We recently mapped a quantitative trait locus (QTL) with a major effect on milk composition-particularly fat content-to the centromeric end of bovine chromosome 14. We subsequently exploited linkage disequilibrium to refine the map position of this QTL to a 3-cM chromosome interval bounded by microsatellite markers BULGE13 and BULGE09. We herein report the positional candidate cloning of this QTL, involving (1) the construction of a BAC contig spanning the corresponding marker interval, (2) the demonstration that a very strong candidate gene, acylCoA:diacylglycerol acyltransferase (DGAT1), maps to that contig, and (3) the identification of a nonconservative K232A substitution in the DGAT1 gene with a major effect on milk fat content and other milk characteristics.[The sequence data described in this paper have been submitted to the GenBank data library under accession number AY065621.]

We recently used a positional cloning approach to identify a nonconservative lysine to alanine substitution (K232A) in the bovine DGAT1 gene that was proposed to be the causative quantitative trait nucleotide underlying a quantitative trait locus (QTL) affecting milk fat composition, previously mapped to the centromeric end of bovine chromosome 14. We herein generate genetic and functional data that confirm the causality of the DGAT1 K232A mutation. We have constructed a high-density single-nucleotide polymorphism map of the 3.8-centimorgan BULGE30 -BULGE9 interval containing the QTL and show that the association with milk fat percentage maximizes at the DGAT1 gene. We provide evidence that the K allele has undergone a selective sweep. By using a baculovirus expression system, we have expressed both DGAT1 alleles in Sf9 cells and show that the K allele, causing an increase in milk fat percentage in the live animal, is characterized by a higher Vmax in producing triglycerides than the A allele.A quantitative trait locus (QTL) with major effect on milk fat composition has been mapped to the centromeric end of bovine chromosome 14 (1, 2). Linkage disequilibrium (LD) was used to refine the map position of the QTL to a 3.8-centimorgan interval bounded by microsatellite markers BULGE30 and BULGE9 (3, 4). A bacterial artificial chromosome (BAC) contig spanning this interval was constructed and shown to contain a very strong positional candidate: diacylglycerol acyl transferase 1 (DGAT1) (5). DGAT1 indeed catalyzes the last step in triglyceride synthesis (6) and abrogates milk yield when knocked out in the mouse (7). By sequencing the DGAT1 gene from individuals with known QTL genotype, a nonconservative lysine to alanine substitution was identified at position 232, and shown to be associated with a major effect on milk yield and composition in several dairy cattle populations and breeds (5,8,9). The DGAT1 K232A mutation was therefore considered to be the likely quantitative trait nucleotide underlying the BTA14 QTL effect.However, based on the available data, we could not formally exclude that the effect observed with the K232A polymorphism was, in fact, caused by another mutation, located in the same or in another gene mapping to the BULGE30-BULGE9 interval that would be in strong LD with K232A. To resolve this issue, we herein (i) describe the development of a high-density map of single-nucleotide polymorphisms (SNP) of the BULGE30-BULGE9 interval and show that the association with milk yield and composition is strongest for the DGAT1 SNPs, thereby strongly incriminating that gene; (ii) present evidence that the K allele has been under positive selection supporting its effect on the functionality of DGAT1; and (iii) demonstrate that the K232A mutation increases the activity of the enzyme in a way that is in agreement with its effect on phenotype.

We report mapping of a quantitative trait locus (QTL) with a major effect on bovine stature to a ∼780-kb interval using a Hidden Markov Model-based approach that simultaneously exploits linkage and linkage disequilibrium. We re-sequenced the interval in six sires with known QTL genotype and identified 13 clustered candidate quantitative trait nucleotides (QTNs) out of >9,572 discovered variants. We eliminated five candidate QTNs by studying the phenotypic effect of a recombinant haplotype identified in a breed diversity panel. We show that the QTL influences fetal expression of seven of the nine genes mapping to the ∼780-kb interval. We further show that two of the eight candidate QTNs, mapping to the PLAG1-CHCHD7 intergenic region, influence bidirectional promoter strength and affect binding of nuclear factors. By performing expression QTL analyses, we identified a splice site variant in CHCHD7 and exploited this naturally occurring null allele to exclude CHCHD7 as single causative gene.

A genome-wide linkage disequilibrium (LD) map was generated using microsatellite genotypes (284 autosomal microsatellite loci) of 581 gametes sampled from the dutch black-and-white dairy cattle population. LD was measured between all marker pairs, both syntenic and nonsyntenic. Analysis of syntenic pairs revealed surprisingly high levels of LD that, although more pronounced for closely linked marker pairs, extended over several tens of centimorgan. In addition, significant gametic associations were also shown to be very common between nonsyntenic loci. Simulations using the known genealogies of the studied sample indicate that random drift alone is likely to account for most of the observed disequilibrium. No clear evidence was obtained for a direct effect of selection ("Bulmer effect"). The observation of long range disequilibrium between syntenic loci using low-density marker maps indicates that LD mapping has the potential to be very effective in livestock populations. The frequent occurrence of gametic associations between nonsyntenic loci, however, encourages the combined use of linkage and linkage disequilibrium methods to avoid false positive results when mapping genes in livestock.Recently, linkage disequilibrium (LD) has received considerable attention as it may be exploited to more effectively map genes underlying both simple and complex (dichotomous and continuously distributed) traits (Terwilliger and Weiss 1998). The potential advantage of LD mapping over conventional linkage analysis performed within families lies in the use of "historical" recombinants, thereby increasing mapping resolution (e.g., Hästbacka et al. 1992;Talbot et al. 1999) and power. To be effective, however, LD-mapping requires a marker density compatible with the distances across which LD extends in the population of interest. Kruglyak (1999) estimated by simulation that useful levels of LD were unlikely to extend beyond an average distance of 3 kb in the human, thereby implying the need for a marker map comprising ∼500,000 SNPs Although experimental LD data are accumulating in the human (e.g., Laan and Pääbo 1997; Nickerson et al. 1998) and some primate species (Crouau-Roy et al. 1996), little is known about the extent of LD in most other mammals, including domestic species. In this paper, we have used genotypes obtained with a panel of 284 microsatellites to measure genome-wide LD in the dutch black-andwhite dairy cattle population. We make the remarkable observation that intrachromosomal LD extends over several tens of centimorgans, and that gametic phase disequilibrium is common between non syntenic loci.

No abstract