Here, we have identified a new transcript variant of the bovine ankyrin 1 gene in muscle. Seven novel SNPs were identified in the promoter region and several associations between the promoter SNPs and haplotypes and meat quality traits were observed. After correction for multiple testing, associations between rSNPs 1 and 4, and haplotype 2 and meat quality were significant.
We have shown that a novel variant of the ankyrin 1 gene is present in muscle. This variant is longer than previously reported variants (NCBI) and increases the overall gene length by 341 bp in the 5' direction. The identification of alternative transcripts of bovine ANK1 is not surprising. Indeed, mammalian genomes express three ankyrin proteins, which are encoded by three separate genes (ANK1-3) [33–35]. In bovine, ANK2 is located on chromosome 6 (13,335,381 - 13,706,921; btau_4.0) and ANK3 is located on chromosome 28 (14,417,416 - 14,535,840; btau_4.0). Unlike ANK1, neither map close to QTL for meat quality traits. However, ANK2 maps close to several QTL for carcass yield traits including longissimus muscle area  and back fat EBV [36, 37]. The functional and molecular diversity of the ankyrin family is a result of differential expression of these three independent genes and their widely alternatively spliced variants, some missing large segments that include whole functional domains [38–42]. It has been reported that ankyrin 1 has the most limited pattern of expression restricted to erythroid, muscle and neural tissue in human, but despite these limitations, tissue-specific isoforms with multiple transcripts are generated by alternative splicing under the control of several tissue-specific promoters [33, 39].
Single nucleotide polymorphisms (SNPs) within regulatory regions can regulate the level and/or timing of gene transcription [26–28]. Evolutionary mutations arising in a promoter can have variable effects depending on the rSNP location relative to TFBS, start codons, transcription start sites etc, and the intracellular molecular environment. In some cases, a SNP may eliminate the natural binding site of a tissue-specific transcription factor or create a new binding site . Therefore, functional rSNPs in TFBS may lead to differences in gene expression  with significant potential to increase phenotypic diversity. In this study, several ANK1 rSNPs were associated with meat quality traits. After correction for multiple testing, two of the SNP associations remained significant. These were the association between rSNP1 and tenderness score and the association between rSNP4 and texture values. While it is possible that these are marker SNPs rather than causative, of interest was that the rSNPs were found within well supported putative transcription factor binding motifs. For instance, presence of the minor T allele of rSNP1 created a glucocorticoid receptor (GR) motif and presence of the C allele of rSNP4 created a Sp1 motif. Glucocorticoids are steroid hormones which regulate a variety of biological processes and can influence many functions of a cell . The GR functions as a hormone-induced transcription factor which has been shown to regulate (either stimulate or repress) gene expression [43–45]. The presence of allele C of rSNP4 led to the creation of a Sp1 motif. Sp1 has been shown to bind to a consensus sequence of the leptin gene promoter and mutations in this region abolished the binding and reduce promoter activity in rat . Adamowicz and co-workers  also reported that a mutation in the Sp1 binding site of the bovine leptin gene affects its expression level. The presence of a G allele of this SNP (rSNP4) created a binding site for the zinc finger nuclear protein, YY1 (Yin Yang 1), also known as NF-E1, UCRBP or CF1, which is known to have fundamental roles in repressing and activating a diverse group of promoters; it not only initiates transcription but also activates or represses it [48, 49]. YY1 is also shown to repress muscle specific expression of sarcomeric alpha-actin and c-myc genes . Walowitz and co-workers also reported that calpains and the 26 S proteasome are involved in YY1 stability in rat muscle indicating that myogenic transcription may be inactivated by developmentally regulated proteolysis to promote muscle development. While it was not the aim of the present study, further functional studies would be valuable to assess the role of these TFBS, in modulating the expression of muscular transcripts of the bovine ankyrin 1 gene.
Few studies have examined variation in ANK1 in relation to muscle biology. However, Wimmers et al.  examined one SNP in the 3' UTR region of the porcine homolog. They have reported that this SNP in the 3' UTR of porcine ANK1 showed associations with shear force and IMF content, as well as several traits related to water-binding capacity in muscle , in several pig breeds.
In the present study, several of these rSNP were linked, therefore association analysis with individual haplotypes was also carried out. HAP4 was significantly associated with measures of tenderness. Sensory tenderness and firmness scores were improved in animals with HAP4 genotypes (approximately 5-10%), as was shear force (10-20%). These effect sizes suggest that the influence on tenderness is quantitative rather than that of a major gene . However, this effect may be important to perceived tenderness. Recent research has shown that beef with shear force values higher than 52.68 N can be reliably classified as 'tough' by sensory panellists and beef with shear force values lower than 42.87 N can be reliably classified as 'tender' . The observed genotypes and haplotypes at these promoter SNPs can thus potentially discern 'tender' from 'intermediate' to tough' beef. The findings reported here may be significant as there are to date, few markers for tenderness that have been widely confirmed and are in commercial use e.g. SNP in calpain and calpastatin gene regions [52–54] and expression level of DNAJA1 [55, 56].
This haplotype (HAP4) was, however, associated with lower scores for texture. This means that selection for improved tenderness may result in a disimprovement in texture. However, it should be noted that sensory texture profile is not as well correlated with sensory tenderness as is Warner Bratzler shear force . Low values of texture can be scored for meat that is too soft and mushy or too firm and unyielding. In contrast, low scores of tenderness and firmness are clearer in their meaning to sensory panellists (tougher and less firm respectively). Additionally, we also observed that homozygous HAP1 individuals also had reduced texture scores in SM muscle compared to individuals with no copy of this haplotype. This association remained significant even after a stringent Bonferroni correction (P = 0.0012), suggesting a possible role in influencing bovine muscle texture. Homozygous individuals with this haplotype also had increased firmness scores in LTL muscle.
Post-mortem tenderisation is influenced by enzymatic degradation of myofibrillar and associated cytoskeletal proteins. These proteins function to maintain myofibrillar integrity and it is known that the calpain system, in particular calpain 1, is responsible for proteolysis of these proteins [10, 12, 13, 15]. The calpain family of Ca2+-dependent proteases are thus key proteins in the regulation of tenderness [58–60]. Proteins which are reported to be proteolysed by calpains can be classified as kinases/phosphatases, membrane associated proteins, some transcription factors and cytoskeletal proteins which include desmin, gelsolin, myosin, nebulin, spectrins, titin, tropomyosin -for review see . Ankyrin is one of the cytoskeletal proteins proteolysed by members of the calpain family [7, 16, 61, 62]. Since ankyrin has functions in maintaining the structural integrity of, and linkages among myofibrils, the degradation of ankyrin could cause further softening of myofibrils and thus, tenderization. The relationship between tenderness (sensory and shear force) and HAP4 or rSNP1, rSNP3 of the bovine ANK1 promoter, may indicate that this haplotype and/or these rSNP might alter the expression of cytoskeletal protein ankyrin 1 in muscle and thus the extent of post-mortem proteolysis by calpains.
An association with IMF content (%) was observed for HAP2 in both muscles. This was supported by sensory juiciness score, which is an indirect sensory measure of intramuscular fat content [31, 32], in that HAP2 showed a tendency towards increased juiciness in SM muscle. Intramuscular fat has been shown to contribute 10 - 15% of the variance in palatability of beef . These results suggest that HAP2 may have potential in selection for IMF content. It is curious that HAP2 was associated with IMF but only differs from HAP4 and HAP1 by a single SNP in each case (rSNP2 and rSNP4, respectively). Neither rSNP2 nor rSNP4 are associated with IMF alone. This suggests that the relationship between HAP2 and this trait may be via linkage to nearby functional SNP. In fact, a number of QTL for marbling and fat thickness have been identified on this chromosome in a mapping population of Bos indicus × Bos taurus (including Hereford, Angus and Charolais) cattle [18, 19]. More recently, 3 QTL for marbling and 2 QTL for fat thickness have been identified on chromosome 27 in commercial Angus cattle . While ANK1 is not located within any of these QTL, its proximity supports the hypothesis that it is a candidate gene for these traits.
Many improvements in meat quality can be achieved by optimising systems management in animal production and meat processing [64–66]. However, in addition to environmental and processing influences on meat quality, there are certainly biological factors which affect the quality of meat [67–70]. The identification of genomic regions influencing economically important traits, in particular tenderness and intramuscular fat, has been made possible due to polymorphic markers and linkage maps in cattle. However, just a few genes have been located within these QTL and found to be influencing these traits. In this study we have found that polymorphisms (SNPs and haplotypes) of the ankyrin 1 gene are in association with these important meat quality traits. This is the first time such associations have been identified, therefore it will be important to confirm these findings in independent populations .