FIELD OF THE INVENTION

&null;0001&null; This invention relates to an application of marker assisted selection of bovine for a quantitative trait loci (QTL) associated with milk production, particularly although by no means exclusively, by assaying for the presence of at least one allele which is associated with increased milk volume as well as improved milk composition. The present invention also relates to the gene associated with the QTL, various polymorphisms within the gene sequence, proteins encoded by these sequences as well as to the application of all of these in the farming industry.

BACKGROUND

&null;0002&null; The genetic basis of bovine milk production is of immense significance to the dairy industry. An ability to modulate milk volumes and content has the potential to alter farming practices and to produce products which are tailored to meet a range of requirements. In particular, a method of genetically evaluating bovine to select those which express desirable traits, such as increased milk production and improved milk composition, would be desirable.

&null;0003&null; To date, bovine genomics are poorly understood and little is known regarding the genes which are critical to milk production. While there have been reports of quantitative trait loci (QTLs) on bovine chromosome 14 postulated to be associated with milk production (Coppieters et al (1998)), the specific genes involved have not to date been identified.

&null;0004&null; Marker assisted selection, which provides the ability to follow a specific favourable genetic allele, involves the identification of a DNA molecular marker or markers that segregate with a gene or group of genes associated with a QTL. DNA markers have several advantages. They are relatively easy to measure and are unambiguous, and as DNA markers are co-dominant, heterozygous and homozygous animals can be distinctively identified. Once a marker system is established, selection decisions are able to be made very easily as DNA markers can be assayed at any time after a DNA containing sample has been collected from an individual infant or adult animal, or even earlier as it is possible to test embryos in vitro if such embryos are collected.

&null;0005&null; The applicants have now identified a gene responsible for the QTL effect on bovine chromosome 14 as well as a number of polymorphisms which are associated with distinct genetic merits of animals for milk composition and volume.

&null;0006&null; It is an object of the present invention to provide an application method for marker assisted selection of this bovine gene, and in particular, of the polymorphisms in the bovine gene which are associated with increased milk volume and altered milk composition; and/or to provide genetic markers for use in such a method; and/or to provide the nucleic acid and amino acid sequences of this gene and encoded polypeptide; and/or to provide animals selected using the method of the invention as well as milk produced by the selected animals; and/or to provide the public with a useful choice.

SUMMARY OF THE INVENTION

&null;0007&null; This invention relates to the discovery of the bovine Diacylglycerol-o-acyltransferase (DGAT1) gene and polymorphisms within the bovine DGAT1 gene which are associated with increased milk yield and altered milk composition. More specifically, several polymorphisms- in the bovine DGAT1 gene have been identified distinguishing multiple DGAT1 alleles in different cattle breeds. These polymorphisms include: K232A (Bases 6829/30 AA-CG nucleic acid change and K-A amino acid change); Nt984&null;8(Base 7438 A-G nucleic acid change); Nt984&null;26(Base 7456 C-T nucleic acid change); Nt1470&null;85(Base 8402 C-T nucleic acid change); Nt191&null;435 (Base 626 T-G nucleic acid change); Nt191-3321 (Base 3512 T-G nucleic acid change); Nt279&null;144 (Base 4040 T-C nucleic acid change); Nt279&null;1067 (Base 4963 A-G nucleic acid change); Nt279&null;1107 (Base 5003 G-A nucleic acid change); Nt358 (Base 5997 C-T nucleic acid change); Nt754&null;3 (Base 6892 G-A nucleic acid change); Nt897&null;32 (Base 7224/5 GG-AC nucleic acid change); Nt1251&null;42 (Base 7987 G-A nucleic acid change) as summarised in Table 1. In particular, DGAT1 alleles characterized by the K232A mutation have been identified as being associated with an increased milk volume and altered milk composition in animals dependent upon whether they are homozygous with or without the mutation or heterozygous carrying one mutated allele. More specifically, the presence of the K232A mutation results in a decrease in milkfat percentage, milkfat yield, solid fat content and milk protein percentage, while increasing milk volume and milk protein yield.

&null;0008&null; The present invention thus relates to the use of the polymorphisms in a method of identification and selection of a bovine having at least one of said polymorphisms as well as to providing markers specific for such identification. Kits comprising said markers for use in marker selection also form part of the present invention as do animals so selected, the milk produced by such selected animals and products produced from such milk, particularly as such milk and milk products affect processing and/or health characteristics for consumers.

&null;0009&null; In particular, the present invention is directed to a method of genotyping cows or bulls for one or more of the polymorphisms disclosed herein, selected cows or bulls so genotyped and milk and semen from said selected cows and bulls respectively.

&null;0010&null; According to a further aspect the present invention is directed to the isolated DGAT1 nucleic acid and allelic nucleic acid molecules comprising polymorphisms as well as to the proteins encoded thereby and their polypeptide sequences. Antibodies raised against said proteins are also contemplated, as are vectors comprising the nucleic acid molecules, host cells comprising the vectors; and protein molecules expressed in said host cells; and the application of all of them in the farming industry.

&null;0011&null; In particular, such applications include methods for modulating milk production and/or composition in a lactating bovine by affecting DGAT1 activity, by reducing the activity of DGAT1 (e.g. by use of specific ribozymes, antisense sequences and/or antibodies, or by transgenic technology to produce a &null;knock out&null; bovine and/or bovine with introduced transgenes containing the DGAT1 gene and/or variations of this gene driven by various promoters).

BRIEF DESCRIPTION OF THE DRAWINGS

&null;0012&null; The invention will now be described with reference to the Figures of the accompanying drawings in which:

&null;0013&null; FIG. 1: Shows a BAC contig spanning the BULGE13-BULGE09 interval relative to a schematic diagram of bovine chromosome 14 and a schematic diagram showing the location of the genetic markers. The most likely position of the QTL is shown as a bar on the FISH-anchored linkage map proximal to BTA14q. The BACs composing the contigs spanning the BULGE13-BULGE09 interval are shown as a series of horizontal lines. The symbols on each BAC indicate their individual STS content: solid circles correspond to STS derived from BAC ends, open boxes to microsatellite markers, and solid triangles to gene-specific Comparative Anchored Tagged Sequences. The arrow heads mark the BACs from which the respective BAC end STS were derived. The length of the lines do not reflect the actual insert size of the corresponding BACs. The BAC contig was aligned with the orthologous human HSA8q24.3 genomic &null;golden path&null; sequence represented according to the Ensembl Human Genome Server (http://www.ensembl.org/): individual sequence contigs are shown in alternating light and dark; a horizontal line indicates a gap in the sequence assembly; genetic markers are indicated under the contig map; the lines and boxes above the contig map represent &null;curated&null;, &null;predicted known&null; or &null;predicted novel&null; genes.

&null;0014&null; FIGS. 2a and 2b: Show the genomic sequence of the bovine DGAT1 gene. FIG. 2a is the 31 base pair sequence upstream but adjacent to the ATG or translation start site and is 5&null; UTR. FIG. 2b is the genomic sequence in the bovine DGAT1 gene from the ATG translation start site (base 1) through to genomic sequence flanking the gene at the 3&null; end. The significant features including intron/exon boundaries, polymorphic sites, polyadenylation signal, and alternate splicing site and some of the primer sequences used in the assays described herein, are indicated;

&null;0015&null; FIG. 3: Shows the genomic organization, four polymorphisms and haplotypes found in the bovine DGAT1 gene. Leader and trailer sequences are shown in light grey, coding sequences in dark grey and intronic sequences as a hollow line. The positions of four of the identified polymorphisms are marked as shown on the gene, and detailed in the underlying boxes including the corresponding sequence traces All the sequence variations are summarised in Table 1. The four DGAT1 haplotypes which were found in the Dutch and New-Zealand Holstein-Friesian population as defined by these polymorphisms are shown and referred to as &null;sHQ-D&null;, &null;sHQ-NZ&null;, &null;sHQ-III&null; for the fat increasing haplotypes and &null;shq&null; for the fat decreasing haplotype;

&null;0016&null; FIG. 4a: Shows the corresponding full length amino acid sequence for DGAT1 sequence of FIG. 2b including annotation of the amino acid substitution;

&null;0017&null; FIG. 4b: Shows the amino acid sequence predicted as a result of alternate splicing with exon VIII;

&null;0018&null; FIG. 5: Shows the multiple peptide alignment of a portion of the DGAT1 protein flanking the K232A substitution from Bos taurus, Bison bison, Ovis aries, Sus scrofa, Homo sapiens, Cercopithecus aethiops, Mus musculus domesticus and Rattus noruegicus showing the evolutionary conservation of the lysine mutated in the bovine K232A polymorphism;

&null;0019&null; FIG. 6: A. Shows the frequency distribution of observed DGAT1 SNP haplotypes in the Dutch and New Zealand Holstein-Friesian dairy cattle populations. B-D. Shows the frequency distribution of the combined microsatellite (BULGE09-BULGE1 1) and SNP DGAT1 haplotypes. The HQ-D and HQ-NZ haplotypes are shown; and

&null;0020&null; FIG. 7: Shows the lod score due to LD when including (&null;) or excluding (&null;) the four DGAT1 polymorphisms shown in FIG. 3 in a combined linkage and LD multipoint maximum likelihood mapping method. The lod score corresponds to the log10 of the ratio between the likelihood of the data assuming LD and linkage between the markers and the likelihood of the data assuming linkage in the absence of LD. The positions of the microsatellites and SNP markers utilized in the analysis are shown on the X-axis, while the position of the DGAT1 SNPs is marked by a red arrow at the top of the figure.

DETAILED DESCRIPTION OF THE INVENTION

&null;0021&null; It has been discovered for the first time that the DGAT1 gene in bovine is associated with the QTL on chromosome 14 which is linked with improved milk production traits. More particularly, a number of novel polymorphisms on the DGAT1 gene have been discovered. It is thought that one or more of these polymorphisms is responsible for these traits.

&null;0022&null; The method used for isolating genes which cause specific phenotypes is known as positional candidate cloning. It involves: (i) the chromosomal localisation of the gene which causes the specific phenotype using genetic markers in a linkage analysis; and (ii) the identification of the gene which causes the specific phenotype amongst the &null;candidate&null; genes known to be located in the corresponding region. Most of the time these candidate genes are selected from available mapping information in humans and mice.

&null;0023&null; The tools required to perform the initial localisation (step (i) above) are microsatellite marker maps, which are available for livestock species and are found in the public domain (Bishop et al., 1994; Barendse et al., 1994; Georges et al., 1995; and Kappes, 1997). The tools required for the positional candidate cloning, particularly the BAC libraries, (step (ii) above) are partially available from the public domain. Genomic libraries with large inserts constructed with Bacterial Artificial Chromosomes (BAC) are available in the public domain for most livestock species including cattle. For general principles of positional candidate cloning, see Collins, 1995 and Georges and Anderson, 1996.

&null;0024&null; Recently, a quantitative trait locus (QTL) with major effect on milk solids composition, located at the centromeric end of bovine chromosome 14, has been reported (Coppieters et al., (1998)). This QTL was shown to effect milk fat content and in particular to significantly affect protein %, volume, protein yield and fat yield of milk. The linkage study as well as subsequent marker assisted segregation analyses allowed for the identification of thirteen Holstein-Friesian sires predicted to be heterozygous &null;Qq&null; for the corresponding QTL (Coppieters et al., (1998); Riquet et al., (1999)).

&null;0025&null; Linkage disequilibrium methods were applied to refine the map position of the QTL to a &null;5 cM interval bounded by microsatellite markers BULGE09 and BULGE30.

&null;0026&null; A bovine DGAT1 nucleotide sequence was determined by the applicants and is shown in FIGS. 2a and 2b with the corresponding amino acid sequences (long and short forms) being shown in FIGS. 4a and 4b respectively. Table 1 sets out all the polymorphisms located to date with reference to the sequence in FIG. 2b. Some of the genetic polymorphisms identified in the bovine DGAT1 gene are reported in FIG. 3. The nucleic acid and protein sequences of the DGAT1 alleles including the K232A mutation are shown in FIGS. 2a and 2b (SEQ ID NOs: 3 and 1), annotated to show the alternatively spliced forms. The cDNA sequence is also set out in SEQ ID NO: 4.

&null;0027&null; The sequence information in the Figures gives rise to numerous, and separate, aspects of the invention.

&null;0028&null; In one aspect, the invention provides a method of determining genetic merit of a bovine with respect to milk composition and volume which comprises the step of determining the bovine DGAT1 genotyping state of said bovine. In particular, this method is useful for genotyping and selecting cows and bulls having the desired genotypic state so that milk and semen may be collected from said cows and bulls respectively. Such semen would be useful for breeding purposes to produce bovine having the desired genotypic and, as a result, phenotypic state. In addition, cows genotyped by the methods of the present invention are also useful for breeding purposes, particularly for breeding with the selected bulls and/or to be artificially inseminated with the semen from selected bulls. The embryos and offspring produced by such cows also form part of the present invention.

&null;0029&null; In one embodiment, the genotypic state is determined with respect to DNA obtained from said bovine.

&null;0030&null; Alternatively, said genotypic state is determined with reference to mRNA obtained from said bovine.

&null;0031&null; In yet a further embodiment, the genotypic state is determined with reference to the amino acid sequence of expressed bovine DGAT1 protein obtained from said bovine.

&null;0032&null; Conveniently, in said method, the genotypic state of DNA encoding bovine DGAT1 is determined, directly or indirectly.

&null;0033&null; Alternatively, in said method the genotypic state of at least one nucleotide difference from the nucleotide sequence encoding bovine DGAT1 is determined, directly or indirectly.

&null;0034&null; More specifically, in said method the genotypic state of bovine DGAT1 allele(s) characterised by one or more of the polymorphisms shown in Table 1 below, is determined, directly or indirectly.

1

TABL 1

Table of polymorphisms in the bovine DGAT1 gene

Start codon (atg); the a residue is denoted as position 1

Base number relative to

Nucleotide distance

Intron/exon

exonic sequence1

from start

substitution

&null;

Nt 191 &null;435

626

T-G

Intron 1

CAGTGCTAGGGG

CAGTGCGAGGGG

Nt 191 &null;3321

3512

T-G

Intron 1

GCATTGCGCT

GCATGGCGCT

Nt 279 &null;144

4040

T-C

Intron 2

TACCCTGGGAC

TACCCCGGGAC

Nt 279 &null;1067

4963

A-G

Intron 2

CTCTTAGCAGC

Nt 279 &null;1107

5003

G-A

Intron 2

ACAGGCAACT

ACAGACAACT

Nt 358

5997

C-T

Exon IV

TGTCTCTGTTC

TGTCTTTGTTC

Nt 692

6829

AA-GC

K232A Exon

GGTAAGAAGGCCAA (Q)

VIII*

GGTAAGGCGGCCAA (q)

Nt 754 &null;3

6892

G-A

Intron VIII

GCGGTGAGGAT

GCGGTAAGGAT

Nt 897 &null;32

7224

GG-AC

Intron X

GGGGGGGGGGGACTCT

GGGGGACGGGGACTCT

Nt 984 &null;8

7438

A-G

Intron XII*

GAGTGACCTGC

GAGTGGCCTGC

Nt 984 &null;26

7456

C-T

Intron XII*

GGACGCGTGGG

GGACGTGTGGG

Nt 1251 &null;42

7987

G-A

Intron XV

GGTGGGGGTGG

GGTGGAGGTGG

Nt 1470 &null;85

8402

C-T

3&null; flanking

CTGGGCGCAGC

region *

CTGGGTGCAGC

The numbers given are far the actual nucleotide or in the case of two nucleotide substitutions to the first nucleotide in the variation (counting 5&null;&null;to 3&null;

*More detail of these polymorphisms is given in FIG. 2b.

1e.g. Nt 191 represents nucletode number 191 from the start site of the coding sequence, &null;&null;435 represents number of nucleotides from and including base 192 in the genomic sequence (intron 1) to the polymorphic nucleotide

The polymorphic nucleotides are shaded

&null;0035&null; Preferably, the invention is directed to a method of determining the genotypic state of bovine DGAT1 allele(s) by determining the presence of the K232A polymorphism, either directly or indirectly.

&null;0036&null; There are numerous art standard methods known for determining whether a particular DNA sequence is present in a sample. An example is the Polymerase Chain Reaction (PCR). A preferred aspect of the invention thus includes a step in which ascertaining whether a polymorphism(s) in the sequence of DGAT1 DNA is present, includes amplifying the DNA in the presence of primers based on the nucleotide sequence of the DGAT1 gene and flanking sequence, and/or in the presence of a primer containing at least a portion of a polymorphism as disclosed herein and which when present results in altered relative milk lipid and protein production, and milk volume.

&null;0037&null; A primer of the present invention, used in PCR for example, is a nucleic acid molecule sufficiently complementary to the sequence on which it is based and of sufficient length to selectively hybridise to the corresponding portion of a nucleic acid molecule intended to be amplified and to prime synthesis thereof under in vitro conditions commonly used in PCR. Likewise, a probe of the present invention, is a molecule, for example a nucleic acid molecule of sufficient length and sufficiently complementary to the nucleic acid molecule of interest, which selectively binds under high or low stringency conditions with the nucleic acid sequence of interest for detection thereof in the presence of nucleic acid molecules having differing sequences.

&null;0038&null; In another aspect, the invention provides a method for determining the genetic merit of bovine with respect to milk content and volume with reference to a sample of material containing mRNA obtained from the bovine. This method includes ascertaining whether a polymorphism(s) in the sequence of the mRNA encoding DGAT1 is present. The presence of such polymorphisms again indicates an association with altered relative milk lipid and protein production and milk volume.

&null;0039&null; Again, if an amplification method such as PCR is used in ascertaining whether a polymorphism(s) in the sequence of the mRNA encoding (DGAT1) is present, the method includes reverse transcribing the mRNA using a reverse transcriptase to generate a cDNA and then amplifying the cDNA in the presence of a pair of primers complementary to a nucleotide sequence encoding a protein having biological activity of wild type DGAT1.

&null;0040&null; In a further aspect, the invention includes the use of a probe in the methods of genotyping according to the invention wherein the probe is selected from any 5 or more contiguous nucleotides of the DGAT1 sequence as shown in FIG. 2b, which is therefore sufficiently complementary with a nucleic acid sequence encoding such bovine DGAT1, or its complement, so as to bind thereto under stringent conditions. Diagnostic kits containing such a probe are also included. Such probes may be selected from ForAA (FAM): CGTTGGCCTTCTTA or DgatADGC (VIC): TTGGCCGCCTTACC. (SEQ ID NOs: 20 and 21 respectively.)

&null;0041&null; The invention further includes isolated nucleic acid molecules encoding the DGAT1 variant proteins i.e. those proteins encoded by SEQ ID NOs: 1 and 4 (FIG. 2b), comprising one or more polymorphisms of SEQ ID NOs. 7 to 19 (Table 1), or a fragment or variant thereof. Particularly, the invention includes an isolated nucleic acid molecule comprising a DNA molecule having in whole or in part the nucleotide sequence identified in FIG. 2b or which varies from the sequence due to the degeneracy of the genetic code, or a nucleic acid strand capable of hybridising with said nucleic acid molecule under stringent hybridisation conditions.

&null;0042&null; The invention includes isolated mRNA transcribed from DNA having a sequence which corresponds to a nucleic acid molecule of the invention.

&null;0043&null; The invention includes isolated DNA in a recombinant cloning vector and a prokaryotic or eukaryotic cell containing and expressing heterologous DNA of the invention.

&null;0044&null; The invention includes a transfected coil line which expreses a protein encoded by the nucleic acid molecules of the invention.

&null;0045&null; The invention also includes a primer composition useful for detection of the presence of one or more polymorphisms associated with improved milk production traits in bovine DNA encoding DGAT1 and/or the presence of DNA encoding a variant protein. In one form, the composition can include a nucleic acid primer substantially complementary to a nucleic acid sequence encoding DGAT1. The nucleic acid sequence can in whole or in part be that identified in FIG. 2b. Diagnostic kits including such a composition are also included.

&null;0046&null; The invention further provides a diagnostic kit useful in detecting DNA encoding a variant DGAT1 protein in bovine which includes first and second primers for amplifying the DNA, the primers being complementary to nucleotide sequences of the DNA upstream and downstream, respectively, of a polymorphism in the portion of the DNA encoding DGAT1 which results in altered relative milk lipid, solid fat content and protein production and milk volume, wherein at least one of the nucleotide sequences is elected to be from a non-coding region of the DGAT1 gene. The kit can also include a third primer complementary to a polymorphism, disclosed herein, located on the DGAT1 gene.

&null;0047&null; The invention includes a process for producing a protein of the invention, including preparing a DNA fragment including a nucleotide sequence which encodes the protein; incorporating the DNA fragment into an expression vector to obtain a recombinant DNA molecule which includes the DNA fragment and is capable of undergoing replication; transforming a host cell with the recombinant DNA molecule to produce a transformant which can express the protein; culturing the transformant to produce the protein; and recovering the protein from resulting cultured mixture.

&null;0048&null; Thus in a further aspect, the invention provides a purified protein encoded by the nucleic acid molecule of the invention and having biological activity of DGAT1. The terms &null;isolated&null; and &null;purified&null; as used herein, each refer to a protein substantially free of cellular material or culture medium when produced by recombinant DNA techniques, or chemical precursors or other chemicals when chemically synthesised. In certain preferred embodiments, the protein having biological activity of DGAT1 comprises an amino acid sequence and variants shown in FIGS. 4a and 4b (SEQ ID NOs: 2, 5 and 6). Furthermore, proteins having biological activity of DGAT1 that are encoded by nucleic acids which hybridise under stringent conditions to a nucleic acid comprising a nucleotide sequence shown in FIG. 2b (SEQ ID NOs: 1 and 4) are encompassed by the invention.

&null;0049&null; Proteins of the invention having DGAT1 activity can be obtained by expression of a nucleic acid coding sequence in a suitable host cell using techniques known in the art. Suitable host cells include prokaryotic or eukaryotic organisms or cell lines, for example, yeast, E. coli, insect cells and COS 1 cells. The recombinant expression vectors of the invention can be used to express a protein having DGAT1 activity in a host cell in order to isolate the protein. The invention provides a method of preparing a purified protein of the invention comprising introducing into a host cell a recombinant nucleic acid encoding the protein, allowing the protein to be expressed in the host cell and isolating and purifying the protein. Preferably, the recombinant nucleic acid is a recombinant expression vector. Proteins can be isolated from a host cell expressing the protein and purified according to standard procedures of the art, including ammonium sulfate precipitation, column chromatography (eg. ion exchange, gel filtration, affinity chromatography, etc.) electrophoresis, and ultimately, crystallisation (see generally &null;Enzyme Purification and Related Techniques&null;. Methods in Enzymology, 22, 233-577 (1971)).

&null;0050&null; Alternatively, the protein or parts thereof can be prepared by chemical synthesis using techniques well known in the chemistry or proteins such as solid phase synthesis (Merrifield, 1964), or synthesis in homogeneous solution (Houbenwcyl, 1987).

&null;0051&null; It will of course be understood that a variety of substitutions of amino acids is possible while preserving the structure responsible for activity of the DGAT1 proteins disclosed herein. Conservative substitutions are described in the patent literature, as for example, in U.S. Pat. No. 5,264,558 or 5,487,983. It is thus expected, for example, that interchange among non-polar aliphatic neutral amino acids, glycine, alanine, proline, valine and isoleucine, would be possible. Likewise, substitutions among the polar aliphatic neutral amino acids, serine, threonine, methionine, asparagine and glutamine could possibly be made. Substitutions among the charged acidic amino acids, aspartic acid and glutamic acid, could probably be made, as could substitutions among the charged basic amino acids, lysine and arginine. Substitutions among the aromatic amino acids, including phenylalanine, histidine, tryptophan and tyrosine would also likely be possible. These sorts of substitutions and interchanges are well known to those skilled in the art. Other substitutions might well be possible. Of course, it would also be expected that the greater percentage of homology ie. sequence similarity, of a variant protein with a naturally occurring protein, the greater the retention of activity.

&null;0052&null; A further advantage may be obtained through chimeric forms of the proteins, as known in the art. A DNA sequence encoding each entire protein, or a portion of the protein, could be linked, for example, with a sequence coding for the C-terminal portion of E. coli &null;-galactosidase to produce a fusion protein.

&null;0053&null; The proteins of the invention, or portions thereof, have numerous applications in turn. By way of example, each protein can be used to prepare antibodies which bind to a distinct epitope in an unconserved region of the protein. An unconserved region of the protein is one which does not have substantial sequence homology to other proteins.

&null;0054&null; Still further, the invention includes an antibody to a bovine DGAT1 variant protein encoded by a nucleotide sequence of the present invention as well as a diagnostic kit containing such an antibody.

&null;0055&null; Conventional methods can be used to prepare the antibodies.. For example, by using a DGAT1 peptide, polyclonal antisera or monoclonal antibodies can be made using standard methods. A mammal, (eg. a mouse, hamster, or rabbit) can be immunised with an immunogenic form of the peptide which elicits an antibody response in the mammal. Techniques for conferring immunogenicity on a peptide include conjugation to carriers or other techniques well known in the art. For example, the peptide can be administered in the presence of adjuvant. The progress of immunisation can be monitored by detection of antibody titers in plasma or serum. Standard ELISA or other immunoassay can be used to assess the levels of antibodies. Following immunisation, antisera can be obtained and, if desired, polyclonal antibodies isolated from the sera.

&null;0056&null; To produce monoclonal antibodies, antibody producing cells (lymphocytes) can be harvested from an immunised animal and fused with myeloma cells by standard somatic cell fusion procedures, thus immortalising these cells and yielding hybridoma cells. Such techniques are well known in the art. For example, the hybridoma technique originally developed by Kohler and Milstein (Kohler, 1975) as well as other techniques such as the human B-cell hybridoma technique (Kozbor, 1983) and screening of combinatorial antibody libraries (Huse, 1989). Hybridoma cells can be screened immunochemically for production of antibodies specifically reactive with the peptide, and monoclonal antibodies isolated.

&null;0057&null; The term &null;antibody&null; as used herein is intended to include fragments thereof which are also specifically reactive with the target protein. Antibodies can be fragmented using conventional techniques and the fragments screened for utility in the same manner as described above for whole antibodies. For example, F(ab&null;)2 fragments can be generated by treating antibody with pepsin. The resulting F(ab&null;)2 fragment can be treated to reduce disulfide bridges to produce Fab&null; fragments.

&null;0058&null; Another method of generating specific antibodies, or antibody fragments, reactive against the target proteins is to screen expression libraries encoding immunoglobulin genes, or portions thereof, expressed in bacteria, with peptides produced from the nucleic acid molecules of the present invention. For example, complete Fab fragments, VH regions and FV regions can be expressed in bacteria using phage expression libraries. See for example Ward et al., Huse et al., and McCafferty et al. (Ward, 1989); Huse 1989; McCafferty, 1990). Screening such libraries with, for example, a DGAT1 protein can identify immunoglobulin fragments reactive with that DGAT1. Alternatively, the SCID-hu mouse developed by Genpharm can be used to produce antibodies, or fragments thereof.

&null;0059&null; The polyclonal, monoclonal or chimeric monoclonal antibodies can be used to detect the proteins of the invention, portions thereof or closely related isoforms in various biological materials. For example, they can be used in an ELISA, radioimmunoassay or histochemical tests. Thus, the antibodies can be used to quantify the amount and location of a DGAT1 protein of the invention, portions thereof or closely related isoforms in a sample in order to determine the role of DGAT1 proteins. Using methods described hereinbefore, polyclonal, monoclonal antibodies, or chimeric monoclonal antibodies can be raised to non-conserved regions of DGAT1 and used to distinguish a particular DGAT1 from other proteins.

&null;0060&null; The polyclonal or monoclonal antibodies can be coupled to a detectable substance or reporter system. The term &null;coupled&null; is used to mean that the detectable substance is physically linked to the antibody. Suitable detectable substances include various enzymes, prosthetic groups, fluorescent materials, luminescent materials and radioactive materials. Examples of suitable enzymes include horseradish peroxidase, alkaline phosphatase, &null;-galactosidase, and acetylcholinesterase; examples of suitable prosthetic group complexes include streptavidin/biotin and avidin/biotin; examples of suitable fluorescent materials include umbelliferone, fluorescein, fluorescein isothiocyanate, rhodamine, dichlorotriazinylamine fluorescein, dansyl chloride and phycoerythrin; an example of a luminescent material includes luminol; and examples of suitable radioactive material include 125I; 131I, 35S and 3H. In a preferred embodiment, the reporter system allows quantitation of the amount of protein (antigen) present.

&null;0061&null; Such an antibody-linked reported system could be used in a method for determining whether a fluid or tissue sample of a bovine contains a deficient amount or an excessive amount of the relevant DGAT1 protein. Given a normal threshold concentration of such a protein, test kits can be developed.

&null;0062&null; The availability of such antibodies gives rise to further applications. One is a diagnostic kit for identifying cells comprising an antibody (such as a monoclonal antibody) which binds to a protein comprising an amino acid sequence shown in FIG. 4a and 4b; means for detecting the antibody when bound to the protein, unreacted protein or unbound antibody; means for determining the amount of protein in the sample; and means for comparing the amount of protein in the sample with a standard. In some embodiments of the invention, the detectability of the antibody which binds to a specific DGAT1 protein is activated by the binding (eg. change in fluorescence spectrum, loss of radioisotopic label). The diagnostic kit can also contain an instruction manual for use of the kit.

&null;0063&null; Antibody-based diagnostics are of course not the only possibility. A further diagnostic kit comprises a nucleotide probe complementary to the sequence, or an oligonucleotide fragment thereof, shown in FIG. 2a and 2b, for example, for hybridisation with mRNA from a sample of cells; means for detecting the nucleotide probe bound to mRNA in the sample with a standard. In a particular aspect, the kit of this aspect of the invention includes a probe having a nucleic acid molecule sufficiently complementary with a sequence identified in FIG. 2a and 2b, or its complement, so as to bind thereto under stringent conditions. &null;Stringent hybridisation conditions&null; takes on its common meaning to a person skilled in the art. Appropriate stringency conditions which promote nucleic acid hybridisation, for example, 6&null;sodium chloride/sodium citrate (SSC) at about 450&null; C. are known to those skilled in the art, including in Current Protocols in Molecular Biology, John Wiley & Sons, N.Y. (1989). Appropriate wash stringency depends on degree of homology and length of probe. If homology is 100%, a high temperature (650&null; C. to 750&null; C.) may be used. If homology is low, lower wash temperatures must be used. However, if the probe is very short (<100bp), lower temperatures must be used even with 100% homology. In general, one starts washing at low temperatures (370&null; C. to 400&null; C.), and raises the temperature by 3-5&null; C. intervals until background is low enough not to be a major factor in autoradiography. The diagnostic kit can also contain an instruction manual for use of the kit.

&null;0064&null; One of the major applications of the present invention is in the marker assisted selection of bovines having a polymorphism in the DGAT1 gene and which are associated with improved milk production traits. The invention therefore provides a diagnostic kit which can be used to determine the DGAT1 genotype of bovine genetic material, for example. One kit includes a set of primers used for amplifying the genetic material. A kit can contain a primer including a nucleotide sequence for amplifyg a region of the genetic material containing one of the polymorphisms described herein. Such a kit could also include a primer for amplifying the corresponding region of the normal DGAT1 gene, i.e. the sequence without polymorphisms. Usually, such a kit would also include another primer upstream or downstream of the region of interest complementary to a coding and/or non-coding portion of the gene. These primers are used to amplify the segment containing the mutation, i.e. polymorphism, of interest.

&null;0065&null; In particular, the invention is directed to the use of.the polymorphisms in the DGAT1 gene in the genotyping of cows and bulls as well as to cows and bulls selected by such genotyping which have one or more of said polymorphisms in the DGAT1 gene. Such bulls so selected are of valuable breeding stock and the invention is also directed to the semen produced by such selected bulls for breeding purposes. Cows so selected are also useful as breeding stock as are their offspring. In addition, such cows may produce valuable dairy herds as the milk produced by such cows is produced in greater volumes than equivalent non-selected cows, and/or has an altered composition in that it comprises less milkfat and more milk protein. Such milk and products made therefrom also form part of the invention. It is also noted that the milk from these selected cows will be valuable as the fat content is not only decreased but is also characterised by being softer. Without being bound by theory, it is thought that this increased fat softness is due to the fatty acid composition being such that there is less saturated and more unsaturated fat in the milk of selected cows. Thus it is anticipated that products made from such milk will have processing advantages, such as in the production of more spreadable butter, as well as having a health benefit on consumers, as generally unsaturated fats are considered to be more &null;healthy&null; than saturated fats. The protein composition of milk produced by such selected cows is also altered. In particular, such milk comprises an altered protein yield compared to milk for nonselected cows and the casein:whey ratio is also altered which makes such milk valuable for cheese production.

&null;0066&null; Thus, the present invention involves genotyping bovine, both cows and bulls, for the DGAT1 polymorphisms disclosed herein, selected cows and bulls so genotyped, milk and semen produced by the selected cows and bulls so genotyped, offspring produced by the selected bovine, including embryos and cells (including cell lines) useful for cloning said selected bovine.

&null;0067&null; The actual genotyping is carried out using primers that target specific polymorphisms as described herein and that could function as allele-specific oligonucleotides in conventional hybridisation, Taqman assays, OLA assays, etc. Alternatively, primers can be designed to permit genotyping by microsequencing.

&null;0068&null; One kit of primers can include first, second and third primers, (a), (b) and (c), respectively. Primer (a) is based on a region containing a DGAT1 mutation such as described above. Primer (b) encodes a region upstream or downstream of the region to be amplified by primer (a) so that genetic material containing the mutation is amplified, by PCR, for example, in the presence of the two primers. Primer (c) is based on the region corresponding to that on which primer (a) is based, but lacking the mutation. Thus, genetic material containing the non-mutated region will be amplified in the presence of primers (b) and (c). Genetic material homozygous for the DGAT1 gene will thus provide amplified products in the presence of primers (b) and (c). Genetic material homozygous for the mutated gene will thus provide amplified products in the presence of primers (a) and (b). Heterozygous genetic material will provide amplified products in both cases.

&null;0069&null; The present invention also contemplates the modulation of milk production and content in non-human animals by modulating the activity of the DGAT1 protein. In particular, this aspect of the invention includes a method of modulating milk production and/or milk content in a lactating bovine, the method comprising administering to the bovine an effective amount of a nucleic acid molecule substantially complementary to at least a portion of mRNA encoding the bovine DGAT1 variant proteins and being of sufficient length to sufficiently reduce expression of said DGAT1, i.e. by use of antisense nucleic acids.

&null;0070&null; Antisense nucleic acids or oligonucleotides (RNA or preferably DNA) can be used to inhibit DGAT1 production in a bovine if this is considered desirable e.g. in order to produce a bovine capable of improved milk production, i.e. increased milk volume and decreased milkfat content. Antisense oligonucleotides, typically 15 to 20 bases long, bind to the sense mRNA or pre mRNA region coding for the protein of interest, which can inhibit translation of the bound mRNA to protein. The cDNA sequence encoding DGAT1 can thus be used to design a series of oligonucleotides which together span a large portion, or even the entire cDNA sequence. These oligonucleotides can be tested to determine which provides the greatest inhibitory effect on the expression of the protein (Stewart 1996). The most suitable mRNA target sites include 5&null;- and 3&null;-untranslated regions as well as the initiation codon. Other regions might be found to be more or less effective.

&null;0071&null; Alternatively, an antisense nucleic acid or oligonucleotide may bind to DGAT1 coding sequences.

&null;0072&null; In yet another embodiment, the invention provides a method of modulating milk production and/or milk content in a lactating bovine, including administering to the bovine an effective amount of a nucleic acid molecule having ribozyme activity and a nucleotide sequence substantially complementary to at least a portion of MRNA encoding a bovine DGAT1 and being of sufficient length to bind selectively thereto to sufficiently reduce expression of said DGAT1.

&null;0073&null; Rather than reducing DGAT1 activity in the bovine by inhibiting gene expression at the nucleic acid level, activity of the relevant DGAT1 protein may be directly inhibited by binding to an agent, such as, for example, a suitable small molecule or a monoclonal antibody.

&null;0074&null; Thus, the invention also includes a method of inhibiting the activity of bovine DGAT1 in a lactating bovine so as to modulate milk production and/or milk solids content, comprising administering an effective amount of an antibody to the relevant DGAT1.

&null;0075&null; The invention still further includes a method of modulating milk production and/or milk solids content by raising an autoantibody to a bovine DGAT1 in the bovine. Raising the autoantibody can include administering a protein having DGAT1 activity to the bovine.

&null;0076&null; In still a further embodiment, nucleic acids which encode DGAT1 proteins can be used to generate transgenic animals. A transgenic animal (eg. a mouse) is an animal having cells that contain a transgene, which transgene is introduced into the animal or an ancestor of the animal at a prenatal, eg. an embryonic stage. A transgene is DNA which is integrated into the genome of a cell from which a transgenic animal develops. In one embodiment, a bovine cDNA, comprising the nucleotide sequence shown in FIG. 2b, or an appropriate variant or subsequence thereof, can be used to generate transgenic animals that contain cells which express the relevant DGAT1. Likewise, variants can be used to generate transgenic animals. &null;Knock out&null; animals can also be generated.

&null;0077&null; Methods for generating transgenic animals, particularly animals such as mice, have become conventional in the art are described, for example, in U.S. Pat. Nos. 4,736,866 and 4,870,009. In such methods, plasmids containing recombinant molecules are microinjected into mouse embryos. In particular, the plasmids can be microinjected into the male pronuclei of fertilised one-cell mouse eggs; the injected eggs transferred to pseudo-pregnant foster females; and the eggs in the foster females allowed to develop to term. (Hogan, 1986). Alternatively, an embryonal stem cell can be transfected with an expression vector comprising nucleic acid encoding a DGAT1 protein, and cells containing the nucleic acid can be used to form aggregation chimeras with embryos from a suitable recipient mouse strain. The chimeric embryos can then be implanted into a suitable pseudopregnant female mouse of the appropriate strain and the embryo brought to term. Progeny harbouring the transfected DNA in their germ cells can be used to breed uniformly transgenic mice.

&null;0078&null; Such animals could be used to determine whether a sequence related to an intact DGAT1 gene retains biological activity of the encoded DGAT1. Thus, for example, mice in which the murine DGAT1 gene has been knocked out and containing the nucleic acid sequence identified in FIG. 2b or fragment or variant thereof could be generated. The animals could be examined with reference to milk production and content.

&null;0079&null; The pattern and extent of expression of a recombinant molecule of the invention in a transgenic mouse is facilitated by fusing a reporter gene to the recombinant molecule such that both genes are co-transcribed to form a polycistronic MRNA. The reporter gene can be introduced into the recombinant molecule using conventional methods such as those described in Sambrook et al., (Sambrook, 1989). efficient expression of both cistrons of the polycistronic mRNA encoding the protein of the invention and the reporter protein can be achieved by inclusion of a known internal translational initiation sequence such as that present in polivirus mRNA. The reported gene should be under the control of the regulatory sequence of the recombinant molecule of the invention and the pattern and extent of expression of the gene encoding a protein of the invention can Accordingly be determined by assaying for the phenotype of the reporter gene. Preferably the reporter gene codes for a phenotype not displayed by the host cell and the phenotype can be assayed quantitatively Examples of suitable reporter genes include lacZ (&null;-galactosidase), neo (neomycin phosphotransferase), CAT (chloramphenicol acetyltransferase) dhfr (dihydrofolate reductase), aphIV (hygromycin phosphotransferase), lux (luciferase), uidA (&null;-glucuronidase), Preferably, the reporter gene is lacZ which codes for &null;-galactosidase. &null;-galactosidase can be assayed using the lactose analogue X-gal (5-bromo-4-chloro-3-indolyl-&null;-D-galactopyranoside) which is broken down by &null;-galactosidase to a product that is blue in colour.

&null;0080&null; Still further transgenic applications of the invention arise from knocking out the endogenous gene encoding DGAT1 in non-human mammals and replacing this with a bovine transgene, in order to obtain a desired effect. This is particularly true in cattle raised for milk production. For example, additional copies of the bovine gene encoding DGAT1 can be inserted as a transgene, or the endogenous gene associated with a high level expression promoter in a transgene. It may also prove advantageous to substitute a defective gene rather than delete the entire sequence of DNA encoding for a protein having DGAT1 activity. A method of producing a transgenic bovine or transgenic bovine embryo is described in U.S. Pat. No. 5,633,076, issued May 27, 1997, for example.

&null;0081&null; These transgenic animals of the invention can again be used to investigate the molecular basis of DGAT1 action. For example, it is expected that mutants in which one or more of the conserved cysteine residues has been deleted would have diminished activity in relation to a DGAT1 protein in which all such residues are retained. Further, deletion of a proteolytic cleavage site would likely result in a mutant lacking biological activity of DGAT1.

&null;0082&null; Transgenic animals of the invention can also be used to test substances for the ability to prevent, slow or enhance DGAT1 activity. A transgenic animal can be treated with the substance in parallel with an untreated control transgenic animal. Substances which could be tested in this way include proteins extracted from foods ingested by the animal, For example, proteins extracted from pastoral grasses and other fodder can be tested to determine their effect on DGAT1 activity, including to determine Whether breed-specific effects can be induced.

&null;0083&null; Thus, in further aspects, the invention provides transgenic non-human animals. These include by way of example only a transgenic bovine having a genome lacking a gene encoding a protein having biological activity of DGAT1 (or indeed any DGAT1 activity at all); a transgenic mouse having a genome containing a gene encoding a bovine protein having biological activity of any DGAT1; and a transgenic bovine having a gene, encoding a bovine protein having biological activity of a bovine DGAT1 and heterologous nucleotide sequence antisense to the gene. The transgenic bovine can include a gene encoding a nucleic acid sequence having ribozyme activity and in transcriptional association with the nucleotide sequence antisense to the gene.

&null;0084&null; The invention further provides a transgenic bovine having a genome which includes additional copies of a gene encoding a protein having biological activity of DGAT1 or copies of a gene encoding a protein having biological activity of DGAT1 under control of a high expression promoter.

&null;0085&null; These are but a selection of the applications of this invention. Others will be apparent to those persons skilled in this art and are in no way excluded. To the contrary, the invention extends to cover not only the specific teaching provided but also all variations and modifications which are within the skill and contemplation of the addressee.

&null;0086&null; The invention will now be defined by specific examples which are illustrative only and are not intended to limit the invention in any way.

&null;0087&null; Experimental

&null;0088&null; 1. Location of the Gene Responsible for the Observed QTL

&null;0089&null; Construction of a BAC contig Spanning the BULGE9-BULGE30 Interval.

&null;0090&null; In order to clone the gene(s) responsible for the observed QTL effect, a BAC contig spanning the corresponding marker interval was constructed. This was accomplished by screening a BAC library by filter hybridisation with the microsatellite markers available for proximal BTA14q, as well as with human cDNA clones mapping to the orthologous chromosome segment on the human RH transcript map: 8q23.3-ter (Riquet et al., (1999)). The ends of the isolated BACs were sequenced, sequence tagged sites (STS) developed from the corresponding sequences, and mapped onto a bovine x hamster whole genome radiation hybrid panel. This STS content mapping approach lead to the construction of the BAC contig shown in FIG. 1.

&null;0091&null; DGAT1 maps to the BULGE9-BULGE30 Interval and is a Strong Positional Candidate for the QTL.

&null;0092&null; A murine gene encoding a protein with Diacylglycerol-o-acyltransferase (DGAT1) activity was identified (Cases et aL, (1998)) and shown to completely inhibit lactation when knocked out in the mouse (Smith et al., (2000)). This gene was reported in the human to map to HSA8qter (Cases et al., (1998)), ie. in the region orthologous to that containing the bovine QTL. Screening the publicly available databases with the published murine and human DGAT1 cDNA sequences allowed identification of (i) a human BAC clone containing the human DGAT1 gene (AF205589), and (ii) three bovine Expressed Sequence Tags (AW446908; AW446985; AW652329) jointly covering approximately two thirds of the bovine gene. Aligning the human DGAT1 genomic sequences with the human and bovine cDNA sequences allowed the corresponding intron-exon boundaries to be identified. Primers were developed to PCR amplify a portion of the bovine DGAT1 gene. Screening the BACs composing the BULGE9-BULGE30 contig clearly indicated that the bovine DGAT1 gene was contained in a subset of the BACs allowing us to accurately position the DGAT1 gene in the contig of FIG. 1.

&null;0093&null; These results demonstrated that the map position of DGAT1 coincided with the most likely position of the chromosome 14 QTL as determined by linkage and linkage disequilibrium analyses. Knowing that the QTL primarily affects fat content, knowing the enzymatic activity of DGAT1 and the effect of a DGAT1 knock-out on lactation, this gene was considered to be a very strong positional candidate for the corresponding QTL.

&null;0094&null; Organisation of the Bovine DGAT1 Gene

&null;0095&null; The organisation of the bovine DGAT1 gene was determined by sequence analysis of one of the DGAT1 containing BACs. Primers were designed based on the available bovine, murine and human cDNA sequences which were either used for direct sequencing of the BAC clone or to generate PCR products corresponding to different parts of the bovine DGAT1 gene from this BAC which were then subjected to cycle-sequencing. All available sequences were then merged using the Phred/Phrap software (Ewing et al., (1998); Ewing & Green, (1998); Gordon et al., (1998)) to yield the consensus sequence shown in FIGS. 2a and b.

&null;0096&null; RT-PCR, 5&null; and 3&null; RACE experiments were performed on mRNA isolated from bovine mammary gland and the obtained PCR products subjected to cycle sequencing. Comparison of the genomic and cDNA sequences showed that the bovine DGAT1 gene spans 8.6 Kb and comprises 17 exons measuring 121.8 bp on average (range: 42 -436 bp) and allowed intron-exon boundaries to be identified (FIGS. 2a, 2b and 3). The cDNA sequence is also set out in SEQ ID NO: 4. While the first two introns are respectively 3.6 and 1.9 Kb long, the remaining 14 introns are only 92.4 bp long on average (range: 70 -215 bp). All introns conform to the GT-AG rule and are strictly conserved between human and bovine. The bovine DGAT1 gene is transcribed in a mRNA comprising >31 bp of 5&null; UTR sequence (FIG. 2a), 1470 bp coding for a protein of 489 amino-acids, and 275 bp of 3&null; UTR sequence including a canonical AATAAA polyadenylation signal. The human and bovine DGAT1 nucleotide (coding) and protein sequences are respectively 89.5% and 92.5% identical (FIGS. 2a, 2b, 4a and 4b). In addition, an alternative splicing variant is predicted in the bovine for exon VIII (FIG. 2b). The corresponding bovine cDNAs are predicted to encode proteins comprising respectively 489 and 467 (alternative splicing variant) amino-acid residues (FIGS. 4a and 4b).

&null;0097&null; The Predicted &null;Q&null; and &null;q&null; QTL Alleles Differ by a Non Conservative Lysine to Alanine Amino-acid Substitution in the DGAT1 Gene.

&null;0098&null; Assuming that DGAT1 is indeed the QTL, it is predicted that the identified &null;Q&null; and &null;q&null; QTL alleles will correspond to functionally distinct DGAT1 alleles, ie. will differ at one or more mutations causing these alleles to be functionally different. To test this hypothesis, the structure of the DGAT1 gene in individuals predicted to be of different QTL genotypes: &null;QQ&null;, &null;Qq&null; and &null;qq&null; was examined. More specifically, the DGAT 1 gene from:

&null;0099&null; (i) two sires with &null;HQ-D/hq&null; genotype as well as two of their &null;HQ-D/HQ-D&null; offspring, two of their &null;hq/hq&null; offspring and one &null;HQ-D/hq&null; offspring, and

&null;0100&null; (ii) one &null;HQ-NZ/hq&null; sire with one of its &null;HQ-NZ/HQ-NZ&null; offspring

&null;0101&null; was analysed wherein HQ-D corresponds to the Dutch Q haplotype and HQ-NZ corresponds to the New Zealand Q haplotype, and primer pairs were designed that allowed for the amplification from genomic DNA of (i) the coding portion of exon I, (ii) exon II, and (iii) the chromosome regions spanning exons III to XVII. The corresponding PCR products from the selected individuals were cycle-sequenced and the resulting sequences examined with the Polyphred software.

&null;0102&null; Additional sequencing analysis, as described above, on DNA from a range of breeds revealed additional polymorphisms included in Table 1 (see Methods section for breeds). Four such polymorphisms were investigated further:

&null;0103&null; (i) K232A: a substitution of a ApA by a GpC dinucleotide in exon VIII (respectively positions 694 and 695 counting from the start codon in the cDNA). The substitution of these two adjacent nucleotides results in a non conservative lysine (hydrophylic basic amino acid) to alanine (hydrophobic amino acid) substitution in the DGAT1 protein. The lysine residue affected by this polymorphism is conserved in the human and murine DGAT1 sequences. Together with the resulting change in the electrical charge of the protein, this strongly suggests that this amino-acid substitution is likely to result in a functional difference between the two corresponding alleles and to be at least partly responsible for the observed QTL effect.

&null;0104&null; (ii) Nt984&null;8(Base 7438 A-G): A A to G substitution in intron 12, eight base pairs downstream of exon XII. Following standard nomenclature, this polymorphism will be referred to as Nt984&null;8(A-G). This polymorphism cannot be predicted as such to modify the functionality of the corresponding alleles although an effect on the splicing mechanism cannot be excluded given its proximity to the intron-exon boundary.

&null;0105&null; (iii) Nt984&null;26(Base 7456 C-T): A C to T substitution in intron 12, 26 base pairs downstream of exon XII. Following standard nomenclature, this polymorphism will be referred to as Nt984&null;26(Base 7456 C-T). Again, this polymorphism cannot be predicted as such to modify the functionality of the corresponding alleles although an effect on the splicing mechanism cannot be excluded given its proximity to the intron-exon boundary.

&null;0106&null; (iv) Nt1470&null;85(Base 8402 C-T): A C to T substitution in the 3&null; UTR. Following standard nomenclature, this polymorphism will be referred to as Nt1470&null;85(Base 8402 C-T). Again, this polymorphism cannot be predicted as such to modify the functionality of the corresponding alleles although an effect on polyadenylation or mRNA stability cannot be excluded.

&null;0107&null; Conclusion

&null;0108&null; These four polymorphisms were shown to assort into three distinct SNP haplotypes referred to as sHQ-D, sHQ-NZ and shq because in the sequenced samples they coincided respectively with microsatellite haplotypes &null;HQ-D, &null;HQ-NZ and &null;hq. The base pair compositions of these three SNP haplotypes are shown in FIG. 3.

&null;0109&null; Because the sHQ-NZ and shq marker haplotypes share the G residue at the DGAT1 Nt984&null;8(Base 7438 A-G) site, the causality of this polymorphism in the determinism of the QTL could be excluded. For the three remaining polymorphic sites, however, the DGAT1 haplotypes associated with marker haplotypes sHED and sHQIz proved identical to each other while different from the shq DGAT1 haplotype. Either of these three polymorphisms could therefore be responsible for the observed QTL effect. The Nt984&null;26(Base 7456 C-T) and Nt1470&null;85(Base 8402 C-T) polymorphisms are a priori more likely to be neutral with respect to DGAT1 activity because of their respective location in an intron and the 3&null; UTR and likewise the other non coding or neutral polymorphism shown in Table 1. A direct effect of the K232A mutation on DGAT1 activity, however, is very plausible. Indeed, the corresponding lysine residue is conserved amongst all examined mammals (i.e. human, mouse, rat, pig, sheep, bison) demonstrating its functional importance (FIG. 5). The evolutionary conservation of this lysine residue also demonstrates that the K residue characterizing the sHQ-D and sHQ-NZ marker haplotypes is more than likely the ancestral state and that it is the A residue characterizing the shq haplotypes that corresponds to a more recently evolved state.

&null;0110&null; 2. Genotype Testing and Analysis I

&null;0111&null; This summarises the genotype testing and subsequent analysis of Holstein-Friesian animals sourced from New Zealand and Holland which were tested for the presence of the K232A polymorphism. Reference to allele &null;Q&null; corresponds to the K residue and allele &null;q&null; to the A residue (as shown in FIG. 3 and Table 1).

&null;0112&null; An oligonucleotide ligation assay (OLA) was developed as described in the method section below that allows for efficient genotyping of the four DGAT1 polymorphisms simultaneously. This OLA-test was used to genotype a previously described (Farnir et. al., 2000) &null;grand-daughter design&null; (i.e. series of paternal half-brother pedigrees) comprising 1,818 Dutch Holstein-Friesian sires as well as a &null;daughter design&null; (i.e. series of paternal half-sister pedigrees) comprising 529 New Zealand Holstein-Friesian cows selected according to phenotype as described below. The marker linkage phase for each individual was determined as described below.

&null;0113&null; FIG. 6 summarizes the frequency distribution of DGAT1 haplotypes encountered in the Dutch and New Zealand populations respectively. Four distinct SNP haplotypes were identified. Three of these correspond to the sHQ-D, sHQ-NZ and shq that were previously identified by sequencing, and jointly account for 99% and 98% of the chromosomes in the Dutch and New-Zealand populations respectively. A fourth minor haplotype was found accounting for the remaining 1% and 2% of the chromosomes. As this haplotype codes for a K residue at position 232 it was assumed to correspond to a fat increasing &null;Q&null; allele and was therefore referred to as sHQ-III (FIG. 3). The observation that the K residue is found on three distinct DGAT1 haplotypes while the A residue is found on a unique DGAT1 haplotype is in agreement with K being the more ancient state.

&null;0114&null; The sHQ-D and sHQ-NZ SNP haplotypes (coding for a K residue at position 232) appear to be in strong linkage disequilibrium (LD) with the flanking microsatellite markers BULGE09 and BULGE11, as they are in essence associated with unique microsatellite haplotypes corresponding respectively to the previously defined &null;HQ-D and &null;HQ-Z haplotypes (FIG. 6C&D). In sharp contrast, the shq haplotype (coding for an A residue at position 232) is nearly evenly distributed across more than ten distinct microsatellite haplotypes (FIG. 6B).

&null;0115&null; These observations are in excellent agreement with the results of the combined linkage and LD analysis (Fernier et. al., 2000). These studies indeed predicted (i) that in the Dutch population the vast majority (estimates ranging from 81% to 92%) of &null;Q&null; allele (&null;K) would reside on the &null;HQ-D microsateflite haplotype, (ii) that in the New Zealand population a large fraction (estimates ranging from 36% to 51%) of &null;Q&null; alleles would reside on haplotype &null;HQ-Nz (we now see that the remainder correspond mainly to the &null;HQ-D microsatellite haplotype) and (iii) that in both populations the &null;q&null; alleles (&null;A) would correspond to multiple marker haplotypes, corresponding to hq.

&null;0116&null; FIG. 7 illustrates the gain in LD signal that could be obtained in the Dutch HolsteinFriesian grand-daughter design when adding the DGAT1 polymorphisms to the previously available markers for proximal BTA14q and performing a joint linkage and LD multipoint analysis (Fernier et. al., 2000) using the sires &null;daughter yield deviations&null; (DYD (Van Raden and Wiggans, 1991) corresponding to half breeding values) for milk fat percentage as phenotype. It can be seen that the lod score attributable to LD essentially doubles (from 3.7 to 7.8), and maximizes exactly at the position of the DGAT1 gene. This result strongly supports the causal involvement of the DGAT1 gene in the QTL effect. The corresponding ML estimates of the &null;Q&null; to &null;q&null; allele substitution effect (&null;/2) (as defined in Falconer and Mackay, 1996), residual standard deviation (&null;), population frequency of the &null;Q&null; allele (fQ), number of generations to coalescence (g) and heterogeneity parameter (&null;) were respectively 0.11% (&null;/2), 0.06% (&null;), 0.20 (fQ), 5 (g) and 0.84 (&null;).

&null;0117&null; Using the same Dutch Holstein-Friesian population, the additive effect of the DGAT1 K232A polymorphism on milk yield and composition was examined. The sons DYDs for milk yield (kgs), protein yield (kgs), fat yield (kgs), protein percentage and fat percentage, were analysed using a mixed model including (i) a regression on the number of K alleles in the genotype (0, 1 or 2), and (ii) a random polygenic component estimated using an individual animal model and accounting for all known pedigree relationships. Table 2 below, reports the obtained results. It can be seen that the K232A mutation has an extremely significant effect on the five analysed dairy traits. The proportion of the trait variance explained by this polymorphism in this population ranges from 8% (protein yield) to 51% (fat percentage), corresponding to between 10% (protein yield) and 64% (fat percentage) of the genetic variance (&null;QTL &null;polygenic).

&null;0118&null; Note that the proportion of the variance explained by the full model (1-r2error) is of the order of 70% for the yield traits and 80% for the percentage traits, which is in agreement with the known reliabilities of the corresponding DYDs (Van Raden and Wiggans, 1991). An interesting feature of this QTL effect is that the &null;q&null; to &null;Q&null; substitution increases fat yield, while decreasing milk and protein yield, despite the strong overall positive correlation characterizing the three yield traits.

2

TABLE 2

Effect of the DGAT1 K232A mutation on sire's daughter

yield deviations (DYDs) for milk yield and composition.

Trait

&null;/2

r2QTL

p-valueQTL

r2polygenic

r2error

Milk yield

&null;&null;158 Kgs

0.18

5.00E&null;35

0.49

0.32

(Kgs)

Fat yield

&null;5.23 Kgs

0.15

1.57E&null;29

0.55

0.30

(Kgs)

Protein

&null;2.82 Kgs

0.08

1.70E&null;15

0.65

0.26

yield (Kgs)

Fat %

0.17%

0.51

4.33E&null;122

0.29

0.19

Protein %

0.04%

0.14

5.05E&null;28

0.66

0.20

(i) &null;/2: QTL allele substitution effect on DYD (half breeding value), corresponding in the mixed model to the regression coefficient on the number of K alleles in the DGAT1 K232A genotype, and to &null;/2, where &null; is defined according to ref. Falconer and Mackay, 1996.

(ii) r2QTL: proportion of the trait variance explained by the DGAT1 K232A polymorphism.

(iii) p-valueQTL: statistical significance of the DGAT1 K232A effect.

(iv) r2polygenic: proportion of the trait variance explained by the random, polygenic effect in the mixed model.

(v) r2error: proportion of the trait variance unexplained by the model.

&null;0119&null; The two previous analyses examined the effect of the DGAT1 polymorphism on estimated breeding values. By definition, this phenotype will only account for the additive component of the DGAT1 effect, and justifies the use of a regression on the number of K alleles in the mixed model. To evaluate the dominance relationship between the DGAT1 alleles, the effect of the K232A genotype on the lactation values (first yield deviations (Van Raden and Wiggans, 1991)) of the cows composing the New Zealand daughter design were analysed. This was achieved by using a mixed model including (i) a fixed effect corresponding to the K232A genotype, and (ii) a random polygenic component accounting for all known pedigree relationships (&null;animal model&null;). Very significant effects of K232A genotype on all examined yield and composition traits were found in this population as well (Table 3, below), accounting for between 1% (protein yield) and 31% (fat percentage) of the trait variance. The observed dominance deviations, d, corresponding to the difference between the genotypic value of the KA genotype and the midpoint between the AA and KK genotypic values (Falconer and Mackey, 1996) are shown in Table 3 below. Genotypic values of the heterozygous genotype are systematically in between alternate homozygotes. None of the d-values proved to be significantly different from zero, indicating an absence of dominance. Average K to A QTL allele substitution effects, &null; (Falconer and Mackey, 1996), were computed from the estimates of a and d-values, as well as the population frequencies of the K and A alleles (Table 3). The predicted substitution effects are in good agreement with those computed from the grand-daughter design: the K allele increases fat yield, fat % and protein %, while decreasing milk and protein yield. The absolute values of a estimated from the grand-daughter and daughter design are in perfect agreement for fat and protein %, while for the yield traits estimates are larger in the grand-daughter design when compared to the daughter design. The exact reasons for this are being explored. It could be due to the fact that the sire population in the grand-daughter design is not representative of the cow population in general, or to intrinsic differences between the Dutch and New-Zealand populations and/or environment.

3

TABLE 3

Effect of the DGAT1 K232A mutation on cows&null; lactation values

for milk yield and composition.

Trait

a

d

&null;

r2QTL

p-valQTL

r2polygenic

r2error

Milk yield (Kgs)

&null;&null;144 Kgs

&null;&null;&null;42 Kgs

&null;&null;161 Kgs

0.03

1.05E&null;8

0.54

0.43

Fat yield (Kgs)

&null;7.82 Kgs

&null;0.89 Kgs

&null;7.46 Kgs

0.09

1.77E&null;20

0.46

0.45

Protein yield (Kgs)

&null;2.34 Kgs

&null;0.76 Kgs

&null;2.64 Kgs

0.01

4.35E&null;2

0.37

0.42

Fat %

0.41%

0.03%

0.42%

0.31

&null;2.5E&null;108

0.49

0.20

Protein %

0.08%

0.03%

0.08%

0.04

1.60E&null;20

0.72

0.24

(i) a: half the difference between the genotypic values of the KK and AA genotypes (Falconer and Mackey, 1996).

(ii) d: dominance deviation (Falconer and Mackey, 1996): deviation of the KA genotypic value from the midpoint between the AA and KK genotypic values; none of these proved to be significantly different from zero.

(iii) &null;: average K to A substitution effect, computed as &null;a &null; d(q &null; p)&null; (Falconer and Mackey, 1996), where q is the allelic frequency of K (&null; 0.7) and p of A (&null; 0.3)

(iv) r2QTL: proportion of the trait variance explained by the DGAT1 K232A polymorphism.

(v) p-valQTL: statistical significance of the DGAT1 K232A effect.

(vi) r2polygenic: proportion of the trait variance explained by the random, polygenic effect in the mixed model.

(vii) r2error: proportion of the trait variance unexplained by the model.

&null;0120&null; Pedigree material and phenotypes. The pedigree material used for the association studies comprised a &null;grand-daughter design&null; (Weller et. al., 1990) counting 1,818 Holstein-Friesian bulls sampled in the Netherlands, as well as a &null;daughter-design&null; (Weller et. al., 1990) counting 529 Holstein-Friesian cows sampled in New Zealand. The phenotypes of the sires were &null;daughter yield deviations&null; (DYD: unregressed weighted averages of the daughter's lactation performances adjusted for 1 systematic environmental effects and breeding values of the daughter's dams and expressed as deviations from the population mean (Van Raden and Wiggans, 1991)) obtained directly from CR-Delta (Arnhem&null;The Netherlands). The phenotypes of the cows were &null;lactation values&null; (first lactation yield deviations (YD), i.e. weighted average lactation performances expressed as deviations from the population mean, adjusted for management group, permanent environmental effects and herd-sire interaction effects (Van Raden and Wiggans, 1991)) obtained directly from Livestock Improvement Corporation (Hamilton&null;New Zealand).

&null;0121&null; Combined linkage and linkage disequilibrium analysis and association studies. The maximum likelihood procedure for combined linkage and linkage disequilibrium analysis is described in detail in Farnir, 2000. The association study in the grand-daughter design was performed using the following model:

Yi&null;&null;30 &null;xi&null;ai&null;ei

&null;0122&null; where yi is the DYD of son i, &null; is the overall population mean, &null; is a fixed regression coefficient estimating the A to K allele substitution effect, xi is an indicator variable corresponding to the number of K alleles in the K232A genotype, ai is a random polygenic component accounting for all known pedigree relationships (&null;animal model&null; Lynch and Walsh, 1997) and ei is a random residual. The association study in the daughter design was performed using the model:

yi&null;&null;&null;gi&null;ai&null;ei

&null;0123&null; where y1 is the lactation value of cow i, gi is a fixed effect corresponding to the DGAT1 genotype (KK, KA, or AA), ai is a random polygenic component accounting for all known pedigree relationships (&null;animal model&null; Lynch and Walsh, 1997) and ei is a random residual. In both instances, maximum likelihood solutions for &null;, gi, &null;i, ei, &null;2a, &null;2e were obtained using the MTDFREML program (Boldman et al, 1997).

&null;0124&null; 3. G Notyp T Sting and Analysis II

&null;0125&null; This summarises the genotype testing and subsequent analysis of Holstein-Friesian, Jersey and Ayrshire animals in a separate population from those presented in genotype testing and analysis I, above.

&null;0126&null; Progeny Tested Sires

&null;0127&null; Each year Livestock Improvement Corporation (New Zealand) progeny test some 200-300 bulls per year. This entails the bulls being genetically evaluated on the basis of 50-85 daughters per sire. The sires are evaluated for milk fat, milk protein, milk volume and 20 non-production traits. Semen has been retained from all progeny tested sires since the early 1970s. DNA was extracted from the semen and genotyped for the K232A DGAT1 polymorphism using the 7900 Taqman system (see Methods section below). Statistical analysis was undertaken on this dataset using Restricted Maximum Likelihood (REML) and the average information algorithm (Johnson and Thompson, 1995). The linear model included the fixed effects of DGAT1 (3 classes; 0, 1 and 2 copies of the Q allele i.e. the K residue) and a covariate corresponding to the proportion of overseas genetics. The random effect was animal with a relationship matrix based on all known relationships. Daughter yield deviations (DYDs), weighted averages of a sire's daughter's lactation performances expressed as deviations from the population mean (van Raden and Wiggans 1991) were used as the phenotypic measurement. The phenotypes were weighted by a weighting factor based on the variance of the DYD for a son being:

1

Var

&null;

&null;

&null;

DYD

=

[

1

+

(

n

-

1

)

&null;

1

4

&null;

h

2

n

]

&null;

&null;

p

2

&null;0128&null; where Var DYD is the variance of son's DYD; n is the number of daughters contributing to the DYD; h2 is the heritability, which was taken as 0.35 for yield traits.

&null;0129&null; The dataset was analysed separately for the 3 major breeds; Holstein-Friesian, Jersey and Ayrshire.

&null;0130&null; Seventeen hundred and thirteen Holstein-Friesian sires were included in the analysis. The effect of the DGAT1 polymorphism was extremely significant for the three milk production traits (Table 4). With each additional Q allele the level of milk fat production increases by approximately 6 kg per lactation, milk protein production decreases by approximately 2.5 kg per lactation and milk volume decreases by approximately 125 litres per lactation.

4

TABLE 4

Effect of the DGAT1 polymorphism on milk production

in the Holstein-Friesian bull population

(kilograms per lactation).

Fat

Protein

Milk

qq

0

0

0

Qq

6.86

&null;2.13

&null;128

QQ

11.83

&null;4.80

&null;266

st. error

0.87

0.68

24

&null;0131&null; The effects for the Jersey and Ayrshire breeds were less significant than those of the Holstein-Friesian breed but were consistent in direction of effects.

&null;0132&null; Daughters for Milk Components

&null;0133&null; Data collection was integrated with LIC's herd testing service using a sample of 102 herds involved in LIC's Sire Proving Scheme (SPS) in 1995. In addition to milk volume from herd testing, the concentrations of fat, crude protein (total nitrogen), casein, whey and lactose were determined. The data was collected from over 3,000 cows born in 1996 and first calving in the 1998 spring season, these being predominantly the daughters of approximately 220 SPS bulls. The milk characteristics were measured at three herd tests on each cow, with each herd having a herd test in each of the Sept/Oct, Nov/Dec and Jan/Feb periods. The Milkoscan FT120, which employs Fourier transform infrared spectrophotometry with enhanced milk calibrations (Foss Electric Application Note Nos. 95, P/N 492280 and 102, P/N 578377), was used to determine the milk component concentrations.

&null;0134&null; Nine hundred and twelve daughters were genotyped for the DGAT1 polymorphism using the OLA system. Analysis was undertaken using SAS (Statistics, Version 5, 1985) fitting a general linear model. The model included sire and maternal grandsire as fixed effects, DGAT1 polymorphism (3 classes; 0, 1 and 2 copies of the Q allele), covariates including 16ths of Holstein-Friesian, Jersey, Ayrshire and other, proportion of overseas genetics within the Holstein-Friesian, Jersey and Ayrshire breeds. Yield deviations that were pre-adjusted for herd, stage of lactation among other fixed effects were used (Johnson et al 2000).

&null;0135&null; The DGAT1 polymorphism is statistically significant for Lactose, casein, beta-casein and whey yield and also for casein and beta-casein percent as outlined in Table 5.

5

TABLE 5

Effect of the DGAT1 polymorphism on milk components.

Trait

qq

Qq

QQ

p-value

Lactose yield*

48

23

0

<0.0001

Casein yield*

11.0

5.8

0

0.01

Casein %

&null;0.13

&null;0.06

0

<0.0001

Whey yield*

6.86

2.31

0

<0.0001

&null;-casein yield*

3.98

2.19

0

0.05

&null;-casein %

&null;0.43

&null;0.23

0

0.0001

*Units &null; g/day for lactose, casein and whey yield and g/litre for &null;-casein yield

&null;0136&null; Daughters for Solid Fat Content

&null;0137&null; Six hundred and ninety-two daughters were phenotyped for solid fat content. Solid fat content of the milkfat is a characteristic which has a major influence on the functionality of milkfat products, and in particular has a significant effect on the hardness of butter (MacGibbon & McLennan, 1987). The solid fat content at 10&null; C. (SFC 10) was used for comparison of the properties of the milkfat as it relates well to the sectility hardness measurement of butter, a major functional property. Thus the performance of milkfat products may be predicted from the characteristics of the milk produced. The solid fat content (SFC) of the extracted fat was determined by pulsed nuclear magnetic resonance (NMR) and expressed as percentage solid fat (MacGibbon & McLennan, 1987). As the milkfat was melted to remove any thermal history, prior to recrystallization under standard conditions, the SFC simply reflects the chemical composition of the milkfat.

&null;0138&null; The 692 daughters were a subset of the 912 daughters that were phenotyped and genotyped for the results presented in Table 2. The solid fat content measures were collected over 2 lactations. Breeding values were calculated using an animal model similar to that of Johnson et al 2000.

&null;0139&null; The same statistical model was fitted for solid fat content as was for the milk component analysis. The DGAT1 polymorphism has a statistically significant effect (p-value <0.0001) on solid fat content, increasing it by approximately by 1% for each addition of the Q allele.

&null;0140&null; This effect was further confirmed in 50 daughters (predominantly Holstein-Friesian) that were farmed at one location and measured for SFC on the same day. The estimated effect for of the DGAT1 polymorphism on SFC was to increase it by approximately 2% per addition of each Q allele. This finding was significant at the five percent threshold level.

&null;0141&null; The genetic standard deviation for SFC is 2.25 (D Johnson personal communication) and thus the effect of DGAT1 is approximately 0.5 of a genetic standard deviation. 4. Relative Transcript Levels of the Splice Variant Real time PCR experiments were conducted using reverse transcribed mRNA isolated from lactating bovine mammary gland(s) (see experimental methods). These experiments revealed that the alternatively spliced transcript as shown on FIG. 2b, was approximately 100 fold less abundant than the full length transcript.

&null;0142&null; METHODS SECTION

&null;0143&null; In order to identify other polymorphisms within the bovine DGAT1 gene, DNA was isolated from sperm, PCR amplified and then using primers designed from the sequence shown in FIGS. 2a and 2b and/or the cDNA sequence (SEQ ID NO: 4) direct sequenced on an ABI 3100. The breeds examined were:

&null;0144&null; Ayrshire, Angler, Belgian Blue, Blond D'Aquitaine, Brown Swiss, Charolais, Red Devon, Devon, Dexter, Friesian, Guernsey, Belted Galloway, Gelbvieh, Hereford, Jersey, Limousin, Longhorn, Maine Anjou, MRI (Meuse-rhine-yssel), Murray Grey, Piedmontese, Romangola, Sahiwal, Santa Gertrudis, Scottish Highland, Shorthorn, South Devon, Sussex, Swedish Red, Simmental, Wagyu, Welsh Black, Angus, and Zebu.

&null;0145&null; All the polymorphisms discovered are listed in Table 1, above.

&null;0146&null; The majority of the primers are also listed in FIG. 2b or contained in the cDNA sequence (SEQ ID NO: 4).

&null;0147&null; Experimental Method for the OLA Analysis of Four SNP's in DGAT1

&null;0148&null; PCR Amplification of the Regions Containing the Polymorphisms

&null;0149&null; Protocol for the PCR amplification of exon VIII, intron XII and 3&null; UTR, the regions containing the four polymorphisms that were initially described in the DGAT1 gene.

6

Component

For 1 sample

Final concentration

HotStar Qiagen Buffer (10 x)

&null;1.5 &null;l

&null;0.7 &null;M

Primer 17F at 100 &null;M

0.07 &null;l

&null;0.7 &null;M

18R at 10 &null;M

0.07 &null;l

&null;0.5 &null;M

Primer 6F at 100 &null;M

0.05 &null;l

&null;0.5 &null;M

AW 446985dn1 at 100 &null;M

0.05 &null;l

&null;0.5 &null;M

Primer InsUp1

0.05 &null;l

&null;0.5 &null;M

14R2

0.05 &null;l

10%

DMSO

&null;&null;1 &null;l

&null;300 &null;M

dNTP 10 mM

&null;0.3 &null;l

&null;0.1 U/&null;l

HotStarQiagen Taq

&null;0.2 &null;l

(CatNr 203205: 5 U/&null;l)

H2O

1.66 &null;l

DNA (5 ng/&null;l)

&null;&null;5 &null;l

Total

&null;&null;10 &null;l

&null;0150&null; Primer sequences are given in the following table as well as the genomic region targeted by them.

7

SNP targeted

Primer name

Primer sequence

5&null;&null;base position

Exon VIII SNP (DG 1)

17F

CCTGAGCTTGCCTCTCCCACAGT

6579

18R

CCAGGAGTCGCCGCAGCAGGAAG

7058

Exon XII SNPs

6F

CCGGCCATCCAGAACTCCATGAAG

7280

(DG 2 and DG3)

AW446985 dn1

TAGAACTCGCGGTCTCCAAAC

7605

InsUp1

TGGCTGTCACTCATCATCGGGCA

8222

3&null;UTR SNP (DG4)

14R2

TTGCACAGCACTTTATTGACACA

8566

&null;0151&null; PCR amplification was performed on MJ PTC100 or PTCT200 cyclers using the following steps:

8

Step

Temperature

Time

Comment

1&null; Activation

94&null; C.

12 minutes

One times

of the enzyme

2&null; Denaturation

92&null; C.

&null;1 minute

Repeat step

3&null; Hybridisation

60&null; C.

&null;1 minute 30 seconds

2 to 4, 35

4&null; Elongation

72&null; C.

&null;1 minute 30 seconds

5&null; Inactivation

99&null; C.

45 minutes

of the enzyme

&null;0152&null; Oligonucleotide Ligation Assay (OLA)

&null;0153&null; The oligonucleotides used in the OLA multiplex reaction are given in the table below.

&null;0154&null; The detection of each mutation relies on the use of two fluorescent-labelled oligonucleotide (SNPx13 FAM and SNPx13 HEX) and one common 3&null; and 5&null; phosphorylated, non-labelled oligonucleotide (SNPx&null;2P)

9

Size of the

5&null;&null;base

Number of spacer

ligation

Locus

Oligo

Sequence

position

phosphoramidites

producta

DG1

SNP1_FAM

AGC TTT GGC AGG TAA GGC

6813

SNP1_HEX

AGC TTT GGC AGG TAA GAA

6813

32

SNP1_2P

GGC CAA CGG GGG AG

6831

0

DG2

SNP2_FAM

GCT GGC GGT GAG TGA

7424

SNP2_HEX

GCT GGC GGT GAG TGG

7424

39

SNP2_2P

CCT GCT GGG TGG GGA

7439

DG3

SNP3_FAM

GCT GGG TGG GGA CGC

7442

SNP3_HEX

GCT GGG TGG GGA CGT

7442

29

SNP3_P

GTG GGG GCG GGT GG

7457

0

DG4

SNP4_FAM

TGC CCC AAC CTG GGT

8388

SNP4_HEX

TGC CCC AAC CTG GGC

8388

36

SNP4_2P

GCA GCA GGA GGA GGC

8403

2

aThe size of the ligation products is the sum of the number of nucleotides of the two ligated oligonucleotides plus 3 bases equivalents per spacer phosphorazmdites molecule, present at the 5&null; end of the common oligonucleotide.

&null;0155&null; For each SNP a mixture of the three oligonucleotides was prepared first, following the dilution guidelines in the table below.

10

Oligonucleotide

Final

SNP mixture

to mix

Quantity

concentration

DG1 (oligo. mixture)

SNP1_FAM 10 &null;M

10 &null;l

1 &null;M

SNP1_HEX 10 &null;M

20 &null;l

2 &null;M

SNP1_2P 10 &null;M

20 &null;l

2 &null;M

H2O

50 &null;l

DG2 (oligo. mixture)

SNP2_FAM 10 &null;M

10 &null;l

1 &null;M

SNP2_HEX 10 &null;M

20 &null;l

2 &null;M

SNP2_2P 10 &null;M

20 &null;l

2 &null;M

H2O

50 &null;l

DG3 (oligo. mixture)

SNP3_FAM 10 &null;M

10 &null;l

1 &null;M

SNP3_HEX 10 &null;M

20 &null;l

2 &null;M

SNP3_2P 10 &null;M

20 &null;l

2 &null;M

H2O

50 &null;l

DG4 (oligo. mixture)

SNP4_FAM 10 &null;M

10 &null;l

1 &null;M

SNP4_HEX 10 &null;M

30 &null;l

3 &null;M

SNP4_2P 10 &null;M

20 &null;l

2 &null;M

H2O

40 &null;l

&null;0156&null; The ligation reaction for one sample was performed as follow:

11

Quantity

Component

persample

DG1 oligonucleotide mixturea (35, 70 and 70 nM)

&null;0.7 &null;l

DG2 oligonucleotide mixturea (12.5, 25 and 25 nM)

0.25 &null;l

DG3 oligonucleotide mixturea (12.5, 25 and 25 nM)

0.25 &null;l

DG4 oligonucleotide mixturea (12.5, 37.5 and 25 nM)

0.25 &null;l

DMSO

&null;&null;2 &null;l

Incubation buffer of the Tsc DNA ligase (Roche, Cat Nr

&null;&null;2 &null;l

1 939 807 or 1 939 815)

Tsc DNA ligase

&null;&null;1 &null;l

H2O

8.55 &null;l

Multiplex PCR (see above)

&null;&null;5 &null;l

Total

&null;&null;20 &null;l

aThe final concentration of the oligonucleotides in the ligation reaction is given between parenthesis (SNPx_FAM, SNPx_HEX and SNPx_2P respectively)

&null;0157&null; The sample was submitted to the following temperature cycling program in a MJ PTC100 or PTC 200 PCR machine.

12

Step

Temperature

Time

Comment

1&null; Initial

98&null; C.

&null;2 minutes

One times

denaturation step

2&null; Denaturation

94&null; C.

30 seconds

Repeat step 2 to 3, 30

3&null; Hybridisation

45&null; C.

&null;3 minutes

times

and ligation

5&null; Inactivation

99&null; C.

45 minutes

of the enzyme

&null;0158&null; Following the LCR, 20 &null;l of H2O was added to the ligation reaction. To 0.5 &null;l of the diluted ligation reaction, either 2 &null;l of loading buffer was added, or 2 &null;l loading buffer containing TAMRA350 internal line size standard.

&null;0159&null; The loading buffer was composed as follows: 1 part of blue dextran (50 mg/ml)/EDTA (25 mM) and 6 parts of formamide

&null;0160&null; The TAMRA350 containing loading buffer was composed as follows: 3 parts TAMRA350 (Applied Biosystems 401736; 8 nM), 10 parts of Blue dextran (50 mg/ml)/EDTA (25 mM) and 60 parts of formamide.

&null;0161&null; TAMRA containing samples was placed alternately with TAMRA free samples when loaded onto the sequencing gel, in order to ease the identification of the lanes on the gel image.

&null;0162&null; The samples may require further dilution in order to avoid a too intense fluorescent signal on the sequencer. It is also very likely that from one primer batch to another, oligonucleotides concentrations will need adjustment.

&null;0163&null; The samples were denatured for 5 minutes at 95&null; C. before loading. The samples were then loaded onto a 6% denaturing acrylamide gel on sequencer ABI 373 or a 4% gel on sequencer ABI 377.

&null;0164&null; In addition to the OLA assays referred to above, genotyping of the DGAT1 polymorphism was carried out by utilizing two different techniques for detection of PCR products.

13

Gel-based Genotyping Assay

Primer sequences 5&null; to 3&null;,

genomic sequence position in brackets:

DGAT1 21:

GTAGCTTTGGCAGGTAAGAA

(6811)

DGAT1 22:

GGGGCGAAGAGGAAGTAGTA

(6984)

DGAT1 23:

TGGCCCTGATGGTCTACACC

(6613)

DGAT1 24B:

GGGCAGCTCCCCCGTTGGCCGC

(6850)

&null;0165&null; The final reaction conditions were 1&null;Gold PCR buffer, 2.5mM MgCl2 (Applied Biosystems), 200 &null;M each dNTP (Roche), 600 nM DGAT1 21 and DGAT1 22, 400 nM DGAT1 23 and DGAT1 24B (1Invitrogen), 10% dimethylsulphoxide (Sigma), 3&null;l DNA template and 2.5 units AmpliTaq Gold DNA polymerase (Applied Biosystems) in a total volume of 50 &null;l.

&null;0166&null; Cycling conditions were a 94&null; C. initial denaturation for 5 minutes, then 35 cycles of denaturation at 94&null; C. for 30 seconds, annealing at 56&null; C. for 30 seconds, extension at 72&null; C. for 20 seconds followed by one extension cycle of 72&null; C. for 2 minutes.

&null;0167&null; Primer positions around polymorphism (in bold) on genomic sequence from 6587 to 6986.

14

&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;DGAT123&null;&null;&null;&null;TGGC CCTGATGGTC TACACC

TGCCTCTCCC ACAGTGGGCT CCGTGCTGGC CCTGATGGTC {right arrow over (TACA)}CCATCC

TCTTCCTCAA GCTGTTCTCC TACCGGGACG TCAACCTCTG GTGCCGAGAG

CGCAGGGCTG GGGCCAAGGC CAAGGCTGGT GAGGGCTGCC TCGGGCTGGG

GCCACTGGGC TGCCACTTGC CTCGGGACCG GCAGGGGCTC GGCTCACCCC

&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;DGAT1 21 GTAGCT TTGGCAGGTA AGAA

CGACCCGCCC CCTGCCGCTT GCTCGTAGCT TTGGCAGGTA {right arrow over (AGAA)}GGCCAA

&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;CGCCGGTT

CGGGGGAGCT GCCCAGCGCA CCGTGAGCTA CCCCGACAAC CTGACCTACC

GCCCCCTCGA CGGG DGAT1 24B

GCGGTGAGGA TCCTGCCGGG GGCTGGGGGG ACTGCCCGGC GGCCTGGCCT

GCTAGCCCCG CCCTCCCTTC CAGATCTCTA CTACTTCCTC TTCGCCCCCA

&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;AT GATGAAGGAG AAGCGGGG DGAT1 22

&null;0168&null; The Q allele has polymorphic sequence AA and is detected by the DGAT1 21&null;22 primers, producing a band of 174 bp. The q allele has polymorphic sequence GC and is detected by the DGAT123&null;24 primers, producing a band of 238 bp.

&null;0169&null; The primers DGAT123 and DGAT122 also successfully PCR the DGAT1 gene producing a product of 372 bp in all reactions. Therefore, a QQ homozygote would have bands at 372 bp and 174 bp, a qq homozygote would have bands at 372 bp and 238 bp and a Qq heterozygote would have all 3 bands at 372 bp, 238 bp and 174 bp.

&null;0170&null; 18 &null;l of PCR product was separated on a 1.2% agarose TAE gel, stained with ethidium bromide and scored independently by two investigators on the basis of the number and size of bands present.

15

TaqMan Allelic Discrimination Genotyping Assay

Primer sequences 5&null;&null;to 3&null;,

genomic sequence position in brackets:

DGAT1forAD:

CGTTGGCCTTCTTA

(6651)

ReverseNZ:

CCGCGGTAGGTCAGGTTGTC

(6890)

Probe sequences 5&null;&null;to 3&null;,

genomic sequence position in brackets:

ForAA (FAM):

CGTTGGCCTTCTTA

(6838)

DGAT1ADGC (VIC):

TTGGCCGCCTTACC

(6836)

&null;0171&null; Both probes use MGB (minor groove binder) as a non-fluorescent quencher.

&null;0172&null; The final reaction conditions are 1&null; Universal PCR Mastermix (Applied Biosystems), 500 nM each primer (Invitrogen), 70 nM ForAA (FAM) probe, 300 nM DGAT1ADGC (VIC) probe (Applied Biosystems) and 2 &null;l of a 1/20 dilution of DNA template in a total volume of 10 &null;l.

&null;0173&null; Cycling conditions were 50&null; C. for 2 minutes, 95&null; C. initial denaturation for 10 minutes, then 37 cycles of denaturation at 94&null; C. for 15 seconds, annealing and extension 60&null; C. for 1 minute.

&null;0174&null; Primer positions around polymorphism (in bold) on genomic sequence from 6587 to 6986.

16

TGCCTCTCCC ACAGTGGGCT CCGTGCTGGC CCTGATGGTC TACACCATCC

&null;&null;&null;&null;DGAT1forAD TTCTCC TACCGGGACG TCAA

TCTTCCTCAA GCTGTTCTCC TACCGGGACG TCAA{right arrow over (CCT)}CTG GTGCCGAGAG

CGCAGGGCTG GGGCCAAGGC CAAGGCTGGT GAGGGCTGCC TCGGGCTGGG

GCCACTGGGC TGCCACTTGC CTCGGGACCG GCAGGGGCTC GGCTCACCCC

&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;F rAA(FAM)A T TCTTCCGGTTGC

CGACCCGCCC CCTGCCGCTT GCTCGTAGCT TTGGCAGGTA AGAAGGCCAA

&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;DGAT1ADGC (VIC)CCAT TCCGCCGGTT

CGGGGGAGCT GCCCAGCGCA CCGTGAGCTA CCCCGACAAC CTGACCTACC

&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;CTGTTG GACTGGATGG

GCGGTGAGGA TCCTGCCGGG GGCTGGGGGG ACTGCCCGGC GGCCTGGCCT

CGCC ReverseNZ

GCTAGCCCCG CCCTCCCTTC CAGATCTCTA CTACTTCCTC TTCGCCCCCA

&null;0175&null; A 240 bp product is produced in this reaction. When the Q allele (AA) is present the FAM-labelled probe binds and fluoresces at 518 nm. When the q allele (GC) is present the VIC-labelled probe binds and fluoresces at 554 nm. After cycling is complete, the plate is scanned on the ABI7900 Sequence Detection System, the fluorescence from each well detected, and a scattergraph is drawn. The scattergraph separates out into 3 clumps with Q homozygotes in the upper left hand corner, q homozygotes in the lower right hand corner and Qq heterozygotes in between. Each clump is circled and the software automatically determines the genotype for each sample. On each plate there are controls with 8 wells each of known Q homozygotes, q homozygotes, Qq heterzygotes and no template controls.

&null;0176&null; Splice Variant Gene Expression

&null;0177&null; To determine the relative gene expression of the splice variants created by insertion/deletion of 66 bp around the polymorphic site by alternate exon usage, RNA was extracted from mammary tissue and reverse transcribed using oligodT primer using a first strand cDNA synthesis kit (Invitrogen). Real time PCR to determine relative quantities of each variant was then carried out.

17

Primer sequences 5&null;&null;to 3&null;,

genornic sequence position in brackets:

DGAT1forRT66:

TCTCCTACCGGGACGTCAAC

(6652)

DGAT1revRT66:

GAGATCGCGGTAGGTCAGGTT

(6964)

DGAT1forRTless66:

GCTGCTTTGGCAGATCTCTACTACTT

(6711)

DGAT1revRTless66:

AAGCGCTTTCGGATGCG

(7038)

Probe sequences 5&null;&null;to 3&null;,

genomic sequence position in brackets:

DGAT1with66 (FAM):

CCGTGAGCTACCC

(6857)

DGAT1less66 (VIC):

CTTCGCCCCCACCCT

(6976)

&null;0178&null; Both probes use MGB (minor groove binder) as a non-fluorescent quencher.

&null;0179&null; Final reaction conditions were 1&null;Universal PCR Mastermix (Applied Biosystems), 60 nM each primer (Invitrogen), 60 nM each probe (Applied Biosystems) and 1 &null;l of template cDNA in a total volume of 10 &null;l.

&null;0180&null; Cycling conditions were 50&null; C. for 2 minutes, 95&null; C. initial denaturation for 10 minutes, then 37 cycles of denaturation at 94&null; C. for 15 seconds, annealing and extension 60&null; C. for 1 minute.

&null;0181&null; Primer positions around 66 bp insertion (in italics) on cDNA sequence. The start of the cDNA sequence is equivalent to position 6479 on the genomic sequence, with the last base of the cDNA equivalent to position 7428 of the genomic sequence.

18

CCGTGGCCTT TCTCCTCGAG TCTATCACTC CAGTGGGCTC CGTGCTGGCC

&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;DGAT1forRT66&null;TCTCCT&null;ACCGGGACGT

CTGATGGTCT ACACCATCCT CTTCCTCAAG CTGTTCTCCT ACCGGGACGT

CAAC&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;DGAT1forRTless66&null;GCTGCTT

CAACCTCTGG TGCCGAGAGC GCAGGGCTGG GGCCAAGGCC AAGGCTGCTT

TGGCAG&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;DGAT1with66(FAM) C CGTGAGCTAC

TGGCAGGTAA GAAGGCCAAC GGGGGAGCTG CCCAGCGCAC CGTGAGCTAC

CC&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;ATCTCTAC TACTT&null;

CCCGACAACC TGACCTACCG CGATCTCTAC TACTTCCTCT TCGCCCCCAC

&null;&null;&null;TTGG ACTGGATGGC GCTAGAGDGAT1revRT66 CT TCGCCCCCAC

CCTGTGCTAC GAGCTCAACT TCCCCCGCTC CCCCCGCATC CGAAAGCGCT

CCT DGAT1less66 (VIC)&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;&null;GCGTAG GCTTTCGCGA

TCCTGCTGCG GCGACTCCTG GAGATGCTGT TCCTCACCCA GCTCCAGGTG

A DGAT1revRTless66

GGGCTGATCC AGCAGTGGAT GGTCCCGGCC ATCCAGAACT CCATGAAGCC

CTTCAAGGAC ATGGACTACT CCCGCATCGT GGAGCGCCTC CTGAAGCTGG

&null;0182&null; This reaction detects the presence of the insertion splice variant by creating a 145 bp product which binds the FAM probe only. The deletion splice variant is detected by a 92 bp product that binds the VIC probe only.

&null;0183&null; The cDNA for each alternate splice variant was cloned into pGemT (Promega). A dilution series of the same, known amount, of each variant plasmid DNA was used to create a standard curve that established the linearity of the PCR reaction over a range of DNA concentrations. The threshold cycle number of the sample variants was converted back to a DNA amount by linear regression and the amounts of each variant present compared.

&null;0184&null; The presence of an alternate spice variant raises the possibility of an alternate function that is at this stage unknown.

&null;0185&null; It will be appreciated that it is not intended to limit the invention to the above examples only, many variations, which may readily occur to a person skilled in the art, being possible without departing from the scope thereof as defined in the accompanying claims.

&null;0186&null; Industrial Application

&null;0187&null; The present invention is directed to a method of genotyping bovine for improved milk production traits. In particular, such traits include increased milk volume and milk protein content and decreased milkfat content and solid fat content. It is anticipated that herds of bovine selected for such a trait will produce milk which will be more easily processed and such milk and products made therefrom may provide health benefits to consumers, as well as producing an increased milk yield.

&null;0188&null; References

&null;0189&null; CASES, S.; SMITH, S. J.; ZHENG,Y-W.; MYERS, H. M.; LEAR, S. R.; SANDE, E.; NOVAK, S.; COLLINS, C.; WELCH, C. B.; LUSIS, A. J.; ERICKSON, S. K.; FARESE, R. V. JR (1998). Identification of a gene encoding an acyl CoA:diacylglycerol acyltransferase, a key enzyme in triacylglycerol synthesis. Proceedings of the National Academy of Sciences, USA. 95(22):13018-23.

&null;0190&null; COPPIETERS, W.; RIQUET, J.; ARRANZ, J.-J.; BERZI, P.; CAMBISANO, N.; GRISART, B.; KARIM, L.; MARCQ, F.; SIMON, P.; VANMANSHOVEN, P.; WAGENAAR, D.;GEORGES, M. (1998) A QTL with major effect on milk yield and composition maps to bovine chromosome 14. Mammalian Genome 9: 540-544.

&null;0191&null; DELOUKAS et al (1998) A physical map of the 30,000 human genes. Science 282, 744-746

&null;0192&null; EWING, B.; HILLIER, L.; WENDL, M. C.; GREEN, P. (1998). Base-calling of automated sequencer traces using Phred. I. Accuracy assessment. Genome Research 8(3):175-185.

&null;0193&null; EWING, B.; GREEN, P. (1998). Base-calling of automated sequencer traces using Phred. II. Error probabilities. Genome Research 8(3):186-194.

&null;0194&null; GORDON, D.; ABAJIAN, C.; GREEN, P. (1998). Consed: a graphical tool for sequence finishing. Genome Research 8(3): 195-202.

&null;0195&null; RIQUET, J.; COPPIETERS, W.; CAMBISANO, N.; ARRANZ, J.-J.; BERZI, P.; DAVIS, S.; GRISART, B.; FARNIR, F.; KARIM, L.; MNI, M.; SIMON, P.; TAYLOR, J.; VANMANSHOVEN, P.; WAGENAAR, D.; WOMACK, J. E.; GEORGES, M. (1999). Identity-by-descent fine-mapping of QTL in outbred populations: application to milk production in dairy cattle. Proceedings of the National Academy of Sciences, USA 96: 9252-9257.

&null;0196&null; FARNIR, F.; GRISART, B.; COPPIETERS, W.; RIQUET, J.; BERZI, P.; CAMBISANO, N.; KARIM, L.; MNI, M.; SIMON, P.; WAGENAAR, D.; GEORGES, M. (2000). Simultaneous mining of linkage and linkage disequilibrium to fine-map QTL in outbred half-sib pedigrees: revisiting the location of a QTL with major effect on milk production on bovine chromosome 14. Ph.D Thesis, University of liege 2000.

&null;0197&null; SMITH, S. J.; CASES, S.; JENSEN, D. R.; CHEN, H. C.; SANDE, E.; TOW, B.; SANAN, D. A.; RABER, J; ECKEL, R. H.; FARESE, R. V. JR (2000). Obesity resistance and multiple mechanisms of triglyceride synthesis in mice lacking DGAT. Nature Genetics 25(1):87-90.

&null;0198&null; BARENDSE, W.; ARMITAGE, S. M.; KOSSAREK, L. M.; SHALOM, A.; KIRKPATRICK, B. W.; RYAN, A. M.; CLAYTON, D.; U, L.; NEIBERGS, H. L.; ZHANG, N.; GROSSE, W. M.; WEISS, J.; CREIGHTON, P.; McCARTHY, F.; RON, M.; TEALE, A. J.; FRIES, R.; McGRAW, R. A.; MOORE, S. S.; GEORGES, M,; SOLLER, M.; WOMACK, J. E.; HETZEL, D. J. S. (1994). A genetic linkage map of the bovine genome. Nature Genet. 6: 227-235.

&null;0199&null; BISHOP, M. D.; KAPPES, S. M.; KEELE, J. W.; STONE, R. T.; SUNDEN, S. L. F.; HAWKINS, G. A.; SOLINAS TOLDO, S.; FRIES, R.; GROSZ, M. D.; YOO, J.; BEATTIE, C. W. (1994). A genetic linkage map for cattle. Genetics 136: 619-639.

&null;0200&null; COLLINS, F. S. (1995). Positional cloning moves from perditional to traditional. Nature Genet. 9: 347-350.

&null;0201&null; GEORGES, M.; ANDERSSON, L. (1996). Livestock genomics comes of age. Genome Research 6: 907-921.

&null;0202&null; GEORGES, M; NIELSEN, D.; MACKINNON, M.; MISHRA, A.; OKIMOTO, R.; PASQUINO, A. T.; SARGEANT, L. S.; SORENSEN, A.; STEELE, M. R.; ZHAO, X.; WOMACK, J. E.; HOESCHELE, I. (1995). Mapping quantitative trait loci controlling milk production by exploiting progeny testing. Genetics 139: 907-920.

&null;0203&null; HOUBENWCYL (1987). Methods of Organic Chemistry, ed. E. Wansch. Vol. 15 I and II. Thieme, Stuttgart.

&null;0204&null; HUSE et al. (1989). Science 246: 1275-1281.

&null;0205&null; KAPPES, S. M.; KEELE, J. W.; STONE, R. T.; McGRAW, R. A.; SONSTEGARD, T. S.; SMITH, T. P. L.; LOPEZ-CORRALES, N. L.; BEATTIE, C. W. (1997) A Second-Generation Linkage Map of the Bovine Genome. Genome Research 7: 235-249.

&null;0206&null; KOHLER and MILSTEIN (1975). Nature. 256: 495-497.

&null;0207&null; KOZBOR et aL (1983). Immunol. Today 4: 72.

&null;0208&null; MERRIFIELD (1964). J. Am. Chem. Assoc. 85: 2149-2154.

&null;0209&null; McCAFFERTY et al. (1990) Nature 348: 552-554.

&null;0210&null; SAMBROOK, J.; FRITSCH, E. F.; MANIATIS, T. (1989). Molecular Cloning: A Laboratory Manual. Cold Spring Harbour Lab Press, Cold Spring Harbour, N.Y.

&null;0211&null; STEWART, A. J.; CANITROT, Y.; BARACCHINI, E.; DEAN, N. M.; DEELEY, R. G. and COLE, S. P. C. (1996). Reduction of expression of the multidrug resistance protein (MRP) in human tumour cells by antisense phophorothioate oligonucleotudes. Biochem Pharmacol. 51: 461-469.

&null;0212&null; VAN RADEN, P. M., AND G. R. WIGGANS, 1991 Derivation, calculation, and use of National Animal Model Information. J. Dairy Sci. 74: 2737-2746.

&null;0213&null; JOHNSON, D. L.; THOMPSON, R. 1995. Restricted maximum likelihood estimation of variance components for univariate animal models using sparse matrix techniques and average information. J. Dairy Sci. 78: 449-456.

&null;0214&null; JOHNSON D. L., S. F. PETCH, A. M. WINKELMAN AND M. BRYANT 2000 Genetics of milk characteristics in New Zealand dairy cattle Proceedings of New Zealand Society of Animal Production 60:318-319.

&null;0215&null; MACGIBBON A. K. H. & MCLENNAN W. D. 1987. Hardness of New Zea-land patted butter: seasonal and regional variations. New Zealand Journal of Dairy Science and Technology, 22: 143-156.

&null;0216&null; FALCONER D. S. and MACKAY T. F. C. Introduction to Quantitative Genetics, 4th Edition. Longman Scientific and Technical, New York, 1996.

&null;0217&null; WELLER, J. I. et. al. Power of daughter and granddaughter designs for determining linkage between marker loci and quantitative trait loci in dairy cattle. J. Dairy Sci. 73, 2525-2537, (1990).

&null;0218&null; LYNCH M and WALSH B (1997). Genetics and analysis of quantitative traits. Sinuaer Associates, Inc. Sunderland, Mass.

&null;0219&null; SAS Institute Inc. 1985 SAS Users Guide: Statistics, Version 5 edition, Cary, N.C. 956p.

&null;0220&null; BOLDMAN K. G.; KRIESE L. A., VAN VLECK L. D.; VAN TASSELL C. P.; KACHMAN S. D. A manual for the use of MTDFREML, 1997.

&null;0221&null; http://aiipl.arsusda.gov/curtvt/mtdfreml.html.