WO2007133075A2 - Igf2 marker assisted selection for porcine reproductive traits - Google Patents

Abstract

The invention relates to the field of domestic animal breeding, especially pig breeding. Methods for increasing litter size of sows are provided, whereby use is made of a positive parental imprinting effect of a DNA region on chromosome 2.

Description

IMPRINTING EFFECTS FOR PORCINE REPRODUCTIVE TRAITS ON SSC2

FIELD OF THE INVENTION

The present invention relates to the field of pig breeding and pig production chain. Provided are methods for making sows which are capable of producing a significantly increased number of piglets per litter. This increase in prolificacy is obtained by combining the genes of a boar having a specific haplotype with the genes of a sow having a specific haplotype, whereby both haplotypes comprise an imprinted DNA region, including the IGF2 gene. Thus, methods are provided for selecting and combining a sire and/or a dam having a specific haplotype, whereby the haplotype comprises an imprinted DNA region comprising the IGF-2 gene.

BACKGROUND OF THE INVENTION

Most traits and disorders have a multifactorial background indicating that they are controlled by environmental factors as well as an unknown number of quantitative trait loci (QTLs). A paternally expressed (maternally imprinted) QTL affecting muscle growth, fat deposition and size of the heart in pigs maps to the IGF-2 (insulin-like growth factor 2) region on chromosome 2 (Jeon et al, 1999, Nature Genet. 21, 157— 158; Nezer et al, 1999, Nature Genet. 21, 155-156).

The IGF-2 locus is, thus, a locus which is described to be paternally expressed (maternally imprinted), meaning that mainly the allele obtained from the sire (i.e. the paternally inherited allele) is actively expressed in certain tissues, while the allele obtained from the dam (i.e. the maternally inherited allele) is not expressed and has therefore no effect. This phenomenon is also referred to as monoallelic expression.

Jeon et al (1999, supra) generated an intercross between the European Wild Boar and Large White domestic pigs. When offspring inherited a 'domestic IGF- 2 allele' from their father (Lp) they showed an increased amount of lean meat in their ham and loin. The heart weight was increased, while the backfat thickness was decreased.

Van Laere et al. (2003, Nature Genet. 425, 832-836) provided a biological explanation for this parental imprinting effect. They showed that there is a nucleotide substitution in intron 3 of the "domestic IGF2-allele". The mutation occurs in an evolutionarily conserved CpG island that is hypomethylated in skeletal muscle.

WO00/36143 describes how the finding that the QTL comprising at least part of the IGF-2 locus is paternally expressed can be used to devise breeding schemes to increase muscle mass and/or fat deposition of slaughter pigs. The suggestion is to use sires which are homozygous for the mutant IGF-2 allele (see page 19-20) to produce slaughter pigs which are heterozygous. Also a method for assessing the genotype of pigs at the IGF-2 locus is provided. A more extensive description of methods and kits for detecting the mutant or wild type IGF-2 allele is described in WO2004/063386. Also a commercial test is available to identify the IGF2-genotype (Dr. Van Haeringen Laboratorium; www.vhlgenetics.com). Blood, hair roots, semen or tissue samples can be tested to identify if an animal is IGF2+ / IGF2+ (homozygous for the mutant IGF-2 allele), IGF2+ / igf2- (heterozygous) or igf2- / igf2- (homozygous for the wild type IGF-2 allele).

In a recent press release by Gentec (entitled "Scientists unveil a major genetic advance in pig productivity") an effect of IGF-2 on fertility and longevity was mentioned. The suggestion was made that the locus is responsible for extra fertility and longevity, allowing "dam line selection for the production of larger litters born or the selection of a greater muscle mass in sire lines." However, it remains unclear which DNA region and allele(s) is/are favorable for fertility traits. There remains, therefore, a need for finding methods which significantly increase prolificacy of sows, so that an increased number of slaughter pigs can be produced.

Both Gentec (www.gentecweb.com) and Newsham Genetics (www.newsham.com) use the IGF2 marker in their breeding schemes, designating the mutant allele as positive '+' allele (IGF2+) and the wild type allele as negative '-' allele (igf2-). They describe that the hybrid female should receive the igf2- allele (i.e. the wild type allele) from its sire, which is exactly opposite to the instant invention, where the hybrid female receives the mutant IGF2 allele (designated herein "1 allele" or "A allele") from its sire. See Fig. 2, right hand side, where the hybrid female has the mutant allele from the sire (i.e. HapA^pat). In contrast, Gentec and Newsham Genetics describe hybrid females having the wild type allele from the sire, i.e. HapG^pat.

Genomic imprinting is a phenomenon which is complex and still little understood at the molecular level, as it involves a complex pattern of spatio-temporal gene regulation, whereby DNA methylation and histone modifications are implicated. See Tycko et al. (2006; Cytogenetic Genomic Res 113: 271-278) for a review on imprinted genes in placental growth.

The present inventors have found that the crossing of pigs having specific haplotypes can be used to significantly increase prolificacy of sows used as dams in the production of slaughter pigs. The positive effect on prolificacy was found to be caused by a parentally imprinted DNA region comprising the IGF-2 locus. In addition, certain combinations of IGF-2 alleles (or of the region comprising the IGF-2 alleles) were found to have a larger beneficial effect on litter size than others, indicating that an interaction between paternal and maternal chromosomes takes place in fertility related tissues and development. Optimal methods are provided herein, which maximize the number of piglets per litter, which in turn is used to produce an increased number of live, lean pigs.

GENERAL DEFINITIONS

"IGF-2 locus" or "IGF-2 gene" refers to the region on Sus scrofa chromosome 2

(SSC2) which comprises an allele of the insulin-like growth factor 2 gene.

"Mutant IGF-2 allele" or "allele 1" or "A allele" refers herein to an allele comprising a mutation in the IGF-2 gene, especially a single nucleotide polymorphism (SNP) in intron-3 of the IGF-2 gene whereby an G (Guanine) is replaced by an A (Adenine), as described in Laere et al (2003, supra). In the prior art the mutant allele is also designated as '+' allele, or IGF2+. "Wild type IGF-2 allele" or "allele 3" or "G allele" refers herein to an allele which does not comprising the mutation from G to A in intron-3 of the IGF-2 gene. In the prior art the wild type allele is also designated as '-' allele, or igf2-. The "genotype" or "allelic make-up" at a specific locus can be denoted by the presence of the two alleles, one derived from the father (denoted on the left), and the other from the mother (denoted on the right). The genotypes for the two alleles present in a diploid cell at the IGF-2 locus can, for example, be denoted as A/A (homozygous for the mutant IGF-2 allele), A/G or G/A (heterozygous for the lean allele, whereby the left allele is inherited from the father and the right allele from the mother) and G/G (homozygous for the wild type allele). The term "geno typing" refers to the determination of the genotype at a locus (or at several loci), using e.g. molecular methods such as PCR based methods or sequencing. The term "haplotype" refers to a set of closely linked genetic markers (such as SNPs and/or other polymorphic DNA markers) present on a chromosome in coupling-phase and which tend to inherit together as a unit (i.e. they are not easily separated by genetic recombination). Thus, the combination of polymorphisms (e.g. SNPs) in a DNA region can be used to define haplotypes. For example, Hap A/Hap A refers to a pig comprising a homozygous haplotype which carries the mutant IGF-2 allele, combined with alleles of other genes linked thereto. This haplotype can, thus, be defined by SNPs in the DNA region and/or by the pig breed comprising this haplotype. A superscript is used to designate from which parent the haplotype originates, such as HapA^pat is a haplotype inherited from the father (paternal), while HapA^mat is a haplotype inherited from the mother (maternal). The same applies to other SNPs in the region which define the haplotype.

"Haplogroup" refers to the group of pigs or pig breeding lines sharing the same haplotype or a very closely related haplotype. An "imprinted DNA region comprising the IGF-2 gene/locus" refers herein to a DNA region on SSC2 which is imprinted. In a preferred embodiment it comprises or consists of the DNA region containing the IGF-2 gene, the Hl 9 gene and/or the INS gene. In another embodiment it comprises or consists of the DNA region from (and including) the Hl 9 gene to the INS gene as shown in Figure 1 (for the corresponding human chromosome region), i.e. part of the distal end of SSC2. Thus, besides the IGF-2 gene, this region comprises other genes, such as the H19 gene (GenBank Accession AY044827) and/or the INS (preproinsulin; GenBank AY044828). Optionally the region extends further to the right, including other imprinted genes such as TSSC6 (also known as PHEMX of PHMX), TRPM5, KCNQl (GenBank AC151879) and/or other imprinted genes in the region (based on comparative genomics with the mouse chromosome 7 / human chromosome region l ip 15.5) such as Mash2 (also named Hash2), Tapal, Litl (also named Kvlqtlas or Kcnqlotl), CDKNlC (also named p57kip2, encoding a cyclin cdk-inhibitor), Msuit, Slc22all (a.k.a. Slc22al8, Imptl, Orctl2, encoding a metabolite transporter), Tssc3 (a.k.a. Phlda2 or IPL, encoding a small PH-domain protein), Nap 1/4 (also named Nap2) and Obphl. For example, the linked maternally expressed H19 locus comprises an allele of the (non-coding) Hl 9 gene. The distance between the IGF-2 locus and the H19 locus is about 100kb. Some of the genes in the imprinted DNA region are known to be maternally expressed (e.g. Hl 9 and KCNqI) and some are known to have an effect on placental growth and development (e.g. H19, Cdknlc, Slc22all and Tssc3); see Tycko (2006, supra). "Maternally inherited" or "paternally inherited" refers to the origin of a hap Io type and an allele from either the mother or the father, respectively. "Maternally expressed" or "paternally imprinted" refers to mainly the maternally inherited allele being expressed, while the paternally inherited allele is expressed to a lower extent or not expressed (monoallelic expression).

"Paternally expressed" or "maternally imprinted" refers to mainly the paternally inherited allele being expressed, while the maternally inherited allele is expressed to a lower extent or not expressed. "Imprinted" or "parentally imprinted" refers to the phenomenon wherein either the paternally inherited DNA region or the maternally inherited DNA region is not expressed in the offspring, while the DNA region inherited from the other parent is expressed. "Domestic" refers to animals, especially mammals, domesticated and bred by humans. "Pig" or "porcine" refers herein to any animal of the species Sus scrofa.

In this document and in its claims, the verb "to comprise" and its conjugations is used in its non-limiting sense to mean that items following the word are included, but items not specifically mentioned are not excluded. In addition, reference to an element by the indefinite article "a" or "an" does not exclude the possibility that more than one of the element is present, unless the context clearly requires that there be one and only one of the elements. Thus, the indefinite article "a" or "an" thus usually means "at least one". DETAILED DESCRIPTION OF THE INVENTION

The present inventors crossed Chinese Meishan pig breeds and five Western pig breeds to study the effects of either maternally or paternally inherited IGF-2 alleles, and haplotypes comprising an imprinted DNA region with these alleles, on growth and finishing traits, meat quality traits and on fertility traits. F₂ crossbred pigs were produced by crossing Fi boars with Fl sows. Genotyping of the animals with respect to the IGF-2 alleles, or haplotypes comprising an imprinted DNA region with particular IGF-2 alleles, allowed statistical analysis models to uncover imprinting effects.

In this study all grand parental boars were homozygous wild type for the IGF-2 allele (G/G) while most grand parental dams of the Large White lines were either heterozygous (A/G) or homozygous mutant type (A/ A) and the grand parental dams of the Landrace lines were all homozygous wild type (G/G) except for 2 sows which were heterozygous (A/G). Fi animals (boars and sows) were either A/A, A/G, G/A or G/G.

It was found that the (haplotype comprising the) maternally inherited wild type allele (HapG^mat) had no effect in F₂ or Fi pigs on growth and finishing traits. Similarly, the (haplotype comprising the) paternally inherited wild type allele (HapG^pat) had a negative effect on weight gain, muscle depth and lipid deposition as shown for ultrasonic backfat (UBF) and Hennesy Grading Probe (HGP) backfat. These results confirmed the earlier finding that IGF-2 is paternally expressed in skeletal muscle and that the paternally inherited mutant allele (A^pat) has a beneficial effect on growth and weight of the offspring, whereby this is attributed to an increase in muscle weight, rather than fat.

Surprisingly, however, it was found herein that both the (haplotypes comprising the) paternally inherited and the maternally inherited alleles of the IGF-2 locus have a positive effect on litter size in pigs. The use of (a haplotype comprising) the paternally inherited mutant IGF-2 allele (A^pat) increased litter size (in at least the first and second parity) by at least about 1 piglet per litter. In addition, the use of (a haplotype comprising) the maternally inherited wild type allele (G^mat) also increased litter size (in at least the first and second parity) by at least about 1 piglet per litter, both in F₂ and surprisingly also in Fi sows. To illustrate the advantage of this finding, it is noted that an increase of two piglets per litter could so far only be achieved by at least 10 years of traditional selection, as on average only an increase of 0.2 piglets per litter per generation could be realized.

This finding can be put to practical use and genotypic assays, combined with appropriate crossing schemes, are provided herein, which provide the most optimal results in terms of growth, fattening and especially fertility traits. Figure 2 shows the optimal breeding method, whereby the imprinted DNA region comprises an IGF-2 genotype of A/A x (A/A x G/G). The finisher pigs, which are used for slaughter, obtain the positive effect of the 'A' allele on growth and fattening traits, while hybrid sows (comprising an imprinted DNA region which contains the 'HapA^pat' allele and the 'HapG^mat' allele) are capable of producing a maximal litter size, producing at least about 1 additional piglet due to the effect of the (haplotype comprising the) paternally inherited 'A' allele and at least about 1 further additional piglet due to the effect of the (haplotype comprising the) maternally inherited 'G' allele. These hybrid sows have, therefore, an overall higher prolificacy due to both a paternal imprinting effect and a maternal imprinting effect. Similarly, use can be made of the maternal expression / paternal imprinting effect on prolificacy of the DNA region comprising the maternally inherited 'G' allele as such, irrespective of the haplotype of the sire used to create the sows, as the sows will be capable of producing larger litter sizes.

In addition, no effect of the IGF-2 alleles on teat number was observed in this study, which is in contrast to the finding of WO2004/063386, wherein it is suggested that the mutant IGF-2 allele has a negative effect on teat number and it is suggested that selection of maternal lines for the wild type allele will enhance teat number. Thus, in one embodiment prolificacy is increased without modifying teat number.

Without limiting the invention in any way, it is thought that the positive effect of the (haplotype comprising the) maternally inherited 'G' allele may not be due to the 'G^mat' allele itself, but may be an indirect effect, resulting from the co-selection of one or more other alleles present in the imprinted DNA region linked to/comprising the 'G' allele, and therefore linked to/comprising the haplotype comprising the 'G' allele (HapG^mat). It is further thought that the positive effect of the paternally inherited 'A' allele on fertility may be due to a better developed reproductive tract in the progeny sows, which have a larger heart and are capable of providing their piglets with more nutrients, allowing more embryos to survive (see e.g. Angiolini et al., 2006, Placenta, Vol. 27, Supplement A, Trophoblast Research, Vol. 20, p 1-5). As a result more total piglets and more live born piglets are produced.

It was also surprisingly found in an interaction model that the positive effect of HapG^mat on litter size differs, depending on the haplotype (and/or pig breed) used, and that especially the Meishan haplotype comprising the wild type IGF-2 allele has a more positive effect on litter size compared to the Large White haplotype comprising the wild type IGF-2 allele. Thus, the Meishan IGF-2 'G' haplotype (referred to as HapG- MS) appears to co-inherit an imprinted DNA region which has a better effect on prolificacy than the Large White IGF-2 'G' haplotype (referred to as HapG-LW) or Land Race IGF-2 'G' haplotype (referred to as HapG-LR), especially when combined with a paternally inherited 'A' allele (HapA^pat). The interaction between the paternally inherited and maternally inherited chromosome regions leads, therefore, to a new combined imprinting effect (i.e. a kind of synergistic effect). Similar interactions between imprinting effects have been described in mice, between placenta and foetus (see Cattanach et al., 2006, Cytogenetic Genome Res 113: 17-23).

The following four basic hap Io types can be distinguished based in the breed and/or polymorphisms (SNPs) in the imprinted DNA region comprising the IGF-2 gene:

Haplotype 1 or HapA-LW: Large White comprising the IGF-2 'A' allele Haplotype 2 or HapG-LW: Large White comprising the IGF-2 'G' allele Haplotype 3 or HapG-MS: Meishan comprising the IGF-2 'G' allele Haplotype 4 or HapG-LR: Landras comprising the IGF-2 'G' allele.

In one embodiment HapG-LW, HapG-MS or HapG-LR are used in methods according to the invention, and most preferably HapG-MS is used, as will be described further below. Methods according to the invention

Thus, in one embodiment of the invention various methods for producing sows capable of producing a (statistically significant) larger number of total and/or a larger number of live piglets per litter are provided.

In a preferred embodiment the method comprising:

(a) crossing a boar which is homozygous for (a haplotype comprising) the mutant IGF-2 allele (HapA/HapA) with a sow which is homozygous for (a haplotype comprising) the wild type IGF-2 allele (HapG/HapG), and (b) selecting one or more sows from the piglets produced in said cross, wherein said sows comprise (a haplotype comprising) said paternally inherited mutant IGF-2 allele (HapA^pat) and a maternally inherited (haplotype comprising) said wild type IGF-2 allele (HapG^mat).

The term "crossing" or "combining" a boar with a sow refers herein preferably to artificial insemination procedures, as known in the art, whereby two individual animals of a race are combined. Significant human intervention is, therefore, required for carrying out the correct crossing/combining procedure (combining the genomes of a selected individual boar with a selected individual sow), as both genotyping methods (using e.g. one or more molecular markers such as SNPs, AFLP, SSR markers, DNA sequencing and/or other) and artificial insemination are involved. Thus, preferably, the haplotypes in step (a) are, or have been, determined by genotyping the boar and/or sow, or one or more of the parental animal(s) used to generate the boar and/or sow, so that the genotype (allelic make-up) at the IGF2 locus is known.

As shown in the Examples, such hybrid sows are capable of producing a larger number of live born and total piglets per litter, especially in the first two parities. Preferably, such sows produce (on average) at least one, more preferably at least two more piglets per litter, whereby part of this effect is attributed to the (haplotype comprising the) paternally inherited allele (HapA^pat) and part to the (haplotype comprising the) maternally inherited allele (HapG^mat). Such hybrid (HapA^pat / HapG^mat) sows, comprising a paternally inherited A allele and a maternally inherited G allele, produce therefore larger litter sizes than sows having a different genotype, such as for example sows derived from a HapG/HapG (sire) x HapG/HapG (dam) cross or from a HapG/HapG (sire) x (Hap A/Hap A) (dam) cross. It is noted that the latter cross produces sows (HapG^pat/HapA^mat) having the lowest prolificacy, while sows derived from a HapG/HapG (sire) x HapG/HapG (dam) cross and having the haplotype HapG^pat/HapG^mat have a higher fertility (producing on average at least one more piglet / litter), and this crossing scheme is also en embodiment of the invention (see further below).

The above method may further comprise the step of

(c) crossing the selected sow with a boar which is homozygous for (a haplotype comprising) the mutant IGF-2 allele (HapA/HapA) and optionally using the offspring as slaughter pigs.

As the sow selected in step (b) is capable of producing more total piglets (and especially also more live piglets) per litter than a sow comprising a (haplotype comprising a) paternally inherited wild type IGF-2 allele and a (haplotype comprising a) maternally inherited mutant IGF-2 allele (HapG^pat/HapA^mat) or a sow which is homozygous (HapG^pat/HapG^mat), and as the A allele (when inherited paternally) confers desired growth and finishing traits (such as increased birth weight, increased end and slaughter weight, decreased ultrasonic backfat, etc., see Examples), this is the optimal crossing scheme for producing larger numbers of lean slaughter pigs.

The homozygous (HapA/HapA) boar is preferably a terminal boar.

In the above method, the sow is capable of producing at least one, but preferably at least two more piglets per litter, compared to e.g. the sows made by alternative crosses described above. In other words, such sows produce (on average) at least one, more preferably at least two more piglets per litter than the average of the populations where the dam and the sire belong to. The (live) piglets born using the methods provided herein comprise a paternally inherited IGF-2 allele HapA^pat. These piglets preferably have an average birth weight which is higher than the average birth weight of piglets comprising a paternally inherited wild type IGF-2 allele (HapG^pat), which lies preferably at least about 30 grams per piglet above the birth weight of piglets comprising a paternally inherited wild type IGF-2 allele (HapG^pat). More preferably the average birth weight is at least 35g, 4Og, 43g, or more, above the birth weight of piglets comprising a paternally inherited wild type IGF-2 allele (HapG^pat). See Table 4. In addition the end weight and slaughter weight are higher (at least about 240Og more and 310Og more, respectively), hgp loin depth is deeper (almost by 2 mm) and the piglets have less ultrasonic backfat (about 1.7 mm less) and hgp backfat (about 2.8 mm less), see Table 4.

The sow used in step (a) preferably has one of the following (homozygous) haplotypes: HapG-LW, HapG-MS or HapG-LR. However, most preferably the sow has the haplotype HapG-MS, because the interaction model (see Examples) showed that the Meishan haplotype has the most beneficial effect on litter size.

To make the appropriate crosses, the allelic make-up (genotype) of the pigs can be determined with respect to the IGF-2 locus and optionally the haplotype can be determined using a number of at least about 3, 5, 10 or more, such as preferably at least 16 SNPs (or more) in the imprinted DNA region comprising the IGF-2 gene. This way sows having a homozygous haplotype for HapG, and especially sows having a homozygous haplotype selected from the group HapG-LW, HapG-MS or HapG-LR, can be identified and selected for use in the crosses.

IGF-2 alleles (mutant and wild type) and haplotypes of pigs can be distinguished by assessing one or more SNPs or other polymorphic markers (AFLPs, SSRs, etc.). For example, an easy assay for distinguishing between the wild type and the mutant IGF-2 allele is to determine the nucleotide of the G/A SNP in intron-3 of the IGF-2 gene, as described in VanLaere et al. (2003, Nature Vol. 425, p832-836, see Figure 1 and GenBank Accessions AY242098 - AY242112). For distinguishing between haplotypes the nucleotide present at one or more of the following 16 SNPs may be determined and used to select pigs having the desired haplotype with respect to the imprinted DNA region comprising the IGF-2 locus:

Legend:

GenBank Accession number AY044827; Sus scrofa Hl 9 gene, complete sequence, gi|21956485 |gb|AY044827.1 GenBank Accession number AY044828; Sus scrofa tyrosine hydroxylase gene, partial cds; and preproinsulin (INS) and insulin-like-growth factor 2 preproprotein (IGF2) genes, complete cds, alternatively spliced, gi|21956486|gb|AY044828.1 GenBank Accession number AC151879; Sus scrofa clone RP44-467B2, WORKING DRAFT SEQUENCE, 3 ordered pieces gi|54019475|gb|AC151879.1, containing TSSC6-TRPM5-KCNQ1.

Thus, the genomic DNA comprising the SNP may be amplified by PCR (e.g. using the PCR primers provided herein), followed by an assay for the nucleotide present at the SNP position. This may for example be done using a primer extension technique, such as single base exteniosn (SBE) wherein a single SBE primer is used to incorporate one of four labeled ddNTPs, depending on the nucleotide present at the SNP position. Alternatively, any other method for determining the SNP genotype may be used, such as described in Vignal et al. (2002, Genet SeI Evol 34: 275-305).

Table 2:

SBE = single base extension sequence, which corresponds to the sequence directly adjacent to the SNP to be detected.

SNP type = nucleotide alternatives detected by extending the SBE primers based on the 5' - 3' sequence.

For certain embodiments it is sufficient to determine which IGF-2 allele is present (G or A), without a need for characterizing the haplotype further. For example if one uses a Meishan breed, or a breed closely related thereto or derived therefrom, it may be sufficient to establish the presence of HapG/HapG, without establishing which nucleotides are present at any of the 16 SNPs of Table 1 below, as these are closely linked and therefore likely those which define the Meishan haplotype HapG-MS. Therefore, in one embodiment of the invention only the IGF-2 genotype needs to be determined. Optionally, also the nucleotide present at at least 2, 3, 5, 10, or more (e.g. all 16) SNP positions, as indicated in Table 1 and 2, may be determined.

Obviously, it is also possible find and use other polymorphic DNA markers in the DNA region comprising the IGF-2 gene as genotypic markers for characterizing the haplotype. For example any marker which is polymorphic between the haplotypes, such as any polymorphic marker (e.g. SNP) found elsewhere on GenBank Accession numbers AY044827, AY04482 or AC151879, may be used to distinguish haplotypes. Thus, when referring to SNPs herein, it is understood that other polymorphic DNA markers are encompassed.

Various methods for determining the allelic-make up (i.e. the nucleotide present) of one or more SNPs at a locus are available in the art. For example, PCR primer pairs may be designed, which amplify the nucleic acid region spanning the SNP (as provided herein for the above 16 SNPs) and the allelic-make up of the SNP (i.e. the nucleotide present at the SNP in a particular pig) may be detected by sequencing, fluorescent labeling assays, primer extension, etc. See for example the SNP genotyping assays by Applied Biosystems (e.g. TaqMan Assays), SNPWave™ (a Multiplex SNP Assay) by Keygene

N.V., Invader® Technology assay (Third Wave Technologies, Inc.), and many others.

See Vignal et al. (2002, supra). The method used for determining which nucleotide is present at the SNP position is not important herein, and any known method may be used. Preferably simple PCR based assays are used for each SNP of Table 1 and 2, so that an easy assay is available for determining the haplotype of one or more pigs. Such a PCR assay and primer pairs for use in this assay are also an embodiment of the invention. A haplotype assays comprises the steps of: a) providing a DNA sample of one or more pigs, b) developing primer pairs capable of amplifying a DNA region comprising a SNP and capable of distinguishing between haplotypes, such as for example the SNPs of Table 1 and 2 or one or more primers/oligonucleotides capable of selectively detecting one of the two alternative nucleotides present at the SNP; and c) detecting which nucleotides are present in the DNA sample of a pig at at least 3, 5, 10, 12 or more SNPs, for example those of Table 1 and 2.

It is understood that the primer pairs provided, which are capable of amplifying a DNA fragment comprising the SNP can also be replaced by other primer pairs or other methods for detecting the SNP and determining the nucleotide at the SNP position.

To select a dam or sire having the desired genotype at the IGF-2 locus and/or having the desired haplotype as defined by the SNPs, preferably by those of Table 1 and 2, any assay which allows the wild type and mutant IGF-2 allele and/or the SNPs to be discriminated may be used. The genotyping may be carried out on any DNA- comprising tissue sample of the pigs.

As the SNPs of Table 1 and 2 lie in a DNA region which is imprinted and comprises besides IGF-2 also other genes, it is also one embodiment of the invention to determine the allelic make-up or (SNP) haplotype for one or more of the following genes, preferably in addition to determining the genotype of the IGF-2 alleles: the Hl 9 gene, the TSSC6 gene, the TRPM5 gene, the KCNQl gene, the Mash2 gene, the Tapal gene, the Litl gene, the CDKNlC gene, the Msuit gene, the Slc22all gene, the Napl/4 gene, the Obphl gene and/or the Tssc3 gene. Without limiting the invention, it is thought that the positive effect on litter size associated with the IGF-2 'G' allele may be due to a parentally imprinted gene genetically linked to the IGF-2 wild type allele (HapG). One can, therefore, generate a sow capable of producing more piglets per litter by selecting a dam having a haplotype comprising those alleles of one or more of the above genes (and SNP combinations characteristic of the DNA comprising particular alleles) which are genetically linked (in cis) to the wild type IGF-2 (G) allele. The dam selected in step a) may thus be characterized by the homozygous haplotype HapG- SNPOOl -SNPO 16 / HapG - SNPOOl -SNPO 16. The hybrid sow produce in step (b) of the method may thus, for example, be characterized by the genotype HapA^pat/HapG^mat - SNPOOl ^mat to SNP016^mat, whereby the selection of the dam in step (a) may be based on any polymorphism in the above genes, especially based on one or more of SNPOOl to SNPO 16 of Table 1 and 2.

Because it was found herein that the wild type IGF-2 allele and/or an imprinted DNA region linked thereto (i.e. the haplotype HapG - SNPOOl -SNPO 16) has a paternal imprinting effect (maternal expression) on litter size, selection of the genotype of the (maternal grand parental) boar is not essentially required for increasing litter size. Therefore, it is an object of the invention to provide a method for producing sows capable of producing a larger number of piglets per litter whereby it is not necessary to select a boar in the maternal breeding line which has a particular genotype.

The method comprises the steps of:

(a) crossing a sow which is homozygous for (a haplotype comprising an imprinted DNA region which comprises) the wild type IGF-2 allele

(HapG/HapG) with a boar, and

(b) selecting sows from the piglets produced in said cross, wherein said selected sows comprise a maternally inherited wild type IGF-2 allele (HapG^mat).

In one embodiment the method further comprises the step of

(c) crossing said selected sow with a boar which is (homozygous for a haplotype comprising) a mutant IGF-2 allele (HapA/HapA) and optionally using the offspring as slaughter pigs.

The HapG^mat allele is again preferably from a Meishan breeding line, and one of the preferred sows used as dam in stem a) is a Meishan breeding line or a Meishan haplotype or haplogroup member. A sow homozygous for a haplotype comprising an imprinted DNA region which comprises the wild type IGF-2 allele (HapG/HapG) can be identified and selected as described above, i.e. by genotyping the allelic make-up of IGF-2 and/or one or more of the genes located in the imprinted DNA region (as mentioned above). Again one or more of the SNPs of Table 1 and 2 may be used, as described above.

Using this method, the sows selected in step (b) are capable of producing at least about one more piglet than, for example, sows produced from a dam comprising alleles HapA^pat/HapA^mat at the IGF-2 locus. Depending on the boar used in step (a), the prolificacy of the sow may be increased even further. However, any boar may be used as sire in step (a), e.g. a boar having one of the following genotypes: HapA/HapA, HapG/HapG, HapA/HapG or HapG/HapA.

The advantages with respect to the growth and fattening characteristics of the piglets obtained, as described above (increased average birth weight, etc.), also apply to this embodiment.

It is also feasible to use a dam which is heterozygous at the IGF-2 locus, such as a dam having the genotype A/G or G/A. However, if such a dam is used, the sows selected in step (b) need to be screened and selected for the presence of the maternally inherited wild type allele (HapG^mat) and/or a maternally inherited allele linked thereto, such as an allele of the genes located in the imprinting region mentioned above. This requires an extra genotyping step, which is therefore more laborious, but not exclude herein.

Again, the allelic make-up or haplotype of the pigs is easily determined by assessing the nucleotide of one or more SNPs in the imprinted DNA region comprising the IGF-2 locus and/or by assessing the allelic make-up of one or more polymorphisms in any of the genes linked (e.g. the Hl 9 locus), using for example the SNPs provided in Table 1 and 2 above. Throughout the invention, it is understood that crosses are preferably carried out using artificial insemination of the genotyped sows, with sperm from genotyped boars, as is standard procedure.

In one embodiment also any of the pigs and piglets obtained or obtainable using the herein described crosses are part of the invention, as are cells thereof and parts thereof. Thus, for example a hybrid sow is provided, whereby the sow comprises the genotype HapA^pat / HapG^mat. The sow is preferably used for the production of slaughter pigs, as shown in Fig. 2.

Because imprinting patterns can be altered by maternal nutrition (especially by a shortage of methyl donors) and other environmental factors (see e.g. Wu et al. 2004, Recent Advances in Nutritional Sciences, Maternal Nutrition and Fetal Development, American Society for Nutritional Sciences page 2169-2172 and Waterland and Jirtle 2004, Nutrition 20: 63-68), the sows are kept under conditions (e.g. maternal nutrition) which ensure correct imprinting and gene expression, so that the increased litter size, resulting from the parentally inherited imprinted DNA region is optimized.

Uses according to the invention In addition to the above methods, it is also an object of the invention to provide the use of (a) a sow which is homozygous for the haplotype comprising the wild type IGF-2 allele (HapG/HapG) and/or which is homozygous for the imprinted DNA region comprising the HapG allele or (b) a sow which comprises a maternally inherited wild type IGF-2 allele (HapG^mat) and/or a maternally inherited imprinted DNA region comprising the HapG^mat allele for producing an increased number of piglets per litter, especially live piglets / litter, compared to a sow which does not have said maternally inherited allele(s).

Especially, the use of the wild type IGF-2 (HapG) allele and/or the imprinted DNA linked to or comprising the wild type IGF-2 allele for increasing litter size due to a maternal imprinting effect is provided. SEQUENCES

SEQ ID NO 1-15: DNA fragment comprising SNPOOl to SNPO 16 of Table 1 and 2. SEQ ID NO 16-45: PCR primers for amplifying the DNA fragments of SEQ ID NO: 1- 15. SEQ ID NO 46-61 : SBE primers for detecting SNPOO 1 to SNPO 16.

FIGURES

Figure 1 illustrates schematically the human imprinted domain and imprinted genes homologous to the pig chromosome SSC2. The following symbols are used: White arrows: biallelic expression

Arrows with many dots: paternally expressed gene (e.g. IGF-2)

Arrows with few dots: maternally expressed gene (e.g. H 19)

Arrow with stripes: imprint status unknown in humans

Filled circle between Hl 9 and IGF-2: maternally methylated imprint center White circle: paternally methylated imprint center

Figure 2 illustrates the optimal breeding scheme and genotypes. "1" denotes the presence of the mutant IGF2 allele (also termed "A" allele), while "3" denotes the presence of the wild type IGF2 allele (also termed "G" allele). The hybrid sow comprises IGF2 alleles ' 1 ' (A) and '3 ' (G), whereby ' 1 ' (A) is inherited from the father (HapA^pat) and '3' is inherited from the mother (HapG^mat).

The following non-limiting Examples describe the use of imprinting effects according to the invention. Unless stated otherwise in the Examples, all DNA techniques are carried out according to standard protocols as described in Sambrook et al. (1989) Molecular Cloning: A Laboratory Manual, Second Edition, Cold Spring Harbor Laboratory Press, and Sambrook and Russell (2001) Molecular Cloning: A Laboratory Manual, Third Edition, Cold Spring Harbor Laboratory Press, NY; and in Volumes 1 and 2 of Ausubel et al. (1994) Current Protocols in Molecular Biology, Current Protocols, USA. EXAMPLES

EXAMPLE 1 - effect of IGF-2 alleles on fattening, fertility and meat quality traits in Meishan F? crossbred pigs

1. 1 - MATERIAL and METHODS Population

F₂ crossbreds between the Chinese Meishan pig breed and five Western pig breeds (two Landrace lines and 3 Large White lines) were available from an experiment involving five Dutch pig breeding companies. Crossbreds were produced in two batches at the same time in all companies during 1989-1991. For each batch, pure bred Western females at five farms were artificially inseminated by a group of 19 pure bred Meishan males, producing 126 crossbred Fl litters. From Fl litters a random selection of young males and females were taken as parents to produce 264 F2 crossbred litters, each female producing one F2 litter, descending from 39 Fl boars. The boars were used across breeding companies through the formation of a central pool and artificial insemination. Fl females remained at the breeding companies and produced a second litter using semen from terminal boar lines, except for 13 Fl sows that were not able to produce a second litter.

From performance tested F2-piglets (n ~ 1200) approximately 350 animals were retained for breeding. Approximately 300 were kept for the experiment, while less than 50 pigs were retained by the companies. The remaining 800 pigs were slaughtered at about 90 kg live weight in a central slaughter-house.

Phenotypes

Fl and F2 piglets were individually tagged and weighed at birth and at weaning (birth weight and weaning weight). The number of teats was counted (number of teats) for each piglet. Most Fl and F2 piglets were performance tested individually or in groups. Except for two companies the pigs were weighed at the start of the test, at approximately 10 weeks of age, (start weight). At the end of the test the live weight (end weight) and ultrasonic backfat (ultrasonic backfat) was recorded. Growth from birth until start of the test (early growth), during the test (test growth) and from birth until end of test (life growth) was calculated for each performance tested pig. Insemination and farrowing data were recorded for Fl- and F2-females that were randomly chosen to reproduce. Total number born included live born, dead born and mummified piglets in first (litter size 1^st parity) and second parity (litter size 2^nd parity). The farrowing interval between parity one and two was calculated in days

(farrowing interval) and scored as normal (score=0) if when ranging from 100 till 160 days and abnormal (score=l) if outside this range (farrowing interval 01). Similarly the interval between farrowing and insemination for the second parity was determined

(farrowing-insemination) in days and also scored as normal (score=0) and abnormal (score=l) if this interval exceeded 45 days (farrowing-insemination 01).

Performance tested F2 pigs that were sent to the slaughter-house were weighed approximately 45 minutes after stunning (slaughter weight) and the backfat thickness and the depth of the m. longissimus dorsi was determined using a Hennesy Grading Probe (hgp backfat and hgp loin depth). The HGP -measurement was taken between the 3^rd and 4^th rib, 6 cm from the spine. Twenty four hours post mortem pH (pH24 loin), drip loss (drip loss), cooking loss (cooking loss), shear force (shearforce) and intramuscular fat percentage (intra muscular fat) were measured in samples of the loin muscle. Additionally pH was measured in the m. semi-membranosus (pH24 ham). Color was measured using a Minolta on loin muscle samples according to the Cielab L*A*B* system, where L8 is an indication of lightness (Minolta L* loin), a* represents the degree of green-redness (Minolta a* loin) and b* represents the degree of blue-yellowness (Minolta b* loin) (MacDougall, D. B. 1986. The chemistry of colour and appearance. Food Chemistry 21:283-299). Details regarding the meat quality measurements can be found in Hovenier et al. (Hovenier, R. 1993. Breeding for meat quality in pigs. PhD-thesis, Animal Breeding and Genetics, Wageningen University, Wageningen, The Netherlands).

Genotypes IGF2 genotypes were obtained using sequence analysis (VanLaere et al. 2003, supra). Combining the flanking markers, as given in Table 3, with these genotypes using Simwalk2 (version 2.89) allowed the construction of haplotypes and for augmentation of those IGF2 samples for which the genotype could unambiguously determined.

Table 3 - Marker map of SSC2.

Statistical analysis

The following models were used to analyze the different traits:

Birth weight

BW = μ + COMP + YS + b*LS + SEX + PE + ANIMAL + P2/M2/P2.M2 + e (1)

where COMP is company (1-5), YS is batch (1-2), LS is litter size (<=7-20) as covariate, PE is random effect of permanent environment (254 sows), ANIMAL is the random polygenic effect (1-6903) including pedigree and genetic groups and P2/M2/P2.M2 is the effect of the paternal or maternal or the interaction among the IGF2 alleles respectively.

Weaning weight

WN = μ + COMP + YS + bl *LS + SEX+ b2*AGEl + PE + ANIMAL + P2/M2/P2.M2 + e

(2) where AGEl is age at farrowing as a covariate. Linearity of litter size and age was tested by including them as factors and studying the least squares estimates.

Test start weight STW = μ + COMP + YS + bl *LS + SEX+ b2*AGE2 + PE + ANIMAL + P2/M2/P2.M2 + e

(3)

Early growth rate model 3 was applied excluding AGE2. AGE2 is the age at the start of the test period. For two companies the start weight of the test was not determined for gilts.

For end weight, slaughter weight, ultrasone backfat and HGP backfat:

EW = μ + COMP.SEXE + YS + bl*LS + b2*AGE3 + PE + ANIMAL + P2/M2/P2.M2 + e

(4) where AGE3 is age at the end of the test. Company by sex interaction was included to account for individual testing of boars at company-3. Model 4 was also applied to growth during the test and life growth rate excluding AGE3. Linearity of AGE3 was tested by including it as a factor and studying the LS-estimates.

For litter size in the first parity, second parity, interval farrowing date and insemination date and interval between the first and second farrowing date:

LSI = μ + COMP.YS + bl*AGElFAR + ANIMAL + P2/M2/P2.M2 (+ PE + PAR) + e

(5)

where AGElFAR is age at first farrowing. Intervals were analyzed as measured, i.e. in days, or as normal or abnormal. Permanent environment, i.e. effect of dam, and effect of parity was included for analysis of litter size where parity 1 and 2 were combined.

1.2 Results

Genotyping and reconstruction IGF2 genotyping was done by sequence analysis (Van Laere et al. 2003). Haplotypes were reconstructed using SIMW ALK2 (version 2.89) (Sobel and Lange 1996 AmJ. Human Genetics 58:1323-1337). All grand parental boars were homozygous wild type (G/G) while most grand parental dams of the Large White lines were either heterozygous (A/G) or homozygous mutant type (A/ A) and the grand parental dams of the Landrace lines were all homozygous wild type (G/G) except for 2 sows which were heterozygous (A/G).

For 23 F2 pigs resulting from a cross of a heterozygous male and female the origin of the alleles could not be established unambiguously and these genotypes were eliminated.

Growth related traits

In Table 4 below the estimated differences are given as well as the significance of the effects on each trait related to growth and fattening. The effect of the paternally and maternally wild type allele (3) is given for F2 pigs where the mutant allele is set at zero. In Fl pigs the contrast between maternally inherited alleles can be calculated. The paternally inherited allele in Fl pigs is coming from the Meishan breed which is homozygous wild type (G/G). The contrast is partly confounded with company in Fl- pigs because Landrace grand dams were mostly homozygous wild type while Large White grand dams were predominantly homozygous mutant (A/ A).

At birth there is already a significant difference which increases with age resulting in a more pronounced difference at weaning. These results indicate that IGF2 is affecting the growth of the piglet in the uterus as well as during lactation. Since piglets do not deposit a lot of fat it is likely that the difference will be mainly due to differences in muscle weight.

During the rearing period and the testing period the difference increases as is shown in the weight at the start of the test, the end of the test and in the slaughter house.

Not only has the paternally inherited wild type allele (G) a negative effect on weight, gain and muscle depth but it also affects lipid deposition as is shown for ultrasonic backfat and HGP backfat. Table 4 confirms the results obtained by Jeon et al (1999 Nature Genetics 21:157-158) and Nezer et al (1999 Nature Genetics 21:155-166) that only the paternally inherited allele affects the growth and fattening characteristics of pigs. Terminal boar lines should therefore preferably be homozygous for the mutant allele.

The interaction between the paternally and maternally inherited allele was not significant for all growth related traits except for backfat, which was borderline significant (p =.049).

Table 4 - Effect of IGF2 on growth related traits

Part 1 describes the effect of the paternally inherited allele in F2-pigs, part 2 the effect of the maternally inherited allele in F2-pigs and part 3 the effect of the maternally inherited allele in Fl -pigs.

trait for F2-sows # obs. ϊgβpat se F-value p-value paternally inherited allele effect birth weight gram 1179 -43 16 7.56 0.006 weaning weight gram 1176 -132 109 1.47 0.226 start weight gram 919 -466 306 2.32 0.128 end weight gram 1115 -2494 844 8.74 0.003 slaughter weight gram 779 -3112 903 11.88 0.001 early growth gr/day 919 -6.1 4.1 2.16 0.142 test growth gr/day 919 -28.9 9.2 10.00 0.002 life growth gr/day 1115 -14.7 5.0 8.51 0.004 ultrasoon back fat mm 1115 1.77 0.26 45.28 0.000 hgp backfat mm 779 2.83 0.51 31.10 0.000 hgp loin depth mm 779 -1.98 0.66 9.08 0.003 trait for F2-sows # obs. igf2mat se F-value p-value maternally inherited allele effect birth weight gram 1176 -4 16.17 0.05 0.823 weaning weight gram 1173 -82 111.6 0.54 0.463 start weight gram 918 123 353.3 0.12 0.729 end weight gram 1112 51 879.51 0.00 1.000 slaughter weight gram 775 -1670 954.1 3.06 0.081 early growth gr/day 918 0.6 4.8 0.02 0.888 test growth gr/day 918 5.3 10.5 0.25 0.617 life growth gr/day 1112 0.6 5.3 0.01 0.920 ultrasoon back fat mm 1112 -0.37 0.27 1.83 0.176 hgp backfat mm 775 0.05 0.57 0.01 0.920 hgp loin depth mm 773 -0.40 0.71 0.32 0.572 trait for Fl-sows # obs. igf2mat se F-value p-value maternally inherited allele effect birth weight gram 287 -22 43 0.27 0.604 weaning weight gram 287 -154 249 0.38 0.538 start weight gram 192 -2 63 0.00 1.000 end weight gram 284 -464 1060 0.19 0.663 early growth gr/day 192 0.2 9.8 0.00 1.000 test growth gr/day 192 -14.2 16.8 0.72 0.397 life growth gr/day 284 -4.2 7.1 0.34 0.560 ultrasoon back fat mm 284 -2.40 4.55 0.28 0.597

Fertility related traits

In Table 5 the test statistics and the estimated effect of the paternally and maternally inherited IGF2-allele are shown with regard to fertility related traits both for F2 and Fl sows. After the F2 animals were tested some randomly chosen females, which met minimal requirements with respect to weight and feet and legs, were kept to generate litter size observations. Around 20% of the F2 sows did not produce a second litter. In most cases boars from a terminal sire line were used, which was not included in the model, for insemination. In the Fl sows the first litter was sired by Fl -boars while the second litter was in most cases sired by boars from a terminal sire line.

Table 5 shows a favorable effect of the IGF-2 allele on litter size both in first and second parity. Significance increased when a repeatability model was applied to the combined data. Surprisingly, a positive effect is observed for the paternally inherited mutant allele (A) in addition to a positive effect of the maternally inherited wild type allele (G). The positive effect of the maternally inherited wild type allele is also observed in Fl sows. This was a surprising result, especially since there is no interaction between the paternally and maternally inherited allele (data not shown). These data show that both the paternally inherited mutant allele and the maternally inherited wild type allele have a beneficial effect on litter size.

The farrowing interval and the interval between farrowing and re-breeding are not normal distributed. The data regarding these traits, however, is censored. The sows that were not re-bred within a limited period were disregarded. In the Fl sows, however, IGF-2 affects the farrowing interval and the interval between farrowing and insemination significantly in contrast with results obtained on F2-sows.

The number of teats was, in the present study, not affected by the IGF2-gene, in contrast with the observation by Jeon et al. (1999, supra) and Nezer et al. (1999, supra). Hirooka et al. (2001 J. of Animal Science 79(9):2320-2326) showed a QTL in the IGF2 region affected number of teats using the same data as in this study. It might indicate that the result by Jeon and Nezer are not due IGF2 but a locus in the vicinity of the IGF2-locus.

Table 5 - Effect of IGF2 on fertility related traits Part 1 describes the effect of the paternally inherited allele in F2-pigs, part 2 the effect of the maternally inherited allele in F2-pigs and part 3 the effect of the maternally inherited allele in Fl -pigs. trait for F2-sows # obs. Igβpat se F-value p-value paternally inherited allele effect littersize 1^st par n 249 -1.02 0 .58 3.10 0.080 littersize 2^nd par n 209 -1.45 0 .65 5.03 0.026 littersize 1^st + 2^nd par n 458 -1.12 0 .47 5.76 0.017 farrowing interval days 208 2.30 2 .07 1.23 0.269 farrowing interval 01 0/1 208 0.03 0 .06 0.16 0.690 farrowing-insemination days 208 2.23 1 .95 1.30 0.256 farrowing-insemination 01 0/1 208 0.06 0 .06 0.88 0.349 number of teats n 1179 -0.15 0 .10 2.35 0.126

trait for F2-sows # obs. Ϊgf2mat se F-value p-value maternally inherited allele effect littersize 1^st par n 250 1.09 0 .90 3.49 0.063 littersize 2^nd par n 211 0.94 0 .64 2.16 0.143 littersize 1^st + 2^nd par n 461 1.23 0 .46 7.10 0.008 farrowing interval days 204 -1.16 2 .11 0.30 0.584 farrowing interval 01 0/1 210 0.01 0 .06 0.03 0.863 farrowing-insemination days 204 -1.02 1 .99 0.26 0.611 farrowing-insemination 01 0/1 210 0.01 0 .06 0.01 0.920 number of teats n 1176 -0.07 0 .10 0.49 0.484

trait for Fl-sows # obs. Ϊgf2mat se F-value p-value maternally inherited allele effect littersize 1^st par n 248 1.05 0 .51 4.14 0.043 littersize 2^nd par n 235 0.96 0 .56 2.88 0.091 littersize 1^st + 2^nd par n 483 1.00 0 .41 5.93 0.015 farrowing interval days 235 5.90 2 .00 8.69 0.004 farrowing interval 01 0/1 235 0.24 0 .05 20.62 0.000 farrowing-insemination days 235 4.33 1 .62 7.09 0.008 farrowing-insemination 01 0/1 235 0.20 0 .06 12.84 0.000 number of teats n 287 0.25 0 .17 2.08 0.150

Meat quality traits

In Table 6 the effect of IGF2 on meat quality parameters are shown. The paternally inherited IGF2 allele affected cooking loss, where the wild type allele showed approximately 1% less loss compared to the mutant allele.

The maternally inherited IGF2 allele affected the pH24 in the ham muscle. The wild type allele decreases the pH which is in not the right direction for cured ham production, where a minimum value of 5.6-5.7 is required. The effect of IGF2 on two meat quality traits is therefore opposite from what is wanted with respect to growth, finishing and reproduction.

Table 6 - Effect of IGF2 on meat quality related traits

Part 1 describes the effect of the paternally inherited allele in F2-pigs and part 2 the effect of the maternally inherited allele in F2-pigs. trait for F2-sows # obs. Ϊgf2pat se F-value p-value paternally inherited allele effect pH24 loin pH-units 778 0.012 0.023 0.30 0.584 pH24 ham pH-units 779 0.009 0.024 0.13 0.719

Drip loss % 111 -0.09 0.15 0.42 0.517

Cooking loss % 778 -1.01 0.29 12.17 0.001

Shearforce newton 778 0.70 0.73 0.91 0.340

Intra muscular fat % 764 -0.14 0.09 2.58 0.109

Minolta L^* Loin units 111 -0.11 0.45 0.06 0.807

Minolta a^* Loin units 119 0.29 0.18 2.72 0.100

Minolta b^* Loin units 778 0.17 0.17 0.99 0.320 trait for F2-sows # obs. Ϊgf2mat se F-value p-value maternally inherited allele effect pH24 loin pH-units 774 -0.003 0.024 0.01 0.920 pH24 ham pH-units 775 -0.057 0.026 4.88 0.027

Drip loss % 773 0.17 0.15 1.31 0.253

Cooking loss % 774 -0.15 0.31 0.23 0.632

Shearforce newton 774 0.19 0.79 0.06 0.807

Intra muscular fat % 760 -0.13 0.09 2.16 0.142

Minolta L^* Loin units 773 -0.79 0.47 2.78 0.096

Minolta a^* Loin units 775 -0.05 0.19 0.00 1.000

Minolta b^* Loin units 774 -0.02 0.18 1.72 0.190

Conclusions from Example 1

There is a significant effect of IGF-2 alleles on important production traits (growth rate, lean percentage, litter size) in pigs. The positive effect of the mutant IGF-2 allele on growth and carcass traits confirms the results previously obtained by Nezer et al (1999, supra) and Jeon et al. (1999, supra).

It was further found that the effect of IGF-2 alleles on litter size in pigs is large and that the parental origin and the type of allele that is inherited determine its effect. It is also shown that IGF-2 alleles might have an effect on cooking loss and pH24 in the ham muscle. The effects of the alleles affecting meat quality positively seem to be opposite to the alleles affecting growth, carcass quality and litter size.

Given the results of Example 1 above, regarding growth, fattening and fertility the ideal cross between lines would be: A x (C x D) with IGF-2 genotype: A/A x (A/A x G/G) as is shown in Figure 2. In this case the finishers obtain the positive effect of IGF2 on growth and fattening traits, while the hybrid sows are heterozygous, having a paternally inherited mutant allele and a wild type maternally inherited allele which resulted in the largest positive effect of IGF2 on litter size.

However, the effect on interval, as was shown in Fl -sows, indicate that the maternally inherited mutant allele (A) might have a negative correlated response with regard to these interval.

EXAMPLE 2 - interaction model

2.1 - Interaction model

IGF-2 mutant A-allele always originates from Large white lines.

IGF-2 wild type G-allele originates from one of the five white lines (G-W), such as Large White or Landras, or from Meishan (G-MS).

There are therefore three alleles for the IGF-2 gene: A, G-LW and G-MS In the data analysis the following interaction model was fitted: For litter size in the first parity, second parity, and combined: LSI = μ + COMP.YS + bl *AGE1FAR + ANIMAL + P2 + M2 (+ PE + PAR) + e LSI = μ + COMP.YS + bl*AGElFAR + ANIMAL + P2.M2 (+ PE + PAR) + e

2.2 - Results

The effect of paternally and maternally inherited haplotypes (based on the origin of each haplotype) is given in the following tables. The first table describes the effect of the paternal and maternal inherited haplotype assuming that they are their expression occurs independently while the second table assumes a complete interaction among the paternal and maternal haplotype. model=p2+m2

The results show that the G-MS allele has a different effect on litter size than the G-LW allele.

Model p2.m2

The results show that the best combination with respect to litter size is a paternally inherited mutant allele (A) together with a maternally inherited wild type Meishan allele (G-MS).

Also, the paternally inherited G-LR allele does not have a positive effect, especially combined with a maternally inherited mutant allele (A).

In conclusion, an interaction between the paternally and maternally inherited allele seems to exist. Although the interaction is quite obvious the standard errors are still large. The combination having the most positive effect on litter size is A^pat/ G-MS^mat

EXAMPLE 3 - effect of haplotvpes on fattening, fertility and meat quality traits in Meishan F? crossbred pigs The analysis of Example 1 is repeated using the 16 SNPs of Table land 2 and IGF-2 geno typing and the results are analysed on the basis of hap Io types 1- 4. The objective of the SNP -testing is to:

1. determine and/or verify the number of haplotypes;

2. determine and/or verify that the differences among the haplotypes occur in an imprinted DNA region comprising the IGF2 and/or Hl 9 and/or INS gene.

3. re-analyse the data sets;

4. determine and/or verify that one of the SNPs/genes provided is the causative SNP/gene that is responsible for the maternally inherited positive effect on litter size.

5. Determine which living populations should be sampled to confirm the results in current commercial populations.

Claims

1. A method for producing sows capable of producing a larger number of piglets per litter, said method comprising:

(a) crossing a boar which is homozygous for a haplotype comprising the mutant IGF-2 allele (HapA/HapA) with a sow which is homozygous for a haplotype comprising the wild type IGF-2 allele (HapG/HapG), and

(b) selecting one or more sows from the piglets produced in said cross, wherein said sows comprise a paternally inherited haplotype comprising said mutant IGF-2 allele (HapA^pat) and a maternally inherited haplotype comprising said wild type IGF-2 allele (HapG^mat), whereby the haplotypes in step (a) are, or have been, determined by genotyping the boar and/or sow, or one or more of the parental animal(s) used to select the boar and/or sow that are combined.

2. The method according to claim 1, further comprising the step of

(c) crossing said selected sow with a boar which is homozygous for a haplotype comprising a mutant IGF-2 allele (HapA/HapA) and optionally using the offspring as slaughter pigs.

3. The method according to claim 1 or 2, wherein HapG^mat is from a Meishan pig breeding line.

4. The method according to any one of the preceding claims, wherein the sow selected in step (b) is capable of producing more piglets per litter than a sow comprising a haplotype comprising a paternally inherited wild type IGF-2 allele (HapG^pat) and/or a haplotype comprising a maternally inherited mutant IGF-2 allele (HapA^mat).

5. The method according to claim 4, wherein said sow is capable of producing on average at least one, preferably at least two more piglets per litter.

6. The method according to claim 5, wherein the piglets are live piglets, preferably having an average birth weight of at least about 30 grams per piglet above the birth weight of piglets comprising a paternally inherited wild type IGF-2 allele (HapG^pat), and preferably being capable of producing a higher end weight and slaughter weight, with a deeper hgp loin depth and less ultrasonic and hgp backfat.

7. The method according to any one of the preceding claims, wherein the hap Io type of the pigs is determined by assessing the nucleotide present at one or more single nucleotide polymorphism (SNP) loci in an imprinted DNA region of chromosome SSC2 comprising the IGF-2 gene, the H19 gene and/or the INS gene.

8. The method according to claim 7, wherein one or more SNPs shown in Table 1 are determined.

9. A method for producing sows capable of producing a larger number of piglets per litter, said method comprising:

(a) crossing a sow which is homozygous for a haplotype comprising an imprinted DNA region which comprises the wild type IGF-2 allele (HapG/HapG) with a boar, and (b) selecting sows from the piglets produced in said cross, wherein said sows comprise a maternally inherited haplotype comprising said imprinted DNA region.

10. The method according to claim 9, further comprising the step of (c) crossing said selected sow with a boar which is homozygous for a haplotype comprising the mutant IGF-2 allele (HapA/HapA) and optionally using the offspring as slaughter pigs.

11. The method according to claim 9, wherein said sows are capable of producing an increased number of piglet per litter, preferably an increased number of live piglets, said live piglets preferably having an average birth weight of at least about 30 grams per piglet above the birth weight of piglets comprising a paternally inherited wild type IGF-2 allele (HapG^pat), and preferably being capable of producing a higher end weight and slaughter weight, with a deeper hgp loin depth and less ultrasonic and hgp backfat.

12. The method according to any one of claims 9 to 11, wherein, wherein the haplotype of the pigs is determined by assessing the nucleotide present at one or more the single nucleotide polymorphisms (SNPs) loci in an imprinted DNA region of chromosome SSC2 comprising the IGF-2 gene, the H19 gene and/or the INS gene.

13. The method according to claim 12, wherein one or more SNPs shown in Table 1 are determined.

14. Use of (a) a sow which is homozygous for a haplotype comprising the wild type IGF-2 allele (HapG/HapG) and/or which is homozygous for an imprinted DNA region comprising the wild type IGF-2 allele, or (b) a sow which comprises a haplotype comprising a maternally inherited wild type IGF-2 allele (HapG^mat) and/or a maternally inherited imprinted DNA region comprising said wild type IGF-2 allele in a cross for producing an increased number of live piglets per litter.

15. A hybrid sow comprising a paternally inherited haplotype comprising a mutant IGF-2 allele (HapA^pat) and a maternally inherited haplotype comprising a wild type

IGF-2 allele (HapG^mat).