AU725544B2 - The porcine heart fatty acid-binding protein encoding gene and methods to identify polymorphisms associated with body weight - Google Patents

Description

WO 97/35878 PCT/NL97/00157 THE PORCINE HEART FATY ACID-BINDING PROTEIN ENCODING GENE AND METHODS TO IDENTIFY POLYMORPHISMS ASSOCIATED WITH BODY WEIGHT

INTRODUCTION

The invention relates to the field of molecular biology as well as to the field of breeding methods for farm animals, in particular pigs. In particular the invention relates to the use of diagnostic methods derived from the field of molecular biology to be applied in breeding programmes that select animals on production traits that improve their breeding value.

By selecting animals on their breeding value calculated mainly from phenotypic measurements of production traits, breeding has greatly improved the genotype for production traits of livestock animals. Thus, traditionally, breeding programmes have selected for phenotypic characteristics of animals. However, more recently selection for genotypic characteristics that are associated with improved production traits have gained interest in the field. Selection for phenotypic characteristics entails mainly selection of the offspring or siblings or other relatives of the animals to be selected whereas selection of specific genotypic characteristics allows for earlier and specific detection of animals of interest.

Within methods that select on specific genotypic characteristics, one may distinguish between methods that detect genetic variation in genes or quantitative trait loci that are merely associated with production traits of animals and methods that detect genetic variation in functional genes that directly influence those production traits. One of the former methods is a marker assisted selection wherein polymorphisms in markers identified in a random manner are associated with production traits.

WO 97/35878 PCT/NL97/00157 2 For instance, meat production is closely linked to embryonic muscle formation, and, consecutively, to the distribution of muscle cells and fat cells. Biologically, production is concentrated in defined tissues of the animal, e.g. muscle tissue for lean meat production. In breeding programmes for optimizing porcine lean meat production, various levels of selection pressure have been applied to different tissues muscle, fat and bone). However, when selecting for lean meat, and thus the abscence of fat, one may lose certain traits that are wanted after all, i.e.

traits that are associated with taste and thus with the consumers perception of the final product.

In pig breeding programs traditionally a lot of emphasis has been put on the fat reduction because of the consumers interest in lean meat. Fat reduction is surveyed as a decrease in backfat thickness and a large reduction has been achieved since the establisment of breeding programmes in pigs. However reduction of the backfat depot also results in less intramuscular fat (IMF). This last depot is the main fat depot in meat and is positively correlated with the taste and thus the acceptance of meat (Wood et al., 1988).

To exclude the IMF depot from further reduction a marker for this trait is necessary because IMF is hardly measurable in living animals. Recently, it has been statistically shown that a single major gene for IMF deposition in pigs must be present (Janss et al., 1994), however, the sequence, location and mode of action of the putative gene were not disclosed. Here we present evidence of a muscle tissue specific candidate gene located on porcine chromosome 6 which is the heart fatty acid-binding protein (H-FABP) gene.

Genetic variation in this gene is responsible for the variation in among others IMF% and other production traits of pigs.

Fatty acid binding proteins (FABP's) are small intracellular proteins involved in fatty acid transport from the membrane to the sites of P oxidation and/or triacylglycerol or phospholipid synthesis (Veerkamp and WO 97/35878 PCT/NL97/00157 3 Maatman, 1995). Furthermore, FABP's modulated the intracellular fatty acid concentration (Veerkamp et al.,1993). Fatty acid metabolism has historically been linked to insuline resistance (Randle, 1963), and therefore mutations in FABP genes may be associated with changes in cellular insulin resistance or dependency, fatty acid oxydation and fatty acid binding. FABP's are members of a family of intracellular lipid binding proteins comprising at least eight structurally distinct types originating from: adipocytes, brain, epidermal cells, heart, intestinal cells, ileal cells, liver and myelin cells.

The heart type FABP (H-FABP) is a 15 kDa protein highly expressed in several tissues with a high demand for fatty acids, such as cardiac and skeletal muscle and lactating mammary gland. H-FABP is identical to MDGI (mammary derived growth inhibitor) a protein which inhibits growth of tumor cells in vitro (Bohmer et al., 1987). Functionally, the H- FABP can induce cardiac myocyte hypertrophy in vitro, when added to the culture (Burton et al.,1994) and also promotes functional differentiation of mammary epithelial cells in vitro (Yang et al.,1994). However, no secretion of H-FABP has been detected so far. On the other hand native and overexpression of H-FABP in mammary epithelium of lactating mice (in vivo) does not correlate with functional differentiation markers of these cells (Binas et al.,1995).

The present invention provides among others an isolated or recombinant pig H-FABP gene specific nucleic acid molucule or pig H-FABP gene specific fragments thereof comprising or hybridising to the nucleotide sequence as shown in figure 1, or its complementary sequence or the RNA equivalents thereof.

The locus of this gene is on porcine chromosome 6. The pig H-FABP gene can be assigned functions in the regulation of intramuscular fat, thereby changing the ratio of fat deposited within the muscle versus fat deposited outside the muscles, i.e. in backfat depots. Since production and deposition of fat is energy consuming and takes away energy WO 97/35878 PCT/NL9/00157 4 for other purposes, such as muscle growth, the regulation of intramuscular fat is correlated to the regulation of growth, and thus body weight and average daily gain and feed efficiency. Also, H-FABP can regulate myocyte (and thus muscle) hypertrophy and thus also muscle regeneration. Since FABP's are involved in fatty acid transport they can influence fatty acid oxidation rates, the metabolism of fatty acid derivatives in the tissue and the fatty acid composition of cells and thus of meat. Furthermore, FABP's may regulate cellular insulin dependency. Also, in pregnant animals, fat storage has an impact on embryo survival, and regulation of H-FABP will influence birth rates and littersize. Since H-FABP regulates functional differentiation of mammary epithelial cells it may be involved in regulating the quantity and composition of the milk available, thus influencing the growth and survival of newborn animals. With the present invention, the genetic variation within the pig H-FABP gene with respect to variation in regulation of expression can now be revealed and analysed for association with above production traits and physiological characteristics.

The present invention further provides a method to generate via recombinant DNA techniques an animal, such as small laboratory animals or farm animals, i.e. a pig, with additional genetic material originating from the pig H-FABP gene. Such animals may than encode wanted alleles of this gene and constitutively or transiently express allelic proteins or fragments thereof that enhance the production or physiological characteristics of those animals.

The invention further provides methods to generate proteins or (poly)peptides comprising various allelic proteins or fragments thereof derived from the pig H-FABP gene. Such peptides, or antibodies specifically directed against such peptides, may be used to influence production traits in the live animal, but may also be used in cellculture systems in vitro. Such (poly)peptides or proteins, or antibodies specifically directed against these, may also WO 97/35878 PCT/NL97/00157 be used in diagnostic test systems to select animals that express wanted forms of allelic proteins or fragments thereof encoded by the pig H-FABP gene.

The invention further provides methods localising, identifying or marking genes or alleles or quantitative trait loci, in particular those corresponding to the pig H- FABP gene, in samples, in particular biological samples, cells or tissues, such as but not limited to hair, skin or blood, of farm animals, in particular pigs, by allowing for specific amplification of genomic fragments of those genes or alleles or quantitative trait loci of pigs. Since marker assisted selection of animals is frequentally based upon geneticvariation that exists within functional genes that influence a production trait directly, i.e. genes such as the pig H-FABP that regulates fatty acid binding, one of the methods that the invention provides is a method that identifies or marks loci or genes and that can distinguish between characteristics of alleles of those genes which characteristics serve as markers in selection programmes for animals with specific versions of those genes that are directly linked with improved production traits.

The invention further provides a method wherein polymorphic restriction sites within functional genes and thus different alleles of those genes are identified by allowing for specific amplification of genomic fragments of those genes, in particular by allowing for specific amplification of fragments of the H-FABP gene. Amplification methods are well known in the art, the best known being PCR.

A short description of the PCR used herein is given in the experimental part. Other primers, enzymes and conditions can of course be applied. After amplification a suitable method of identifying wanted alleles is a restriction endonuclease treatment. Suitable restriction enzymes for pig H-FABP alleles are MspI, HaeIII or HinfI, but others may also be used. By these methods large numbers of pigs can be rapidly genotyped for studies in which genotypic variation can be 6 associated with growth characteristics and other production traits of pigs.

However, there are many other methods identifying polymorphisms in alleles, both at the nucleic (DNA/RNA) level and at the product (protein) level. In particular at the protein level there are many possibilities using immunoassays, whereas at the nucleic acid levels there are many assays which all include some kinD of hybridisation step of for instance primers or labelled nuclei acids. A very good possibility would be mismatch PCR. Primers to be used in the invention can be identified by the person skilled in the art, the sets given in the experimental part are for illustrative purposes only.

Furthermore, the methods according to the invention can be developed into diagnostic assays or kits by which selection of pigs with alleles of interest can be performed in routine screening protocols employed in breeding programmes. With such protocols better results of selection can be expected when genes responsible for regulation of commercially interesting body tissues can be rapidly identified and controlled.

In the specific case of the pig H-FABP gene, such testing protocols can be used to identify, select and breed farm animals, such as pigs, which have better production traits, such as IMF% or backfat thickness or average daily weight gain or feed efficiency, than the average animal in the population. Better production traits such as BW or daily weight gain will increase the production per year expressed as amount of meat per animal raised. A population of animals with a higher and less variable IMF will result in a more homogenous product (meat) which is also better appreciated by putative customers because of a better taste.

Furthermore, selection for higher IMF% may be possible while at the same time selection against fat deposition in other depots, such as backfat, can be K^R. performed.

H: \Bkrot\Keep\speci\21803-97.doc 18/04/00 6a Throughout the description and claims of this specification, the word "comprise" and variations of the word, such as "comprising" and "comprises", means "including but not limited to", and is not intended to exclude other additives, components, integers or steps".

a. a a a a.

a a a a.

a a. a a.

a.

a aW.

a a a a a a a a a.

a a H:\Bkrot\Keep\speci\21803-97.doc 18/04/00 WO 97/35878 PCT/NL97/00157 7

EXPERIMENTAL

The porcine H-FABP gene has been isolated, characterized and chromosomally localized. Poylymorphisms in this gene have been identified. To test the association between bodyweight (BW) and percentage of intramuscular fat (IMF animals with different polymorphisms were selected, their bodyweight was measured and the amount of IMF after slaughter was measured.

MATERIALS AND METHODS Isolation of H-FABP containing phage clones.

A porcine genomic DNA EMBL3/SP6/T7 library (Clontech Laboratories Inc. Palo Alto, CA) was screened using the plaque hybridization method (Sambrook et al. ,1989). The human H-FABP cDNA cloned in the pSP65 vector (Peeters et al.,1991) and labeled with 32P-dCTP by nick translation (Sambrook et al.,1989) was used as a probe Briefly, 500.000 plaques were transferred to replica nitrocellulose filters and incubated in denaturation buffer (1.5 M NaC1, M NaOH) for 2 minutes, neutralisation buffer (1.5 M NaC1, 0.5 M Tris-HCl pH 8.0) for 5 minutes and fixation buffer (0.2 M Tris-HCl pH 7.5, 2X SSC(0.3 M NaC1, 0.03 M Sodium citrate)) for 30 s. The filters were air dried and the DNA was irreversibly bound by baking the filters at 80 0

C

for 2 h.

The filters were prehybridized (6X SSC, 0.5% SDS, 5X Denhardt's and 100ug/ml NaOH treated salmon sperm DNA) for two hours at 67-C and hybridized in identical buffer with the addition of the radioactively probe at 67 0

C

overnight. The filters were washed four times with 2X SSC, 0.1% SDS for 30 min at room temperature. Twenty plaques that showed positive signals on both replica filters were isolated and each subjected to two additional rounds of low density plaque purification.

WO 97/35878 PCT/NL97/00157 8 DNA of these clones was isolated using the plate lysate method (Sambrook et al.,1989).

Polymerase chain reactions.

PCR amplifications were performed on 1 .1 of a 1:1000 dilution of phage DNA preparations or 50 ngr of genomic DNA in 50 .1 containing 0.2 U Super Tth polymerase (SphaeroQ, Leiden, N1) in 10 mM Tris-HCl pH 9.0, 50 mM KC1, 1.5 mM MgCl, 0.1% gelatin, 1% Triton X-100, 0.5 pM of each primer (Pharmacia Biotechnologies, Uppsala, Sweden) and 0.2 pM of each dNTP (Boehringer Mannheim, Mannheim, Germany).

After 3 min of denaturation at 94 0 C, 33 cycles of amplification were carried out: 94 0 C for 1 min, the indicated annealing temperature (Tables 1 and 2) for 1 min and 72 0 C for the time considering the length of the expected fragment (approximately 1 min for every kb).

The primer sequences as used for the poly-T microsatellite amplification are [Hex]-5'TCTGGGCTTCAACTTACTCTG3' and 5'CTAGCGCTTCAGCTCTGATTG3' PCR was performed as described before with the following cycling conditions, 20 s at 94 0 C, 40 s at 570 and 1 s at 72 0 C followed by a final extension step for 10 min at 72 0

C,

PCR products were analysed on a 8% polyacryl amide gel in an ABI373A (Perkin Elmer, Foster City, CA, USA) using Genescan 1.1.1.2 software (Perkin Elmer). Allele sizes were estimated by comparison with a commercial Tamra-labeled marker (Perkin Elmer).

DNA sequence analysis PCR#1 (Table 1) was performed on DNA of all the phage clones to detect porcine H-FABP intron 1 specific fragments.

Two of the three phage clones containing the H-FABP gene were used to subclone the various SacI and KpnI (Boehringer WO 97/35878 PCT/NL97/00157 9 Mannheim, Mannheim, Germany) restriction digestion fragments of the gene region in pBS. Unfortunately, neither of these phage clones contained exon 4 and the 3' untranslated region as detected by restriction analysis.

The intron 3/exon 4 splice junction was cloned as PCR#2 (Table 1) product using porcine (Great Yorkshire) genomic DNA as a template. The products of two independent PCR reactions were cloned to identify errors by the Super Tth polymerase upon analysis.

The 3' untranslated region was isolated using the 5'/3' RACE-PCR kit (Boehringer Mannheim, Mannheim, Germany) with porcine (Meishan) muscle cDNA as the template and porcine H- FABP exon 1 or 3 (Table 1) specific primers in combination with the provided poly-A primer.

PCR products were cloned in the pT7Blue vector (Novagen Inc.,Madison, All H-FABP (sub)clones were transformed and the recombinant plasmid DNA was isolated and purified with the Wizard Maxiprep kit (Promega, Madison, WI, The nucleotide sequence was determined by dideoxy sequencing, partially by cycle sequencing (Perkin Elmer,) or autoread sequencing (Pharmacia Biotechnologies, Uppsala, Sweden) and the analysis was performed on a ABI 373 (Applied Biosystems) or ALF DNA sequenator (Pharmacia Biotechnologies, Uppsala, Sweden) respectively.

RFLP screening Porcine genomic DNA was isolated as described (Sambrook et al.,1989) from EDTA treated blood stored at -80 0 C. One hundred ng of genomic DNA was used for PCR amplification in gl reaction as described before. The primer sequences and its corresponding product size and annealing temperature for each combination are given in table 2. Fifteen .l of the PCR reaction was used for restriction digestion with 2 units of HaeIII, Hinfl or MspI (Boehringer Mannheim) in a total volume of 20 ul. For HaeIII and HinfI the recommended buffer WO 97/35878 PCT/NL97/00157 conditions were additionally used whereas MspI was added directly to the PCR buffer. Restriction digestion fragments were loaded on a 2% (MspI) or 3% (HaeIII and HinfI) agarose (Sigma, St Louis, MO, gel and after electrophoresis the RFLP patterns were scored by two persons, independently.

Chromosomal localisation Two independently established pig/rodent somatic cell hybrid panels (Panel A: Rettenberger et al.,1994a,1994b,1994c, 1995a and Panel B: Zijlstra et al.,1994a,1994b,1994c, in prep.) were used to assign the H- FABP gene to a specific chromosome by PCR.

DNA from each cell hybrid containing porcine chromosomes in various combinations was used in PCR#3 (Table 1) which unambiguously amplified porcine H-FABP intron 3 sequences. The obtained data was statistically evaluated according to Chevalet and Corpet (1986) in comparison with the cytogenetically and/or reference loci data of both panels.

Analysis of bodyweight and intramuscular fat in relation to polymorphisms in H-FABP genotypes.

One hundred Duroc pigs were selected on the basis of the amount of IMF in their slaughtered relatives. The animals came from the test farms Someren (N=50) and Herpen Blood samples were used to isolate DNA which was used to determine the genotypes of the three polymorphisms of each animal. Of all animals BW was measured.

Animals which where not used in the regular breeding program where slaughtered and analyzed for the amount of IMF. Only animals were slaughtered. For three of these animals it was not possible to determine the genotype of one or more polymorphisms. In total 42 animals were used in the analyses.

WO 97/35878 PCT/NL97/00157 11 An additional experiment was performed to further study the production traits in relation to genetic variation in the H-FABP gene. Therefore thirteen boars and seventy-two dams were selected for this investigation from two Duroc populations housed at separate test stations. Selection was based on heterozygosity for each H-FABP PCR-RFLP. Progeny was housed in groups and fattened with ad libitum food access until slaughterweight (110 kg).

Performance traits recorded for each pig were live weight at 180 days of age backfat thickness (BFT) and for each dam the number of piglets produced alive in first (FPP) or second parity (SPP), respectively. At slaughter, meat quality traits i.e. cooking loss, drip loss, intamuscular fat percentage, minolta colour, pH and shear force were measured in a subset of the slaughtered animals.

Blood or hairroots were collected from each animal to isolate genomic DNA for H-FABP PCR-RFLP genotyping.

The final dataset comprises information from in total 2345 pigs including pedigree. For 823 pigs H-PABP genotype information is available for at least one of the PCR-RFLPS.

Analysis was done with the statistical program Statistical Analyses System (SAS, 1990) and with the program Prediction and Estimation (PEST, Groeneveld, 1990). The latter program uses family information to estimate Best Linear Unbiased Predictions (BLUP) for the influence of, in this case different genotypes on the amount of IMF and BW.

The mean amount of IMF% of the Duroc population is 3.20, its standard deviation is 0.84 (Hovenier, 1992). PEST was also used in combination with the GeneProb program (Kerr and Kinghorn, 1996) that estimates missing genotypes for animals based on genetical and or phenotypical information for the trait of interest. Thus, the effect of the H-FABP genotypes on performance, and meat quality traits was studied with the following model: WO 97/35878 PCT/NL97/00157 12 Trait= int test station*test year*test month sexe litter P(XX) P(Xx) P(xx) individual covariable residual effect where P(XX), (xx) are the estimated chance for each genotype for each animal.

BW was standarized to weight at 180 days (STD-BW) BFT was standarized to a weight of 110 kg (STD-BFT) For IMF age was included as a covariable and analysis was also performed with STD-BFT and STD-BW as covariables.

Heritability (h 2 estimates for each trait in this Duroc population were assumed to be similar to the estimates described by Hovenier et al. (1992) for the Duroc breed.

RESULTS

H-FABP gene sequence determination and analysis Twenty H-FABP positive phage clones were identified and the corresponding DNA isolated and examined for the presence of the H-FABP gene. Using PCR#1 (Table 1) three phage clones appeared to contain intron 1. The rest of the phage clones contained H-FABP pseudogene-like sequences because of the absence of intron 1 and 3 in the amplification product of PCR#1 and #2 (Table 1) respectively. Sequence analysis of these pseudogene specific amplification products showed various nucleotide substitutions in comparison with the H- FABP gene coding sequences. Furthermore a 27 bp internal duplication was detected in the PCR#3 amplification product of one H-FABP pseudogene containing phage clone (data not shown). However this particular pseudogene specific PCR#3 fragment was not detected in the main pig breeds of our panel.

Further PCR analysis of the three H-FABP gene containing phage clones revealed that neither contained the exon 4 and 3' untranslated region of the gene. Using PCR#2 WO 97/35878 PCT/NL97/00157 13 on porcine genomic DNA a 1500 bp fragment was amplified and cloned for intron 3 and exon 4 sequence analysis. The 3' untranslated region (3'UTR) was amplified on porcine whole muscle cDNA, cloned and sequenced.

The coding sequences with the flanking intronic sequences and also 1600 bp of the 5' upstream region were determined (Fig.l). The exon-intron splice junctions were located in comparison with the porcine H-FABP cDNA and the murine H-FABP/MDGI (Treuner et al.,1994) gene sequence. A potential TATA-box was located 92 bp upstream the ATG start codon and in the (3'UTR) a consensus poly-A signal sequence was identified (see Fig.l). The coding sequences showed 92%, 91%, 87% and 85% identity to the bovine, human, mouse and rat H-FABP sequences at the nucleotide level and the deduced amino acid sequence were 92%, 90%, 87% and 86% identical, respectively (Billich et al., 1988; Peeters et al., 1991; Binas et al., 1992; Claffey et al., 1987).

Detection of genetic variation A panel comprising genomic DNA of 7 pig breeds each represented by unrelated animals (see table 3) was used to detect genetic variation in the 5' upstream region, intron 2 and intron 3 of the porcine H-FABP gene. Therefore, part of the 5' upstream region was amplified on DNA of this panel using PCR (Table 2) and digested with the restriction enzyme HinfI. The HinfI digestion showed two alleles a single fragment of 256 bp (allele h) or two fragments of 197 and 59 bp (allele Similarly intron 3 (PCR#3,Table 1) and intron 2 (Table 2) were tested for genetic variation with the enzymes MspI and HaeIII respectively, and both showed genetic variation in intron 2. HaeIII showed one fragment of 850 bp (allele D) and/or fragments of 400 and 450 bp (allele Accurate size determination revealed that these three fragments were 684 bp, 278 bp and 406 bp. MspI showed a fragment of 850 bp (allele a) and/or fragments of 750 and bp Accurate size determination revealed that these WO 97/35878 PCT/NL97/00157 14 fragments were 814 bp, 703 bp and 111 bp. Both sites of genetic variation are approximately 300 (285) bp apart.

Sequence analysis of the porcine H-FABP gene sequence (Fig. 1) revealed a 25 thymidine-nucleotide (poly-T microsatellite) stretch in the first intron. To investigate genetic variation in this poly-T stretch this region was amplified by PCR. In the Duroc pigs at least 3 alleles (HI, 215-bp, H2:220-bp and H3:221-bp) were detected. Obviously, these alleles showed a complete linkage with H-FABP PCR-RFLP alleles which are located within a 1,5 kb region.

Table 3 represents the allele frequencies of the different PCR-RFLPs in the different pig breeds tested.

The mendelian inheritance pattern of the three PCR- RFLPs was analysed in a porcine family comprising 3 generations of a Great Yorkshire breed. The genotypes of the individual pigs show consistent patterns of inheritance in this family.

Chromosomal localisation The porcine H-FABP gene was chromosomally localized using a porcine H-FABP gene intron 3 specific PCR which amplified no rodent homologous. Amplification on DNA of two independently established pig/rodent cell hybrid panels and comparison with the cytogenetically (panel A and B) and reference loci data (panel A) revealed a single significant association of the H-FABP gene with chromosome 6 (Table 4) for both cell hybrid panels.

Analyses of bodyweight and intramuscular fat in relation to polymorphisms in H-FABP genotypes.

Table 5 shows the result of mean values and their standard deviations of IMF and bodyweight for different fixed effects which were taken into account in this analyses.

WO 97/35878 PCT/NL97/00157 Statistical analyses using SAS The data was analyzed using the General Linear Models procedure (GLM) from SAS (Sas, 1990). A large model (lrg) with IMF as dependant variable contained the (fixed) effects test farm, sexe, MspI, HaeIII and HinfI. Also models containing the different effects separately (ind) and one containing the combined genotype of these three polymorphisms were analyzed. Table 6 shows the significance values for the different effects for the models analyzed.

Table 9 shows the effects of the three polymorphisms on intramuscular fat%.

Statistical analyses using PEST The PEST program was used to be able to use family information in the analyses of the different fixed effects.

The used model contained the same fixed effect as the model with SAS but also contains a random animal effect. Also, a pedigree file was used containing family relations up to two generations back. Table 7, 10, 11, 12 and 13 show predicted values for the different fixed effects and their standard errors.

Hypothesis testing The different values and their standard errors, found for the fixed effects where used in a Chi-squared hypothesis test. Critical values were calculated for a 90% and 95% twosided confidence limit. Calculation was performed using formula 1 and 2. The p1, g2, s.e.l and s.e.2 are taken from PEST output.

The value tl-1/2T is taken from a confidence table and has a value of 1.96 for a 95% confidence interval.

yl-y2 Rl g2 ±tl-1/2T*V(s.e.2) s.e.2 (nl*s.e.12 n2*s.e.22)/(nl+n2-2) WO 97/35878 PCT/NL97/00157 16 tl-1/2T 1.96 (95% confidence limits) difference between two genotypes of one polymorphism Table 8 shows the difference between different genotypes of each polymorphism and its 90% and 95% critical values.

The differences between the values for the different genotypes on IMF are not significant Only the difference between the homozygote combination (aaddHH) and the heterozygote combination (AaDdHh) showed a significant difference but was not significant with the 95% two sided confidence limits.

Bodyweight is significantly different for the different genotypes in all three polymorphisms.

When the different genotype combinations are analysed for bodyweight the heterozygote combination (AaDdHh) did not differ significantly from both homozygote genotypes. Both homozygote genotypes (AADDhh-aaddHH) show a significant difference of 9.11 kg (P<0.05).

Conclusions The genotypes of the three polymorphisms tested (MspI, HaeIII and Hinfl) show a significant difference in bodyweight All three polymorphisms can be used in selection for bodyweight. The genotypes of the three polymorphisms show a distinct, albeit non-significant difference in IMF percentage. If there is a difference between different genotypes of 0.20, 50 animals of the least frequent genotype (AADDhh) would be needed to reach a significant difference of 0.2. Tables 9, 10 and 11 show that when more animals are tested, statistically significant differences among the three polymorphisms can indeed be found, for instance for IMF, backfat thickness and

BW.

WO 97/35878 PCT/NL97/00157 Also, tables 12 and 13 show that the effect on IMF, as measured by RFLP testing, can stil be found when the effects are corrected for backfat thickness and/or growth.

WO 97/35878 PCT/NL97/00157 18 Brief description of the drawings Figure 1. The porcine H-FABP gene sequence including 1632 bp of the 5'upstream region and 200 bp of the 3' untranslated region. Exons are represented by bold capital letters and the deduced amino acid sequence is shown directly beneath it. Standard one letter amino acid symbols are used. The putative TATA-box, the polyadenylation signal in the 3'UTR and the 13 nucleotide element are underlined.

The size of the nondepicted intron sequences is shown between arrowheads. The polymorphic HaeIII (GGCC), HinfI (GATTC), MspI (CAGG) sites and the polymorphic microsatellite sequence (poly-T) are depicted bold and underlined.

WO 97/35878 PCT/NL97/00157 19

REFERENCES

Archibald, A.L.,Haley, C.S.,Brown, et al. (1995) The PIGMaP consortium linkage map of the pig (Sus scrofa). Mamm.

Genome 6: 157-175.

Billich, Wissel, Kratzin, Hahn Hagenhoff, Lezius, A.G. and Spener F. (1988) Cloning of a full length complementary DNA for fatty acid-binding protein from bovine heart. Eur. J. Biochem. 175: 549-556 Binas, Spitzer, Zschiesche, Erdmann, Kurtz, Mueller, Niemann, Blenau, W. and Grosse, R.

(1992) Hormonal induction of functional differentiation and mammary derived growth inhibitor expression in cultured mouse mammary gland explants. In vitro Cell Dev. Biol.

28A:625-634.

Binas, Gusterson,B., Wallace, and Clarck A.J. (1995) Epithelial proliferation and differentiation in the mammary galnd do not correlate with cFABP gene expression during early pregnancy. Dev. Gen. 17:167-175.

Burton, Hogben, Jouannou, CLarck, Hsuan, Totty, Sorensen, Evans, R.W.

and Tynan, M.J. (1994) Heart fatty acid binding protein is a novel regulator of cardiac myocyte hypertrophy. Biochem.

Biophys. Res. Comm. vol 205 no 3:1822-1828.

Chevalet, C. and Corpet, F. (1986) Statistical decision rules concerning synteny or independence between markers.

Cytogenet. Cell Genet. 43:132-139.

Claffey, Herrera, Brecher, P. and Ruiz-Opazo, N.

(1987) Cloning and tissue distribution of rat heart fatty acid-binding protein mRNA: Identical forms in heart and skeletal muscle. Biochemistry 26:7900-7904.

Groeneveld E (1990) PEST Users'Manual. Report Department of Animal Sciences, University of Illinois, Urbana USA, 72 p.

Janss, Van Arendonk Brascamp, E.W. (1994) Identification of a single gene affecting intramuscular fat in Meishan crossbreds using Gibbs sampling. World Conference on genetics applied to livestock production, Guelph.

WO 97/35878 PCT/NL97/00157 Hovenier, Kanis, Van Asseldonk, Th. and Westerink, N.G. (1992) Genetic parameters of pig meat quality traits in a halothane negative population. Livest. Prod. Sci. 32:309- 321.

Kerr R.J. and Kinghorn B.P. (1996). An efficient algorithm for segregation analysis in large populations. J. Anim.

Breed. and Genet. 113:457-469 Peeters, Veerkamp, Geurts van Kessel, Kanda, T. and Ono, T. (1991) Cloning of the cDNA encoding human skeletal muscle fatty acid binding protein, its peptide sequence and chromosomal localization. Biochem. J. 276:203- 207.

Randle, Hales, Garland, and Newsholme, E.A. (1963) The glucose fatty acid cycle: its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet i:785-789.

Rettenberger, G.,Fries,R., Engel, Scheit, K.H.,Dolf, G.

and Hameister, H. (1994a) Establisment of a partially informative porcine somatic cell hybrid panel and assignment of the loci for transition protein 2 (TNP2) and protamine 1 (PRM1) to chromosome 3 and polyubiquitin (UBC) to chromosome 14. Genomics 21: 558-566.

Rettenberger, G.,burkhardt, e.,Adham, Engel, W., Fries, Klett, C. and Hameister, H. (1994b) Assignment of the Leydig-insulin like hormone to porcine chromosome 2q12q13 by somatic cell hybrid analysis and fluoresence in situ hybridisation. Mamm. Genome 5:307-309.

Rettenberger, G.,Fredholm M. and Fries, R. (1994c) Chromosomal assignment of porcine microsatellites by use of a somatic cell hybrid mapping panel. Anim. Genet. 25:343- 345.

Rettenberger, Bruch, Beattie, Moran, C., Fries, R. and Hameister H. (1995a) Chromosomal assignment of 17 porcine microsatellites and genes by use of a somatc cell hybrid mapping panel. Anim. Genet. 26: 269-273.

Rettenberger, Klett, Zechner, Kunz, Vogel, W. and Hameister H. (1995b) Visualization of the WO 97/35878 PCT/NL97/00157 21 conservation between pigs and humans by heterologous chromosomal painting. Genomics 26:372-378.

Sambrook,J., Fritsch, E.F. and Maniatis, T. (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

Treuner, Kozak, Gallahan, Grosse, R. and Muller, T. (1994) Cloning and characterization of the mouse gene encoding mammary-derived growth inhibitor/ heart fatty acid-binding protein. Gene 147:237-242.

Veerkamp, Van Kuppevelt,T.H.M.S.M., Maatman, R.G.H.J.

and Prinsen C.F.M. (1993) Structural and functional aspects of cytosolic fatty acid binding proteins. Prostaglandins Leukot. Essent. Fatty Acids 49:887-906 Veerkamp, J.H. and Maatman, R.G.H.J.(1995) Cytoplasmic fatty acid binding proteins: their structure and genes. Prog.

Lipid Res. vol 34 no 1:17-52.

Wood, Enser, Moncrieff, C.B. and Kempster, A.J.

(1988) Effects of carcass fatness and sex on the composition and quality of pigmeat. 34 th International Congress of Meat Science and Technology. august 29- september 2, 1988 Brisbane Australia. 562-564.

Yang, Spitzer, Kenny, Zschiesche, Li, M., Kromminga, Muller, Spener, Lezius, Veerkamp, Smith, Salomon, D.S. and Grosse, R. (1994) Members of the fatty acid-binding protein family are differentiation factors for the mammary gland. J. Cell Biol.

vol 127 no 4:1097-1109.

Yerle, Lahib-Mansais, Mellink, Goureau, et al. (1995) The PIGMaP consortium cytogenetic map of the domestic pig (Sus scrofa domestica). Mamm. Genome 9:176-186.

Zijlstra, C.,Bosma, A.A. and de Haan, N.A. (1994a) Comparative study of pig-rodent somatic cell hybrids. Anim.

Genet. 25: 319-327.

Zijlstra, C.,Bosma, de Haan, Mellink, C.H.M. and Groenen, M.A.M. (1994b) Confirmation of the chromosomal assignment of eight polymorphic microsatellites in the pig WO 97/35878 PCT/NL97/00157 22 using a panel of cytogenetically characterized pig-rodent somatic cell hybrids. Anim. Genet. 25 supl.2:58.

Zijlstra, C.,Bosma, de Haan, N.A. and Mellink, C.H.M.

(1994c) Cytogenetic and biochemical characteristics of a partly informative panel of pig-rodent somatic cell hybrids.

Proc. 11th Europ. Coll. Cytogenet. Domest. Anim., Copenhagen, 170-173.

Zijlstra, C.,Bosma, de Haan, N.A. and Mellink, C.H.M.

(in prep.) Construction of a cytogenetically characterized porcine somatic cell hybrid panel and its use as a mapping tool.

WO 97/3587 1PrrrfNn.Q'7InnIrZ'7 Table 1: 0 23 Primer sequiences and the corr~esponding size and annealing temperatures f or PCP. reactions of different reg~ions of the porcine H-FABP gene.

PC poitI primer sequlence r I size ion exon 51 GC'AGCATCACTM~rTGGACGCTTTrC PCR""1 S-1 4000 exon 5' CTIAAAGCTGATC-CTGTGTtC exon IS' IGAGGCAACTTGTTCACCTC PCR#t2 3 57 1600 exon S 'TCTTTCTCGTAAGTGCGAGTIGC exon I GA3ccAAAcTTGTT6CACCTGC PCRZ#3 3 62 1500 intro 5' GTACTGGGAGCACTCTTCACTC WO 97/35878 PCTfNL97/00157 Table 2: The primer sequences and combiLnations used for PCR- RFLP detection with the corresponding annealing RFLP I-Primer sequence 3AT TC CGTG-, G717AG 4ae77- /MsOI 5, TCAGGAATGGGrAGTATTGG :-;GGACCAGATGCCTACGCC3~ ST-Gc 'ITCTGACCAAGAGG WO 97/358 Table 3: W97/38 PCTINL97/00157 25 Allele frequencies of the porcine E-FABP gene HaeIII, MspI and the Hinf I RFLPs in unrelated animals of different pig breeds.

DL (Du [Gy HS He Pi I 34 6 5 AA 0.95 0.20 0.65 1.00 1.00 0.80 0.40 MspI JA a 0.05 0.40 0.32 0.00 0.00 0.20 0.60 a a 0.00 0.40 0.03 0.00 0.00 10.20 0.00 DD 0.05 0.20 0.06 1.00 1.00 0.40 0.00 HaeIII Dd 0.55 j0.40 050 0.00 0.00 0.20 0.20 dd 0.40 0.40 j0.44 0.00 0.00 0.40 0.80 HH 0.45 0.60 0.94 0.17 0.18 0.60 0.80 Hinf I Hh 0.50 0.20 0.06 0.33 0.55 0.20 0.20 hh 0.05 0.20 0.00 0.50 0.27 0.20 0.00 Abroviations for the various pigbreeds are DL: dutch Landrace; DU: Durac; GY: Great Yorkshire; HS: Hampshire; ME: ideishan; PI: Pietz-ain; WP: Wild Pig. The second line indicates the number of unrelated animals tested.

WO 97/35878 PCT/NL97/00157 Table 4 Chromosomal localization of the porcine hFABP gene by two independent cell hybrid panels.

The percentage concordance was detearmined by the total of equal absence/presence in both data sets divided by 20. The 9 value represents the correlation between both datasets. A gene can be assigned to a chromosome when the concordance in high and the p is more than 0.74 (syntenic) Values for between 0.59 and 0.74 give no validation of the assignment. The rest is asyntanic. Using these conditions the probability of the assignment is 97.5% correct.

PANEL A PANEL B CHR.NR locus %CON p %CON p 1 IFNA 62 0.16 46 -0.10 2 S0091 81 0.59 71 0.39 3 APOB 67 0.39 66 0.45 4 S0001 48 -0.03 66 0.34 S0092 71 0.45 50 0.06 6 RYR1 95 0.89 92 0.83 7 TNFB 62 0.33 62 0.10 8 ALB 48 0.31 62 0.36 9 S0095 57 0.40 50 0.15 S0038 57 0.28 50 0.03 11 CGT9 48 -0.03 58 0.17 12 GH 57 0.40 50 -0.03 13 S0076 48 -0.03 67 0.29 14 DAO 62 0.16 71 0.45 S0088 52 0.03 75 0.53 16 CGT6 57 0.28 58 0.10 17 ENDO 67 0.30 50 -0.15 18 CGT12 57 0.00 58 0.08 X S0022 52 0.12 75 0.54 Y SRY 76 0.39 58 0.00 Panel A: Rettenberger et al., 1994a Panel B: Zijlstra et al., 1994a WO 97/35878 PCT/NL97/00157 27 Table 5 Number of animals their mean and standard deviations (std) for intramuscular fat (IMF) and bodyweight (BW) difference from 110 kg at 180 days) for different fixed effects (effect).

Effect N IMF(%) (std) BW (kg) (std) test farm Someren Herpen Slaughter 1 2 4 6 24 18 week 5 19 6 5 7 2.77 (1.66) 2.00 (0.79) 2.02 2.25 2.82 2.20 3.09 (0.54) (0.77) (0.70) (1.11) (3.02) Slaughter month jan 35 2.31 (0.79) feb 7 3.09 (3.02) Sexe boar gilt 2.02 (0.71) 2.99 (1.85) -4.87 -6.73 -4.76 -5.83 -6.12 -8.65 -3.34 -6.13 -3.34 -3.44 -8.63 -10.27 -7.23 -1.79 -10.27 -7.26 -1.01 -4.04 -6.50 -10.27 (8.35) (7.85) (8.34) (6.81) (8.80) (11.61) (9.63) (7.83) (9.63) (7.90) (7.57) (6.78) (7.47) (8.01) (6.78) (7.53) (8.12) (9.06) (7.00) (6.78) Polymorphisms Mspl AA 5 Aa 24 aa 13 HaeIII

DD

Dd dd HinfI

HH

Hh hh 5 5 2.58 2.54 2.18 2.58 2.57 2.19 (0.69) (1.73) (0.80) (0.69) (1.80) (0.76 (0.72) (2.04) (0.69) 2.02 3.00 5 2.58 WO 97/35878 PCT/NL97/00157 28 Table 6 Values of significance for different fixed effects influencing intramuscular fat (IMF) and bodyweight (BW) using a large model (lrg) and using seperate models per fixed effect (ind) (correcting for fixed effects test farm and sexe).

Effect IMF (Pr F) BW (Pr >F) Irg ind Irg ind regio 0.07 0.99 sexe 0.02 0.03 MspI 0.52 0.38 0.39 0.03 HaeIII 0.60 0.50 0.56 0.04 HinfI 0.55 0.43 0.59 0.30 Combined 0.80 0.11 WO 97/35878 PCT/NL97/00157 Table 7 Predicted values for intramuscular fat (IMF) and bodyweight (BW) for the different fixed effects (effect) and their standard errors Effect IMF s.e. BW s.e.

sexe boar -0.61±0.18 4.78±2.52 test farm Someren 0.48±0.24 2.56±3.05 MspI* AA -0.02±0.32 -3.78±4.39 Aa 0.00±0.00 0.00±0.00 aa -0.22±0.23 5.34±3.07 HaeIII* DD 0.02±0.32 -4.32±4.36 Dd 0.00±0.00 0.00±0.00 dd -0.12±0.22 4.35±2.87 HinfI* HH -0.30±0.21 2.16±2.84 Hh 0.00±0.00 0.00±0.00 hh -0.08±0.32 -5.16±4.36 the different polymorphisms are analysed using one polymorphism in each model (the fixed effects test farm and sexe are included in all models).

WO 97/35878 PCT/NL97/00157 Table 8 Differences (diff) between genotypes and their and 95% (95) confidence limits for intramuscular fat (IMF) and bodyweight (BW).

IMF

BW

genotype diff 90 diff aa-AA -0.20 ±0.39 9.11 ±8.23 dd-DD -0.14 ±0.39 8.56 ±6.48 HfI-hh -0.22 ±0.38 7.32 ±6.43 WO 97/35878 WO 9735878PCT/NL97/00157 31 Table 9 Effect of different genotypes and the difference between the homozygotes in IMP% for each RFLP Genotype

MSPI

AA

Aa aa HaeIII

D

Dd dd HinfI

HH

Hh hh effect -0.373 -0.250 0 .000 -0.379 -0.274 0.000 0.402 0 .048 0. 000 difference -0.373 ±0.195 -0.379 ±0.191 0.402 ±0.191 p=0 0.06 0.05 0.04 WO 97/35878 PTN9/05 PCT/NL97/00157 Table 10 Effect of thickness at 110 kg different genotypes on backf at (difference from population mean in mm) Genotype Msp I

AA

Aa aa HaeITI

DD

Dd dd HinfI

HH

Hh hh effect -0.559 -0.346 0 .000 -0.598 -0.342 -0.000 0.380 0.092 0 .000 difference -0.559 ±0.209 -0.598 ±0.207 0.380 ±0.213 P=0 0.01 0. 004 0 .07 WO 97/35878 WO 9735878PCT/NL97/00157 Table 11 Effect of different genotypes on (difference from population mean days in kg bodyweight at 180 Genotype MspI

AA

Aa aa Ha elII

DD

Dd dd Hinf I

HH

Hh hh effect -2.311 -2.229 0 .000 -2.691 -2.300 0 .000 0 .404 -0.621 0 .000 difference -2.311 1.275 -2.691 1.258 0.404 1.298 P=0 0 .07 0 .03 0.76 WO 97/35878 PCT/NL97/00157 34 Table 12. Effect of different genotypes and the difference between the homozygotes in IMF% for each RFLP when corrected for backfat thickness.

Genotype MspI

AA

Aa aa HaelII

DD

Dd dd HinfI

HH

Hh hh Effect -0.226 -0.166 0.000 -0.234 -0.195 0.000 0.316 0.006 0.000 Difference -0.226 0.188 -0.234 0.184 0.316 0.182 WO 97/35878 PCT/NL97/00157 Table 13 Effect of different genotypes and the difference between the homozygotes in IMF% for each RFLP when corrected for backfat thickness and growth Genotype MspI

AA

Aa aa HaeIII

DD

Dd dd HinfI

HH

Hh hh Effect -0.245 -0.192 0.000 -0.250 -0.214 0.000 0.316 0.001 0.000 Difference -0.245 0.188 -0.250 0.184 0.316 0.182

Claims (15)

1. A method for localising, identifying or marking quantitative trait loci that are associated with production traits of pigs, using an isolated or recombinant pig H-FABP gene specific nucleic acid molecule or pig H-FABP gene specific fragments thereof comprising or hybridising to the nucleotide sequence as listed in figure 1 or its complementary sequence or the RNA equivalents thereof.

2. A method according to claim 1 distinguishing between alleles of the H-FABP gene of pigs.

3. A method according to claim 2 by detecting specific restriction sites in an allele of the H-FABP gene Sof pigs. S4., A method according to claim 3 whereby an MspI restriction site is detected. A method according to claim 3 whereby an HaeIII restriction site is detected. 0

6. A method according to claim 3 whereby an HinfI restriction site is detected.

7. A method according to any one of claims 1-6 for localising, identifying or marking genes or alleles or quantitative trait loci in biological samples, cells or tissues, such as but not limited to hair, skin or blood of pigs, by allowing for specific amplification of genomic fragments of those genes or alleles or quantitative trait loci.

8. A method using an isolated or recombinant pig H- FABP gene specific nucleic acid molecule or pig H-FABP gene specific fragments thereof comprising or hybridising to the H:\Bkrot\Keep\speci\21803-97.doc 18/04/00 37 nucleotide sequence as listed in figure 1 or its complementary sequence or the RNA equivalents thereof, for localising, identifying or marking genes or alleles or quantitative trait loci in biological samples, cells or tissues, such as but not limited to hair, skin or blood, by allowing for specific amplification of genomic fragments of a pig H-FABP gene.

9. A method according to any one of claims 7 or 8 in which the method of amplification is the polymerase chain reaction.

10. A method according to any one of claims 1-9 identifying differences between alleles of the pig that 15 are associated with differences in production traits of pigs.

11. A method according to claim 10 identifying alleles of the pig that are associated with improved 20 production traits of the pig.

12. Use of a method according to any one of claims 1- 11 in marker assisted identification of pigs or in marker assisted selection of pigs.

13. Use of a method according to claims 10 or 11 in breeding programmes.

14. A kit when used for localising, identifying or marking a quantitative trait locus on DNA in a sample, comprising at least one nucleic acid capable of hybrising to a pig H-FABP gene which comprises all or part of the nucleotide sequence as listed in figure 1 or its complementary sequence or the RNA equivalents thereof. A kit according to claim 14 wherein said nucleic acid is a primer for use in specific amplification of a E:\Simeona\Keep\Speci\21803- 9 7.doc 21/07/00 38 fragment of the pig H-FABP gene.

16. A kit according to claim 15, wherein the method of amplification is the polymerase chain reaction.

17. A kit according to claim 14 or 15 wherein said quantitative trait locus is associated to a production trait.

18. A kit according to claim 17, wherein the production trait is selected from the group consisting of intramuscular fat percentage, backfat thickness, average daily weight gain and feed efficiency. S15 19. A method according to claim 1 substantially as hereinbefore described with reference to any one of the 0* S* examples. Dated this 21 day of July 2000 PIG GENES B.V.I.O. By their Patent Attorneys GRIFFITH HACK Fellows Institute of Patent and Trade Mark Attorneys of Australia E:\Simeona\Keep\Speci\21803-9 7 .doc 21/07/00