AU2002301557A1 - Methods for Analyzing Animal Products - Google Patents

Description

-1-

AUSTRALIA

PATENTS ACT 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT

ORIGINAL

Name of Applicant/s: Pig Improvement Company UK Limited Actual Inventor/s: Leif Andersson and James Kijas and Elisabetta Giuffra and Gary Jon Evans and Richard Wales and Graham Stuart Plastow Address for Service: BALDWIN SHELSTON WATERS MARGARET STREET SYDNEY NSW 2000 CCN: 3710000352 Invention Title: Methods for Analyzing Animal Products Details of Original Application No. 76646/98 dated 27 May 1998 The following statement is a full description of this invention, including the best method of performing it known to me/us:- File: 26123AUP01 la- METHODS FOR ANALYZING ANIMAL PRODUCTS The present application is a divisional application of Australian Application No.

76646/98, which is incorporated in its entirety herein by reference.

The present invention relates to methods for analysing animals and their products.

In particular, the invention relates to methods for differentiating animal products on the basis of breed origin, determining or testing the breed origin of an animal product and for validating an animal product, as well as to kits for carrying out such methods. In addition, the present invention provides methods for the determination of pig genotype with respect to coat colour.

Introduction Animal breeds For thousands of years, selective pressure has been applied by humans in the course of animal husbandry to produce livestock exhibiting certain desirable characteristics. These characteristics have been selected to meet aesthetic, technical, ritual, social and economic needs. The result has been the production of a large number of different animal breeds.

The term "breed" is a term of art used to define a homogenous, subspecific group of domestic livestock with definable and identifiable external characteristics that enable it to be separated by visual appraisal from other similarly defined groups within the same species. The term therefore defines a group of animals to which selective pressure has been applied by humans to give rise to a uniform appearance that is inheritable and distinctive with respect to other members of the species.

As breeds become established, their integrity is maintained by breed societies, herdbooks and pedigree records.

Breed selection Conventional breed selection methods are based on direct measurement of the phenotype of an animal and/or its relatives. Thus, the implementation of breeding.

2 schemes requires extensive phenotypic record keeping. For example, dairy herd improvement programs in the United States and Western Europe relied in part on the collection of individual records (milk yield and composition, type traits, health traits, etc.) performed on a monthly basis for millions of cows. Likewise, breeding companies carefully monitor their pig and poultry breeding stock for a whole range of phenotypic measurements.

However, some important characteristics are not immediately apparent at the level of the living animal. For example, many parameters of meat quality are determined by subtle physiological or biochemical characteristics which are not readily apparent and so cannot serve as the basis for efficient artificial selection.

Breeding for qualities of this type has relied in part upon selection for other (more readily apparent traits) which are to some extent coinherited (linked or associated) with the desirable characteristics. For example, in the pig industry lop ears have in the past been associated with mothering ability and so have been used as a marker for this trait Conventional breed selection methods are limited by the fact that some phenotypes are expressed only in one sex or at a specific developmental stage. Moreover, some phenotypes are difficult and costly to measure. Indirect detection of such phenotypic traits via DNA-based diagnosis (for use in marker-assisted selection or MAS) is therefore seen as a desirable alternative to direct measurement of phenotypic parameters (see Georges and Andersson (1996), Livestock genomics comes of age, Genome Research, VoL 6: 907-921). However, the gene sucture-function relationships underlying many of the desirable traits are often highly complex and not yet sufficiently well-established to make such an approach feasible in practice.

Breed identification The definition of animal breeds is currently at a watershed Whereas previously they have been defined by overt physical characteristics and pedigree records, in the future as new breeds are developed from specific breeding lines they will be defined by sets of DNA markers. The work described herein allows not only the most accurate approach 3 to breed determination currently possible in a range of products but also allows the integration in a common format of breed determinant information obtained through use of the present invention with that which will be used in the future. The present invention therefore allows not only the determination of source breed in the current environment but also links this to the development of future breeds and their unique identification- It is generally recognized that the only definitive way to identify a particular animal as a representative of a given breed is through its pedigree. Thus, despite the fundamental importance of overt phenotypic traits in the breeding process and in the maintenance of breed purity, those skilled in the art generally consider that breed identity cannot be definitively characterized on the basis of visual inspection of such traits. By way of example, the genetic factor causing the belt phenotype in pigs is dominant to the nonbelted form- Thus, a belted animal may result from an animal of a belted breed such as Hampshire being crossed with a non-belted breed.

As stated in PIGS A handbook to breeds of the world, V. porter, Helm mInf. Ltd, ISBN 1- 873403-17-8, 1993, page 16, "What is a Breed?": Appearances can be deceptive: never judge apig breed by its coat! However, in many circumstances breed identification on the basis of direct evidence of pedigree is difficult or impossible. Thus, in practice, so-called "breed markers" may be used to determine breed identity.

The term "breed madrker" is a term of art which defines a measurable characteristic which on the basis of empirical data appears to be breed specific. Breed markers include genotypic features such as DNA polymorphisms, chemical features such as protein and water contents of meats, epigenetic/biochemical features (such as protein polymorphisms), chromosome structure, gene copy number, DNA fingerprinting, microsatellite analysis and RAPD DNA markers.

4 Other useful markers include breed determinants. The term "breed determinant" is used herein to indicate an overt phenotypic characteristic which is used (at least in part) as the basis of artificial selection during breeding programmes. It is used in contradistinction to the term "breed marker", which (as explained above) is used herein to define other characteristics which appear to be breed specific on the basis of empirical data. The term "breed determinant gene" is used to indicate a gene which is involved (at least in part) in the expression of the corresponding overt phenotypic Some breed determinants coat colour) have traditionally been used as breed "trademarks", and so have long served as an indication of pedigree (and breed identity).

Other breed determinants that have also been selected for in breed development include features such as ear carriage, face shape and general anatomical conformation The advantage of breed determinants relative to simple breed markers is the inseverable link between the characteristics of the breed and the determinant Biochemical and enetic tests for breed identity Many of the breed markers discussed above can be characterized using biochemical or genetic tests. Such markers include genotypic features DNA polymorphisms), biochemical features protein polymorphisms), chromosome structure, gene copy number, DNA fingerprints, microsatellite patterns and RAPD DNA markers.

However, there are significant problems associated with such tests, as discussed below: Tests based on the chemical composition of animal products meat or seminal plasma) may be compromised by the fact that the chemical profile varies between sites in the animal (ie different muscles)and is affected by diet, age, sex and sample storage conditions. Moreover, the results obtained are usually quantitative in nature, leading to problems with interpretation and comparison between different test sites.

Tests based on protein polymorphisms are limited by the fact that the distribution of any given protein is unlikely to be uniform, so that the protein of interest is absent in certain tissues. Thus, a number of different polymorphism markers may be required to check all products of interest for breed provenance. Moreover, such tests are based on antibody assays, and a significant investment is also required to -develop the reagents Ibr a specific antibody test Chromosome structure analyses are compromised by the high level of skcill required for cytogenetical methodology and interpretation and the elaborate precautions and care required for sample preservation. Such markers are poorly applicable on anything: but materials derived from living or newly rdeceased animals Classical DNA fingerprinting is based upon regions of repeated DNA sequence that due to their structure show a large degree of variation in length within a. population. Such regions are ofien present in a number of copies within the DNA of an individual, thus increasing the potential for individual variation. By separating fr-agments of the total DNA according to size and then defining the position (and so size) of the hypervariable region using a specific probe, a fingerprint of a series of bands for a particular individual can be obtained. A number o f probes for hypervariable regions of DNA have been examined in pigs (including M13 viral sequences and human minisatelite probes) and it is claimed that specific bands were found in each breed.

Random Amplified Polymorphic DNA (RAPD) markers are based upon PCR amplification of DNA fr-agments using primers of random sequence. Such reactions generally give rise to a number of DNA fragments which can. be characterised according to size by gel electrophoresis. If the products of reactions based upon DNA 2 5 from different breeds are examined there is the possibility of finding certain DNA bands which are breed specific. However, there is in most cases no direct link betwee the, alleles of such repeat series present and the features determining the actual nature of the breed. This, combined with the hypervaiiahle nature of these regions of DNA, results in them rarely being breed specific (similar alleles being found in a number of different 3 0 breeds). As there i's no link to the phenotype of the breed there is a greater risk that cross specific alleles could exist or arise in a breed, whereas this is unlikely with breed determinants as they define the phenotype itself Given the large number of populations of animals of specific breeds that exist extensive research would have to be carried out to exclude a DNA marker from breeds other than that with which it is claimed to-be' linked.

However, a major drawback with this approach is that RAPD markers are considered to be unreliable and found to be subject to variation between laboratories. Such problems are exacerbated when samples of different types and history must be analyzed and compared.

There is therefore a need for reliable breed markers which can be used as the basis for rapid and inexpensive methods for identifying the breed provenance of various animal products and for validating animal products (such as foodstuffs and semen for use in breeding programmnes).

It has now been recognized that breed determinants as hereinbefore defined (such as coat colour) have unexpected advantages as breed identifiers or breed specific markers.

In particular, it has surprisingly been discovered tha the use of overt phenotypic characteristics as the basis for selection over long periods of time has led to particular alleles becoming fixed in most breeds. Such breed markers can be used to provide 2 0 industry standard profiles for a particular breed that has application to all materials derived from a particular species.

Thus, it has now been found that many breeds are in fact genetically homogenous with respect to breed determinant genes (as hereinbefore defined), so thLat these genes may 2 5 serve as the basis of reliable breed-specific markers (contrary to the prejudice in the art mentioned earlier regarding the utility of breed determinants per se, such as coat colour, in breed identification).

Moreover, it has surprisingly been found that the nature of the breed determinant genes 3 0 (or alleles thereof) underlying any one breed determinant (such as coat colour) may be highly polymorphic. Thus, variation in breed determinant genes and/or alleles between different breeds may exist, notwithstanding the fact that the different breed determinant genes/alleles may contribute to the expression of the same overt phenotypic characteristic.

Prior to the present invention, it was assumed that" the corresponding genetic s determinants would be insufficiently polymorphic to provide a useful basis for distinguishing between breeds. For example, coat colour was known to be shared among different breeds of pig and (as mentioned above) was therefore not regarded as a good candidate for a breed specific marker. However, the present inventors have found that the alleles underlying the coat phenotype in such breeds are in falct highly polymnorphic and often distinctive (and so useful as the basis for breed identification).

Similar considerations apply to other overt physical traits (breed determinanits), which may therefore be shared by different breeds while nevertheless associated with distinct genes/alleles in each breed. An example of this is seen in cattle exhibiting the double 315 muscled phenotype. Work by Kambadur etcilia (1997, Genome Research 7, 910-915) and Grobet el alia (1997, Nature Genetics 17, 71-74) illustrates that the double muscled phenotype of cattle is caused by mutations in the myostatin gene. However, in the Belgian Blue and Asturiana breeds, this gene contains an I11 bp deletion whereas in the Piedmiontese breed a G to A transition is present. Thus, as with porcine coat colour a 2 0 single selected characteristic is caused by a number of potential polymorphisms.

However, the nature of the arisal and selection history for such overt physical characteristics leads to the fixation of particular alleles within the breeds contributing to the breed specific profile of determinants.

In the light of these findings, it has now been recognized that genetic analysis of breed determinants (such as coat colour) provides an effective means for validating animal prdcs(e.g. foodstuffs) and may advantageously be incorporated into animal product food) processing lines to monitor and maintain product quality and into quality control protocols in the food industry.

Coat colour Pig breeds show a variety of coat colours and these are often associated with 8 particular production characteristics. For exanmple, white is the predominant coat colour among European commercial breeds e.g. Large White and Landrace, and these breeds are associated with larger litters and good mothering ability. However, there are a number of commercially important coloured breeds, demonstrating a number of colours and combinations. The Duroc, associated with meat tenderness, is red, the Pictrain, a heavily muscled animal which produces a very lean, carcass, is spotted, and the Hampshire, also heavily muscled, is black with a white saddle over its shoulders.

In addition, there may be other useful local breeds which have traits of potential commercial interest, and which arc coloured. For example, the Chinese Meishan breed has been imported into Europe and the US because of its very large litter size.

The European Wild Boar is brown when adult and striped when juvenile, and this breed is utilised to satisfy consumer demand for traditional meat products. It is also claimed that other local breeds or landraces are important because of their adaptation to local environmrents, e.g. temperature, endemic diseases and local feedstuffs.

Coat colour is important to the pig industry for a number of reasons. Firstly, gross variation in appearance a range of coat colours) of pigs claimed to be genetically consistent for traits other than coat colour can lead to questions about the consistency and quality of t he animals in the mind of pig-producers. Thus, the coat colour of the 2 0 pig is often used as a trademark of the breed and the breeders want to ensure tbat their animals br~ed true for colour. For example, in several. markets, local, traditional, coloured breeds are marketed for their meat quality or in terms of the production system used to rear tenm- However, this is not a trivial task since the coat colour is controlled by a number of genes. The inheritance is also complicated by the presence 2 5 of dominance and interaction between genes. There is also an application in the assessmnt of the purity of the genetics of traditional breeds used as the basis for modern synthetic lines and the confirmation of the derivation of the latter.

Secondly, in a number of markets there is a preference for white skinned meat This 3 0 is due to the fact that pork is often marketed with the skin still attached, and sin from coloured pigs, even if dehaired, can still exhibit colour, which can lead to negative perception by the consumer partly, since the surface of the meat may appear to be spotted by mould. It is therefore necessary in these markets to remove the skin from such carcasses, entailing additional cost. For example, in the US, coloured carcassesare associated with approximately 1% skin defects requiring dehairing and skinning to remove pigment. As a result of this, coloured pig carcasses are generlly d iscounted.

One example of the problem concerns the presence of black pigmented spots occurring in production animals that are crossbreds between a white and a pigmented line. This may -occur because the dominant white gene inherited from the white breed is not always fully dominant in the heterozygous. condition which occurs in this cross.

A possible solution to this problem would be to ensure that the production animals are homozygous for the recessive red allele present in breeds such as the Duroc. In this case the pigmented spots would be red instead of black and much less conspicuous. To achieve this one needs to breed the recessive red allele to homozygosity in both the white and pigmented line used for cross-breeding.

However, this would be very difficult using phenotypic selection as selection for a red background colour in a white line could only be accomplished with very expensive progeny testing schemes.

In addition, pig breeders would like to be able to be in a position to ensure consistency 2 0 in breeding populations. Breeders may wish to ensure that progeny produced by breeding croisses were always white. Alternatively, a breeder of Duroc or Hampshire pigs may wish to ensure that breeding crosses always produced the characteristic Duroc or Hampshire colouring. Traditional animal breeding practices have in the past, been used to attempt to eliminate untypical colour from pig lines. For example 2 5 purebred breeders must submit potential boars for progeny testing in order to demonstrate that they are suitable for inclusion in the breeding herd. This procedure incurs significant cost, including the substantial delay to conflnn sufficient matings and progeny have been produced before the animal can be used commercially.

Therefore, selection based on a diagnostic DNA test for mutations in coat colour genes would be a major advance compared with phenotypic selection.

Coat colour is determined by the action of a number of different gene loci. For example, the gene determining whether a pig is white or coloured is designated I (for inhibition of coat colour). The version of the gene preventing the expression of any colour (1I) is dominant to that which allows colour to develop Traditional selection for white animals has reduced the frequency of i, but it still remains in the population of white heterozygous carrier animals. Recently, a number of structural differences in the alleles of the KIT gene were identified and found to be involved with this aspect of coat colour determination which allowed the development of methods of distinguishing between alleles at this locus.

However, animals which carry two copies of the recessive allele, i, at this locus have non-white coat colours (Johansson-Moller et al., Mamm Genome, 7:822-830 (1996), WO-A-97/05278, the disclosure of which is incorporated herein by reference). Pigs of this type can be all one colour, such as the Duroc (which is red), or have combinations of colours (particularly spotted or striped or banded patterns, such as the Pietain and Hampshire, respectively). Many other combinations are possible and are observed (see the table, below): Genotype Colour 1/I White li White i/i Coloured f/f White with coloured spots f/li White F/i White with coloured patches 2S The non-white colour in such animals may be varying shades of red or black. The type of colour expressed is determined by the action of a second gene which is designated E (for extension of coat colour). Based on the literature, animals which contain the E version of the gene are completely black, and this version of the gene is dominant to that which results in red coat colour Patched or spotted animals, such as the Pietrain breed, contain a third version of the gene designated This version of the gene is dominant to e but not to E. For example, black animals may have the 11 genotype, UEe, iiEE P or iiEE. A black sow and a black boar which were both heterozygous at the E locus and which were of the genotype iiEe would produce both black and red piglets in the ratio 3:1. The black piglets would be iiEE or iiEe and the red piglets would be iee.

The density and coverage of coat colour and the position of bands of white are determined by additional loci, one of which, the belt locus, is discussed later in this application. For example, the Hampshire breed is background black with a white band across its shoulders, the width of this band may vary, however, the colour should be black. There is evidence that some Hampshire animals are derived from herds that have been crossed previously with red breeds, such as the Duroc. In this situation, the red version of the gene can be maintained silently in the heterozygous state. When two heterozygotes are crossed 25% of the offspring will contain red. In some cases such pigs will have the appearance of the Duroc breed, being solid red, however, in other cases, the animals will have the white band inherited from the Hampshire and have the appearance of red Hampshires. It is the presence of the atypical coat colour rather than the pattern that is important in this situation.

The extension locus is known in other breeds of domestic animali such as the horse, where e is associated with chestnut colour (Adalsteinsson, J. Hered 65:15-20 (1974)), cattle (Klungland et al., Mammalian Genome 6: 636-639 (1995)), the fox (Adalsteinsson, J Hered 78:15-20 (1987)) and the mouse (Jackson, Ann Rev'Genet. 28: 189-217 (1994)). The extension locus encodes the alpha melanocyte-stimulafing hormone receptor (aMSHR). It has been shown that recessive alleles at this locus do not express a fmctional aMSH receptor (Robbins et al, Cell, 72: 827-834, Klungland et al, Mammalian Genome 6: 636-639 (1995)) and these workers have identified mutations in the sequence of the aAESHR gene in these species associated with different coat colours.

3 0 Classical segregation analyses have identified a minimum of three alleles at the pig extension locus: E for uniform black, for black spotting and e for uniform red (Ollivier and Sellier, Ann. Gdndt. SdL Anita., 14:481-544, (1982)). The dominance 12 relationship among the three alleles is as follows E>i>e. We have now found that these coat colour variations are associated with sequence polymorphism in the aMSHR gene in the pig. We have analysed the DNA sequence of this gene using samples from the following breeds with different coat colour: Wild Boar which is wild type coloured, Meishan and Hampshire which carry alleles for uniform black Pietrain and Large White which carry alleles for black spotting (EP) and Duroc which is uniform red In Large White the patches or spots of colour that might be expected due to the presence of the if allele are hidden as this breed also carries the dominant white gene which prevents any expression of colour. Five different aMSHR sequences were obtained one from the Wild Boar, one from Meishan, one from Duroc one from Hampshire, and one found in Pietrain and Large White. We have designated the allele found in the Wild Boar as E and assume that the presence of this allele is necessary for the expression of the wild type colour. The E alleles for uniform black carried by Mcishan and Hampshire pigs were associated with different aMSHR IS sequences. We have denoted these two alleles E" and Eh respectively. The DNA sequence associated with the allele for black spotting found in the Pietrain was denotedif. The similarity of the A' and E' alleles suggests that they are derived from a common origin. The sequence differences presented here can be used as the basis of methods and kits to determine the genotype of pigs in relation to coat colour.

2 0 Alternatively, alleles of linked markers, such as microsatellite or AFLP markers, found to be 'n linkage disequilibrium with these alleles could be used to predict colour genotype. In conclusion we have found fivedifferen ctMSHR sequences associated with five different extension alleles i.e. E E E P and e.

Except for the 2 base pair insertion at the 5' end of the A P allele and the I bp deletion in the 3' untranslated region of E" allele the DNA sequence differences identified in the aMSHR gene are single base pair changes. Some of these are silent, however, a number lead to changes in the amino acid sequence of the aMSHR protein. For example, the differences between e and the other alleles are two missense mutations 3 0 in the coding sequence of the aMSHR gene. Importantly, the differences in the pig gene are different from that found in other species. The cattle and mouse e mutations are one base pair deletions (Robbins et al, Cell, 72: 827-834 (1993); Klungland et a!, 13 Mammalian Genome 6: 636-639 (1995), Joerg et al, Genome 7: 317-318 (1996)), whilst the mutations identified here include a missense mutation (G727 changed to A) in a region which is conserved among human, mouse, cattle and horse gene sequences (Wikberg et al, WO 94/04674 (1994), Valverde et al, Nature Genet. 11: 328-330 (1995), Robbins et al, Cell, 72: 827-834 (1993), Klungland et al, Mammalian Genome 6: 636-639 (1995), Joerg et al, Genome 7: 317-318 (1996), Markdund et al, Mamm. Genome, 7:895-899 (1996)). The ff group has a dinucleotide insertion in the end of the gene after nucleotide position 66 of the Wild Boar sequence which leads to the creation of a stop codon further into the gene resulting in a predicted mutant polypeptide of only 54 amino acids. Finally, the Meishan allele shows four amino acid changes in the protein. Two of these differences are in the same region of the gene which is altered in cattle.

The colours of a series of pig breeds, the classical genotypes for I and E and the determined genotypes for E based on sequencing and testing studies are shown in the table below: Breed Hampshire 2 0 Large White Landrace Pietrain Berkshire Meishan Duroc Wild Boar I locus i/i

I/I

I/I

i/i i/i i/i i/i A4 04 £4 E locus Colour EA/E Black with white belt Ef/F White Fl/Pf White F/fE P White with black patches f/F' Black with white points Black or black with white points e/e Red Brown (banded hair) 3 0 Thus, it is possible to distinguish between the alleles of r, E m E, and e and so determine the genotype of individual pigs (or the genetic provenance of products derived therefrom) with respect to non-white coat colour. Interestingly, the white 14 breeds that have been examined all appear to be fixed for alleles f at the E locus.

There is considered to be potentially some modifying effect of theE locus o ii-the

U-

phenotype conferred by the I locus. While the basis of this is not established, the fixing of L in these lines illustrates the subtle effects on loci involved with coat colour upon selection for breed characteristics thus providing more determinants among such loci than might be expected.

Associations can be determined between extension locus genotype and linked markers, eg microsatellite sequences which are linked to the gene. A number of microsatellite markers have been located to the region of porcine chromosome 6 to which the oMSHR gene has been mapped.

A number of pig breeds characteristically show the belt phenotype consisting of a continuous white belt over the shoulders and white fore legs. Examples of breeds demonstrating this characteristic are the British Saddleback (derived from the Wessex and Essex breeds) and the Hampshire, which show a white belt upon a black back ground and the Bavarian Landschwein characterised by a white belt upon a red background. The characteristic is controlled by a dominantly acting locus Belt designated Be for which there are thought to be two alleles (Legault 1997 in The Genetics of the Pig, Ed Rothschild M.F. and Ruvinsky A, Publ. CAB International) (Ollivier arid Sellier, Ann. Gdn&t. Sel. Anim., 14:481-544, (1982)). Be giving rise to a belt and be which in the homozygous form leads to the absence of a belt. The heterozygous animal Be/be carries a belt but in this genotype the belt is generally narrower in character.

To identify the actual genetic basis of the belted and non belted phenotype studies were carried out using animals from a Pietrain x (Pietrain x Hampshire) cross. The Pietrain is be/be while the Hampshire is Be/Be. Thus the F1 generation all have the genotype Be/be. Further crossing of the F 1 back to the Pietrain (be/be)leads to the segregation of 3 0 the Be allele between offspring, giving rise to Be/be animals showing belts and be/be non belted offspring. A correlation was then established between the inheritance of the belted condition and certain microsatellite markers within these pedigrees. This work surprisingly identified the actual gene involved as the KIT gene also described- above as involved in dominant white. Further analysis showed correlation of the phenotype in this pedigree with a polymorphismn at KIT nucleotide 2678 with a C or T occurring at this position. The presence of a C creates a restriction site for Aci I which is absent when a T is present- Based upon these unexpected findings a number of approaches can be taken to the determination of the genotype for an animal at the belt locus using either single nucleotide polymorphisms or linked markers including microsateltes or other single nucleotide polymorphisms. Thus animals can be genotyped by a number of approaches to determine their genetic status for this particular overt characteristic.

Sunmmary of the invention According to the present invention there is provided a method for differentiating animals and animal products on the basis of breed origin, for determining or testing the is breed origin of an animal product or for validating an animal prqduct, wherein the method comprises the steps of: providing a sample of the animal product; and (ii) analyzing the allele(s) of one or more breed determinant genes present in the sample.

As explained above, the breed determinant is an overt phenotypic trait. As used herein, 2 0 an overt phenotypic trait is one which can be viualy recognized.

Differentiation of animal products on the basis of breed origin involves the partition of members of a class of different animal products into a number of different products sharing the same breed origin, It does not necessarily imply identification of the nature of the breed source. Animal product differentiation on this kind of basis may be sufficient where the consistency of source of animal products must be monitored (but its actual breed provenance is not important).

In contjAt determination of the breed origin of an animal product implies identification 3 0 of the breed source, while testing the breed origin implies analysis sufficient to determine whether a breed source other tha that desir-ed has been used (without necessarily identifying such other breed sources in case where they are indicated).

Validating an animal product implies confirming that it meets stipulated specifications as to breed. provenance. Such validation may involve differentiation, determination and/or testing, depending on the circumstances under which the analysis is performed s and the nature and extent of ancillary data which may be available.

The sample for use-in the invention may be in any convenient form In many cases, the sample will be a sample of a food meat product). For most Applications, the sample is wre-treated extracted, purified and/or fractionated) in such a way so as to make the alleles of a breed determinant gene or genes available for analysis (either at the level of nucleic acids (such as RNA or DNA) and/or proteins). The sample is preferably a nucleic acid sample, in which case the analysing step (ii) comprises DNA or RNA analysis. Alternatively, the sample may be a protein sample (where the namure of the protein reflects a breed determinant allele), in which case the analysing step (ii) compnises protein analysis.

The breed deterninant of the invention may be a monogenic or polygenic trait.

Monogenic traits are preferred, since the genes conferring such traits are relatively -easily identified and analyzed. However, in some cases it may be useful to analyze the alleles of polygenic traits traits which are controlled by a plurality of genes), since the underlyiirg allele polymorphism is often greater in such cases (so increasing the potential for breed differentiation).

Typically, overt phenotypic traits are those traits which have been used as the basis for artificial selection during the breeding programme. The overt phenotypic trait is preferably a behavioural or morphological, physiological or behavioural trait.

The overt phenotypic trait may vary qualitatively or quantitatively between breeds.

Preferred are traits which vary qualitatively between breeds, since such traits are often reflected by qualitative differences in the alleles of the corresponding breed determinant gene(s). In such cases, analysis yields relatively robust positive-negative results, which are easily interpreted and compared between testing stations/laborat ories.

The breed determinant gene analysed in step (ii) may be any suitable breed determinant gene. Such genes may be identified and analysed by methods well known in the art using routine trial and error. Preferably, they are selected from any of a coat colour, pattern, texture, density or length gene; an ear aspect gene; a double muscling gene; a horn morphology gene; a tusk morphology gene; an eye colour gene; a plumage gcnw, a beak* colour/morphology gene; a vocalization barking) gene; a comb or wattle gene; and/or a gene controlling display behaviour.

In preferred embodiments, the breed determinant gene is the KIT or aMSHR coat colour gene (for example, the pig KIT and/or aMSHR gene).

The analysis step (ii) may comprise any of a wide range of known nucleic acid/protein analytical techniques. The nature of the analytical technique selected is not critical to the practice of the invention, and those skilled in the art can readily determine the appropriate technique according to the circumstances in which the analysis is to be conducted and the type of data required.

Preferably, the analysis step (ii) comprises selectively amplifying a specific fragment of nucleic acid by PCR), testing for the presence of one or more restriction endonuclease sites within the breed determinant gene(s) restriction fragment length polymorphism (RYLP) analysis), detennining the nucleotide sequence of all or a portion of the breed detemninant gene(s), probing the nucleic acid sample with an allele-specific DNA or RNA probe, or carrying out one or more PCR amplification cycles of the nucleic acid sample using at least one pair of suitable primers and then carrying out RFLP analysis on the amplified nucleic acid so obtained.

Alternatively, the analysis step (ii) comprises probing the protein sample with an antibody a monoclonal antibody) specific for an allele-specific cpitope, electrophoretic analysis, chromatographic analysis, amino-acid sequence analysis, proteolytic cleavage analysis or epitope mapping. For example the 96 allele might be distinguished by any method capable of detecting an alteration in the size of the encoded protein.

In particularly preferred embodiments, the analysis step (ii) comprises determining the nucleotide sequence of the KIT and/or aMSHR gene or the amino acid sequence of the KIT and/or aMSHR protein. Here, the analysis may comprise establishing the presence or absence of at least one mutation in the KIT and/or cMSHR gene. Any method for identfying the presence of the specific sequence change may be used, including for example single-strand conformation polymorphism (SSCP) analysis, ligase chain reaction, mutagenically separated PCR, RFLP analysis, heteroduplex analysis, denaturing gradient gel electrophoresis, temperature gradient electrophoresis, DNA sequence analysis and non-gel based systems such as TaqMan m (Perkin-Elmer).

In the TaqMan T M system, oligonucleotide PCR primers are designed that flank the mutation in question and allow PCR amplification of the region. A third oligonucleotide probe is then designed to hybridize to the region containing the base subject to change between different alleles of the gene. This probe is labelled with fluorescent dyes at both the 5' and 3' ends. These dyes are chosen such that while in this proximity to each other the flourescence of one of them is quenched by the other and cannot be detected. Extension by Taq DNA polymerase from the PCR primer positioned 5' on the template relative to the probe leads to the cleavage of the dye attached to the 5' end of the annealed probe through the 5' nuclease activity of the Taq DNA polymerase. This removes the quenching effect allowing detection of the florescence from the dye at the 3' end of the probe. The discrimination between different DNA sequences arises through the fact that if the hybridization of the probe to the template molecule is not complete there is mismatch of some form), then cleavage of the dye does not take place. Thus only if the nucleotide sequence of the oligonucleotide probe is completely complementary to the template molecule to which it is bound will quenching be removed. A reaction mix can contain two different probe sequences each designed against different alleles that might be present thus allowing the 3 0 detection of both alleles in one reaction.

Although the TaqMan

T

system is currently capable of distinguishing only two alleles, 19 labelled probe primer sets could be developed in which the probes for certain target allele(s) are labelled with a different fluorescent dye from non target alleles. For example, if one wished to confirm that a group of Duroc breed pigs carried only allele e one could have a probe present capable of detecting this allele labelled with one S fluorescent dye and probes capable of detecting all the other alleles labelled with the second dye. Thus one would detect the presence of any non Duroc type alleles at this locus. Such probe sets could be designed and labelled according to the needs of the experiment.

The analysis step (ii) may further comprise determining the association between one or more microsatellite marker alleles linked to the KIT and/or aASHR gene and to particular alleles of the KIT and/or aAMSfHR gene.

Alternatively, the analysis step (ii) may be based on the identification of microsatellite markers present in the nucleic acid sample.

The analysis step (ii) preferably comprises: determining the association between one or more microsatellite marker alleles linked to the KIT and/or aASHR gene and to Sparticular alleles of the KIT and/or aMSHR gene; determining which microsatellite 2 0 marker allele or alleles are present in the nucleic acid sample.

The analysis step (ii) preferably further comprises the step of determining the genotype of at least one additional locus, for example an additional breed determinant coat colour) locus. Particularly preferred as an additional locus is the KIT gene locus the pig KIT gene locus).

The analysis step (ii) preferably comprises PCR using at least one pair of suitable primers. In the case where the gene is the pig aMLSHR gene, the at least one pair of suitable primers is: aMSHR Forward Primer 1: (5'-TGT AAA ACG ACO CGCC AGT RGT GCC TGG AGG TGT-3') aMSHR Reverse Primer 5: (5'-CGC CCA GAT GGC CGC GAT GGA CCG-3'); or aMSHR Forward Primer 2: (5'-CGG CCA TCT GGG CGG GCA GCG TGC-3') aMSHR Reverse Primer 2: (5'-GGA AGG CGT AGA TGA GGG GGT CCA-3); or aMSHR Forward Primer 3: (5'-GCA CAT CGC CCG GCT CCA CAA GAC-3) aMSHR Reverse Primer 3: (5'-GGG GCA GAG GAC GAC GAG GGA GAG-3) The analysis step (ii) may also comprise restriction fragment length polymorphism (RFLP) analysis, for example involving digesting the pig nucleic acid with one or more of the restriction enzymes BstUI, Hhal and/or BspHI. In cases where the gene is the pig aMAHR gene, this analysis may involve identification of a polymorphism at any of the nucleotide positions shown to be polymorphic including 283, 305, 363, 370, 491, 727, 729, 1162 or between nucleotide positions 60 and 70 or between nucleotide positons 1005 and 1010of the pig CMSHR gene.

The analysis step (ii) may involve carrying out one or more PCR amplification cycles of the nucleic acid sample using at least one pair of suitable primers and then carrying out RFLP analysis on the amplified nucleic acid so obtained to determine the KIT or aMSHR genotype of the pig. Here, when the gene is the pig aMSHR gene the at least one pair of suitable primers is as defined above.

The animal product preferably comprises or consists of meat processed and/or canned meat), egg, egg swab or washing, semen, blood, serum, sputum, wool, biopsy sample or leather. It may comprise genomic DNA, RNA or mitochondrial DNA.

The animal is preferably a mammal pig, cattle, dog, cat, horse, sheep, rodent or rabbit), fish salmon or trout) or bird chicken or turkey).

The invention may be used with extremely small samples and can be used to screen large numbers of samples quickly and inexpensively. The invention may be adapted to 3 0 yield absolute results, and quantification is not essential. Moreover, only small fragments of nucleic acid are required, and the same tests can be used on the majority of animal products.

Aplications The invention finds application in a number of areas. For example, certain breeds are considered to yield meat of higher eating quality, and a number of retailers now market products which claim to be derived from specific or traditional breeds (for example, Wild Boar crosses). The invention enables consumer organisations to validate these claims and also permits retailers to monitor the quality of the products with which they are being supplied perform product validation). The invention finds particular application in validation studies carried out and used by retailers to support consumer confidence, since the linkage between a genetic marker and an overt physical feature is more readily grasped by the lay person than the concept of breed specific markms. This makes the use of such breed determinants attractive and also offers marketing opportunities for retailers to underpin validation schemes.

There are also a number of reports of breed influences on the quality of hams produced by various meat processing techniques. For example, in one report hams from three different pig breeds were reliably classified on the basis of sensory descriptors of marbling, saltiness and dry cure flavour. The breed identification processes of the invention enables producers to validate raw materials as part of quality controL The ability to enforce and validate raw material source uniformity also yields improved process control, lower costs and greater product consistency, since it has now been found that heterogeneity in chemical composition of products from different breeds is an important factor in flavour profile variation and there may also be differences in the functionality of other meat components between breeds.

The invention also finds utility in the maintenance of stock purity by animal pig) breeders. The small size of traditional breed populations means that the maintenance of a gene pool of sufficient size to avoid the effects of inbreeding requires the importation 3 0 and movement of stock between separate populations. A risk of genetic contamination is associated with such movements, and the invention may be used to reduce or eliminate these risks. The maintenance of biodiversity and the rare breeds providing the 22 reservoir for this diversity provides an increasing need for breed identifiers. There is for example a problem for breeders of the British Saddleback. Certain bloodlin'of this breed carry a higher frequency of the be allele of the belt locus which can result in the production of belt-less animals which do not reach the required breed standard and decrease the value not only of that individual animal but of the whole litter. The ability to select against this allele when new bloodlines are introdu to an existing population would enable breeders to increase the genetic diversity with out the risk of lowering the relative standard of the particular population to that of the breed in general.

The invention may also be used as part of a breeding programme to confirm particular crosses. This may be of enormous value in the establishm of pyramid breeding schemes. Particular breed characteristics such as coat colour, body shape and ear aspect arm ohen altered in such crosses, yet there is a need to be able to confirm the presence of genetics of the desired parents.

Such visible breed characteristics for the visible confimation of crosses are also absent in the use of artificial insemination, where semen may be supplied from pigs in distant geographical locations.

In addition, the skilled person will appreciate that based on the information described herein, it is possi le to provide tests for determining pig genotype, with respect to coat colour. Thus, the present invention also provides a method of detcrmining the coat colour genotype of a pig which comprises: obtaining a sample of pig nucleic acid; and (ii) analysing the nucleic acid obtained in to determine which allele or alleles of the aMSHR gene are present.

In one embodiment of this aspect of the invention the determination in step (ii) is carried out by determining the nucieotide sequence of the zxMSHR gene and, in particular, is based on determining -which missense, insertion or deletion mutation is 23 present in the coding region of the gene.

In another embodiment one could first determine the association between microsatellite or other linked marker alleles linked to the aLMSHR gene and particular alleles of the aMSHR gene. Thus, the determination in step CHn) would be based on identification of microsatellite marker alleles present in the nucleic acid sample In a further aspect, therefore, the present invention provides a method of determining the coat colour genotype of a pig which comprises: determining the association between one or more microsatellite or other linked marker alleles linked to the CIMSHR gene and particular alleles of the CZMSHR gene; (ii) obtaining a sample of pig nucleic acid; and (iii) analysing the nucleic acid obtained in (ii) to determine which microsatdflite or other linked marker allele or alleles are present.

The determination of the alleles at the extension locus will indicate the background colour of the animal and in some cases the pattern of mixed colouration, i.e. spotting, but will not necessarily determine the coat colour of resulting progeny. This will be dependent on the genotype at other loci such as the dominant white locus, I. The genotype at the Ilocus can be determined separately as described in WO-A-97/05278.

Thus, suitably, the methods as described above may further comprise the step: 24 (iii)/(iv) determining the genotype of additional coat colour loci.

An example of such an additional coat colour locus is the belt locus.

In a preferred method PCR is carried out using primers that amplify a region of the KIT gene containing nucleotide 2678. An example of a suitable pair of primers is: forward primer LA93 5" GAGCAGCCCCTACCCCGGAATGCCAGTTGA -3' and the reverse primer KT56 5' CTTTAAAACAGAACATAAAAGCGGAAACATCATGCGAAGG 3'.

ioThe method of analysis enables determination of the presence of a C orT at position 2678. Suitably, the restriction enzyme Acil can be used since the presence ofa C creates a restriction site which is absent when a T is present Similar examinations within a pedigree of will allow the determination of the genotype of offspring.

Thus, in additional aspect; the present invention provides A method of determining the coat colour genotype of a pig which comprises: obtaining a sample of pig nucleic acid: and (ii) analysing the nucleic acid obtained in to determine whether the KIT gene carries any polymorphism associated with Belt genotype.

Preferably, the method comprises RFLP analysis which is suitably carried out on a sample of pig genomic DNA which has been amplified using PCR and a pair of suitable primers.

Preferred methods for identifying the presence of the specific sequence change are described above in relation to breed determinants.

Brief description of the Figures Figure 1 Partial nucleotide sequence and the derived amino acid sequence of the porcine aMSH-R gene as determined from a number of pig breeds. Position numbers for the nucleotide sequence are based upon nucleotide 1 being the A of the ATG initiation codon. Numbers of the amino acids are in accordance with the bovine BDF3 sequence (Vanetti et al. FEBS Lett. 348: 268-272 (1995)) to allow comparison.

Figure 2: Agarose gel electrophoresis of DNA fragments obtained by digestion of DNA fragments amplified from the porcine aMSH-R gene with BstUIl or HhaI. Lanes labelled M contain DNA markers of 50, 150, 300,500, 750, l000bp. The other samples were derived from: 1. Pietrain 2. Pietrain 3. Large White 4. Large White 5. Large White 6. Duroc 7. Duroc 8. Hampshire 9. Mcishaii 10. Berkshire 11. Berkshire Figure 3: Agarose gel electrophoresis of DNA fragments obtained by digestion of DNA fragments amplified from the porcine aMSH-R gene with BstUI (lanes labelled B) or Hhal (lanes labelled Lanes labelled M contain DNA markers of 50, 150, 300,500, 750, 1000bp. The other samples were derived from: 1. Retailer 1. Skin 2. Retailer 1. Fat 3. Retailer 1. Muscle 4. Retailer 2. Fat Retailer 2. Muscle Figure 4: Electropherogram agarose) showing RT-PCR products of KIT exoa 16- 19 with the primers KITIF and KITR. The samples 1-3 and 4-6 are Swedish Large White and Hampshire pigs respectively. The size difference between the 424 and 301 bp fragments is due to lack of exon 17 in the latter fraction. The two upper bands of the Yorkshire pigs were interpreted as heteroduplexes (HD).

Figure 5: A 48 bp sequence is shown comprising 21 bp of KIT exon 17 and 27 bp of KIT intron 17. The position of the intron/exon border is marked with a vertical line and the splice site mutation (ntl o-

A

indicated with a vertical arrow. Identical bases in alleles t and i are marked with a dot.

Figure 6: hNaIII PCR RFLP test used to detect the presence of a splice site mutation in intron 17 of the KIT gene. Figure 6A shows the position of two NIMa recognition sites within the PCR product amplified using primer pair KIT21 and KIT35. All distances are given in base pairs. Figure 6B shows the size of fragments which result following NIMa digestion of either normal KIT or splice mutant KIT. Figure 6C illustrates use of the PCR RFLP test Lane 1 shows the KIT2I/KIT35 amplified fragment undigested- Digestion was performed on PCR products amplified from, in Lane 2: a clone which contains the splice site mutation; Lane 3: a clone which contains the normal splice site sequence; Lane 4: genomic DNA from a coloured pig; Lane 5: genomic DNA from a white pig. Fragment sizes are given in base pairs.

Figure 7: Comparison of the ratio of normal to splice mutant KIT for three classes of genotype.

Figure 8: Comparison of the ratio of normal to splice mutant KIT for two breeds of pig.

Figure 9: SSCP analysis of the KIT gene in Swedish Landrace (lanes 1-8) and Wild Boar (lanes 9 10) breeds. The two polymorphic bands are indicated.

27 Figure 10: Nucleotide sequence of the porcine KIT cDNA from an animal of the Hampshire breed. The sequence is numbered with the first nucleotide of the N terminal methionine codon taken as 1.

Fiure 11: Polyacrylamide gel electrophoresis of PCR-RFLP analysis of KIT gene at polymorphic nucleotide 2678 in a number of animals. Lanes: 1 2, Hampshire Wild Boar respectively, both homozygous for the C at position 2678. Lanes 3-7 and 9 unrelated Large White sows all homozygous for T at position 2678. Lane 11, a Pietrain, homozygous for T at this position and lane 8 a Large White sow heterozygous for C and T Lane 12 contains undigested PCR product and lane M DNA size standards Fiure 12 Nucleotide sequence of the 3' end of the porcine aMSHR coding region and adjacent 3' untranslated region. The TGA stop codon is highlighted in bold, the primer binding sites for EPIGI4 is shown in italics. Numbering is based on the system used in figure la in which nucleotide 1 is the A of the ATG initiation codon of the Wild Boar sequence. Bases in common with the European Wild Boar are marked with a dash.

Missing bases are marked with a:.

Examples Example 1: Determination of the sequence of the aMSHR gene The DNA sequence of the porcine aMSHR gene was determined through the DNA sequencing of a combination of PCR products and cloned portions of porcine DNA.

Preparation of template DNA for PCR DNA can be prepared from any source of tissue containing cell nuclei, for example white blood cells, hair follicles, ear notches and muscle. The procedure here relates to blood cell preparations; other tissues can be processed similarly by directly suspending material in K buffer and then proceeding from the same stage of the blood procedure. The method outlined here produces a cell lysate containing crude DNA which is suitable for PCR amplification. However, any method for preparing purified, or crude, DNA should be equally effective.

Blood was collected in 50mM EDTA pH 8.0 to prevent coagulation. 50pl of blood 28 was dispensed into a small microcentrifuge tube (0.5ml Eppendorf or equivalent).

450l of TE buffer was added to lyse the red blood cells (haem groups inhibit PCR) and the mix vortexed for 2 seconds. The intact white and residual red blood cells were then centrifuged for 12 seconds at 13,000 g in a microcentrifuge. The s supernatant was removed by gentle aspiration using a low pressure vacuum pump system. A further 450pl of TE buffer was then added to lyse the remaining red blood cells and the white blood cells collected by centrifugation as before. If any redness remained in the pellet, this process was repeated until the pellet was white. After removal of the last drop of supernatant from the pelleted white blood cells, 100pl of K buffer containing proteinase K was added and the mixture incubated at 55 degrees C for 2 hours. The mixture was then heated to 95-100 degrees C for 8 minutes and the DNA lysates stored at -20 degrees C until needed.

Reagents T.E. Buffer 10mM TRIS-HCI 1s ImM EDTA K Buffer: 50mM KCI TRIS-HCl pH8.3 MgCl2 Tween PCR to produce DNA sequencing template The aMSHR gene was amplified for sequence analysis using three primer pairs.

Primers MSHR Forward Primer 1: (5'-TGT AAA ACG ACG GCC AGT RGT GCC TGG AGG TGT CCA and MSHR Forward Primer 5: (5'-CGC CCA GAT GGC CGC GAT GGA CCG-3') amplify a 428 bp fragment from the 5' half of the gene.

29 Primers MSHR Forward Primer 2: (5'-CGG CCA TCT GGG CGG GCA GCG TGC-3'); and aMSHR Reverse Primer 2: (5'-GGA AGG CGT AGA TGA GGG GGT CCA-3) amplify a 405 bp fragment the 3' half of the gene.

S As these two fragments are non-overlapping a third primer pair aMSHR Forward Primer 4 (5'-TGC GCT ACC ACA GCA TCG TGA CCC TGC-3); and uMSHR Reverse Primer 4 (5'-GTA GTA GGC GAT GAA GAG CGT GCT-3) were used to amplify a 98 bp fragment which spans the 50 bp gap. PCR was carried out on a DNA thermal cycler (Perkin Elmer 9600) in a total volume of 20 pl containing 25 ng genomic DNA, 1.0 mM MgC12, 50 mM KC1, 10 mM Tris-HCI, pH 8.3,200 (M dNTPs, 0.5 U AmpliTaq Gold (Perkin Elmer) and 10 pmol of both forward and reverse primer. To activate AmpliTaq Gold, initial heat denaturation was carried out'at 94 degrees C for 10 minutes followed by 32 cycles each consisting of sec at 94 degrees C, 45 sec at 53 degrees C and 45 sec at 72 degrees C. The final extension lasted for 7 min at 72 degrees C. PCR products were cloned into vector pUC 18 using the SureClone ligation kit (Pharmacia).

Preparation of plasmid DNA Plasmid DNA was purified from overnight bacterial culture using the Jetstar plasmid midi kit 50 (Genomed) and the resulting DNA diluted to 150 ng/pl.

Sequencing of plasmid DNA Cloned plasmid inserts were sequenced using dye primer chemistry. Each cycling reaction was prepared with template and ready reaction mix containing fluorescently labelled M 13 forward or reverse primer as described in the ABI Prism protocol P/N 402113 (Perkin Elmer). Cycling and sample pooling was performed using a Catalyst 800 Molecular Biology Workstation (ABI) following the instruments user manual (Document number 903877, Perkin Elmer). The resulting extension products were purified, loaded and analysed using the 377 ABI Prism DNA sequencer as described by the instrument protocol (Perkin Elmer protocol P/N 402078).

Dye Terminator Seauencing of PCR products Dye terminator DNA sequencing requires purification of PCR product free from excess dNTPs and residual primers. This was achieved by passage of the template DNA through QiaQuick spin columns (Qiagen) before the purified DNA was diluted to 15 ng/pl. Dye terminator cycle sequencing was performed using AmpliTaq DNA polymerase FS in accordance with the ABI Prism protocol P/N 402078 (Perkin Elmer). Cycle sequencing reactions were performed in a total reaction volume of This comprised 1.6 pmole of either the forward or reverse primer used to amplify the target fragment from genomic DNA, 20 ng of purified template DNA and terminator ready reaction mix (Perkin Elmer) which contains each of four dye terminators, dNTPs, Tris-HCl (pH MgCl 2 thermal stable pyrophosphate.and AmpliTaq DNA polymerase FS. Cycle sequencing was performed with a GeneAmp 9600 machine (Perkin Elmer) over 25 cycles, each consisting of 10 sec at 96 degrees C, 5 sec at 50 degrees C and 4 min at 60 degrees C. Extension products were purified for gel separation using ethanol precipitation, loaded and run on a 377 ABI Prism DNA sequencer as described by the instrument protocol (Perkin Elmer protocol P/N 402078).

Results The partial coding region DNA sequence of the porcine aMSHR gene sequence from a number of pig breeds is given in figure l a combined with sequence determined in example 22. The derived amino acid sequence is shown in figure lb.

Example 2: PCR-RFLP based discrimination.of alleles at the E locus DNA preparation for PCR As in example 1.

PCR

Reactions were set up in a 20p1 reaction volume in thin walled 0.25ml tubes (Perkin Elmer) with the following components: 20pi reaction volume: 24 1 template DNA mMMgCI2 200pM each dNTP, 3pM each of forward and reverse primers 0.5 U AmpliTaq Gold (Perkin Elmer) MSHR Forward primer 3 sequence: 5' GCA CAT CGC CCG GCT CCA CAA GAC 3' MSHR Reverse primer 3 sequence: 5' GGG GCA GAG GAC GAC GAG GGA GAG 3' The reaction tubes were placed on a Perkin Elmer 9600 thermal cycler preheated to 94 degrees C and PCR carried out according to the regime below: Initial denaturation step of 94 degrees C for 10 min.

33 cycles: 94 degrees C 45 sees 53 degrees C 45 sees 72 degrees.C 45 sees The last cycle is followed by a final elongation of 72 degrees C for 7 min. Samples are stored at 4 degrees C until required.

Restriction Enzyme Digestion and Electrophoresis The PCR amplification product is 148bp in length. To test for polymorphism in the amplified products the reaction is split into two aliquots of 10W each of which is digested with Hhal (GIBCO-BRL) or BstUI (New England Biolabs). The reactions are set up and incubated as below: BsUI digest amplified DNA BstUI degrees C 60 minutes Hhal digest 10gl amplified DNA 2.5p Hhal 0.5pl I Ox React 2 buffer (GIBCO-BRL) 37 degrees C 60 minutes Following digestion, 2±l of loading dye is added to each reaction (100mM Tris 100mM Boric Acid, ImM EDTA, 50% glycerol, 0.02% w/v Orange G) and the mixes loaded on a 4% agarose gel NuSieve/l% Seakem, FMC Bioproducts) in 0.5x TBE (44.5mM Tris pH8.0,44.5 mM boric acid and EDTA) and electrophoresed for I hour at 150v.

Products are visualised by ethidium bromide staining.

Results BstUI and Hhal digestion each result in bands of 61 and 87bp. The relationship of digestion to the possible allele is as shown in the table below: Relationship of restriction digest profiles to individual alleles at the E locus Allele

E"

e Digestion with BstUI Yes No No Digestion with Hhal Yes Yes No 2 5 If the uncut alleles are designated as allele I and the alleles digesting with each enzyme as allele 2 the various genotypes will be as shown in the table below: 'Actual E genotypes and associated scores Genotype

EOIEO

E"/EP or E"/f Ele EP or F/i or E/e or tie e/e BstUI 1/1 1/2 1/1 2/2 1/2 1/1 hal 2/2 2/2 1/2 2/2 1/2 1/1 Note: The results for animals carrying the allele E will be the same as those carrying

*P

Samples were prepared from a number of pigs and tested according to the above protocol. The results are shown in the table below and figure 2 illustrates the patterns seen upon electrophoresis.

E genotvpes determined for a range of breeds using the BstUI/Hhal digestion system Breed Hampshire Large White Landrace 20 Pietrain Berkshire Bazna No Tested Genotype (see note 1) f/EN

EPL

ele

EN/E

BstUI 2/2 2/2 2/2 2/2 2/2 1/2 1/1 1/1 IIl aMSHR type Hhal 2/2 2/2 2/2 2/2 2/2 2/2 2/2 1/1 2/2 Duroc Meishan Note 1. Thegenotype cannot be distinguished from E or e in this particular test.

As can be seen from the results above the genotypes determined fit with those expected from the sequencing data given in figure I a for Hampshire, Large White, Meishan and Duroc. The additional breeds typed here show the genotypes expected from their phenotype and descriptions in published literature (Ollivier and Sellier, Ann. G&t S6l. Anim., 14: 481-544, (1982)). The Pietr-ain is a white breed with black patches of varying extent and has long been considered to be A (in agreement with the result here). The Berkshire, originally a spotted breed, is now a mainly black animal with white 'socks' again generally considered to be EP as was found here. The Landrace is a white animal due to it carrying the dominant white allele at the I locus, however its genotype at the E locus has been shown to be EP from classical breeding studies. Once again this is in agreement with the results obtained here. The Bama is a Romanian breed having black base colour with a white belt It was developed from the Bcrkshire and Mangalitza, a Hungarian breed with a number of colour variations including black (Porter, Pigs, a handbook to breeds of the world, publ: Helm Information, ISBN 1-873403-17-8 (1993)). The ancestory of the Bazna being based upon a black breed potentially carrying a similar allele to the Meishan, and the Berkshire carrying F, is in agreement with the alleles found to be present in the breed in this workL Example 3: Validation of source breeds of retail meats DNA preparation DNA was prepared from different parts of pork chops from two separate retailers.

The DNA was prepared from skin (1 retailer only), fat and muscle using the Promega Wizard Genomic DNA preparation kit according to the manufacturers instructions.

Approximately 4mm 3 of each tissue was cut into small fragments for the extraction- PCR and Restriction Digest Analysis This was carried out exactly as in example 2.

Results The results are shown in figure 3. It can be seen that DNA extracted from a range of tissue types can be utilised for this DNA based test with results being obtained here for muscle, fat and skin. The genotype of the pig with regard to the (MSHR gene can then be determined. In this case the material from both retailers was derived from an animal of test type BstUI 1/2 and Ihal 1/2 using the nomenclature as in example 2.

This translates into genotype E'/e or Et/e. Based on our current knowledge of the distribution of the alleles in commercial pig breeds the conclusion can be drawn that both source animals contain genetic material derived from the Duroc.

Example 4: Validation of source breeds of processed meat samples Method DNA was prepared from heat treated meat samples according to the method of Meyer et al. (Journal of AOAC International, 78 1542-1551). Meat samples were minced with a scalpel and 0.3g transferred to a sterile 1.5ml eppendorf tube containing 430pl of extraction buffer (10mM Tris-HCI pH 8.0, 150mM NaC1, 2mM EDTA, and 1% w/v sodium dodecyl sulphate). Fifty microlitres of 5M guanidine hydrochloride and of 20mg/ml proteinase K (Boehringer) were added and mixed by inversion followed by incubation at 57°C for 3h. After digestion samples were centrifuged for min at 13,000 x g, and 450 1 l of the aqueous phase added to lml Wizard DNA purification resin (Promega). The mixture was mixed by gentle inversion and following the Wizard DNA clean-up procedure carried out according to the manufacturers instructions the purified DNA was eluted with 50l of 70°C water. 1l of a 1:10 dilution was then used as template in a 10l PCR.

PCR was carried out as described in the previous example.

2 Results Meat samples from a Large White based line and a Duroc based line heated at for 30 mins could be differentiated on the basis of their genotype at the E locus with the Large White samples giving a pattern characteristic of the E' allele and the Duroc 36 samples a pattern characteristic of the e allele.

Example 5: Validation of source breeds of semen Genomic DNA was isolated from porcine semen. Iml of semen was centrifuged for 2 min at 13,500 x g and the supernatant removed. I ml of 2xSSC was added and the mix vortexed to resuspend the sperm. The mix was then centrifuged as before and the supernatant removed. 400gl of 0.2M NaOAc pH 7.0 was added and the mix vortexed followed by the addition of 34l of B-mercaptoethanol. The mixture was incubated at for 30 min followed by the addition of 100pl of 10% w/v sodium dodecyl sulphate and 50pl of 15 mg/ml Proteinase K (Boehringer) and further incubation at 40°C for 3 hours. 500 p phenol equilibrated with Tris-HCl pH 8.0 was added and the mix vortexed twice followed by centrifugation at 13,500 x g for 4 min. 400pl of the aqueous phase was removed and 800gl of ethanol added. DNA was allowed to precipitate for 5 min at room temperature followed by centrifugation at 13,500 x g for min. The pellet was washed with 800pl 70% ethanol v/v and air dried followed by resuspension in 200pl of Wizard DNA resuspension buffer (Promega). 1pl of a 1/10 dilution was used in a 10pl PCR PCR was carried out as described in example 2.

Results Semen form a Hampshire based line and a Duroc based line could be differentiated on 20 the basis of their genotype at the E locus with the Hampshire samples giving a pattern characteristic of the t allele and the Duroc samples a pattern characteristic of the e allele.

Example 6: Discrimination of allele E from alleles FY/ DNA preparation DNA was prepared as described in example 1.

PCR

37 Reactions were set up in a 201l reaction volume in thin walled 0.25ml tubes (Perkin Elmer) with the following components: 1 l reaction volume: 2pl template DNA 2.5 mM MgC12 200pM each dNTP, each of forward and reverse primers U AmpliTaq Gold (Perkin Elmer) Forward primer sequence: 5' CTG CCT GGC CGT GTC GGA CCT G 3' Reverse primer sequence: 5' CTG TGG TAG CGC AGC GCG TAG AAG 3' The reaction tubes were placed on a Strategene Robocycler and PCR carried out according to the regime below: Initial denaturation step of 94 0 C for 10 min.

cycles: 94°C 60 sees 61"C 60 sees 72°C 60 sees The last cycle is followed by a final elongation of 72"C for 7 min. Samples are held at 6°C until required.

Restriction Enzyme Digestion and Electrophoresis.

The PCR amplification product is 228 in length. To test for polymorphism in the amplified products the reaction is digested with BspI (New England Biolabs). The reactions are set up and incubated as below: BspHI digest 104l amplified DNA lul 1Ox React 2 (NEB New England Biolabs) deionised water units BstUI 37 0C 60 minutes Following digestion, 2dI of loading dye is added to the reaction (100mM Tris pHS.0, 100mM Boric Acid, ImM EDTA, 50% glycerol, 0.02% w/v Orange G) and the mix loaded on a 4% agarose gel NuSieve/l% Seakem, FMC Bioproducts) in TBE (445mM Tris pHS.0, 4 4 5 mM boric acid and 0.5mM EDTA) and electrophoresed for I hours at 150v.

Products are visualised by ethidium bromide staining.

Results BspHI digestion each result in bands of 124 and 104bp. The relationship of digestion 0oto the possible allele is as shown below: Relationship of restriction digest profiles to individual alleles at the E locus Allele Digestion with BspHI lE"/f Yes E* No Samples were prepared from a number of pigs and tested according to the above protocol and the results are shown below: E genotypes determined for a range of breeds using the BspHI digestion system Breed No Tested Genotype (see note 1) Number Wild Boar x 3 E/ 3 Swedish Landrace Large White Landrace Pietrain

EPIEP

Ef/E

EFIE

4 1 3 Note 1. Where the genotype EP is listed this cannot be distinguished from E in this particular test.

Example 7: Discrimination of cattle products by breed DNA was prepared from cattle muscle samples as described in example 4. PCR was then carried out in a 100pl reaction using the primer pair.

5'-TGAGGTAGGAGAGTTITGGG-3' and 5'-TCGAAATTGAGGGGAAGACC-3' as described in Kambadur et al. Genome Research 7:910-915 (1997) at a concentration of 500nM with other reaction components being 2.5mM MgCI2, 200ipM dNTPs, 50mM KC1, 10mM Tris-HCI pH 8.3, 5 units AmpliTaq Gold (Perkin Elmer).

I pl of bovine genomic DNA was used as template. Denaturation was carried out for 12 min at 94C followed by 30 cycles of 94°C for 1 min, 55*C for 1 min, 72" 1.5 min followed by 5 min at 72C. Following PCR 2.0pl of loading dye (44.5mM Tris pH 44.5mM boric acid, 0.5mM EDTA, 50%w/v glycerol, 0.02% w/v Orange G) was s added to 10pl of product and analysis carried out by electrophoresis on a 2% agarose gel prepared-in 0.5x TBE buffer (445mM Tris pH 8.0, 44.5mM boric acid, EDTA) for 1 hour at 100V.

The remainder of the PCR was analysed for DNA sequencing using ABI dye terminator chemistry as described in example 1.

2o Results Bovine mvostatin DNA polymorphisms and related phenotype Breed Phenotype at position 941 length PCR product (bp) Belgian Blue Double muscle G 482 Piedmontese normal A 493 Holstein-Friesian Double muscle G 493 Example 8 RT-PCR of porcine KIT exon 16-19 i. mRNA purification from blood samples Fresh blood samples were collected in citrate tubes from coloured Hampshire pigs and Large White pigs. Leukocytes were isolated from 5 ml blood using Ficoll 100 (Pharmacia Biotech). Isolation of mRNA from leukocytes was then carried out using the Quickprep Micro mRNA purification kit (Pharmacia Biotech). The mRNA was stored as a precipitate under ethanol at -70 0 C for up to one month before use in reverse transcriptase (RT)-PCR.

o1 ii.RT-PCR of KIT exon 16-19 First-strand cDNA synthesis was accomplished using the First-Strand cDNA Synthesis kit (Pharmacia Biotech) so that -100 ng mRNA was randomly primed by 0.1 gig pd(N6) in a total volume of 15 l. Two Il of the completed first cDNA strand reaction was then directly used per 12 til PCR reaction by adding 10 l PCR mix is containing 10 pmol each of the mouse/human derived primers KITIF and KIT7R TCR TAC ATA GAA AGA GAY GTG ACT C and S'-AGC CTT CCT TGA TCA TCT TGT AG, respectively; Moller et al. 1996, supra), 1.2 l 10 x PCR-buffer mM Tris-HCi, pH 8.3, 50 mM KCI) and 0.5 U of AmpliTaq polymerase (Perkin- Elmer) incubated with an equal amount Taqstart antibody (Clontech) at 25 °C for min to achieve a hot start PCR. The reaction was covered with 20 pi mineral oil and thermocycled in a Hybaid Touchdown machine (Hybaid) with 40 cycles at 94°C for 1 min, 55-48 C (touchdown one degree per cycle the first seven cycles and then 48°C in the remaining cycles) for 1 min and 72°C for 1 min. After PCR 2l loading dye was added to each sample which were then loaded on 4% agarose gel (Nusieve/Seakem 3:1, FMC Bioproducts) and electrophoresed with 100V for 80 min.

Products were visualised by ethidium bromide staining and UV-illumination.

iii.Cloning and sequencing of RT-PCR-products The RT-PCR products representing KIT exon 16-19 were purified by extraction from 2% agarose gels using the QIAEX gel extraction kit (QIAGEN) and cloned into the 41 pUC 18 vector using the Sureclone ligation kit (Pharmnnacia Biotech). Plasmids were isolated using the QIAFilter plasmid Midi kit (QIAGEN). Cloned plasmid inserts were sequenced using dye primer chemistry. Each cycling reaction was prepared with plasmid template DNA and ready reaction mix containing fluorescently labelled M13 forward or reverse primer as described in the ABI Prism protocol P/N 402113 (Perkin Elmer). Cycling and sample pooling were performed using a Catalyst 800 Molecular Biology Workstation (ABI) following the instruments user manual (Document number 903877, Perkidn Elmer). The resulting extension products were purified, loaded and analysed using the 377 ABI Prism sequencer as described by the 1o instrument protocol P/N 402078 (Perkin Elmer).

iv Results and discussion A 424 bp fragment including KIT cDNA exon 16-19 was amplified from all pigs. The Hampshire pigs did not show any additional products whereas the Large White pigs (eight tested) all showed a 301 bp truncated cDNA fragment (Fig Sequence 15 analysis revealed the 424 bp fragment was identical in the two breeds whereas the whole exon 17 (123 bp) was missing from the 301 bp fragment Apparent differences between individuals regarding the relative amounts of these two products may have been caused either by different genotypes containing differing numbers of copies of the KIT gene sequence, individual differences in mRNA expression levels or random RT-PCR effects.

The two upper fragments present in Large white pigs represent heteroduplexes between the 301 and 424 bp fragments (Fig. This was shown by an experiment where these slow migrating fragments were generated by pooling homoduplexes of the 424 and 301 bp which were then heat denatured and cooled to 25°C. Moreover, cloning of the lower heteroduplex fraction of a Large White pig resulted in clones with insert length corresponding to either of the two homoduplexes.

Example 9 PCR Amplification and Sequencing of KIT Exon 17-Intron 17 Splice Site) 42 i. PCR to produce DNA Sequencing Template A 175 bp region including the boundary between exonl7 and intronl7 of the KIT gene was amplified for sequence analysis using forward primer KIT21 GTA TTC ACA GAG ACT TGG CGG C and reverse primer KIT35 AAA CCT GCA AGG AAA ATC CTT CAC GG PCR was carried out on a DNA thermal cycler (Perkin Elmer 9600) in a total volume of 20 l containing 25 ng genomic DNA, 1.0 mM MgCl 2 50 mM KC, 10 mM Tris-HCI, pH 83,200 pM dNTPs, 0.5 U AmpliTaq Gold (Perkin Elmer) and 10 pmol of both KIT21 and KIT35 primer. To activate AmpliTaq Gold, initial heat denaturation was carried out at 94 0 C for minutes followed by 32 cycles each consisting of 45 sec at 94 0 C, 45 sec at 55°C and sec at 72 0 C. The final extension lasted for 7 min at 72°C. PCR products were cloned into vector pUC 8 using the SureClone ligation kit (Pharmacia Biotech).

ii. Preparation ofPlasmid DNA Plasmid DNA was purified from overnight bacterial culture using the Jetstar plasmid midi kit (Genomed) and the resulting DNA diluted to 150 ng/pl.

iii. Sequencing of plasmid DNA DNA was sequenced as in example 8.

iv. Results A portion of the DNA sequence from exon 17 and intron 17 of the KIT gene was determined and compared between animals with each of these three alleles. Figure shows that the Iallele carries a splice site mutation at position 1 of intron 17. This G to A base substitution is present in one of the two gene copies carried on each chromosome. The base substitution occurs in the invariant GT dinucleotide which characterises 5' exon/intron boundaries. Analysis of the allele showed the splice site mutation was not present in either the normal (KITI) or duplicated copy of the gene (KIT2). We have found the splice site mutation is unique to the I alleles, and therefore makes it possible to distinguish the I-KIT2 sequences.

Example Testing For the Presence of the Splice Site Mutation with PCR RFLP To easily test for the presence of the G to A splice site mutation, restriction endonuclease Maf (CATG) was used to exploit the point substitution identified at position 1 ofintron 17 (Figure The Ma recognition sites in the fragment s amplified from KIT and the expected restriction products are illustrated in Figure 6A and 6B respectively.

i PCR to produce DNA for RFLP Test The PCR to produce DNA for RFLP analysis was performed exactly as described in example 9.

ii. Restriction Enzyme Digestion and Electrophoresis The PCR amplification product is 175 bp in length. To test for polymorphism at position 1 ofintron 17, digestion reactions were set up as below l PCR amplified DNA 10 X NEBuffer4 0.1 ll BSA 100 pg/ml 0.1 plNMaf l0 U/pl 5.8 pd (1 X NEBuffer 4 (New England Biolabs) contains 50 mM potassium acetate, 20 mM Tris acetate, 10 mM magnesium acetate and I mM DT). Following incubation at 37"C for 90 minutes each 10 pl reaction volume had 2 pl of loading dye added and the mix loaded on a 8% native polyacrylamide gel (Protogel, 375:1 acrylamide:bisacrylamide, National Diagnostics, Atlanta) in 0.5 X TBE (44.5 mM Tris pH 8.0,44.5 mM boric acid and 0.5 mM EDTA) and electrophoresed for 3 hours at 200V in a vertical slab unit (SE600 Hoefer Scientific Instruments). Products were visualised by ethidium bromide staining.

iii Results A PCR RFLP protocol was designed to test for the presence of the splice site mutation as the substitution occurs within the recognition site for restriction endonuclease MaH. Figure 6B illustrates that presence of the G to A base substitution at position 1 of KIT intron 17 results in restriction at each of two NMaII recognition sites within the 175 bp DNA fragmentL Following electrophoresis, this results in fragments of sizes bp, 54 bp and 41 bp. Where the splice site mutation is absent however, incubation with NMI[ results in digestion only at recognition site 1. Following electrophoresis this results in fragments of 134 bp and 41 bp. The invariant NIaH recognition site 1 serves as an internal control to ensure complete digestion has taken place. Results of lOthis PCR RFLP analysis are illustrated in Figure 6C. Analysis was performed on fragments amplified from clones which either carry the splice site mutation (lane 2) or carry the normal splice site sequence (lane Lane 4 shows the result of analysis where DNA amplified from the genomic DNA of a coloured animal was used. Lane shows the resulting bands where a white animal was tested- The test was used to 2 oanalyse 121 individuals from seven different breeds of pig. The splice site mutation was found only in the 97 animals with the dominant white phcnotype or and none of the 24 coloured (f or 0) examples (see table below). This analysis confirms I and I* to be unique in that they are the only alleles to carry the splice site mutation.

Distribution of the Splice Site Mutation Between Different Breeds and Coat 25 Phenotype Breed Coat Assumed Animals Normally spliced Splice Colour Genotype' Tested K• l Mutation 2 Large White white 33 33 33 Landrace white 1- 56 56 56 Hampshire coloured i/i 5 5 0 Duroc coloured i/i 5 5 0 Pietrain coloured i/i 8 8 0 Meishan coloured i/i 5 5 0 Wild Boar coloured i/i 1 1 0 Wildd Boar white 8 8 S x Large White Totals white white coloured 89 8 24 89 8 24 White animals may be homozygous or heterozygous for the I allele 2 Presence of the splice site mutation determined by NMaI PCR RFLP test Example 11 Quantification of Normal KIT and Splice Mutant KIT (ntron 17 ntl CA) As the splice site mutation is present in only one of the duplicated regions off and not in the duplicated region of/I, the various genotypes can be expected to have the attributes described in the table below: Genotype Copies of Normal Copies of KIT Ratio of normal KIT to KIT containing the splice splice mutant KIT mutation

I/I

IA

ii I/ip 1 P r, Due to the dominance of allele I, three of the genotypes in the table are carried by white animals and therefore can not be identified by phenotypic characterisaion- Quantification of the relative amounts of the normal KIT gene and the splice mutant KIT gene allows the ratio between the two to be calculated, and therefore the genotype of individual animals predicted. This was achieved by quantification of two DNA fragments following NaUI digestion. The amount of 134 bp fragment, representative of the normally spliced KIT gene, and of 54 bp fragment, representative of the splice 46 mutant KT, were measured following electrophoresis using GeneScan software.

i. PCR to Produce DNA for Quantifiation As described in example 9 section i. The reverse primer K1T35 is labelled with the ABI fluorescent dye FAM at the 5' end.

ii Restriction Enzyme Digestion As described in example 9 section ii.

iii Electrophoresis and Ouantification ofDNA Framents Following digestion, 0.5 W of the reaction volume was mixed with 2,5 pl of deionised formamide, 0.5 Il of GS350 DNA standard (ABI) and 0.4 pl blue dextran solution before being heated to 90 0 C for 2 minutes and rapidly cooled on ice. Three pI of this mix was then loaded onto a 377 ABI Prism sequencer and the DNA fragments separated on a 6% polyacrylamide gel in 1 X TBE buffer for 2 hours at 700 V, 40 mA, 32 W. The peak area of fragments representative to both the normal and splice mutant forms of KIT were quantitated using the GeneScan (ABI) software.

is iv. Ratio Calculations The peak area value of the 134 bp fragment (normal K17) was divided by twice the peak area value of the 54 bp fragment (splice mutant K1I) in order to calculate the ratio value for each sample.

v_ Results Analysis was performed on animals from the Swedish wild pig/Large White inter=cross pedigree for which genotypes at I have been determined by conventional breeding experiments with linked markers. Figure 7 and the table below show the ratio of normal to mutant KIT calculated for animals from each of the three genotype classes, I (expected ratio i (expected ratio 2:1) and Yf (expected ratio The results are entirely consistent with the expected ratio values and indicate that the three genotype classes can be distinguished using this method.

Ratio of the Two KIT Forms in Different Dominant White Genotypes in a Wild Pig/Large White Intercross Genotype Phenotype Expected Observed Ratio Number Ratio (Normal:Mutant) Tested (Normal: Mutant)

SE

I/I white 1:1 1.15 0.075 13 FI" white 3:1 3.11 0.084 12 I/I white- 2:1 2.23 0.109 14 Figure 7 illustrates that the range of ratio values calculated for the two genotypes I and /f do not overlap. This enables animals carrying the f allele to be identified and the frequency of the allele within different pig breeds determined. Ratio values were calculated for 56 Landrace and 33 Large White animals and the results are shown in Figure 8. A clearly bimodal distribution is observed with 7 Landrace and 3 Large White individuals having a ratio value of approximately 3 or above, suggesting them to be heterozygous carriers for the f allele (genotype This means f has gene frequency estimates of 6.25% (7/112 chromosomes tested) and 4.5% (3/66 chromosomes tested) within the Landrace and Large White breeds respectively.

EXAMPLE 12 DNA Preparation DNA can be prepared from any source of tissue containing cell nuclei, for example white blood cells, hair follicles, ear notches and muscle. The procedure outlined here relates to blood cell preparations; other tissues can be processed similarly by directly suspending material in K buffer and then proceeding from the same stage of the blood procedure. The method outlined here produces a cell lysate containing crude DNA which is suitable for PCR amplification. However, any method for preparing purified, or crde, DNA should be equally effective.

Blood was collected in 50 mM EDTA pH 8.0 to prevent coagulation. 50 Rl of blood 48 was dispersed into a small microcentrifuge tube (0.5 ml Eppendorf or equivalent).

450 pi of TE buffer was added to lyse the red blood cells (haem groups inhibit PCR) and the mix vortexed for 2 seconds. The intact white and residual red blood cells were then centrifuged for 12 seconds at 13,000 g in a microcentrifuge. The supernatant was removed by gentle aspiration using a low pressure vacuum pump system. A further 4 5 0 pi of TE buffer was then added to lyse the remaining red blood cells and the white blood cells collected by centrifgation as before. If any redness remained in the pellet, this process was repeated until the pellet was white. After removal of the last drop of supernatant from the pelleted white blood cells, 100 p1 of K buffer containing proteinase K was added and the mixture incubated at 55°C for 2 hours. The mixture was then heated to 95-100°C for 8 minutes and the DNA lysates stored at -20 0 C until needed.

Reagents TE buffer: 10 mM TRIS-HC1 pH 1 mM EDTA K buffer 50 mM KCI mM TRIS-HCI pH 8.3 mM MgCI 2 Tween Prior to use for lysates, 10 pl of 20 mg/ml proteinase K (Molecular Probes Inc.) per ml of K buffer was added.

(ii) PCR Reactions were set up as follows in thin walled 0.25 ml tubes (Perkin Elmer): pl 5 pM CRC Forward primer, 4.0 pl 5 pM CRC Reverse primer, pi 5 pM K TI-REV primer, pl 5 pM KITI-FOR primer pl 2 mM dNTPs (Pharmacia); pl 35 mM MgCl2.

A wax bead (PCR Gem 50, Perkin Enler) was added and the tube placed in a Perkin Elmer 9600 thermal cycler. The tube was then raised to 80°C for 15 seconds followed by cooling to 4°C. A second set of reagents was then added to each tube as below:p1 10x buffer, 9.6 pi sterile deionised water, 0.4 pl (0.5 units) AmpliTaq DNA polymerase (Perkin Elmer); 2 pI DNA lysate.

o1 Reaction tubes were then placed on a Perkin Elmer 9600 thermal cycler preheated to 94°C and PCR carried out according to the regime indicated below:- 940C for 4 minutes; cycles of 94°C for 30 sees, 62°C for 30 sees and 72°C for 30 sees; 0°C until required.

s1 The number of cycles may vary depending upon the tissue used as the DNA source.

KIT rimers Forward Reverse GAATATTGTTGCTATGGTGATCTCC KTI-FOR CCGCTTCTGCGTGATCTTCCTG KITI-REV CRC primers Forward Reverse CTGGATGTCCTGTGTTCCCTGT CRC-FORWARD AGGTTTGTCTGCAGCAGAAGCTC CRC-REVERSE The reverse KITprimer and the forward CRC primer are labelled with the ABI fluorescent dye FAM at the 5' end.

(iii) Electrophoresis and Ouantitation of DNA Fragments 1 pI of the PCR was mixed with 2.5 pl of deionised formamide, 0.5 pl of OS350 DNA standards, 0.4 pl blue dextran solution, heated at 90 0 C for 2 minutes followed by rapid cooling on ice. 3 Pl of this mix were then loaded onto an AB1373 DNA sequencer and DNA fragments separated on a 6% polyacrylamide gel in 1 x TBE buffer for 2 hours at 700 V, 40 mA, 32 W. The fragments corresponding to the products from the KIT and CRC genes were quantitated using GeneScan software, the peak area for each of the bands being determined.

0 (iv) Results The data given in the table belowv represents the results obtained from an experiment in which DNA lysates were produced from each of 23 animals, with two PCR tests being carried out on each lysate. The ratio of KIT peak area to CRC peak area was calculated for each PCR and the average taken of those samples from the same animal.

Animal Genotype KIT/CRC peak area ratio i n 3.25 2 Ii 2.45 3 I 2.94 4 ii 1.16 ii 134 6 ii 1.20 7 li 2.18 8 Ii 2.19 9 II 2.88 ii 130 11 Ii 1.84 12 II 2.84 13 ii 1.50 14 ii 1.30.

Ii 2.07 16 ii 1.31 17 ii 1.14 18 Ii 2.02 19 Ii 1.87 Ii 2.00 21 ii 0.99 22 ii 1.15 23 11 2.80 The upper and lower limits for the ratio values from animal of the different genotypes HI, Ii and ii in this experiment are as below: Genotype

I/I

Ili i/i Uper Limit 3.25 2.45 1.50 Lower Limit 2.80 1.84 0.99 These results illustrate differentiation of the genotypes using this test.

EXAMPLE 13 The second test utilises unique sequences of DNA that are present at one end of the duplication (or both ends if the duplicated region is reversed relative to the rest of the gene or if the duplicated region does not occur in direct tandem with the nonduplicated region). Oligonucleotide primers for use in PCR are designed such that at the annealing temperatures used in the PCR process, they will anneal only to the junction regions at the end of the duplicated region. A PCR is then carried out using two pairs of oligonucleotides. One pair consists of the aforementioned primer spanning the junction region and a second primer a suitable distance away which allows amplification to occur only from I allele containing duplication. The second pair of primers allow amplification of a sequence present only as a single 52 copy in the haploid genome. The product of this reaction, carried out in the same tube, functions as an internal standard as in the previous test. The ratio of product from the reaction specific to the junction region is measured relative to that from the single copy control sequence.

In this test there is a larger difference between the predicted ratios of the products from the different genotypes. The relative levels of product and their ratios are illustrated below:- Junction Control Genotype Product Product Ratio II 2 2 1:1 H 1 2 1:2 ii 0 2 0:2 These larger ratios allow greater differentiation between the ranges of results obtained from the different genotypes, reducing risks of miss-scoring animals.

EXAMPLE 14 DNA Preparation DNA can be prepared as described in example 12.

(ii) PCR Reactions were set up as follows in thin walled 0.25 ml tubes (Perkin Elmer): 2.0 pl 5 mM K7/DEL2-FOR primer; pl 5 mM K7TDEL2-REV primer, il 2 mM dNTPs (Pharmacia); 1.2 pl 25 mM MgC12 p 1 Ox buffer (without MgCl2) 0.1 pl (0.5 units) AmpliTaq DNA polymerase (Perkin Elmer); pl DNA lysate; 9.7 pi sterile deionised water.

Reaction tubes were then placed on a Perkin Elmer 9600 thermal cycler and PCR carried out according to the regime indicated belowfor 1 minute; 3 cycles of 95°C for 15 sees, 50°C for 20 sees and 72*C for 40 sees; 27 cycles of 94*C for 15 secs, 50*C for 20 sees and 72°C for 50 sees; 72*C for 5 minutes; 4°C until required.

The number of cycles may vary depending upon the tissue used as the DNA source.

KITprimers Forward GAAAGTGA(CrT)GTCTGGTCCTAT(C/G)GGAT K1DEL2-

FOR

Reverse AGCCTCACTTGATCATCTTGTAG KIDEL2-REV (iii) Electrophoresis 1i 1 1 l of the PCR product was mixed with 3 pl loading buffer (95% deionised formamide, 10mM NaOH, 20mM EDTA, 0.05% bromophenolblue, 0.05% Xylene-cyanol), heated to 95°C for 3 minutes followed by rapid cooling on ice.

The sample was then loaded on an 8% native polyacrylamide gel (Protogel, 37.5:1 Acrylamide:bisacrylamide, National Diagnostics, Atlanta) in I x TBE buffer (89mM Tris, 89mM boric acid, 2mM EDTA.Na2). The DNA fragments were separated by electrophoresis for 4.5 hours at 6W with a constant temperature of and 0.6 x TBE as running buffer in a vertical slab unit (SE600 Hoefer Scientific Instruments, San Francisco).

(iv) Visualisation of DNA fragments by silver staining After electrophoresis the gel was incubated, with gentle agitation, in the fix 54 solution for 20 minutes or until the tacking dyes were no longer visible. The gel was rinsed three times (2 minutes each with agitation) in deionised water. The gel was then incubated in the staining solution for 40 minutes, with gentle agitation, followed by a brief wash (5-10 seconds) in deionised water and direct transfer to the developing solution. The gel was incubated in the developing solution until bands were clearly visible and then the development was terminated by adding an equal volume of fix solution. Finally, the gel was rinsed for 2 minutes in deionised water.

Reagents Fix solution: 10% glacial acetic acid in deionised water Staining solution: 2 g silver nitrate (AgNO3) 3 ml 37% formaldehyde 2 litres deionised water Developing solution: 60 g sodium carbonate (Na2CO3) dissolved in 2 liters deionised water. Immediately before use add 3 ml 37% formaldehyde and 400 ml sodium thiosulfate (10 mg/ml).

The solution should be at a temperature of 10-12°C when used.

Results This SSCP analysis reveals an informative polymorphism so far only found in

J

animals with the dominant white phenotype (Fig. In lanes 1 to 8 the analysis was carried out on DNA from Swedish Landrace pigs carrying the dominant white colour and in lanes 9 and 10 DNA was from wild pigs of wild type colour. The polymorphic bands are indicated. The polymorphism is characterised by two unique fragments only present in animals carrying a duplicated KITgene of allele type L. The fragments represent heteroduplexes of DNA strands from PCR products of unequal length representing the duplicated and non-duplicated copy of the KIT gene. The results ofa screening test with this marker using 40 unrelated animals representing five breeds and 190 F2 animals from a Large White/Wild pig intercross are presented in the table below: BREED COLOUR NO. OF HETERODUPLEX

ANIMALS

NOT

PRESENT PRESENT SWEDISH WHITE 10 10 0

LANDRACE

SWEDISH LARGE WHITE 8 8 0

WHITE

SWEDISH COLOURE 10 0 HAMPSHIRE

D

SWEDISH COLOURE 10 0 DUROC D WILD PIG COLOURE 2 0 2

D

LARGE WHITE/ WHITE 131 106 WILD PIG PATCH 9 0 9 INTERCROSS COLOURE 50 0

D

The results show that this particular polymorphism is very closely associated with the presence of the KIT duplication. It is not completely associated with the duplication as some white animals did not show the heteroduplex pattern. The polymorphism is therefore an example of a closely linked genetic marker which by itself or in combination with other linked markers can be used to differentiate genotypes as regards the dominant white coat colour.

EXAMPLE i) DNA extraction 56 DNA was prepared as in example 12.

ii) PCR Reactions were set up in 0.25ml thin walled reaction tubes (Perkin Elmer) as follows: 0.51l 5 jM K/TDELl-FOR primer 5 M KITDELl-REV primer 1.Op1 2mM dNTPs (Pharmacia) 1.l 15mM MgCl2 1.Ol I OX buffer 4 .9l Sterile distilled water 0.1 p AmpliTaq DNA polymerase DNA lysate Reaction tubes were then placed in a Perkin Elmer 9600 thermal cycler and PCR carried out according to the regime 94*C for 4 minutes; 21 cycles of 94"C for 30 sec, 60"C for 30 sec, and 72C for 30 sec 72*C for 4 min; 4"C until required.

The number of cycles used may vary depending on the tissue used as the source of the DNA.

Primers forward TGTGGGAGCCTCTCTCTCTTAGG KTDEL1-FOR reverse CCAGCAGGACAATGGGAACATCT KT1DEL1-REV The reverse primer was labelled with the ABI fluorescent dye FAM at the 5' end.

57 iii) Electrophoresis and quantitation of DNA fragments 1l of the PCR was mixed with 1.5fl of deionised formamide, 0.25 l of GS350 DNA standards, 0.25pJ loading buffer (50mg/ml blue dextran, 25mM EDTA) and heated at 90C for two minutes followed by rapid cooling on ice. 1.75pl of this was then loaded onto an ABI 377DNA sequencer and DNA fragments separated on a 4.12% polyacrylamide gel in lx TBE buffer for two hours at 3000V, 200W and 48'C. The 97bp and 93bp fragments corresponding to the products from the KIT gene template lacking the deletion and containing the deletion respectively were quantitated using GeneScan software, the peak area for each of the bands being detennined- Results The data given in the table below represents the results obtained from an experiment in which DNA lysates were produced from each of 20 animals of known genotype with one PCR test being carried out on each lysate. The ratio of the peak area of the product from the DNA template not containing the four base pair deletion to-that containing the deletion was calculated.

ANIMAL GENOTYPE Non del/del peak area ratio 1 I 1.347 2 II 1.21 3 H 1_33 4 II 2.267 H 0.444- 6 II 0.713 7 H 8.387 8 H 0.994 9 H 1.673 II 1.056 11 Ii 1.751 1 li 1.73 13 Ii 1.83 14 1 0.631 i 1.975 16 I 2.147 17 li 1.901 18 li 1.749 19 l 2.103 Ii 2.026 For this small sample the value of 1.5 which is midway between the predicted ratio values for each genotype (expected ratio=2 for i and 1 for might be used as the dividing line for scoring the animals to either genotype. It can be determined from the table that 7/10 Hand 9/10 li are identified as the correct genotype.

Examnle 16 Sequencing of KIT cDNA clones mRNA was isolated from peripheral blood leukocytes from white (Landrace/Large White) and coloured (Hampshire) pigs using the Message Maker mRNA isolation system (Gibco BRL) with one mRNA selection from total RNA. 100ng poly(A) mRNA was revese-transcribed with random primers (First-Strand cDNA Synthesis kit, Pharmacia Biotech) and the product was used at a 1:10 dilution for RT-PCR using the proof-reading Advantage KlenTaq Polymerase (Clontech) according to the manufacturer's recommendation. The following primers were used to amplify almost the entire coding sequence and some of the 5' untranslated region: KIT40 TCT GGG GGC TCG GCT TTG C) corresponding to the 5'untranslated region and KIT22S TCA GAC ATC TTC GTG GAC AAG CAG AGG) corresponding to exon 21; both primers had been designed using consensus sequence of the human and mouse KIT sequences in the GENBANK database. The RT-PCR products were gel purified and cloned using the pGEM-T vector system (Promega). Plasmid clones were sequenced using a set of internal primers and the ABI Pris T M dRhodamie Terminator Cycle Sequencing Kit (PE Applied Biosystems). Two subclones representing each type of KIT sequence were initially sequenced and in those-cases where a discrepancy was observed (possibly due to PCR errors) additional clones were sequenced over those particular nucleotide sites. RT-PCR analysis of KIT exon 16-19 was carried out with the primers KITIF (5'-TCR TAC ATA GAA AGA GAY GTG ACT C) and KIT7R(5'-AGC CTT CCT TGA TCA TCT TGT AG).

Results The sequence of the KIT gene coding region derived from an animal of the Hampshire Breed is shown in Figure 10. Differences between KIT cDNA sequences cloned from a Hampshire and a Yorkshire/Landrace pig, respectively are shown in the table below. The sequence comparison includes the whole open reading frame 2919 bp, except for the last 27 bp occupied by the reverse PCR primer. Exon and base pair position number as well as amino acid codon are given for each difference. Polymorphic bases are shown in bold. A dash indicates identity with the Hampshire allele.

Breed Coat Mannead. Sequoia Sp~dna Sp&Iq Rims S Zion B Exa 6 Zion 6 zion B tong KiW zo zi Colour Gumolp yarat -m 821 123 978 84' 1008 1dM 18 19 .14 17 25 w67 Hampshire wioure! 14 KIT! '0101 Normal wand AGO ACA MAC GOA GAG ACO caT oco Arg Itt Asa Gly Gin i pro Mla Yorkshirv Vwte 111' urfI'zi Normal Pnral MT 0CC CM ACA CCC 0Th /Lmnt Aim City Giun i Pro Val KI*I0202 Narmal normal AAG ACO G OAk ACA CCC 0Th Ar; 7hr Ohs iTu Pro V.1 KITZ'101 Normal skippe MAT OW GMA ACA CCC 070 An fly Glu iTk pro Val 1( 1p. Skipped skiped MAT 0CC GMA ACA CCC 070 Variant Amn Oly GiW iTh 2 Genot V/I, 1/0 or 10/10 MOere by th pig being a mow tha got it1001% wte furow foilow n rdtng to allHamir (iI bl 2nTh skipping of eon 14 (151 bp) caus a nonsense translation wit termdnation. atposlton 2161.

Example 17 DNA preparation Genomic DNA was prepared as described in example 12.

PCR

A 158bp fragemnt covering 99bp of the end of exon 19 and 59bp of the KIT gene was amplified using forward primer LA93 (5'-GAG CAG CCC CTA CCC CGG AAT GCC AGT TGA-3') and reverse primer KIT56 (5'-CTT TAA AAC AGA ACA TAA AAG CGG AAA CAT CAT GCG AAG PCR was carried out on a Perkin Elmer 9600 Thermal Cycler in a total volume of 20il containing 25ng genomic DNA, MgCl2, 50mM Kcl, 10mM Tris-HCI, pH 8.3, 200M dNTPs, 0.5u AmpliTaq Gold (Perkin Elmer) and 10 pmol of both LA93 and KIT56 primer. To activate AmphiTaq Gold, initial heat denaturation was carried out at 94 0 C for 10 minutes followed by 32 cycles each consisting of 45 sec at 94°C, 45 sec at 55 0 C and 45 sec at 72 0

C.

Restriction digestion and electrophoresis The PCR amplification product is 158 bp in length. To test for polymorphism at position 93 of this product (corresponding to position 2678 of the KIT cDNA sequence) digestion reactions were set up and incubated as below: l PCR product 1.Opl 10x reaction buffer 3 (New England Biolabs) 02pl AcIl 2.8p deionised water Following digestion at 37C for 120 minutes each 0ll reaction volume had 2il of loading dye aded and the mix was loaded on an 8% native polyacrylamide gel (Protogel, 37.5:1 acrylamide:bisacrylamide, National Diagnostics, Atlanta) in 0.5 x TBE (44.5 mM Tris pH8.0, 44.5 mM boric acid and 0.5 mM EDTA) and electrophoresed for 3 hours at 200v in a vertical slab gel unit (SE600 Hoefer Scientific Instruments). Products are visualised by ethidium bromide staining.

Results The reverse primer is designed such that an Acid site is introduced into the amplified sequence. This results in digestion of amplicon with Acil releasing a fragment of 23bp that allows confirmation of the digestion process. Digestion of the remaining 135 bp fragment into fragments of 92 and 43 bp is dependant on the nucleotide at the position corresponding to position 2678 of the KIT cDNA sequence. T at this position prevents digestion while a C at this position allows digestion. Gel resolution is not sufficient to allow resolution of the 23bp fragment but comparison to undigested product allows confirmation of the process.

10 Figure 11 illustrates the results obtained with animals of a range of genotypes.

The test was used to analyse a total of 66 unrelated individuals from seven breeds of pig. The results are shown in the table below: Breed No. KIT Genotype' Genotype at pos'n 2678 C/C CIT T/T Hampshire 4 Vi 1 1 2 Polish Wild Boar 13 ii 0 1 12 Duroc 11 /i 0 1 Pietrain 1 ii 0 0 1 Swedish Wild Boar 1 i/i 1 0 0 Swedish Landrace 12 1I 0 0 12 I/T 0 2 3 Swedish Yorkshire 14 I/I 0 1 13 I/ 0 1 4 1 Genotype based on MaI RFLP analysis as described in example 11.

Example 18 Determination of genotype at the Ilocus using a rapid DNA based test Crude DNA lysates were prepared from hair samples from animals of three breeding lines, a Hampshire based line, a Large White line, and white animals from a cross bred line originally produced from the two former lines. Four hair follicles were placed into 00l of K buffer (50mM KC1, 10mM Tris-HCl pH 83, 2.5mM MgC2, 0.5% w/v Tween 20) and pl Proteinase K (15 mg/ml) (Boehringer) added. This mix was incubated for 2 hours at 55"C followed by 16 min at 95C. DNA was also prepared as described in example 12.

Allelic discrimination reactions were set up using the PE Applied Biosystems to TaqMan m system. 25pl reactions contained the primers E19FOR GAGCAGCCCCTACCCCGGAATGCCAGTGA-3') and E19REV CTITAAAACAGAACATAAAAGCGGAAACATCATGCGAAGG-3') at 300nM, 8% glycerol IX TaqMan T buffer A (PE Applied Biosystems), 5mM MgCI 2 200LM dATP, dGTP, dCTP and dUTP, 0.65 units AmpliTaq Gold T m (PE Applied Biosystems), 0.25 units AmpEraseu UNG (PE Applied Biosystems) and the TaqMan probes E19PC CATACATITCCGCAGGTGCATGC-FAM) and E19PT TCATACATITCCACAGGTGCATGC-TET) at a concentration of 100mM. -I lp. of crude lysate DNA was used as template. PCR amplification was carried out using a PE9600 thermal cycler (PE Applied Biosystems) or a the ABI7700 Prism (PE Applied Biosystems) with a thermal cycling regime of 50°C for 2 min followed by 95°C for 10 min followed by 40 cycles of 950C 15 sec, 62°C 1 min. 8 control samples of each homozygote genotype, 2678C and 2678T, and 8 no template controls where deionized water was substituted for template controls were used per 96 well plate. Allele identification based on these reactions was carried out using the allelic discrimination function of the ABI7700 Prism (PE Applied Biosystems).

Results The test was used to analyse a total of 20 unrelated individuals from four breeds of pig.

The results are shown in the table below: Breed No. Assumed KIT Genotype at pos'n 2678 Genotype 64 C/C C/T TfT Hampshire 5 i/i 1 1 3 Landrace 5 1/I 0 0 Duroc 5 i/i 0 0 Pietrain 1 i/i 0 0 Example 19 Complete cosegregation of the Belt coat colour locus and KIT Method Hampshire pigs have a characteristic coat colour phenotype with a white belt on a solid black background. Belt is determined by a dominant allele The segregation of the Belt locus was investigated in a backcross between Hampshire (Be/Be) and Pietrain (be/be) pigs. FI sows (Be/be) were back-crossed to pure-bred Pietrain (be/be) boars. DNA preparations were carried out exactly as described in Example 3.

KIT exon 19 PCR RFLP i) PCR to produce DNA for the RFLP test A 158 bp fragment covering 99 bp of the 3' end of exon 19 and 59 bp of intron 19 of the KIT gene was amplified using the following primers: forward LA93 GAGCAGCCCCTACCCCGGAATGCCAGTTGA-3' and reverse KIT56(5' -CTITAAAACAGAACATAAAAGCGGAAACATCATGCGAAGG PCR was carried out in a total volume of 20 pl containg 25 ng genomic DNA, mM MgCI2, 50 mM KC1, 10 mM tris-HCl, pH 8.3,200 iM dNTPs, 0.5 U AmpliTaq Gold (Perkin Elmer) and 10 pmol of both LA93 and KIT56 primer. To activate Amplitaq Gold, initial heat denaturation was carried out at 94 0 C for 10 minutes followed by 32 cycles each consisting of 45 sec at 94 0 C, 45 sec at 55°C and 45 sec at 72 0

C.

ii) Restriction enzyme digestion and electrophoresis The PCR amplification product is 158 bp in length. To test for polymorphism at position 93 of this product, digestion reactions were set up and incubated as follows: pi PCR amplified DNA 1.0 llOXNEBuffer3 0.2 pAca (5 U4/pl) 2.8 1 (1 X NEBuffer (New England Biolabs) contains 100 mM sodium chloride, 50 mM Tris-HCI, 10 mM magnesium chloride, and 1 mrM DTI). Following digestion at 3T7C for 120 minutes, two p1 loading dye was added to each sample and the mix loaded on a 12 native polyacrylamide gel in 0.5 TBE (44.5 mM Tris pH 8.0, 44.5 boric acid and 0.5 mM EDTA) and electrophoresed for 3 hours at 200 V in a vertical slab unit. Products were visualised by Ethidium bromide staining.

Results KIT nucleotide 2678 is polymorphic and a C or T occurs at this position. The presence of a C creates a restriction site for Aci I which is absent when a T is present A second Acil site has been engineered into the reverse primer KIT56 to serve as an internal control of digestion and is therefore invariant The polymorphism can be detected by a simple PCR-RFLP analysis as described in the table below.

Detection of KIT single nucleotide polymorphism (SNP) at position 2678 Size in bp of DNA Nucleotide fragments after digestion C 23+43+92 T 23+135 The cosegregation of the Belt and KIT loci in this pedigree is summarised in the table below.

Cosegregation between KIT and Belt in a Hampshire/Pietrain backcross Animal No tested Phenot Belt locus KIT SNP2678 Fl sows 14 Belt Be/be C/T Pietrain sires 2 non-Belt be/be TT Offspring 41 Belt Be/be C/T Offspring 41 non-Belt be/be T/T The complete cosegregation between the Belt phenotype and the KIT polymorphism shows that this phenotype most likely is controlled by a mutation at the KIT locus.

This means that detection of KIT polymorphism can be used to identify animal products derived from Hampshire pigs since the Belt is the most important breed determinant in Hampshire pigs. It is likely that the Belt phenotype present in Saddleback and Hannover-Braunschweig pigs is controlled by the same locus.

Example Determination of the sequence of the 5' untranslated and 5' coding region of the aMSHR gene The entire coding region of the aMSHR gene was determined and compared between pig breeds known to carry the different at the E locus, Eh, Et and f. Hampshire carries E h and has a solid black body interrupted with a white belt. This belt is the result of another coat colour locus. The Wild Boar which carries allele If has a wildtype phenotype while the Pietrain breed carries allele EF and is characterized by having black spots on a white body.

PCR to produce DNA for clone construction The entire coding region of the aMSHR gene was amplified from genomic DNA using primers EPIG10 and EPIG16. These primers have sequence: EPIGI0 5' GGT CTA GAT CAC CAG GAG CAC TGC AGC ACC 3' EPIG16 5' GGG AAG CTT GAC CCC CGA GAG CGA CGC GCC 3' PCR was carried out on a DNA thermal cycler (Perkin Elmer 9600) in a total volume 67 of 20 l containing 25 ng genomic DNA, 1.5 mM MgCl2, 50 mM KC1, 10 mM Tris- HC, pH 8.3, 200 pM dNTPs, 5.0 DMSO (dimethyl sulfoxide), 0.5 U AmpliTaq Gold (Perkin Elmer) and 10 pmol of both EPIG10 and EPIG16. To activate AmpliTaq Gold, initial heat denaturation was carried out at 96°C for 10 minutes followed by 32 s cycles each consisting of 45 sec at 94°C, 45 sec at 55°C and 45 sec at 72 0 C. The final extension lasted for 7 min at 72°C.

Cloning of PCR products To facilitate cloning of PCR products, both primers were designed with restriction endonuclease recognition sites located.at the 5' end. Primer EPIGI0 has sequence to TCTAGA which is cut using enzyme Xbal and EPIGI6 contains sequence AAGCTT which is cut using enzyme HindlI. Following PCR as described above, the entire reaction volume was electrophoresed and purified using the Qiaex II gel extraction kit.

following the manufacturers instructions (Qiagen). The purified PCR product was digested prior to ligation as follows: is PCR product 17.0 pl Hindil (Amersham) 1.0 pl Xba (Amersham) 1.0 pl xl0 reaction buffer M (Amersham) 3.0 l xl0 bovine serum albumin (Amersham) 3.0 pl H20 5.0 pl The reactions were incubated at 37 degrees C for 16 hours before the digested DNA was purified by passage through a QIAquick spin column following the manufacturers instructions (Qiagen). PCR products were ligated into 100 ng of vector pRc/CMV (Invitrogen) using 400 U T4 DNA ligase (New England Biolabs) in a total reaction volume of 20 pl containing 10 mM MgCl, 50 mM Tris-HCl, pH 7.5, 10 mM dithiothreitol, 1 mM ATP and 25 pg/ml bovine serum albumin. Ligation reactions proceeded at 16 degrees for 16 hours.

Preparation and Sequencing ofPlasmid DNA Plasmid DNA was purified from overnight bacterial culture using the Jetstar plasmid midi kit 50 (Genomed) and the resulting DNA diluted to 15 ng/pl. Dye terminator 68 cycle sequencing was performed using AmpliTaq DNA polymerase in accordance with the ABI Prism protocol P/N 402078 (perkin Elmer). Cycle sequencing reactions comprised 1.6 pmole of either T7 or SP6 sequencing primer (Promega), 15 ng of plasmid DNA and the terminator ready reaction mix (Perkin Elmer). The cycle sequencing reactions were performed in a GeneAmp 9600 machine over 25 cycles, each consisting of 10 sec at 96 0 C, 5 sec at 50"C and 4 min at 60°C. Extension products were purified for gel separation using ethanol precipitation, loaded and run on a 377 ABI Prism DNA sequencer as described by the instrument protocol (Perkin Elmer protocol P/N 402178).

Results A 2bp insertion was identified in the aMSHR gene of pigs of the Pietrain breed which carry the allele between nucleotide positions equivalent to 66 and 67 in the Wild Boar aMSHR sequence. This results in a shift in the translation frame and creates a TGA stop codon at nucleotide positions equivalent to 161 to 163 in the Wild Boar zMSHR sequence. The 5' portion of the aMSHR coding sequence compared between three breeds is shown below. This comparison illustrates the two base pair insertion present within the alleles carried by the Pietrain animal when compared with either the Hampshire or Wild Boar alleles.. The ATG start codon is highlighted in bold, the 3' end of primer EPIG16 is shown in italics and bases in common with the Pietrain sequence are marked with a dash. Missing bases are marked with Pietrain CGACGCGCCC TCCCTGCTCC CTGGCGGGAC GATGCCTGTG CTTGGCCCGG Meishan--- Wild Boar Pietrain AGAGGAGGCT GCTGGCTTCC CTCAGCTCCG CGCCCCCAGC CGCCCCCCCC Meishan Wild Boar Pietrain GCCTCGGGCT GGCCGCCAAC CAGACCAACC AGACGGGCCC CCAGTGCCTG Meishan Wild Boar Pietrain GAGGTGTCCA TT Meishan Wild Boar These results are also incorporated into figure la.

Example 21 69 A rapid DNA test for the presence of the 2bp insertion mutation in the porcine aMSHR gene allowing rapid distinction of the allele from all other alleles identified at this locus PCR was conducted with forward primer EPIG16 (see above) and reverse primer MC1R121A exactly as described above. The reverse primer was labeled with ABI dye Hex and has sequence: MCIR121A 5' Hex- GGA CTC CAT GGA GCC GCA GAT GAG CAC GGT 3' Following PCR cycling, 0.2 ld of the reaction volume was mixed with 2.5 pl of deionised formamide, 0.5 pl of GS500 DNA standard (ABI) and 0.4 pl blue dextran solution before being heated to 90°C for 2 minutes and rapidly cooled on ice. 1 pl of this mix was then loaded onto a 377 ABI Prism sequencer and the DNA fragments separated on a 6% polyacrylamide gel in 1 X TBE buffer for 2 hours at 700 V, mA, 32 W. The length of the resulting PCR products were determined using the GeneScan software (ABI).

Results A test was devised to assay genomic DNA directly for the presence of the identified 2 bp insertion. Primers EPIG16 and MCIR121A were used to PCR amplify 448 bp of the 5' portion of the wildtype porcine MC1R gene. To facilitate fluorescent detection of amplified products, the ABI dye HEX was covalently attached to the 5' end of primer MCIRI21A. PCR was conducted on a number of unrelated individuals from three breeds and the resulting PCR products size determined using the GeneScan software (ABI). The results are presented in the table below: Breed Number Extension Size of PCR product tested Genotype 448 bp 450 bp Pietrain 5 EIE 0 Large White 3 FPME 0 3 Hampshire 5 E/IE 5 0 Analysis of the length of PCR poducts amplified between individuals showed either a 44 bp or 450 bp product resulted The xpected 448 bp fiagment was amplified fiom each Hampshir animal, however a product 2 bp longer was detected from each Pietrain and Large White animal tested. This indicates the 2 bp insertion identified via sequence analysis to be present in the genomic DNA of Pierain and Large White, two breeds ascribed the B' allele, but not in Hampshire which carcs E.

EXAMPLE 22 In addition to the coding region of the aMSHR gene, DNA sequence polymorphism may exist between breeds within the untranslated regions (UTR). Sequence to information was collected from the 3' UTR and compared between six breeds of pigs which display a variety of coat color phentypes.

PCR to produce DNA for sequencing A 454 bp product containing 38 coding nucleotides from the 3' portion of the molecule and 416 bp of 3' untranslated region (not incluing primer binding sites) was amplified using primers EPIG3 and EPIG14. These primers have sequence: EPIG13 5' GCA AGA CCC TCC AGG AGG TG 3' EPIG14 5' CAC TGA GCC GTA GAA GAG AG 3' PCR was caried out on a DNA thermal cycler (Perkin Elmer 9600) in a total volume of 20 pl containing 25 ng genomic DNA, 1-5 mM MgCl, 50 mM KCI, 10 mM Tris- HCI, pH 83, 200 pM dNTPs, 05 U AmphTaq Gold (Pekin Elmer) and 10 pmol of both EPIG13 and EPIG14. To activate AmphiTaq Gold, initial heat denaturation was carried out at 96°C for 10 minutes followed by 32 cycles each consisting of 45 sec at 94°C, 45 sec at 550C and 45 sec at 72*C. The final extension lasted for 7 min at 72C.

Sequencing of PCR products PCR products were sequenced using dye terminator chemisty. This first requires 71 purification of PCR product free from excess dNT s and primers. This was achieved by passage of the template DNA through a QlAquick spin column following the manufacturers instructions (Qiagen). Cycle sequencing reactions were performed on ng of purified template using either EPIGI13 or EPIGI4 as described in Example I.

Results Primers EPIG13 and EPIGI4 were used to amplify a 454 bp region of DNA which comprises both the 3' terminal coding region of the MCI) gene and the immediately adjacent 3' UTRt The sequence information collected is displayed in figure 15 and two polymorphic positions were identified.

lo The first polymorphism identified is a I bp deletion common to Meishan and Large Black which occurs seven positions downstream from the stop codon at the equivalent of nucleotide position 1007 in the Wild Boar sequence. As this deletion occurs outside the translated region it is not expected to alter the amino acid composition of the resulting receptor molecule, however its influence on mRNA stability and 3' end 1- formation through endonucleolytic cleavage is unknown. It is unique to two breeds which carry E and which carry a variant of the a&ISHR gene not found in any other breed examined to date.

The second polymorphism is a base substitution at nucleotide position 1162 unique to the European Wild boar. Figure 15 shows five breeds to have a G base at this position and the Wild boar to contain an A. This sequence difference offers the possibility to distinguish the European Wild boar from the other breeds analysed with a DNA based test.

Claims (33)

1. A method for ditfferentiating animals and animal products on the basis of breed origin; or detennining aor testing the breed oarigin of an animal product or validating an animal prodnuct compriSing the steps of providing a ample of the animal prduct; and (ii) analysing the allele(s) ofoan or more breed deteninant genes present in the sample.

2. The method of claim I wherein the breed determinant is a monogenic trait '0 3. The method of claim 1 wherein the breed deteminant is a polygenic trait

4. The method of any one of claims 1-3 wherein the overt phenotypic trait is a behavioural or morphological trait. Thc method of claim 3 or claim 4 wherein the overt phenotypic trait varies qualitatively or quatitntively betwaeen breeds.

6. The method of any one of the preceding claims wherein the breed detaninant gone analysed in step (ii) is selected from any oE a coat colour gene; and/or a coat pattm gene; andlor a coat texture gene; and/or a cot density gene; andor a coat length gate; and/or an ear aspect gene; and/or a double muscling gene; and/or a horn morphology gene; and/or i) a tusk morphology gene; andlor an eye colour gene; and/or a plumage gene; and/or a beak coloudmarphology gene; and/or a vocalization barking) gene; and/or a comb or wattle gaen; and/or a gene controlling display behaviour.

7. The method of claim 6(a) wherein the coat colour gne is the KT or AxSHl gne (for example, the pig KIT or aAdUIR gene).

8. The method of any one of the preceding claims wherein the sample is a nucleic acid sample and the analysing step (ui) comprises DNA or RNA analysis. to 9. The method of any one of claims 1-7 wherein the sample is a protein sample and the analysing step (ii) comprises protein analysis. A method of detmnining the coat colour genotype of a pig which comprises: obtaining a sample of pig nucleic acid; and (ii) analysing the nucleic acid obtained in 0i) to determine which allele a alleles of the oMSHR gane islare present

11. The method of claim 8 or claim 10 wherein the analysis step (ii) comprises selectively amplifying a specific fragment of nucleic acid by PCR)- and/or testing for the presence of one or more restriction endonulease sites within the breed determinant gene(s)YMSHR gene resuiction fragment length polymorphism (RFLP) analysis); and/or determining the nucleotide sequence of all or a portion of the breed determinant gene(s)MSHR gene; and/or probing the nucleic acid sample with an allele-specific DNA or RNA probe; and/or carrying out one or more PCR amplification cycles of the nucleic acid 74 sample using at least one pair of suitable primers and then carrying out RFLP analysis on the amplified nucleic acid so obtained.

12. A method of determining the coat colour genotype of a pig which comprises: obtaining a sample of pig aM R protein; and (ii) analysing the protein obtained in step to determine the amino acid sequence at those positions associated with coat colour genotype or the size of the protein

13. The method of claim 9 or claim 12 wherein the analysis step (ii) comprises: probing the protein sample with an antibody a monoclonal antibody) specific for an allelc-specific cpitope; and/or electrophoretic analysis; and/or chromatographic analysis; and/or amino-acid sequence analysis; and/or proteolytic cleavage aalysis; and/or epitope mapping and or translating a copy of the DNA or RNA of the gene produced by PCR or other means in an in-vitro tracription/translation system

14. The method of claim 7 wherein the analysis step (ii) comprises determining the nucleotide sequence of the KIT or a~dHR gene or the amino acid sequence of the KIT or a&CIR protein. The method of claim 7 or claim 14 wherein the analysis step (ii) comprises establishing the presence or absence of at least one nucleotide change in the KITor aA&HR gene and/or their flanking regions.

16. The method of claim 10 or claim 11 wherein the determination in step (i) involves identifying the presence or absence of at least one missense mutation, insertion or deletion in the the oMSHR gene and/or it's associated flanking regidns.

17. The method of any one of claims 7, 10, 11,14 and 15 wherein the analysis step (ii) further comprises determining the association between one or more microsatellite or other linked marker alleles linked to the KUT or aMSHR gene and to particular alleles of the KIT or aMSHR gene, I S. The method of claim 17 wherein the analysis step (ii) is based on the identification of microsatellite markers present in the nucleic acid sample.

19. The method of claim 7 wherein the analysis step (ii) comprises: determining the association between one or more microsatellite or other linked marker alleles linked to the KIT or aASHR gene and to particular alleles of the KIT or aMSHR gene; determining which microsatellite or other linked marker allele or alleles are present in the nucleic acid sample. A method of determining the coat colour genotype of a pig which comprises: determining the association between one or more microsatellite or other linked marker alleles linked to the aMSHR gene and particular alleles of the aMSHR gene; (ii) obtaining a sample of pig nucleic aci&d: and (iii) analysing the nucleic acid obtained in (ii) to determine which microsatellite or other linked marker allele or alleles arc present

21. The method of any one of claims 7, 10, 11 and 14-20 wherein the analysis step (ii) further comprises the step of determining the genotype of at least one additional locus. .76

22. The method of claim 21 wherein the additional locus is an additional coat colour locus.

23. The method of claim 22 wherein the additional coat colour locus is the KIT gene locus the pig KIT gene locus).

24. The method of claim 23 wherein the KIT gene locus is analysed to determine whether it carries any polymorphism associated with Belt genotype. The method of claim 24 wherein the determination comprises RFLP analysis. 26 A method of determining the coat colour genotype of a pig which comprises: obtaining a sample of pig nucleic acid: and (ii) analysing the nucleic acid obtained in (7i) to determine whether the KiT gene carries any polymorphism associated with Belt genotype.

27. A method as claimed in Claim 26 wherein step (ii) comprises RFLP analysis

28- A method as claimed in claim 26 or claim 27 wherein a sample of pig genomic DNA is amplified using PCR and a pair of suitable primers.

29. The method of claim 21 wherein the additional locus is a breed determinant gene locus selected from any of those genes specified in claim 6. 3 0. The method of claim 21 wherein the additional locus is. a breed specific marker. 3 1. The method of claim 30 wherein the breed specific marker is a microsatellite marker.

32. The method of any one of claimns 7, 10, 11, 14-23 and 28-31 wherein the analysis step (ii) comprises PCR using at least one pair of suitable primers. 77

33. The method of claim 32 wherein the gene is the pig aMSHR gene and at least one pair of suitable primers is: aMSHR Forward Primer 1: (5'-TGT AAA ACG ACG GCC AGT RGT GCC TGG AGG TGT MSHR Reverse Primer 5: (5'-CGC CCA GAT GGC CGC GAT GGA CCG-3); or aMSHR Forward Primer 2: (5'-CGG CCA TCT GGG CGG GCA GCG TGC-3') aMSHR Reverse Primer 2: (5'-GGA AGO CGT AGA TGA GGG GGT CCA-3'); or aMSHR Forward Primer 3: (5'-GCA CAT CGC CCG GCT CCA CAA GAC-3) aMSHR Reverse Primer 3: (5'-GGG GCA GAG GAC GAC GAG GGA GAG-3').

34. The method of any one of claims 7,10, 11, 14-23 and 28-33 wherein the analysis step (ii) comprises restriction fragment length polymorphism (RFLP) analysis, for example involving digesting the pig nucleic acid with one or more of the restriction enzymes BstUI, Hhal and/or BspH. The method of claim 34 wherein the gene is the pig aMSHR gene and the analysis involves identification of a polymorphism at nucleotide position 283,305, 363, 370, 491, 727, 729 1162 or between nucleotide positions 60 and 70 or between nucleotide positions 1005 and 1010 Oof the sequence of the pig aASHR gene.

36. The method of claim 7 wherein the analysis step (ii) carrying out one or more PCR amplification cycles of the nucleic acid sample using at least one pair of suitable primers and then carrying out RFLP analysis on the amplified nucleic acid so obtained to determine the KIT or aMSEHR genotype of the pig.

37. The method of claim 36 wherein the gene is the pig aMEHR gene and the at least one pair of suitable primers is as defined in claim

38. The method of claim 30 or 31 wherein the gene is the pig KIT or aAISHR gene and the RFLP analysis is as defined in claim 28.

39. The method of any one of claims 1-9, 11, 13-19,21-25 and 29-38 whecin the animaproduct is meat processed and/or canned meat eg egg swab or washing, semen, wool or leather. The method of any one of claims 1-9,11, 1 3 -19 and 21-39 wherein the sample comprises genomic DNA, RNA or mitochndrialDNA.

41. The method of any one of claims 1-9, 11, 13-19 and 21-40 wherein the animal is a mammal pig, cattle, dog, cat horse, sheep, rodent or rabbit), fish salmon or trout) or bird (chicken or tukey).

42. A kit for differentiating animal products on the basis of breed origin; or determining or testing the breed origin of an animal product; or validating an animal product comprising one or more reagents for analysing the allele(s) of one or more breed determinant genes present in the sample.

43. A kit for determining the coat colour genotype of a pig, comprising one or more reagents for analysing the aMAFHR genotype of the pig.

44. A kit as claimed in claim 42 or claim 43 which is adapted to be used with a sample of pig genomic DNA. A kit as claimed in any on of claims 42 to 44 comprising one or more reagents for carrying out at least one cycle of PCR together with at least one pair of suitable primers.

46. A kit as claimed in claim 45 wherein the atleast one pair of suitable primers is: oMSHR Forward Primer 1: (5-TGT AAA ACG ACG GCC AGT RGT GCC TGG AGG TGT CCA T-3) 79 caMSHR Reverse Primer 5: (5'-CGC CCA GAT GGC CGC GAT GGA CCG-3'); or aMSHR Forward Primer 2: (5'-CGG CCA TCT GGG CGG GCA GCG TGC-3) aMSHR Reverse Primer 2: (5'-GGA AGG CGT AGA TGA GGG GGT CCA-3); or cMSHR Forward Primer 3: (5'-GCA CAT CGC CCG GCT CCA CAA GAC-3' aMSHR Reverse Primer 3: (5'-GGG GCA GAG GAC GAC GAG GGA GAG-3'.

47. A kit as claimed in any one of claims 42 to 46 which comprises one or more reagents for RFP analysis of pig nucleic acid- DATEDt-this 18th Day of October 2002 EIG IMPROVEMENT COMPANY UK LIMITED Attorney: IVAN A. RAJKOVIC Fellow Institute of Patent and Trade Mark Attorneys of Australia of BALDWIN SHELSTON:WATERS