B. J. Isler*,
K. M. Irvin1*,
S. M. Neal‡,
S. J. Moeller*,
M. E. Davis*,
and D. L. Meeker*
*The Ohio State University Department of Animal Sciences
‡The Ohio State University Agricultural Technical Institute
1For more information, contact at: The Ohio State University,
110F Animal Science Building, 2029 Fyffe Road,
Columbus, OH 43210; 614-292-6407; 614-292-2929 fax;
e-mail : irvin.3@osu.edu.
Previous studies have shown that a specific allele (B allele) of the estrogen receptor (ESR) locus is associated with increased litter size in swine. At this time, research is lacking in the examination of the association between ESR genotype and the reproductive system itself. The objective of this current study was to investigate the association between ESR genotype and reproductive components in swine. The ESR genotype of 322 Yorkshire (Y¥Y), Large White (LW¥LW), and crossbred (LWxY, YxLW) animals was determined to be either AA, AB, or BB using a PCR-RFLP procedure. Of this group, 107 females were selected and mated to Hampshire boars. At approximately day 75 of gestation, the females were slaughtered and their reproductive tracts collected. Data collected included ovulation rate, horn length, number of fetuses, fetal mass, uterine mass, number of mummies, fetal sex, fetal placement, fetal survival, and fetal space. Data were analyzed using a model that included ESR genotype, breed, parity, and significant two-way interactions. Uterine horn was also included in some analyses. ESR genotype was not found to be a significant (P > 0.05) effect for any of the traits studied. Some traits displayed favorable, but not statistically significant, trends with respect to ESR genotype - fetal survival, total uterine length, total fetal weight, total number of mummies, fetuses per horn, horn length, and fetal space. The ESR gene, therefore, appears to be positively associated with several reproductive traits. Parity and breed also affected some reproductive traits. Animals of parity => 3 had both a significantly (P < 0.05) larger ovulation rate per horn (+1.78 ova) and a lower fetal space per horn (-16.01 cm) than animals of parity 1. Also, animals with a Large White dam had an increased number of fetuses per horn, increased fetal weight per horn, and a decreased fetal space per horn.
For many years, scientists and producers alike have made tremendous improvements in our common livestock species using traditional methods of genetic selection. New discoveries in the field of molecular genetics now allow for the isolation and study of specific regions of the genome that influence important traits. Animals that contain these "marker" regions can then be selected for inclusion in a marker-assisted selection program. This approach has shown special promise for those traits that are of low heritability and act in a sex-limited manner, such as the reproductive traits. Due to the large part reproductive traits play in determining the efficiency of production in livestock species, a great deal of research has focused on the search for genes that influence these traits. An especially promising group of genes that has been investigated are those genes that are associated with the steroid hormones. The role of estrogen and the estrogen receptor in reproduction has been especially well studied. It has been shown that mutations in the estrogen receptor gene (ESR) can produce considerable phenotypic changes in the mammalian reproductive system, including cancer (Lehrer et al., 1990) and infertility (Korach, 1994). Based on observations such as these, it was hypothesized that the ESR gene could influence reproductive traits in swine.
Initial studies of the ESR gene in swine utilized animals of the Chinese Meishan breed. The Meishan breed is historically known for its large litter sizes (Haley et al., 1992). Studies using Meishan and Meishan crosses discovered variation at the ESR locus in these swine (Rothschild et al., 1991). Subsequent studies have found an association between a favorable ESR allele and reproductive traits in several breeds of swine. This advantageous allele has a positive additive effect on total number born and number born alive in swine. The effect of this allele has been shown to range from 1.25 pigs/litter in Meishan crosses to 0.4 to 0.6 pigs/litter in Large White and Large White crosses (Rothschild et al., 1994; Short et al., 1997). Researchers have also tried to find an association between this locus and other traits in swine, such as backfat depth and teat number (Rothschild et al., 1994; Short et al., 1997). One area that has not been studied, however, is the association between the ESR gene and reproductive tract traits. If the ESR gene influences traits such as litter size, it should follow that this gene also influences the reproductive system itself. The purpose of this study is to determine the effect of the ESR gene on several of these reproductive tract traits in swine.
Three hundred twenty-two purebred Yorkshire (Y x Y), purebred Large White (LW x LW), and crossbred (LW x Y, Y x LW) animals were selected for use in this study. All animals were raised at the Western Branch of The Ohio State University's Ohio Agricultural Research and Development Center (South Charleston, Ohio). Animals consisted of related and unrelated animals of both sexes and varying ages.
For each animal, DNA was extracted from lymphocytes and the ESR gene amplified using a polymerase chain reaction protocol. This protocol has been outlined previously (Short et al., 1997). Amplified products were digested with PvuII restriction endonuclease, separated on a 4% agarose gel, and visualized under UV light after ethidium bromide staining. Two ESR alleles (A and B) were identified, and each animal was classified as either AA, AB, or BB with respect to ESR genotype.
Of the original 322 animals genotyped, 107 females were selected for reproductive tract analysis. Females selected were of all four breed combinations and varying parities. All females were bred to Hampshire boars. All females were slaughtered at approximately 75 days of gestation in a commercial slaughter facility. Animals were slaughtered in four separate groups, with approximately 30 animals in each group. Following slaughter, gravid uterine tracts were collected and analyzed. Data collected on these tracts included ovulation rate, horn length, number of fetuses in each horn, fetal weight, uterine weight, number of mummies, fetal sex, fetal placement, fetal survival ([number of fetuses / ovulation rate] * 100), and fetal space (uterine length / [number of fetuses + number of mummies]).
Allele frequency analysis was performed using Excel 6.0. Allele and genotype frequencies were calculated within each of the breed subgroups and within each of the two larger (n = 322 and n = 107) groups. Expected genotype frequencies were calculated based on the Hardy-Weinberg equation. Expected and observed values were compared using a chi-square test to determine the presence of Hardy-Weinberg equilibria in each studied population.
All reproductive tract data were analyzed using the General Linear Model Procedures of SAS (1990). Data were analyzed using a model that included the effects of ESR genotype, breed, parity, and all significant two-way interactions. Uterine horn was also included in some analyses. Linear contrasts were used to determine the presence of individual heterosis and maternal breed effects.
Allele and genotype frequencies for the population of slaughtered animals are shown in Table 1. The A allele was more frequent than the B allele in all breed groups except for the LW x Y group. All breed groups were also tested for the presence of Hardy-Weinberg equilibria (data not shown). All groups were found to be in equilibrium, except for the Y x LW group; this group had a larger number of animals with the AB genotype than was expected based on the Hardy-Weinberg principle. The small number of animals in each of the breed groups, however, makes the determination of Hardy-Weinberg equilibria very difficult; in small groups such as these, the effects of sampling error could be quite large.
ESR genotype was not found to be a significant (P > 0.05) effect for any of the traits studied. P-values for the effect of ESR genotype in the model were generally very high, ranging from 0.2 to 0.8. Some traits did show notable, but statistically nonsignificant, trends with respect to ESR genotype (Table 2). Note the trend for animals with additional copies of the ESR B allele to have increased fetal survival, increased fetal weight, increased uterine length, and decreased fetal space. These trends agree with previous reports that have showed the ESR B allele to be associated with an increased number of pigs born per litter. To determine these effects, however, large numbers of litter records were required. In a study by Short et al. in 1997, more than 9,000 litter records were required in order to find the small effect of the ESR B allele. In contrast, our current study only utilized 107 records to determine the effect of the ESR gene. Therefore, it seems logical that observed trends were not found to be significant. The future addition of more animals to the study will help verify the validity of these trends.
| Table 1. Estrogen Receptor Gene (ESR) Allele and Genotype Frequencies for the Slaughtered Animal Population. | ||||||
|---|---|---|---|---|---|---|
| Breed of Animala | N | ESR Allele Frequencies | ESR Genotype Frequencies | |||
| A | B | AA | AB | BB | ||
| YxY | 36 | 0.51 | 0.49 | 0.22 | 0.58 | 0.19 |
| YxLW | 26 | 0.52 | 0.48 | 0.12 | 0.81 | 0.10 |
| LWxLW | 28 | 0.64 | 0.36 | 0.36 | 0.57 | 0.07 |
| LWxY | 17 | 0.38 | 0.62 | 0.06 | 0.65 | 0.29 |
| Total Population | 107 | 0.53 | 0.47 | 0.20 | 0.65 | 0.15 |
| a YxY = Yorkshire sire x Yorkshire dam, YxLW Yorkshire x Large White dam, LWxLW = Large Whire sire x Large White dam, LWxY = Large White sire x Yorkshire dam. | ||||||
| Table 2. Least-Squares Means and Standard Errors for All Reproductive Traits That Showed Notable Trends With Respect to the Estrogen Receptor Gene (ESR) Genotype. | |||||
|---|---|---|---|---|---|
| Reproductive Traita | N | Least-Squares Means and Standard Errors for
Animals With Specified ESR Genotype. |
P-Valueb | ||
| AA | AB | BB | |||
| FETSRV | 100 | 52.8 ± 4.7 | 59.1 ± 2.5 | 61.3 ± 5.2 | 0.38 |
| TNOFET | 100 | 10.20 ± 0.83 | 11.05 ± 0.44 | 11.58 ± 0.92 | 0.50 |
| UTLTH | 100 | 543 ± 23 | 567 ± 12 | 582 ± 25 | 0.48 |
| TNOMUM | 106 | 0.27 ± 0.15 | 0.27 ± 0.07 | 0.57 ± 0.22 | 0.45 |
| TFETWT | 100 | 3,735 ± 158 | 3,889 ± 87 | 4,004 ± 172 | 0.49 |
| HNLTH | 204 | 271.3 ± 8.4 | 282.5 ± 4.5 | 290.6 ± 9.1 | 0.27 |
| NOFET | 211 | 5.11 ± 0.30 | 5.57 ± 0.16 | 5.84 ± 0.34 | 0.23 |
| FETSPC | 204 | 58.3 ± 3.8 | 54.9 ± 1.9 | 52.1 ± 5.5 | 0.61 |
| a FETSRV = percentage of ova per uterus that survive to day 75 of gestation, TNOFET = number of fetuses per uterus, UTLTH = total length of uterine horns (cm), TNOMUM = total number of mummies per uterus, TFETWT = total fetal weight (g) per uterus, HNLTH = length of uterine horn (cm), NOFET = number of fetuses per horn, FETSPC = amount of uterine space (cm) available per fetus. b Significance level of effect of ESR genotyoe on specified trait. | |||||
Parity significantly affected several of the reproductive traits (Table 3). Animals of a higher parity have both a larger ovulation rate and a lower fetal space. This could reflect the increased reproductive efficiency of older animals, which ovulate more eggs, carry more piglets to farrowing, and have a reduced fetal space (Hughes and Varley, 1980). However, parity was not significantly associated with an increased number of fetuses (TNOFET, P = 0.220; NOFET, P = 0.0925).
Breed was also significantly associated with several of the traits studied (Table 4). Linear contrasts detected the presence of a maternal breed effect for several traits (Table 5). Animals with a Large White dam had an increased number of fetuses per horn, increased fetal weight per horn, and a decreased fetal space per horn. Linear contrasts did not detect the presence of positive individual heterosis for any of the traits studied. However, negative heterosis was detected for several of the traits (Table 5), with the purebred animals having a greater performance than the crossbred animals. Possibly, the low number of crossbred animals (compared to purebred) included in the study may affect the true determination of heterosis.
| Table 3. Least-Squares Means and P-values for Reproductive Traits Associated With Parity. | |||||
|---|---|---|---|---|---|
| Reproductive Traitc |
N | Least-Squares Means for Animals of Specified Parity | P-valued | ||
| 1 | 2 | => 3 | |||
| TOV | 100 | 17.98a | 18.49a | 21.65b | 0.0006 |
| OV | 204 | 9.00a | 9.21a | 10.78b | 0.055 |
| FETSPC | 204 | 61.6a | 58.0ab | 45.6b | 0.044 |
| a-b Means within a row without a common subscript are significantly different (P < 0.05). c TOV = total number of corpora lutea per animal, OV = number of corpea lutea per horn, FETSPC = amount of uterine space (cm) available per fetus per horn. d Significance level of effect of parity on specified trait. | |||||
| Table 4. Least-Squares Means and P-Values for Reproductive Traits Associated With Breed | ||||||
|---|---|---|---|---|---|---|
| Reproductive Traitd |
N | Least-Squares Means for Animals of Specified Breede | P-valuef | |||
| YxY | YxLW | LWxLW | LWxY | |||
| UTLTH | 100 | 598a | 548b | 586ab | 522b | 0.036 |
| TFETSPC | 100 | 61.8a | 50.8b | 50.5b | 56.3ab | 0.054 |
| HNLTH | 204 | 298.9a | 274.6bc | 292.1ab | 260.4c | 0.002 |
| FETWT | 211 | 1,839a | 1,949b | 1,993b | 1,731a | 0.0001 |
| FETSPC | 204 | 62.3a | 50.3b | 50.6b | 57.1ab | 0.006 |
| a-c Means within a row without a common subscript are significantly different (P < 0.05). d UTLTH = total length of uterine horns (cm), TFETSPC = amount of uterine space (cm) available per fetus per uterus, HNLTH = length of uterine horn (cm), TFETWT = total fetal weight (g) per horn, FETSPC = amount of uterine space (cm) available per fetus per horn. e YxY = Yorkshire sire x Yorkshire dam, YxLW Yorkshire x Large White dam, LWxLW = Large Whire sire x Large White dam, LWxY = Large White sire x Yorkshire dam. f Significance level of effect of parity on specified trait. | ||||||
| Table 5. Least-Squares Means and P-values for the Determination of Breed Effects for Reproductive Traits. | |||||
|---|---|---|---|---|---|
| Breed of Animala | Least-Squares Means and Standard Errors for Selected Reproductive Traitsb | ||||
| TFETSPC | HNLTH | NOFET | FETWT | FETSPC | |
| YxY | 61.84 ± 3.18 | 298.9 ± 6.3 | 5.35 ± 0.23 | 1839 ± 38 | 62.33 ± 3.0 |
| LWxLW | 50.53 ± 3.89 | 292.1 ± 7.5 | 5.80 ± 0.27 | 1993 ± 44 | 50.63 ± 4.5 |
| YxLW | 50.84 ± 4.02 | 274.6 ± 8.0 | 5.94 ± 0.29 | 1949 ± 43 | 50.29 ± 3.5 |
| LWxY | 56.33 ± 4.85 | 260.4 ± 9.5 | 4.94 ± 0.34 | 1731 ± 53 | 57.07 ± 4.3 |
| Pures versus Crosses P-valuec | 0.50 | 0.0002 | 0.31 | 0.06 | 0.38 |
| Maternal Breed Effect P-valued | 0.03 | 0.63 | 0.007 | 0.0001 | 0.004 |
| a YxY = Yorkshire sire x Yorkshire dam, YxLW Yorkshire x Large White dam, LWxLW = Large Whire sire x Large White dam,LWxY = Large White sire x Yorkshire dam. b TFETSPC = amount of uterine space (cm) available per fetus per uterus, HNLTH = length of uterine horn (cm), NOFET = number of fetuses per horn, FETWT = total fetal weight (g) per horn, FETSPC = amount of uterine space (cm) available per fetus per horn. c For the linear contrast for heterosis, where H0 = no differences in the indicated trait between purebred and crossbred animals. d For the linear contrast for maternal effects, where H0 = no differences in the indicated trait between animals with the same breed of dam. | |||||
The results of this study allow us to begin to construct a preliminary picture of how the ESR gene positively influences the reproductive performance of the female pig. Adding copies of the ESR B allele appears to increase the fetal survival, total number of fetuses, total fetal weight, and number of fetuses per horn in the pregnant female. The overall effect of the B allele would therefore be an increase in reproductive performance, which has previously been demonstrated with both litter traits (Short et al ., 1997) and placental traits (Van Rens and van der Lende, 1998). The addition of more animals to this study should allow a final determination of the true validity and significance of these trends.
Haley, C. S., E. D'Agaro, and M. Ellis. 1992. Genetic components of growth and ultrasonic fat depth traits in Meishan and Large White pigs and their reciprocal crosses. Animal Production. 266:105.
Hughes, P. E. and M. A. Varley. 1980. Reproduction in the Pig. Butterworth & Co., Ltd., Boston.
Korach, K. S. 1994. Insights from the study of animals lacking functional estrogen receptor. Science. 266:1524.
Lehrer, S., M. Sanchez, H. K. Song, J. Dalton, F. Levine, P. Savoretti, S. N. Thung, and B. Schachter. 1990. Oestrogen receptor B-region polymorphism and spontaneous abortion in women with breast cancer. Lancet. 335:622.
O'Malley, B. 1990. The steroid receptor superfamily: more excitement predicted for the future. Mol. Endocrinol. 4:363.
Rothschild, M. F., R. Larson, C. Jacobson, and P. Pearson. 1991. PvuII polymorphisms at the porcine oestrogen receptor locus (ESR). Anim. Genet. 22:448.
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.
Short, T. H., M. Rothschild, O. Southwood, D. McLaren, A. de Vries, H. van der Steen, G. Eckardt, C. Tuggle, J. Helm, D. Vaske, A. Mileham, and G. Plastow. 1997. Effect of the estrogen receptor locus on reproduction and production traits in four commercial pig lines. J. Anim. Sci. 75:3138.
SAS. 1990. SAS/STAT User's Guide. SAS Institute, Inc. Cary, N.C.
Van Rens, B. T. T. M. and T. van der Lende. 1997. The estrogen receptor gene and prolificacy in the pig. Proceedings of the Wageningen Institute of Animal Science Seminar, "What Have We Learned from the Meishan Pig."