Molecular Genetics and Meat Quality1

Anchie C. Clutter, Oklahoma State University

As we consider the genetic improvement of meat quality it becomes clear that these characteristics will be very difficult to approach with traditional selection methods. Although results from the NPPC National Genetic Evaluation Program and several other experiments indicate that quality characteristics are generally moderate in heritability, muscle quality is difficult if not impossible to measure in the live animal, and very expensive to measure completely in samples from the carcass. Consequently, and especially with rapid developments in the molecular genetics of livestock, we look hopefully toward nontraditional methods to attack this important problem.

Where are we in terms of understanding the molecular genetics of meat quality, what kind of research is underway or planned and what are the prospects for finding molecular information that will be useful in genetic improvement programs? My objectives today are to review completed, ongoing and planned research in this area and suggest the important factors for future research success.

Most of the progress in this area to date has been associated with major gene effects. The discovery of a mutation in the ryanodine receptor gene related to porcine stress syndrome, and new results describing the linked effects on carcass composition and quality, make this the greatest area of success so far in terms of molecular genetics actually applied to pork production. There is also new information regarding the Rn gene segregating in Hampshire pigs and their crosses, and so I'll start today by discussing those two loci.

But the potential for longterm genetic improvement assisted by molecular techniques lies in the discovery of those genes that have quantitative or relatively small effects on quality traits, i.e. finding the quantitative trait loci (QTL) involved. Research in this area as it relates to muscle quality is in the early stages, but there have been QTL reported that affect composition, so I'll review those reports as well as describe studies ongoing and planned that include quality traits.

It is important to point out that all of these efforts depend on development of genetic markers, or DNA sequences that mark specific locations in the genome, i.e. a genetic marker map of the pig genome The majority of these markers are not functional genes themselves, but allow us to track areas of the genome so that important genes can be located. Progress in molecular pig genetics is accelerating because development of a working marker map is complete.

MAJOR GENE EFFECTS

Halothane susceptibility is by now a trait the industry is very familiar with. The reaction, malignant hypothermia, is triggered in the pig by stress or inhalation of halothane and has been known for some time to be transmitted through a recessive allele. The stress reaction is associated with a calcium release channel, the ryanodine receptor, in skeletal muscle, and both the halothane reaction and receptor gene were mapped to chromosome 6. The identification of the mutation in the receptor gene associated with the stress reaction (Fujii et al., 1991) led to a PCRbased DNA test for the hal gene and the ability to eradicate the gene through selection. The gene is additionally important in production because of linked effects on carcass composition and quality, and I'll leave the description of those effects to Rodney Goodwin in the next talk.

Rendement Napole is trait that estimates the yield of processed Paris ham. Le Roy et al. reported in 1990 that there was a major gene segregating in Hampshire pigs and their crosses that affected this yield. Specifically, a dominant form or allele (Rn) causes a significantly greater level of glycogen in the muscle (up to 70% greater) and a reduction of yield (i.e. an unfavorable dominant allele). It is thought that this allele may explain much of the socalled "Hampsliire effect" on muscle quality of high glycogen and associated low pH.

With the ability to classify animals for presence or absence of Rn by glycogen content, and the availability of genetic markers throughout most of the genome, three research groups set out to map the Rn locus. The Kiel project in Germany (Rudat et al., 1995) produced 564 progeny from matings of Hampshire x Pietrain sires known to be heterozygous for Rn, with homozygous normal females (rn/rn). A Swedish study (Mariani et al., 1996) produced 105 progeny from a single heterozygous boar and 22 homozygous normal females. A similar design was also implemented by a French research group (Milan et al,, 1995).

In each case, genetic markers were identified for which the sires were also heterozygous. progeny were phenotyped for Rn (glycogen level) and marker genotype. By following the inheritance of Rn phenotype and markers from throughout the genome, linkage of Rn with the markers on chromosome 15 was determined. Mapping results from Mariani et al. (1996) are presented in Figure 1.

Now that the approximate location of Rn has been determined, two steps are pursued: 1) typing with additional (anonymous) markers in this region to determine a more precise location of the Rn locus so that selection against the unfavorable Rn allele can begin, and 2) use of comparative mapping to identify candidate genes in the region.

Comparative mapping is based on the knowledge that species share much of the same DNA sequence and, therefore, a region of a chromosome in one species corresponds to alike(homologous) regions on chromosomes in other species. The human and mouse are considered 'maprich' species in which very detailed genomic maps have been developed, including the locations of many functional genes. By lining up the region of interest in the pig with the homologous regions in those species, a wealth of information regarding functional genes in the area may be revealed. For example, this region of chromosome 15 in the pig is homologous with a region of chromosome 2 in the human. Two functional genes from the human map of this region that are involved in glycogen metabolism will be investigated further as candidates for the Rn gene.

Comparative mapping is not only valuable in attempts to locate major genes like Rn, but will also be useful in the search for QTL that affect important traits like muscle quality. For effective comparative mapping to occur, a sufficient number of functional genes must be mapped in the pig so that species maps can be correctly aligned. Chris Tuggle at Iowa State has begun a new project to develop markers for functional genes in regions of the pig genome where there are 'gaps' in comparative information between the pig and human and mouse. At Oklahoma State, in collaboration with Daniel Pomp at Nebraska, we are developing markers in the pig for candidate genes associated with feed intake (CCK and CCK receptors Clutter et al., 1996) and previously associated with obesity in the mouse (ob and agouti). Once developed, markers are mapped based on linkage with previously mapped loci in the PiGMaP Consortium reference families.

QUANTITATIVE TRAIT LOCI (QTL)

Conceptually, studies to locate QTL that affect important production traits are following the inheritance of alleles of markers from throughout the genome, and determining the association of those marker alleles with performance, regions of the genome harboring QTL can be identified. If enough evenly spaced markers are used, wherever a QTL resides it will be linked to one of the markers. The most effective design is often one in which segregating marker alleles are studied in an F2 'resource' family produced by crossing very divergent grandparent stock (e.g. wild boar x domestic).

In Figure 2 is a simple example involving a single marker and linked QTL. The cross of divergent lines should result in an Fl that is heterozygous for marker and QTL alleles. The expectation for the F2 is the familiar 1:2:1 ratio for the three possible genotypes at each locus. By comparing average performance of the marker genotype groups for traits of interest (e.g. growth, fat, marbling), the existence of a linked QTL can be detected.

A study in which this design was implemented to find QTL in pigs was reported in a 1994 issue of prestigious journal science (Andersson et al.). A wild boar x Large White cross was used to produce the F2 resource family. QTL on chromosome 4 affected growth, backfat (~.5 cm difference between homozygotes) and abdominal fat. QTL explained ~18% of F2 phenotypic variation in both measurements of fat.

F2 resource families (Chinese x Domestic) were also used in an Iowa State study reported by Yu et al. (1995). However, rather than a screening of the genome with anonymous markers, the effects associated with a single candidate gene were determined. The "candidate gene" approaoh is appropriate when a gene is known to function in such a way that it may explain genetic variation in traits of interest, and was essentially the only method available before completion of a workable marker map.

The candidate gene in the study reported by Yu et al. (1995) is PITl which codes for a transcription factor that regulates the action of several hormones, including growth hormone. Segregation of markers in the PITl gene were associated with differences in backfat of from .4 to .5 cm, and in loin muscle area (LMA) of 3.9 cm2. No effects on loin marbling, color or firmness were observed.

Although Yu et al. studied some characteristics of muscle quality, reports of QTL affecting quality are still rare. In a study at the University of Illinois (Clamp et al., 1992), a halfsib family was produced from a single Duroc boar heterozygous for four DNA markers. A total of 186 progeny were genotyped for marker alleles and measured for growth, carcass composition, loin color and firmness, and ham muscle mass. A marker (PGD) on chromosome 6 was associated with loin muscle firmness, but the effect may have been linked to the ryrl locus which is also located on chromosome 6.Another study at Iowa State (Rothschild et al., 1996) used a combination of the candidate gene approach and a broader screening with anonymous markers. Previous work by the group revealed an association between markers in the swine major histocompatability complex on chromosome 7 and market pig performance (Jung et al., 1989). Consequently, segregation of several markers on chromosome 7 was studied in the Chinese x Domestic resource families. A QTL (P<.01) for backfat was detected between TNF and anonymous marker S0102. A QTL (P<.05) for LMA was linked to anonymous marker S0066. There was also a tendency (P<.10) for an effect in the same region on loin muscle firmness, marbling and color.

Several QTL studies aimed at, or including, meat quality traits are underway or planned. In a collaborative study with Brian Kirkpatrick at Wisconsin, lines at Oklahoma State that have undergone 10 generations of selection for fast (F) or slow (S) postweaning gain have been crossed to produce F 1 boars. These boars are expected to be heterozygous for QTL alleles targeted by selection for gain. Two of these sires have been mated with an unrelated population of females to produce halfsib families (n~150 each) in which the alleles are segregating. A comprehensive screening of DNA from the progeny is being completed using candidate gene and anonymous markers to locate QTL affecting growth, backfat, LMA, and loin and ham pH. At Iowa State, in addition to the continued analyses of marker segregation and muscle quality in Chinese x Domestic families, production of a resource family aimed specifically at the location of QTL affecting quality is underway. From 400 to 500 F2 progeny will be produced by crossing Berkshire and Yorkshire grandparent stock, and several characteristics of carcass quality measured for each. Berkshire sires that transmit the "Berkshire effect" on meat quality (high pH, tenderness, juiciness, firmness, and low cooking loss) were identified for use in the study. A comprehensive screening of the gnome will be used to locate the important QTL related to muscle and eating quality.

The US group of scientists involved in swine breeding (formerly NC206) has submitted a new proposal that includes an objective to find QTL for important production traits (see also the paper by Larry Young in these proceedings). At Oklahoma State, we plan to produce an F2 resource family with the same selection lines (I7 x S) used in the collaborative study with Wisconsin. At Auburn, there are plans to produce a resource family by crossing a line selected for fast growth and a control. North Carolina State plans to cross lines that have undergone divergent selection for plasma testosterone level and express correlated changes in body composition. In each of these projects, there is the potential for collection and analyses of muscle quality data.

The study by Andersson et al. (1994) mentioned earlier was part of the PiGMaP Consortium of researchers first referred to as "the European Laboratory without walls". More recently the group has been expanded to include laboratories in the US and Australia. There are QTL studies ongoing in five PiGMaP resource populations produced by either Chinese x Domestic or Wild Boar x Domestic crosses. Characteristics of muscle quality are also being studied in these QTL projects.

Summary. Limitations to the improvement of muscle quality through traditional methods suggest the need for molecular techniques. Success in locating QTL with important effects on these characteristics that can be used to assist selection will depend on appropriately designed resource families. Families must originate from lines that differ significantly in muscle quality. For example, the F and S lines at Oklahoma State seem wellsuited to locate (QTL affecting feed intake, growth and fatness (differences of N4, 3 and 3 standard deviations, respectively), but are less divergent in the quality traits that have been measured Differences of .9 and .6 standard deviations in loin marbling and firmness, respectively). Families such as the Berkshire x Yorkshire F2 at Iowa State, and from greatly divergent backgrounds (e.8. wild boar x domestic)will probably prove most informative for quality traits.

Although the benefits from information generated by this research may be tremendous, especially with increasing market demands for quality, the costs of these experiments to collect complete quality data and genotype 400 to 500 animals for an adequate number of markers is also great. The unique populations of animals are available, and the expertise and technology exists, but the bestdesigned research will benefit no one without sources of adequate funding.

ACKNOWLEDGMENTS

I would like to acknowledge my collaborators in this area of research: Brian Kirkpatrick, Daniel Pomp and Chris Tuggle. I would also like recognize Max Rothschild not only as a collaborator, but also for his work as US Swine Genome Coordinator. Alan Archibald and Chris Haley are coordinators of the PiGMaP Consortium.

REFERENCES

Andersson, L., C.S. Haley, H. Ellegren, S.A. Knott, M. Johansson, K. Andersson, L. AnderssonEklund, I. EdforsLilja, M. Fredholm, I. Hansson, J. Hakansson, and K. Lundstrom. 1994. Genetic mapping of quantitative trait loci for growth and fatness in pigs. Science 263:1771-1774.

Clamp, P.A., JOE. Beever, R.L. Femando, D.G. McLaren, and L.B. Schook. 1992. Detection of linkage between genetic markers and genes that affect growth and carcass traits in pigs. J. Anim. Sci. 70:2695-2706.

Clutter, A.C., S. Sasaki, and D. Pomp. 1996. PCRbased polymorphism in the porcine cholecystokinin (CCK) gene and assignment to chromosome 13. Anim. Genet. (submitted)

Fujii, J., K. Otsu, F. Zorzato, S. De Leon, V.K. Khanna, J E. Wciler, P.J. O'Brien, and D.H. MacLennan. 1991. Identification of a mutation in porcine ryanodine receptor associated with malignant hypothermia. Science 253:448-451.

Jung. Y.C., M.F. Rothsehild, M.P. Flanagan, L.L. Christian, and C.M. Warner. 1989. Association of restriction fragment length polymorphisms of swine leukocyte antigen class I genes and production traits of Duroc and Hampshire boars. Anim. Genet. 20:79-91.

Le Roy, P. J., Naveau, J.M. Elsen and P. Sellier. 1990. Evidence for a new major gene influencing meat quality in pigs. Genet. Res. Camb. 55:33-40.

Mariani, P., K. Lundstrom, U. Gustafsson, A.C. Enfalt, R.K. Juneja, and L. Andersson. 1996. A major locus (RN) affecting muscle glycogen content is located on pig chromosome 1i. Mamm. Genome (in press).

Milan, D., P. Le Roy, N. Woloszyn, J.C. Caritez, J.M. Elsen, P. Sellier, and J. Gellin. 1995. The RN locus for meat quality maps to pig chromosome 15. Genet. Sel. Evol. 27: 195-199.

Rothschild, M.F., H.C. Liu, C.K. Tuggle, T.P. Yu, and L. Wang. 1996. Analysis of pig chromosome 7 genetic markers for growth and carcass performance traits J. Anim. Breed. and Genet. (in press)

Rudat, I., C. Looft, N. Reinsch, and E. Kalm. 1995. The Kiel RN project mapping a major locus for meat quality. Proc. of the 46th Annual Meeting of the European Association for Animal Production.

Yu, T.P., C.K. Tuggle, C.B. Schmitz, and M.F. Rothschild. 1995. Association of PITl polymorphisms with growth and carcass traits in pigs. J. Anim. Sci. 73:12821288.

__________________
1 Presented to the National Swine Improvement Federation, Des Moines, Iowa, December 1, 1995.