| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL GENETICS |
,2
* Department of Animal Science, Iowa State University, Ames 50011;
and
Sygen International, Franklin, KY 42134;
and
Sygen International, Rosmalen 5241 LN, The Netherlands; and
and
Sygen International, Kingston Bagpuize OX13 5AS, U.K.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Key Words: Calpastatin • Haplotype • Meat Quality • Pig • Tenderness
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
).
Calpastatin (CAST) is a specific inhibitor of µ- and m-calpain (EC 3.4.22.17) proteases. There is evidence indicating that in different species, calpastatin activity postmortem is highly related to meat tenderness (Koohmaraie et al., 1991; Sensky et al., 1998; Parr et al., 1999).
Starting with Lonergan et al. (1995), several association studies have investigated sequence variation in calpastatin as a source of genetic markers that may influence meat tenderness. Although some suggestive associations were reported in cattle (Chung et al., 1999, 2002; Barendse, 2002) and pigs (Kocwin-Podsiadla et al., 2003), none of these studies found any AA sequence variation or changes in the potential regulatory regions that could explain phenotypic differences in meat quality.
Using a Berkshire x Yorkshire pig family (Malek et al., 2001b), suggestive QTL for raw firmness score, cooked meat quality traits, including average Instron force, and the sensory traits juiciness, chewiness, and tenderness scores were revealed on swine chromosome 2 (SSC2). Based on its function and location, CAST was considered to be a good candidate gene for these QTL. The objective of this study was to evaluate this positional candidate gene for association with variation in pork quality.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
), as well as blood samples and muscle tissue from the ham and loin muscles from more than 10 individual F3 animals for RNA/DNA isolation purposes. A large set of tissue samples from four different commercial purebred lines of pigs were collected at a single plant over a 3.5-yr period before 2001. These animals were produced at two different farms and slaughtered over 73 slaughter dates. The lines of pigs included Large White, Duroc, Duroc/Large White Synthetic, and a Composite line. Several of these lines were used to produce a set of crossbred slaughter pigs. Tissue samples were collected for RNA/DNA isolation purposes from the longissimus dorsi of eight Large White individuals and one Meishan. Genomic DNA was isolated either from whole blood by standard salting out procedures or from ear notches using the DNeasy tissue kit (Qiagen, Valencia, CA). Total RNA was extracted from ham (semimembranosus) and loin muscle tissue (longissimus dorsi) using the TRIzol reagent method according to manufacturer instructions (Invitrogen-Life Technologies, Carlsbad, CA).
Phenotypic Trait Measurement
B x Y F2 Reference Family.
Meat quality measures were made on the longissimus dorsi and included firmness on raw meat, several sensory traits (juiciness, tenderness, and chewiness score) on cooked meat, and average Instron force. For trait descriptions and methods used, see Malek et al. (2001a, b).
Commercial Purebred Pig Lines.
Subjective firmness was measured and scored from 1 to 5 on the longissimus dorsi, with higher values indicating greater firmness. Scores were determined visually on raw meat at a packing plant 24 h after slaughter on individuals sampled from the four purebred commercial pig lines (Berg, 2000).
Commercial Crossbred Slaughter Pigs.
Longissimus dorsi pH, instrumental tenderness (shear force), cooking loss, and sensory measurements (juiciness and tenderness) were evaluated at the University of Illinois after 14 d aging. Shear force was measured with an Instron 1222 Universal Testing Machine (Instron, Canton, MS) fitted with a Warner-Bratzler shear attachment. Sensory juiciness and tenderness were scored from 1 to 10, with higher values indicating juicier or more tender meat. For cooking loss measurements, a chop was cut from the longissimus dorsi muscle after aging, weighed, cooked to 70°C in a Farberware open-hearth electric grill (Farberware Inc., Westbury, NY), refrigerated until cool, and reweighed. During cooking, the temperature in each chop was monitored in its center with thermocouples. Cooking loss was calculated from weights taken before and after cooking and was expressed as a percent.
Linkage Mapping and QTL Analysis
An intercross between B x Y pig breeds, yielding 525 F2 offspring, was generated to map QTL for meat quality, growth, and carcass composition (Malek et al., 2001a,b). Additional microsatellite markers were subsequently genotyped across the three-generation family to increase the number of informative meioses in the region of the CAST gene, which seemed to harbor QTL for firmness and related traits (Thomsen et al., 2002). Mapping of the CAST gene to the SSC2 linkage map was performed by using a previously reported CAST MspI substitution (Ernst et al., 1998), and was accomplished using CRI-MAP (Green et al., 1990).
PCR, RT-PCR, and Polymorphism Discovery
Based on the CAST pig cDNA sequence available in GenBank (M20160), primers were designed to amplify the entire coding region of the type III CAST skeletal muscle isoform. The cDNA obtained from B x Y F3 individuals and from Large White and Meishan individuals were used in polymorphism discovery. The PCR was performed using AmpliTaq Gold DNA polymerase (Applied Biosystems, Foster City, CA) and standard PCR protocols. Reverse transcription of total RNA was performed by random hexanucleotide priming and Superscript II (Gibco BRL, Gaithersburg, MD) according to the manufacturer’s protocol, and PCR primers were designed to amplify the full coding region of the type III CAST isoform (primers: set A, forward 5' CCGAGCCC AACCAGGAATGC 3' and reverse 5' TTTGTCCTTGA CCTCTTCTAATC 3'; set B forward 5' CCAGAAGATG GAAAGCCTGTG 3' and reverse 5' TGAAGATTGTCT ATGCAGATACC 3'). Amplicons were sequenced using dye terminators on an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems). The Sequencher software (Gene Codes, Ann Arbor, MI) was used to assemble the sequences and to identify polymorphisms. The sequence polymorphisms were confirmed using restriction enzyme tests.
Genotyping and PCR-RFLP Analysis
The region flanking each newly discovered missense mutation (Ser66Asn, Arg249Lys, and Ser638Arg) was PCR amplified (primers: Ser66Asn forward 5' GTAAA GCCAAAGGAACACCCAG 3' and reverse 5' TTTTTA TTTCTCTGATGTTGGCTGTGCA 3'; Arg249Lys, forward 5' AAATCTACTGGAGAGGTTTTGAA 3', and reverse GACTTCTCCCGAATCAGTTCC; Ser638Arg, forward 5' AAACCTATTTTCAGGGATATGGG 3' and reverse 5' CCTTTGTTGTGTTCTCTGAGG 3') and then digested with ApaLI (Ser66Asn), Hpy188I (Arg249Lys), and PvuII (Ser638Arg). The Ser66Asn reverse primer was modified relative to the original swine cDNA sequence to create an ApaLI restriction polymorphism to be able to differentiate the alleles by PCR-RFLP. The Ser66Asn substitution was not used in these association studies because initial analysis with 200 individuals from eight Western breeds revealed that it was in complete linkage disequilibrium with Arg249Lys. Potential phosphorylation sites for adenosine cyclic 3', 5'-monophosphate dependent protein kinase (PKA) were identified using NetPhos 2.0 prediction server (Blom et al., 1999).
Statistical Analyses
Berkshire x Yorkshire F2 Population Analysis.
Haplotypes defined by the CAST Arg249Lys and CAST Ser63-8Arg polymorphisms were assigned in the B x Y population based on complete genotyping of the parents. The substitution effects of the haplotypes were estimated using a mixed model (SAS PROC MIXED, SAS Inst., Inc., Cary, NC) that included litter as a random effect, and slaughter date and sex as fixed effects. One variable was included for each haplotype, with values –1, 0, and 1 corresponding to the animal having 0, 1, or 2 copies of the haplotype in question. The effect of the haplotype variables was tested by calculating the likelihood ratio of this full model and a reduced model without the haplotype variables. Significance P-values were obtained from a 
2 distribution with 2 df for the values of –2 times the log likelihood ratio. Haplotype substitution effects were estimated for traits affected (P < 0.10 or P< 0.20 in more then one independent data set) by the CAST haplotypes.
Commercial Purebred Lines Analyses.
Haplotypes defined by the CAST Arg249Lys and CAST Ser638Arg polymorphisms were predicted based on inferences derived from the genotypes of homozygous individuals. Haplotype analysis was performed (as described in the previous paragraph) on the combined data from four commercial lines with a mixed model (SAS PROC MIXED) that included sire (n = 165) as a random effect. Potential relationships between the sires were ignored in the analysis. Slaughter date and line were included as fixed effects. Sex and farm were not included because all traits were measured only in females and only one farm was represented on each slaughter date. Associations between the individual CAST polymorphisms CAST Arg249Lys and Ser638Arg and meat quality traits also were tested by replacing the haplotype variables with marker genotype as a fixed effect (P < 0.10). Line x genotype interactions were tested in this model and not found to be significant. Least Squares (LS) means for the three genotype classes were obtained for both polymorphisms and contrasts tested for significance (nominal P < 0.05).
Commercial Crossbred Pig Data Analysis.
No pedigree information was available for this data set, but haplotypes were predicted based on inferences derived from the genotypes of homozygous individuals. Analysis of haplotype associations was based on a model with the breed composition of the six crossbred products included as a fixed effect and haplotypes as described under the statistical analysis for the B x Y F2 population.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). Using additional markers and a previously reported CAST polymorphism (CAST – MspI; Ernst et al., 1998), a new SSC2 map was developed. Extended QTL analysis of this B x Y family for growth and meat quality traits (Thomsen et al., 2002) using the new map confirmed a QTL significant at the 5% chromosome-wise level for firmness (Figure 1; Table 1), and a nominally significant QTL for average Instron force and juiciness in the CAST region (Figure 1; Table 1). The firmness QTL was located approximately 13 cM away from the main group of QTL. The favorable alleles at these QTL each had an additive effect and were derived from the Berkshire breed, which is generally regarded as having very good meat quality (Table 1).
|
|
,b), the CAST gene was a good candidate gene in this chromosomal region. By sequencing the entire coding region of the CAST gene, including parts of 5' and 3' untranslated areas, in B x Y F3 individuals with extreme values for meat quality, three missense mutations (CAST Ser66Asn located in Domain L, CAST Arg249Lys in Domain 1, and CAST Ser638Arg in Domain 4) were identified (see Figure 2). In the B x Y family and also in 200 individuals from several commercial lines, the CAST Ser66Asn polymorphism was in complete linkage disequilibrium with the CAST Arg249Lys polymorphism. Five silent mutations were also found in the B x Y resource family. Additional sequencing using several individuals from a Large White line and a Meishan individual revealed four additional polymorphisms. These included one in the 5' untranslated region (UTR), a Leu306Val missense mutation located in Domain 2 (in a nonconservative position within Subdomain B), and two Meishan specific mutations: one silent and one missense, the last one being located in Domain 3, Subdomain C (Figure 2). Differences in allele frequency at each polymorphic site between the founders of the intercross family were intermediate (e.g., 0.75 and 0.39 for the Arg249Lys and 0.75 and 0.50 for the Ser638Arg in the Berkshire and Yorkshire founders, respectively).
|
). Four haplotypes were determined (B x Y and sequencing work) and/or predicted to be present in the populations used here, three were found at a significant frequency in most of the Western type commercial breeds (Haplotypes 1, 2 and 3), and the fourth was Meishan specific (Haplotype 4). Although additional sequencing may reveal more haplotypes, their frequency is expected to be low so that they would not be useful for analysis of this data set.
Association Study
Berkshire x Yorkshire F2 Population Analysis.
Two of the polymorphisms, Arg249Lys and Ser638Arg, were considered to be the most interesting because of their missense characteristics and because these two polymorphisms, assembled together in haplotypes, specified the three individual haplotypes revealed in the CAST gene in the commercial populations (Figure 2). These three haplotypes (1, 2, and 3) were also present in the founders of the B x Y resource family. As with the individual substitution sites, the differences in haplotype frequency were not large between the founders of the intercross family (e.g., 0.75 and 0.39 for Haplotype 1, 0 and 0.11 for Haplotype 2, and 0.25 and 0.5 for Haplotype 3 in the Berkshire and Yorkshire founders, respectively).
Individual substitution effects were analyzed and significant differences were detected for both polymorphisms (results not shown); however, because of the significant linkage disequilibrium between the CAST polymorphisms, the haplotypes were tested instead of the individual polymorphisms and found to have an effect on firmness (P < 0.01), Instron force (P < 0.01), juiciness score (P < 0.05), and a trend was seen for chewiness score (P < 0.10). Effects were estimated as contrasts between the variables representing the observed haplotypes (Table 2). Substitution effects were discovered in raw muscle for firmness between Haplotypes 1 and 2 (P < 0.01) and between Haplotypes 2 and 3 (P < 0.05). A substitution effect was also revealed for meat quality in cooked meat between Haplotypes 1 and 3 for subjective juiciness (P < 0.01), chewiness (P < 0.05), and average Instron force (P < 0.01). Tenderness score was not significant but haplotype substitution effects were included in Table 2 because the direction of the substitution effects were the same as for the other sensory traits and because the same trait was found at P < 0.20 in this data set as well as in the commercial crossbred pig data set (see Table 5). Haplotype 1 was the favorable haplotype for all traits measured on cooked meat, as it is associated with higher juiciness and lower chewiness and average Instron force. Haplotypes 1 and 3 were associated with a softer muscle compared with Haplotype 2.
|
|
), and Haplotype 1 associated with a lower firmness. Haplotype 1 was also associated with less firmness compared with Haplotype 3 in the B x Y population. In addition, the differences between haplotype effects were relatively small for firmness. In this data set, we also tested the Arg249Lys and Ser638Arg substitutions separately. The overall significance for both polymorphisms was very close to the 0.05 significance level and LS means of genotypes across lines (Table 4) showed significant differences for firmness. The 249Lys/249Lys and 638Arg/638Arg genotypes were associated with lower firmness, which is in agreement with the results of the B x Y haplotype association analysis. The LS mean of the 249Lys/249Lys genotype was different from the LS mean of 249Lys/249Arg genotype (P < 0.05) and the 638Arg/638Arg and 638Arg/638Ser genotypes were different from the homozygote for the 638Ser allele (P < 0.05).
|
|
). Haplotype effects were found for percentage of cooking loss (P < 0.001) and juiciness score (P < 0.05), trends were found for longissimus dorsi pH (P < 0.10) and shear force (P < 0.10) and effects for tenderness score were found consistent with the results from the B x Y population (Table 2).
The effect of Haplotype 1 was different from the effect of Haplotype 2 for shear force (P < 0.05) and percentage of cooking loss (P < 0.001; Table 5). Also the effect of Haplotype 1 differed from that of Haplotype 3 for longissimus dorsi pH (P < 0.05), percentage of cooking loss (P < 0.001), and subjective juiciness (P < 0.01). Haplotypes 1 and 3 tended to differ for subjective tenderness (P < 0.10). For all traits, Haplotype 1 was the favorable haplotype and it was associated with higher pH, better subjective tenderness and juiciness, and significantly lower shear force and percentage of cooking loss.
Phosphorylation Site Prediction
Using NetPhos 2.0, six potential adenosine cyclic 3', 5'-phosphate-PKA phosphorylation sites were predicted in CAST, of which four were at serine AA residues. Experimentation has revealed that for most protein kinases, phosphorylation of serine is generally preferred over threonine (Kennelly and Krebs, 1991). Interestingly, CAST Ser66Asn, which was in complete linkage disequilibrium with Arg249Lys, and Ser638Arg both affect the PKA consensus sequence for two of these predicted phosphorylation sites (Figure 3). Specifically, 638Ser may eliminate a predicted PKA recognition sequence, replacing the required arginine with a serine from the –2/–3 position relative to the potentially phosphorylated serine. In the L domain, the 66Ser allelic variant introduces a potential PKA recognition sequence.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). Even though their substrates exclude major key structural proteins, such as actin, myosin, and 
-actinin, there is significant evidence linking the calpains and, indirectly, calpastatin, a calpain inhibitor, to tenderization in beef, pork, and lamb (Sensky et al., 2001). Also Ca2+ stimulates contractions in pre-rigor muscle and may stimulate more rapid glycolysis and pH decline. The results reported in this study provide important evidence in favor of the presence of new alleles/haplotypes of the CAST gene associated with several important meat quality traits in pigs. This conclusion is based on the following facts: 1) the known role of calpastatin as a calpain inhibitor; 2) observation of suggestive QTL for firmness, average Instron force and related subjective traits such as juiciness, chewiness, and tenderness scores in a B x Y F2 family on SSC2 in the region where the CAST gene is located; 3) association results between the identified substitutions and/or haplotypes that explain most of the CAST coding sequence variation, with a number of relevant meat quality traits in several different commercial pig lines and data sets; and 4) the potential effect of the CAST Ser66Asn and Ser638Arg substitutions on phosphorylation of CAST by PKA as a result of changes in the PKA recognition sequence.
The evidence of the suggestive QTL for Instron force and related subjective meat quality traits on the SSC2 in the B x Y resource family was followed by mapping of the CAST gene in the same linkage map. The position of CAST under the QTL peaks suggested that CAST might be responsible for the observed phenotypic variation. By sequencing the entire coding region in B x Y F3 individuals with extreme values for meat quality, eight mutations, including three missense mutations, were identified. It is interesting to note that the differences in allele frequencies at each polymorphic site between the Berkshire and Yorkshire founders of the intercross family were not extreme. This may explain why the QTL was only significant for firmness and the QTL effect was smaller than the haplotype substitution effect (Haplotype 1 vs. 2) from association analysis, as the model used evaluates the difference in effects associated with an average allele derived from each of the parental breeds.
Two of the missense mutations, CAST Arg249Lys and Ser638Arg, are located in or close to Subdomain C of their respective domains (see Figure 2). This subdomain potentiates the inhibitory activity of CAST (Takano and Maki, 1999). It has been suggested that single mutations in conserved regions of any of the A or C subdomains might affect CAST activity (Ma et al., 1994), even though they do not have inhibitory properties like Subdomain B. Both these substitutions (Arg249Lys and Ser638Arg) are outside the most conserved area of Subdomain C: KPxxEDDxIDALSxDF (reviewed by Takano and Maki, 1999), but the Ser638Arg mutation is separated by just one AA from this sequence. Recently, Tompa et al. (2002) suggested that the A and C subdomains are activators of calpains, which, if confirmed, would open new directions for the study of the calpastatin-calpain system.
The Ser66Asn polymorphism is situated in Domain L. The function of this domain is not clear even though its sequence is well conserved between mammalian species. Recently, Hao et al. (2000) demonstrated the role of the L domain in regulation of L-type Ca2+ channels in guinea pig cardiac myocytes, suggesting a role of CAST in restoring Ca2+ channel activity, which facilitates Ca2+ influx and subsequently activation of calpain.
The major haplotypes observed can be separated into two groups based on sequence variation; one group is represented by Haplotypes 1 and 4 and is characterized by the presence of the same blocks of allelic variants for the adjacent substitutions located in the codons 194 to 211 and 237 to 249. The second group is represented by Haplotypes 2 and 3 (Figure 2). The association results presented do not include Haplotype 4, but significant contrasts are seen between Haplotype 1 and either Haplotype 2 or 3, and with one exception not within the group of Haplotypes 2 and 3. It will be interesting to investigate whether these two different haplotype groups share a different genetic origin and whether the substitution effect of Haplotype 4 is similar to Haplotype 1. These haplotypes can be characterized by the missense variants identified in the L domain (Ser66-Asn) and the C subdomains in Domains 1 (Arg249Lys), 3 (Ala504Thr) and 4 (Ser638Arg). Haplotypes 2 and 3 differ at the substitution in Subdomain 4C with Haplotype 2 (in common with Haplotypes 1 and 4) having an arginine at this position, whereas Haplotype 3 has a serine. It is interesting to note that a serine is found at this position in human, bovine, rabbit, rat, and murine sequences. Haplotypes 2 and 3 also differ at position 306; however, this is regarded as conservative change (Leu306Val) that is unlikely to be significant. The other haplotypes differ from each other at two or three positions. For example, Haplotype 1 differs from the other three haplotypes in having 66Asn in the L domain and from Haplotypes 2 and 3 by having 249Lys at Subdomain 1C. Finally, Haplotypes 1 and 3 also differ at Subdomain 4C at the codon 638 substitution, where Haplotype 1 (and Haplotype 2) has an arginine instead of a serine in the case of Haplotype 3. These haplotypes can be summarized as NKAR, SRAR, SRAS, and SKTR at these positions (Haplotypes 1 to 4, respectively). Two of the polymorphisms were therefore sufficient to define the three major haplotypes present in the commercial breeds (Figure 2) and therefore the CAST Arg249Lys and Ser638Arg polymorphisms were used for both single substitution and haplotype association analyses.
In the B x Y family the haplotype analysis showed important differences between the effects of Haplotypes 1 and 3 for average Instron force, juiciness, and chewiness scores. For firmness, significant differences were revealed between the effects of Haplotypes 1 and 2 and also between Haplotypes 2 and 3. Haplotype 1 was demonstrated to be the favorable haplotype, and it was associated with juicier and tender pork that was also less firm (Tables 2 and 3). Low firmness scores are often considered to be unfavorable because they are generally associated with extreme values such as those observed in PSE pork. In this case, lower firmness may relate more to the higher tenderness score; however, the LS means of the analyzed genotypes are in the acceptable range for firmness. Sensory tenderness, which did not show significant differences between the haplotypes did, however, show a similar hierarchy, with Haplotype 1 being most favorable. Haplotype 1 (NKAR) differs from Haplotype 3 (SRAS) at three of the substitutions, whereas it only differs from Haplotype 2 (SRAR) at the first two positions. Haplotypes 2 and 3 only differ at the last position (Ser638Arg). Thus, there is some agreement between the number of AA differences (for these three domains) and the size of the differences between the meat quality traits (Table 2). Finally, similar results were obtained with commercial crossbred pigs, with Haplotype 1 being the favorable haplotype for loin pH, shear force, percent cooking loss and subjective tenderness and juiciness, with the most significant difference been between Haplotypes 1 and 3 (except for shear force and percent cooking loss; Table 5).
In the single allele substitution association analyses, P-values very close to significance (0.05) were obtained for both substitution sites (Table 4). Significant differences were obtained between the LS means estimated for each CAST genotype, with the 249Lys/249Lys and 638Arg/638Arg genotypes being associated with a lower firmness as expected. In the 249 codon substitution analysis, Haplotype 1 was contrasted with Haplotypes 2 and 3, and in the 638 substitution analysis, Haplotype 3 was contrasted with Haplotype 1 and 2. Although initial examination of these observations (Table 4) points to the larger effects being associated with the substitution at codon 638, some caution is needed because genotype comparisons have considerable differences in numbers of observations.
The associations observed mean that selecting for Haplotype 1 could have a significant effect on meat quality. The effect of Haplotype 1 on percent cooking loss is estimated to correspond to a decrease of 2.64% per copy of Haplotype 1 compared with Haplotype 2. For shear force, an effect of 0.17 kg per copy of Haplotype 1 was detected (even after 14 d of aging) compared with Haplotype 2. Significant differences were also found between Haplotypes 1 and 3 for longissimus dorsi pH (0.05 per copy), percent cooking loss, and subjective juiciness (0.38 per copy). The directions of the effects of the different traits follow the known correlations for these traits, supporting the suggestion that the observed haplotype associations are real.
Using an additional small data set of 64 crossbred Duroc and Yorkshire animals a significant effect was seen of the CAST Arg249Lys and Ser638Arg on Warner-Bratzler Shear force at d 3 and 5 postmortem. At d 5, an effect of 0.72 kg per copy of the 638Arg allele was observed (data not shown).
A comparison of CAST sequences from different mammals shows that the sequence around the putative phosphorylation sites is reasonably well conserved. For example, for the Ser638Arg position a serine is found in sequences of human (GenBank, BAA03747), cattle (AAA19643), rabbit (AAA31186), and mouse (BAA84768), although an alanine is present in sheep (AAB07483). At position 66, a serine or asparagine is present in pigs, and in this case, both AA have been reported with a serine in the case of sheep and cattle, asparagine in the mouse, and alanine in humans. However, the pig sequence is unique in that it is predicted to have full recognition sequences at these two sites for PKA.
Calpastatin solubility and cellular location are influenced by phosphorylation by PKA (Averna et al., 2001). Adachi et al. (1991) showed that phosphorylation increased the proportion of CAST bound to membranes and Salamino et al. (1997) demonstrated that phosphorylation decreased CAST inhibitory efficiency. Calpastatin phosphorylation could therefore influence proteolysis and ultimately have an effect on tenderness and other related meat quality traits. Analysis of the variation in porcine CAST suggests that both CAST Ser66-Asn and Ser638Arg could affect a PKA recognition consensus sequence. Haplotype 1 is the only haplotype (of Haplotypes 1 to 3) that has the 66Asn variant (as well as 249Lys), and it may therefore encode a peptide that will not be phosphorylated at the 66th AA (Figure 3). Haplotype 1 also contains an arginine at position 638, as part of the potential PKA recognition sequence, which might promote phosphorylation. Cong et al. (1998) demonstrated PKA phosphorylation of a truncated cattle CAST peptide consisting of Domains 1 to 4, which includes codon 638. Results from our association analysis demonstrate that Haplotype 1 is associated with higher tenderness and juiciness scores and this might be a result of advanced proteolysis of several key proteins. Haplotype 3 differs from Haplotype 1 at both of the potential phosphorylation sites discussed here (at AA 66 and 638), and encodes a peptide that will promote phosphorylation based on the 66Ser allele but not based on 638Ser. For this reason, Ser638Arg might be an explanation for the significant differences observed between Haplotypes 1 and 3. Based on this hypothesis, we expected to see the same differences between Haplotypes 2 (which has same 638Arg variant, like Haplotype 1) and 3, but did not. Therefore, we cannot attribute all of the differences to 638 substitution, and a different hypothesis must be proposed. Further biochemical study will be needed to test whether decreased PKA phosphorylation does occur at the CAST 66Asn and/or CAST 638Ser sites. The other substitution contained in the main haplotypes is at position Arg249Lys, and an arginine is found at this position in human, ovine, bovine, rabbit, rat, and murine sequences. It is not clear at this point which of the observed variants may be causative as the haplotypes contain more than one different substitution variant, and the associations also may be due to undetected genetic variation in the regulatory regions of CAST or by genes whose effects have been captured by haplotype analysis due to linkage disequilibrium. Two of the variants identified, Ser66Asn and Ser638Arg, based on their position, AA change, and potential effect on PKA phosphorylation of CAST, may influence postmortem proteolysis. At least in the commercial lines, the Ser66-Asn polymorphism was in linkage disequilibrium with the Arg249Lys polymorphism, even though they are located a significant distance apart in genomic DNA. Of the haplotypes present in these lines, the variants 66Asn and 249Lys of these substitutions are specific to Haplotype 1. The Ser66Asn is a nonconservative substitution and is located in a conserved area of Domain L. This may suggest an important functional role that might be associated with the observed phenotypic differences in meat quality.
|
Implications |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
2 Correspondence: 3033 Nashville Rd., Franklin, KY 42134 (phone: 270-598-7442; fax: 270-586-0312; e-mail: daniel.ciobanu{at}sygeninternational.com).
Received for publication September 24, 2003. Accepted for publication July 6, 2004.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Davoli and S. Braglia Molecular approaches in pig breeding to improve meat quality Brief Funct Genomic Proteomic, January 21, 2008; (2008) elm036v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Stearns, J. E. Beever, B. R. Southey, M. Ellis, F. K. McKeith, and S. L. Rodriguez-Zas Evaluation of approaches to detect quantitative trait loci for growth, carcass, and meat quality on swine chromosomes 2, 6, 13, and 18. I. Univariate outbred F2 and sib-pair analyses J Anim Sci, July 1, 2005; 83(7): 1481 - 1493. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A. M. van der Steen, G. F. W. Prall, and G. S. Plastow Application of genomics to the pork industry J Anim Sci, June 1, 2005; 83(13_suppl): E1 - 8. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
