A parallel mechanism underlying frizzle in domestic chickens

Dear Editor, The feather, a highly keratinized tissue with variations in the shape, distribution, pigmentation, and structure, is an attractive topic in developmental and evolutionary biology (Boer et al., 2017). One phenotype noted by Darwin (Darwin, 1868), frizzle, consists of feather rachis and barbs curling outwards. It was previously reported that a 69-bp deletion in KRT6A (formerly named as KRT75 in the original paper; see Supplementary Materials and methods for details) was responsible for frizzle in chicken (Ng et al., 2012). Nevertheless, a recent screening of KRT6A in the Qilin chicken, a frizzle breed from Southern China (Supplementary Materials and methods, and Figures S1 and S2), failed to detect the 69-bp deletion (Tao et al., 2015). This raises a possibility that there is an independent genetic mechanism determining frizzle in Qilin chicken. To dissect this issue, we adopt a comparative population genomic strategy to investigate the genetic basis underlying frizzle in Qilin chickens. This strategy has been shown efficient in studying the phenotypic evolution in chicken (Wang et al., 2016, 2017). We sequenced the whole genomes of 20 Qilin individuals. Together with published data, a total of 62 chicken genomes were analyzed (Supplementary Table S1). After mapping sequencing reads to the chicken reference genome (Galgal5), we obtained >2.8 million indels and 18 million SNPs (Supplementary Table S2). Compared with free-ranging Yunnan and Tibetan native chickens, Qilin and Yuanbao breeds showed lower levels of nucleotide diversity and slower decay rates of linkage disequilibrium (Supplementary Figure S3). In the maximum-likelihood tree and the neighbor-joining tree, Qilin and Yuanbao clustered together and then grouped with Yunnan and Tibetan chickens (Supplementary Figure S4). Principal component analysis and ADMIXTURE revealed a high level of homogeneity in the Qilin population (Supplementary Figures S5 and S6). In total, 249, 349, and 157 genes were identified with signature of selection using ΔPi, XP-EHH, and LSBL, respectively (Figure 1A and Supplementary Tables S3–S5). The enrichment analysis for the detected selective genes revealed some overrepresented functional categories being associated with hair phenotypes, such as ‘woolly hair’ and ‘sparse hair’ (Supplementary Tables S6–S8). Several potentially selective genes are involved in feather development. For example, Pcdh9, a member of the delta-protocadherins (Pcdhs) family, is expressed in the dermis of the feather bud and may play a variety of roles during avian feather bud formation (Lin et al., 2013); SOBP gene is expressed in many tissues such as the feather follicle, and has an important role in regulating feather development (Liu and Li, 2012). A region in chromosome 33 exhibits the strongest signatures within chromosome 33 in the Qilin chicken. (A) Sliding window analysis for XP-EHH, ΔPi, and LSBL (50-kb window with 25-kb step increment). (B) The comparison of nucleotide diversity and population divergence in Qilin and other (i.e. Yuanbao, Yunnan, and Tibetan) chickens. (C) Haplotype pattern for the region from 1.23 M to 1.34 M of chr33 in Qilin and other chickens. Alternative alleles are labeled in blue. A region from 1.17 M to 1.39 M within chromosome 33 harboring 15 genes presented the strongest selective signatures in each of three approaches (Figure 1B). Further haplotype analysis showed Qilin chickens carrying a haplotype pattern that differs strikingly from those observed in other chickens (Figure 1C). Accordingly, we focused on this region in subsequent investigations. Breeding experiments revealed the Mendelian autosomal dominant inheritance of frizzle in Qilin chickens, most likely directed by a single locus (Supplementary Materials and methods). Therefore, we screened for variants spanning the region of chr33:1.17M–1.39 M across the 62 chicken genomes (Supplementary Tables S10 and S11) and variants (SNPs or indels) in chickens with normal feathers as heterozygotes were excluded. Also, variants in introns or synonymous were ignored. Considering the potential errors in sequencing and variants calling, we set the minor allele frequency >0.1 as the threshold to the existence of heterozygote variants (SNPs or indels) in the population. After the filtrations, two variants remained: a missense mutation in ENSGALG00000031831 (1274879, A/G) and a 15-bp deletion in KRT75L4, remained (highlighted in Supplementary Tables S10 and S11). To check the association between the variants and frizzle, we genotyped these two variants in 208 chickens. The missense mutation in ENSGALG00000031831 (1274879, A/G) occurred in chickens with normal feathers and, thus, was disregarded (Supplementary Table S12). The 15-bp deletion in KRT75L4 segregated completely with the frizzle phenotype in all the frizzle offspring (Supplementary Table S12). KRT75L4 is a member of α-keratin genes (Supplementary Figure S7) that have been demonstrated to play important roles in feather development of chickens (Ng et al., 2012). Sequence alignment showed that the 15-bp deletion caused a deletion of five amino acids that are conserved among 41 avian species (Supplementary Figure S8). These results favor the hypothesis that the deletion in KRT75L4 is the causative mutation for frizzle in Qilin chickens. The qPCR results indicated that the expression of KRT75L4 in feather follicles of adult Qilin chickens (frizzled) was similar to the normal controls (Supplementary Figure S9). This implied that malfunction of the protein caused the frizzle phenotype. To discern whether the 15-bp deletion in KRT75L4 was responsible for frizzle feathers, we conducted a feather regeneration experiment in both adult Qilin (frizzled) and Huaixiang (normal) chickens. After plucking the adult flight feathers, we overexpressed KRT75L4-WT (wild-type), KRT75L4-MT (15-bp deletion), and KRT75L4-Null by injecting the lentivirus into the corresponding follicles of chickens (Supplementary Materials and methods). No significant changes were observed in the regenerated feathers without injections from both right and left wings in both normal and frizzle chickens (Supplementary Figure S10). These regenerated feathers without injections served as the control in the comparisons with those with injections. The injections of overexpressed KRT75L4-Null generated little or mild changes in normal and frizzle chickens (Supplementary Figure S11), respectively. In chickens with normal feathers, the overexpression of KRT75L4-MT made the regenerated feathers curved (3/10) or with twisted rachis and wear off barbs (2/10) (Supplementary Table S13, and Figures S12 and S13). We did not observe obvious changes in the barbules and hooklets (Supplementary Figure S14). In comparison, overexpression of KRT75L4-WT in frizzle Qilin chickens made rachis less curved (Supplementary Figure S15 and Table S13), without affecting the microstructure (Supplementary Figure S16). Our results suggest that KRT75L4 plays substantial roles in the formation of feathers. The 15-bp in-frame deletion could lead to the frizzle phenotype in Qilin chickens. Taken together, our study provides a new case of phenotypic evolution in chicken based on parallel genetic mechanisms. During the past 5 years, studies revealed parallel genetic mechanisms for blue eggshell (Wang et al., 2013), dwarfism (Wang et al., 2017), and high-altitude adaptation (Wang et al., 2015) in different chicken populations. Thus, it is essential to consider the genetic background of different chicken populations even though all chickens were derived from domestication and subsequent breeding events within the Holocene. Moreover, our study presents a paradigm for exploring Mendelian traits within an evolutionary genomic approach. [Supplementary material is available at Journal of Molecular Cell Biology online. We thank Zhen-Hua Gao and Jin-Jun Chen (Guangdong Ocean University, Zhanjiang, China) for their technical assistance. This work was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB13020600), the National Natural Science Foundation of China (31321002, 31271339, and 31771415), the Innovative School Project of Department of Education of Guangdong Province (GDOU2013050222), the Science and Technology Program of Guangdong Province (2012B020305008), and the Key Project of Modern Agriculture in Zhanjiang City (2016A03010). This work was also supported, in part, by the Chinese Academy of Sciences President’s International Fellowship Initiative (2017VBA0003) and the National R&D Infrastructure and Facility Development Program—Special animal germplasm resources sharing platform—Guinea fowl and Houdan chicken preservation project (201720).]

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.