Background

Marbling has an important effect on the economics of beef production [1–4]. Thus, it is greatly interesting to obtain better knowledge on the molecular architecture of marbling and to generate new opportunities for more effective marker-assisted selection.

The endothelial differentiation, sphingolipid G-protein-coupled receptor, 1 (EDG1) gene, which is involved in blood vessel formation [5], exhibited higher expression levels in high-marbled steer group than in low-marbled steer group in musculus longissimus muscle across all ages of the test period [6, 7]. We have recently showed that single nucleotide polymorphisms (SNPs) in the EDG1 were associated with marbling in Japanese Black beef cattle [8].

From the result of our previous differential-display PCR (ddPCR) analysis in musculus longissimus muscle [6], the titin (TTN) gene exhibited higher expression levels in low-marbled steer group than in high-marbled steer group in the late stage of the test period. Since TTN is known to be involved in myofibrillogenesis [9], the differential expression suggested that the decrease in the TTN expression might promote proliferation, differentiation, or maturation of adipocyte-lineage cells by weakening the structural integrity of the sarcomere, thereby resulting in high levels of marbling.

We have also located the TTN gene within genomic region of a quantitative trait locus for marbling [10, 11], which is mapped in a half-sib family of Japanese Black cattle to bovine chromosome 2 region [12]. Thus, the TTN gene was regarded as a positional functional candidate for the gene responsible for marbling.

TTN gene structure

We screened the NCBI bovine genome sequence database (National Center for Biotechnology Information, Bethesda, MD) with the publicly available bovine TTN cDNA sequence [GenBank:XR_027685] including 101,593 bp, and obtained 5,046,004-bp bovine genomic sequence [GenBank:NW_001494580] containing the TTN gene. By comparing the genomic and cDNA sequences, we showed that the bovine TTN gene spans 276,648 bp and comprises 211 bp of the first exon, 1,315 bp of the last exon, 100,067 bp of 313 internal exons, and 175,055 bp of 314 introns.

SNP detection

Two Holstein steers and two somatic nuclear-derived cloned steers [13] from a Japanese Black Itofuku sire with a very high estimated breeding value for marbling [14], which were assigned for low-marbled and high-marbled steer groups, respectively, in our previous ddPCR analysis [6], were used for SNP detection in this study. The two low-marbled Holstein steers and two high-marbled cloned steers were previously shown to have different TTN gene expression patterns [6]. The details of DNA samples from these steers were described previously [6]. We used two high-marbled cloned steers to confirm the authenticity of newly discovered SNP in the TTN gene.

PCR primers were designed to target ~7-kb proximal promoter and the first and last exon regions, to screen polymorphisms in the TTN gene between two low-marbled Holstein steers and two high-marbled cloned steers (Table 1). We selected these genomic regions for SNP detection, because it is likely that these regions are involved in TTN gene expression. PCR amplifications were performed using 25 ng of the prepared DNA as template in a final volume of 100 μl containing 1 μM of each primer, 0.25 mM of each dNTP, 2.5 U of Go Taq polymerase (Promega, Madison, WI), and 1 × Go Taq buffer (Promega). The PCR conditions were carried out as follows: 94°C for 2 min, 35 cycles of 94°C for 30 s, the annealing temperature indicated in Table 1 for 30 s, and 72°C for 1 min, followed by a further 5-min extension at 72°C. PCR products were examined by electrophoresis through a 1.0% agarose gel to determine the quality and quantity for DNA sequencing. DNA sequencing of PCR-amplified products was performed by the direct sequencing with an ABI377 sequencer (ABI, Foster City, CA) following standard Big Dye protocols (ABI). Primers used for PCR amplifications and obtained from primer walking were used as sequencing primers. Nucleotide polymorphisms were identified by comparison of sequence traces between the 2 steer groups, and designated according to nomenclature for the description of sequence variations in the HGVS (Human Genome Variation Society, Fitzroy, VIC, Australia) [15].

Table 1 Primers designed for SNP detection in the bovine TTN gene
Full size table

We sequenced the 6,615-bp 5' flanking region, the 211-bp first exon region, and the 1,315-bp last exon region of the TTN gene for the 4 DNA samples. This sequence analysis revealed only one SNP in the TTN gene between the 2 steer groups: a C to T transition located 652 bp upstream of the putative transcription initiation site in the promoter region (g.231054C>T). The g.231054C>T has been submitted to the dbSNP with ss accession number 120037472. The two Holstein steers were homozygous for C allele at the g.231054C>T site, whereas the two cloned steers heterozygous for C and T alleles at the g.231054C>T site.

Association study

Samples and data

We performed 3 experiments for association of the g.231054C>T with marbling and subcutaneous fat thickness. We used 101 Japanese Black sires in experiment 1. The sires used to be or are used in the Oita Prefectural Institute of Animal Industry (Oita, Japan). There was no strong bias for a specific father or a specific maternal grandfather of the sires, and the sire panel likely represents a variety of the sire lines. In experiment 2, 848 paternal half-sib Japanese Black progeny steers (55 to 445 steers per sire) produced from 5 sires heterozygous for the g.231054C>T, with dams considered to represent a random sample of the female population, were used. We used 820 paternal half-sib Japanese Black progeny steers (88 to 535 steers per sire) produced from 3 sires homozygous for C allele at the g.231054C>T, with dams considered to be a random mating population, in experiment 3. These progeny steers (in experiments 2 and 3) were fattened, and shipped to the carcass market in the Oita prefecture. Semen or blood of the sires and adipose tissues of the progeny steers were collected for SNP genotyping. DNA samples were prepared from the materials according to standard protocols.

The predicted breeding values of the sires and the progeny steers for marbling score and subcutaneous fat thickness were obtained from the Oita recording system for beef cattle reported by Sasaki et al. [16], and used as phenotypic values in this study. In the recording system, the breeding values were predicted from carcass records of Japanese Black steers and heifers, fattened in the Oita prefecture. The fattened animals were shipped to various carcass markets from 1988 to 2003, where they were slaughtered and their carcasses evaluated. The data were edited to connect across subclasses such that each market-year subclass had 50 or more animals and each farm had 10 or more animals. The final number of animals was 48,045, and there were 89 market-year subclasses, 332 farms, and 228 sires.

Marbling score and subcutaneous fat thickness were measured on carcasses dissected at the sixth and seventh rib section, according to the Japanese meat grading system by certified graders from the Japan Meat Grading Association (Tokyo, Japan) [4]. Marbling score was scored from 1 to 12 (beef marbling standard number) with a standard model panel, in which higher scores correspond to more marbling. Score 1 likely corresponds to "Practically Devoid", "Traces", "Slight" and "Small", score 2 or 3 "Modest", "Moderate" and "Slightly Abundant", and score 4 or 5 "Moderately Abundant". Subcutaneous fat thickness was measured at the rib interface perpendicular to the outside surface at a point 3-fourths the length of the longissimus muscle from its chine bone end.

Data were analyzed by the REML method using the MTDFREML programs [17], and genetic and environmental variances were estimated. The BLUP option in the programs using the estimated variance components was chosen to predict the breeding values of animals with a single trait model. Sex, market-year, and farm were considered fixed effects. Fattening period and slaughter age were also considered as up to quadratic covariates. The fattening period denotes the period from the start of fattening to shipping to market for each animal. These fixed effects were all significant (P < 0.001). Random effects included additive genetic effect of the individuals; that is, the animal model was adopted to predict the breeding values.

This study conformed to the guidelines for animal experimentation of the Graduate School of Agriculture, Kyoto University (Kyoto, Japan).

SNP genotyping

The g.231054C>T was genotyped using PCR-restriction fragment length polymorphism (RFLP) method. PCR primers used for PCR-RFLP were 5'-TCATCTCCTAACTACTTCCCA-3' and 5'-ACAAAATCTGAACCTGGCTT-3' (Nucleotide positions relative to the putative transcription initiation site of the TTN gene were -842 to -822 and -616 to -635, respectively). PCR amplifications were carried out as described in SNP detection section, using a final volume of 20 μl and the annealing temperature of 55°C. An aliquot of PCR-amplified products was digested at 37°C for 1 h with restriction enzyme Hpy CH4III, and electrophoresed on a 3.0% agarose gel. Agarose gels were stained with ethidium bromide and photographed under an ultraviolet light. Using this method, 227-bp PCR fragments containing the SNP site were amplified, and Hpy CH4III-digested into 36- and 191-bp fragments at the C allele, but not at the T allele: the CC homozygotes, the TT homozygotes, and the CT heterozygotes resulted in 2 bands (36 and 191 bp), 1 band (227 bp), and 3 bands (36, 191, and 227 bp).

Experiment 1

Genotyping 101 sires for the g.231054C>T revealed 50 animals homozygous for C allele, 47 animals heterozygous for C allele and T allele, and 4 animals homozygous for T allele. The 4 TT homozygotes were too small sample size to give reliable estimates, and then excluded from our statistical analysis. Statistically significant differences among the genotypes of the SNP were detected for marbling (P = 0.004), but not for subcutaneous fat thickness (P = 0.984), by the analysis with the model that included the SNP genotype as the fixed effect and the sire (father of the sire) as the random effect (Table 2). The marbling score was significantly higher in the CT heterozygotes than in the CC homozygotes (Table 2).

Table 2 Effect of the SNP genotypes on marbling and subcutaneous fat thickness in experiment 1
Full size table

Experiment 2

To better estimate the effect of the g.231054C>T T allele on marbling and subcutaneous fat thickness, we performed experiment 2. The interaction between the SNP genotype and the sire was not statistically significant (P = 0.320 for marbling, and P = 0.192 for subcutaneous fat thickness) in the model that included the SNP genotype and the sire as the fixed effects and their interaction, and was excluded from our statistical model. In the model without the interaction effect, the SNP genotype had the statistically significant effect on marbling (P = 0.046), but not subcutaneous fat thickness (P = 0.749) (Table 3). The marbling score was significantly higher in the TT homozygotes than in the CC homozygotes, and that of the heterozygotes intermediate between those of the 2 homozygotes (Table 3).

Table 3 Effect of the SNP genotypes on marbling and subcutaneous fat thickness in experiment 2
Full size table

Experiment 3

To further verify the association of the g.231054C>T with marbling, we carried out experiment 3. The steers in this experiment could be grouped only according to the alleles that they received from their dams, which are considered to be a random sample of a general population in Japanese Black beef cattle. Therefore, this experiment likely allowed a linkage disequilibrium estimate of the effect of the SNP. The interaction between the SNP genotype and the sire was not statistically significant (P = 0.208 for marbling, and P = 0.980 for subcutaneous fat thickness) in the model that included the SNP genotype and the sire as the fixed effects and their interaction, and was excluded from our statistical model. In the model without the interaction effect, the SNP genotype effect reached marginal significance (P = 0.051) for marbling, but not for subcutaneous fat thickness (P = 0.310) (Table 4). The marbling score exhibited a tendency to be higher in the CT heterozygotes than in the CC homozygotes (Table 4).

Table 4 Effect of the SNP genotypes on marbling and subcutaneous fat thickness in experiment 3
Full size table

Based on 3 experiments, we showed that the g.231054C>T is associated with marbling in Japanese Black beef cattle, with the T allele resulting in high levels of marbling. Especially, in experiments 2 and 3, because the dams can be considered to represent a random sample of the Japanese Black population, the association is likely to be true. We used the predicted breeding values as phenotypic values, and then sire contributions to the predicted breeding values were the same for all offspring, even if the heterozygous sires were used in experiment 2. Thus, it is likely to be desirable to use larger numbers of dams in the analyses using paternal half-sib progeny steers such as experiments 2 and 3. We should note that the experiments in this study were the case. Further, the association of the g.231054C>T with marbling was corroborated by an independent study using 104 paternal half-sib Japanese Black families with a total of 821 progeny steers (S. Sukegawa, unpublished data).

Based on the association of the g.231054C>T with marbling, together with TTN expression difference between low-marbled (with g.231054C>T C allele) and high-marbled steer groups (with g.231054C>T C and T alleles) and co-localization of the marbling quantitative trait locus with the TTN, we can hypothesize that the SNP in the promoter region might have an impact on TTN expression and also marbling by affecting TTN promoter activity. However, the SNP is not located within any of as yet identified canonical sequences involved in gene transcription. Thus, a more likely event is that the TTN SNP is in linkage disequilibrium with an unidentified and truly relevant mutation, rather than functional and a causal mutation for marbling.

The effect of genotypes of the SNP was not statistically significant (P > 0.05) for subcutaneous fat thickness. Further, the marbling quantitative trait locus corresponding to chromosomal position of the TTN did not have statistically significant effect on subcutaneous fat thickness [10, 11]. Thus, it is likely that the TTN SNP is not associated with subcutaneous fat thickness in Japanese Black beef cattle. This might be supported by the fact that Japanese Black breed exhibits low genetic correlation between marbling and subcutaneous fat thickness [7].

Several previous studies have reported polymorphisms associated with beef marbling using beef cattle breeds other than Japanese Black [18–27]. We have recently reported that SNPs in the EDG1 gene were associated with marbling in Japanese Black breed [8]. Thus, our present study seems to be the second report to show polymorphisms associated with marbling using Japanese Black breed.

The information on the TTN SNP obtained in this study, as well as the EDG1 SNPs, may be applied to effective marker-assisted selection to increase the levels of marbling in Japanese Black beef cattle. Further replicate studies will be needed to confirm the allelic association observed here, and to expand the results to evaluate all possible genotypic combinations of alleles.