M.H.P.W. Visker

Milk Fatty Acid Unsaturation : Genetic Parameters and Effects of Stearoyl-CoA Desaturase (SCD1) and Acyl CoA: Diacylglycerol Acyltransferase 1 (DGAT1)

With regard to human health aspects of milk fat, increasing the amount of unsaturated fatty acids in milk is an important selection objective. The cows diet has an influence on the degree of unsaturation, but literature suggests that genetics also plays a role. To estimate genetic variation in milk fatty acid unsaturation indices, milk fatty acid composition of 1,933 Dutch Holstein Friesian heifers was measured and unsaturation indices were calculated. An unsaturation index represents the concentration of the unsaturated product proportional to the sum of the unsaturated product and the saturated substrate. Intraherd heritabilities were moderate, ranging from 0.23 +/- 0.07 for conjugated linoleic acid (CLA) index to 0.46 +/- 0.09 for C16 index. We genotyped the cows for the SCD1 A293V and DGAT1 K232A polymorphisms, which are known to alter milk fatty acid composition. Both genes explain part of the genetic variation in unsaturation indices. The SCD1 V allele is associated with lower C10, C12, and C14 indices, and with higher C16, C18, and CLA indices in comparison to the SCD1 A allele, with no differences in total unsaturation index. In comparison to the DGAT1 K allele, the DGAT1 A allele is associated with lower C10, C12, C14, and C16 indices and with higher C18, CLA, and total indices. We conclude that selective breeding can contribute to higher unsaturation indices, and that selective breeding can capitalize on genotypic information of both the SCD1 A293V and the DGAT1 K232A polymorphism.

Effects of milk protein variants on the protein composition of bovine milk.

The effects of beta-lactoglobulin (beta-LG), beta-casein (beta-CN), and kappa-CN variants and beta-kappa-CN haplotypes on the relative concentrations of the major milk proteins alpha-lactalbumin (alpha-LA), beta-LG, alpha(S1)-CN, alpha(S2)-CN, beta-CN, and kappa-CN and milk production traits were estimated in the milk of 1,912 Dutch Holstein-Friesian cows. We show that in the Dutch Holstein-Friesian population, the allele frequencies have changed in the past 16 years. In addition, genetic variants and casein haplotypes have a major impact on the protein composition of milk and explain a considerable part of the genetic variation in milk protein composition. The beta-LG genotype was associated with the relative concentrations of beta-LG (A >> B) and of alpha-LA, alpha(S1)-CN, alpha(S2)-CN, beta-CN, and kappa-CN (B > A) but not with any milk production trait. The beta-CN genotype was associated with the relative concentrations of beta-CN and alpha(S2)-CN (A(2) > A(1)) and of alpha(S1)-CN and kappa-CN (A(1) > A(2)) and with protein yield (A(2) > A(1)). The kappa-CN genotype was associated with the relative concentrations of kappa-CN (B > E > A), alpha(S2)-CN (B > A), alpha-LA, and alpha(S1)-CN (A > B) and with protein percentage (B > A). Comparing the effects of casein haplotypes with the effects of single casein variants can provide better insight into what really underlies the effect of a variant on protein composition. We conclude that selection for both the beta-LG genotype B and the beta-kappa-CN haplotype A(2)B will result in cows that produce milk that is more suitable for cheese production.

Effect of polymorphisms in the FASN, OLR1, PPARGC1A, PRL and STAT5A genes on bovine milk-fat composition

The aim of our study was to estimate effects of polymorphisms in the ATP-binding cassette G2 (ABCG2), fatty acid synthase (FASN), oxidized low-density lipoprotein receptor 1 (OLR1), peroxysome proliferator-activated receptor-gamma coactivator-1alpha (PPARGC1A), prolactin (PRL) and signal transducer and activator of transcription 5A (STAT5A) genes on milk production traits and detailed milk-fat composition. Milk-fat composition phenotypes were available for 1905 Dutch Holstein-Friesian cows. First, the presence of each SNP in the Dutch Holstein-Friesian population was evaluated by direct sequencing of the PCR product surrounding the SNP in 22 proven Dutch Holstein-Friesian bulls. The ABCG2 SNP did not segregate in the bull population. Second, we genotyped the cows for the FASN(g.16024G>A), FASN(g.17924A>G), OLR1(g.8232C>A), PPARGC1A(c.1790+514G>A), PPARGC1A(c.1892+19G>A), PRL(g.8398G>A) and STAT5A(g.9501G>A) polymorphisms, and estimated genotype effects on milk production traits and milk-fat composition. FASN(g.17924A>G) and OLR(g.8232C>A) had a significant effect (P < 0.05) on milk-fat percentage. However, we were not able to confirm results reported in the literature that showed effects of all evaluated polymorphisms on milk-fat percentage or milk-fat yield. All polymorphisms showed significant effects (P < 0.05) on milk-fat composition. The polymorphisms in FASN and STAT5A, which had an effect on C14:0 and were located on chromosome 19, could not fully explain the quantitative trait locus for C14:0 that was previously detected on chromosome 19 in a genome-wide scan using linkage analysis.

Genome-wide association of milk fatty acids in Dutch dairy cattle

BackgroundIdentifying genomic regions, and preferably individual genes, responsible for genetic variation in milk fat composition of bovine milk will enhance the understanding of biological pathways involved in fatty acid synthesis and may point to opportunities for changing milk fat composition via selective breeding. An association study of 50,000 single nucleotide polymorphisms (SNPs) was performed for even-chain saturated fatty acids (C4:0-C18:0), even-chain monounsaturated fatty acids (C10:1-C18:1), and the polyunsaturated C18:2cis9,trans11 (CLA) to identify genomic regions associated with individual fatty acids in bovine milk.ResultsThe two-step single SNP association analysis found a total of 54 regions on 29 chromosomes that were significantly associated with one or more fatty acids. Bos taurus autosomes (BTA) 14, 19, and 26 showed highly significant associations with seven to ten traits, explaining a relatively large percentage of the total additive genetic variation. Many additional regions were significantly associated with the fatty acids. Some of the regions harbor genes that are known to be involved in fat synthesis or were previously identified as underlying quantitative trait loci for fat yield or content, such as ABCG2 and PPARGC1A on BTA 6; ACSS2 on BTA 13; DGAT1 on BTA 14; ACLY, SREBF1, STAT5A, GH, and FASN on BTA 19; SCD1 on BTA26; and AGPAT6 on BTA 27.ConclusionsMedium chain and unsaturated fatty acids are strongly influenced by polymorphisms in DGAT1 and SCD1. Other regions also showed significant associations with the fatty acids studied. These additional regions explain a relatively small percentage of the total additive genetic variance, but they are relevant to the total genetic merit of an individual and in unraveling the genetic background of milk fat composition. Regions identified in this study can be fine mapped to find causal mutations. The results also create opportunities for changing milk fat composition through breeding by selecting individuals based on their genetic merit for milk fat composition.

Genetic parameters for major milk proteins in Dutch Holstein-Friesians

The objective of this study was to estimate genetic parameters for major milk proteins. One morning milk sample was collected from 1,940 first-parity Holstein-Friesian cows in February or March 2005. Each sample was analyzed with capillary zone electrophoresis to determine the relative concentrations of the 6 major milk proteins. The results show that there is considerable genetic variation in milk protein composition. The intraherd heritabilities for the relative protein concentrations were high and ranged from 0.25 for beta-casein to 0.80 for beta-lactoglobulin. The intraherd heritability for the summed whey fractions (0.71) was higher than that for the summed casein fractions (0.41). Further, there was relatively more variation in the summed whey fraction (coefficient of variation was 11% and standard deviation was 1.23) compared with the summed casein fraction (coefficient of variation was 2% and standard deviation was 1.72). For the caseins and alpha-lactalbumin, the proportion of phenotypic variation explained by herd was approximately 14%. For beta-lactoglobulin, the proportion of phenotypic variation explained by herd was considerably lower (5%). Eighty percent of the genetic correlations among the relative contributions of the major milk proteins were between -0.38 and +0.45. The genetic correlations suggest that it is possible to change the relative proportion of caseins in milk. Strong negative genetic correlations were found for beta-lactoglobulin with the summed casein fractions (-0.76), and for beta-lactoglobulin with casein index (-0.98). This study suggests that there are opportunities to change the milk protein composition in the cows milk using selective breeding.

Whole-genome association study for milk protein composition in dairy cattle

Our objective was to perform a genome-wide association study for content in bovine milk of α(S1)-casein (α(S1)-CN), α(S2)-casein (α(S2)-CN), β-casein (β-CN), κ-casein (κ-CN), α-lactalbumin (α-LA), β-lactoglobulin (β-LG), casein index, protein percentage, and protein yield using a 50K single nucleotide polymorphism (SNP) chip. In total, 1,713 Dutch Holstein-Friesian cows were genotyped for 50,228 SNP and a 2-step association study was performed. The first step involved a general linear model and the second step used a mixed model accounting for all family relationships. Associations with milk protein content and composition were detected on 20 bovine autosomes. The main genomic regions associated with milk protein composition or protein percentage were found on chromosomes 5, 6, 11, and 14. The number of chromosomal regions showing significant (false discovery rate <0.01) effects ranged from 3 for β-CN and 3 for β-LG to 12 for α(S2)-CN. A genomic region on Bos taurus autosome (BTA) 6 was significantly associated with all 6 major milk proteins, and a genomic region on BTA 11 was significantly associated with the 4 caseins and β-LG. In addition, regions were detected that only showed a significant effect on one of the milk protein fractions: regions on BTA 13 and 22 with effects on α(S1)-CN; regions on BTA 1, 9, 10, 17, 19, and 28 with effects on α(S2)-CN; a region on BTA 6 with an effect on β-CN; regions on BTA 13 and 21 with effects on κ-CN; regions on BTA 1, 5, 9, 16, 17, and 26 with effects on α-LA; and a region on BTA 24 with an effect on β-LG. The proportion of genetic variance explained by the SNP showing the strongest association in each of these genomic regions ranged from <1% for α(S1)-CN on BTA 22 to almost 100% for casein index on BTA 11. Variation associated with regions on BTA 6, 11, and 14 could in large part but not completely be explained by known protein variants of β-CN (BTA 6), κ-CN (BTA 6), and β-LG (BTA 11) or DGAT1 variants (BTA 14). Our results indicate 3 regions with major effects on milk protein composition, in addition to several regions with smaller effects involved in the regulation of milk protein composition.

Genome-wide scan for bovine milk-fat composition. I. Quantitative trait loci for short- and medium-chain fatty acids

A genome-wide scan was performed to identify quantitative trait loci (QTL) for short- and medium-chain fatty acids (expressed in wt/wt %). Milk samples were available from 1,905 cows from 398 commercial herds in the Netherlands, and milk-fat composition was measured by gas chromatography. DNA was available from 7 of the paternal half-sib families: 849 cows and their 7 sires. A genetic map was constructed comprising 1,341 SNP and 2,829 cM, with an average information content of 0.83. Multimarker interval mapping was used in an across-family regression on corrected phenotypes for the 7 half-sib families. Four QTL were found: on Bos taurus autosome (BTA) 6, a QTL was identified for C6:0 and C8:0; on BTA14, a QTL was identified for fat percentage, all odd-chain fatty acids, and C14:0, C16:0, C16:1, and their unsaturation indices; on BTA19, a QTL affected C14:0; and on BTA26, a QTL was identified for the monounsaturated fatty acids and their unsaturation indices. The QTL explained 3 to 19% of phenotypic variance. Furthermore, 49 traits with suggestive evidence for linkage were found on 21 chromosomes. Additional analyses revealed that the QTL on BTA14 was most likely caused by a mutation in DGAT1, whereas the QTL on BTA26 was most likely caused by a mutation in the SCD1 gene. Quantitative trait loci that affect specific fatty acids might increase the understanding of physiological processes regarding fat synthesis and the position of the causal genes.

Novel polymorphisms in the bovine β‐lactoglobulin gene and their effects on β‐lactoglobulin protein concentration in milk

The aim of our study was to detect new polymorphisms in the bovine beta-lactoglobulin (beta-LG) gene with significant effects on beta-LG protein concentration. Genomic DNA samples from 22 proven bulls were screened for polymorphisms in the coding and promoter regions of the beta-LG gene. In total, 50 polymorphisms were detected. Two single nucleotide polymorphisms (SNPs) (g.1772G>A and g.3054C>T) lead to amino acid changes and are the causal genetic polymorphisms of beta-LG protein variants A and B. Forty-two polymorphisms were in complete linkage disequilibrium (LD) with beta-LG protein variants A and B. Any of these 42 polymorphisms can be involved in the differential expression of the respective A and B alleles of the beta-LG gene. The eight polymorphisms not in complete LD with beta-LG protein variants A and B and the two polymorphisms causing the amino acid changes were genotyped in a set of 208 cows: 106 animals homozygous for beta-LG protein variant A and 102 animals homozygous for beta-LG protein variant B. Of these eight polymorphisms, six SNPs segregated only within the cows homozygous for beta-LG protein variant A and two SNPs segregated only within the cows homozygous for beta-LG protein variant B. One of the eight polymorphisms had a significant effect on beta-LG protein concentration. This SNP, g.-731G>A, segregated only within the 106 cows homozygous for beta-LG protein variant A. Within these cows, adjusted relative beta-LG protein concentration was reduced by 1.22% (w/w) in animals homozygous g.-731AA compared with animals homozygous g.-731GG.

Short communication: Genome-wide scan for bovine milk-fat composition. II. Quantitative trait loci for long-chain fatty acids

We present the results of a genome-wide scan to identify quantitative trait loci (QTL) that contribute to genetic variation in long-chain milk fatty acids. Milk-fat composition phenotypes were available on 1,905 Dutch Holstein-Friesian cows. A total of 849 cows and their 7 sires were genotyped for 1,341 single nucleotide polymorphisms across all Bos taurus autosomes (BTA). We detected significant QTL on BTA14, BTA15, and BTA16: for C18:1 cis-9, C18:1 cis-12, C18:2 cis-9,12, CLA cis-9,trans-11, C18:3 cis-9,12,15, the C18 index, the total index, total saturated fatty acids, total unsaturated fatty acids (UFA), and the ratio of saturated fatty acids:unsaturated fatty acids on BTA14; for C18:1 trans fatty acids on BTA15; and for the C18 and CLA indices on BTA16. The QTL explained 3 to 19% of the phenotypic variance. Suggestive QTL were found on 16 other chromosomes. The diacylglycerol acyltransferase 1 (DGAT1) K232A polymorphism on BTA14, which is known to influence fatty acid composition, most likely explains the QTL that was detected on BTA14.

Comparison of information content for microsatellites and SNPs in poultry and cattle

Data were available for 12 poultry microsatellites and 29 poultry single nucleotide polymorphisms (SNPs), and for 34 cattle microsatellites and 36 cattle SNPs. Stochastic permutation was used to determine the number of SNPs needed to obtain the same average information content as a given number of microsatellites. For poultry, the information content averaged 0.71 for the 12 microsatellites compared to 0.72 for the 29 SNPs. For cattle, the information content averaged 0.92 for the 34 microsatellites compared with 0.79 for the 36 SNPs. This study shows that, for each microsatellite, three SNPs are needed to obtain the same average information content.