| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL GENETICS |
* Graduate School of Agricultural Science, Tohoku University, 1-1 Tsutsumidori–Amamiyamachi, Aoba-ku, Sendai, Miyagi 981-8555, Japan;
and
National Institute of Livestock and Grassland Science, 2 Ikenodai, Tsukuba 305-0901, Japan;
and
National Livestock Breeding Center, Nishigo, Fukushima 961-8511, Japan;
and
School of Food, Agricultural and Environmental Sciences, Miyagi University, 2-2-1 Hatatate, Taihaku-ku, Sendai, Miyagi 982-0215, Japan; and
# Miyagi Prefecture Animal Industry Experiment Station, Hiwatashi 1, Iwadeyama-cho, Tamatsukuri-gun, Miyagi 989-6445, Japan
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
Key Words: Duroc pig • meat quality • multiple generation • quantitative trait loci • selected population
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
; Haley et al., 1994; Bidanel and Rothschild, 2002). However, the QTL information detected in these resource populations would be difficult to use directly for improvement of a purebred population, because their genetic backgrounds are different. Results of recent studies indicate that some QTL that were detected are segregating in purebred populations (Evans et al., 2003; Nagamine et al., 2003; Vidal et al., 2005). These QTL analyses within breeds were conducted using half-sib families. It has been suggested that a complex multigenerational pedigree design is more powerful than the half-sib or full-sib family designs for detecting QTL (Williams et al., 1997; Slate et al., 1999). George et al. (2000) described a method by which large complex pedigrees can be analyzed using a variance component approach. Slate et al. (2002) detected QTL in a wild red deer population with a 6-generation pedigree using a variance component approach.
Sato et al. (2003) reported a significant QTL region for intramuscular fat (IMF) on SSC 7 using a Duroc x Meishan crossbred population. Suzuki et al. (2005b) reported selection during 7 generations for meat production traits and IMF in the loin. The IMF increased, and some meat quality traits changed with selection (Suzuki et al., 2005a). Therefore, the experiment described herein was conducted to confirm the existence of additional QTL related to IMF and the existence of additional QTL that might be specific to this Duroc population, which has a complex multigenerational pedigree structure, including 8 generations. This study examined the genotypes of 10 microsatellite markers that are distributed on SSC 7. Subsequently, QTL analyses of several economically important traits were conducted.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
Experimental Animals
The Duroc pigs used in this experiment had been selected over 7 generations during 1995 to 2001 at the Miyagi Prefecture Animal Industry Experiment Station. Details of the selection method were described by Suzuki et al. (2005b). The selection traits were daily gain from 30 to 105 kg of BW (DG), LM area (LEA), and backfat thickness at 105 kg of BW measured using ultrasound technology and IMF content measured in slaughtered sib pigs. Gilts farrowed only once and boars were retained for use for one 4-to 6-wk breeding period. Thereby, a new generation was obtained each year. Pigs were weaned at 4 wk of age. Principally based on BW at 8 wk of age, 1 to 2 male piglets (total of 50 piglets) and 2 to 4 female piglets (total of 100 piglets) were selected from each litter at 8 wk as candidates for boars and gilts. In all, about 80 piglets, mainly comprising boars, with a few gilts when boars were not available from each litter, were selected from each litter for full-sib testing in each generation. This first stage of selection was conducted within litters. Boars used for full-sib tests were subsequently castrated. Performance tests began when BW of 30 kg was attained and ended when the pigs had reached a BW of 105 kg. Pigs were provided ad libitum access to a commercial diet (15% CP, 78% total digestible nutrients, and 0.76% lysine content on a DM basis) in testing periods, during which their BW increased from 30 to 105 kg. Pigs had free access to water. Boars were reared individually in performance-testing pens. Gilts and barrows were reared in growing pens in a concrete-floored building with 8 pigs per pen, which allowed 1.2 m2 of floor space for each pig. Animals were fed in groups.
In all, 1,004 pigs from the first to seventh generation were used for QTL analysis. This population comprised 1 family (in a complex multigenerational pedigree) such that all individuals were related. Table 1 shows the number of animals by generation. Although 110 sires and 351 dams were included in this family, only 99 sires and 286 dams were used to produce the next generation. Consequently, 286 full-sib families and 99 half-sib families were produced. Half-sib families produced an average of 9.6 offspring each.
|
All pigs were measured for 8 body scale traits when their BW had reached 105 kg. The measured traits were body length, the circumference of the chest and cannon bone, chest width, chest depth, withers height, and hip height. In addition, physiological traits such as serum IGF-1 were measured. Details of the procedure for measuring traits have been described by Suzuki et al. (2004).
Carcass and Meat Quality Measurement
Before being manually slaughtered using low-voltage (200 V) electrical stunning, pigs for full-sib tests (barrows and gilts) were kept for feed removal for 24 h but with free access to water. Processed dressed carcasses were placed in a refrigerator as soon as possible. Carcasses were placed in a conventional chiller at 4°C for 24 h. The carcass measurements included weight, lengths (I, II, and III), backfat thickness, and number of vertebra. Carcass lengths I, II, and III, respectively, denote the lengths from the first cervical bone to the pubic bone, from the first rib to the pubic bone, and from the first rib to last lumbar vertebra. For subsequent measurement of the meat quality in the LM, a 7- to 10-cm-long piece of the loin (2 thoracic vertebra sections above the last rib) was taken from the left half of each pig carcass. At that time, the meat color was assessed using the pork color standard (PCS; PCS2, 1 = light to 6 = dark; Nakai et al., 1975). Moreover, the loin meat color was assessed according to the fifth-sixth thoracic vertebra (PCS1). The chops were then moved to a laboratory for measurement of meat quality traits. External loin adipose tissue was removed, and the meat was cut vertically along the length of the loin. The sliced meat (about 50 g) was hung by a wire in a cylindrical plastic case of 15-cm diameter and 20-cm height. Drip loss was determined by weighing the sliced meat stored at 4°C in the refrigerator after 24 h and was calculated as a percentage of the original weight of the sliced meat. Lightness (L*), redness (a*), and yellowness (b*) were measured using a spectrophotometer (CM-2002; Minolta Co. Ltd., Osaka, Japan) after cutting and blooming for more than 15 min; pH was also measured. The remaining loin meat section was cut into 2 sections along the muscle fiber and was used to analyze cooking loss, tenderness, and pliability. Two pieces (2 x 2 x 5 cm) of meat were cut from each section, weighed, and packaged in polyethylene bags. They were vacuum-packaged and heated in a warm water bath at 70°C for 30 min. Then, after cooling to room temperature, the meat was dried and weighed again. Cooking loss was determined by measuring the driploss as a percentage of the original meat weight. Furthermore, 2 cooked pieces per animal were cut into pieces of 1 x 1 x 5 cm. Tenderness (kgf/cm2) was measured using a Tensipresser (TTP-50BXII; Taketomo Electric Corp., Tokyo, Japan) developed by Nakai et al. (1992). This machine was developed to evaluate meat tenderness accurately using an up-and-down motion to simulate chewing action. Two minced loin meat samples of about 20 g were analyzed using the Soxhlet method to determine IMF.
Genotyping of Markers
In all, 10 informative microsatellite markers on SSC 7 from the USDA Meat-Animal Research Center linkage map (Rohrer et al., 1996) were selected: SW2564, S0064, TNFB, SW1856, SW175, SW147, SW1083, SWR773, SW581, and SW2108. The DNA was extracted from ear tissue or blood, and the DNA concentration was adjusted to 20 ng/µL. Polymerase chain reaction primers for microsatellite markers were labeled with fluorescent dyes: 6-FAM, HEX, and TET (Perkin-Elmer, Foster City, CA). The PCR was carried out in a total volume of 15 µL containing 20 ng of genomic DNA, 6.25 pmol of each primer, 0.2 mM each deoxy-nucleoside triphosphate, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, and 0.375 U of recombinant Taq polymerase (Takara, Kyoto, Japan). Reaction mixtures were denatured at 94°C for 4 min, cycled 30 times (94°C for 30 s, 55°C for 30 s, and 72°C for 30 s), and incubated at 72°C for 5 min. The PCR product sizes were measured using an ABI 377 sequencer and analyzed using Genescan software and Genotyper software (Perkin-Elmer).
QTL Analysis
The identical by descent (IBD) for these markers was estimated for the multigenerational family using pedigree information from 8 generations. Data from all genotyped animals and CRI-MAP mapping software (Green et al., 1990) were used to estimate the distance between markers. Table 2 shows the map that was calculated using the Kosambi and Haldane map function. First, using the linkage map, we conducted a multi-point variance components analysis (Amos, 1994; Almasy and Blangero, 1998) to test for linkage between QTL and the phenotypic value, using a maximum-likelihood method implemented in the Sequential Oligogenic Linkage Analysis Routines package (SOLAR; Almasy and Blangero, 1998). To conduct the QTL analysis, it is necessary to correct the phenotypic value for any fixed nongenetic effects. The best linear unbiased estimator for generation and sex was estimated using the VCE4.25 program (Neumaier and Groeneveld, 1998). Then, QTL analysis was conducted using the multipoint variance component approach on sex- and generation-adjusted phenotypes.
|
); they were then imported into SOLAR (Almasy and Blangero, 1998). The LOKI program uses a Markov chain Monte Carlo approach to obtain IBD scores in complex pedigrees with missing marker data and unknown haplotypes. In this calculation, we did not evaluate the convergence of the Markov chain Monte Carlo sampler. Instead, we used 100,000 iterations for every position with the beginning time of output and output frequencies at iteration 10. All multipoint variance-component approaches were carried out according to the procedures outlined in SOLAR (Almasy and Blangero, 1998). Variance component analysis was performed at 1- to 5-cM intervals along every candidate region. For use in linkage analysis, the covariance matrix for a pedigree can be expressed as:
where 
= the covariance matrix for a given pedigree. Furthermore, in that equation, 
= a matrix with elements (
ijl) providing the expected proportion of alleles at the specific chromosomal location of the quantitative trait locus at which individuals j and l share IBD, which is estimated using genetic marker data; 
q2 = the additive genetic variance attributable to the major locus; 
= the kinship matrix; g2 = the variance attributable to residual additive genetic effects; I = an identity matrix; and e2 = the variance attributable to random environmental effects.
Using the variance component model, the null hypothesis that the additive genetic variance attributable to the ith QTL, q2, equals zero (no linkage) was tested by comparing the likelihood of this restricted model to that of a model in which the variance attributable to the ith QTL was estimated. The difference between the 2 log10 likelihoods produced a likelihood of odds (LOD) score that was the equivalent of the classical LOD score of linkage analysis. Twice the difference loge likelihoods of these 2 models yielded a test statistic that is distributed asymptotically as a 50:50 mixture of a 
variable and a point mass at zero (Self and Liang, 1987). Xu and Atchley (1995) found that the empirical distribution of the likelihood ratio test statistic follows a 
2 distribution with between 1 and 2 df when a chromosomal interval is being tested. Therefore, 2 distributions with 1 () and 2 (
) df were used to provide threshold values (Xu and Atchley, 1995). Significance tests were based on a likelihood ratio, which was obtained by multiplying the LOD score by 4.605.
Heritability Estimates
The variance component approach can include a polygenic random effect and can partition the total phenotypic variance into additive genetic variance attributable to the QTL, residual polygenic additive genetic variance, and the variance attributable to random environmental effects. Consequently, the QTL genotypic heritability, hq2 [= q2/(q2 + g2 + e2)], and residual polygenic heritability, hg2 [= g2/(q2 + g2 + e2)], were obtained.
|
RESULTS AND DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
Table 3 shows means and standard deviations for 33 traits of the Duroc purebred population. The significant QTL position and the heritability of the QTL and the polygene region at the 0.05 level with 1 df are presented in Table 4. Because results are presented for multipoint LOD score, a correction of statistical significance would be appropriate. Type I error was considered by Bonferroni correction for multiple testing as follows: as 10 different marker comparisons were done with each trait, the Bonferroni correction that gives an overall 
= 0.05 has a threshold for each comparison P < 0.05/10 = 0.005. The threshold of P = 0.005 of the 2 test with 1 df is an LOD score of 1.711 (7.879/4.605) and 2.302 (10.6/4.605) with 2 df. Significant QTL positions were detected for growth, body scale, meat quality, and carcass traits. A total of 9 QTL for 18 traits were found to be significant. Of these 9 QTL, 7 were significant at the 0.005 level with 2 df; 2 were also significant at the 0.005 level with 1 df. No significant QTL were detected in this population for IMF on SSC 7. Many QTL on SSC 7 that affect growth were reported with crossbred animals (Bidanel and Rothschild, 2002) and with purebred animals (De Koning et al., 2003; Nagamine et al., 2003).
|
|
. Edwards et al. (2008) also reported a QTL for carcass LM area on SSC 7 in a population of Duroc x Pietrain crossbreds, but their QTL was located in a more distant position (108 cM). A QTL for the circumference of the cannon bone was detected in this study. Sato et al. (2003) also reported a suggestive QTL for circumference of the cannon bone that had the greatest P-value. This QTL was mapped at 56.2 cM on SSC 7 near marker SW1856.
A QTL for PCS1, a meat color trait, was detected in this population. A moderate QTL heritability of 0.32 and a high LOD score were shown for PCS1. Pork color standard can be evaluated by comparing meat with a meat color mode devised by Nakai et al. (1975). It has a character that differs from optical L*, a*, and b* measurements. Moreover, PCS1 is measurement of the loin meat in front of the fifth-sixth thoracic vertebrae and has a larger phenotypic variance compared with that of PCS2 (Table 3
). Recently, Edwards et al. (2008) reported a QTL for L* and subjective color score ranging from 1 (pale pinkish gray) to 6 (dark purplish red) at the same region on SSC 7 as that reported for our study. On SSC 7, QTL of Minolta L* and Minolta a* were reported by Ovilo et al. (2002) in an F2 cross between divergent breeds; a QTL for Minolta b* was reported in a commercial population by Vidal et al. (2005). A QTL for Minolta b* was suggested in a position near that identified in a report by Vidal et al. (2005). These results collectively suggest that the QTL on SSC 7 that influence meat color are segregating within breeds.
A suggestive QTL for a serum concentration of IGF-I at 8 wk was detected in this study; the QTL heritability of this trait was 0.26. The IGF-I gene is a candidate gene because of the role of IGF-I in growth and body composition (Hossner et al., 1997); positive genetic correlations were estimated between plasma concentration of IGF-I at 8 wk and DG, LEA, and IMF (Suzuki et al., 2004). The possibility remains that QTL associated with IGF-I that relate to growth are segregating, because this population is selected for DG (Suzuki et al., 2005b). However, the IGF-I gene is located on SSC 5 (Rohrer et al., 1994). Perhaps what we see here is a locus interacting with the IGF-I locus.
Significant QTL related to carcass traits were detected. A high LOD score of 27.29 was obtained for the number of thoracic vertebrae, and traits related to this, such as carcass length and loin length, also had a high LOD score. Sato et al. (2003) and Mikawa et al. (2005) reported a significant QTL position on SSC 7 and SSC 1 for the number of thoracic vertebrae. Although they found that these QTL are located in the 99.4-cM position of SSC 7, we found them to be located in the 120-cM position. However, when this position was calculated using the Kosambi map function (Table 2), the same region (99.5 cM) as Sato et al. (2003) was shown. The large effect genes for thoracic vertebra are located on the same chromosome across breeds. Mikawa et al. (2005) used nine 3-generation families produced by crossing 2 breeds. Significant effects were detected around the same region on SSC 7 in 6 families, although no significant effects were detected in 3 families. As an explanation for not detecting the significant effects on SSC7 in 3 families, they considered that the allele that increases the number of vertebrae was not fixed in the European parental pigs in the experimental families. Using 1 sire with homozygous QTL as a grand sire in the experimental family, significant QTL were not detected. In this respect, because approximately 16 sires and 50 gilts per generation were used in this closed Duroc population, the high heterozygosity of this population was maintained and the possibility of detecting the QTL increased. In this Duroc population, the number of vertebrae is not fixed. The frequency of thoracic vertebra numbers of 14, 15, and 16 were 8, 68, and 23%, respectively, at first generation, and the values in the seventh generation have changed dramatically to 35, 60, and 5%. The phenotypic mean has also changed from 15.3 at the first generation to 14.7 by the seventh generation. These results confirm that QTL analysis of a multigenerational population is effective. A significant position can be detected using QTL analysis related to other chromosome regions for the trait examined in this study using a similar method.
Although Sato et al. (2003) in a Duroc x Meishan F2 population and Bidanel et al. (2002) in a Large White x Meishan F2 resource population reported a significant QTL on SSC 7 that influences IMF, no significant QTL position was detected in the present study. The population used in this investigation was a population in which IMF was part of a selection scheme. We expected that it would be easier to locate IMF on SSC 7 because of selection. Other QTL related to IMF have been reported in chromosomal regions (e.g., SSC 6) other than SSC 7 (Bidanel and Rothschild, 2002). It will be necessary to investigate other chromosomal regions, because the possibility exists that important QTL related to IMF have been separated from SSC 7.
Many QTL have been reported for fatness on SSC 7 using an F2 hybrid population (Bidanel and Rothschild, 2002). Moreover, the possibility that QTL that influence fatness are located on SSC 7 within breeds was inferred from studies of a commercial group and a purebred population (Nagamine et al., 2003; Vidal et al., 2005). However, no significant QTL region related to fatness was detected in the present research.
Importance of QTL Analysis Within a Closed Population
For this study, we performed an outbred QTL analysis using a closed Duroc population selected for 4 traits over 7 generations. Numerous studies have been performed to locate QTL in F2 crossbred populations, because the power of detection of segregating QTL using linecross data is greater than that using within-population data (Nagamine et al., 2004). However, QTL information detected in crossbred populations cannot necessarily be used for breeding of purebred populations, because the genetic backgrounds of these populations differ. Nagamine et al. (2003) reported that QTL can be detected in highly selected commercial populations of the second generation. Vidal et al. (2005) also reported that commercial purebred populations retain significant genetic variation, even for traits that have been selected over many generations. However, the present study is the first QTL analysis for a closed Duroc population selected over 7 generations. Consequently, some important QTL were detected. In Japan, several local animal experiment stations have performed closed linebreeding for 5 to 7 generations with Landrace, Large White, Berkshire, and Duroc breeds as described in the report by Suzuki et al. (2005b). In these experiments, selected traits were DG, backfat thickness, LEA, meat quality, and reproduction. In these cases, QTL analyses within breeds will be possible if DNA from these populations can be sampled from blood, with examination of DNA microsatellite polymorphisms. The possibility exists of detecting QTL that are directly useful for marker-assisted selection. This report describes QTL analysis limited to the seventh chromosome of the closed selection population. The results suggest that QTL analysis with a closed multigenerational purebred population is possible by variance component analysis using the SOLAR program. We are now performing QTL analysis of other chromosomes of this Duroc population.
1 Corresponding author: k1suzuki{at}bios.tohoku.ac.jp
Received for publication May 23, 2007. Accepted for publication June 6, 2008.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
