Advances in Genetic Improvement of Meat Quality Traits in Pig

This article reviews the advances in molecular genetics that have led to the genetic improvement in meat quality in pigs. During the past few decades, genes, or markers associated with genes have been identified that affects meat quality. The huge amount of information emerging from genomic studies is opening up new opportunities for the genetic selection of pigs. Several genes that influence meat quality have already been, or are close to being identified. Some of them have already been implemented into breeding programs by marker-assisted selection. Fields like functional genomics and proteomics are helping in better understanding of the function and regulation of genes and how these participate in complex networks controlling the phenotypic characteristics of a trait. An integrated approach of genomics and proteomics with bioinformatics tools will further exploit the available molecular genetics information. This will allow animal breeders to make progress more rapidly to meet the demand for high quality pork.


Introduction
During the last 20 years, the genetics underlying pork quality traits has become subject of increased interest of research due to the processors', packers' and consumers' demands for food with better quality [1].Meat quality traits are recognized as quantitative traits, which are affected by genetic and environmental factors [2,3].While some pork quality traits can be divided in either technological (water-holding capacity, firmness, glycogen content) or sensorial (tenderness, juiciness, flavor), there is also some overlap for some traits that belong to both categories (pH, colour, muscle fibre characteristics) [4].Generally, good pork quality is associated with darker (more red) meat, increased water holding capacity, juiciness, tenderness and better texture.
From the beginning of animal domestication, man have been trying to get progeny from the animals that seemed best fit for the purpose, changing the genetic design of animals without knowledge of the underlying genes.For most of this period selection was empirical and directed by, what Darwin refers to as, unconscious selection.
Selection in pig breeding went through different phases.Initial phase of selection focused on easily measurable parameters such as body weight, growth rate, number of teats, number of piglets born alive, number of piglets weaned, duration of the interval weaning to serving, number of productive cycles per sow, and so on.Later on, internal parameters were included.Using ultrasonic equipments, back fat thickness and loin eye area were measured on living animals and taken into account in the selection programme.Computer technology was also introduced.Databases were created which were then analysed through BLUP (Best Linear Unbiased Prediction) programmes to give an estimated breeding value for each individual [5].
It was not until the late 18 th and 19 th century and in particular after the 1950s, the pig breeding increasingly based on scientific genetic theory.The pig breeding programmes have been very successful in effecting genetic improvement of economically important traits, especially daily gain (+100%), backfat thickness (-75%), and feed efficiency during the last decade [6].

Genetic Maps, Gene Discovery and the Molecular Age
In the early 1990s, the PiGMap consortium was formed to better understand the pig genome [7].Since then, geneticists have come a long way to generate genetic maps of the porcine genome.During initial years, genetic linkage maps consisting of anonymous genetic markers and a limited number of known genes were developed.Development of comparative genome maps also aided in the quest for potentially useful genes in pigs [8].
The development of molecular markers and genome maps facilitated application of molecular genetics approaches, increasing the rate of genetic gain and identifying genes and polymorphisms controlling variation in traits of interest.The hunt for QTL in pigs has been ongoing for more than two decades now, beginning with the first report of QTL for fatness on pig chromosome-4 [9].Interestingly, it was the first QTL reported in livestock.As of 17 th May, 2017 there are 17,955 pig QTL in the database from 576 publications representing 635 pig traits (Table 1).
The population structures used for the majority of these studies involved experimental crosses using pig breeds exhibiting extremes for the phenotypes of interest, with the expectation that alleles for QTL controlling these phenotypes would be segregating.The class of genetic markers used for genotyping most populations was microsatellites because there were genome-wide microsatellite marker sets available, providing the ability to detect QTL using linkage analysis methods typically with 100-200 microsatellites [10].Significant effects of major genes and candidate genes have been reported [11].For example, the MC4R (melanocortin 4 receptor) gene was found to significantly affect growth rate by 7 to 9% by influencing feed intake.The MC4R gene maps to chromosome 1 close to a significant QTL.Due to the MC4R effect on feed intake, variation in this gene is also significantly associated with 5 to 8% differences in back fat and relates to one QTL for back fat thickness on chromosome-1 [12].The work done so far on some candidate genes for their associations with pork quality traits is presented in table 2.

Marker Assisted Selection (MAS)
Marker-assisted selection (MAS) combines the genetic information with the classical performance records and genealogical information to increase selection accuracy.There are two types of MAS: one is the candidate gene approach which uses causative mutations with a major effect on a particular trait, and another uses linkage disequilibrium (LD) between markers with QTL.The breeding value is predicted based on the combined information of polygenic effects and significant markers.MAS can be more useful for traits that are expressed late in the life of the animal, have low heritability, are sex-limited, expensive to measure or controlled by a few genes.However, for most recessive alleles with lethal or semi-lethal effects, natural selection will maintain their frequencies very low making MAS unnecessary.
The impact of MAS in breeding programs has been modest because the QTL that exceeds the chosen significance thresholds usually accounts only for a minor fraction of the trait variance.However, it would be a question of time whether availability of the number of markers was the only limitation for the success of MAS [45].

Potential of Mas
Information at DNA level can assist in the selection of quantitative traits including those that can be selected by traditional means.Molecular information increases the accuracy of selection and therefore the selection response.Importantly, these responses can be sustained if new markers are continually identified.New markers can be added to the selection index as old markers begin to reach fixation.Many studies on QTL and candidate analyses have yielded interesting results.The QTL scans have identified several chromosomes that are now targets for further confirmation of the chromosomal region, advanced fine mapping of the QTL and positional comparative candidate gene analysis.MAS can accelerate genetic improvement of the herd by leading to decisions that predict improved performance of the animals [46].For complex traits controlled by multigenes and the environment, it should be realized that markers for MAS are not exhaustive and many other genes contribute towards the trait.The presence or absence of the numerous other ''unmarked'' genes and the production environment will determine whether an animal actually displays the desired phenotype [47].Implementation of MAS requires careful consideration of issues ranging from sample collection and storage, genotyping and data analysis [48].Furthermore, MAS should be seen as a tool to assist with, not as a replacement for traditional selection techniques.

Genomic Selection
Recent developments in technology have removed limitations of MAS, by using high-density of markers across the genome to capture a larger percentage of the genetic variance for the trait of interest [49].Advances in genome sequencing, identification of large numbers of SNPs, and the cost-effective high-throughput genotyping of tens of thousands of such SNPs, combined with the further development of statistical analysis methods has led to a paradigm shift in the strategy of using genetic markers for the prediction of breeding values for genomic selection (GS).GS assumes that each QTL is in LD with at least one closely linked marker and the effects of chromosome segments will be the same within the population.Therefore, the density of markers must be high enough to ensure that all QTL are in LD with at least one marker [50].The commercially available SNP genotyping panels for the pig is presented in table 3.
A reference population with genotypic and phenotypic records is needed to carry out genomic selection to develop a genomic estimated breeding value (GEBV) prediction equation [52].This prediction equation is then applied to the candidate population with genotypic records but without phenotypic information to predict GEBV for each individual, and this is used to select the best animals for breeding in the selection population [49,52].
Development of a first commercial SNP panel for high throughput genotyping [53] and sequencing of the pig genome [54] were the starting points for the application of genomic selection in pigs.The first SNP panel commercially available from Illumina contained about 60K SNPs that cover all autosomal and X chromosomes [53].

Perspectives of Genomic Selection in pigs
Genomic selection in pigs has only recently been implemented and it is still not so common as in dairy cattle.However, field applications of genomic selection are opening new opportunities also in pig breeding.Pig breeding schemes currently have developed a very efficient data recording scheme which easily could include genomic information.
Breeding programmes should be eventually redesigned to fully capitalise the benefits of genomic selection [55].Selecting animals based on their GEBV or GS would allow selection of animals at an early stage of their life and increase accuracy of prediction.Consequently, this would increase the genetic gain by decreasing generation interval, increasing the selection accuracy and the frequency of favourable alleles [52].Although GS is revolutionizing, it faces challenges like requirement of large reference population size to predict GEBV accurately.Whereas, it is known that larger reference populations result in higher genomic prediction accuracy, in most of the studies so far small reference populations have been used for genomic prediction [52].Involvement of multiple breeds in livestock industries poses another major challenge in GS.For divergent breeds, it has been shown that large reference population sizes and more than 300,000 SNPs are needed [56].Using non-additive genetic effects is another challenge in using genomic selection.It might be beneficial to improve the selection accuracy by including non-additive genetic effects such as dominance and epistatic effects.Methods for genomic prediction like estimation of interactions between high density of markers [57], estimation of both dominance and epistatic effects by using single-marker Bayesian approach [58], and the dominance model [59] allow us to estimate non-additive genetic effects.
Quite soon, it is expected that SNP genotyping technologies will be substituted by sequencing-based approaches in genomic selection, as sequencing costs should decrease with the introduction of new sequencing approaches and technologies.However, despite the decrease in costs, sequencing might still be too expensive for implementation on a large number of animals [51].

Genes Affecting Meat Quality Traits in Pigs Myogenic regulatory factors (MRF) gene family
Myogenesis and postnatal muscle tissue growth are regulated by the myogenic regulatory factors (MRF) gene family.The MRF gene family consists of five structurally related transcription factors: myogenin, MyoD1, MRF-4, MYF-5 and MYF-6.They regulate both skeletal muscle fiber development and postnatal hypertropic growth.The MyoD and MYF-5 genes regulate proliferation of myoblasts and satellite cells (postnatal type) by having the ability to fuse with existing myofibers, but lacking the ability to form new myofibers.Myogenin is expressed during fusion of cells to form multinucleated myofibers.The role of MYF-6 has been mainly described as maintaining the muscle tissue.The MRF gene family is considered as strong candidate gene for skeleton muscle mass, i.e., meat mass.The candidate gene approach generally comprises of recognizing the gene of a trait on the basis of physiological function, followed by study of genomic variation of that gene to detect the ultimate causal mutation leading to modified protein with a single changed activity or slightly change in the level of expression of an unmodified protein.Finally, a candidate gene for the given trait can be validated by means of an association study.In general, an association study is comprised of investigation of resource families with large population (for example fullsib families) by genotyping of the polymorphic site at candidate gene.

The stress susceptibility gene mutation, RYR1
One of the first discovered gene mutation affecting pig meat quality was ryanodine receptor gene mutation (RYR1), also known as the Halothane gene (HAL) or malignant hyperthermia gene (MH).The ryanodine receptor belongs to the calcium channels.Among several different ligands for this receptor, anaesthetic halothane can activate them causing releasing of Ca 2+ , especially rapid releasing calcium ion is observed in hypersensitive mutated form.Animals carrying such mutation are characterized by increased muscle mass and reduced fat content.Since effective selection programme for reduced back fat and increased lean yield was introduced, this mutation has widespread influence among commercial pig breeds.This mutation is responsible for the increase of the occurrence of porcine stress syndrome (PSS) and pale soft exudative pork (PSE).Porcine stress syndrome manifests as a loss of consciousness or even death of animals caused by environmental stress, such as transport from farms to slaughterhouses.PSE is described as pale and watery pork, and is associated with rapid drop of pH and strong muscle post-mortem rigor contraction.Halothane sensitivity has also been shown to reduce the value of pork for the production of cooked hams and curried dried hams.Pig breeds differ in occurrence of these pork defects.This mutation is most frequent in Pietrain breed compared to other pig breeds.Since carriers of this mutation can also produce undesirable PSE pork, a test for detection of this mutation has been in use in pig breeding programs.

The rendement napole (RN) gene
Rendement Napole (RN) mutation is related with higher level of muscle glycogen and causes a paler colour, poor ham quality and high drip loss.RN mutation causes a 70% higher level of glycogen.Studies aimed at localizing and describing the causal gene mutation responsible for this pork defect, revealed a gene of muscle specific monophosphate-activated protein kinase protein (PRKAG3).A single nucleotide mutation (G→A) that inhibits the activity of this enzyme and responsible for breaking down storage of glycogen has been discovered.Use of such mutation has found favor in pig industry.The RN is most frequent in Hampshire breed, but it was estimated that 80% of the high glycogen pork came from commercial white-breed pig population.The high value of glycolytic potential was not due to the presence of the RNallele (200Q) in the tested commercial pigs [26].However, their data confirmed the effect of other mutations at the PRKAG3 locus (T30N and G52S) on meat quality.Moreover, they observed that muscle glycogen content significantly depended on PGAM2 (phosphoglycerate mutase 2) genotype.

Marbling fat HFABP and AFABP genes
Intramuscular fat content has a major influence on meat quality in pigs.Amount of intramuscular fat decides of marbling of the muscle.It was determined that pork loin should have at least 2% fat in the lean meat; otherwise it is too dry after cooking.Fatty acid-binding proteins (FABPs) are members of the superfamily of lipidbinding proteins.There are nine different tissue-specific FABPs that have been identified: L (liver), I (intestinal), H (muscle and heart), A (adipocyte), E (epidermal), II (ileal), B (brain), M (myelin) and T (testis).The primary role of all the FABP family members is regulation of fatty acid uptake and intracellular transport.Because of their physiological properties, some FABP genes were tested in order to identify mutations altering lipid metabolism.Heart fatty acid binding protein (H-FABP) and adipocyte fatty acid binding protein (A-FABP) are involved in transport and metabolism of fatty acids.These two genes were chosen as candidate genes for improvement of muscle marbling content but they did not increase the back fat.In H-FABP, the tested mutation causing improvement of marbling content was placed in promoter region.A-FABP gene was analyzed using microsatellite sequence placed in the second intron.
The HFABP and AFABP were responsible for an increase of fat content by 1% in Duroc and Meishan pig breed [60][61][62].Polymorphisms in the adipocyte and heart fatty acid-binding protein genes, A-FABP and H-FABP, are significantly associated with genetic variation of intramuscular fat content in a Duroc pig population.This was supported by information about the role of H-FABP but not for A-FABP in a Meishan crossbred pig population.In view of that, the effect of closely linked genes could not be excluded.

Calpastatin (CAST) gene
The rate and extent of skeletal muscle growth depends mainly on three factors: rate of muscle protein synthesis, rate of muscle protein degradation, and the number and size of skeletal muscle cells.Studies have shown that calpain activity is required for myoblast fusion [63] and cell proliferation in addition to cell growth [64].A number of studies have shown that the calpain system is also important in normal skeletal muscle growth.Increased rate of skeletal muscle growth can result from a decreased rate of muscle protein degradation, and this is associated with a decrease in activity of the calpain system, primarily due to a large increase in calpastatin activity [65].These observations suggest that genes coding for calpains and calpastatin may be considered as candidate genes for lean content of carcass in pigs.
By investigating the impact of CAST HinfI, RsaI and MspI genotypes on carcass traits in different pig breeds, it was shown that CAST/HinfI genotype had no impact on carcass traits, whereas, CAST/RsaI and CAST/MspI affected some of the meat and fat deposition traits in pigs [66].By analysing meat quality traits influenced by CAST/MspI genotype, it was observed that animals with BB genotype at this locus were characterized by the most profitable values of all analysed traits [67].

Growth hormone (GH) and growth hormone releasing hormone (GHRH) gene
The GHRH is an endogenous stimulator of somatotropin secretion.It is released by hypothalamus and stimulates the proliferation of pituitary somatotroph cells during their development regulating production and secretion of GH.GH is a peptide hormone with about 190 residues which regulates growth, development and various metabolic activities.The AluI polymorphism in the third exon of GHRH gene was found to be significantly associated with fat thickness and meat content of carcass in different breeds of pigs [68] and with average daily gain and fat thickness [69].Although most amino acids of the GH protein are conserved, there are still many single nucleotide polymorphisms which are reported in the traits of growth, lean rate and milk production.The GH DdeI polymorphisms are associated with fat thickness (P=0.0326) and average daily gain (P=0.0127)[69].Different DdeI, NarI and BsmNI polymorphisms have been found in the coding sequence of GH in different breeds.
Peroxisome proliferator-activated receptor gamma coactivator-1 (PPARGC1) Fat deposition and its body distribution in pigs is economically important for production of pigs that are low in back fat but having intramuscular fat, which contributes to the improvement of sensory traits of meat.Recent research has focused on the identification of candidate gene for predicting meat quality and muscle tissue development.The peroxisome proliferatoractivated receptor-gamma coactivator-1 (PPARGC1) is a transcriptional coactivator of many nuclear hormone receptors including peroxisome proliferator-activated receptor-gamma, which has been shown to be involved in lipid metabolism and play central regulatory role in adipocyte differentiation [70].Because of its role in body composition and fat distribution, PPARGC1 is a candidate gene for fatness and leanness.
The complete coding sequence of the porcine PPARGC1 gene including a single nucleotide polymorphism in exon 8 has been determined in recent times.A T/A substitution at nucleotide position 1378 results in an amino acid substitution (Cys→Ser) at position 430 in the amino acid sequence.Differences of T/A allele distribution between Chinese and Western pig breeds were described [71].T (Cys) allele was present in all 13 analysed pig breeds.However, A (Ser) allele was present only in Western pig breeds, and in one Chinese (Taoyuan) breed.

Conclusion
Demand for high-value animal protein is expected to increase fuelled by increase in global human population, urbanization and socio-economic mobility that will require efficient livestock production without compromise on food quality.This shall be harnessed by application of new technologies to accelerate genetic gain.Gene expression profiling approach has helped understand the biology of porcine muscle growth and development better, and important genes and signalling pathways are being identified that may affect meat quality [72].Emergence of "-omics" will open up new vistas that has already uncovered associations capable of generating new biological hypotheses.Nutrigenomic approach may also help in identifying nutrition-responsive major genes for meat quality trait formation [73].Although rapid genetic improvement for both existing and new traits will be a challenging task, integration of new and emerging technologies with conventional breeding is expected to help achieve progress more rapidly than was possible in the past to ensure that the demand for quality pork is satisfied.