As Edea et al. (2013) suggested, the absence of any significant inbreeding effects may be a reflection of the high gene flow between the populations, as supported by high Nm values (Table S1), the large population from which the samples were drawn, and the fact that related individuals were purposely avoided. In AMOVA analysis, when only Chinese populations were analyzed as one group, 29% of the total genetic variation corresponded to differences between populations and the remaining 71% was the result of variation between individuals within populations. De Marchi et al. (2006) used the Gst index (Nei, 1973) toestimate the percentage of the total variation in Veneto chicken breeds and reported a value of 0.40 for Gst across loci, indicating that 40% of the total gene diversity was observed between breeds, while the remaining 60% was accounted for by the within-breed component of variation. As expected, within-individual variation diminished to 52% when Chinese populations were grouped based on 2 eggshell color types, while differences between groups increased to 43%, and 5% of total variation also

9 1784 DALIRSEFAT ET AL. corresponded to the between-population within-group variation. This is apparently a result of higher genetic similarity between each eggshell color type group, which is in agreement with low genetic distance between populations within these 2 groups (Table 3). When White Leghorn chickens were included in the analysis, the proportion of variation component was not influenced and showed a very close proportion to those of the second data set. The number of migrants per generation (Nm) values showed relatively high influence of gene flow and genetic drift between populations with same eggshell color. Among populations, Lushi Blue and Jingmen Blue showed a substantial value of gene flow, which is consistent with the geographical locations of these 2 populations (Figure 1). However, a relatively high degree of gene flow between Lindian Blue and Yimeng Blue seems to contradict their geographical distribution in 2 distinct provinces in China. The similarity of these 2 blue-shelled chicken populations may be explained by the famous historical events of human mass immigration from Shandong province to the northeast of China since 1644 in the Qing Dynasty. Genetic variation will give rise to considerable differentiation where Nm < 1 but not where Nm > 1 (Slatkin, 1987). In the present study, the estimated number of migrants (Nm) between populations within each group based on eggshell color type was significantly higher than an earlier estimate for 15 Chinese indigenous chicken breeds using 29 microsatellite markers (Chen et al., 2008), and implies substantial gene flow between these populations, resulting in low measures of genetic differentiation and inbreeding. The UPGMA tree based on Reynolds genetic distances (Figure 2) clearly distinguished the blue-shelled populations from the brown-shelled populations, which further confirmed the results obtained from PCoA and STRUCTURE-based clustering. In the UPGMA tree, Lushi Blue and Jingmen Blue chickens clustered together and were supported by a low value of F ST and high value of gene flow (Table 3), indicating a close genetic relationship between the 2 populations. In a lower level of relationship, these 2 populations together with Wulong Blue and Dongxiang Blue located in a sub-cluster. A possible explanation is that Henan (Lushi chickens), Hubei (Jingmen chickens), Chongqing (Wulong chickens), and Jiangxi (Dongxiang chickens) Provinces are geographically close to each other, raising the possibility of interbreeding. The high gene flow and low F ST values between these populations (Table 3) supported this close clustering of the populations. The 4 populations did not separate during the STRUCTURE runs from K = 2 to 6. This close genetic association may arise from a common genetic background or migration. Considering the geographical distribution of the populations, one would have expected a higher level of differentiation between the Yimeng Blue (from Shandong Province) and Lindian Blue (from Heilongjiang Province). Instead, based on results from all analyses in this study, Yimeng Blue and Lindian Blue chickens together formed a subcluster which is supported by extremely low genetic distance (0.008) and genetic differentiation (F ST = 0.004) with a considerably high valueofgeneflow(nm = 64) between the 2 populations (Table 3). Furthermore, there was high genetic variability and low inbreeding coefficients (Table 2) in both populations. This hypothesis was also confirmed by PCoA and STRUCTURE results, as the populations did not separate from K =2to7. As expected, Dongxiang Brown and Lushi Brown chickens grouped in a different subcluster from all blue-shelled chickens. These 2 populations have same origin as Dongxiang Blue and Lushi Blue, but a different oocyan genotype resulting to a different phenotype. Divergence between chicken populations with different types of eggshell color is obviously due to the SLCO1B3 gene controlling eggshell pigmentation. Substantial SNP dissimilarities between blue eggshell and brown eggshell types within same population may explain why the brown-shelled chickens formed a different cluster at lower K values. White Leghorn chicken as an outgroup population was expected to form a fully separate cluster from the Chinese chickens, but it clustered with brown-shelled chickens. STRUCTURE and PCoA results further supported this result, which may be due to more SNPs in the SLCO1B3 gene of White Leghorn being shared with brown-shelled rather than blue-shelled types. In our study, the STRUCTURE analysis clustered individuals into separate populations or groups of closely related populations, and suggested that the blue-shelled populations are mixture populations (Figure 4). The obvious mixed nature of these populations is consistent with the high estimates of gene flow between them (Table 3). Aligning EAV HP insertions of Chinese blue-shelled chickens in this study revealed no polymorphism, whereas Wragg et al. (2013) identified 5 polymorphisms through aligning their EAV HP sequence of Dongxiang (GenBank Accession No.: KC632577) to that of Wang et al. (2013; GenBank Accession No.: JF837512), resulting in a sequence divergence of 0.1%. It is noteworthy that our EAV HP sequences of 5 Chinese chickens were fully identical to that of Dongxiang sequenced by Wragg et al. (2013), but not to that sequenced by Wang et al. (2013). In addition, different host integration sites and DNA sequences have been identified for the Chinese chickens and Mapuche/Araucana fowls (Wragg et al., 2013). With the knowledge of distinct genomic insertion sites, our results agree with recent studies (Wragg et al., 2013; Wang et al., 2013) and clearly indicate the independent acquisition of the oocyan phenotype in native Asian and South American chickens. As cited by Wragg et al. (2013), according to historical evidence, the oocyan phenotype has been present since at least 500 years ago in the Dongxiang chicken in Asia (Gao et al., 2008), and since the late 19th century with a wide geographic distribution in South America

10 PHYLOGENETIC ANALYSIS OF CHINESE CHICKENS 1785 (Castello, 1924). The lack of divergence in the Long terminal repeat (LTR) sequences of the EAV HP insertions within the Chinese breeds and the Mapuche fowl supports a relatively recent integration event on both continents (Wragg et al., 2013). In conclusion, the present study confirms that SNP information derived from the whole genomic region of the SLCO1B3 gene is able to discriminate blueshelled from nonblue-shelled chickens, and distinguish populations within each group. Relatively low genetic diversity was observed in the 8 Chinese indigenous chicken populations. The populations that contribute to each eggshell color type are threatened by uncontrolled breeding between them, and therefore are at risk to become genetically uniform in the future. As Ibeagha- Awemu and Erhardt (2005) suggested, this unfortunate situation can be avoided by putting in place effective breeding measures and management practices aimed at controlling high genetic exchanges between the populations and preserving the typical/special characteristics of each population. An initial step towards achieving these goals might be to geographically define the breeds while allowing for normal traditional systems of management (Ibeagha-Awemu and Erhardt, 2005). We also found no polymorphism in EAV HP insertion sequences within Chinese blue-shelled populations, demonstrating their failure to differentiate these populations. However, this information might be useful to identify variation between Chinese blue-shelled chickens and Mapuche/Araucana fowls from South America and Europe. ACKNOWLEDGMENTS The National 863 Project (2013AA102501) and the Natural Science Foundation of China ( ) are acknowledged for financial support of this study. We also thank breed conservation farm and the contacts (Qigui Wang in the Poultry Institute of the Chongqing Academy of Animal Science, Hui Li and Li Leng in Northeast Agricultural University, Xiangtao Kang and Guirong Sun in Henan Agricultural University, Jiansheng Xu in Jiangxi Donghua livestock Co., Ltd., Junjiang Long in Shandong Longsheng agriculture and animal husbandry Group Co., Ltd., Yan Yang in Hubei Shendi Agriculture Science and Trade Co., Ltd.) in the 5 provinces of China for providing samples of Wulong, Lindian, Lushi, Dongxiang, Yimeng, and Jingmen chicken. Our appreciation goes to Mohsen Falahati (University of Tehran, Iran) for his helpful technical assistance during STRUCTURE analyses. SUPPLEMENTARY DATA Table S1. F-Statistics and estimates of Nm among all population for each locus Table S2. Output of the Evanno method results. Yellow highlight shows the largest value in the Delta K column. Table S3. Proportion of analyzed chicken populations in each cluster (K = 7). Figure S1. Plot of mean likelihood L(K) and variance per K value from STRUCTURE on a dataset containing 318 individuals genotyped for 21 polymorphic SNP loci. Figure S2. Evanno et al. (2005) plots for detecting the number of K groups that best fit the data Figure S3. Partial schematic diagram of LTR sequence alignment in six Chinese blue eggshell and Mapuche/Araucana chickens. Forward strand multiple sequence alignment of the host (Gallus gallus OATP1B3) sequence upstream (a) and downstream (b) of each integration site is highlighted in green, with the LTR sequence of the EAV HP highlighted in yellow, and the target site duplication highlighted in red. REFERENCES Berima, M. A., I. A. Yousif, H. Eding, S. Weigend, and H. H. Musa Population structure and genetic diversity of Sudanese native chickens. Afr. J. Biotechnol. 12: Castelló, S El Gallus inauris y la gallina que da el huevo azul. (Gallus inauris and the hen that lays the blue egg). Segundo Congreso y Exposición Mundiales de Avicultura. (2nd World Aviculture Congress and Exhibition). Barcelona, Spain. pp Chen, G. H., W. B. Bao, J. Shu, C. L. Ji, M. Q. Wang, H. Eding, F. Muchadeyi, and S. Weigend Assessment of population structure and genetic diversity of 15 Chinese indigenous chicken breeds using microsatellite markers. Asian-Australas. J. Anim. Sci. 21: Chen, G. H., K. H. Wang, J. Y. Wang, C. Ding, and N. Yang Poultry Genetic Resources in China. 1st ed. Shanghai Scientific and Technological Press, Shanghai, China. Crawford, R. D Origin and history of poultry species. Pages 1 42 in Poultry Breeding and Genetics, R. D. Crawford, ed. Elsevier, Amsterdam, Netherlands. Crooijmans, R. P., A. F. Groen, A. J. Van Kampen, S. Van der Beek, J. J. Van der Poel, and M. A. Groenen Microsatellite polymorphism in commercial broiler and layer lines estimated using pooled blood samples. Poult. Sci. 75: De Marchi, M., C. Dalvit, C. Targhetta, and M. Cassandro Assessing genetic diversity in indigenous Veneto chicken breeds using AFLP markers. Anim. Genet. 37: Earl, D. A., and B. M. vonholdt STRUCTURE HAR- VESTER: A website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 4: Edea, Z., H. Dadi, S. W. Kim, T. Dessie, T. Lee, H. Kim, J. J. Kim, and K. S. Kim Genetic diversity, population structure and relationships in indigenous cattle populations of Ethiopia and Korean Hanwoo breeds using SNP markers. Front. Genet. 4:35. doi: /fgene Eltanany, M., U. Philipp, S. Weigend, and O. Distl Genetic diversity of ten Egyptian chicken strains using 29 microsatellite markers. Anim. Genet. 42: Evanno, G., S. Regnaut, and J. Goudet Detecting the number of clusters of individuals using the software structure: A simulation study. Mol. Ecol. 14: Excoffier, L., G. Laval, and S. Schneider Arlequin (version 3.0): An integrated software package for population genetics data analysis. Evol. Bioinform. Online. 1: Gao, Y., Y. Tu, H. Tong, K. Wang, X. Tang, and K. Chen Genetic variation of indigenous chicken breeds in China and a

11 1786 DALIRSEFAT ET AL. Recessive White breed using AFLP fingerprinting. S. Afr. J. Anim. Sci. 38: Granevitze, Z., J. Hillel, G. H. Chen, N. T. K. Cuc, M. Feldman, H. Eding, and S. Weigend Genetic diversity within chicken populations from different continents and management histories. Anim. Genet. 38: Hagenbuch, B., and C. Gui Xenobiotic transporters of the human organic anion transporting polypeptides (OATP) family. Xenobiotica. 38: Hillel, J., M. A. M. Groenen, M. Tixier-Boichard, A. B. Korol, L. David, V. M. Kirzhner, T. Burke, A. Barre-Dirie, R. P. M. A. Crooijmans,K.Elo,M.W.Feldman,P.J.Freidlin,A.Mäki- Tanila, M. Oortwijn, P. Thomson, A. Vignal, K. Wimmers, and S. Weigend Biodiversity of 52 chicken populations assessed by microsatellite typing of DNA pools. Genet. Sel. Evol. 35: Ibeagha-Awemu, E. M., and G. Erhardt Genetic structure and differentiation of 12 African Bos indicus and Bos Taurus cattle breeds, inferred from protein and microsatellite polymorphisms. J. Anim. Breed. Genet. 122: Jakobsson, M., and N. A. Rosenberg CLUMPP: A cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics. 23: Krukenberg, C. F. W Die Farbstoffe der Vogel Eierschalen. Ber. Phys. Med. Ges. Wurzbg. 17: Lang, M. R., and J. W. Wells A review of egg shell pigmentation. World s Poult. Sci. J. 43: Leroy, G., B. B. Kayang, I. A. Youssao, C. V. Yapi-Gnaoré, R. Osei-Amponsah, N. E. Loukou, J. C. Fotsa, K. Benabdeljelil, B. Bed hom, M. Tixier-Boichard, and X. Rognon Gene diversity, agroecological structure and introgression patterns among village chicken populations across North, West and Central Africa. BMC Genet. 13:34. doi: / Lin, B. Z., S. Sasazaki, and H. Mannen Genetic diversity and structure in Bos taurus and Bos indicus populations analyzed by SNP markers. Anim. Sci. J. 81: Liu, K., and S. V. Muse PowerMarker: Integrated analysis environment for genetic marker data. Bioinformatics. 21: McKay, S. D., R. D. Schnabel, B. M. Murdoch, L. K. Matukumalli, J. Aerts, W. Coppieters, D. Crews, E. D. Neto, C. A. Gill, C. Gao, H. Mannen, Z. Wang, C. P. V. Tassell, J. L. Williams, J. F. Taylor, and S. S. Moore An assessment of population structure in eight breeds of cattle using a whole genome SNP panel. BMC Genet. 9:37. doi: / Morales, J., J. J. Sanz, and J. Moreno Egg color reflects the amount of yolk maternal antibodies and fledging success in a songbird. Biol. Lett. 2: Nei, M Analysis of gene diversity in subdivided populations. Proc. Natl. Acad. Sci. USA. 70: Orlóci, L Multivariate analysis in vegetation research. 2nd ed. Dr. W. Junk. The Hague. 451 pp. Peakall, R., and P. E. Smouse GenAlEx 6.5: Genetic analysis in Excel. Population genetic software for teaching and research An update. Bioinformatics 28, Popovic, M., R. Zaja, and T. Smital Organic anion transporting polypeptides (OATP) in zebrafish (Danio rerio): Phylogenetic analysis and tissue distribution. Comp. Biochem. Physiol. A. 155: Pritchard, J. K., M. Stephens, and P. Donnelly Inference of population structure using multilocus genotype data. Genetics. 155: Pritchard, J. K., X. Wen, and D. Falush Documentation for structure software: Version 2.3. ( edu/structure.html). Punnett, R. C Genetic study in poultry IX. The blue egg. Genetics. 27: Qu, L. J., X. Y. Li, G. F. Xu, K. W. Chen, H. J. Yang, and L. C. Zhang Evaluation of genetic diversity in Chinese indigenous chicken breeds using microsatellite markers. Sci. Chin. C Life Sci. 49: Reynolds, J., B. S. Weir, and C. C. Cockerham Estimation of the coancestry coefficient: Basis for a short-term genetic distance. Genetics. 105: Rosenberg, N. A Distruct: A program for the graphical display of population structure. Mol. Ecol. Notes. 4: Slatkin, M Gene flow and the geographic structure of natural populations. Science. 236: Sneath, P. H. A., and R. R. Sokal Numerical Taxonom. W. H. Freedman and Company, San Francisco. Sorby, H. C On the coloring-matters of the shells of birds eggs. Proceeding of the Zoological Society of London Tadano, R., A. Nakamura, and K. Kino Analysis of genetic divergence between closely related lines of chickens. Poult. Sci. 91: Tamura, K., N. Peterson, G. Stecher, M. Nei, and S. Kumar MEGA5. Molecular evolutionary analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28: Templeton, A. R., and B. Read Inbreeding: One word, seveal meaning, much confusion. Experientia Supplementum. 68: Underwood, T. J., and S. G. Sealy Adaptive significance of egg coloration. Pages in Avian Incubation: Behaviour, Environment and Evolution, D. C. Deeming, eds. Oxford University Press, Oxford, UK. Vignal, A., D. Milan, M. Sancristobal, and A. Eggen A review on SNP and other types of molecular markers and their use in animal genetics. Genet. Sel. Evol. 34: Wang, Z., L. Qu, J. Yao, X. Yang, G. Li, Y. Zhang, J. Li, X. Wang, J. Bai, G. Xu, X. Deng, N. Yang, and C. Wu An EAV-HP Insertion in 5 flanking region of SLCO1B3 causes blue eggshell in the chicken. PLoS Genet. 9:e Weir, B. S., and C. C. Cockerham Estimating F-statistics for the analysis of population structure. Evolution. 38: Wicke, W Uber des pigment in den eishalen der vogel. Naumannia. 8: Wragg, D., J. M. Mwacharo, J. A. Alcalde, C. Wang, J. L. Han, J. Gongora, D. Gourichon, M. Tixier-Boichard, and O. Hanotte Endogenous retrovirus EAV-HP linked to blue egg phenotype in Mapuche fowl. PLoS ONE 8:e Wright, S Evolution and the Genetics of Populations Variability within and among Natural Populations. 4th ed. University of Chicago Press, Chicago, IL. Yang, W., X. Kang, Q. Yang, Y. Lin, and M. Fang Review on the development of genotyping methods for assessing farm animal diversity. J. Anim. Sci. Biotechnol. 4:2. doi: / Zanetti, E., C. Dalvit, M. De Marchi, R. Dal Zotto, and M. Cassandro Genetic characterization of Italian chicken breeds using a panel of twenty microsatellite markers. Agriculture. 13: Zhao, R., G. Y. Xu, Z. Z. Liu, X. Y. Li, and N. Yang A study on eggshell pigmentation: Biliverdin in blue-shelled chickens. Poult. Sci. 85: