Genetic studies have shown that the paternally inherited and maternally inherited genomes in eutherian mammals are not functionally equivalent. This is most aptly illustrated by genetic (or ‘genomic’) imprinting, a form of epigenetic regulation that results in the preferential expression of an allele from one of the two parentally inherited chromosomes in a parent-of-origin manner (McGrath & Solter, Reference McGrath and Solter1984; Surani et al. Reference Surani, Barton and Norris1984). To date, nearly 100 imprinted mammalian genes have been identified, many of which play important roles in fetal growth and development (Bartolomei, Reference Bartolomei2009; Feil, Reference Feil2009).
The insulin-like growth factor 2 gene (IGF2), which encodes a fetal mitogenic protein (IGF-II) structurally related to insulin (O'Dell & Day, Reference O'Dell and Day1998), has been the most extensively studied imprinted mammalian gene owing to its pivotal role in the regulation of embryonic development and relationship to disease (Reik et al. Reference Reik, Constancia, Dean, Davies, Bowden, Murrell, Feil, Walter and Kelsey2000; Rodriguez et al. Reference Rodriguez, Gaunt and Day2007). In mammals, IGF2 is expressed preferentially from the paternally inherited allele in most embryonic tissues and forms a conserved imprinted gene cluster with the reciprocally imprinted, non-protein coding H19 gene, which is highly expressed in embryonic and fetal tissue but whose function remains unclear (Bartolomei et al. Reference Bartolomei, Zemel and Tilghman1991; Rachmilewitz et al. Reference Rachmilewitz, Goshen, Ariel, Schneider, de Groot and Hochberg1992; Giannoukakis et al. Reference Giannoukakis, Deal, Paquette, Goodyer and Polychronakos1993; Feil et al. Reference Feil, Khosla, Cappai and Loi1998; McLaren & Montgomery, Reference McLaren and Montgomery1999; Dindot et al. Reference Dindot, Farin, Farin, Romano, Walker, Long and Piedrahita2004a; Dindot et al. Reference Dindot, Kent, Evers, Loskutoff, Womack and Piedrahita2004b; Zhang et al. Reference Zhang, Kubota, Yang, Zhang, Page, O'Neill, Yang and Tian2004; Li et al. Reference Li, Bin, Curchoe, Yang, Feng, Jiang, O'Neill, Tian and Zhang2008). The detection of biallelic IGF2 expression in certain mammalian post-natal and adult tissues, however, does suggest that monoallelic expression of this gene is both tissue- and developmental-stage specific (Dindot et al. Reference Dindot, Kent, Evers, Loskutoff, Womack and Piedrahita2004b; Curchoe et al. Reference Curchoe, Zhang, Bin, Zhang, Yang, Feng, O'Neill and Tian2005; Goodall & Schmutz, Reference Goodall and Schmutz2007; Chao & D'Amore, Reference Chao and D'Amore2008; Li et al. Reference Li, Bin, Curchoe, Yang, Feng, Jiang, O'Neill, Tian and Zhang2008).
The majority of studies involving IGF2 have focused on mouse models and human biomedical disorders (Chao & D'Amore, Reference Chao and D'Amore2008); however, there is growing interest in the role of IGF2 in domestic livestock. For example, DNA sequence polymorphisms within IGF2 have been shown to contribute to variation in complex production traits, notably muscle mass and fat deposition in pigs (Jeon et al. Reference Jeon, Carlborg, Tornsten, Giuffra, Amarger, Chardon, Andersson-Eklund, Andersson, Hansson, Lundstrom and Andersson1999; Nezer et al. Reference Nezer, Moreau, Brouwers, Coppieters, Detilleux, Hanset, Karim, Kvasz, Leroy and Georges1999; Nezer et al. Reference Nezer, Moreau, Wagenaar and Georges2002; Van Laere et al. Reference Van Laere, Nguyen, Braunschweig, Nezer, Collette, Moreau, Archibald, Haley, Buys, Tally, Andersson, Georges and Andersson2003). Since these early studies, a range of SNPs distributed across the porcine IGF2 gene have been shown to be associated with carcass traits, meat production, body size, fertility and survival traits in pigs (Vykoukalova et al. Reference Vykoukalova, Knoll, Dvorak and Cepica2006; Stinckens et al. Reference Stinckens, Van den Maagdenberg, Luyten, Georges, De Smet and Buys2007; Heuven et al. Reference Heuven, van Wijk, Dibbits, van Kampen, Knol and Bovenhuis2009; Oczkowicz et al. Reference Oczkowicz, Tyra, Walinowicz, Rozycki and Rejduch2009; Hou et al. Reference Hou, Wang, Guan, Zeng, Huang and Ma2010; Stinckens et al. Reference Stinckens, Mathur, Janssens, Bruggeman, Onagbesan, Schroyen, Spincemaille, Decuypere, Georges and Buys2010).
According to the Ensembl database (http://www.ensembl.org), the current fully annotated bovine IGF2 gene is located at the telomeric end of chromosome 29 (BTA29) and consists of five exons, the first two of which are untranslated (Ensembl gene ID ENSBTAG00000013066; Ensembl transcript ID ENSBTAT00000017372). Two alternative RNA transcripts have also been reported from this gene (Ensembl transcript IDs ENSBTAT00000044139 and ENSBTAT00000044140). Recently, there have been a number of publications detailing associations between DNA sequence polymorphisms in the bovine IGF2 gene and meat and milk production traits in beef and dairy cattle, respectively (Flisikowski et al. Reference Flisikowski, Adamowicz, Strabel, Jankowski, Switonski and Zwierzchowski2007; Goodall & Schmutz, Reference Goodall and Schmutz2007; Sherman et al. Reference Sherman, Nkrumah, Murdoch, Li, Wang, Fu and Moore2008; Bagnicka et al. Reference Bagnicka, Siadkowska, Strzalkowska, Zelazowska, Flisikowski, Krzyzewski and Zwierzchowski2010), hence we hypothesized that DNA sequence polymorphisms within this gene may also be associated with performance traits within a population of progeny-tested Irish Holstein-Friesian artificial insemination (AI) sires.
Previously, we validated 15 SNPs in the bovine IGF2 gene (Magee et al. Reference Magee, Berkowicz, Sikora, Berry, Park, Kelly, Sweeney, Kenny, Evans, Wickham, Spillane and MacHugh2010). In the present study, we investigated genotype-phenotype associations between four of these SNPs distributed across a ~31 kilobase (kb) region of Bos taurus chromosome 29 (BTA29) encompassing the bovine IGF2 gene and genetic merit for a range of performance traits in 848 Irish Holstein-Friesian AI sires, estimated from progeny performance.
Materials and Methods
SNP validation
The methods used to validate DNA sequence polymorphisms within, or proximal to, the bovine IGF2 gene have been discussed in detail elsewhere (Magee et al. Reference Magee, Berkowicz, Sikora, Berry, Park, Kelly, Sweeney, Kenny, Evans, Wickham, Spillane and MacHugh2010). Briefly, high-fidelity polymerase chain reaction (PCR) amplicons spanning putative IGF2-associated SNPs on BTA29 (as per build BTAU_4.0 of the bovine genome in the Ensembl database: http://www.ensembl.org) were generated for a panel of 26 animals of wide geographic provenance and sequenced bi-directionally (Macrogen Inc., Seoul, Korea). The MEGA 4.0 software package (Tamura et al. Reference Tamura, Dudley, Nei and Kumar2007) was used to analyse all resulting DNA sequences, validate the SNPs reported in Ensembl and identify novel SNPs in the re-sequenced regions.
In the current study, four validated SNPs distributed across a 30 827 base pair (bp) region spanning the bovine IGF2 gene (Ensembl gene ID ENSBTAG00000013066) were selected for high-throughput genotyping: rs42196909, IGF2.g-3815A>G, rs42194733 and rs42196901 (Table 1). These SNPs were selected as they are distributed across the full length of the annotated bovine IGF2 gene and displayed minor allele frequencies (MAF) ⩾0·18 in a panel of 138 European B. taurus animals screened previously by us (Magee et al. Reference Magee, Berkowicz, Sikora, Berry, Park, Kelly, Sweeney, Kenny, Evans, Wickham, Spillane and MacHugh2010). Based on the current annotation of the bovine IGF2 gene in the Ensembl database, two SNPs were located upstream of the IGF2 gene (rs42196909 and IGF2.g-3815A>G), one SNP (rs42194733) was located in an intron between the 2nd and 3rd exon of the IGF2 gene, and one SNP (rs42196901) was located downstream of the IGF2 gene. All SNPs represented transitions.
Table 1 Summary statistics for the IGF2-associated SNPs analysed across 848 Holstein-Friesian sires analysed in this studyFootnote †
† Genotype and allele frequencies, and the significance of deviations from Hardy-Weinberg equilibrium (HWE) based on P-values obtained from χ2-test results are shown for all four IGF2-associated SNPs. All SNP nucleotide positions (on BTA29) were obtained from the Build 4.0 of the B. taurus genome sequence on the Ensembl database (http://www.ensembl.org, release 58, as per Ensembl gene ID ENSBTAG00000013066) or the UCSC genome browser (http://genome.ucsc.edu). All SNPs genotyped are located on BTA29. The open reading frame (ORF) gene model positions for each SNP are given
‡ Details of the SNP nomenclature used is given in the Materials and Methods section of the manuscript
§ The dbSNP accession for the analysed SNP
It is important to note that the IGF2.g-3815A>G is not currently deposited within the dbSNP database (http://www.ncbi.nih.gov/projects/SNP); however, this SNP has previously been reported by us (Magee et al. Reference Magee, Berkowicz, Sikora, Berry, Park, Kelly, Sweeney, Kenny, Evans, Wickham, Spillane and MacHugh2010), where it was listed as SNP IGF2_08. In the current study we have recoded this SNP using the guidelines of the Human Genome Variation Society (http://www.hgvs.org). This SNP was named by first listing the nearest gene (i.e. IGF2) followed by ‘.g’ to denote genomic DNA, ‘-3815’ to denote the position of this SNP relative to the transcriptional start site (i.e. 3815 bp upstream of the IGF2 gene based on Ensembl transcript ID ENSBTAT00000017372) and finally ‘A>G’ to denote the alleles present at this locus.
DNA samples, DNA extraction, high-throughput SNP genotyping and SNP data filtering
Genomic DNA from 914 Irish Holstein-Friesian AI bulls was purified using a Maxwell™ 16 automated apparatus (Promega Corp., Madison WI, USA) according to the manufacturer's instructions. These bulls have been used to produce progeny in Ireland and are representative of the commercial germplasm used in Irish dairy herds in recent years. All four IGF2 SNPs were genotyped in all 914 sires (together with an additional 25 independently extracted, duplicate samples that were included for genotype quality control purposes) using the MassARRAY® iPLEX™ Gold genotyping platform provided by Sequenom Inc. (San Diego CA, USA; http//:www.sequenom.com). This SNP genotyping method discriminates between SNP alleles using single base primer extension technology after which primer extension products are analysed using matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectroscopy (http://www.sequenom.com/iplex). Following quality control on all genotype data (Waters et al. Reference Waters, McCabe, Howard, Giblin, Magee, MacHugh and Berry2010) genotypes for 848 bulls remained. The SNP genotype concordance rate between technical replicate for these SNPs was 100%.
A range of phenotypic traits were analysed in this study and were categorized into seven broad categories: (1) milk production traits (milk yield, fat yield, protein yield, milk fat and protein percentage); (2) udder health (somatic cell count, SCC); (3) carcass traits (cow carcass weight, progeny carcass weight, subcutaneous carcass fat level and carcass conformation score); (4) growth and size related traits in live animals (stature, chest width, body depth, rump width); (5) subjectively assessed subcutaneous fat level on live animals (angularity and body condition score); (6) calving traits (direct calving difficulty, maternal calving difficulty, perinatal mortality); and (7) fertility and survival (calving interval and functional survival). Sire predicted transmitting ability (PTA) was the dependent variable for all traits with the exception of the milk production traits, including SCC, which were daughter yield deviations (DYDs) expressed on a PTA scale.
Models used in genetic evaluations in Ireland, as well as variance components, have been previously described in detail by Berry et al. (Reference Berry, Kearney and Harris2009) and summarized by Waters et al. (Reference Waters, McCabe, Howard, Giblin, Magee, MacHugh and Berry2010). All PTAs were de-regressed using the procedure outlined by Berry et al. (Reference Berry, Kearney and Harris2009). Only sires with a reliability score, less parental contribution, of >60% were retained for inclusion in the association analysis. A total of 742 sires fulfilled this criterion for inclusion in the analysis of milk, fat and protein yield as well as milk fat and protein concentration and the number of sires included for calving interval and survival was 501 and 477, respectively. The numbers of sires for direct calving difficulty, maternal calving difficulty, and perinatal mortality were 575, 506 and 201, respectively. The number of sires with a reliability of >60% for the carcass traits was 446 and the number of sires with a reliability of >60% for the size linear type traits varied from 484 to 551.
The association between each SNP and performance was quantified using weighted mixed linear models in ASReml (Gilmour et al. Reference Gilmour, Cullis, Welham and Thompson2009), with individual included as a random effect, and average expected relationships among individuals accounted for through the numerator relationship matrix. Year of birth (divided into five-yearly intervals) and percent Holstein of the individual sire were included as fixed effects in the model. In all instances the dependent variable was de-regressed PTA or DYD, weighted by their respective reliability, less the parental contribution. Genotype was included in the analysis as a continuous variable coded as the number of copies of a given allele. The Haploview package (Barrett et al. Reference Barrett, Fry, Maller and Daly2005) was used to measure r 2 and D' values of linkage disequilibrium (LD) between pair-wise combinations of segregating SNPs (Lewontin, Reference Lewontin1964; Hill & Robertson, Reference Hill and Robertson1968).
Results and Discussion
Summary statistics for the four analysed IGF2 locus SNPs
Summary statistics, including genotype and allele frequencies together with deviations from Hardy-Weinberg equilibrium (HWE), for each of the four genotyped IGF2 SNPs are presented in Table 1 and the location of all four SNPs is depicted graphically in Fig. 1. Minor allele frequencies (MAF) for all analysed loci were ⩾0·12. Observed heterozygosity (i.e. the proportion of heterozygous individuals within the analysed sire population) ranged from 0·19 (rs42196901) to 0·45 (rs42196909 and IGF2.g-3815A>G) with a mean observed heterozygosity of 0·37 across all four loci. Only one SNP (rs42196901) exhibited a significant deviation from HWE (P=0·02). This was due to an observed excess of homozygous individuals for the ‘C’ allele (observed proportion of C/C individuals=0·021; expected proportion=0·013) and may be explained by random sampling.
Fig. 1. Location of the four bovine IGF2-associated SNPs analysed in this study based on ORF model for the annotated IGF2 gene transcript (Ensembl transcript ID ENSBTAT00000017372; BTA29) as per Ensembl release 58 of the B. taurus genome.
Inspection of the r 2 values of LD for all pairwise SNP combinations (Table 2) demonstrated that the rs42196909 and IGF2.g-3815A>G SNPs, which are separated by a total of 3243 bp, were in strong LD (r 2=0·995), suggesting that they represent an IGF2 haplotype. This is further supported by the identical allele and genotype frequencies observed at these markers (Table 1). All remaining pairwise SNP combinations displayed r 2 values <0·230 suggesting the existence of discrete haplotype blocks within the 30·8 kb genomic region analysed. LD statistics for each pairwise SNP combination are presented in Table 2.
Table 2. Pairwise SNP linkage disequilibrium (LD) statistics for the four IGF2-associated SNPs analysed in this study
Associations with carcass and body conformation traits
Significant phenotype-genotype associations for all SNPs analysed are detailed in Table 3. In the current study, one of the four bovine SNPs (rs42194733) within the bovine IGF2 locus was significantly associated (P<0·01) with progeny carcass weight: a T-to-C allele substitution at this intronic SNP was associated with gain in progeny carcass weight. The rs42194733 was not associated with any of the other performance traits analysed in this study. In addition, the two SNPs located upstream of the bovine IGF2 gene (rs42196909 and IGF2.g-3815A>G), both of which were in strong LD, were associated (P⩽0·05) with overall carcass conformation—a visually assessed measure of animal muscularity (Butterfield, Reference Butterfield1988). Both the G-to-A substitutions at the rs42196909 and IGF2.g-3815A>G loci were associated (P⩽0·05) with small increases in progeny carcass conformation, which are indicative of increased muscle mass. With the exception of these associations, none of the four assayed SNPs were associated with any of the other carcass, body conformation and body size traits analysed (listed in the Materials and Methods section). Also, although a recent study reported associations between SNPs within the porcine IGF2 gene and prolificacy and longevity in pigs (Stinckens et al. Reference Stinckens, Mathur, Janssens, Bruggeman, Onagbesan, Schroyen, Spincemaille, Decuypere, Georges and Buys2010), no associations with fertility or survival were observed here.
Table 3 Significant allele substitution effects between the four IGF2 SNPs and milk traits and carcass traits. se for each trait is shown in parentheses.
‡ A value of 1 prior to multiplication by 1000 equates to 1 percentage unit
§ See Materials and Methods for details
Significance of difference from zero:
† P⩽0·10; *P⩽0·05; **P⩽0·01
These observations suggest that the bovine IGF2 locus may harbour a QTL for muscle mass in cattle, an assertion which is supported by previous animal genetic studies. QTL mapping studies in cattle initially identified BTA29— the chromosome to which the bovine IGF2 locus maps— as containing a QTL influencing muscle mass in cattle (Casas et al. Reference Casas, Shackelford, Keele, Koohmaraie, Smith and Stone2003). More recently, genetic studies have revealed association between a single IGF2 C-to-T SNP [designated IGF2c.-292C>T (Goodall & Schmutz, Reference Goodall and Schmutz2003)] and meat traits (including rib eye area and body fat content) and body weight in beef cattle (Goodall & Schmutz, Reference Goodall and Schmutz2007; Sherman et al. Reference Sherman, Nkrumah, Murdoch, Li, Wang, Fu and Moore2008). Also, Schmutz & Goodall (Reference Schmutz and Goodall2005) reported that the C allele of this SNP was associated with lighter birth weight, while Sherman et al. (Reference Sherman, Nkrumah, Murdoch, Li, Wang, Fu and Moore2008) have reported that Aberdeen Angus animals with the TT genotype for this marker displayed increased average daily weight gain.
Studies in pigs have also shown that DNA sequence variation in the porcine IGF2 gene directly contributes to growth and carcass traits. Notably, a single G-to-A substitution within a regulatory region of the 3rd intron of the IGF2 gene (termed ‘IGF2 intron3 g.3072G>A’) which is directly responsible for a QTL influencing muscle mass and fat deposition in pigs; this SNP has since been classified as a quantitative trait nucleotide [QTN] (Van Laere et al. Reference Van Laere, Nguyen, Braunschweig, Nezer, Collette, Moreau, Archibald, Haley, Buys, Tally, Andersson, Georges and Andersson2003). It is considered likely that the ‘A’ allele prevents binding of a transcriptional repressor protein to the IGF2 gene; hence individuals inheriting a sire-derived ‘A’ allele at this locus display increased muscle growth and a corresponding reduction in body fat due to increased expression of padumnal IGF2 mRNA (Van Laere et al. Reference Van Laere, Nguyen, Braunschweig, Nezer, Collette, Moreau, Archibald, Haley, Buys, Tally, Andersson, Georges and Andersson2003; Stinckens et al. Reference Stinckens, Van den Maagdenberg, Luyten, Georges, De Smet and Buys2007). Other studies have demonstrated similar phenotypic effects attributable to this QTN in other pig populations (Jungerius et al. Reference Jungerius, van Laere, Te Pas, van Oost, Andersson and Groenen2004; Estellé et al. Reference Estellé, Mercadé, Noguera, Perez-Enciso, Ovilo, Sanchez and Folch2005; Stinckens et al. Reference Stinckens, Van den Maagdenberg, Luyten, Georges, De Smet and Buys2007).
The observed phenotypic associations between DNA sequence polymorphisms within the IGF2 locus and animal carcass and growth traits, as reported in the current study and elsewhere, are not surprising as it encodes an important fetal mitogen (DeChiara et al. Reference DeChiara, Robertson and Efstratiadis1991; Giannoukakis et al. Reference Giannoukakis, Deal, Paquette, Goodyer and Polychronakos1993). While functional genetic studies have identified a causal mutation for muscle mass and fat deposition in pigs (Van Laere et al. Reference Van Laere, Nguyen, Braunschweig, Nezer, Collette, Moreau, Archibald, Haley, Buys, Tally, Andersson, Georges and Andersson2003), no such causal mutations have yet been identified within the orthologous IGF2 locus in cattle. It has been proposed that the observed associations between the IGF2c.-292C>T SNP and muscle mass and fat content could be due to the location of the SNP in a regulatory region of the IGF2 gene, which may alter the efficiency of IGF2 mRNA translation and stability (Goodall & Schmutz, Reference Goodall and Schmutz2007). Initially, this SNP was reported within the untranslated exon 2 of the bovine IGF2 gene (Goodall & Schmutz, Reference Goodall and Schmutz2003; Goodall & Schmutz, Reference Goodall and Schmutz2007); however, inspection of the currently annotated bovine IGF2 gene in the Ensembl database (Ensembl gene ID ENSBTAG00000013066; Ensembl release 58 of the B. taurus genome) locates the IGF2c.-292C>T SNP 54 bp before the start of the 5′ UTR of the IGF2 gene (Ensembl transcript ID ENSBTAT00000017372). As per Build 4.0 of the B. taurus genome (Ensembl release 58), the IGF2c.-292C>T SNP is located at nucleotide position 51 257 871 on BTA29 and has not yet been deposited within the bovine dbSNP database (http://www.ncbi.nih.gov/projects/SNP). Despite the relocation of this SNP from an untranslated exon sequence to 54 bp upstream of the gene, it is possible that this polymorphism influences, or is linked to a polymorphism that regulates IGF2 expression, thereby accounting for the observed associations with bovine carcass traits.
Notably, none of the SNPs analysed in this study were located in amino-acid coding exons of the IGF2 gene or within untranslated IGF2 exonic sequences. Indeed, only one SNP [(rs42194733; located in IGF2 intron 2 based on the annotation of the bovine IGF2 gene (Ensembl ID ENSBTAG00000013066)] was located within the IGF2 gene region; all other SNPs were located upstream or downstream of the gene. As with the intronic IGF2 QTN described in pigs by Van Laere et al. (Reference Van Laere, Nguyen, Braunschweig, Nezer, Collette, Moreau, Archibald, Haley, Buys, Tally, Andersson, Georges and Andersson2003), it is possible that the intronic rs42194733 SNP has a similar function in regulating IGF2 expression, possibly through interference with repressor proteins; however, functional genetic studies are required to confirm this. It is more plausible, however, that this SNP is itself associated with a regulatory SNP (or set of SNPs) located proximal to, or within the IGF2 locus.
Associations with milk performance traits
In the current study, significant associations between all four assayed SNPs and milk performance traits were detected (Table 3). The A to G … SNP was positively associated with both milk yield and milk protein yield, and was negatively associated with milk protein percentage. A tendency to be associated (P⩽0·10) with milk fat percentage was also evident at this locus; however, this SNP did not display any association with milk fat yield. Similar phenotypic associations with milk yield were also observed at the rs42196909 locus, where an A-to-G substitution was associated (P⩽0·05) with improved milk yield and milk protein yield and tended to be associated (P⩽0·10) with reduced milk protein percentage. Again, the similar phenotypic associations support the existence of an upstream IGF2 haplotype incorporating both the IGF2.g-3815A>G and rs42196909 SNP loci. Finally, the A-to-G allele substitution at the rs42196901 SNP, located downstream of the IGF2 gene, was the only SNP associated (P⩽0·05) with increases in milk fat yield and milk fat percentage. A tendency to be associated (P⩽0·10) with increased milk protein percentage was also observed for this marker; however, this SNP was not associated with milk yield or milk protein yield.
Recently there have been a number of publications indicating that the bovine IGF2 locus may also harbour a QTL for milk production traits. Ashwell et al. (Reference Ashwell, Heyen, Sonstegard, Van Tassell, Da, VanRaden, Ron, Weller and Lewin2004) detected a QTL associated with milk production traits on BTA29; however, this QTL was localized to the centromeric region of BTA29, whereas IGF2 maps to the telomeric region of this chromosome. More recently, Flisikowski et al. (Reference Flisikowski, Adamowicz, Strabel, Jankowski, Switonski and Zwierzchowski2007) observed associations between an insertion/deletion polymorphism in the non-translated exon 6 of the bovine IGF2 gene and estimated breeding values (EBV) for milk yield, milk fat yield and milk protein yield in Holstein-Friesian bulls. Furthermore, associations with milk traits including milk yield and milk constituents have been reported at haplotypes constructed from the C-to-T SNP located at position 51 257 871 and a G-to-T non-amino acid changing substitution in the translated exon 10 of the IGF2 gene (nucleotide position 51 273 733 on BTA29) in Polish Holstein-Friesian cows (Bagnicka et al. Reference Bagnicka, Siadkowska, Strzalkowska, Zelazowska, Flisikowski, Krzyzewski and Zwierzchowski2010).
The open reading frame gene model location of the SNPs analysed here (two upstream, one intronic and one downstream) does not immediately suggest that these polymorphisms are functional. Instead, our results suggest that these IGF2 SNPs are associated with a causal regulatory mutation (or set of mutations) located proximal to, or within the IGF2 locus that has not yet been identified. It is noteworthy that previous studies have implicated IGF2 as playing a role in mammary gland development. For example, local infusion of the IGF-II protein has been shown to increase milk production in goats (Prosser et al. Reference Prosser, Davis, Farr, Moore and Gluckman1994) while studies in mice have shown that locally secreted IGF-II mediates the effect of prolactin on mammary gland development (Hovey et al. Reference Hovey, Harris, Hadsell, Lee, Ormandy and Vonderhaar2003). Another possible candidate gene harbouring a QTL for milk production on BTA29 is the insulin (INS) gene, which encodes the insulin hormone peptide and is located ~9·3 kb upstream of the bovine IGF2 gene. It is possible that the observed associations reported here between SNPs within the IGF2 locus may in fact be due to strong LD between the SNPs described here and unidentified functional polymorphisms associated with the INS gene. This hypothesis is supported by the role of the insulin protein in mediating mammalian gland development and lactation in dairy cattle (Akers, Reference Akers2006).
Conclusion
The identification of DNA sequence variations associated with body composition and milk performance traits make the imprinted bovine IGF2 gene an attractive target for future dairy cattle breeding programmes. The detection of association between a single genotyped SNP in the current study and traits related to animal growth also support this locus as harbouring a potentially important QTN(s) for beef breeding as observed in other cattle genetic studies and studies in other domestic species, most notably pigs. Furthermore, the results presented here add to an accumulating body of research showing that imprinted genes contribute to many complex performance traits in cattle and other livestock species. These findings, together with the documented biological roles of mammalian imprinted genes in mediating growth and development suggest that they represent an important reservoir of molecular markers for future genetic improvement of dairy and beef cattle populations.
This work was supported by a Research Stimulus Grant from the Irish Department of Agriculture, Fisheries and Food (project number: RSF-06-406, RSF-06-0353; RSF-06-0409) and Investigator Programme Grants from Science Foundation Ireland (SFI/01/F.1/B028 and SFI/08/IN.l/B2038). MPM was funded by Science Foundation Ireland research grant number 07/SRC/B1156.
