WO2007070965A1

WO2007070965A1 - Quantitative trait loci for bovine net feed intake

Info

Publication number: WO2007070965A1
Application number: PCT/AU2006/001970
Authority: WO
Inventors: Wayne Pitchford; Cynthia Bottema; Madan Naik
Original assignee: Adelaide Research & Innovation Pty Ltd; Commonwealth Scientific And Industrial Research Organisation; Agriculture Victoria Services Pty Ltd; Meat And Livestock Australia Limited; The State Of Queensland Through Its Department Of Primary Industries And Fisheries; The University Of New England; The Department Of Primary Industries For And On Behalf Of The State Of New South Wales
Priority date: 2005-12-22
Filing date: 2006-12-22
Publication date: 2007-06-28

Abstract

The present invention relates generally to quantitative trait loci (QTL) in animals. More particularly, the present invention relates to QTL for improved feed efficiency, measured as net feed intake, in bovine animals. The QTL identified in accordance with the present invention represent regions of the bovine genome where genetic variation is correlated with the level of feed intake and/or net feed intake in bovine animals.

Description

QUANTITATIVE TRAIT LOCI FOR BOVINE NET FEED INTAKE

FIELD OF THE INVENTION

The present invention relates generally to quantitative trait loci (QTL) in animals. More particularly, the present invention relates to QTL for improved feed efficiency, measured as net feed intake, in bovine animals.

BACKGROUND OF THE INVENTION

Reference to any prior art in this specification is not, and should not be taken as, an acknowledgment or any form of suggestion that this prior art forms part of the common general knowledge in any country.

In typical beef cattle production systems, the breeding herd accounts for 65- 85% of the total feed requirements and 65-75% of this is used for maintenance. Primarily, this maintenance requirement is because cattle are a large, slowly maturing species with a low annual reproductive rate. Furthermore, only a single product is harvested (meat). Essentially, the 'machinery' of production represented by the breeding cow requires a proportionately higher level of raw 'inputs' to maintain itself than is required to produce the actual 'product', represented by the cow's offspring. The large maintenance requirement is in contrast to other production systems such as pigs or poultry, where the breeding animal has a small intake relative to the total intake of all progeny. Any improvement in the efficiency with which breeding cows maintain body weight will result in an increase in total meat production for a given amount of feed. In addition to the costs of cow maintenance, long-fed cattle for the Japanese market have a large maintenance feed cost because they are close to their mature weight and are fed on a very expensive diet. For all classes of cattle that have high feed costs, improvements in feed efficiency are desirable.

In addition to the costs of cow maintenance, young cattle (less than 3 years) that are fed expensive feed, either for finishing (period immediately prior to slaughter) or for maintaining weight during drought, are likely to be more profitable if they have lower feed requirements. There are various measures of efficiency of cattle. All are a function of growth or weight and the amount of feed eaten. Thus, by lowering feed requirements, all measures of efficiency are likely to improve.

Attempts have been made previously to select animals for gross efficiency (weight gain/feed intake). However, due to a strong genetic correlation between gross efficiency and growth rate, this resulted in faster growing animals that grew to larger mature sizes with correspondingly higher feed intakes at maturity. Thus, there is a need for a means of selecting animals at a young age to reduce feed intake without affecting mature size. With this in mind, studies have been undertaken to examine genetic variation in feed intake that is independent of body size and growth rate, termed 'residual feed intake' or 'net feed intake' or 'net feed efficiency'. These definitions of efficiency can be used interchangeably since net feed intake and residual feed intake are two terms for the identical measure. Net feed efficiency is opposite in sign but equal in magnitude to net feed intake such that animals with low net feed intake are said to have high net feed efficiency. It would also be desirable to have a means for selecting animals for meat production on the basis of feed conversion (feed/weight gain) or gross efficiency as well as selecting breeding animals for net feed intake.

Selection for net feed intake would provide an opportunity to significantly reduce feed costs in livestock breeding programs. However, direct measurement of net feed intake is expensive, for example at present direct measurement of net feed intake costs around AUD$500 ($300 labour plus $200 feed) per animal. Thus, direct measurement of net feed intake is not an economical test for producers. An alternative would be to use a genetic DNA test for markers associated with intake, as such DNA tests currently cost less than AUS$100 for traits such as marbling and tenderness. DNA testing can also significantly reduce the generation interval because animals can be selected at a very early age without measuring the trait.

Thus, there is a need to locate regions in animal genomes, particularly the cattle genome, that contain genomic nucleotide sequences that affect net feed intake. Identification of these regions called "quantitative trait loci" (QTL) will allow the identification of specific alleles or other genomic nucleotide sequences that segregate in specific breeds and cause variation in feed intake and efficiency. These "markers" can then be tested in selection programs and then used for marker assisted selection in breeding programs.

SUMMARY OF THE INVENTION

The present invention is predicated, in part, on the identification of quantitative trait loci, or QTL, on bovine autosomes 1 , 8, 1 1 and 20, which are associated with feed intake and/or net feed intake in bovine animals.

The QTL identified in accordance with the present invention represent regions of the bovine genome where genetic variation is correlated with the level of feed intake and/or net feed intake in bovine animals.

Thus, by selecting a bovine animal having a known level of feed intake, and/or net feed intake, and then determining a genomic nucleotide sequence of the animal at one or more of the subject QTL, it is possible to identify a genomic nucleotide sequence at the QTL, which is associated with the particular level of feed intake of the bovine animal. - A -

Accordingly, in a first aspect, the present invention provides a method for identifying a genomic nucleotide sequence associated with a particular level of feed intake and/or net feed intake in a bovine animal, the method comprising: selecting a bovine animal having a known level of feed intake and/or net feed intake; and determining a genomic nucleotide sequence of the animal at a QTL on one or more of bovine autosomes 1 , 8, 1 1 and 20; wherein the genomic nucleotide sequence at the QTL is associated with the level of feed intake and/or net feed intake of the bovine animal.

In one embodiment, the first aspect of the invention provides a method for identifying a genomic nucleotide sequence that is associated with low feed intake and/or low net feed intake in a bovine animal.

In a second aspect, the present invention also provides a genomic nucleotide sequence that is associated with a particular level of feed intake and/or net feed intake in a bovine animal, wherein the genomic nucleotide sequence is identified in accordance with the first aspect of the invention.

Once one or more genomic nucleotide sequences at the QTL have been identified as being associated with a particular level of feed intake and/or net feed intake, it is then possible, for example, to screen for one or more of these genomic nucleotide sequences in animals of unknown feed intake and/or net feed intake in order to predict the level of feed intake and/or net feed intake of the animal.

Accordingly, in a third aspect, the present invention also provides a method for predicting the feed intake and/or net feed intake of a bovine animal, the method comprising determining a genomic nucleotide sequence of one or more cells of the bovine animal at a QTL on one or more of bovine autosomes 1 , 8, 1 1 or 20, wherein the genomic nucleotide sequence at the QTL is indicative of the feed intake and/or net feed intake of the bovine animal. The QTL, and/or genomic nucleotide sequences at the QTL, identified in accordance with the present invention can also be used for the development of marker-assisted breeding or rapid assessment techniques to identify and select for animals having a desired level of feed intake and/or net feed intake.

Accordingly, in a fourth aspect, the present invention provides a method of producing a bovine animal with a desired level of feed intake and/or net feed intake, the method comprising the steps of: selecting a first bovine animal comprising a genomic nucleotide sequence that is associated with a desired level of feed intake and/or net feed intake at a QTL on one or more of bovine autosomes 1 , 8, 1 1 or 20; sexually or asexually reproducing the animal; and selecting progeny arising from the sexual or asexual reproduction which comprise the genomic nucleotide sequence associated with the desired level of feed intake and/or net feed intake at the QTL.

In a fifth aspect, the present invention provides a bovine animal, the bovine animal being a progeny of the breeding method of the fourth aspect of the invention.

In a sixth aspect, the present invention also contemplates the use of the bovine animal of the fifth aspect of the invention, or reproductive material therefrom, for breeding.

In a seventh aspect, the present invention also provides a method for modulating the overall feed intake and/or overall net feed intake of a herd of bovine animals, the method comprising: selecting bovine animals for breeding within the herd, wherein the bovine animals selected for breeding comprise a genomic nucleotide sequence that is associated with a desired level of feed intake and/or net feed intake at a QTL on one or more of bovine autosomes 1 , 8, 1 1 or 20; and breeding the herd from those selected animals which comprise the genomic nucleotide sequence associated with the desired level of feed intake and/or net feed intake at the one or more QTL.

In an eighth aspect, the present invention provides a herd of bovine animals wherein the herd has altered overall feed intake and/or altered overall net feed intake, wherein the herd is produced according to the method of the seventh aspect of the invention.

Throughout this specification, unless the context requires otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element or integer or group of elements or integers but not the exclusion of any other element or integer or group of elements or integers.

DESCRIPTION OF PREFERRED EMBODIMENTS

It is to be understood that the following description is for the purpose of describing particular embodiments only and is not intended to be limiting with respect to the above description.

As set out above, the present invention is predicated, in part, on the identification of quantitative trait loci, or QTL, that are associated with feed intake and/or net feed intake in bovine animals.

As referred to herein, "a bovine animal" includes at least all species within the genus Bos, including Bos taurus, Bos indicus, Bos frontalis, Bos grunniens, Bos javanicυs, Bos mυtυs and Bos sauveli, as well as hybrids of any of these species. In one embodiment, a bovine animal refers to a member of the species Bos taurus or Bos indicus or a hybrid of these species. As used herein, "feed intake" should be understood as the actual feed intake of an animal per unit time (eg. per day, week, month, year etc.). In one embodiment, feed intake may be measured as average daily feed intake.

"Net Feed Intake" or "NFI" should be understood as the difference between the actual feed intake of an animal and the predicted feed intake of the animal, wherein the predicted feed intake of the animal is determined on the basis of the body weight of the animal and/or the average daily gain of the animal. Accordingly, a relatively low net feed intake (high feed efficiency) is considered an actual feed intake that is lower than the predicted feed intake, whereas a relatively high net feed intake (low feed efficiency) is a feed intake that is higher than the predicted feed intake.

Bigger and faster growing cattle generally eat more than small, slow growing cattle. Accordingly, intake may be modelled (eg. the multiple regression model below) as a function of average weight and weight gain during a test period.

For example, intake data may be modelled as a regression of weight as below:

Y, = μ + βi(average weight) + β₂(average daily weight gain) + R,

Where Y, is the animal's feed intake, μ is the mean intake of the test group, βi is the regression coefficient of intake on average weight, β₂ is the regression coefficient of intake on weight gain, and R, is the residual (net) feed intake.

The present invention utilises genetic marker technology to statistically correlate the presence or absence of particular genetic markers in bovine animal populations with the magnitude of a particular physiological trait, eg. feed intake and/or net feed intake. This statistical association is based on the technique of multiple simultaneous linear regressions of trait data with genetic marker allele data using computer software. In accordance with the present invention, the cattle genome was "scanned" for groups of markers that exhibited covariance with the trait of interest, ie. feed intake and/or net feed intake. These groups of markers were then classified as bounding a QTL at least partially controlling that trait. The degree of association between the markers and the trait can then be used to estimate the "strength" of the QTL, i.e., the percentage of the trait variance that the particular QTL can account for.

The QTL identified in accordance with the present invention may include, but are not necessarily limited to, the particular gene or genes involved in the control of feed intake and/or net feed intake in bovine animals. Rather, the QTL of the present invention represent regions of the bovine genome that include genes, transcriptional control sequences (such as promoters, enhancers, repressors and the like) or other genomic nucleotide sequences that are potentially involved in the control of feed intake and/or net feed intake in bovine animals, and/or include genetic markers linked to these traits.

Accordingly, as referred to herein, a "quantitative trait locus" or "QTL" should be understood to refer to a genetic locus or region of genomic DNA in an organism wherein genetic variation at the locus is associated with variation in a phenotypic trait. Therefore, the QTL of the present invention comprise genetic loci in the bovine genome where genetic variation is associated with variation in the feed intake and/or net feed intake of the bovine animal.

Thus, the identification of the QTL in accordance with the present invention identifies regions of the cattle genome that are associated with feed intake and/or net feed intake. These genomic regions may include, for example: a genomic nucleotide sequence which is directly or indirectly involved in determining the feed intake and/or net feed intake of a bovine animal; and/or a genomic nucleotide sequence which may not be involved in determining the feed intake and/or net feed intake of a bovine animal, but which nonetheless exhibits linkage with this trait. By selecting a bovine animal having a known level of feed intake and/or net feed intake and then determining a genomic nucleotide sequence of the animal at one or more of the QTL identified in accordance with the present invention, it is possible to identify a genomic nucleotide sequence at the QTL, which is associated with the particular level of feed intake and/or net feed intake of the bovine animal.

The QTL of the present invention may be further defined, for example, with reference to the genetic markers that delimit the region of genomic DNA which defines the QTL and/or the genetic markers that are encompassed by the QTL.

In a first embodiment, the QTL is on bovine autosome 1.

In another embodiment, the QTL is a region of bovine autosome 1 that is delimited by microsatellite marker CSSM032 and microsatellite marker

BMS2263; and/or is between BTA1 Flanking marker 1 and BTA1 Flanking marker 2 as defined in Table 3.

In yet another embodiment, the QTL is a genomic region proximal to microsatellite marker BM1824 on bovine autosome 1 ; and/or proximal to the BTA1 peak marker as defined in Table 3. In a yet further embodiment, the QTL on bovine autosome 1 comprises one or more of: the alpha-HS-glycoprotein gene (ASHG), the glucose transporter 2 gene (SLC2A2), the growth hormone factor 1 /pituitary specific positive transcription factor 1 gene (P0LJ1 F1 ), the interleukin 12A gene (IL12A), the uridine monophosphate synthetase gene (UMPS), or the NADH dehydrogenase (ubiquinone) 1 -alpha subcomplex, 10, 75kDa gene (Tables 1-2).

In a second embodiment, the QTL is on bovine autosome 8.

In another embodiment, the QTL is a region of bovine autosome 8 that is delimited by microsatellite marker BMS1341 and microsatellite marker BMS836; and/or is between BTA8 Flanking marker 1 and BTA8 Flanking marker 2 as defined in Table 3.

In yet another embodiment, the QTL is a genomic region proximal to the LPL gene and/or microsatellite marker BMS71 1 on bovine autosome 8; and/or proximal to the BTA8 peak marker as defined in Table 3.

In a yet further embodiment, the QTL on bovine autosome 8 comprises one or more of: the cathepsin B gene (CTSB), the lipoprotein lipase gene (LPL) and the endothelial tyrosine kinase gene (TEK) (Tables 1-2).

In a third embodiment, the QTL is on bovine autosome 1 1.

In another embodiment, the QTL is a region of bovine autosome 1 1 that is delimited by microsatellite marker RM096 and microsatellite marker RM363; and/or is between BTA1 1 Flanking marker 1 and BTA1 1 Flanking marker 2 as defined in Table 3.

In yet another embodiment, the QTL is a genomic region proximal to microsatellite marker BM1048 on bovine autosome 1 1 ; and/or proximal to the BTA1 1 peak marker as defined in Table 3. In a yet further embodiment, the QTL on bovine autosome 1 1 comprise one or more of: the argininosuccinate gene (ASS), the follicle stimulating hormone receptor gene (FSHR) and the proopiomelanocortin (POMC) gene, the lnterleukin 1 β (IL1 B) gene, the or the NADH dehydrogenase (ubiquinone) 1- alpha subcomplex, 10, 19kDa gene or the ATPase, V1 subunit G2 gene (Tables 1 -2).

In a fourth embodiment, the QTL is on bovine autosome 20.

In another embodiment, the QTL is a region of bovine autosome 20 that is delimited by microsatellite marker TGLA126 and microsatellite marker BM5004; and/or is between BTA20 Flanking marker 1 and BTA20 Flanking marker 2 as defined in Table 3.

In yet another embodiment, the QTL is a genomic region proximal to microsatellite marker BMS703 on bovine autosome 20; and/or proximal to the

BTA20 peak marker as defined in Table 3.

In a yet further embodiment, the QTL on bovine autosome 20 comprise one or more of: the 5'-AMP-activated protein kinase, catalytic alpha-1 chain gene (PRKAA1 ) gene, the follistatin precursor gene (FST), the growth hormone receptor (GHR) gene, the microtubule-associated protein 1 B (MAPI B) gene, the phosphatidylinositol 3-kinase P-85-alpha subunit (PI3KR1 ) gene, or the prolactin receptor gene (PRLR) (Tables 1-2).

As referred to herein, a QTL comprising a region which is "proximal" to a particular marker refers to genomic sequence within about 5OcM, 2OcM, 1 OcM or 5cM of the marker, as determined by genetic linkage analysis. In some embodiments, the region may be genomic sequence within 2cM, within 1 cM or within 0.5 cM, 0.2 cM or 0.1 cM of the marker. Details of the specific microsatellite markers referred to herein, including their position in the genome, heterozygosity, number of alleles and suitable primers for amplification are provided by the US Meat Animal Research Centre (MARC) at

and by the National Center for Biotechnology Information (NCBI) Cow Genome Resources at http://www.ncbi.nlm.nih.gov/mapview/map search. cgi?taxid=9913. Methods for identifying and amplifying microsatellite markers, such as those referred to above, as well as methods for the determination of polymorphisms thereof are known in the art. For example, suitable methods are described by Casas ef a/. {Anim. Genet. 35: 2, 2003) and lhara et al. {Genome Res. 14: 1987-1998, 2004).

The genomic nucleotide sequence associated with a particular level of feed intake and/or net feed intake in a bovine animal at the one or more QTL, as determined in accordance with the present invention, may comprise any nucleotide sequence present in the genomic region defined by the QTL. Furthermore, the determined genomic nucleotide sequence may be the entire genomic sequence in the region defined by the QTL or, more typically, the determined genomic sequence may be a part of the genomic sequence in the region defined by the QTL.

For example, the genomic nucleotide sequence may include one or more of: a protein encoding nucleotide sequence, a 5' or 3' UTR region, a transcriptional control sequence, a non-transcribed RNA encoding genomic sequence, or a non-transcribed genomic nucleotide sequence within the QTL.

In some embodiments of the invention, the genomic nucleotide sequence associated with a particular level of feed intake and/or net feed intake in a bovine animal at the one or more QTL comprises a mutation, sequence variant or SNP. In one embodiment, the genomic nucleotide sequence at the QTL may comprise a genomic nucleotide sequence that directly or indirectly influences the net feed intake of a bovine animal, for example a sequence, sequence variant, mutation or SNP which is in a gene or associated regulatory sequence that influences the feed intake and/or net feed intake of the bovine animal.

Alternatively, the genomic nucleotide sequence at the QTL may comprise a genomic nucleotide sequence that does not influence the feed intake and/or net feed intake of the animal, but rather is in linkage disequilibrium with a sequence that does influence the feed intake and/or net feed intake of the bovine animal.

"Linkage disequilibrium" (LD) refers to the non-random association of alleles at two or more loci. The loci in linkage disequilibrium may be on the same chromosome, but are not necessarily on the same chromosome. LD describes a situation in which some combinations of alleles or genetic markers occur more or less frequently in a population than would be expected from a random formation of haplotypes from alleles based on their frequencies.

Methods for identifying a genomic nucleotide sequence at a QTL are well known in the art and include, for example, the methods of positional cloning, the candidate gene approach and combinations thereof (for examples see Coppieters et ai, Archiv fur Tierzucht - Archives of Animal Breeding 42: 86-92, 1999; and Goddard, Animal Science 76: 353-365, 2003).

Furthermore, in order to determine the genomic nucleotide sequence of an animal at the one or more QTL, a DNA sample is generally first collected from the test animal. The DNA sample may be derived from any suitable material such as a hair, blood, semen or other tissue or cell sample derived from the test animal. Once a sample has been collected, DNA may be extracted from the sample using any suitable method known in the art. For example, the method of Miller et al. {Nucl. Acids Res. 16: 1215, 1988) may be used to isolate a DNA sample from the blood of a test animal. In some embodiments, the first aspect of the invention provides a method for identifying a genomic nucleotide sequence, at the one or more QTL of the present invention, which is associated with low feed intake and/or low net feed intake in a bovine animal.

As set out in the Examples, the method of the first aspect of the invention was applied to identify a number of exemplary 'candidate genes' and specific SNPs in these candidate genes which are associated (either directly or via linkage disequilibrium) with the level of feed intake and/or net feed intake in bovine animals. For example, Table 4 in the Examples identifies a number of exemplary SNPs in genes within the QTL of the present invention which are associated with the level of feed intake and/or net feed intake in bovine animals.

Accordingly, in a second aspect, the present invention also provides a genomic nucleotide sequence that is associated with a particular level of feed intake and/or net feed intake in a bovine animal, wherein the genomic nucleotide sequence is identified in accordance with the first aspect of the invention.

In one embodiment, the genomic nucleotide sequence of the second aspect of the invention comprises a genomic nucleotide sequence associated with low feed intake and/or low net feed intake. In an alternate preferred embodiment, the genomic nucleotide sequence of the second aspect of the invention comprises a genomic nucleotide sequence associated with high feed intake and/or high net feed intake.

Once one or more genomic nucleotide sequences at the QTL have been identified as being associated with a particular level of feed intake and/or net feed intake, it is then possible, for example, to screen for one or more of these genomic nucleotide sequences in animals of unknown feed intake and/or net feed intake in order to predict the level of feed intake and/or net feed intake of the animal. Accordingly, in a third aspect, the present invention also provides a method for predicting the feed intake and/or net feed intake of a bovine animal, the method comprising determining a genomic nucleotide sequence of one or more cells of the bovine animal at a QTL on one or more of bovine autosomes 1 , 8, 1 1 or 20, wherein the genomic nucleotide sequence at the QTL is indicative of the feed intake and/or net feed intake of the bovine animal.

The one or more cells used to determine a genomic nucleotide sequence at the QTL may be any cell from the animal, including, for example, somatic cells, blood cells, sperm cells, egg cells and the like.

In one embodiment, the genomic nucleotide sequence at the QTL, which is indicative of the feed intake and/or net feed intake of the animal, comprises a genomic nucleotide sequence identified in accordance with the method of the first aspect of the invention. For example, the genomic nucleotide sequence at the QTL, which is indicative of the feed intake and/or net feed intake of the animal, comprises may comprise a SNP as set out in Table 4 of the Examples.

The method of the third aspect of the invention may be used, for example, to identify animals having a relatively low feed intake and/or low net feed intake or a relatively high feed intake and/or high net feed intake by screening for genomic nucleotide sequences at the QTL that are associated with relatively low- or relatively high feed intake and/or net feed intake, respectively.

Accordingly, the method of the third aspect of the invention may further comprise the step of identifying a bovine animal having a feed intake level and/or net feed intake level of interest on the basis of the genomic nucleotide sequence of the animal at a QTL defined by the present invention.

Genetic identification of animals having a relatively low feed intake and/or low net feed intake in accordance with the third aspect of the invention provides, among other things, a significantly more economical method for identifying animals having these traits than current feeding-trial based tests.

As such, the QTL, and/or genomic nucleotide sequences at the QTL, identified in accordance with the present invention can also be used for the development of marker-assisted breeding or rapid assessment techniques to identify and select animals having a desired level of feed intake and/or net feed intake.

For example, by selecting animals for breeding on the basis of genomic nucleotide sequences that are associated with low feed intake and/or low net feed intake at one or more of the QTL identified herein, overall decreases in feed intake and/or overall decreases in net feed intake may be achieved in cattle herds.

For example, the genetic variation at each of the QTL on chromosomes 1 , 8, 11 and 20 identified in accordance with the present invention accounts for approximately 8% of the variance observed in net feed intake in cattle. Therefore, in combination, the genetic variation at these QTL could account for up to about 30% of the variance in net feed intake observed in bovine animals.

The method of the third aspect of the invention may be used to identify and select animals having a desired level of feed intake and/or net feed intake for any purpose. For example, the selected animals may be used in breeding programs, may be selected for use in a feedlot, or the like.

Therefore, as set out above, the present invention also provides methods for producing bovine animals to produce progeny having low feed intake and/or low net feed intake.

In one embodiment "selecting a first bovine animal comprising a genomic nucleotide sequence that is associated with a desired level of feed intake and/or net feed intake at a QTL on one or more of bovine autosomes 1 , 8, 11 or 20" may include selecting an animal on the basis of the animal including a particular genomic nucleotide sequence at the QTL which is associated with a desired level of feed intake and/or net feed intake. For example, an animal may be selected on the basis of it having a genomic nucleotide sequence at the QTL which is known to be associated with low feed intake and/or low net feed intake. However, in another embodiment "selecting a first bovine animal comprising a genomic nucleotide sequence that is associated with a desired level of feed intake and/or net feed intake at a QTL on one or more of bovine autosomes 1 , 8, 1 1 or 20" may include selecting an animal on the basis that it does not contain a genomic nucleotide sequence at the QTL which is known to be associated with a non-desired level of feed intake and/or net feed intake. For example an animal may be selected on the basis of it not having a genomic nucleotide sequence at the QTL which is known to be associated with high feed intake and/or net feed intake.

In another embodiment, the selected bovine animal is identified in accordance with the third aspect of the invention. In another embodiment, the selected bovine animal has low feed intake and/or low net feed intake.

"Sexually or asexually reproducing the animal" contemplates any means by which a selected bovine animal may be reproduced. For example, this term includes natural service mating, in-vitro fertilisation (including intracytoplasmic sperm injection techniques), cloning (including nuclear transfer cloning) and the like.

As referred to herein, the term "reproductive material" includes a gamete from a bovine animal including a sperm or egg. Also, the term "reproductive material" may include a somatic cell, or the nucleus thereof, if such a cell or nucleus is used to clone a bovine animal using methods such as nuclear transfer. For examples of nuclear-transfer mediated cloning see Wilmut ef a/. {Nature 385: 810-813, 1997), Westhusin ef a/. {Nature 415: 859, 2002) and Hochedlinger and Jaenisch (Nature 415: 1035-1038, 2002).

Furthermore, the present invention also provides methods for decreasing the overall feed intake and/or net feed intake of a herd of bovine animals by selecting animals comprising genomic nucleotide sequences associated with low feed intake and/or low net feed intake at one or more of the QTL defined herein for breeding within the herd.

In one embodiment the animals selected for breeding may on the basis of the animal(s) including a particular genomic nucleotide sequence at the QTL that is associated with a desired level of feed intake and/or net feed intake. For example, animals may be selected on the basis of having a genomic nucleotide sequence at the QTL which is known to be associated with low feed intake and/or low net feed intake. However, in another embodiment animals may be selected on the basis that they do not contain a genomic nucleotide sequence at the QTL that is known to be associated with a non-desired level of feed intake and/or net feed intake. For example animals may be selected on the basis of not having a genomic nucleotide sequence at the QTL which is known to be associated with high feed intake and/or net feed intake.

As referred to herein, a decrease in the overall feed intake and/or net feed intake of a herd of includes, for example, a decrease in the mean or median feed intake and/or net feed intake of one or more of the bovine animals in the herd. Furthermore, a decrease in the overall feed intake of the animals in the herd includes, for example, an overall decrease in the amount of feed eaten by the herd.

In one embodiment, one or more of the selected bovine animal(s) are identified in accordance with the third aspect of the invention. In another embodiment, one or more of the selected bovine animal(s) have low feed intake and/or low net feed intake.

Finally, reference is made to standard textbooks of molecular biology that contain methods for carrying out basic techniques encompassed by the present invention. See, for example, Maniatis et al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Laboratory Press, New York, 1982) and Sambrook et al., Molecular Cloning: A Laboratory Manual, 2^nd Ed. (Cold Spring Harbor Laboratory Press, New York, 1989).

The present invention is further exemplified by the following non-limiting examples:

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows the results of QTL mapping for NFI on bovine autosomes 1 , 8, 1 1 and 20 in the Trangie Angus selection lines. Log ratio of greater than 3.84 is considered significant. Panels A to D show the results for BTA1 , BTA8, BTA1 1 and BTA20, respectively, "fcr" is feed conversion ratio (feed intake / weight gain during the 70 day test period), "sdfi70" is standardised (to 10MJ/kg DM) daily feed intake during the 70 test period, "nfi70" is net feed intake during the 70 day test period, and "adg" is average daily weight gain during the 70 day test period.

Figure 2 is a graphical representation of the QTL significance for net feed intake with and without SNPs in the analysis and also comparing results with those in Angus cattle. Panels A to D show the results for BTA1 , BTA8, BTA1 1 and BTA20, respectively. Solid line micro-satellite (SSR) markers only, Dashed line including SNPs.

Figure 3 is a graphical representation showing AMPK enzyme activity in liver cells of high and low efficiency animals. Figure 4 is a graphical representation showing the activity of oxidative phosphorylation complexes in liver mitochondria of high and low efficiency animals. A. Complex I. B. Complex II. C. Complex IV. Bar indicates the mean.

Figure 5 shows a 2D electrophoresis gel of liver mitochondrial proteins from high and low efficiency animals. Panel A shows an example of a gel run with Cy3 and Cy5 fluorescently labelled proteins. Panel B shows an example of a protein spot that varies in concentration by more than 1.5-fold, as measured by the difference in signal intensity.

EXAMPLE 1 Materials and Methods

(i) General trial design

The trial design involved two of the more extreme Bos taurus dam breeds, Jersey (J) and Limousin (L), mated to first cross (JxL or LxJ) sires to produce back-cross calves. A total of about 400 heifer and steer progeny were generated in three calf crops (born 1996-98) with a total of 323 records available for analysis.

(H) Animal measurements

Calves were weaned at approximately 8 months of age, and raised on pasture for a further 12 months. The animals were then transported to an intensive feeding facility where feed intake was measured using Ruddweigh electronic feeders. Cattle in a feeding pen (8-10 per pen) were tagged with electronic ear tags that produce a signal with a unique number. When an animal enters the feeding box, an infra-red beam is broken recording of feed intake and body weight (1996 born cattle were weighed weekly in separate system). When the animal leaves the feeding box, the infra-red beam is reset. The weight of food that an animal has eaten is then calculated as the weight of the food bin after feeding has finished subtracted from the weight of the feed bin before feeding commenced.

Feed intake data was processed by calculating least-squares means for each animal over a test period. Day was included in the model to allow for weather, personnel, time of feeding and any other factors that could affect the intake. Metabolic mid-weight (MMWT) was calculated as the average weight over the test period raised by the power 0.73. Average daily weight gain (ADG) was calculated as the regression coefficient of weight against day of test. Net feed intake (NFI) was calculated after modelling daily feed intake for MMWT and ADG while on the feed intake test. The initial equation used to calculate NFI included the main effect of cohort (6 combinations of year of birth and sex of calf) and interactions between MMWT and ADG. Thus, a model comprising MMWT, ADG and cohort was used.

(Hi) Genotyping

CRI-MAP (Green et al., Documentation for CRI-MAP, version 2.4, Washington University School of Medicine, St. Louis MO, 1990) was used to detect any errors in the genotyping of the cattle and to confirm marker order in the linkage maps. Non-Mendelian inheritance is detected when CRI-MAP analyses pedigree and genomic nucleotide sequence data file (^*.gen file). However, errors in genotyping do not always lead to non-Mendelian inheritance (the perceived error rate is less than the true error rate) so once the linkage map was built and confirmed with the published map (Bovine ArkDB; M£;//texaMbeiir^^^ ; accessed 20^th January 2002), the chrompic function of CRI-MAP was used. The chrompic function finds the maximum likelihood estimates of the recombination fractions of the specified locus order; these estimates are then used to find the particular phase for each sire family and the grand-parental and grand-maternal phases (Green et al., 1990, supra). Chrompic also gives the number of recombination events. Any individuals with double or triple recombination events were carefully examined for genotyping or pedigree errors, as double recombinations are unlikely when markers are spaced at 2OcM intervals.

All genotyping was done on contract by AgResearch, New Zealand. Genomic DNA was amplified by PCR using proprietary primers flanking the microsatellite markers. The PCR products were radiolabeled and separated by polyacrylamide gel electrophoresis. Autoradiographs of the polyacrylamide gels were independently scored twice. Sire-derived alleles were determined for a total of 253 informative microsatellite loci (an average of 185 loci per sire group) spread across all bovine autosomes (BTA). There were 3-9 markers typed per chromosome for the 29 autosomes, at approximately 2OcM intervals. The 3

sires were genomic nucleotide sequenced to ensure they were heterozygous with the alleles coded "A" and "B". Thus, progeny inherited either an "A" or "B" allele from their sire. The genomic nucleotide sequences of the progeny were coded AA, AB, BB, AC or BC with the "C" representing any other allele. In addition, 43 SNP markers in 22 candidate genes were sequenced by primer extension. The MARC2004 marker locations were used for the linkage analysis and for candidate genes not on the MARC map, locations were based on CRIMAP analyses.

(iv) Statistical analysis

Genomic nucleotide sequence probabilities were calculated using QTL Express (Seaton et ai, Bioinformatics 18: 339-340, 2002; http://qtl.cap.ed.ac.uk ; accessed November 2002 then regularly during 2004-5) for every 1cM on each chromosome. Animals were assigned a value of either 0 (which represents the "A" allele) or 1 ("B" allele) or 0.5 if uninformative. The genomic nucleotide sequence probability was then calculated between 0.5 and 1 , depending on the level of confidence. Phenotypes were regressed against the genomic nucleotide sequence probabilities for every chromosome using "Haley and Knott regression" (Haley and Knott, Heredity 69: 312-324, 1992). Cohort (combination of sex and breed) and breed of dam (Jersey or Limousin) were included as factors in the model and the regression was nested within sire. Additional models that included two QTL or QTL by breed of dam interactions were also tested for a number of traits. Experiment-wise threshold values were calculated according to Lander and Kruglyak (Nature Genetics 1 1 : 241-247, 1995) and results reported are at least suggestive of linkage.

EXAMPLE 2 QTL identified on Bos taurus autosomes 1 , 8, 1 1 and 20

As shown in table 1 , QTL for net feed intake in cattle were identified on Bos taurus autosomes 1 , 8, 1 1 and 20.

The QTL were identified as regions of relatively high F-ratios in the offspring of one or more of the sires (F-ratio (361 ), F-ratio (368) or F-ratio (398)) or a relatively high F-ratio across the offspring of the three different sires.

For example, the data shown in table 1 indicates:

(i) a QTL on chromosome 1 flanked by microsatellite markers CSSM032 and BMS2263, with an F-ratio that peaks at microsatellite marker

BM 1824;

(ii) a QTL on chromosome 8 proximal to microsatellite marker BM711 ;

(iii) a QTL on chromosome 1 1 flanked by microsatellite markers RM096 and RM363, with an F-ratio that peaks at a SNP marker in the POMC gene; and

(iv) a QTL on chromosome 20 flanked by a SNP marker in the ITGA2 gene and microsatellite marker BMS521 , with an F-ratio that peaks at microsatellite marker BMS703. TABLE 1 - QTL on chromosomes 1 , 8, 1 1 and 20 in cattle

^* Denotes SNP markers in genes.

Table 2, below, sets out the size of effect (in kg feed/day) of the QTL associated with each of the markers and Table 3 documents the comparative map locations of the flanking markers for each QTL.

TABLE 2 - Size of effect (kg feed/day) of QTL associated with markers.

^* Denotes a SNP marker in a gene.

TABLE 3 - Comparative map locations of QTL markers (from USDA website (www.animalgenome.org/cattle/maps) and Ensembl website (www.ensembl.org/index.html) as accessed from 12/05-12/06)

EXAMPLE 3 QTL in other cattle breeds

In 1993, a research project (MLA Project DAN.075) commenced at the Agricultural Research Centre in Trangie, NSW to investigate the potential for genetic improvement in post-weaning feed efficiency as a means of improving whole beef production system efficiency.

Starting with the 1993 born animals, feed intake and efficiency tests were conducted each year using an automated feeding system that delivers and records individual animal feed intake. Feed efficiency was measured immediately post-weaning over a 70-day test and was calculated as net feed intake (NFI). NFI measures whether an animal consumes more or less feed than that predicted for growth and maintenance, with low NFI being superior. In 1994, high feed efficiency (low NFI) and low feed efficiency (high NFI) divergent selection lines were established. Starting with the 1993 born animals, the females were allocated to the high efficiency line and the low efficiency line based on their individual NFI values. The three most efficient bulls born in 1993 were allocated to the high efficiency line and the three least efficient bulls to the low efficiency line. Throughout the experiment, the sole selection criterion for all replacement bulls and heifers in the high and the low efficiency lines was individual NFI. This design was chosen to provide a rapid divergence in NFI between the high and low selection lines.

Only 200 animals could be tested in the Trangie feed efficiency testing facility at any one time, and for this project, a maximum of 100 males and 100 females were tested per year. Therefore, there was very little selection in the females due to limited numbers. In the males, however, three to six bulls were selected per line each year, depending upon the number of females available to be mated. Throughout the project, bulls and heifers were mated at 14 months of age, and bulls were used for only one mating season except for the 1997 and 1998 mating seasons, where for each selection line, one bull from the previous year was used again. Animals from each selection line were grazed together throughout the year, except during mating. Allocation of mates within selection line was completely random, except for the avoidance of half-sib and son-dam matings. All matings were by natural service. The first progeny of selected parents were born in 1995 and the last in 1999. Calves were nursed by their dams until weaning, and the breeding herd was on pasture all year round, with supplementary feed (lucerne hay and wheat) offered during times of limited pasture growth. The animals were brought to the efficiency testing facility a few weeks (generally 4 to 6 weeks) after weaning. Records taken during the test were used to calculate NFI for each animal.

At the efficiency testing facility, a pre-test adjustment period of at least 21 days was allowed for the animals to adapt to the feeding system and diet, followed by a 70-day test. The average age at the start of test was 268 days. During the test, animals had ad libitum access to a pelleted ration of approximately 10.5 MJ/kg dry matter and 16% crude protein. Records taken during the test were used to calculate NFI for each animal.

The growth of each animal during the test was modelled by linear regression of weight on time (days), and the regression coefficient represented average daily gain (ADG). The mean weight (MWT) of an animal during the test was computed as the average of the start and end of test weights. Metabolic body weight (MMWT) was calculated as MWT^{0 73}. Feed intake (Fl) was standardised to a concentration of 10 MJ ME/kg dry matter. Feed conversion ratio was calculated as Fl divided by ADG. A linear regression model of Fl on MMWT and ADG for each animal, with test group and sex included as class variables, was fitted to the data. The regression coefficients from this model were used to obtain expected feed intake of all animals based on ADG and MMWT. NFI was calculated as the actual (measured) feed intake minus that predicted using the regression equation. Following results of QTL mapping in the Davies population, four chromosomes were selected for a mapping experiment in the Trangie population. These were chromosomes 1 , 8, 1 1 , 20. There were between 4 to 7 microsatellite markers per chromosome. The markers were selected to be evenly spaced along the chromosome. The sire families that were genotyped were selected across the Trangie pedigree, with the main criterion being sufficient numbers of progeny per sire to provide adequate power to detect QTL of moderate to large effect. Of the progeny, 1600 animals were genotyped for each marker.

The microsatellite data and the NFI phenotypes were analysed in a two-step approach. In the first step, the inheritance of the markers is traced through the complex pedigree that connects all the animals using a Gibbs sampling method. This computer program calculates the probability that any two animals inherited QTL alleles, at a specific point on the chromosome, that are identical by descent. The second step uses these probabilities in a linear model that includes the QTL, polygenic inheritance and fixed effects. A likelihood ratio test was performed to evaluate if the QTL was significant at each point along the chromosome. Let L₀ be the likelihood value for the model under H₀, where the QTL is not included in the model. Then Li is the likelihood value for the alternative model, that is, the QTL are included in the model. The test statistic was defined as - 2(lnL₀ - InL₁) . This statistic is significant at P<0.05 when it exceeds 3.84.

As shown in Figure 1 , in the Trangie Angus selection line population, there were clearly significant QTL on BTA 1 , 8, 1 1 and 20. Results from Trangie Angus are compared with those in Davies population in Figure 2. Specifically, QTL were identified at 0-4OcM and 70-10OcM on BTA 1 , +40-10OcM on BTA8, 80-10OcM on BTA11 and 0-15cM and 45-6OcM on BTA20. EXAMPLE 4

Identification of genomic nucleotide sequences associated with low feed intake and/or low net feed intake

After locating a QTL, the corresponding region of the bovine genome is aligned with the human and/or mouse genomes. Based on this comparative mapping, the human and/or mouse genome databases are examined for genes within the aligned regions that potentially affect net feed intake (for example see O'Brien et al., Science 286: 458).

Genes identified as potential candidates for controlling net feed intake are then sequenced from the genomic DNA of the three mapping sires. Sequence variants observed among the candidate genes in the three sires are analysed to determine whether they could potentially be functional, that is, whether the sequence variation is likely to affect gene expression or activity.

Potential functional sequence variants are then genotyped in the progeny of the sires and statistical analyses are performed to determine if the sequence variant is associated with high or low net feed intake. For example, a SNP marker in the GHR gene, which effects an amino acid substitution, addition or deletion in the growth hormone receptor, may lead to altered net feed intake in animals which carry the SNP marker.

EXAMPLE 5

Genotype effects on daily and net feed intake

QTL mapping was also conducted in mouse net feed intake selection lines. The QTL for net feed efficiency that mapped to homeologous or equivalent regions in cattle and mouse were examined for candidate genes by comparative mapping. By using the available gene mapping data in cattle, comparative chromosomal maps for human and mouse from the 4 cattle QTL map locations were prepared (e.g. Table 3). Genes within the homeologous human and mouse genomic regions were selected as candidate genes based on their known function.

For each candidate gene, primers were designed to amplify the coding regions of the gene and the PCR conditions were optimised to obtain a single genomic product from the primers. The genomic DNA from the 3 F1 mapping sires was amplified and sequenced. Sequence from the three sires were aligned using Sequencher software and any sequence variant (single nucleotide polymorphisms, in/del, etc) was noted. Sequence variants were confirmed by 1 ) sequencing the sire PCR products in the opposite direction and 2) sequencing PCR products amplified using genomic DNA from the grandparents to verify Mendelian inheritance.

All confirmed, potentially functional DNA variants were genotyped (Tables 4-5). In addition, wherever possible, two confirmed single nucleotide polymorphisms (SNPs) for each gene were genotyped so that each candidate gene would have 3 possible haplotypes. SNPs were genotyped by amplifying approximately 200 bases flanking the SNP and using primer extension (Applied Biosystems) to discriminate the SNP alleles by fluorescence polarisation.

TABLE 4 - Sequence context of the candidate gene SNPs

Where SNPs are denoted by IUB single letter codes: K = G/T, M = A/C, R = A/G, S = C/G, W = A/T, Y = C/T

² This column indicates the nucleotide base at the SNP which is associated with low net feed intake

An association (linkage disequilibrium) analysis was performed in the Davies herd by a number of different methods. Initially, all 38 SNPs were coded as factors with 3 levels (e.g. aa, ab, bb) and included in a mixed model across all animals with daily feed intake and net feed intake records. In addition, when there were 2 SNPs per gene, haplotypes were formed and tested to see if they were more informative (as expected) than single SNPs (Tables 6-7).

One of the important parameters to estimate is the variance accounted for each SNP. This is a function of both the size of effect (α) and the allele frequency (p) as follows:

Variance = 2p(1-p)α²

The advantage of using the mixed model for testing SNPs is that the variance is estimated, indicating that both the size of effect may be useful and the gene frequency intermediate (Tables 6-7). Both these criteria must be met if the SNP is to be commercially valuable. If the SNP accounts for little variance, it converges to zero and is automatically not included in the final model. However, these variance estimates are poorly estimated with only 3 classes and for a number of the SNPs, there are unequal numbers across the genotypes. An alternative would be to include the SNPs as fixed effects, but this approach results in a model that is highly parameterized with possibly overestimated size of effects and, thus, was not used.

In addition to examining single SNPs as genotypic variances, when there were 2 SNPs per gene, haplotypes were formed and tested to see if they were more informative (as expected) than single SNPs. This was done by testing whether there was variance associated with the interaction between two SNPs in additional to variance associated with their main effect. TABLE 5 - Single nucleotide polymorphism discovery for association tests

# Genes E_HflUa # Potentially functional SNPs

TABLE 6 - Variances for SNPs when fitted as genotype or additive effects on daily feed intake

^aOnly non-zero variances presented.

The genotype effects for NFI were similar, although not the same as for daily feed intake.

TABLE 7 - Variances for SNPs when fitted as genotype effects on net feed intake

^aOnly non-zero variances presented.

EXAMPLE 6 Additive gene effects (breeding value) on net feed intake

The SNP genotype information may be used to estimate breeding values. Thus, the genotypes can be coded as covariates with values -1 , 0 or 1 representing aa, ab and bb, respectively. Fitting the SNPs in this form either saves a degree of freedom for each SNP (if fixed) or minimizes bias (if random). Results from this analysis are presented in Table 8.

TABLE 8 - Variances and effects for SNPs when fitted as additive effects on net feed intake

Only non-zero variances presented. ^bAII effects reported as positive but could be selected in either direction. The effect is for a single copy of the allele, so the difference between homozygotes is double this value. ^*Based on a simple t-test of significance of difference from zero, P<0.05.

Most genes that were associated with variation in net feed intake had similar variance estimates when fitted as a genotype (Example 5, Table 7) or additive effect (Table 8). The exception was IL12A. It is assumed that this exception arose because there were only 6 individuals with one of the homozygous genotypes, and the lack of balance was less severe when fitting as an additive effect. If the actual allelic effect was estimated and based on a simple t-test, three SNPs (UMPS, MAPI B and PRKAA1 ) also had a statistically significant effect on net feed intake.

EXAMPLE 7 Linkage mapping with SNPs in candidate genes

In addition to the association studies, confirmed SNPs were selected as markers for additional linkage analysis in the QTL regions. The SNPs markers were chosen such that 3 haplotypes/gene were present and, generally, 3 - 4 genes were informative in the sire of interest for that QTL (Table 9). Thus, two

SNPs per gene were chosen for both linkage analysis and association testing on BTA 1 , 8, 1 1 and 20. The lack of genes mapped on BTA11 resulted in fewer informative genes for this chromosome.

Further, some SNPs were likely to be functional variants. That is, the change in the DNA sequence would be likely to affect the expression or activity of the gene product. All functional SNPs were also genotyped in the gene mapping progeny. The SNP markers were placed on linkage maps using CRIMAP and positions verified where possible from other sources (eg USDA MARC 2004 map). TABLE 9 - Number of genes containing SNPs included in linkage analysis

¹ Genes heterozygous for sires 361 , 368, and 398

The results of mapping the candidate gene SNPs verified the QTL identified by microsatellite only linkage analysis (Table 10, Figure 2). The significance (F- value) of the QTL and the size of effect also increased with the addition of the candidate genes to the linkage mapping.

TABLE 10 - QTL mapping results for net feed intake with and without SNP data.

iHRϋlfl

EXAMPLE 8

Differential enzyme (AMPK) activity and oxidative phosphorylation between high and low efficiency Angus cattle

Biochemical and proteomic experiments were conducted to determine if high and low efficiency animals differ in these pathways, and thereby, validating candidate genes in these pathways. The initial experiments focused on two of these pathways, oxidative phosphorylation and malonyl coenzyme A / long chain fatty acid synthesis.

To examine the malonyl coenzyme A / long chain fatty acid pathway, the activity of a key rate-limiting enzyme in the pathway, AMP-activated protein kinase (AMPK), was studied. AMPK is a crucial regulator of fat metabolism and a trigger for feed intake. AMPK is primarily activated by changes in the cellular AMP:ATP ratio. When activated by AMP, AMPK increases ATP concentration by phosphorylating downstream peptide substrates, such as acetyl CoA carboxylase (ACC), which subsequently changes the energy balance.

ATP + peptide AMPK P-peptide + ADP AMP*

The AMPK gene, PRKAA 1, is located within the net feed efficiency QTL on BTA20 in cattle. Hence, PRKAA 1 may be a gene involved in determining the feed intake and/or net feed intake in bovine animals. Therefore, in addition to investigating the association between net feed efficiency and SNPs within the PRKAA 1 gene, the activity of the enzyme was measured in liver samples from high and low efficiency animals

Muscle and liver samples were collected from Angus animals at the abattoir and frozen immediately in liquid nitrogen. The samples were stored at -80⁰C until analysed. The samples came from animals in the Angus Elite Progeny Testing Program, a joint venture between Angus Australia and Meat and Livestock Australia (MLA). The animals were derived from the Trangie Angus net feed intake selection lines and measured for net feed efficiency at Tullimba. The animals differed in net feed efficiency by up to 6 kg feed/day. Samples from the 20 most extreme high and low efficiency animals were used in the assays to ensure the maximum physiological difference.

Frozen muscle and liver samples from these animals were used to prepare the cytosol and mitochondria. The cytosol (for the AMP-activated protein kinase assay) and the mitochondria (for the complex I, complex Il and complex IV assays) were prepared using standard protocols. The mitochondrial preparations also were used in the proteomic experiments.

In the cell, acetyl-CoA carboxylase and other proteins are phosphorylated by AMPK. However, a synthetic peptide (SAMS) is a specific substrate for AMPK and can be used to measure AMPK activity in vitro in the presence of AMP. In the absence of additional AMP, there is no specific activation and only background activity is observed. Therefore, AMPK enzyme activity was determined as the level of phosphorylation measured by using ³²P-ATP to label SAMS in the presence and absence of AMP.

The activity of the AMPK enzyme is increased in the highly efficient animals (p < 0.05, Figure 3). None of the SNPs discovered in the PRKAA1 gene are likely to be functional DNA variants that could explain this difference observed in activity. Nevertheless, the analysis of the SNP data from the PRKAA1 gene did indicate an association between AMPK and net feed intake.

To examine the relationship between oxidative phosphorylation and net feed intake, three of the mitochondrial complexes (complex I, complex Il and complex IV), involved in oxidative phosphorylation, were assayed. The same set of liver and muscle samples from the Trangie Angus progeny test high and low efficiency animals were used. For the complex I assay, the activity of the NADH-ubiquinone oxidoreductase/NADH dehydrogenase was measured spectrophotometrically by following the reduction of 2,6-dichlorophenolindophenol (DCIP) by NADH at 600nm absorbance. The sensitivity of the assay was verified by using the complex I inhibitor rotenone.

For the complex Il assay, the activity of the succinate-ubiquinone oxidoreductase/succinate dehydrogenase was measured spectrophotometrically by following the reduction of 2,6- dichlorophenolindophenol (DCIP) by decylubiquinone (ubiquinone analogue) at 600nm absorbance. The sensitivity of the assay was verified by using complex Il inhibitors (malonate and thenoyltrifluoroacetone).

For the complex IV assay, the activity of the ferrocytochrome c oxygen oxidoreductase/cytochrome c oxidase was measured spectrophotometrically by following the oxidation of reduced cytochrome c at 550nm. The sensitivity of the assay was verified by using the complex IV inhibitor potassium cyanide.

The activity of these three respiratory enzyme complexes involved in oxidative phosphorylation (complex I, complex Il and complex IV) were examined in the liver and muscle mitochondria of these low and high intake animals. The data indicated that although there is no difference in the activity of complexes Il and IV, the activity of the rate-limiting complex I does differ by 1.5-fold between liver mitochondria of low and high intake animals (p < 0.02) (Figure 5). Two genes that code for subunits in complex I (NADH dehydrogenase (ubiquinone) 1 -alpha subcomplex, 5, 19kDa and NADH dehydrogenase (ubiquinone) 1- alpha subcomplex, 10, 75kDa) are located within the QTL for net feed intake on chromosomes 1 1 and 1 , respectively. EXAMPLE 9 Differential protein activity between high and low efficiency Angus cattle

To have a more global understanding of the potential differences in oxidative phosphorylation between high and low efficiency animals, a proteomic approach was also taken. Proteomics involves using 2D gel electrophoresis to separate proteins based on charge and mass by iso-electrical focusing. A differential gel electrophoresis (DIGE) methodology was applied. The steps for each experiment include:

1. Sample preparation and fluorescent labelling

2. First dimensional separation

3. Equilibration

4. Second dimensional separation 5. Scanning

6. Image analysis

7. Protein identification

The same mitochondrial preparations from the high and low net feed efficiency Trangie Angus test progeny were used. 50 μg (as measured by the Bradford assay) from each sample was used for each 2-D gel. The proteins were dissolved under denaturing conditions (urea and detergents) and labelled with either Cy3 or Cy5 for samples from high and low efficiency animals. The dyes were then swapped for the samples from high and low efficiency animals in a second experiment. Cy2 is used to label a pool of all samples as a standard.

The proteins are then separated in an electrical field according to their isoelectric points (pi values) using pre-cast immobilised-pH-gradient (IPG) gel strips and carrier ampholytes. The proteins migrated to a position where their net charge is 0. The focusing range was wide (pH 3-1 1 ), so resolution was approximately 0.5 pH units. The gels are then scanned by densitometry to detect differences in quantity of specific proteins between the samples based on the spot intensity. The spots representing proteins that varied by more than 1.5-fold concentration between the samples from high and low efficiency animals were located. After this iso- electrical focusing, the Cy3 and Cy5 intensities of the protein spots were measured to determine the relative quantity of each protein in the high versus low efficiency animals (Figure 5).

These proteins were excised from the gel, digested and identified using mass spectrometry. Analysis of the gels indicated that out of the some 550 mitochondrial proteins detected, 27 of the proteins differed in concentration by more than 1.5-fold between the high and low efficiency animals (Table 12). These proteins have been identified and at least 5 of the proteins are subunits of the oxidative phosphorylation complex I, 1 is a subunit of complex II, 2 are subunits of complex III, and 3 are subunits of complex V. Three of the proteins are coded by genes within the QTL regions defined by the present invention (Table 11 ).

TABLE 1 1 - List of genes coding for differentially expressed proteins within QTL chromosomal regions

TABLE 12 - List of differential expressed mitochondrial proteins in high and low net feed efficiency cattle.

Protein iverage IaILMJMJ BTA ccession ratio³ position number

Oxidative phosphorylation

Complex I

NDUFA10 gil74267974 -1.32 Involve in electron BTA 3, transfer chain and 85 Mb cellular energy production

EXAMPLE 10

Marker Assisted Cattle Selection

Once a particular sequence variant, for example a sequence variant of a growth hormone receptor gene, is identified as being associated with net feed intake, a population screen is conducted. In this screen, a large number of animals representing many breeds are genotyped and the magnitude of effect of the sequence variation is determined. When a large magnitude of effect is observed, then the sequence variant may be used as a marker for marker assisted selection, for example as described by Georges {Theriogenology 55(1 ): 15-21 , 2001 ) or Dekkers and Hospital (Nature Reviews Genetics 3(1 ): 22-32, 2002). Animals selected by genotyping markers for net feed intake may be used for breeding or managed to enhance production (for example in a feedlot).

Those skilled in the art will appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. It is to be understood that the invention includes all such variations and modifications. The invention also includes all of the steps, features, compositions and compounds referred to, or indicated in this specification, individually or collectively, and any and all combinations of any two or more of the steps or features.

Also, it must be noted that, as used herein, the singular forms "a", "an" and "the" include plural aspects unless the context already dictates otherwise. For example, reference herein to "a quantitative trait locus" or "a QTL" refers to a single quantitative trait locus or QTL or two or more quantitative trait loci or QTL; similarly reference to "a SNP marker" refers to a single SNP or two or more SNPs; and so forth.

Claims

THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:

1 . A method for identifying a genomic nucleotide sequence associated with a particular level of feed intake and/or net feed intake in a bovine animal, the method comprising: selecting a bovine animal having a known level of feed intake and/or net feed intake; and determining a genomic nucleotide sequence of the animal at a QTL on one or more of bovine autosomes 1 , 8, 1 1 and 20; wherein the genomic nucleotide sequence at the QTL is associated with the level of feed intake and/or net feed intake of the bovine animal.

2. The method of claim 1 wherein the QTL is on bovine autosome 1.

3. The method of claim 2 wherein the QTL is a region of bovine autosome 1 that is delimited by microsatellite marker CSSM032 and microsatellite marker BMS2263.

4. The method of claim 2 wherein the QTL is a genomic region proximal to microsatellite marker BM 1824 on bovine autosome 1.

5. The method of any one of claims 2 to 4 wherein the QTL comprises one or more of: the alpha-HS-glycoprotein gene (ASHG), the glucose transporter 2 gene (SLC2A2), the growth hormone factor 1 /pituitary specific positive transcription factor 1 gene (P0LJ1 F1 ), the interleukin 12A gene (IL12A), the uridine monophosphate synthetase gene (UMPS), or the NADH dehydrogenase (ubiquinone) 1 -alpha subcomplex, 10, 75kDa gene.

6. The method of claim 1 wherein the QTL is on bovine autosome 8.

7. The method of claim 6 wherein the QTL is a region of bovine autosome 8 that is delimited by microsatellite marker BMS1341 and microsatellite marker BMS836.

8. The method of claim 6 wherein the QTL is a genomic region proximal to the LPL gene and/or microsatellite marker BMS1341 71 1 on bovine autosome 8.

9. The method of any one of claims 6 to 8 wherein the QTL comprises one or more of: the cathepsin B gene (CTSB), the lipoprotein lipase gene (LPL) and the endothelial tyrosine kinase gene (TEK).

10. The method of claim 1 wherein the QTL is on bovine autosome 11 .

1 1 . The method of claim 10 wherein the QTL is a region of bovine autosome 1 1 that is delimited by microsatellite marker RM096 and microsatellite marker RM363.

12. The method of claim 10 wherein the QTL is a genomic region proximal to microsatellite marker BM 1048 on bovine autosome 1 1.

13. The method of any one of claims 10 to 12 wherein the QTL comprises one or more of: the argininosuccinate gene (ASS), the follicle stimulating hormone receptor gene (FSHR) and the proopiomelanocortin (POMC) gene, the lnterleukin 1 β (IL1 B) gene, the or the NADH dehydrogenase (ubiquinone) 1- alpha subcomplex, 10, 19kDa gene or the ATPase, V1 subunit G2 gene.

14. The method of claim 1 wherein the QTL is on bovine autosome 20.

15. The method of claim 14 wherein the QTL is a region of bovine autosome 20 that is delimited by microsatellite marker TGLA126 and microsatellite marker BM5004.

16. The method of claim 14 wherein the QTL is a genomic region proximal to microsatellite marker BMS703 on bovine autosome 20.

17. The method of any one of claims 14 to 16 wherein the QTL comprises one or more of: the 5'-AMP-activated protein kinase, catalytic alpha-1 chain gene (PRKAA1 ) gene, the follistatin precursor gene (FST), the growth hormone receptor (GHR) gene, the microtubule-associated protein 1 B (MAPI B) gene, the phosphatidylinositol 3-kinase P-85-alpha subunit (PI3KR1 ) gene, or the prolactin receptor gene (PRLR).

18. The method of any one of claims 1 to 17 wherein the genomic nucleotide sequence is associated with low feed intake and/or low net feed intake in a bovine animal.

19. A genomic nucleotide sequence identified in accordance with the method of any one of claims 1 to 18.

20. A method for predicting the feed intake and/or net feed intake of a bovine animal, the method comprising determining a genomic nucleotide sequence of one or more cells of the bovine animal at a QTL on one or more of bovine autosomes 1 , 8, 1 1 or 20, wherein the genomic nucleotide sequence at the QTL is indicative of the feed intake and/or net feed intake of the bovine animal.

21 . The method of claim 20 wherein the QTL is on bovine autosome 1.

22. The method of claim 21 wherein the QTL is a region of bovine autosome 1 that is delimited by microsatellite marker CSSM032 and microsatellite marker BMS2263.

23. The method of claim 21 wherein the QTL is a genomic region proximal to microsatellite marker BM1824 on bovine autosome 1.

24. The method of any one of claims 21 to 23 wherein the QTL comprises one or more of: the alpha-HS-glycoprotein gene (ASHG), the glucose transporter 2 gene (SLC2A2), the growth hormone factor 1 /pituitary specific positive transcription factor 1 gene (POU1 F1 ), the interleukin 12A gene (IL12A), the uridine monophosphate synthetase gene (UMPS), or the NADH dehydrogenase (ubiquinone) 1 -alpha subcomplex, 10, 75kDa gene.

25. The method of claim 20 wherein the QTL is on bovine autosome 8.

26. The method of claim 25 wherein the QTL is a region of bovine autosome 8 that is delimited by microsatellite marker BMS1341 and microsatellite marker BMS836.

27. The method of claim 25 wherein the QTL is a genomic region proximal to the LPL gene and/or microsatellite marker BMS1341 71 1 on bovine autosome 8.

28. The method of any one of claims 25 to 27 wherein the QTL comprises one or more of: the cathepsin B gene (CTSB), the lipoprotein lipase gene (LPL) and the endothelial tyrosine kinase gene (TEK).

29. The method of claim 20 wherein the QTL is on bovine autosome 1 1.

30. The method of claim 29 wherein the QTL is a region of bovine autosome 1 1 that is delimited by microsatellite marker RM096 and microsatellite marker RM363.

31 . The method of claim 29 wherein the QTL is a genomic region proximal to microsatellite marker BM 1048 on bovine autosome 1 1.

32. The method of any one of claims 29 to 31 wherein the QTL comprises one or more of: the argininosuccinate gene (ASS), the follicle stimulating hormone receptor gene (FSHR) and the proopiomelanocortin (POMC) gene, the lnterleukin 1 β (IL1 B) gene, the or the NADH dehydrogenase (ubiquinone) 1- alpha subcomplex, 10, 19kDa gene or the ATPase, V1 subunit G2 gene.

33. The method of claim 20 wherein the QTL is on bovine autosome 20.

34. The method of claim 33 wherein the QTL is a region of bovine autosome 20 that is delimited by microsatellite marker TGLA126 and microsatellite marker BM5004.

35. The method of claim 33 wherein the QTL is a genomic region proximal to microsatellite marker BMS703 on bovine autosome 20.

36. The method of any one of claims 33 to 35 wherein the QTL comprises one or more of: the 5'-AMP-activated protein kinase, catalytic alpha-1 chain gene (PRKAA1 ) gene, the follistatin precursor gene (FST), the growth hormone receptor (GHR) gene, the microtubule-associated protein 1 B (MAPI B) gene, the phosphatidylinositol 3-kinase P-85-alpha subunit (PI3KR1 ) gene, or the prolactin receptor gene (PRLR).

37. The method of any one of claims 20 to 36 wherein the genomic nucleotide sequence at the one or more QTL, which is indicative of the net feed intake of the animal, comprises a genomic nucleotide sequence identified according to the method of any one of claims 1 to 18.

38. The method of any one of claims 20 to 37 wherein the method further comprises identifying a bovine animal having a net feed intake level of interest on the basis of the predicted net feed intake.

39. The method of claim 38 wherein the bovine animal identified has a low feed intake and/or low net feed intake.

40. A method of producing a bovine animal with a desired level of feed intake and/or net feed intake, the method comprising the steps of: selecting a first bovine animal comprising a genomic nucleotide sequence that is associated with a desired level of feed intake and/or net feed intake at a QTL on one or more of bovine autosomes 1 , 8, 1 1 or 20; sexually or asexually reproducing the animal; and selecting progeny arising from the sexual or asexual reproduction which comprise the genomic nucleotide sequence associated with the desired level of feed intake and/or net feed intake at the QTL.

41 . The method of claim 40 wherein the selected bovine animal is identified according to the method of claim 38 or 39.

42. The method of claim 40 or 41 wherein the selected bovine animal has low feed intake and/or low net feed intake.

43. A bovine animal, wherein the bovine animal is progeny of the breeding method of the method of any one of claims 40 to 42.

44. Use of the bovine animal of claim 43, or reproductive material therefrom, for breeding.

45. A method for modulating the overall feed intake and/or overall net feed intake of a herd of bovine animals, the method comprising: selecting bovine animals for breeding within the herd, wherein the bovine animals selected for breeding comprise a genomic nucleotide sequence that is associated with a desired level of feed intake and/or net feed intake at a QTL on one or more of bovine autosomes 1 , 8, 1 1 or 20; and breeding the herd from those selected animals which comprise the genomic nucleotide sequence associated with the desired level of feed intake and/or net feed intake at the one or more QTL.

46. The method of claim 45 wherein one or more of the selected bovine animal is identified according to the method of claim 38 or 39.

47. The method of claim 45 or 46 wherein one or more of the selected bovine animals has low feed intake and/or low net feed intake.

48. A herd of bovine animals having decreased overall net feed intake and/or decreased overall feed intake, wherein the herd is produced according to the method of any one of claims 45 to 47.