A 31-bp indel in the 5′ UTR region of GNB1L is significantly associated with chicken body weight and carcass traits

Background G-protein subunit beta 1 like (GNB1L) encodes a G-protein beta-subunit-like polypeptide. Chicken GNB1L is upregulated in the breast muscle of high feed efficiency chickens, and its expression is 1.52-fold that in low feed efficiency chickens. However, no report has described the effects of GNB1L indels on the chicken carcass and growth traits. Results This study identified a 31-bp indel in the 5′ untranslated region (UTR) of GNB1L and elucidated the effect of this gene mutation on the carcass and growth traits in chickens. The 31-bp indel showed a highly significant association with the body weight at 8 different stages and was significantly correlated with daily gains at 0 to 4 weeks and 4 to 8 weeks. Similarly, the mutation was significantly associated with small intestine length, breast width, breast depth and breast muscle weight. Moreover, DD and ID were superior genotypes for chicken growth and carcass traits. Conclusions These results show that the 31-bp indel of GNB1L significantly affects chicken body weight and carcass traits and can serve as a candidate molecular marker for chicken genetics and breeding programs.

efficiency chickens [9]. However, no report has described the effects of GNB1L indels on chicken growth and carcass traits.
Gene variants, such as single-nucleotide polymorphisms (SNPs) and insertion-deletions (indels), are widely distributed in an animal's genome, as reported in many studies on humans and livestock animals [10,11]. Compared with SNPs, the genotyping of large-fragment indels has higher efficiency [12]. Indel mutations also play important roles in many aspects of animal economic traits. A 10-bp indel exists in the PAX7 promoter region, which is located at the binding site of ZNF219, and the homozygous deletion genotype upregulated the expression and promoter activity of PAX7, which, in turn, affected the early growth traits of cattle [13]. A 19bp indel mutation in the PLAGI intron region affected the growth traits of Chinese cattle [14]. The 16-bp indel in the 5′ untranslated regions (UTR) of ZNF132 significantly affected the body length of the Hainan black goat [3]. Recent studies have revealed that the 11-bp indel in the DNMT3B intron region was significantly correlated with the litter size at first parity in the goat [15]; two indels (P2-16 bp and P14-15 bp) of DSCAML1 were significantly related to sperm quality in the male goat, and three indels of DSCAML1 were significantly correlated with the litter size at first parity in female goat [16]. A study has shown that the 13-bp indel mutation in the 3′ UTR of DGAT2 affected its expression and fat deposition in porcine [17]. In poultry research, two novel indels in QPCTL significantly affected chicken carcass traits and body weight at 5 different weeks of age [18]; the indel of CDKN3 was significantly associated with chicken carcass and growth traits [19]; a 22-bp indel of ZNF764L was significantly related to chicken birth weight, body slanting length, chest breadth and subcutaneous fat weight [20]. A 65-bp indel in the chicken GOLGB1 intron significantly affected body weight and carcass traits at 13 weeks [21]. The 80-bp indel in PRLR was significantly associated with chicken leg weight, body weight and shank length [22].
In the present study, we confirmed a 31-bp indel in GNB1L from 10× whole-genome resequencing data of ten XH and ten RW chickens (data unpublished) (EVA accession number: PRJEB36864). Chicken GNB1L is located on chromosome 15 and comprises 15 exons. Furthermore, 80 indels were found in GNB1L in the Ensembl database (http://asia.ensembl.org/Gallus_gallus/ Gene/Variation_Gene/ Table?db=core;g=  ENSGALG00000001925;r=15:1232691-1273276;t=  ENSGALT00000002979). However, there are no report and verification about the Indel of chicken GNB1L. This study was aimed to verify the indel mutation of GNB1L, clarify the effect of the GNB1L indel on chicken economic traits, and analyze GNB1L expression in different tissues and in leg and breast muscle tissues at different embryonic development stages. Additionally, we examined the distribution of the 31-bp indel in different populations. These results indicated that the 31-bp indel mutation in GNB1L could be used as a candidate molecular marker for chicken growth traits and provided a reference for the molecular breeding of chickens.

Polymorphism detection and genotyping
A novel 31-bp indel polymorphism in the 5′ UTR region of GNB1L was observed by DNA sequencing ( Figure S1) (TSINGKE, Guangzhou, China). All PCR amplification products were detected using 3.0% agarose gel electrophoresis, which revealed three genotypes, the 301-bp homozygous DD genotype, the heterozygous ID genotype (332 bp and 301 bp) and the 332-bp homozygous II genotype ( Figure S2).

Genetic diversity of the 31-bp indel in different populations
The genetic parameters, allele frequencies and genotype frequencies of seven different breeds and the F2 population were analyzed ( Table 1). The results suggest that the D allele frequency is lower than that of I in all breeds, except in LS chicken. Additionally, we counted different genotype distributions among the dual-purpose chickens (ND, GX, WC, QY and LS), F2 population, commercial broilers (RW) and commercial layers (ISA). The percentage of the DD genotype was the lowest in all breeds ( Figure S3). The results of χ 2 test suggest that the genotype frequencies of F2, ND and RW are not in HWE (P < 0.05), and WC, QY, ISA and LS were in HWE (P > 0.05). The values of He were from 0.46 to 0.50, and those of Ne were from 1.85 to 1.99. The smallest and largest values of PIC were 0.35 and 0.37, respectively. The results revealed that the 31-bp indel of GNB1L represents intermediate polymorphism and the lack of high genetic diversity among all the populations (Table 1).

Genetic differentiation of the 31-bp indel
The results of differential selection suggest medium genetic differentiation between LS and QY and between LS and GX (0.05 < Fst < 0.15). Moreover, we observed little genetic differentiation among the other breeds (Fst < 0.05; Table S1).
Correlation between the GNB1L 31-bp indel and economic traits The mixed model was used to analyze the correlation between the genotypes and economic traits. Three genotypes showed a significant correlation with 11 chicken growth traits, and a highly significant association with 9 growth traits ( Table 2). In particular, different genotypes showed a highly significant correlation with body weight at 7, 14, 21, 28, 35, 42, 49 and 56 weeks and daily gains at 0 to 4 weeks (P < 0.01) and were significantly related to daily gains at 4 to 8 weeks and shank length at 49 weeks (P < 0.05) ( Table 2). Importantly, the DD and ID genotypes were greater than the II genotype in all the related growth traits. Notably, the 31-bp indel displayed a highly significant correlation with breast width, breast depth, breast muscle weight and small intestine length in carcass traits, and was significantly correlated with fat cingula width (Table 3). Interestingly, the DD and ID genotypes were greater than the II genotype in all the related carcass traits. In the association analysis of the 31-bp indel and meat quality traits, the different genotypes showed a significant correlation with the dry matter content of leg muscle and a critical correlation with the crude fat content of leg muscle (Table S2).

GNB1L expression in chickens
GNB1L expression in 12 tissues of 20-week-old QY spotted-brown chickens was detected by qPCR. GNB1L was found to be relatively highly abundant in the heart, breast muscle, leg muscle, kidney and ovary and showed low expression levels in the small intestine, spleen, liver, lung, and abdominal fat (Fig. 1). Furthermore, GNB1L expression increased first and then decreased in breast muscle at different embryonic stages, and the expression level decreased first and then increased in leg muscle at different embryonic stages (Fig. 2a, b).

Transcription factor prediction in the GNB1L 31-bp indel
The transcription binding sites in the 31-bp indel of GNB1L were analyzed using an online prediction website, and the results revealed five potential transcription factors (NF-1, SP1, T3R, RAR-α and GR) ( Figure S4).

Discussion
The allelic frequency of genes can reflect the genetic diversity between different groups, indicating that new mutations are introduced to some extent [2,23]. In recent decades, the breeding of commercial broilers and layers has focused on growth and reproductive traits, respectively. In these commercial breeds, dominant genotypes for specific traits may be selected for breeding. Moreover, manual selection also determines the distribution and amount of genetic variation during domestication [23]. In this study, the I allele was the predominant allele in the ND, GX, WC, QY, F2, RW and ISA chickens, but not in LS chickens. The results showed that LS chickens might undergo different selection pressure during evolutionary processes compared with other chickens. Interestingly, the LS chicken is the only breed that can produce blue eggs among these breeds [24]. The body weight of chickens is a heritable trait with approximately 0.24-0.47% heritability during growth [25]. Compared with commercial broilers, Chinese domestic broilers have a relatively low growth rate and body weight. Therefore, we studied the correlation between the 31-bp indel in the GNB1L 5′ UTR region and F2 population carcass and growth traits. The 31-bp indel showed a highly significant correlation with the body weight at 8 different stages (Table 2). Moreover, the three different genotypes were significantly correlated with daily gains at 0 to 4 weeks and 4 to 8 weeks, and shank length at 49 weeks ( Table 2). The DD and ID genotypes were greater than the II genotype in all the related growth traits, with the DD genotype showing the highest weights at 7, 14, 21, 28, 35, 42 and 49 weeks, but not at 56 weeks. Interestingly, DD is the dominant genotype for daily gains at 0 to 4 weeks and 4 to 8 weeks. We hypothesized that the DD and ID genotypes might have a higher feed conversion ratio during chicken development. In summary, the II genotype is a disadvantaged genotype in all growth traits.
Chinese domestic chickens have a good carcass yield, with breast muscles accounting for approximately 30% of the carcass weight and the weight of muscles accounting for approximately 40% of the weight of the carcass [26]. Therefore, individuals with a larger breast width, breast depth and breast weight are also the breeding direction of local yellow-feathered broilers. The mutation was significantly related to breast width, breast muscle weight, breast depth and small intestine length of carcass traits (Table 3). Similarly, the DD and ID genotypes were greater than the II genotype in all the related carcass traits. Growing evidence suggests that the small intestine is mainly responsible for the efficient absorption and metabolic processing of nutrients, and the small intestinal villi are the main sites for absorbing nutrients [27]. Perhaps the longer length of the small intestine is helpful in improving the efficiency of animal absorption of food. We speculate that GNB1L 31-bp indel may affect the conversion efficiency of feed by affecting the length of the small intestine, ultimately leading to differences in carcass and growth traits of individuals with different genotypes. Previous research results also indicate that GNB1L is related to higher feed efficiency [9].
Studies have demonstrated that mutations in the 5′ UTR of some genes can affect gene expression [28,29]. Furthermore, TFs are essential factors that regulate gene expression, and the prediction results of TFs showed five  potential TFs (NF-1, SP1, T3R, RAR-α and GR) associated with the 31-bp GNB1L indel. We speculate that these TFs may be involved in the transcription of GNB1L, which leads to differences in the phenotype among the three genotypes. In this study, the expression of GNB1L was relatively highly abundant in the heart, leg muscle, breast muscle, kidney and ovary; other tissues had relatively low expression levels. Moreover, GNB1L expression increased first and then decreased in breast muscle at different embryonic stages, and decreased first and then increased in leg muscle. These results showed that GNB1L might be related to embryonic muscle development.

Conclusion
For the first time, we found that GNB1L, a candidate gene for high feed efficiency, has a 31-bp indel in its 5′ UTR that is significantly related to chicken carcass and growth traits. Moreover, DD and ID are superior genotypes for the carcass and growth traits of chickens. Furthermore, we confirmed that GNB1L might be a candidate gene for a higher feed conversion rate. In summary, this study showed that GNB1L might be involved in chicken embryonic development and growth, and the 31-bp indel of GNB1L can serve as a candidate molecular marker for the genetics and breeding programs of chicken.

Animal samples and trait measurement
The Additionally, the F2 resource population is a hybrid strain of RW and Xinghua (XH) chickens; XH chickens represent a slow-growing Chinese domestic chicken. In our laboratory, 2 mL of 5% pentobarbital was injected intraperitoneally into the chicken (No. 57-33-0; Chinese Academy of Sciences, Beijing Siyuan Technology Co., Ltd.). After 2-3 min, the chicken was sacrificed by bleeding through the carotid artery. Data records about economic traits, as well as detailed information on the measuring methods, were available for the F2 population, as previously described [30]. Twelve different tissues were obtained from four QY chickens. Additionally, the breast muscle of six embryonic periods (E10-15) and leg muscle of four embryonic periods (E12-15) were used to detect relative GNB1L expression.
cDNA synthesis and qRT-PCR RNA was extracted using TRIzol (Takara, Dalian, China). Next, the RNA was reverse transcribed using the cDNA reverse transcription kit (Takara, Dalian, China) and then was subjected to PCR. Relative gene expression was calculated using the 2 -ΔΔCt method, and significance was determined using ANOVA followed by Duncan's test. All the reactions were performed using three biological and technical repetitions. The relative expression of GNB1L using different tissues and embryo ages were analyzed by qRT-PCR. The qRT-PCR primers used for GNB1L and the internal control β-actin are listed in Table S3.

Indel detection and diversity analysis of different breeds
A 31-bp indel was identified in GNB1L from wholegenome resequencing data of ten XH and ten RW chickens (unpublished data). Genotyping of the GNB1L 31-bp indel was performed by PCR amplification and gel electrophoresis in eight diverse populations. Blood samples were used to extract DNA, and the final DNA concentration used for amplification was diluted to 50 ng/μL. The GNB1L PCR primers based on the genome are listed in The genotype and allele frequencies of the mutation were calculated directly in different breeds. Hardy-Weinberg equilibrium (HWE) was analyzed using the SHEsis website (http://analysis.biox.cn). Moreover, the allele numbers (Ne), genetic indices of heterozygosity (He), effective polymorphism information content (PIC) and population differentiation were analyzed using Pop-Gene software (Version 1.3.1) [31,32].

Transcription factor prediction
The transcription factors (TFs) in the 31-bp indel mutation of the 5′ UTR region of GNB1L were predicted using AliBaba software (Version 2.1) [24].

Statistics
Association analysis of the F2 population was performed using SPSS 22.0 software, and two different models were used in the analysis. All the growth traits used Model I (Yijkl = μ + Gi + Sj + Hk + fl + eijkl), and all the carcass traits used Model II (Yijkl = μ + Gi + Sj + Hk + fl + b (Wijkl -W ) + eijkl); the carcass weight served as a concomitant variable of Model II. Yijkl represents the observed value, μ is the overall population mean, fl is the fixed effect of family, Gi is the fixed effect of genotype, Hk is the fixed effect of hatch, Sj is the fixed effect of sex, b is the regression coefficient for carcass weight, W is average slaughter weight, Wijkl represents the individual slaughter weight, and eijkl represents the random error in the two models. Significance was set at a P-value < 0.05, and Bonferroni's test was performed for multiple comparisons [18].
Additional file 1: Figure S1. DNA sequencing files of the GNB1L 31-bp indel. (a) Partial sequence of the D allele. (b) Partial sequence of the I allele. Figure S2. Electrophoresis (3.0%) patterns showing the amplification results for GNB1L. DD, ID and II are the three different genotypes, and M represents DL2000. Because the gel did not melt sufficiently, a white stain appears in the picture. Figure S3. Percentages of the DD (blue), ID (red), and II (gray) genotypes in four types. Figure S4. Transcription factor binding sites predicted in the GNB1L 31-bp indel. AliBaba 2.1 online website parameters were used such as cons = 75% and classification level K = 4. Table S1. Pairwise fixation index (Fst) of GNB1L in different chickens. Note: F2: F2 resource population (F2; n = 360); ND: Ningdu chickens; RW: Recessive white Rock chickens; ISA: ISA brown laying hen; GX: Guangxi chickens; WC: Wenchang chickens; QY: Qingyuan chickens; LS: Lushi chickens. Table S2. Association analysis of the GNB1L 31-bp indel with meat traits in the Xinghua × Recessive White Rock F2 populations. Note: SE = standard error of the mean; BMSF = breast muscle shear force; LMSF = leg muscle shear force; RWL = rate of water loss; CLMF = crosssectional area of leg muscle fibers; CBMF = cross-sectional area of breast muscle fibers; BMDC = breast muscle dry matter content; BMDC = leg muscle dry matter content; BMFC = breast muscle fat content; LMFC = leg muscle fat content. Table S3. Details of primer pairs. Note: "-" represents a primer that is not used for genotyping.