| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL GENETICS |
,2
* Laboratoire de Génétique Cellulaire, INRA, BP27, 31326 Castanet-Tolosan, France;
and
Station de Génétique Quantitative et Appliquée, INRA, 78352 Jouy-en-Josas, France;
and
ADN, 29190 Pleyben, France;
and
Domaine Expérimental de Bourges, 18390 Osmoy, France; and
and
# Domaine Expérimental du Magneraud, Saint-Pierre d’Amilly, 17700 Surgères, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Key Words: Backcross • Fatness • Growth • Pig • Quantitative Trait Loci • Swine Leukocyte Antigens
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
; Bidanel et al., 2001; de Koning et al., 2001), backfat thickness (Rohrer and Keele, 1998; Bidanel et al., 2001; Malek et al., 2001), carcass composition (Milan et al., 2002), and meat quality (Grindflek et al., 2001; Quintanilla et al., 2003). Most of these studies mapped the QTL close to the swine leukocyte antigens (SLA), with a most likely position in the SSC 7 p12–q12 region. In all studies, Meishan alleles at the SSC 7 QTL were associated with better performance and had dominant effects over European alleles.
Previous comparative mapping studies suggested that the 15-cM region limited by LRA1 and S0102 contains hundreds of genes (Genet et al., 2001; Demeure et al., 2003). Because many of these genes can be considered good candidates, decreasing the QTL location interval seems to be a necessity. In addition, the QTL effects evidenced on different traits in this interval could be associated with one pleiotropic gene or different co-localized genes. Genetic chromosome dissection through a backcrossing program has been shown as a successful approach to confirm QTL effects and improve mapping resolution (Marklund et al., 1999).
A backcross program was developed as described by Sanchez et al. (2005) to improve the resolution of QTL located on several chromosomal regions, particularly SSC 7. More specifically, on chromosome 7, the precise mapping of the QTL relative to the SLA complex is of particular interest because its effect on adaptability and natural selection. The use of a backcross program also allowed epistatic effects to be estimated in animals with a European genetic background. Finally, effects of the QTL on 44 additional meat quality, growth, and fatness traits were investigated.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
The recombinant animals were obtained with three different designs, using animals from experimental farms or breeding companies (Figure 1). All QTL in this study had favorable dominant effects of Meishan (MS) alleles over Large White (LW) alleles, making it interesting to test the effects of Meishan alleles in a European genetic background. In the first design (LW Family 1), backcross animals were obtained by crossing Large White x Meishan F1 boars used in the INRA QTL program (Bidanel et al., 2001) with Large White females located in two INRA experimental herds (Avord, Cher and Le Magneraud, Charente-Maritime). Fathers from the second and third designs came from commercial crosses of the ADN Breeding Co. (Pleyben, France). First-generation backcross pigs (BC1) were obtained by crossing either Large White x Meishan F1 males with Large White females (LW Family 2) or Landrace x Meishan F1 males with Landrace females (LR Family). For both LW Family 2 and LR Family, a second generation of backcross pigs (BC2) was produced by crossing BC1 boars with Large White sows from INRA herd.
|
). Due to the different Meishan regions conserved, these animals permitted testing the QTL location in the chromosomal regions represented in black on Figure 2. Only two different Meishan haplotypes (termed A and B) were segregating in the five tested animals. Both Meishan haplotypes were previously tested in the INRA F2 QTL program families and they showed similar effects on growth and fatness in contrast to Large White haplotypes. The progeny testing principle is detailed in Sanchez et al. (2005). Briefly, it is based on a comparison of piglet performance depending on which allele they received from the tested sire. A sire is considered to be heterozygous for the QTL if the paternal allele significantly affects the value of the trait in his progeny (P < 0.05) and homozygous for the QTL if the probability of such an effect is weak (not significant at P < 0.05).
|
A total of 44 traits (7 growth, 12 fatness, 17 carcass composition, and 8 meat quality traits) were measured (Table 1).
|
The microsatellites markers used were mostly public markers available on the USDA Web site (http://www.marc.usda.gov/genome/genome.html). Information about the DAXX, WAF, and NFY microsatellites are available in Demeure et al. (2003). Those three markers have been integrated in the genetic map by genotyping 245 animals from the INRA QTL program families. Mapping analyses were completed with the 2.4 version of CriMap software (Green et al., 1990). The amplifications were performed on ABI9700 PCR machines (Applied Biosystems, Foster City, CA). Thermal cycling parameters were denaturation at 95°C for 5 min, followed by 30 cycles of 95°C for 30 s; annealing temperature for 30 s; 72°C for 45 s; and a final amplification was performed at 72°C for 15 min. The PCR amplification analysis was done on ABI377 automatic sequencer (Applied Biosystems). Results were analyzed with ABI Genescan and Genotyper (Applied Biosystems) and scored and validated in GEMMA (Iannuccelli et al., 1996).
Statistical Analyses
The data were first adjusted for environmental effects using the GLM procedure of SAS (SAS Inst., Inc., Cary, NC). A contemporary group (i.e., a slaughter date effect for meat quality traits and a batch effect for the other traits) and sex (except for Minolta coordinates) were included in the model for all traits. Some traits were additionally corrected for a covariate: birth weight for ADG1; weaning weight for ADG2; weight at the beginning of the test for ADG3; and weight at the measurement for ultrasonic backfat thickness and body composition traits.
These adjusted data were then used to test within each sire family for the presence of a QTL in the chromosomal region investigated. Analyses were performed with the half full-sib model developed by Le Roy et al. (1998). The hypothesis of one QTL linked to the set of markers considered was compared with the hypothesis of no QTL at the same location. Under the one QTL hypothesis, a QTL with a gene substitution effect for each sire and for each dam was fitted to the data. Likelihoods were then maximized under each hypothesis and the test statistic was computed as the ratio of likelihoods. The location with the highest ratio of likelihoods was considered as the most likely position of the QTL. Approximate confidence intervals of QTL position were determined empirically using the "Drop-off" method. More details on the model and methodology can be found in Le Roy et al. (1998) and Bidanel et al. (2001). Chromosome-wide significance thresholds were determined assuming a polygenic infinitesimal model and a Normal distribution of performance (Le Roy et al., 1998). A total of 1,000 simulations were done for each sire x trait combination. Because many tests were carried out, two experiment-wide thresholds were defined in addition to those corresponding to the usual 5 and 1% chromosome-wide significance levels. As traits were correlated, an equivalent number of independent traits was estimated through a canonical decomposition of the correlation matrix. The first 20 eigenvalues explained more than 95% of the total variability, leading to an approximate number of 100 independent tests (20 traits x five sires). A Bonferroni correction was used to compute probabilities corresponding to 10 and 5% experiment-wide significance levels (i.e., 10-3 and 5.10-4). A total of 5, 1, 0.1, and 0.05 false positives were expected from 0.05, 0.01, 0.001, and 5 x 10–4 significance levels, respectively.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). The distances calculated under the Kosambi formula for Sw1856-DAXX, DAXX-WAF, WAF-NFY, and NFY-S0102 intervals are 0.8, 1, 3.7, and 4.1 cM, respectively.
The LW Family 1 BC1 (LW1-BC1) boar had an entire Meishan chromosome (with a haplotype called A) and a full LW chromosome (Figure 2
). This animal allowed QTL effects to be tested and estimated in a 7/8 Large White background. Under an additive model, QTL effects were expected to be roughly similar to those detected in F2 pigs as Meishan alleles were found to be dominant over LW alleles. Indeed, QTL effects were identified for several fatness and carcass composition traits; however in contrast to F2 results, no significant effects were found for growth and meat quality traits (Table 2).
|
). The next chromosomal region tested was the distal part of the QTL position confidence interval obtained in F2 pigs. As shown in Table 2, only two traits (ADG3 and BFT) showed chromosome-wide significant QTL effects, suggesting that this animal does not carry the MS allele of the QTL identified in its father. The LW Family 2 BC2 (LW2-BC2) boar carried the same MS haplotype (A) as LW Family 1 pigs. The associated QTL allele was thus assumed to be the same. Again, only suggestive QTL effects were found for a limited number of traits (LMR, G2, M6, and head weight), with QTL locations estimated around 50 cM. The lack of effect on growth and fatness traits observed in LW1-BC2 and LW2-BC2 progeny led us to decrease the QTL position to the Sw1856 to Sw1701 interval. However, as this interval was deduced from homozygous animals, we had to confirm it by testing animals heterozygous for QTL in this region.
The two Landrace backcross pigs that were progeny tested shared the B MS haplotype. The Landrace BC1 male had a MS haplotype from Sw1369 to WAF and a Landrace haplotype from NFY to Sw632. The progeny test revealed very important QTL effects on many growth, fatness, and carcass composition traits (Table 3; Figure 3), leading us to conclude that the boar was heterozygous for QTL with most likely positions around 65 cM. Suggestive QTL effects also were found for meat quality, with positive effects of MS allele on WHC (+3.8 s, P < 0.05) and longissimus muscle ultimate pH (+0.1 units; P < 0.05). The second Landrace backcross boar (LR-BC2) was the son of Landrace BC1. Its MS haplotype was decreased to a small region from Sw2019 to WAF. The LR-BC2 boar also seemed to be heterozygous for the main QTL effects (Table 3, Figure 3). These two progeny tests gave a new QTL interval decreased from TNFß to NFY (around 10 cM). Considering the LW2-BC2 progeny tests, the proximal limit might even be Sw1856, thus defining a 6-cM interval. Only further experiments on complementary animals could definitively exclude the TNFß - Sw1856 region.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
; Milan et al., 2002).
The first backcross (LW1-BC1) boar confirmed the existence of SSC 7 QTL effects on fatness and carcass composition traits in a Large White genetic background. Conversely, no significant QTL effects could be found for growth traits, which was surprising as the Meishan haplotype of this animal came from the F2 design (Bidanel et al., 2001). This result is likely to be due to the limited progeny size (only 29 offspring), which was large enough to detect the very strong effects of the chromosomal region on fatness traits, but not for the more moderate effects on growth rate. This hypothesis seems to be confirmed by progeny-test results from Landrace BC1 and BC2 boars, where larger sizes (118 and 89 offspring) allowed significant effects to be observed on at least some growth traits.
Few meat quality traits were found to be significantly affected by the chromosomal region investigated, most of them being only suggestive QTL (P < 0.05). In addition, none of the detected QTL was located in the previously defined interval, except the pH24 of the longissimus (in the Landrace BC1 boar progeny) and the biceps femoris L* (in the Landrace BC2 boar progeny) QTL, mapped at 65 and 64 cM, respectively. Previous studies found significant effects of this chromosomal region on meat quality, but in different crossbred populations: (Duroc x Landrace) x (Landrace x Yorkshire) pigs (Grinflek et al., 2001); and Meishan x Wild boar F2 pigs (Yue et al., 2003). Thus, it can be concluded that the chromosomal region investigated in this study has little or no effect on meat quality traits in crosses between Meishan and Large White or Landrace breeds.
The backcross experimental design also allowed testing of the hypothesis that several loci can affect traits as different as growth and fatness. Indeed, comparing the observed QTL effects in the progenies of different sires makes it possible to evaluate whether a recombination event has broken up a QTL haplotype resulting in markedly decreased QTL effects in some families. Results from the progeny of boar LW1-BC2 suggest that no additional major QTL segregates in the region distal to Sw1701, except for ADG3 and BFT, 60 cM away from the other QTL most likely positions. The effect on ADG might be due to the segregation in our crosses of a QTL affecting ADG similar to the one previously described in a Pietrain x Large White intercross around Sw352 (Nezer et al., 2002).
The SLA region was a particularly interesting target because, in humans, HLA heterozygosity confers a selective advantage against multiple-strain infections (Penn et al., 2002). Selecting QTL located in the SLA region might decrease its variability and result in decreased animal adaptability, with dramatic consequences for breeders. The progeny test of an LW2-BC2 boar revealed possible additional QTL in the SLA region or on the p arm of SSC 7 for a few carcass composition traits, but no QTL affecting growth or fatness traits has been detected. Wada et al. (2000) and Rohrer (2000) pointed out that it is likely that an additional region on the p arm should contain a QTL influencing also fatness traits. Results from the progeny of Landrace boars BC1 and BC2 showed similar QTL effects on fatness traits in both families (see Figure 3 for backfat thickness measured after slaughter), except for the QTL affecting fatness at 100 and 120 d of age, which was lost in boar LR-BC2. This could be due to the loss of a second QTL located in the TNFß-Sw2019 interval or to the progeny size difference, the QTL significance levels being globally lower for the LR-BC2. Conversely, results for carcass composition traits are different between the two boars, with a loss of the effect on LMC in the LR-BC2 progeny, whereas this same male has a new and strong QTL effect (P < 0.01) on carcass length. The loss of the QTL effect on LMC might be explained by an epistatic effect or by a second QTL proximal to Sw2019. The effect on carcass length observed in the LR-BC2 progeny indicates that the QTL involved in its control could be segregating in the LW and/or LR breeds. Globally, the large effect on fatness traits observed both in the Large White (where Meishan Haplotype A segregates) and Landrace families (where Meishan Haplotype B segregates) were similar, which confirms that both Chinese haplotypes also studied in the INRA F2 original cross have similar effects on fatness. However, we could not exclude the fact that the limited differences observed among the different families could be due to the segregation of additional QTL within European Large White and Landrace herds or limited difference among Meishan Haplotypes A and B.
In the analysis of the results of this backcross program, we should not formally exclude the risk that the favorable QTL allele of Meishan origin also could segregate within European Large White or Landrace herds. In that case, the observation of a contrast of effects between the two chromosomes of a tested sire will lead to a false conclusion about the location of the studied QTL. To limit the risk of a false conclusion on the localization of the QTL, we mainly based our deduction on the analysis of sires found to be heterozygous at the QTL and carrying part of a Meishan haplotype, the effect of which has already been evaluated in other families. Indeed, when a sire is identified as heterozygous, the shape of the likelihood ratio test curve, the estimate of the QTL effect, and their comparison with previous results allow increased confidence in the reliability of results. Considering both animals identified as heterozygous and homozygous at QTL, we decrease the QTL region to a 6-cM interval (Sw1856 to NFY), but if we only wish to consider animals identified as heterozygous at QTL, the QTL is located in a larger interval: TNFß to NFY. A recent multiple-QTL study, performed on the INRA QTL program, revealed that three QTL might be located in the Sw1856 to Sw1701 interval (Gilbert et al., 2004). The next generations of backcross animals will allow one to confirm the 6-cM interval, refine it, and test the multiple-loci determinism of growth and fatness traits suggested by the results of Gilbert et al. (2004). In parallel, recent high throughput and low costs for SNP genotyping methods should facilitate new approaches. In particular, fine-mapping strategies based on linkage disequilibrium will be evaluated. The first one is based on the study of allele frequency evolution in a population subjected to selection for growth and fatness traits. A second approach would be a progeny test using animals from different commercial or experimental programs, in order to find fragments that are identical by descent.
|
Implications |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
2 Correspondence—phone: +33-1-34-65-22-84; fax: +33-1-34-65-22-10; e-mail: bidanel{at}dga.jouy.inra.fr.
Received for publication March 30, 2005. Accepted for publication June 7, 2005.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Demars, J. Riquet, M.-P. Sanchez, Y. Billon, J.-F. Hocquette, B. Lebret, N. Iannuccelli, J.-P. Bidanel, D. Milan, and F. Gondret Metabolic and histochemical characteristics of fat and muscle tissues in homozygous or heterozygous pigs for the body composition QTL located on chromosome 7 Physiol Genomics, August 20, 2007; 30(3): 232 - 241. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Kim, M. Kong, Y. J. Nam, and C. Lee A Quantitative Trait Locus for Oleic Fatty Acid Content on Sus scrofa Chromosome 7 J. Hered., September 1, 2006; 97(5): 535 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-P. Sanchez, J. Riquet, N. Iannuccelli, J. Gogue, Y. Billon, O. Demeure, J.-C. Caritez, G. Burgaud, K. Feve, M. Bonnet, et al. Effects of quantitative trait loci on chromosomes 1, 2, 4, and 7 on growth, carcass, and meat quality traits in backcross Meishan x Large White pigs J Anim Sci, March 1, 2006; 84(3): 526 - 537. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
