The F₂ population of 100-day-old Sanhuang chickens (n = 439) used for the slaughter experiments in this study were reared under the same environmental and management conditions by Wens Food Group Co., Ltd. (Yunfu, China). The animal experiments in this study were approved by the Animal Care Committee of South China Agricultural University (permit number: SCAU#2017015, 13 September 2017) (14). The fat-related traits including AFW, AFR and ST were measured and calculated according to the Performance terminology and measurements for poultry (NY/T823-2020).

Blood samples were collected from veins under the wings of all Sanhuang chickens (n = 439), and DNA extraction was performed according to the instructions included in the DNA extraction kit (OMEGA, Georgia, United States). All DNA samples were used to amplify the dsDNA of the RGS16 5′UTR, 3′UTR and CDS with 2× Taq MasterMix (CWBIO, Nanjing, China). The PCR reaction conditions were as follows: pre-denaturation at 94°C for 2 min, amplification at 94°C for 30 s, 60°C for 30 s, and 72°C for 1 min, repeated for 30 cycles, final amplification at 72°C for 2 min, and hold at 4°C indefinitely. Finally, the PCR products were purified and sequenced by Sangon Biotech (Shanghai, China).

The immortalized chicken preadipocyte 1 (ICP-1) cells used in this study were provided by Li Hui’s research group from Northeast Agricultural University (Heilongjiang, China). The basal medium used was DMEM/F12 (Gibco, United States) supplemented with 15% fetal bovine serum (FBS) (Gibco, United States) and 1% streptomycin/penicillin solution (Gibco, United States), the cells were cultured at 37°C and 5% CO₂ in a calculator (15).

To construct a chicken RGS16 overexpression plasmid, the complete CDS region of RGS16 was cloned into the EcoRI and BamHI sites of pcDNA3.1 (Promega, New York, United States), and the plasmids were extracted according to the instructions included with the HiPure Plasmid/BAC EF Mini Kit (Magen, Guangzhou, China). The knockdown of RGS16 (5′-GGACCATTGATGGCCATAA-3′) was performed using specific siRNA designed and synthesized by RiboBio Co., Ltd. (Guangzhou, China).

Total RNA was extracted from tissues or cells using Magzol Reagent (Magen, Guangzhou, China) following the manufacturer’s instructions. PrimeScript RT Reagent Kit with gDNA Eraser (TaKaRa, Otsu, Japan) was used for cDNA preparation from total RNA. RT-qPCR was performed using a ChamQ Universal SYBR qPCR Master Mix (Vazyme, Nanjing, China) in a Bio-Rad CFX96 Real-Time Detection System. Data were analyzed using the 2^−△△Ct method with chicken GAPDH as the reference gene (16). The RT-qPCR primers were designed in Primer-BLAST ¹ and their detail were listed in Supplementary Table 1.

For oil red O staining, the ICP-1 cells were washed with PBS (Gbico, New York, United States) and then fixed with 4% paraformaldehyde solution (Sangon Biotech, Shanghai, China) for 30 min at room temperature. After fixation, the cells were rinsed twice with PBS, and oil red O staining solution (Sangon Biotech, Shanghai, China) was added followed by the incubation of cells for 60 min at room temperature. Then, the staining solution was removed, and the cells were washed three times with PBS. Image analysis was carried out using a DMi8 microscope (Leica, Wetzlar, Germany). The oil red O dye in stained cells was extracted with isopropanol solution and the absorption value was measured at 510 nm using a microreader (Bio-Tek, Vermont, United States).

EdU staining was performed using a Cell-Light EdU Apollo 488 In Vitro Kit (RiboBio, Guangzhou, China). For the EdU assay, ICP-1 cells were treated with EdU medium (1:1,000; RiboBio, Guangzhou, China) for 2 h at 37°C and then fixed in 4% paraformaldehyde solution (Yike, Guangzhou, China) for 30 min. The cells were subsequently permeabilized using 0.1% Triton X-100 solution (Gbico, New York, United States). The cells were incubated with the staining solution for 30 min at room temperature in the dark. Finally, the stained cells were scraped off with a cell scraper, collected in a 1.5 mL centrifuge tube and resuspended in 1 mL PBS. A BD Accouri C6 flow cytometer (BD Biosciences, San Jose, CA, United States) was utilized for the analysis of stained cells (16).

For the CCK-8 assay, ICP-1 cells were inoculated in 96-well plates and transfected with plasmids or siRNA. Cell proliferation and viability were then monitored at 12, 24, 36, and 48 h using the CCK-8 kit (TransGen Biotech, Beijing, China) following the manufacturer’s instructions. This analysis was performed using a microreader (Bio-Rad, Hercules, CA, United States) to measure the absorption value at 450 nm.

The MLM package in IBM SPSS software (version 26, IBM: International Business Machines Corporation) was used to analyze the association of gene polymorphisms and haplotypes with chicken fat-related traits. All results were represented as mean ± SEM. The MLM model is as follows:

Where Y represents the observed value, μ represents the mean, G represents the fixed effect of genotypes, S represents the fixed effect of sex, and e is the random error. In addition, we used Haploview v.4.2 software for linkage disequilibrium analysis (17, 18).

Statistical analysis and plots generation were conducted using GraphPad Prism v.9.4 (CA, United States) and R studio v.4.2.1 software (MA, United States). Correlation coefficients were calculated using Pearson correlation analysis. An unpaired t-test (two-tailed) was used to assess the statistical significance between two groups. In multiple comparisons, significant differences were assessed using the LDS method.

The descriptive statistics of fat-related traits in the F₂ population of Wens Sanhuang chickens (n = 439) was listed in Table 1. Among of them, average AFW was 98.43 g, the max value was 193.00 while the min value was 28.40 indicating a particularly evident inter-individual differences and a wide range of variation, with a coefficient of variation exceeding 30%. Average AFR was 7.72%, with a min value of 2.79% and a max value of 13.68%, and a CV value of 22.8%. Its CV value was much lower than that of AFW, indicating a potential positive correlation between AFW and body weight. For ST, the average value was 7.53 mm, with a min value of 1.16 mm and a max value of 12.83 mm. The CV of ST was 21.87%, which was closed to that of AFR. In general, the CV of fat-related traits in this population was large, which could be used to validate the association analysis between nucleotide polymorphism and fat traits.

To screen for polymorphisms, the flanking and exon regions of RGS16, with a length of 3,053 bp, were, respectively, amplified (Figure 1A). A total of 30 SNPs were identified in this region from the F₂ population of Wens Sanhuang chickens, and these were mapped to the genome sequence (Version: GRCg6a) in the Ensembl Database for identification (Table 2). The sequencing files of the 439 samples were analyzed using SnapGene v4.1.8 software to identify their genotypes for every individuals. Among of the 30 SNPs, the 6 SNPs were located in exon region but they were all synonymous variant. The statistics of genotype frequencies and allele frequencies are shown in Table 3. The 30 identified SNPs could all be divided into three genotypes. A chi-square test confirmed that all identified SNPs were in accordance with Hardy–Weinberg equilibrium (p > 0.05). Pearson correlation coefficients were used to analyze the correlations between fat-related traits, including AFW, AFR, and ST. As shown in Figure 1B, AFW was significantly and positively correlated with AFR (ρ = 0.934, p < 0.001), while ST is significantly and positively correlated with AFW and AFR (ρ = 0.290 for AFW, ρ = 0.234 for AFR, p < 0.001).

To perform association analysis of genotypes with fat-related traits in chickens, the MLM package in IBM SPSS software was utilized. The full results of this analysis were listed in Supplementary Table 2. The SNPs significantly associated with one or more fat-related traits were shown in Table 4. In terms of ST, the CC and TC genotypes of both rs735029742 and rs731455923 were found to be significantly higher than the TT genotype (p < 0.05). Additionally, the TT genotype of rs738408023 was significantly higher than the TC genotype (p < 0.05). For AFW and AFR, the TT and TC genotypes were significantly higher than CC genotypes for rs741662332 and rs315021359, respectively, while the CC genotypes were significantly higher than TT and TC genotypes for rs733023691 (p < 0.05). The CC and TC genotypes of rs735761612 were found to have higher AFR than the TT genotype (p < 0.05), and the AA and AG genotypes of rs16622145 were found to have higher AFR than the GG genotype (p < 0.05).

To better understand the relationship between SNPs and fat-related traits, the eight SNPs significantly associated with fat-related traits were further analyzed for linkage disequilibrium (LD) using Haploview software (Figure 1C). The LD plot showed two haplotype blocks, with block 1 including rs735029742, rs738408023, and rs735761612, and block 2 including rs741662332, rs731455923, rs315021359, and rs16622145. All SNP pairs within each block had high LD values ranging from 0.93 to 1.00. Both blocks were analyzed using the MLM model in relation to three fat-related traits. Supplementary Table 3 showed that block 1 was not significantly associated with any traits, while block 2 was significantly associated with AFR and ST. Table 5 presented the traits and relevant data indicating a significant association with block 2 haplotypes. For AFR, both H3H3 and H4H4 genotypes were significantly lower than H1H1, H1H2, H1H3, H1H4, H2H3, and H4H3 (p < 0.05). As for ST, both H2H2 and H3H3 genotypes were significantly lower than H1H1, H1H4, and H4H2 (p < 0.05). In addition, the H2H2 genotype was significantly lower than H1H3 (p < 0.05). These associations demonstrated that RGS16 may have a potential role in regulating chicken fat deposition.

In our previous study, we found that RGS16 was highly expressed in individuals with high level of abdominal fat (accession ID: PRJNA656618). Here, the RGS16 mRNA level difference between high- and low-fat groups was verified using RT-qPCR (Figure 2A). To investigate the function of RGS16 in preadipocytes, ICP-1 cells were transfected with the RGS16 overexpression plasmid and siRNA. The RT-qPCR results showed that the expression of RGS16 mRNA could be upregulated by more than 400 times and knocked down by about 40%, respectively (Figures 2B,C). CCK-8 assay was performed to measure the proliferation viability of ICP-1 cells after 12-, 24-, 36-, and 48-h transfection. From these data, it is evident that RGS16 overexpression significantly inhibits cell proliferation at 24 and 48 h, whereas RGS16 knockdown significantly promotes cell proliferation at 36 h (Figures 2D,E). To further assess the function of RGS16, the cell cycle phase was detected using flow cytometry after 48-h transfection. The results showed that the overexpression of RGS16 prolongs the G1 phase, preventing cells from entering the S phase and inhibiting their proliferation (Figures 2F,G). Furthermore, we also detected the mRNA levels of several cell-cycle-associated genes, CCNE1, CCNA1, and CCND1 using RT-qPCR and found that the overexpression of RGS16 significantly inhibits the expression of CCNA1, CCNE1, and CCND1, while RGS16 knockdown had the opposite effect, revealing its inhibitory effect on cell proliferation (Figures 2H,I). We performed an EdU assay to verify this conclusion. Forty-eight h after transfection, EdU staining was detected using flow cytometry, and the results showed that the proportion of EdU-stained cells significantly decreases with RGS16 overexpression and increases with RGS16 knockdown (Figures 2J–M). All of the above experimental results indicated that RGS16 is able to inhibit preadipocyte proliferation.

In addition, we hypothesized that RGS16 may be involved in regulating the process of chicken preadipocyte differentiation, based on its mRNA level difference between high- and low-fat individuals. ICP-1 differentiation was induced with 80 μM sodium oleate, and cells were collected at five different time points (12, 24, 36, 48, and 60 h) for RNA extraction. Interestingly, we found that RGS16 expression was significantly increased during ICP-1 cell differentiation (Figure 3A). The cells were collected for RNA extraction after 48-h RGS16 overexpression and RGS16 knock down, and RT-qPCR was performed to detect the expression level of genes associated with preadipocyte differentiation (including PPARγ, CEBPβ, and APOA1). The RT-qPCR results showed that RGS16 overexpression increases the expression of preadipocyte differentiation-related genes, whereas RGS16 knockdown decreases their expression (Figures 3B,C). After 12 h of transfection, the cells were induced in 80 μM sodium oleate medium for 48 h and stained with oil red O. The results show that RGS16 overexpression promotes lipid droplet formation, whereas RGS16 knockdown inhibits lipid droplet formation (Figures 3D–G). Our results demonstrated that RGS16 indeed is capable of driving preadipocyte differentiation and lipid droplet formation.