Effect of two non-synonymous ecto-5'-nucleotidase variants on the genetic architecture of inosine 5'-monophosphate (IMP) and its degradation products in Japanese Black beef

Abstract

Background: Umami is a Japanese term for the fifth basic taste and is an important sensory property of beef palatability. Inosine 5'-monophosphate (IMP) contributes to umami taste in beef. Thus, the overall change in concentration of IMP and its degradation products can potentially affect the beef palatability. In this study, we investigated the genetic architecture of IMP and its degradation products in Japanese Black beef. First, we performed genome-wide association study (GWAS), candidate gene analysis, and functional analysis to detect the causal variants that affect IMP, inosine, and hypoxanthine. Second, we evaluated the allele frequencies in the different breeds, the contribution of genetic variance, and the effect on other economical traits using the detected variants.

Results: A total of 574 Japanese Black cattle were genotyped using the Illumina BovineSNP50 BeadChip and were then used for GWAS. The results of GWAS showed that the genome-wide significant single nucleotide polymorphisms (SNPs) on BTA9 were detected for IMP, inosine, and hypoxanthine. The ecto-5'-nucleotidase (NT5E) gene, which encodes the enzyme NT5E for the extracellular degradation of IMP to inosine, was located near the significant region on BTA9. The results of candidate gene analysis and functional analysis showed that two non-synonymous SNPs (c.1318C > T and c.1475 T > A) in NT5E affected the amount of IMP and its degradation products in beef by regulating the enzymatic activity of NT5E. The Q haplotype showed a positive effect on IMP and a negative effect on the enzymatic activity of NT5E in IMP degradation. The two SNPs were under perfect linkage disequilibrium in five different breeds, and different haplotype frequencies were seen among breeds. The two SNPs contribute to about half of the total genetic variance in IMP, and the results of genetic relationship between IMP and its degradation products showed that NT5E affected the overall concentration balance of IMP and its degradation products. In addition, the SNPs in NT5E did not have an unfavorable effect on the other economical traits.

Conclusion: Based on all the above findings taken together, two non-synonymous SNPs in NT5E would be useful for improving IMP and its degradation products by marker-assisted selection in Japanese Black cattle.

Keywords: GWAS; IMP; Japanese Black cattle; Meat quality; NT5E.

Fig. 1

Schematic representation of the pathway of formation and degradation of inosine 5′-monophosphate (IMP) in muscle. Each abbreviation is adenosine triphosphate (ATP), adenosine diphosphate (ADP), adenosine 5′-monophosphate (AMP), ecto-5′-nucleotidase (NT5E), and purine nucleoside phosphorylase (PNP). Black and red characters indicate products, and blue and black characters in italics indicate enzymes. This schematic representation is based on previous reports [–12]

Fig. 2

Genome-wide plots of p-values (˗log₁₀) for significantly associated loci. a Inosine 5′-monophosphate (IMP), b Inosine, and c Hypoxanthine. The x-axis indicates the chromosome number, and the y-axis indicates p-values (˗log₁₀). Dashed red line indicates the threshold of the Bonferroni 5% significance level

Fig. 3

The significant region and linkage disequilibrium (LD) from 65.4 to 66.6 Mbp on BTA 9. a The regional plots of the locus are associated with inosine 5′-monophosphate (IMP), inosine, and hypoxanthine. The x-axis indicates the Mbp, and the y-axis indicates p-values (˗ log₁₀). The gene loci and their strand were annotated based on Btau4.6 assembly from the bovine genome database (http://bovinegenome.org/). The dashed red line indicates the threshold of the Bonferroni 5% significance level. The p-values of SNPs on the SNP array (unfilled points) and detected variants in ecto-5′-nucleotidase (NT5E) (filled points) were plotted. b LD coefficients (r²) between the SNPs in this region. Black fields display r² values >0.80, and white and gray fields display r² values <0.80

Fig. 4

The schematic structure and the SNP features of ecto-5′-nucleotidase (NT5E). a Schematic representation of the positions of variants from the 5′-upstream region to the 3′-UTR in NT5E. The detailed positions and names of the variants are shown in Table 1. The Q and q haplotypes are defined by the genotypes of the three non-synonymous SNPs on exon 7 and exon 8 and two SNPs on 3′-UTR of exon 9. The Q haplotype has a positive effect on inosine 5′-monophosphate (IMP). Bovine reference (Ref) allele from Gene Bank accession no. NM_174129.3 is also shown. b The IMPase activity of ecto-5′-nucleotidase in COS-7 cells. Q-Q-Q, the construct with Q haplotype; q-q-q, the construct with q haplotype; Q-q-q, the construct mutated from q haplotype to Q allele in c.1318C > T; q-Q-q, the construct mutated from q haplotype to Q allele in c.1475 T > A; q-q-Q, the construct mutated from q haplotype to Q allele in c.1526A > G. The superscript letters indicate significant differences among five constructs tested by analysis of variance followed by a Tukey HSD (honestly significant difference) multiple comparison test (p-value <0.05). c The allelic imbalance test for levels of NT5E mRNA in the heterozygotes. The y-axis shows the ratio of peak height of the Q allele over the q allele in the genomic DNA (gDNA) and the complementary DNA (cDNA) from the same animal. Red bars indicate the mean expression. No significant (N.S.) difference was shown between them. d Multiple sequence alignment of the regions flanking p.His440Tyr, p.Val492Glu, and p.Gln509Arg. Q haplotype and q haplotype sequences of the cattle are shown on the top and other mammalian sequences are shown below

Fig. 5

Predicted structure of ecto-5′-nucleotidase (NT5E). a Predicted structure of NT5E modeled by the SWISS-MODEL server [18, 19]. b Docking of the predicted NT5E structure with inosine 5′-monophosphate (IMP) ligand. Three amino acids based on the three non-synonymous SNPs (p.His440, p.Val492, and p.Arg509) are shown as red sticks, and IMP ligand is shown as a blue stick

Fig. 6

Proposed model for the difference of ecto-5′-nucleotidase (NT5E) enzyme activity between Q and q haplotypes in beef. IMP, inosine 5′-monophosphate; Hx, hypoxanthine