|
|
||||||||
ANIMAL GENETICS |



* Departament de Ciència Animal i dels Aliments, Facultat de Veterinària, Universitat Autònoma de Barcelona, 08193, Bellaterra, Spain;
and
Àrea de Producció Animal, Centre UdL-IRTA, C/Alcalde Rovira Roure 177, 25198, Lleida, Spain;
and
Institut Català de Recerca i Estudis Avançats, C/Lluis Companys 23, 08010, Barcelona, Spain; and
and
Dpto. Mejora Genética Animal SGIT-INIA, Ctra. De la Coruña Km. 7, 28040, Madrid, Spain
| Abstract |
|---|
|
|
|---|
Key Words: Fatness Growth IGF2 Pig Quantitative Trait Loci
| Introduction |
|---|
|
|
|---|
The causal mutations of the majority of QTL in domestic animals are thus far unknown (Andersson and Georges, 2004
). One of the exceptions is the QTL localized in the SSC 2 IGF2 region. A paternally expressed QTL affecting muscle growth, fat deposition, and heart size was localized in the IGF2 region (Jeon et al., 1999
; Nezer et al., 1999
) and was confirmed by de Koning et al. (2000)
and Thomsen et al. (2004)
. Recently, Van Laere et al. (2003)
reported that the IGF2-intron3-G3072A substitution, localized in a highly conserved regulatory region that binds a repressor nuclear factor, is the causal mutation of that QTL. Jungerius et al. (2004)
reported that the IGF2 substitution explains a backfat thickness QTL.
Evans et al. (2003)
and de Koning et al. (2003)
analyzed the Large White commercial population for the presence of QTL, but inconsistent results for fatness traits were obtained in the IGF2 region. Previous results in the Iberian x Landrace intercross reported a QTL in SSC 2, but not within the IGF2 region (Varona et al., 2002
).
The objectives of the present work were to test the IGF2 substitution effect in the Large White commercial population and in the Iberian x Landrace F2 intercross and to determine whether additional QTL are segregating on SSC 2 in the F2 population.
| Materials and Methods |
|---|
|
|
|---|
Both the Large White and F2 cross, together with the phenotypes recorded, have been described extensively elsewhere (Varona et al., 2002
; de Koning et al., 2003
; Evans et al., 2003
). In brief, the Large White population (COPAGA Cooperative, Lleida, Spain) consisted of five sires, 54 females, and 387 offspring; the F2 cross pedigree was derived from 34 parents (three Iberian boars and 31 Landrace sows), 79 F1, and 321 F2 animals (NOVA GENÈTICA, S. A., Lleida, Spain). Animals were maintained under intensive conditions, and feeding was ad libitum with a cereal-based commercial diet (13.4 MJ of ME /kg, 17.5% CP, 1% lysine; as-fed basis). The objectives of the two experiments were quite different: in the Large White population, the purpose was to confirm, in a commercial population, the segregation of QTL previously described in experimental crosses, whereas the F2 cross was designed to perform a genome scan for the detection of QTL for growth, fatness, and meat quality traits.
The traits recorded in the Large White and F2 populations were subcutaneous backfat thickness at 6 cm off the midline between the third and fourth ribs using the SFK Fat-o-Meater (SFK Technology A/S, Herlev, Denmark) probe (G2), backfat thickness measured with a ruler at the last rib (BF2), carcass weight (CW), LM area (LMA), ham weight (HW), shoulder weight (SW), and pH at 24 h postmortem in the semimembranosus muscle (pH24h). Measurements of LM were available only in the Iberian x Landrace F2 intercross.
Genotyping
The genotyping of the IGF2 substitution was done as described by Van Laere et al. (2003)
in a PSQ HS 96 system (Pyrosequencing AB, Uppsala, Sweden). Once the sires genotypes were obtained, the offspring from heterozygous boars and homozygous sows were genotyped so that the allele transmitted could be determined unambiguously. The offspring of AA and GG sires were assigned the A or G haplotypes, respectively. Ten micro-satellites and the IGF2 substitution (marker IGF2) were used in the F2 population. The positions (in cM) of the markers obtained with CRIMAP 2.4 (Green et al., 1990
) were IGF2 (0.0) SWC9 (0.1) S0141 (30.1) SW240 (42.2) SW1201 (49.2) SW395 (68.1) S0226 (75.3) SW1517 (78.1) SW1695 (83.7) S0378 (96.3) SWR308 (139.1). Three additional microsatellites (SW1201, SW1517, and SW1695) used by Varona et al. (2002)
were genotyped. Microsatellite PCR reactions were carried out in an automatic PCR ABI PRISM 877 integrated thermal cycler (Perkin Elmer, Foster City, CA) and analyzed with fluorescent detection in an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA). The genotypes were determined using the GeneScan 3.7 analysis software (Applied Biosystems) and stored in the Gemma database (Iannuccelli et al., 1996
).
Statistical Analyses
The general model used to test the IGF2 substitution effect in both populations and to perform a QTL scan on SSC 2 in the F2 cross was
![]() | [1] |
where yi is the individual phenotype i; batchi is the date of slaughter; ß is the covariate coefficient of ci (CW, except for CW itself, which was corrected by age at slaughter); Pai is the probability of the individual being homozygous for alleles of Iberian origin minus the probability of being homozygous for alleles of Landrace origin at the position of interest; Pdi is the probability of being heterozygous; aQTL and dQTL are the QTL additive and dominant effects, respectively;
i is an indicator variable with values
and
, depending on whether allele G or A has been received from the sire at the IGF2 substitution; aIGF2 is the IGF2 substitution effect; ui is the infinitesimal genetic effect (treated as random with covariance matrix A
2u; A is the numerator relationship matrix); and ei is the residual.
The SSC 2 region genotyped in the Large White population was restricted to the IGF2 region, whereas a chromosome-wide scan was carried out in the F2 population. Thus, for the Large White population, the QTL effects aQTL and dQTL were not fitted, and only the effect of the IGF2 substitution was tested using a likelihood ratio test. This reduced model was also applied to the F2 data for the sake of symmetry. In contrast, the full model [1] was fitted every cM along SSC 2 in the F2 data. Three tests were performed; the IGF2, the QTL, or both QTL and IGF2 effects were tested by removing the appropriate effect(s) from the model. The dominant QTL effect (dQTL) was excluded from the model when it was not significant (P > 0.05).
In all cases, nominal P values were obtained using the
2 approximation to the distribution of the log-likelihood ratios. Maximum likelihood and parameter estimates were obtained via the Qxpak package (Pérez-Enciso and Misztal, 2004
). It should be noted that maximum likelihood estimates of random effects are slightly biased with respect to REML, but they allowed us to test fixed effects such as the IGF2 substitution. The Qxpak first estimates the Pa and Pd coefficients via a MCMC algorithm and, in a second step, uses these values to estimate the rest of the parameters. Regarding significance of P-values, permutation tests cannot be applied to the model used here because the infinitesimal effects estimates are meaningless when phenotypes are permuted between related individuals. We have shown previously (Mercadé et al., 2005
) that a 1% chromosome-wise significant P-value corresponds roughly to a nominal P-value of 0.001. Although we cannot claim that this is a completely satisfactory solution, here we considered that nominal P-values < 0.001 were significant for the QTL test.
| Results |
|---|
|
|
|---|
Table 1
shows the results of testing the IGF2 mutation with model [1] without the QTL effects for the sake of comparison between Large White and F2 populations. The IGF2 substitution effect was significant (P < 0.05) in the Large White population for all traits analyzed, except pH. In contrast, the IGF2 effect was significant only for G2, LMA, and HW in the F2. Regarding the allele effects, the A allele decreased G2 and BF2 and increased CW, LMA, HW, and SW.
|
|
|
| Discussion |
|---|
|
|
|---|
In previous analyses of the F2 cross, we had identified a significant QTL for LMA and LM depth and suggestive QTL for growth and fatness in SSC 2 (Varona et al., 2002
). To characterize the contribution of the IGF2 substitution further, as well as other loci to the phenotypic differences in the intercross population, we performed a series of QTL analyses for the same traits used in the IGF2 substitution effect validation and retested the IGF2 substitution effect with a QTL in the model. The results agree with those obtained without a QTL effect in the model (Table 1
), so we can conclude that, in the F2 population, the IGF2 substitution has significant effects on G2, LMA, and HW (P < 0.01). The lack of significance of the IGF2 substitution effect in the other traits, which were significant in the Large White population, could be due to the small size of the F2 family where the substitution is segregating. We also report additional QTL segregating in this chromosome in the Iberian x Landrace F2 cross. Taken together, these two facts show that there exists at least one more locus, in addition to the IGF2 substitution, affecting the analyzed traits. The position of the CW QTL is close to that reported for ADG (Bidanel et al., 2001
; Malek et al., 2001b
). Bidanel et al. (2001)
, in a cross between Meishan and Large White breeds, showed that the Meishan alleles in SSC 2 were associated with a higher growth rate; in our analyses, the Iberian alleles increased the CW. Thus, the effect is in the opposite direction of phenotype differences in both Meishan and Iberian breeds (differences between Iberian and Landrace breeds are described in Serra et al., 1998
). The QTL for LMA detected in the same population by Varona et al. (2002)
has been confirmed, and the genotyping of three additional markers allowed the refinement of its localization, which mapped very close to the newly genotyped SW1517 microsatellite. As expected, given the phenotypic differences between breeds, the Landrace alleles increase LMA. As in the case of the SSC 4 QTL (Mercadé et al., 2005
), the Iberian alleles of the HW and SW QTL in SSC 2 decreased HW and SW, which agrees with the phenotypic differences between breeds. The HW and SW are corrected for CW so, as argued by Mercadé et al. (2005)
, they can be considered measurements of the shape of the animals because they measure how total weight is distributed along the different body regions. Malek et al. (2001a)
found QTL for water-holding capacity and color traits near the pH24h QTL described here. Because both traits are correlated with pork ultimate pH (Malek et al., 2001a
), it can be hypothesized that both QTL may correspond to the same gene. Su et al. (2004)
described a suggestive pH QTL in a cross between Large White and Meishan in a position similar to that described in our study. Evans et al. (2003)
and de Koning et al. (2003)
also described a QTL for pH in SSC 2, but it was within the IGF2 region. In a previous genome scan for meat quality traits using the same Iberian x Landrace F2 cross, no QTL for pH was detected within SSC 2, although SSC 3 showed one (Ovilo et al., 2002
). In the report of Ovilo et al. (2002)
, there were fewer markers in SSC 2. In particular, two additional markers have been genotyped in the pH24h QTL region. Thus, the appearance of this new QTL could be due to an increased marker density in its region. The breed allelic effects on pH24h are as expected regarding the differences between the two breeds. Finally, Varona et al. (2002)
detected a suggestive QTL for fatness in SSC 2 that corresponds with the QTL profile for BF2 presented in Figure 1
. Here, the BF2 QTL still remains only suggestive; so, as in Jungerius et al. (2004)
, additional fatness QTL segregating in this chromosome can be neither confirmed nor excluded.
In conclusion, we have shown that the IGF2 substitution has a large effect in the Large White population and in the F2 cross, despite the small size of the F2 family. Our results confirm the intron-3 G3072A substitution as the causal factor of the QTL detected in the IGF2 region with the estimation of the effects in the expected direction. This study has been performed in populations unrelated to those of the original report (Van Laere et al., 2003
) and in the Dutch report (Jungerius et al., 2004
), conferring additional validity to the relevance of this mutation. An intriguing question is why such a large effect mutation is segregating at intermediate frequencies in highly selected lines such as Landrace or Large White, even more when the causative mutation was evident at high allele frequency in breeds strongly selected for lean growth (Van Laere et al., 2003
), and both the IGF2 substitution and the SWC9 microsatellite (within the IGF2 gene) are in Hardy-Weinberg equilibrium in the Large White parental population (data not shown). A possibility is that it has pleiotropic unwanted effects in other traits or that there are counterbalanced QTL with opposite effects in the same chromosome, diminishing the effective selection pressure. The results concerning BF2 in the F2 cross support this tentative hypothesis (Table 2
). Another possibility is that the introduction of the A allele had been recent, and thus a selective sweep is being produced at the current time. This is the hypothesis suggested by Van Laere et al. (2003)
. In other breeds, such as Pietrain, we have observed that the mutant A allele was fixed, which is in agreement with the IGF2 region results of the Spanish Pietrain population in Evans et al. (2003)
and de Koning et al. (2003)
, perhaps as a result of higher selective pressure for conformation traits and smaller effective sizes than either Large White or Landrace. Thus, the situation may be different from breed to breed. Finally, in addition to confirming the effect of the IGF2 substitution, we also have found that there are other loci segregating in SSC 2 of the Iberian x Landrace cross (at least for traits CW, LMA, and pH24h), showing that this chromosome still has valuable mutations to be discovered.
| Implications |
|---|
|
|
|---|
| Footnotes |
|---|
2 Correspondencephone: 34 93 581 4260; fax: 34 93 581 2106; e-mail: Jordi.Estelle{at}uab.es.
Received for publication March 18, 2005. Accepted for publication August 11, 2005.
| Literature Cited |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Estelle, F. Gil, J. M. Vazquez, R. Latorre, G. Ramirez, M. C. Barragan, J. M. Folch, J. L. Noguera, M. A. Toro, and M. Perez-Enciso A quantitative trait locus genome scan for porcine muscle fiber traits reveals overdominance and epistasis J Anim Sci, December 1, 2008; 86(12): 3290 - 3299. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Ojeda, L.-S. Huang, J. Ren, A. Angiolillo, I.-C. Cho, H. Soto, C. Lemus-Flores, S. M. Makuza, J. M. Folch, and M. Perez-Enciso Selection in the Making: A Worldwide Survey of Haplotypic Diversity Around a Causative Mutation in Porcine IGF2 Genetics, March 1, 2008; 178(3): 1639 - 1652. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. W. Zhang, H. C. Cheng, C. D. Jiang, C. Y. Deng, Y. Z. Xiong, F. E. Li, and M. G. Lei Imprinted status of pleomorphic adenoma gene-like I and paternal expression gene 10 genes in pigs J Anim Sci, April 1, 2007; 85(4): 886 - 890. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |