J. Anim. Sci. 2005. 83:2308-2312
© 2005 American Society of Animal Science
Divergent selection for uterine capacity in rabbits. III. Responses in uterine capacity and its components estimated with a cryopreserved control population1
M. L. Mocé2,
M. A. Santacreu,
A. Climent and
A. Blasco
Departamento de Ciencia Animal, Universidad Politécnica de Valencia, 46071 Valencia, Spain
 |
Abstract
|
|---|
This work evaluated the response to 10 generations of divergent selection for uterine capacity (UC) in rabbits to determine whether this response was symmetric by contrasting both lines against a cryopreserved control population. Animals came from the 13th generation of an experiment of divergent selection for UC and from a cryopreserved control population. The two UC lines were divergently selected for 10 generations, and selection was relaxed from the 11th generation until the 13th generation. Uterine capacity was estimated as litter size (LS) in unilaterally ovariectomized (ULO) does. To create the control population, embryos from the base generation were vitrified and stored in liquid N2 for 10 generations. Data from 461 pregnancies produced by 134 ULO does were used: 62 does from the high UC line, 55 females from the low UC line, and 17 females from the control line. The following traits were analyzed: ovulation rate (OR); number of implanted embryos (IE); (UC), estimated as total number of rabbits born; number born alive (NBA); prenatal survival (PS), estimated as UC/OR; embryo survival (ES), estimated as IE/OR; and fetal survival (FS), estimated as UC/IE. Ovulation rate, IE, PS, ES, and FS were measured by laparoscopy only in the second parity. Uterine capacity and NBA were measured over four parities. Responses in UC and its components were estimated as differences between the selected lines and the control line using a Bayesian approach. Selection for UC led to differences of 1.01 kits between the high and low lines, but this response was asymmetric. No differences were found between the high and control lines (high – control = –0.08), whereas the low and control lines differed by 1.08 kits, with a probability of the difference being greater than zero of 0.98. Difference between the high and low lines and between the control and low lines was one-half of the difference reported for correlated response in LS in previous studies. No differences in OR were detected among lines. The control and low lines differed by 1.06 IE, with a probability of the difference being higher than zero of 0.84. Prenatal survival for the low line was less than that of the control line. In summary, selection for UC was asymmetric, which was mainly due to a correlated response in PS. Response in UC was one-half of the difference reported for correlated response in LS in previous studies.
Key Words: Control Population • Ovulation Rate • Prenatal Survival • Rabbits • Selection Response • Uterine Capacity
|
Introduction
|
|---|
Selection for uterine capacity (UC) has been proposed as an alternative method to improve litter size (LS) (Blasco et al., 1994
). Selection for UC has been successful in mice (Kirby and Nielsen, 1993) and in rabbits (Blasco et al., 2005). Moreover, a correlated response in LS when selecting for UC was observed in both mice (Kirby and Nielsen, 1993) and rabbits (Santacreu et al., 2005). Correlated response in LS was asymmetric (Santacreu et al., 2005), whereas direct response to selection for UC was symmetric (Blasco et al., 2005). Analyses performed by Blasco et al. (2005) were based on predicted genetic trends. Cryopreserved control populations are useful in divergent selection experiments because they can detect asymmetric responses. Moreover, these control populations are not affected by drift and allow increased selection pressure on the selected lines because they do not require experimental facilities.
This study was conducted to evaluate the response to divergent selection for UC in rabbits and to determine whether this response was symmetric by contrasting both lines against a cryopreserved control population.
|
Materials and Methods
|
|---|
Animals
Animals came from a divergent selection experiment for UC and from a cryopreserved control population. The two UC lines were divergently selected for 10 generations, and selection was relaxed from the 11th until the 13th generation. Animals from the 13th generation were used in this experiment. A detailed description of the selection procedure has been given by Blasco et al. (2005). To create the control population, embryos from the base generation were vitrified and stored in liquid N2 and subsequently thawed and transferred. The control population was contemporary to the 11th generation of the high and low UC lines. The control population was maintained for two generations, and in this experiment, we used animals from the 2nd generation, which were contemporary to animals from the 13th generation. To avoid inbreeding, planned matings were used during these two generations. Embryo recovery, cryopreservation, and transferring techniques were as described by Vicente and García-Ximénez (1996). Unilateral ovariectomies as described by Blasco et al. (1994) were performed on all females at 14 to 16 wk of age (ULO [unilaterally ovariectomized] females). Does were first mated at 18 wk of age and thereafter 10 d after parturition. A laparoscopy was performed on all does at d 12 of their second gestation to record number of corpora lutea and number of implanted embryos (IE). Details of the laparoscopic technique can be found in Santacreu et al. (1990).
Data from 461 pregnancies produced by 134 ULO females were used: 62 does from the 13th generation of the high line, 55 females from the 13th generation of the low line, and 17 females from the control line. Animals were housed at the experimental farm of the Universidad Politécnica de Valencia in individual metal cages. Animals were kept under controlled (16 h light:8 h dark) photoperiods and were fed ad libitum with a commercial diet (barley and wheat as the primary grains, wheat bran, barley straw, and alfalfa hay as the fiber source).
Traits
The following traits were analyzed: ovulation rate (OR), estimated as number of corpora lutea; number of IE, estimated as number of implantation sites; UC, estimated as total number of rabbits born; number born alive (NBA); prenatal survival (PS), estimated as UC/ OR; embryo survival (ES), estimated as IE/OR; and fetal survival (FS), estimated as UC/IE. Ovulation rate, IE, PS, ES, and FS were measured only in the second parity. Uterine capacity and NBA were measured over four parities.
Statistical Analyses
Traits were analyzed using a Bayesian approach. Uterine capacity and NBA were analyzed using a model with the effects of line (high, low, and control), year-season, lactation state with five levels (nulliparous does, lactating and nonlactating does of second parity, and lactating and nonlactating does with more than two parities), and an effect of doe. To analyze the traits recorded only in the second parity, a model including the effects of line, year-season, and lactation state (lactating and nonlactating females) was used.
Traits were assumed to be normally distributed. Bounded uniform priors were used for all unknowns, with the exception of doe effect, which was considered normally distributed with mean 0 and variance I
Priors for variances also were bounded uniform. Features of the marginal posterior distribution of the differences between lines were estimated using Gibbs sampling. After some exploratory analyses, we used a single long chain of 120,000 samples with a burn-in period of 20,000. Only one sample from each 10 samples was saved for inferences. Convergence was tested using the Z criterion of Geweke (Sorensen and Gianola, 2002).
|
Results and Discussion
|
|---|
Means and SD for the traits measured in the control line are shown in Table 1. Values obtained for all of the traits are in the usual range reported by others (Blasco et al., 1994, 2005). Santacreu et al. (2005) found an LS for the control line of 9.2 kits, but they used females with both uterine horns functional. Thus, ULO females, with only one functional uterine horn, produced litters smaller than those of intact females. This result agrees with the findings of Blasco et al. (1994), who observed that ULO rabbits produced litters at 77% of the size of intact females. In contrast, Johnson (1970), using only 10 ULO females, did not find any difference between LS of intact and ULO rabbit females. In mice, results are similar to those reported for rabbits; ULO mice produced litters approximately 85% of the size of intact females (Johnson, 1970; Lamberson et al., 1989). In pigs, in contrast to rabbits and mice, intact sows showed a LS that was twice the size of ULO females (Fenton et al., 1970; Père et al., 1997).
View this table:
[in this window]
[in a new window]
|
Table 1. Means, SD, and number of litters for uterine capacity (UC), number born alive (NBA), ovulation rate (OR), number of implanted embryos (IE), embryo survival (ES), fetal survival (FS), and prenatal survival (PS) in the control linea
|
|
Features of the estimated marginal posterior distributions of the traits UC and NBA are presented in Table 2. Monte Carlo SE were small, and the Geweke test did not detect lack of convergence in any case. Selection for UC led to differences of one kit between the high and low lines, but this response was asymmetric. The high and control lines had similar UC, whereas the low and control lines differed by 1.08 kits, with a probability of the difference being greater than zero of 0.98. Santacreu et al. (2005), using females from the high, low, and control lines, reported an asymmetric correlated response for LS in intact females with two functional uterine horns, but they reported a low correlated response for LS in the line selected for high UC, which was not detected in our experiment. Correlated response in NBA also was asymmetric.
View this table:
[in this window]
[in a new window]
|
Table 2. Features of the marginal posterior distributions of the differences between high and control line (H – C), control and low line (C – L), and high and low line (H – L) in uterine capacity (UC) and number born alive (NBA)
|
|
This asymmetric response might be explained by the presence of a major gene controlling UC. Argente et al. (2003) performed a complex segregation analysis using data from the 10 generations of this divergent selection experiment and found evidence for a QTL with moderate effect on UC. This result may explain both the asymmetric response and the large response found by Blasco et al. (2005) in the first and the second generations of selection. Frequency of the favorable allele on the base population was estimated to be 0.7 (Argente et al., 2003), which could explain the greater response obtained when selecting for low UC than when selecting for high UC; however, this asymmetric response was not detected when analysis of predicted trends was performed (Blasco et al., 2005). Results obtained by Blasco et al. (2005) are more precise than the results obtained in our experiment, but they are more dependent on the model used for the analysis because predicted genetic trends are highly dependent on the estimated genetic parameters (Sorensen and Johansson, 1992).
A hypothesis of the model of UC proposed by Blasco et al. (1994) is that the response obtained in females with only one functional uterine horn will be duplicated in intact females. Santacreu et al. (2005) reported a difference between the high and low lines and between the control and low lines that was twice the difference observed for UC in this experiment, which agrees with the previous hypothesis. Conversely, Kirby and Nielsen (1993) reported that, in mice, direct response in UC (1.8 pups) was similar to correlated response in LS (1.7 pups).
Correlated responses in components of LS were calculated using a smaller number of observations; thus, inferences from data reported in Tables 3 and 4 should be made with caution. A correlated response on OR was not detected, and the three lines showed similar OR (Table 3). The control and low lines differed by 1.06 IE, with a probability of the difference being greater than zero of 0.84. Although when working with P-values, this result is considered to be nonsignificant, in a Bayesian framework, it gives the actual value of the probability of the difference between lines. Thus, 0.84 is a high probability of the control line being different from the low line. In mice, Clutter et al. (1994) reported a slight correlated response for OR after 21 generations of selection for greater UC, but this correlated response was much smaller than the response obtained when selection was on LS or on an index based on OR and PS. The high and low lines differed by only 0.46 IE, with a probability of the difference being greater than zero of 0.78. As previously observed for UC, the difference between the lines in number of IE in ULO females was one-half of the difference reported by Santacreu et al. (2005) in intact females.
View this table:
[in this window]
[in a new window]
|
Table 3. Features of the marginal posterior distributions of the differences between high and control line (H – C), control and low line (C – L), and high and low line (H – L) in ovulation rate (OR) and number of implanted embryos (IE)
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Features of the marginal posterior distributions of the differences between high and control line (H – C), control and low line (C – L), and high and low line (H – L) in embryo survival (ES), fetal survival (FS), and prenatal survival (PS)
|
|
Results obtained for survival (Table 4
) are not in disagreement with the previous results published by Blasco et al. (2005) and Santacreu et al. (2005). The high and low lines differed in PS, and these differences were due to differences in FS and ES. Once again, it was observed that the differences between the lines for PS using ULO females were smaller than those observed when using intact females (Santacreu et al., 2005).
|
Implications
|
|---|
In rabbits, response in uterine capacity for unilaterally ovariectomized females is one-half of the response obtained in intact females; thus, uterine capacity measured in a single horn is a good indicator of total uterine capacity of the female. Litter size is the main goal for the rabbit industry. Selection for greater uterine capacity, which would be an indirect way to improve litter size, was not effective in producing response; however, selection for less uterine capacity was successful. Response for low uterine capacity was mainly due to a correlated response in prenatal survival. This trait deserves more research for a better understanding of the genetic control of litter size.
|
Footnotes
|
|---|
1 This study was supported by the Comisión Interministerial de Ciencia y Tecnología CICYT-AGL2001-3068-C03-01 and CICYT-AGL2002-04383-C02-02). This research was approved by the Ethical Committee of the Universidad Politecnica de Valencia and also has fulfilled the ethical requirements of the Ministerio de Educación y Ciencia for animal experimentation. 
2 Correspondence: Departamento de Producción Animal y Ciencia y Tecnología de los Alimentos, Universidad Cardenal Herrera-CEU, Edificio Seminario, 46113 Moncada, Valencia, Spain (phone: 34961369000; fax: 34961395272; e-mail: mmoce{at}uch.ceu.es).
Received for publication March 3, 2005.
Accepted for publication July 1, 2005.
|
Literature Cited
|
|---|
Argente, M. J., A. Blasco, J. A. Ortega, C. S. Haley, and P. M. Visscher. 2003. Analyses for the presence of a major gene affecting uterine capacity in unilaterally ovariectomized rabbits. Genetics 163:1061–1068.[Abstract/Free Full Text]
Blasco, A., M. J. Argente, C. S. Haley, and M. A. Santacreu. 1994. Relationships between components of litter size in unilaterally ovariectomized and intact rabbit does. J. Anim. Sci. 72:3066–3072.[Abstract]
Blasco, A., J. A. Ortega, M. A. Santacreu, and A. Climent. 2005. Divergent selection for uterine capacity in rabbits. I. Genetic parameters and response to selection. J. Anim. Sci. 83:2297–2302.[Abstract/Free Full Text]
Clutter, A. C., Y. L. Kochera Kirby, and M. K. Nielsen. 1994. Uterine capacity and ovulation rate in mice selected 21 generations on alternative criteria to increase litter size. J. Anim. Sci. 72:577–583.[Abstract]
Fenton, F. R., F. W. Bazer, O. W. Robison, and L. C. Ulberg. 1970. Effect of quantity of uterus on uterine capacity in gilts. J. Anim. Sci. 31:104–106.
Johnson, A. D. 1970. Limitation of fetus number in the rat, mouse and rabbit. J. Anim. Sci. 30:978–983.
Kirby, Y. L., and M. K. Nielsen. 1993. Alternative methods of selection for litter size in mice. III. Response to 21 generations of selection. J. Anim. Sci. 71:571–578.[Abstract]
Lamberson, W. R., R. M. Blair, and C. R. Long. 1989. Effects of unilateral ovariectomy on reproductive traits of mice. Anim. Reprod. Sci. 20:49–55.
Père, C., J. Y. Dourmad, and M. Etienne. 1997. Effect of number of pig embryos in the uterus on their survival and development and on maternal metabolism. J. Anim. Sci. 75:1337–1342.[Abstract/Free Full Text]
Santacreu, M. A., M. L. Mocé, A. Climent, and A. Blasco. 2005. Divergent selection for uterine capacity in rabbits. II. Correlated response in litter size and its components estimated with a cryopreserved control population. J. Anim. Sci. 83:2303–2307.[Abstract/Free Full Text]
Santacreu, M. A., M. P. Viudes de Castro, and A. Blasco. 1990. Evaluation par coelioscopie des corps jaunes et des embryons. Influence sur la taille de portée chez la lapine. Reprod. Nutr. Dev. 30:583–588.
Sorensen, D., and D. Gianola. 2002. Likelihood, Bayesian, and MCMC Methods in Quantitative Genetics. Springer, New York, NY.
Sorensen, D., and K. Johansson. 1992. Estimation of direct and correlated responses to selection using univariate animal models. J. Anim. Sci. 70:2038–2044.[Abstract]
Vicente, J. S., and F. García-Ximénez. 1996. Direct transfer of vitrified rabbit embryos. Theriogenology 45:811–815.
Top
Abstract
Introduction
