Polymorphisms of KLF3 gene coding region and identification of their functionality for abdominal fat in chickens

Abstract KLF3 is a member of the Kruppel‐like factor (KLF) family of transcription factors, and plays an important role in several biological processes, including adipogenesis, erythropoiesis and B‐cell development. The purposes of this study are to search for polymorphisms of KLF3 coding region and to provide functional evidence for abdominal fat in chickens. A total of 168 SNPs in KLF3 coding region were detected in a unique chicken population, the Northeast Agricultural University broiler lines divergently selected for abdominal fat content (NEAUHLF). Of which three single nucleotide polymorphisms (g.3452T > C, g.8663A > G and g.10751G > A) were significantly correlated with abdominal fat weight (AFW) and abdominal fat percentage (AFP) of 329 birds from the 19th generation of NEAUHLF (FDR < 0.05). The reporter gene assay was performed to verify functionality of these three SNPs in both ICP‐1 and DF1 cells. Results showed that the luciferase activity of G allele was significantly higher than that of A allele in g.10751G > A (p < 0.05). However, there were no significant differences between different alleles of others two SNPs in luciferase activity. Overall, KLF3 is an important candidate gene that affects chicken abdominal fat content, and the g.10751G > A is a functional variant that potential would be applied to marker‐assisted selection (MAS) for selective breeding programme.


| INTRODUC TI ON
With the rapid development of broiler industry, the problem of excessive accumulation of broiler body fat (especially abdominal fat) is becoming increasingly serious. Excessive fat deposition in broilers has adverse effects on feed conversion, carcass yield, hatching rate and fertility rate (Chen et al., 2019). Therefore, controlling the excessive accumulation of fat in chickens and further improving the feed conversion efficiency and carcass quality of broilers are important problems that need to be studied and solved urgently (Demeure et al., 2013;Zhang et al., 2018).
Identification of functional variations associated with abdominal fat deposition in chickens is helpful for implementation of marker-assisted selection (MAS) of abdominal fat traits and understanding of the genetic mechanism underlying fat growth and development. DOI: 10.1002/vms3.422 Kruppel-like factors (KLF) is a class of transcription factors with zinc finger structure, and widely involved in the regulation of cell proliferation, apoptosis, differentiation and embryonic development, and other life activities (Suske et al., 2005). KLF3, a member of the KLF family, is involved in the regulation of adipogenesis and lipid metabolism (Pearson et al., 2011). In vitro, KLF3 is regarded as a repressor that depresses the core gene CCAATenhancer binding protein alpha (CEBPα) in 3T3-L1 adipogenesis.
In vivo, KLF3 knockout mice have smaller and fewer adipocytes and are leaner than wild-type mice Sue et al., 2008).
A lot of studies showed that KLF3 also plays an important role in adipogenesis and lipid metabolism in agricultural animals. The transcriptional activity of KLF3 is promoted by KLF15 in bovine adipocytes . Polymorphisms in transcription factor binding sites of the KLF3 promoter region affect intramuscular fat content in Qinchuan cattle . Our previous studies found that chicken KLF3 is expressed in chicken adipose tissue, and regulates important genes involved in adipogenesis and lipid metabolism, more specifically, KLF3 overexpression suppress chicken CCAAT/enhancer binding protein alpha (C/EBPα), fatty acid binding protein 4 (FABP4), fatty acid synthase (FASN) and lipoprotein lipase (LPL) promoter activities, but increase chicken peroxisome proliferator-activated receptor gamma (PPARγ) promoter activity .
Based on the function of KLF3, it may be an important gene influencing the chicken abdominal fat content. To date, however, the relationship between sequence variants and chicken abdominal fat traits remains unclear. The purposes of this study are to analyse the polymorphisms in the coding region of KLF3, to investigate the association of SNP markers with abdominal fat traits and to identify their functionality.

| Experimental populations and phenotype measurements
The Northeast Agricultural University broiler lines divergently selected for abdominal fat content (NEAUHLF) have been established since 1996, using abdominal fat percentage (AFP) [AFP = abdominal fat weight (AFW)/body weight at 7 weeks of age (BW 7 )] and plasma very low-density lipoprotein (VLDL) levels as selection criteria (Guo et al., 2011;Leng et al., 2009). A total of 329 male birds (159 fat birds and 170 lean birds) from the 19th generation of NEAUHLF were used in this study. All male birds were slaughtered at 7 weeks of age. The blood was collected from wing veins before slaughtered and genomic DNA was extracted according to a procedure based on phenol/chloroform extraction and ethanol precipitation and properly kept (Leng et al., 2009).
The body weight at 7 weeks of age (BW 7 ) was measured before slaughtered and AFW were measured after slaughter. Abdominal fat percentage was calculated as the ratio of AFW to BW 7 (AFP = AFW/BW 7 ).

| Single nucleotide polymorphism (SNP) detection
We constructed individual 350bp DNA libraries for 329 male birds of 19th-generation population in NEAUHLF, and conducted wholegenome sequencing. After alignment, variant calling was performed for all samples by using the Unified Genotyper function in GATK 3.3 software. SNPs were selected by using the Variant Filtration parameter in GATK . A total of 168 SNPs were detected in the coding region of KLF3. Genotypes of these SNPs were available by sequencing of 329 birds.

| Cell culture
Two cell lines were utilized for promoter luciferase reporter assay, one is DF1 cell line that is widely used for cellular and molecular studies in chickens (Mannstadt et al., 2008;Wang et al., 2013) and the other is an immortalized chicken pre-adipocyte cell line (ICP-1) from our laboratory (W. Wang et al., 2017).

| Transfection and activity detection of dual luciferase reporter vectors
For transfection of the luciferase reporter plasmid, cells were seeded in 24-well plates, at 70%-80% confluence were washed with PBS, transfected with pGL3-Promoter vector containing the SNP and pRL-TK Renilla luciferase vector (Promega) as an internal control using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). After 48h of transfection, cell lysates were collected and operated according to the instructions of Promega's double luciferase detection kit.
Firefly luciferase activity was normalized to Renilla luciferase.

| Statistical analysis
Allele frequencies were calculated and the difference of allele frequencies between the lean and fat lines was analysed by Chi-square test. A value of p < 0.05 was used as the significant difference between the lean and fat lines. Population parameters, including observed heterozygosity (Ho), gene heterozygosity (He), effective allele numbers (Ne) and polymorphism information content (PIC), were calculated by Nei's methods through the MSR call online platform (http://www.msrca ll.com).
The associations between the SNPs and abdominal fat content (AFW and AFP) were analysed by using JMP 11.2 software (SAS Institute, Cary, NC, USA). The models used were as follows: Where Y was the dependent variable for the traits measured in the population (AFW and AFP), μ was the overall population mean of the traits, G was the genotype fixed effect, Sire was the gender fixed effect, Line was the Line fixed effect, BW at 7 weeks (BW 7 ) as a linear covariate (except for the AFP% traits), e was the residual random error, Sire (Line) was the random effect of Sire nested within Line and Dam (Line, Sire) was the random effect of Dam nested within Line and Sire. FDR was calculated from the distribution of p-values for the 168 SNPs, and FDR < 0.05 was considered as statistically significant for association (Matsuoka et al., 2015). Multiple comparisons between different genotypes were carried out by Tukey HSD, and significance level was set at p < 0.05. The variance explained by the SNP 2 SNP was estimated for each SNP-trait test as follows: where RMS is the residual of the reduced model (SNP effect excluded), and FMS is the residual of the full model (SNP effect included) (Gorshkova et al., 2006).

| Polymorphisms in coding region KLF3 gene
A total of 168 SNPs were detected in coding region of KLF3. Among them, 165 SNPs were located in introns, and three SNPs were located in exons. The distribution of SNPs on KLF3 was shown in the Figure 1 (Table S1). Totally, 329 male birds from 19th generation (G19) were genotyped for these SNPs by sequencing.

| Association analysis between SNPs and abdominal fat traits
We carried out the correlation analysis between the 168 SNPs and abdominal fat traits. The results were shown in Table 1. Totally, three SNPs (g.3452T > C in intron 2, g.8663A > G in intron 4 and g.10751G > A in intron 4) were significantly associated with AFW and AFP (FDR < 0.05). Percentage of additive genetic variance explained (PVE) by these three SNPs varied from 2.01% to 8.81%.
Multiple comparisons between their different genotypes were shown in Table 1. Birds with homozygous wild genotype were significantly lower than the other two genotype individuals for two SNPs (g.3452T > C, g.8663A > G) in AFW and AFP (p < 0.05). Birds with homozygous mutation genotypes had lower AFW and AFP for g.10751G > A (p < 0.05).
The genotype frequencies and allele frequencies of the three SNPs (g.3452T > C, g.8663A > G and g.10751G > A) of NEAUHLF were analysed, the results indicated that the allele frequencies of the three SNPs were extremely significantly different between the lean and fat lines (p < 0.01). Moreover, polymorphism information content (PIC) indicated that the g.3452T > C, g.8663A > G and g.10751G > A were lowly polymorphic in the lean line, and moderately polymorphic in the fat line (Table 2).

| Dual luciferase report gene assay
In order to identify functionality of three SNPs (g.3452T > C,

| Bioinformatics analysis
In order to investigate potential molecular mechanism underlying the association of abdominal fat content with chicken KLF3 g.10751G > A, we carried out in silico analysis of the transcription factor binding site of KLF3. The results showed that g.10751G > A was located in multiple transcription factor binding regions (Table 3).

| D ISCUSS I ON
Excessive abdominal fat deposition in broilers not only reduces reproductive performance and causes metabolic diseases but also reduces meat quality (Chen et al., 2019). Under this circumstance, molecular marker-assisted selection (MAS) is an efficient approach to enhance selection efficiency and further improve production performance.
Previous studies have shown that several members of the KLF family play a vital role in mammalian fat deposit (Farmer, 2006;Lefterova & Lazar, 2009;Rosen & MacDougald, 2006). Kruppellike factor 3 (KLF3) is an important regulator of fatty acid synthesis, lipid secretion and degradation, which are critical in mammalian lipid metabolism Zhang et al., 2009Zhang et al., , 2011Zhang et al., , 2013.
Researches showed that KLF3 also plays an important role in adipogenesis and lipid metabolism in agricultural animals Pertille et al., 2015). Our previous study indicated that KLF3 was expressed ubiquitously in chicken tissues and expressed consecutively during the chicken adipose tissue development from 1 to 12 weeks of age. In addition, the relative KLF3 mRNA expression levels of lean males were higher at 7 weeks of age and lower at 10 weeks of age than those of fat males, suggesting that the difference in KLF3 expression in abdominal adipose tissue might contribute to the fatness trait difference between fat and lean birds . Therefore, in the present study, the KLF3 was selected as a candidate gene to investigate associations of gene polymorphisms with chicken abdominal fat in NEAUHLF and functional variants in vitro.

F I G U R E 1 The distribution of SNPs in coding region of KLF3
Among 168 SNPs detected in coding region of KLF3, three SNPs (g.3452T > C, g.8663A > G and g.10751G > A) were highlighted significantly associated with abdominal fat traits (AFW and AFP, FDR < 0.05). As far as application is concerned, homozygous wild genotype for g.3452T > C, g.8663A > G and homozygous mutated genotypes for g.10751G > A are favourable in that the birds with these genotypes have lower AFW or AFP. In addition, there were significant differences between the two lines in allele frequencies of these three SNPs (Table 2) (p < 0.01). Generally speaking, artificial selection for important traits of domesticated animals is accompanied with alteration of allele frequencies of gene, frequencies of the alleles favourable for human are more likely to be increased, whereas unfavourable alleles tend to be decreased (Zhou et al., 2020).
NEAUHLF used in the present study has been divergently selected for AFP since 1996, exhibiting significant differences in AFP after 4th generation. It is expected that changing trend of alleles affecting AFP is similar to that of AFP. Therefore, three SNPs (g.3452T > C, g.8663A > G and g.10751G > A) are likely subjected to selection with abdominal fat due to extremely significant differences in allele frequencies between fat and lean lines. Regretfully, however, it is only in 19th generation that differences in alleles between two lines were investigated in this study. It is probably more reasonable to probe changing trend of alleles of these three SNPs between fat and lean lines in multiple and consecutive generations.
In addition to association analysis, we also provided functional evidence supporting a role for the SNPs in abdominal fat deposition. transcription factor NF-Y (Gorshkova et al., 2006). We performed luciferase reporter assays to investigate whether these three intronic SNPs (g.3452T > C, g.8663A > G and g.10751G > A) affected transcriptional activity. In the chicken DF1 and ICP-1 cell, the luciferase activity of allele G was significantly higher than that of allele A of the g.10751G > A. There was no significant difference in luciferase activity between different alleles at other SNPs (g.3452T > C, g.8663A > G). These results indicated that g.10751G > A was a functional variant affecting KLF3 transcriptional efficiency in vitro.
Studies have reported that the binding of transcription factors to intronic regions is modulated by intronic SNPs within transcription factor binding sites (Tsukada et al., 2006). In silico analysis suggested that the g.10751G > A located in multiple transcription factor binding regions, which may affect the transcription efficiency of the KLF3 by binding to some transcription factors (Table 3). CREB1 has the function in lipid storage or fat cell regulation (Xu et al., 2017). It has been reported that AP1 can affect adipocyte differentiation by regulating PPARγ (Luther et al., 2014). Transcription factor OCT1 is highly expressed in mouse liver and plays a major role in glucose and lipid metabolism in liver (Chen et al., 2014). P53 functions in adipose tissue metabolism and homeostasis (Krstic et al., 2018). In cancer cells, Sp1 promotes fat production by up-regulating the expression of sterol regulatory element binding protein-1c (srebp-1c) and FASN (Lu & Archer, 2010). Phosphorylation of GATA2 by Akt increases adipose tissue differentiation and reduces adipose tissue-related inflammation (Menghini et al., 2005)

| CON CLUS IONS
Three SNPs in KLF3, including g.3452T > C, g.8663A > G and g.10751G > A, may be important molecular markers that affect chicken abdominal fat traits. Taking into consideration that its strongest association with AFW (PVE = 7.15%) and AFP (PVE = 8.81%), and significant differences in luciferase activity of different alleles in vitro, g.10751G > A is considered as a functional SNP for abdominal fat in chickens. These findings will provide useful molecular markers for genetic improvement of abdominal fat content in chicken MAS programme. Rural Affairs in China (No. 19190526).

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflicts of interest. Associated with lipid storage or fat cell regulation To play a role in glucose and lipid metabolism in liver To affect adipocyte differentiation by regulating PPARγ Related to adipose tissue metabolism and homeostasis To promote fat production Xu et al., 2017Chen et al., 2014Luther et al., 2014Krstic et al., 2018Lu et al., 2010 A GATA2 To increase adipose tissue differentiation Menghini et al., 2005