HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sato, S. Articles by Kobayashi, E. Search for Related Content PubMed PubMed Citation Articles by Sato, S. Articles by Kobayashi, E.

Identification of quantitative trait loci affecting corpora lutea and number of teats in a Meishan x Duroc F₂ resource population

S. Sato^*,1,2, K. Atsuji^*, N. Saito, M. Okitsu, S. Sato^*, A. Komatsuda, T. Mitsuhashi^*,2, K. Nirasawa^*,3, T. Hayashi, Y. Sugimoto^# and E. Kobayashi^*

^* National Livestock Breeding Center, Nishigo, Fukushima 961-8511, Japan; and Ibaraki Branch of National Livestock Breeding Center, Fujigaya, Ibaraki 308-0112, Japan; and Miyazaki Branch of National Livestock Breeding Center, Kobayashi, Miyazaki 886-0004, Japan; and National Institute of Agrobiological Sciences, Tsukuba, Ibaraki 305-8602, Japan; and and ^# Shirakawa Institute of Animal Genetics, Nishigo, Fukushima 961-8061, Japan

	Abstract

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 
Understanding of the genetic control of female reproductive performance in pigs would offer the opportunity to utilize natural variation and improve selective breeding programs through marker-assisted selection. The Chinese Meishan is one of the most prolific pig breeds known, farrowing 3 to 5 more viable piglets per litter than Western breeds. This difference in prolificacy is attributed to the Meishan’s superior prenatal survival. Our study utilized a 3-generation resource population, in which the founder grandparental animals were purebred Meishan and Duroc pigs, in a genome scan for QTL. Grandparent, F₁, and F₂ animals were genotyped for 180 microsatellite markers. Reproductive traits, including number of corpora lutea (number of animals = 234), number of fetuses per animal (n = 226), number of teats (n = 801), and total number born (n = 288), were recorded for F₂ females. Genome-wide significance level thresholds of 1, 5, and 10% were calculated using a permutation approach. We identified 9 QTL for 3 traits at a 10% genome-wise significance level. Parametric interval mapping analysis indicated evidence of a 1% genome-wise significant QTL for corpora lutea on SSC 3. Nonparametric interval mapping for number of teats found 4 significant QTL on chromosomes SSC3 (P < 0.01), SSC7 (P < 0.01), SSC8 (P < 0.01), and SSC12 (P < 0.05). Partial imprinting of a QTL affecting teat number (P < 0.10) was detected on SSC8. Using the likelihood-ratio test for a categorical trait, 2 QTL for pin nipples were detected on SSC2 and SSC16 (P < 0.01). Fine mapping of the QTL regions will be required for their application to introgression programs and gene cloning.

Key Words: corpora lutea • imprinting • number of teats • pig • quantitative trait loci

	INTRODUCTION

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 
Reproductive performance in sows is regarded as one of the most economically important traits in production of pork. Litter size is a major trait for evaluation of reproductive capability and is affected by component traits, such as ovulation rate, fertilization rate, embryo survival, and uterine capacity (Christenson et al., 1987). The candidate gene approach and QTL analysis have been utilized often to understand the genetic components of such reproductive traits. Rothschild et al. (1996), using a candidate gene approach, were the first to report that the genotype of the estrogen receptor gene was associated with a significant difference in the litter size of pigs. On the other hand, a study that used a whole-genome scan to identify the QTL for ovulation rate in swine (Rathje et al., 1997) detected 4 regions with considerable evidence for QTL segregating on SSC 4, SSC8, SSC13, and SSC15. Rohrer et al. (1999) reported the mapping of QTL for 2 components of litter size (uterine capacity and ovulation rate) using a whole-genome scan: a likely QTL was detected on SSC8, 2 possible QTL for ovulation rate were detected on SSC3 and SSC10, and evidence suggesting the presence of a QTL affecting uterine capacity was found on SSC8.

Our aim was to gain a better understanding of the mechanisms determining litter size by detecting the genomic regions affecting some reproductive traits. Here we report the QTL results obtained from analyses of female reproductive traits such as number of corpora lutea (CL) and number of teats by using parametric and nonparametric interval mapping methods, assuming Mendelian and imprinted QTL models.

	MATERIALS AND METHODS

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 
All the procedures involving animals followed the guidelines for the care and use of laboratory animals established by National Livestock Breeding Center.

Phenotype Measurement

Construction of an F₂ resource population and measurement of phenotypes were performed at the Ibaraki and Miyazaki Branches of the National Livestock Breeding Center. The F₁ generation, consisting of 27 males and 25 females, was produced from crossing a Meishan sow and a Duroc boar, and then a total of 801 F₂ progeny (414 males and 387 females) were produced from 3 F₁ males and 18 F₁ females over 5 farrowings, in which the same F₁ sows were always mated to the same F₁ boars (Sato et al., 2003). Because all F₂ animals were produced in 5 independent groups in different parities and on 2 farms, possible phenotypic differences derived from groups, parities, and farms were removed by preadjusting the phenotypic observations using the appropriate mixed model least squares estimates provided by the maximum-likelihood computer program, as described by Harvey (1977).

The 5 reproductive traits measured are presented in Table 1. Number of teats was measured for 801 F₂ animals at 60 d. Extremely small and nonfunctional nipples were defined as pin nipples (Pn). Number of teats was recorded as the sum of the number of functional nipples and Pn. For Pn, each animal was classified into 1 of 2 classes, class 1 and class 2, corresponding to the absence and presence of Pn, respectively. The Pn were thus treated as a categorical trait. Of the 387 F₂ sows, 288 sows (with at least 1 randomly sow selected from each full-sib family) were mated to Large White boars to measure the total number born (TNB) at first parity. The F₂ sows were then mated again to Large White boars and were slaughtered at 4 wk of pregnancy. We measured the number of CL and number of fetuses (NF) per animal at this second parity. Nonpregnant F₂ sows were removed to measure of CL and NF, resulting in 234 and 226 F₂ sows measured for CL and NF, respectively.

We located 180 informative microsatellite markers at approximately 20-cM intervals in the whole pig genome. These markers were selected from those on the USDA-MARC linkage map (Rohrer et al., 1996). We genotyped the 180 markers for all animals in the resource population (Sato et al., 2003). Extraction of DNA and marker genotyping were carried out as described by Sato et al. (2003).

Linkage Analysis

Linkage maps were constructed by using CRI-MAP (Green et al., 1990) for the 18 autosomes and the method described by Sato et al. (2003) for the sex chromosome. A sex-averaged map was used for the whole-genome scan of QTL. The information content was calculated over the whole-genome region using the method described by Knott et al. (1998).

QTL Analyses

We attempted to detect QTL affecting number of teats, Pn, TNB, CL, and NF. Number of teats, TNB, CL, and NF were discrete traits, and Pn was recorded as a category. The phenotypic values of TNB, CL, and NF were scattered widely and symmetrically, ranging from 2 to 15, from 0 to 21, and from 0 to 19, respectively (Table 1), and were approximately normally distributed. Therefore, these 3 traits were analyzed by a method of interval mapping based on the least-squares method (Haley et al., 1994) by using the following linear model for the phenotypic value of a trait for the ith F₂ individual, y_i, taking imprinting effect into consideration, as described by Knott et al. (1998):

where µ is the overall mean, u_i, v_i, and w_i are coefficients for the additive effect, a; dominance effect, d; and imprinting effect, f, respectively, of the QTL at the tested position, and e_i is the residual error. Denoting the probability of an individual having the genotype AB (i.e., the paternally inherited allele is A and the maternally inherited allele is B) as P(AB), with the genotype of the Duroc boar being QQ and that of the Meishan sows being qq, we can calculate u_i, v_i, and w_i as u_i = P(QQ) – P(qq), v _i= P(Qq) + P(qQ), and w_i = P(Qq) – P(qQ), where it is assumed that the grandpa-rental breeds were fixed for alternative alleles at the QTL. The residual error was assumed to follow a normal distribution with a mean of 0.

Using the least-squares method, we obtained 3 F-ratios, F⁽¹⁾, F⁽²⁾, and F⁽³⁾, calculated from the residual sums of squares under the null model, M₀, assuming no QTL; under the full model, M₁, including an imprinting effect in addition to the additive and dominance effects of the QTL; and the Mendelian model, M₂, including only additive and dominance effects without an imprinting effect. We provided F⁽¹⁾, F⁽²⁾, and F⁽³⁾ as the F-ratios for 3 comparisons, M₀ vs. M₁, M₀ vs. M₂, and M₂ vs. M₁, respectively. If either or both F⁽¹⁾ and F⁽²⁾ were significant, a QTL significantly affecting a trait was considered to be detected at the tested position. For the detected QTL, we used F⁽³⁾ as a test statistic to see whether the QTL was imprinted or not. When the value of F⁽³⁾ was significant with a genome-wide significance level of 5%, the detected QTL was decided to be imprinted.

For number of teats, the values of which covered only a small range (from 11 to 19, as shown in Table 1), the linear model described above was inappropriate. Therefore, in our analyses of QTL for number of teats we utilized the nonparametric interval mapping (NPIM) method, taking imprinting effect into consideration, which is an extension of the method proposed by Kruglyak and Lander (1995), which included only additive and dominance effects. In the NPIM, we considered the correlation between the rank by phenotypic value and the QTL genotype of an F₂ individual. We defined Y_a, Y_d, and Y_f as

where rank(i) denotes the rank by phenotype of the ith F₂ individual. Denoting the variance of a quantity Y by Var(Y), we obtained

Under the null model, if a = d = f = 0, then Z_a, Z_d, and Z_f are independently normally distributed with a mean of 0 and a variance of 1. If either or both of 2 statistical values, X₁²= Z_a²+ Z_d²+ Z_f² and X₂² = Z_a²+ Z_d², were significant, we determined that a significant QTL was detected. Moreover, we investigated the value of Z_f² to see whether the imprinting effect, f, of the detected QTL was significant. The statistics used in the NPIM, X₁², X₂₂², and Z_f², were regarded as test statistics for comparison of models M₀ vs. M₁, M₀ vs. M₂, and M₂ vs. M₁, respectively.

For the analysis of Pn, which was recorded as a category, an interval mapping method for unordered categorical traits was applied, as described by Hayashi and Awata (2006). In this method, the probabilities of individuals being classified into each category, denoted by P(Class i), where i = 1, 2, were modeled with a logistic regression, where Log[P(Class 2)/P(Class 1)] was expressed as a linear model using the genotype of the QTL affecting Pn as a covariate. The log likelihood-ratio test (LRT) statistic was calculated from the maximum likelihood of the null model assuming no QTL and the alternative model assuming the presence of QTL affecting Pn. For the analysis of Pn, the imprinting effect was not incorporated. Moreover, we investigated the segregation ratio of alleles at significant QTL for Pn in Class 2 (the presence of Pn) to determine which alleles derived from Duroc or Meishan caused Pn.

In the analyses of all traits, sex-specific regions on the Y chromosome were assumed to include no QTL and were thus excluded from the analyses. The genome-wide significance thresholds of F-ratios, statistics in NPIM, and LRT were obtained by permutation test (Churchill and Doerge, 1994) of 1,000 repetitions (Table 2). In the detection of QTL, 5% genome-wide thresholds were adopted.

	RESULTS AND DISCUSSION

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

The number of animals measured for each analyzed trait is listed in Table 1, along with the means and standard deviations. Measurements were not available for all traits for all animals because of reproductive problems such as infertility.

Results of QTL Analyses

The significant QTL detected are summarized in Table 2. The results of the analyses for each trait are described below.

Total Number Born and Number of Fetuses per Animal. No significant QTL affecting TNB and NF were detected. Rohrer et al. (1999) detected several QTL for female reproductive traits, such as ovulation rate and uterine capacity, in a multigeneration Meishan–White composite population. However, they also could not detect QTL for litter size with a threshold set by a nominal significance level of P = 0.05. Cassady et al. (2001) reported that the strongest evidence for a number of stillborn QTL was on SSC13 (P < 0.05, n = 370). Subsequently, Holl et al. (2004) reanalyzed the same F₂ data and reported that Mendelian QTL on SSC13 (P < 0.05), SSC5 (P < 0.10), and SSC12 (P < 0.10) affecting number of stillborn were identified in single QTL models (n = 370). King et al. (2003) mapped a QTL with a nominal significance level (F-ratio = 4.79) for litter size close to marker SPP1 on SSC8q in 3 Meishan x Large White cross-populations (n = 152).

Number of Corpora Lutea. We mapped a significant QTL for CL in the interval between SW72 and SWR1637 on SSC3 (36.6 cM, F⁽¹⁾ = 8.11 and F⁽²⁾ = 12.17; Figure 1). This QTL, however, showed no significant imprinting effect (F⁽³⁾ = 0.90). The estimated additive and dominance effects were a = –1.27 (±0.305) and d = –1.18 (±0.445), which indicated that the Meishan allele was associated with an increase in CL for the QTL. Rohrer et al. (1999) reported a QTL for ovulation rate on SSC8 and in the same region on SSC3, as a result of genomic scanning in a population derived from a cross between Meishan and a White composite. Rathje et al. (1997), Wilkie et al. (1999), and Braunschweig et al. (2001) reported that putative ovulation rate or CL QTL were observed on SSC8 at 105, 101, and 99 cM. Cassady et al. (2001) and Holl et al. (2004) observed an ovulation rate QTL on SSC9 at 1 cM. In a previous study, we mapped a significant QTL for testicular weight near SWR1637 on SSC3 in the same population (Sato et al., 2003). These results together indicate that a region around SWR1637 is important for reproductive function, i.e., spermatogenesis in the male and follicular maturation in the female.

View larger version (27K): [in this window] [in a new window]   Figure 1. Plot of the F-ratio from multilocus least squares analysis (Haley et al., 1994). The x-axis indicates the relative position in the linkage map. The left y-axis and right y-axis represent the F-ratio from the Mendelian model and information content, respectively. A triangle on the x-axis indicates a marker position. Horizontal lines indicate threshold values for genome-wise 5% level (thin line) and genome-wise 1% level (thick line). The curved line indicates information content, and • = CL.

View larger version (29K): [in this window] [in a new window]   Figure 2. Plot of the X12 from nonparametric QTL mapping with Mendelian model (Kruglyak and Lander, 1995). The x-axis indicates the relative position in the linkage map. The left y-axis and right y-axis represent the 2 statistic (X12) from the Mendelian model and information content, respectively. A triangle on the x-axis indicates a marker position. Horizontal lines indicate threshold values for genome-wise 5% level (thin line) and genome-wise 1% level (thick line). The curved line indicates information content, and + = number of teats.

View larger version (29K): [in this window] [in a new window]   Figure 3. Plot of the X12 and Zf2 from nonparametric QTL mapping with the Mendelian and imprinting models, respectively (Kruglyak and Lander, 1995). The x-axis indicates the relative position in the linkage map. The left y-axis and right y-axis represent the 2 statistics (X12 and Zf2) and information content, respectively. A triangle on the x-axis indicates a marker position. Horizontal lines indicate threshold values for genome-wise 5% level (thin line), genome-wise 1% level (thick line). The dash line indicates threshold values for genome-wise 5% level with the imprinting model. The curved line indicates information content, + = number of teats with the Mendelian model, and = number of teats with the imprinting model.

Using LRT, we detected 3 QTL affecting Pn on SSC2, 3, and 16. On the markers adjacent to the highly significant QTL (P < 0.01), SW1879 and SW2192 on SSC2 and SW403 and SWR2086 on SSC16, the ² test indicated that segregation at QTL in all F₂ animals followed the expected ratio (1:2:1) but that the segregation in Pn animals was significantly different from the expected ratio with the increased frequencies of Duroc alleles (Table 3). This result might indicate that a Duroc founder had a causative gene for Pn, although it is difficult to come to a clear conclusion because of small number of Pn animals (30 in 801 animals). Identification of the causative gene for Pn could be of practical use in pig breeding programs for excluding genetic diseases, as has been done in the case of mutations in the ryanodine receptor 1 (RYR1) gene (Fujii et al., 1991).

	IMPLICATIONS

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 
Our findings provide basic information on the transmission of some porcine quantitative traits. Clarification of the genes responsible for reproductive traits requires the use of algorithms to analyze gene interactions because these traits are multifactorial. Several of the quantitative trait loci identified in this study could be promising targets for marker-assisted selection.

	Footnotes

 
² Present address: National Institute of Agrobiological Sciences, Tsukuba, Ibaraki 305-8602, Japan.

³ Present address: National Institute of Livestock and Grassland Sciences, Tsukuba, Ibaraki 305-0901, Japan.

¹ Corresponding author: sshuji{at}affrc.go.jp

Received for publication March 24, 2006. Accepted for publication July 6, 2006.

	LITERATURE CITED

Top Abstract INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION IMPLICATIONS LITERATURE CITED

 


Braunschweig, M. H., A. A. Paszek, J. I. Weller, Y. Da, R. J. Hawken, M. B. Wheeler, L. B. Schook, and L. J. Alexander. 2001. Generation and exploration of a dense genetic map in a region of a QTL affecting corpora lutea in a Meishan x Yorkshire cross. Mamm. Genome 12:719–723.[CrossRef][Medline]

Cassady, J. P., R. K. Johnson, D. Pomp, G. A. Rohrer, L. D. Van Vleck, E. K. Spiegel, and K. M. Gilson. 2001. Identification of quantitative trait loci affecting reproduction in pigs. J. Anim. Sci. 79:623–633.[Abstract/Free Full Text]

Christenson, R. K., K. A. Leymaster, and L. D. Young. 1987. Justification of unilateral hysterectomy-ovariectomy as a model to evaluate uterine capacity in swine. J. Anim. Sci. 65:738–744.[Abstract/Free Full Text]

Churchill, G. A., and R. W. Doerge. 1994. Empirical threshold values for quantitative trait mapping. Genetics 138:963–971.[Abstract]

Fujii, J., K. Otsu, F. Zorzato, S. de Leon, V. K. Khanna, J. E. Weiler, P. J. O’Brien, and D. H. Maclennan. 1991. Identification of a mutation in porcine ryanodine receptor associated with malignant hyperthermia. Science 253:448–451.[Abstract/Free Full Text]

Green, P., K. Falls, and S. Crooks. 1990. Documentation for CRI-MAP, Version 2.4. Washington University School of Medicine, St. Louis, MO.

Haley, C. S., S. A. Knott, and J. M. Elsen. 1994. Mapping quantitative trait loci in crosses between outbred lines using least squares. Genetics 136:1195–1207.[Abstract]

Harvey, W. R. 1977. User’s guide for LSML 76, mixed model least-squares and maximum likelihood computer program. Mimeo. The Ohio State University Press, Columbus, OH.

Hayashi, T., and T. Awata. 2006. Interval mapping for loci affecting unordered categorical traits. Heredity 96:185–194.[CrossRef][Medline]

Holl, J. W., J. P. Cassady, D. Pomp, and R. K. Johnson. 2004. A genome scan for quantitative trait loci and imprinted regions affecting reproduction in pigs. J. Anim. Sci. 82:3421–3429.[Abstract/Free Full Text]

King, A. H., Z. Jiang, J. P. Gibson, C. S. Haley, and A. L. Archibald. 2003. Mapping quantitative trait loci affecting female reproductive traits on porcine chromosome 8. Biol. Reprod. 68:2172–2179.[Abstract/Free Full Text]

Knott, S. A., L. Marklund, C. S. Haley, K. Andersson, W. Davies, H. Ellegren, M. Fredholm, I. Hansson, B. Hoyheim, K. Lundström, M. Moller, and L. Andersson. 1998. Multiple marker mapping of quantitative trait loci in a cross between outbred wild boar and large white pigs. Genetics 149:1069–1080.[Abstract/Free Full Text]

Kruglyak, L., and E. S. Lander. 1995. A nonparametric approach for mapping quantitative trait loci. Genetics 139:1421–1428.[Abstract]

Rathje, T. A., G. A. Rohrer, and R. K. Johnson. 1997. Evidence for quantitative trait loci affecting ovulation rate in pigs. J. Anim. Sci. 75:1486–1494.[Abstract/Free Full Text]

Rohrer, G. A., L. J. Alexander, Z. Hu, T. P. L. Smith, J. W. Keele, and C. W. Beattie. 1996. A comprehensive map of the porcine genome. Genome Res. 6:371–391.[Abstract/Free Full Text]

Rohrer, G. A., J. J. Ford, T. H. Wise, J. L. Vallet, and R. K. Christenson. 1999. Identification of quantitative trait loci affecting female reproductive traits in a multigeneration Meishan-White Composite swine population. J. Anim. Sci. 77:1385–1391.[Abstract/Free Full Text]

Rothschild, M., C. Jacobson, D. Vaske, C. Tuggle, L. Wang, T. Short, G. Eckardt, S. Sasaki, A. Vincent, D. McLaren, O. Southwood, H. van der Steen, A. Mileham, and G. Plastow. 1996. The estrogen receptor locus is associated with a major gene influencing litter size in pigs. Proc. Natl. Acad. Sci. USA 93:201–205.[Abstract/Free Full Text]

Sato, S., Y. Oyamada, K. Atsuji, T. Nade, S. Sato, E. Kobayashi, T. Mitsuhashi, K. Nirasawa, A. Komatsuda, Y. Saito, S. Terai, T. Hayashi, and Y. Sugimoto. 2003. Quantitative trait loci analysis for growth and carcass traits in a Meishan x Duroc F₂ resource population. J. Anim. Sci. 81:2938–2949.[Abstract/Free Full Text]

Wilkie, P. J., A. A. Paszek, C. W. Beattie, L. J. Alexander, M. B. Wheeler, and L. B. Schook. 1999. A genomic scan of porcine reproductive traits reveals possible quantitative trait loci (QTLs) for number of corpora lutea. Mamm. Genome 10:573–578.[CrossRef][Medline]

This article has been cited by other articles:

				S. Tetzlaff, E. Murani, K. Schellander, S. Ponsuksili, and K. Wimmers Differential expression of growth factors and their receptors indicates their involvement in the inverted teat defect in pigs J Anim Sci, November 1, 2009; 87(11): 3451 - 3457. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sato, S. Articles by Kobayashi, E. Search for Related Content PubMed PubMed Citation Articles by Sato, S. Articles by Kobayashi, E.