TGIF1 and SF1 polymorphisms are associated with litter size in Small Tail Han sheep

Abstract TGF‐β induced factor homeobox 1 (TGIF1) and splicing factor 1 (SF1) are important for mammalian reproduction; however, the effects of these genes on litter size in sheep remain unexplored. In this study, we genotyped 768 ewes from seven sheep breeds at two loci: g.37871539C>T, a synonymous mutation of TGIF1; and g.42314637T>C, a 3′UTR variant of SF1. Our analysis of polymorphism revealed only two genotypes at locus g.37871539C>T in TGIF1, with most sheep populations being moderately polymorphic (0.25 < PIC < 0.5) at this site. In contrast, most breeds exhibited low polymorphism (PIC ≤0.25) at the SF1 locus g.42314637T>C. The association analysis revealed that a synonymous mutation at g.37871539C>T in TGIF1 was highly associated with litter size in Small Tail Han sheep, in which it causes a significant decrease in litter size. Conversely, while the SF1 3′UTR variant g.42314637T>C was also highly associated with litter size in sheep, it causes a significant increase in the number of litter size. Combined, these data provide valuable information regarding candidate genetic markers for sheep breeding programs.


| INTRODUC TI ON
Understanding the genetics of reproduction is important for the sheep industry, and recent researches have revealed that the regulation of litter size in sheep and goats is multiallelic and polygenic Zhang et al., 2020). For example, FecB is a prominent gene affecting litter size, and ewes with the genotype of FecB BB carry two copies of a FecB mutation. This increases their ovulation rate and litter size by 3 and 1.5, respectively, over ewes with the genotype of FecB++. Likewise, ewes with one copy of the FecB mutation (FecB B+) have an ovulation rate and litter size 1.5 and 1 greater than FecB++ ewes, respectively . Mutations in a number of other genes have also been demonstrated to be useful genetic markers for sheep breeding. These genes include bone morphogenetic protein 15 (BMP15; Calvo et al., 2020), growth differentiation factor 9 (GDF9; Våge, Husdal, Kent, Klemetsdal, & Boman, 2013), melatonin receptor 1A (MTNR1A; , BMP7 and BMP2 . Such findings indicate that the identification of genetic markers can facilitate the progress of molecular marker-assisted breeding in sheep. TGF-β induced factor homeobox 1 (TGIF1) is a member of the three-amino-acid loop-extension (TALE) superfamily that is highly expressed in human (Hu, Yu, Shaw, Renfree, & Pask, 2011) and sheep (Jiang et al., 2014) ovaries, and highly conserved in mammals (Shen & Walsh, 2005). This suggests that TGIF1 plays a critical role in mammalian reproduction. Studies have also revealed that TGF-β/ SMAD signalling is important in reproductive processes, including in follicular activation (Yin, Chang, Yi, Yao, & Leung, 2020), ovarian follicle development (Knight & Glister, 2006) and oocyte maturation (Yin et al., 2020). Indeed, the inhibition or mutation of key members of the TGF-β/SMAD signalling pathway can cause reproductive problems (Chand, Robertson, Shelling, & Harrison, 2007), including sterility (Li et al., 2017). Significantly, TGIF1 in human and zebra fish functions as a repressor, and reversibly modulates important members of the TGF-β/SMAD signalling pathways that are activated by TGF-β, such as SMAD2 and SMAD4 (Hu et al., 2011;Hyman, Bartholin, Newfeld, & Wotton, 2003;Wotton, Lo, Lee, & Massagué, 1999). Therefore, TGIF1 may be a crucial factor in the reproduction of sheep and other mammalian species.
In this study, we examined the association between TGIF1 and SF1 polymorphisms and litter size in Small Tail Han sheep. This is the first analysis of such data, and our results provide useful genetic markers for future sheep breeding programs.  sheep (n = 60) and Tan sheep (n = 23); these are lower prolificacy breeds (Table 1). All ewes used in this study were randomly selected and had been fed in the same regions since they were born, with no sire effects. All sheep were freely mated within their breeds.

| Primer design and genotyping
Primers to genotype the loci g.37871539C>T of TGIF1 and  Table 2); subsequently, single-base extended PCR reactions are performed using specific primers ( Table 2). The molecular weight of the amplified products differs after extension because different bases are present at polymorphic sites. These differences are then

| Statistical analyses
Based on genotyping data, we calculated allele and genotype frequencies, polymorphism information content (PIC), heterozygosity (HE), and the number of effective alleles (NE) and p values for chisquare tests. If a particular ewe population had p > .05 (chi-square test), it was considered to be under Hardy-Weinberg equilibrium.
Regarding an association between TGIF1 and SF1 polymorphisms and litter size, chi-square test was performed to analyse the association of litter size with genotypes in each parity of Small Tail Han sheep.

| Prediction of protein interaction networks via the STRING database
As proteins modulate most physiological activities, identifying the interactions between proteins can enhance our understanding of complex traits such as litter size. To investigate the mechanisms by which TGIF1 and SF1 may modulate ovulation and litter size, we predicted the protein interaction networks of TGIF1 and SF1 using the STRING database (https://strin g-db.org).

| Genotyping and population genetic analysis of g.37871539C>T in TGIF1 and g.42314637T>C in SF1 in seven sheep breeds
The genotyping results were showed in (Figure 1). The results suggested that only two genotypes were detected at locus g.37871539C>T (TGIF1), while three genotypes were detected at locus g.42314637T>C (SF1. Our population genetic analysis (

| Protein interaction networks involving TGIF1 and SF1
To investigate the mechanisms by which TGIF1 and SF1 might affect ovulation and litter size in sheep, we predicted protein interaction  networks involving TGIF1 and SF1 via the STRING database. This analysis suggested that both TGIF1 and SF1 interact with ten other proteins ( Figure S1).

| TGIF1 and SF1 polymorphism in sheep
TGIF1 and SF1 may be important regulators of reproduction (Corioni et al., 2010;Hu et al., 2011;Hyman et al., 2003;Schoen et al., 2013;Wotton et al., 1999); however, the extent and effects of their polymorphisms in sheep are poorly understood. At the loci studied here, we detected only two TGIF1 genotypes in all ewe populations and two SF1 genotypes in some ewe populations (Table 3); some of these populations also exhibited low polymorphism (PIC≤0.25). These data may be limited by the number of sheep studied, and thus increasing the sample size could enable three genotypes to be captured, which would also increase the PIC value. Importantly, considering the negative effects of mutation on litter size, the lethality is the most likely reason for lack of TT genotype (g.37871539C>T) in all sheep breeds. Furthermore, we found that several sheep breeds were not in Hardy-Weinberg equilibrium (p ≤ .05) at the studied loci (Table 3), which may be a consequence of selection.

| Analysis of the association between TGIF1 and SF1 polymorphisms and litter size
Missense mutations cause the substitution of one amino acid for another and can therefore exert great influence on complex traits. TGIF1 and SF1 act as regulators in reproductive processes (Corioni et al., 2010;Hu et al., 2011;Hyman et al., 2003;Schoen et al., 2013;Wotton et al., 1999); however, their effects on litter size remain to be explored. Here, we conducted an association analysis of TGIF1 and SF1 polymorphisms with litter size in Small Tail Han sheep. Regarding locus g.37871539C>T in TGIF1, only two genotypes were identified, and litter size in ewes with the genotype CT was significantly lower than that of ewes with the wild-type genotype CC. This suggests that this synonymous mutation may be harmful. Furthermore, we failed to detect the mutant homozygous genotype TT, and the lethality was the most likely reason for lack of TT genotype (g.37871539C>T) in all sheep breeds. TGIF1 can negatively regulate TGF-β-activated genes such as SMAD2 and SMAD4 (Hu et al., 2011;Hyman et al., 2003;Wotton et al., 1999), and TGF-β/SMAD signalling plays crucial roles in mediating reproduction (Knight & Glister, 2006;Liu, Chang, Yi, Yao, & Leung, 2019;Rossetti et al., 2020;Yin et al., 2020). For example, the disruption of TGF-β/SMAD signalling can cause reproductive problems (Chand et al., 2007), including sterility (Li et al., 2017). Therefore, while on one hand, our failure to detect the third genotype at the g.37871539C>T locus may be due to inadequate ewe numbers; on the other hand, we hypothesize that this synonymous mutation may enhance the negative regulatory effects of TGIF1 on TGF-β/SMAD signalling, and even cause death in embryos with the homozygous TT genotype.
SF1 is mainly involved in alternative splicing events (Corioni et al., 2010), thus polymorphism in the 3′UTR of SF1 could modulate gene stability (Akdeli et al., 2014) and influence protein abundance.
Considering the key roles reported for alternative splicing in sheep reproduction (Miao et al., 2018), we speculate that the SF1 3′UTR TA B L E 4 The mean, standard deviations (SD) and p value (chi-squared test) of litter size in Small Tail Han sheep in the loci g.37871539C>T (TGIF1) and g.42314637T>C (SF1)

| Proteins interacting with TGIF1 and SF1
Most physiological processes are underpinned by multiple proteins; therefore, building interaction networks of proteins can enhance our understanding of complex traits such as reproduction. In our analysis, TGIF1 was predicted to interact with SMAD2, which has been reported to participate in reproductive processes.
Specifically, several studies have suggested that SMAD2 may be a candidate gene influencing reproduction in sheep with different fecundity (Zhang, Tang, et al., 2019;Zheng et al., 2019), and the inactivation of SMAD2 in mouse can lead to endometrial dysregulation and infertility (Kriseman et al., 2019). Previous study highlighted that the TGF-β induced factor homeobox 2 (TGIF2) has the same DNA binding homeodomains as TGIF1, suggesting that these two proteins may bind the same regulatory sequences (Melhuish, Gallo, & Wotton, 2001). As TGIF2 has been detected in tammar granulosa and theca cells, indicating a role in folliculogenesis (Hu et al., 2011), we speculate that TGIF1 may also be involved in this process.
SF1 was predicted to interact with ten proteins, including cell division cycle 5-like protein (CDC5L) and U2 small nuclear RNA auxiliary factor 2 (U2AF2). While CDC5L is a key factor regulating the cell cycle (Zhang, Kaur, Akhter, & Legerski, 2009), it is also highly abundant in BCB-positive GV oocytes (Liu et al., 2018) and follicles larger than eight mm-including prematuration oocytes (Dieleman et al., 2002). CDC5L could therefore act as a regulator promoting oocyte development and maturation. U2AF2 plays an important role in splicing decisions and has been reported to participate in alternative splicing events (Sutandy et al., 2018), such events have also been associated with reproduction (Miao et al., 2018).

| CON CLUS IONS
This study reported the first analysis of polymorphisms at two loci in TGIF1 and SF1, and suggested that these loci (g.37871539C>T in TGIF1 and g.42314637T>C in SF1) were highly associated with litter size in sheep. All in all, these data provide valuable genetic markers for sheep breeding.

CO N FLI C T O F I NTE R E S T
None of the authors have any conflict of interest to declare.

DATA AVA I L A B I L I T Y
Data sharing is not applicable to this article as no new data were created or analysed in this study.