Genetic selection for ovulation rate and litter size in rabbits: estimation of genetic parameters, direct and correlated responses

INTERNATIONAL MASTER ON ANIMAL BREEDING AND REPRODUCTION BIOTECHNOLOGY Genetic selection for ovulation rate and litter size in rabbits: estimation of genetic parameters, direct and correlated responses Thesis of Master Valencia, 7th June 2012 Chiraz Ziadi Thesis supervisors: Dra. Mª Antonia Santacreu Jerez Dra. Mª Lorena Mocé Cervera ♥♥♥ A Mes Très Chers Parents ♥♥♥ Tous les mots du monde ne sauraient exprimer l’immense amour que je vous porte, ni la profonde gratitude que je vous témoigne pour tous les efforts et les sacrifices que vous n’avez jamais cessé de consentir pour mon instruction et mon bien-être. C’est à travers de vos encouragements et que j’ai opté pour cette noble profession, et c’est à travers de vos critiques que je me suis realisée. J’espère avoir répondu à aux espoirs que vous avez fondé en moi. Je vous rends hommage par ce modeste travail en guise de ma reconnaissance éternelle et de mon infini amour. Vous résumez si bien le mot parents qu’il serait superflu d’y ajouter quelque chose. Que Dieu tout puissant vous garde et vous procure santé, bonheur et longue vie pour que vous demeuriez le flambeau illuminant le chemin de vos enfants. We know very little, and yet it is astonishing that we know so much, and still more astonishing that so little knowledge can give us so much power. - Albert Einstein- ACKNOWLEDGMENTS In the name of Allah, the Most Gracious and the Most Merciful I’m very much indebted to Dra. María Antonia Santacreu and Dra María Lorena Mocé for being an outstanding advisors and excellent professors. Their constant encouragement, support, and invaluable suggestions made this work successful. I’m so grateful for their patience with all the problems which rinse during this work. They have been everything that one could want in an advisor. I am deeply indebted to my committee members Dr. Manuel Baselga and Dr. Duní Gabiña. Moose for their time and effort in reviewing this work. I would like to thanks to Dr. Agustín Blasco for being excellent profesor, always available for scientific discussions and for being patience all the time repeating for me the same things for a good understanding of their fields. I would like to express my gratitude to my colleagues in the Department of Animal Science, especially in the fourth floor (Mohamed, Pilar, Paty, Ayman, Carlos and Ahmad) for creating and maintaining a cheerful atmosphere in our group. Special thanks go to Cristina and Vero for being not only excellent colleagues but also very good friends, for their understanding and help in all the occasions. I would like to express my thanks to my Palestinian lovely friend Rima, who always give help in any time, encouragement and hear to my problems. My sincere thanks to my Egyptian friends: Abeer, Hanan, Hadeer, Afaf, Tarek, for supporting, encouragement me and for the good time we shared in our apartment. I’m most grateful to the Mediterranean Agronomic Institute of Zaragoza (IAMZ), for supporting me with a full grant to study two years for having the international master degree, which helped me to exchange powerful scientific ideas with persons from different countries and upgrade my knowledge. Last but not least, my deepest gratitude goes to my beloved parents and also my lovely sister for their endless love, prayers and encouragement. Table of Content Chapter 1 General Introduction I. General introduction to rabbit production................................................... 2 II. Litter size and its components: ovulation rate and prenatal survival ......... 4 1. Ovulation rate................................................................................................ 4 1.1 Estimation of ovulation rate.............................................................................4 1.2 The mechanism of ovulation............................................................................4 1.3 Timing of ovulation.........................................................................................5 1.4 Oocyte quality.................................................................................................5 2. Prenatal survival ........................................................................................... 6 2.1 Components of prenatal survival: embryonic and fetal survival .......................6 2.2 Estimation of embryonic and fetal survival......................................................6 2.3 Timing and extent of prenatal mortality...........................................................7 2.4 Factors associated with prenatal mortality .......................................................7 III. Genetic improvement for litter size .............................................................. 8 1. Conventional selection for litter size............................................................. 8 2. Selection for the components of litter size .................................................... 9 2.1 Selection for ovulation rate..............................................................................9 2.2 Selection for prenatal survival .......................................................................12 2.3 Selection for uterine capacity.........................................................................13 2.4 Index selection for components of litter size: Ovulation rate and prenatal survival .........................................................................................................14 2.5 Two-stage selection for ovulation rate and litter size......................................16 Chapter 2 Objectives................................................................................................... 28 Chapter 3 Selection for ovulation rate and litter size in rabbits: genetic parameters, direct and correlated responses............................................................. 30 Chapter 4 General Discussion.....................................................................................52 Chapter 5 Conclusions.................................................................................................58 Index of Tables Table 1.1: Direct responses in number of kits born alive (NBA) or number of kits weaned (NW) and correlated responses in ovulation rate (OR) and prenatal survival (PS) estimated per generation in rabbits, with their standard errors (in parenthesis). Table 1.2a: Estimated heritabilities (h2) of ovulation rate (OR), litter size (LS) and prenatal survival (PS) and phenotypic and genetic correlations (rpand rg, respectively) between these traits and litter size (LS) at the day of gestation indicated (DG) in mice, pigs and rabbits. Table 1.2b: Estimated heritabilities (h2) of ovulation rate (OR), litter size (LS) and prenatal survival (PS) and phenotypic and genetic correlations (rpand rg, respectively) between these traits and litter size (LS) at the day of gestation indicated (DG) in mice, pigs and rabbits. Table 1.3: Direct response in ovulation rate (OR) and correlated responses in total number born (TNB) and prenatal survival (PS) with their standard errors (SE) estimated per generation in the experiments of selection for OR in mice and pigs. Table 1.4: Responses in prenatal survival (PS), ovulation rate (OR) and litter size (LS) estimated per generation in pigs and mice selected for prenatal survival with their standard errors (in parenthesis). Table 1.5: Responses to selection to increase uterine capacity (UC) and correlated responses in ovulation rate (OR), litter size (LS) and prenatal survival (PS) estimated per generation. Table 1.6: Responses to selection in ovulation rate (OR), litter size (LS) and prenatal survival (PS) estimated per generation in pigs and mice selected for an index of ovulation rate and prenatal survival, with their standard errors (in parenthesis). Table 3.1: Descriptive statistics for litter size (LS), ovulation rate (OR), implanted embryos (IE), embryo survival (ES), fetal survival (FS), and prenatal survival (PS). Table 3.2: Means and SD (in parentheses) for ovulation rate (OR), litter size (LS), number of implanted embryos (IE), embryo survival (ES), fetal survival (FS) and prenatal survival (PS) in generations 0 to 13. Table 3.3: Features of the marginal posterior distributions of the heritability (h2) and the repeatability (r) of litter size (LS), ovulation rate (OR), number of implanted embryos (IE), embryo survival (ES) and the heritability of fetal survival (FS) and prenatal survival (PS). Table 3.4: Features of the marginal posterior distributions of the genetic correlation between the traits analyzed: litter size (LS), ovulation rate (OR), number of implanted embryos (IE), embryo survival (ES), fetal survival (FS) and prenatal survival (PS). Table 3.5: Features of the marginal posterior distributions of the phenotypic correlation between the traits analyzed: litter size (LS), ovulation rate (OR), number of implanted embryos (IE), embryo survival (ES), fetal survival (FS) and prenatal survival (PS). Index of Figures Figure 1.1: Rabbit carcass meat production in Europe in 2010. Figure 1.2 Rabbit census in Spain in 2010. Figure 1.3: Rabbit carcass meat production in Spain in 2010. Figure 1.4: Meat consumption in Spain in 2001. Figure 3.1: Genetic trends for ovulation rate (OR), litter size (LS) and number of implanted embryos (IE) of line OR_LS. Figure 3.2: Genetic trends for embryo survival (ES), fetal survival (FS) and prenatal survival (PS) of line OR_LS. Chapter 1 6 competence along gestation following embryo transfer. However, this is usually not possible because of economic and technical factors. The study of oocyte morphology is relatively quick and simple; however, it is unreliable if it is not accompanied by other methods (Balaban and Urman, 2006). Other methods to assess oocyte quality that have been proposed are: measurement of ATP, an important energy source for maintaining protein synthesis and other cellular functions (reviewed by Krisher, 2004); measurement of glutathione (GSH), the main compound that protects the cell against the oxidative stress (reviewed by Luberda, 2005; Rausell and Tarín, 2005); quantification of mitochondrial DNA; quantification of oocyte mRNA and proteins (reviewed by Krisher, 2004). Besides, some authors have studied the expression of genes in the granulosa cells or in the oocyte itself, looking for specific molecular markers of oocyte quality, or have performed polar body biopsy to screen oocytes with chromosomal abnormalities deriving from errors in the two meiotic divisions (reviewed by Revelli et al., 2009). 2. Prenatal survival Prenatal survival is an important character in animal production, a high prenatal survival leads to increased litter size at birth and consequently to greater economic benefits (Santacreu, 2006). 2.1 Components of prenatal survival: embryonic and fetal survival Prenatal survival is by definition the proportion of ova shed giving birth to young. It comprises two periods: the embryonic and the fetal period. In rabbits, it has been accepted to call embryonic period to the period before implantation (d 7) and fetal period to the period from implantation until birth (d 30) (Mocé et al., 2010). 2.2 Estimation of embryonic and fetal survival Embryonic survival is calculated as the proportion of implanted embryos from the number of corpora lutea and fetal survival is calculated as the proportion of kits born from the number of implanted embryos. Besides, prenatal survival is the product of both embryonic and fetal survival; it is calculated as the proportion of kits born from the number of corpora lutea. The estimation of survival during gestation requires the counting of implanted embryos or fetuses. In rabbits the laparoscopic method permits the estimation of embryonic and fetal survival in the same female without affecting litter size (Santacreu et al., 1990). However, in pigs, it is not possible to estimate ovulation rate, embryo survival and foetal survival in the same females without General Introduction 7 compromising litter size (Neal and Johnson, 1986) because in this species implantation sites cannot be determined by observation of the external surface of the uterus. This make the rabbit a particularly useful model for examine the relationships between litter size and its components in the same females. 2.3 Timing and extent of prenatal mortality Prenatal mortality is the mayor limiting factor of litter size in rabbits like in pigs and mice. In rabbits, prenatal mortality is around 30% (Adams, 1960a, b; García and Baselga, 2002), 1014% corresponding to the embryonic period, and 20-22% to the fetal period. Prenatal mortality in mice is lower, around 20%. This percentage is almost equally distributed between the pre and the post-implantation period (reviewed by Wilmut et al., 1986). In pigs, a prenatal loss of 40 to 60% has been reported (reviewed by Foxcroft et al., 2006); the largest proportion of it occurs before d 30-35 of gestation of the 114day gestation period. 2.4 Factors associated with prenatal mortality Physiologically, prenatal survival is a complex trait which depends on a series of events ranging from gamete maturation to the birth of viable offspring (Blasco et al., 1993). Prenatal survival depends on the genotype of the dam, the embryo and their interaction. It seems that the maternal genotype plays the most important rol whereas the embryo genotype has a minor effect (reviewed by Bradford, 1969 and Blasco et al., 1993). The part of prenatal survival due to the female is called uterine capacity. Uterine capacity has been defined as the maximal number of fetuses a female can carry to term when the number of potentially viable embryos is not limiting (Bennett and Leymaster, 1989). Thus, uterine capacity depends on both embryo survival and fetal survival. Different causes have been suggested to explain the prenatal losses related to the maternal genotype in early stages of gestation: an increased number of immature oocytes when ovulation rate is high (Torres, 1982 in rabbit, Koenig et al., 1986 in pigs) and an inadequate secretion of certain proteins and hormones necessary for the development of the embryo (Beier, 2000; Daniel, 2000 in rabbit; Bagchi et al., 2001 in humans and rats; Vallet et al., 1998 in pigs). In the later stages of gestation, it has been suggested that the main cause of mortality is competition among embryos for the availability of space and nutrients when number of embryos in the uterus a large Chapter 1 8 (Adams, 1960b; Hafez, 1969). The most characteristics studied are the length, weight and degree of vascularization of the uterus (Argente et al., 2003). III. Genetic improvement for litter size 1. Conventional selection for litter size The most common criteria used in selection programs of rabbit maternal lines are litter size at birth or at weaning (for a review by Khalil and Al-Saef, 2008). Litter size has a low heritability (reviewed by Mocé and Santacreu, 2010 in rabbits; Rotschild and Bidanel, 1998 in pigs). Direct selection on litter size in closed populations have led to a response much lower than expected, around 0.1 young per generation (Table 1.1, for a review in rabbits; Ollivier and Bolet, 1981; Bolet et al., 1989; Holl and Robison, 2003 in pigs). However, in mice, direct selection for litter size has obtained a higher response than in pigs and rabbits, 0.15 to 0.20 young per generation (Bradford, 1968, 1969; Falconer, 1971; Bakker et al., 1978; Gion et al., 1990). Table 1.1: Direct responses in number of kits born alive (NBA) or number of kits weaned (NW) and correlated responses in ovulation rate (OR) and prenatal survival (PS) estimated per generation in rabbits, with their standard errors (in parenthesis). Responses Line G Cr Method NBA/NW OR PS Gó mez et al., Prat 3 NW BLUP/REML 0.09 /year - - Rochambeau et al., 1998 1077 18 NW BLUP/REML 0.08 0.06* - Control 0.08 - - 2066 18 NBA BLUP/REML 0.13 - - García and Baselga, 2002a V 0 - 21 NW BLUP/REML 0.09 - - 15 - 21 Control 0.09 0.18 0.06% García and Baselga, 2002b A 1 - 26 NW BLUP/REML 0.18 - - 17 - 26 Contr ol 0.09 0.01 0.41% G: Generations; Cr: Criterium of selection * Response estimated by Brun et al. (1992) after 13 generations of selection. Modified from Laborda (2011). The puzzling results obtained from selection experiments for litter size in close populations led to the search for alternative methods of selection: experiments of General Introduction 9 selection for ovulation rate, uterine capacity and prenatal survival for improving indirectly litter size. 2. Selection for the components of litter size An approach to increasing litter size is to select for its components, ovulation rate and prenatal survival. The efficiency for improving litter size through its components is highly dependent on their genetic parameters. There is also little information of the heritabilities of these traits and their correlations (Tables 1.2a, b). Ovulation rate has higher heritability than litter size, but this heritability is lower than in pigs and mice (reviewed by Blasco et al., 1993b). Genetic correlation between ovulation rate and litter size is positive and low. Prenatal survival has a low heritability similar to litter size, and a high correlation with litter size. Few selection experiments for components of litter size have been reported in rabbits: two divergent selection experiments for uterine capacity and one selection experiment for ovulation rate. 2.1 Selection for ovulation rate In rabbits, and another species like pigs and mice, the increase in litter size is basically associated to an increase in ovulation rate (Bolet et al., 1989 in pigs; Brun et al., 1992; García and Baselga, 2002a in rabbits; Bakker et al., 1978; Gion et al., 1990 in mice). This phenomenon, in addition to the fact that ovulation rate presents a higher heritability than litter size (Blasco et al., 1993b), and that both traits were correlated, led to propose selection for ovulation rate as an indirect way to improve litter size. Moreover, ovulation rate sets the upper limit for litter size and it could be easily counted by laparoscopy, laparotomy or after slaughter. The first experiments of selection for ovulation rate were proposed in mice by Bradford (1969) and Land and Falconer (1969) and in pigs by Zimmerman and Cunningham (1975). There are six selection experiments for ovulation rate in prolific species, three in pigs (Cunningham et al., 1979; Leymaster and Christenson, 2000; Rosendo et al., 2007), two in mice (Bradford,1969 and Land and Falconer, 1969) and only one selection experiment for ovulation rate has been carried out in rabbits (Laborda et al., 2011, 2012). The estimated responses to selection in these experiments are summarized in Table 1.3. In these experiments, ovulation rate responded to selection but no correlated Chapter 1 10 Table 1.2a: Estimated heritabilities of ovulation rate (OR), litter size (LS), and prenatal survival (PS) and phenotypic and genetic correlations between these traits and litter size (LS) at the day of gestation indicated (DG) in mice, pigs and rabbits. Heritability Phenotypic correlation Genetic correlation Species DG OR LS PS OR, LS OR, PS PS, LS OR, LS OR, PS, LS Land and Falconer, 1969 Mice - 0.31 - - - - - - - - Bradford, 1969 Mice - 0.10 - - - - - - - - Clutter et al., 1990aMice 17 0.33 0.18 0.15 0.45 -0.04 0.86 0.81 0.06 0.60 Long et al., 1991 Mice Birth 0.18 (0.07) 0.33 (0.13) - - - - 0.62 (0.24) - - Young et al., 1977 Pigs 30 0.21 (0.20) 0.39 (0.17) - - - - - -0.26 - Young et al., 1978 Pigs Birth 0.59 (0.12) 0.72 (0.22) - 0.06 - - -0.01 (0.46) - - Cunningham et al., 1979 Pigs Birth 0.42 (0.06) - - - - - 0.07 - - Bolet et al., 1989 Pigs Birth 0.21 (0.12) 0.03 (0.08) - - - - 0.85 - - Bidanel et al. 1992 Pigs 30 0.11 (0.02) -0.03 (0.03) 0.41 (0.04) -0.13 (0.04) 0.87 (0.01) 0.98 (0.33) -0.13 0.99 Haley and Lee, 1992 Pigs Birth 0.30 (0.10) 0.09 (0.06) 0.00 0.21 (0.05) - 0.28 (0.05) 0.87 (0.01) 0.98 (1.00) * * General Introduction 11 Table 1.2b: Estimated heritabilities of ovulation rate (OR), litter size (LS), and prenatal survival (PS) and phenotypic and genetic correlations between these traits and litter size (LS) at the day of gestation indicated (DG) in mice, pigs and rabbits (continuation of Table 1.2a). * Not estimated because the estimate of the heritability of PS was zero. aStandard errors range from 0.05 to 0.06 for the heritabilities and from 0.06 to 0.66 for the genetic correlations. Litter size was estimated as the number of fetuses at d 17 of gestation. b They measure prenatal loss instead of prenatal survival. cStandard errors range from 0.01 to 0.03 for the heritabilities and from 0.03 to 0.13 for the correlation. Modified from Laborda (2011). Heritability Phenotypic correlation Genetic correlation Species DG OR LS PS OR, LS OR, PS PS, LS OR, LS OR, PS PS, LS Bidanel et al., 1996 Pigs 30 0.27 (0.02) - 0.08 (0.03) - - 0.12 (0.04) - - - 0.11 (0.15) - Johnson et al., 1999 Pigs 50 0.24 0.16 0.14 0.03 -0.47 0.48 0.24 -0.86 0.36 Ruiz-Flores and Johnson, 2001 bPigs Birth 0.42 (0.06) 0.18 (0.08) 0.12 (0.09) 0.16 0.59 -0.69 0.52 0.83 -0.04 Rosendo et al., 2007 cPigs Birth 0.34 - 0.14 0.06 -0.18 0.82 0.41 -0.26 0.66 Blasco et al., 1993a Rabbits Birth 0.21 (0.11) 0.27 (0.21) 0.23 (0.10) 0.25 (0.06) -0.30 (0.05) 0.84 (0.02) 0.36 (0.31) -0.14 (0.35) 0.87 (0.08) Bolet et al., 1994 Rabbits - 0.24 (0.04) 0.11 (0.03) - - - - - - - Chapter 1 12 response on litter size at birth was obtained. The lacking correlated response in litter size was associated with an increase in prenatal mortality. There is little information about the timing of prenatal mortality in the experiments of selection for ovulation rate, probably due to the difficulties in measuring the number of fetuses in live animals in pigs and mice. In all cases, fetal survival has decreased with selection for ovulation rate in these three species. Table 1.3: Direct response in ovulation rate (OR) and correlated responses in total number born (TNB) and prenatal survival (PS) with their standard errors (SE) estimated per generation in the experiments of selection for OR in mice and pigs. Species G Response in OR (SE) Response in TNB (SE) Response in PS (SE) Land and Falconer, 1969 Mice 12 0.40 bno clear changes bBradford, 1969 Mice 11 0.26 (0.11) a 0.07 (0.05) a - 0.12 b 0.02 b - 0.7% b Cunningham et al., 1979 Pigs 91 0.38 (0.08) a,1 0.15 (0.13) a - 0.49 (0.10) b,1 0.06 (0.07) b-1.6% (0.5%) b,2 Leymaster and Christenson, 2000 Pigs 10 0.29 b0.06 bRosendo et al., 2007 Pigs 6 0.49 (0.10) c 0.08 (0.11) c -1.0% (0.9%) c 0.51 (0.10) b 0.06 (0.11) b -1.6% (0.9% b G: number of generations; Parity: parity number for litter size; a Regression of line means on generation number; bResponse estimated with a control population; cREML estimate 1Johnson et al., 1984, responses estimated at generation 10. 2Geisert et al., 1978: response per generation in survival at d 30 and at d 70, 0.5% and 1.1%, respectively. Modified from Laborda (2011). 2.2 Selection for prenatal survival There are two experiments of selection for prenatal survival in polytocous species, one in pigs (Rosendo et al., 2007) and the other one in mice (Bradford, 1969). In pigs, the selection criterion was the average prenatal survival over the first two parities corrected for ovulation rate (prenatal survival + 0.018 x ovulation rate). In mice, selection was General Introduction 13 based on [(number of normal fetuses at d 16 / ovulation rate) x number of normal fetuses at d 16]. The number of normal fetuses at d 16 was used as an estimator of litter size at birth. The objective in both experiments was to select for prenatal survival avoiding selection against ovulation rate. Responses to selection in pigs and mice are presented in Table 1.4. Selection for prenatal survival increased litter size both in pigs and mice compared to a control line, and a correlated response in ovulation rate was observed in mice. In mice, the increases in ovulation rate and litter size in the line selected for prenatal survival nearly equalled those of two contemporarily lines directly selected for ovulation rate and for litter size, respectively. In pigs, it is not possible to determine if the estimated response was higher than response to direct selection for litter size due to the high standard error of the estimate and to the absence of a contemporary line selected for litter size. Summarizing, selection for prenatal survival increased litter size, but it was not more effective than direct selection for litter size. Table 1.4: Responses in prenatal survival (PS), ovulation rate (OR) and litter size (LS) estimated per generation in pigs and mice selected for prenatal survival, with their standard errors (in parenthesis). Species Pigs Mice Generations 6 11 Method Control population 1REML 1Control population 2Regression 2* RESPONSE PS (%) 1.0 (0.9) 0.8 (0.9) 0.8 0.4 (0.4) OR (ova) 0.04 (0.11) 0.11 (0.11) 0.15 0.23 (0.09) LS (kits) 0.21 (0.11) 0.24 (0.11) 0.20 0.25 (0.06) 1 Rosendo et al., 2007; 2 Bradford, 1969 * Regression of generation mean on generation number. Modified from Laborda (2011). 2.3 Selection for uterine capacity Selection for increased uterine capacity has been proposed as an indirect way of improving litter size (Bennett and Leymaster, 1989, 1990). In rabbits, Blasco et al. (1994) proposed using unilateral ovariectomy to measure uterine capacity. Chapter 1 14 There is little information on genetic parameters of uterine capacity. Heritability of uterine capacity was low (0.05, and 0.11 in rabbits reported by Bolet et al. (1994) and Blasco et al. (2005) respectively; 0.08 in mice (Kirby and Nielsen, 1993). In pigs, there is only one experiment of selection for uterine capacity (Leymaster and Christenson, 2000) and their results have not been fully published yet. There are three more experiments of selection for uterine capacity: two experiments of divergent selection in rabbits (first experiment: Bolet et al., 1994; Santacreu et al., 1994; second experiment: Blasco et al., 2005; Mocé et al., 2005; Santacreu et al., 2005), and one experiment in mice (Clutter et al., 1990; Gion et al., 1990; Kirby and Nielsen, 1993). The estimated responses to selection in these experiments are summarized in Table 1.5. In rabbits, in the first experiment (Bolet et al., 1994), selection was performed on number of dead fetuses from implantation to birth. After 4 generations of selection it was observed that the number of dead fetuses did not change and no significant response was obtained in litter size and its components. The second experiment consisted in selection on litter size in unilateral ovariectomized females, which includes both embryo and fetal survival (Blasco et al., 2005; Mocé et al., 2005; Santacreu et al., 2005). After 10 generations of selection for uterine capacity, correlated response to selection in litter size was not symmetric and a response was detected in the low line. A divergence of 2.35 kits was found between the high and low lines, mainly because of a higher correlated response in the low line. In mice, Gion et al. (1990) and Kirby and Nielsen (1993) found a favorable correlated response in litter size when selecting for high uterine capacity, but selection for uterine capacity was not more effective than direct selection for litter size. In conclusion, direct responses to increase uterine capacity and correlated responses in litter size were low or close to zero in rabbits and mice. 2.4 Index selection for components of litter size: Ovulation rate and prenatal survival Cunningham et al. (1979) suggested that litter size could be regarded as a natural index of ovulation rate and embryonic survival. Johnson et al. (1984) used this idea to develop a model in which litter size is determined by the product of ovulation rate and embryonic survival and an index was constructed to optimize weights on component traits. Selection on the optimum index was predicted to increase ovulation rate General Introduction 15 restricting the decrease in embryonic survival and to increase litter size more than direct selection. Table 1.5: Responses to selection to increase uterine capacity (UC) and correlated responses in ovulation rate (OR), litter size (LS) and prenatal survival (PS) estimated per generation. Species Rabbits (1 st exp.) Rabbits (2 nd exp.) Pigs Mice Generations 10 4 11 13 6, 21 7 Method Control Population1,2 Genetic Trends3Genetic Trends 4Control population5Control population RESPONSE UC (kits) - 0.01 1 0.08 - 0.15 0.11 0.10 (0.02) OR (ova) -0.03 20.03 -0.3 0.00 0.03 6 LS (kits) 0.05 2- - 0.08 0.00 7 PS (%) 0.5 20.4 0 - 0.3 6 1 Mocé et al., 2005; 2 Santacreu et al., 2005; 3 Blasco et al., 2005; 4 data calculated from results presented in Santacreu et al., 1994, assuming a symmetric response; 5 Leymaster and Christenson, 2000; 6 Gion et al., 1990; 7 Kirby and Nielsen, 1993; *Standard errors (SE) in parenthesis. Modified from Laborda (2011). To our knowledge, there are only two experiments of selection for an index of ovulation rate and prenatal survival: one in pigs (Johnson et al., 1984; Neal et al., 1989; Bennett and Leymaster, 1989, 1990; Casey et al., 1995; Johnson et al., 1999), and the other one in mice (Clutter et al., 1990; Gion et al., 1990; Kirby and Nielsen, 1993; Ribeiro et al., 1997a, b). Selection was efficient in increasing litter size when compared to the control line (Table 1.6) but this response was similar to the observed response to direct selection for litter size in other experiments. In pigs, the index was recalculated during the experiment to optimize response to selection. In mice, response to selection was estimated by comparison with a control line and with a line selected for litter size (Gion et al., 1990). As in pigs, litter size increased with selection compared with the control line (Table 1.6), but it increased at a similar rate to the line selected for litter size. The increase in litter size in the mice line selected for the index was due to a higher ovulation rate and prenatal survival in the selected line than in the control line. The index was used along Chapter 1 22 Gion, J. M., Clutter, A. C., and Nielsen, M. K. 1990. Alternative methods of selection for litter size in mice: II. Response to thirteen generations of selection. J. Anim. Sci. 68: 3543-3556. Gómez, E. A., Rafel, O., Ramón, J., and Baselga, M. 1996. A genetic study of a line selected on litter size at weaning. In Proc. 6th World Rabbit Congress, Toulouse, France 2: 289-292. Hafez, E. S. E. 1969. Fetal survival in undercrowded and over crowded unilaterally pregnant uteri in the rabbit. VI Congr. Reprod. Anim. Paris. Francia. 1:575. Haley, C. S., and Lee, G. J. 1992. Genetic factors contributing to variation in litter size in British Large White gilts. Livest. Prod. Sci. 30: 99-113. Hirsch, E., Otto, T., Blanchard, R., and Rosenberg, J. O. 1999. Mouse laparoscopy. J. Am. Assoc. Gynecol. Laparosc. 6(2): 173-177. Holl, J. W., and Robison, O. W. 2003. Results from nine generations of selection for increased litter size in swine. J. Anim. Sci. 81: 624-629. INRA-SAGA, Institut National de la Recherche Agronomique-Station d’amélioration génétique des animaux. http://www.avicampus.fr/PDF/PDFlapin/selectionlapin1.pdf. Johnson, R. K., Nielsen, M. K., and Casey, D. S. 1999. Responses in ovulation rate, embryonal survival and litter traits in swine to 14 generations of selection to increase litter size. J. Anim. Sci. 77:541-557. Johnson, R. K., Zimmerman, D. R., and R. J. Kittok. 1984. Selection for components of reproduction in swine. Livest. Prod. Sci. 11: 541-558. Joakimsen, O., and Baker, R.L. 1977. Selection for litter size in mice. Acta Agriculturae Scandinavica. 27: 301-318. Khalil, M. H., and Al-Saef, A. M. 2008. Methods, criteria, techniques and genetic responses for rabbit selection: a review. 9th World Rabbit Congress, Verona, Italy. Pages 1-22 Kirby, Y. K., and Nielsen, M. K. 1993. Alternative methods of selection for litter size in mice: III. Response to 21 generations of selection. J. Anim. Sci. 71: 571-578. General Introduction 23 Koenig, J. L. F., Zimmerman, D. R., Eldrige, F. E., and Kopf, J. D. 1986. The effect of superovulation and selection for high ovulation rate on chromosomal abnormalities in swine ova. J. Anim. Sci. 63(Suppl.1):202 (Abstr.). Krisher, R.L. 2004. The effect of oocyte quality on development. J. Anim. Sci. 82 (E. Suppl.): E14-E23. Laborda, P. 2011. Selection for ovulation rate in rabbits. Universidad Politécnica de Valencia. Doctoral thesis. Laborda, P., Mocé, M. L., Blasco, A., and Santacreu, M. A. 2012. Selection for ovulation rate in rabbits: genetic parameters and correlated responses on survival rates. J. Anim. Sci. 90:439-446. Laborda, P., Mocé, M. L., Santacreu, M. A., and Blasco, A. 2011. Selection for ovulation rate in rabbits: Genetic parameters, direct response, and correlated response on litter size. J. Anim. Sci. 89: 2981-2987. Land, R. B., and Falconer, D. S. 1969. Genetic studies of ovulation rate in the mouse. Genet. Res. 13: 25-46. Leymaster, K. A., and Christenson, R. K. 2000. Direct and correlated responses to selection for ovulation rate or uterine capacity in swine. J. Anim. Sci. 78(Suppl. 1):68. (Abstr.). Long, C. R., Lamberson, W. R., and Bates, R. O. 1991. Genetic correlations among reproductive traits and uterine dimensions in mice. J. Anim. Sci. 69: 99-103. Luberda, Z., 2005. The role of glutathione in mammalian gametes. Biol. Reprod. 5: 517. Mocé, M. L., Santacreu, M. A., Climent, A., and Blasco, A. 2005. Divergent selection for uterine capacity in rabbits. III. Responses in uterine capacity and its components estimated with a cryopreserved control population. J. Anim. Sci. 83: 2308-2312. Mocé, M. L., Blasco, A., and Santacreu, M.A. 2010. In vivo development of vitrified rabbit embryos: Effects on prenatal survival and placental development. Theriogenology 73: 704-710. Mocé, M. L. and Santacreu, M. A. 2010. Genetic improvement of litter size in rabbits: a review. CD-Proc. 9th World Congr. Genet. Appl. Livest. Prod. Leipzig, Germany. Chapter 1 24 Neal, S. M., Johnson, R. K., and Kittok, R. J. 1989. Index selection for components of litter size in swine: Response to five generations of selection. J. Anim. Sci. 67: 1933-1945. Ollivier, L., and G. Bolet. 1981. Selection for prolificacy in the pig: results of a ten generation selection experiment. Journee Rech. Porc. France 13: 261-268. Pascual, M., Serrano, P., Torres, C., and Gomez, E. 2011. Algunos conceptos para la mejora de la rentabilidad en explotaciones cunículas. 36 Symposium de cunicultura de ASESCU, Peñíscola, Spain. Pages 18-23. Peiró, R., Santacreu, M. A., Climent, A., and Blasco, A. 2007. Early embryonic survival and embryo development in two divergent lines of rabbits selected for uterine capacity. J. Anim. Sci. 85: 1634-1639. Pope, W. F. 1988. Uterine Asynchrony: A Cause of Embryonic Loss. Biol. of Reprod. 39: 999-1003. Pope, W. F., Xie, S., Broermann, D. M., and K. P. Nephew. 1990. Causes and consequences of early embryonic diversity in pigs. J. Reprod. Fertil. Suppl. 10: 251260. Rausell, F., and Tarín, J. J. 2005. Función del glutatión reducido durante la maduración y fecundación de ovocitos y desarrollo pre-implantatorio de embriones in vitro de mamíferos. Revista Iberoamericana de Fertilidad. Vol. 22nº 6. REGA, Registro General de Explotaciones Ganaderas. 2010. http://www.marm.es/app/vocwai/documentos/Adjuntos_AreaPublica/INDICADOR ES%20ECON%C3%93MICOS%20SECTOR%20CUN%C3%8DCOLA%202010.p df Revelli, A., Delle Piane, L., Casano, S., Molinari, E., Massobrio, M., and Rinaudo, P. 2009. Follicular fluid content and oocyte quality: from single biochemical markers to metabolomics. Reprod. Biol. Endocr. 7: 40-53. Ribeiro, E. L., Nielsen, M. K., Bennett, G. L., and Leymaster, K. A. 1997a. A simulation model including ovulation rate, potential embryonic viability, and uterine capacity to explain litter size in mice: I. Model development and implementation. J. Anim. Sci. 75: 641-651. General Introduction 25 Ribeiro, E. L., Nielsen, M. K., Leymaster, K. A., and Bennett, G. L. 1997b. A Simulation Model Including Ovulation Rate, Potential Embryonic Viability, and Uterine Capacity to Explain Litter Size in Mice: II. Responses to Alternative Criteria of Selection. J. Anim. Sci. 75: 652-656. Rochambeau, H., Duzert, R., and Tudela, F. 1998. Long term selection experiment in rabbit. Estimation of genetic progress on litter size at weaning. Proc. 6th World Congr. Genet. Appl. Livest. Prod., Armidale, Australia, Vol. 26. Rosell, J. M. 2000. Biología. In: Enfermedades del conejo. Ed. Mundi-Prensa, Madrid, España. Pages 55-127. Rosendo, A., Druet, T., Gogue, J., and Bidanel, J. P. 2007. Direct responses to six generations of selection for ovulation rate or prenatal survival in Large White pigs. J. Anim. Sci. 85: 356-364. Rothschild, M. F., and Bidanel, J. P. 1998. Biology and genetics of reproduction. In: The Genetics of the Pig. M.F. Rothschild and A. Ruvinsky (eds.) Wallingford (UK): CAB International. Pages 313-343. Ruiz-Flores, A., and Johnson, R. K. 2001. Direct and correlated responses to two-stage selection for ovulation rate and number of fully formed pigs at birth in swine. J. Anim. Sci. 79: 2286-2299. Santacreu, M. A., Viudes, M. P., and Blasco, A. 1990. Evaluation par coelioscopie des corps jaunes et des embryons. Influence sur la taillé de portée chez la lapine. Reprod. Nutr. Dev. 30: 583-588. Santacreu, M. A., Argente, M. J., Climent, A., Blasco, A., and Bolet, G. 1994. Divergent selection for uterine efficiency in unilaterally ovariectomized rabbits. II. Response to selection. Proc. 5th World Congr. Genet. Appl. Livest. Prod. 19: 265267. Santacreu, M. A., Mocé, M. L., Climent, A., and Blasco, A. 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. Santacreu, M. A. 2006. La supervivencia prenatal en la coneja reproductora. XXXI. Symposium de cunicultura de ASESCU, Lorca, Murcia, Spain. Vol. I: 229-236. Chapter 1 26 Soede, N. M., Wetzels, C. C. H., Zondag, W., de Koning, M. A. I., and Kemp, B. 1995. Effects of time of insemination relative to ovulation, as determined by ultrasonography, on fertilization rate and accessory sperm count in sows. J. Reprod. Fertil. 104: 99-106. Theau-Clement, M., Salvetti, P., Bolet, G., Saleil, G., and Joly, T. 2009. Influence de l´intervalle entre le sevrage et l´insemination sur la production d´embryons et leur qualité chez la lapine. 13émes Journées de la Recherche Cunicole. INRA-ITAVI, Le Mans, France. Pages 125-128. Torres, S. 1982. Etude de la mortalité embryonnaire chez la lapine. III. Journées de la Recherche Cunicole. Paris. Francia. N° 15. Torres, S., Hulot, F., and Meunier, M. 1984. Étude comparée du développement et de la mortalité embryonnaire chez deux genotypes de lapines. 3rd World Rabbit Congress. Rome, Italy 2: 417-425. Vallet, J. L., Christenson, R. K., Trout, W. E., and Klemcke, H. G. 1998. Conceptus, progesterone, and breed effects on uterine protein secretion in swine. J. Anim. Sci. 76: 2657-2670. Wilde, M. H., Xie, S., Day, M. L., and Pope, W. F. 1988. Survival of small and large littermate blastocysts in swine after synchronous and asynchronous transfer procedures. Theriogenology 30: 1069-1074. Wilmut I., Sales, D. I., and Ashworth, C. J. 1986. Maternal and embryonic factors associated with prenatal loss in mammals. J. Reprod. Fert. 76: 851-864. Xie, S., Broermann, D. M., Nephew, K. P., Geisert, R. D., and Pope, W. F. 1990. Ovulation and Early Embryogenesis in Swine. Biol. Reprod. 43: 236-240. Young, L. D., Johnson, R. K., and Omtveldt, I. T. 1977. An analysis of the dependency structure between a gilt's prebreeding and reproductive traits. I. Phenotypic and genetic correlations. J. Anim. Sci. 44: 557-564. Young L. D., Pumfrey, R. A., Cunningham, P. J., and Zimmerman, D. R. 1978. Heritabilities and genetic and phenotypic correlations for prebreeding traits, reproductive traits and principal components. J. Anim. Sci. 46: 937-949. General Introduction 27 Zimmerman, D. R., and Cunningham, P. J. 1975. Selection for ovulation rate in swine: population, procedures and ovulation response. J. Anim. Sci. 40: 61-69. 28 Chapter 2 Objectives C. Ziadi Instituto de Ciencia y Tecnología Animal, Universidad Politécnica de Valencia, 46071 Valencia. Spain. 29 Objectives The objectives of this Thesis are: 1. To study the phenotypic and genetic parameters of ovulation rate, litter size, embryonic, fetal and prenatal survival rates in a rabbit population selected for six generations for ovulation rate and then for seven generations for both ovulation rate and litter size. 2. To estimate genetic responses to selection for ovulation, litter size, implanted embryos and survival rates in the same rabbit line. 30 Chapter 3 Genetic selection for litter size and ovulation rate in rabbits: estimation of genetic parameters, direct and correlated responses C. Ziadi Instituto de Ciencia y Tecnología Animal, Universidad Politécnica de Valencia, 46071 Valencia. Spain. Chapter 3 31 Abstract The aim of this work was to estimate direct and correlated responses in survival rates in an experiment of selection for ovulation rate and litter size in rabbits (OR_LS line). The experiment consisted of 2 periods of selection. In period 1, selection was performed for ovulation rate during 6 generations. In period 2, line underwent a two-stage selection for ovulation rate and litter size during 7 generations. Selection in period 1 was based on the phenotypic value of ovulation rate estimated at d 12 of gestation by laparoscopy. Two-stage selection was based on the phenotypic value of ovulation rate and the average litter size over the first two parities. Total selection pressure was about 30%. The line had approximately 17 males and 75 females per generation. Traits recorded were: ovulation rate (OR) estimated as the number of corpora lutea in both ovaries; number of implanted embryos (IE), estimated as the number of implantation sites; litter size (LS), estimated as total number of rabbits born recorded at each parity; embryo survival (ES) estimated as IE/OR, fetal survival (FS) estimated as LS/IE, and prenatal survival (PS) estimated as LS/OR. Data were analyzed using Bayesian methodology. The estimated heritabilities of LS, OR, IE, ES, FS and PS were 0.07, 0.21, 0.10, 0.07, 0.12 and 0.16 respectively. The estimated repeatabilities of LS, OR, IE and ES were 0.16, 0.27, 0.20 and 0.14 respectively. In the first period of selection, OR increased 1.36 ova in 6 generations, but no correlated response was observed in LS due to a decrease on fetal survival. Correlated responses for implanted embryos, embryo, fetal and prenatal survival in the first selection period were 1.11, 0.00, -0.04 and -0.01 respectively. After 7 generations of two-stage selection for ovulation rate and litter size, OR increased 1.0 ova and correlated response on LS was 0.9 kits. Correlated responses for implanted embryos, embryo, fetal, and prenatal survival in the second selection period were 1.14, 0.02, 0.02, and 0.07 respectively. Two-stage selection for ovulation rate and litter size could be a promising procedure to improve litter size in rabbits. Key words: Rabbit, two-stage selection, litter size, ovulation rate, survival rates. Selection for OR and LS 38 Table 3.2. Means and SD (in parentheses) for ovulation rate (OR), litter size (LS), number of implanted embryos (IE), embryo survival (ES), fetal survival (FS) and prenatal survival (PS) in generations 0 to 13. Generation 0 1 2 3 4 5 6 7 8 9 10 11 12 13 N85 75 92 80 65 59 102 67 82 74 62 49 76 76 S *,a 2.71 2.43 3.65 2.91 3.85 1.65 1.65 1.51 1.15 1.55 1.01 1.53 1.76 - S *,b 0.44 0.68 0.65 0.37 0.09 0.23 2.13 2.34 2.41 1.86 2.64 1.94 2.50 - ORa14.9 (2.2) 15.5 (2.7) 15.8 (2.6) 16.4 (2.4) 15.8 (2.7) 15.5 (2.4) 16.3 (2.3) 15.9 (2.5) 15.7 (2.4) 15.5 (2.7) 16.4 (2.4) 16.1 (2.5) 16.5 (2.9) 16.1 (2.6) LSb8.1 (3.0) 8.5 (2.6) 9.1 (2.8) 9.1 (3.0) 8.6 (2.9) 8.7 (3.1) 9.3 (2.9) 9.1 (3.3) 8.8 (3.5) 9.4 (3.0) 9.4 (3.5) 9.6 (3.1) 9.1 (3.1) 9.6 (3.2) IEc12.5 (3.1) 12.6 (3.6) 12.5 (3.6) 12.1 (3.6) 11.1 (4.1) 11.5 (4.3) 13.1 (3.4) 12.1 (4.1) 12.5 (4.0) 12.9 (3.3) 12.6 (4.5) 11.5 (3.7) 12.4 (3.7) 11.5 (4.4) ES 0.82 (0.18) 0.81 (0.20) 0.79 (0.20) 0.75 (0.21) 0.71 (0.24) 0.74 (0.26) 0.79 (0.18) 0.75 (0.23) 0.76 (0.21) 0.82 (0.18) 0.76 (0.24) 0.73 (0.24) 0.76 (0.20) 0.71 (0.26) FS 0.72 (0.19) 0.73 (0.20) 0.78 (0.17) 0.68 (0.22) 0.75 (0.16) 0.69 (0.18) 0.74 (0.14) 0.75 (0.19) 0.75 (0.18) 0.79 (0.15) 0.80 (0.15) 0.79 (0.14) 0.74 (0.16) 0.79 (0.17) PS 0.59 (0.19) 0.57 (0.18) 0.62 (0.20) 0.51 (0.22) 0.58 (0.19) 0.55 (0.21) 0.58 (0.17) 0.58 (0.19) 0.57 (0.19) 0.66 (0.18) 0.61 (0.22) 0.58 (0.19) 0.57 (0.19) 0.63 (0.24) N: number of females at each generation. S*: Selection differential applied to animals at generations 0 and consecutively to the other generations for ovulation rate [superscript a] and litter size [superscript b]. a Unit = ova. bUnit = kits. cUnit = embryos. Chapter 3 39 OR, recent studies in pigs reported higher heritability estimates than in rabbits (RuizFlores and Johnson, 2001; Rosendo et al., 2007). The repeatability estimate of LS was 0.16 with HPD95% [0.13, 0.20] (Table 3.3). Repeatability estimate for LS agrees with estimates reported for a maternal line by Khalil (1993), but are lower than the ones reported by Lukefahr and Hamilton (1997) (r = 0.23) and Rastogi et al. (2000) (r = 0.30). Ovulation rate and IE had a moderate repeatabilities estimates (0.27 and 0.20 for OR and IE respectively). The repeatability estimate of ES was 0.14 with HPD95% [0.08, 0.21] (Table 3.3). No repeatability or p2 estimates for the traits IE and ES have been reported in the literature. These repeatability estimates lead to an estimated ratio of the permanent environmental variance to the phenotypic variance (p2) of 0.09, 0.06, 0.07 and 0.07 for LS, OR, IE and ES, respectively. Our estimates of p2are inside the range reported for litter size in rabbits (reviewed by Garreau et al., 2004). Table 3.3. Features of the marginal posterior distributions of the heritability (h2) and the repeatability (r) of litter size (LS), ovulation rate (OR), number of implanted embryos (IE), embryo survival (ES) and the heritability of fetal survival (FS) and prenatal survival (PS). Traits h2HPD95%(h2) P0.10 k r HPD95%(r) LS 0.07 0.02, 0.12 0.16 0.03 0.16 0.13, 0.20 OR 0.21 0.13, 0.29 1.00 0.14 0.27 0.21, 0.35 IE 0.10 0.05, 0.17 0.60 0.06 0.20 0.14, 0.26 ES 0.07 0.02, 0.12 0.19 0.03 0.14 0.08, 0.21 FS 0.12 0.06, 0.21 0.69 0.07 - - PS 0.16 0.10, 0.20 0.99 0.11 - - HPD95%: high posterior density interval at 95%. P0.10: probability of the heritability being higher than 0.10. k: limit for the interval [k, +∞) of the heritability having a probability of 95%. Features of the estimated marginal posterior distributions of the genetic correlations are summarized in Table 3.4. The estimate of the genetic correlation between LS and OR was positive (P = 0.92; Table 3.4), but imprecise (HPD95%, Table 3.4). Estimated genetic correlations between LS and the remaining traits were positive (value P; Table 3.4) being moderate with ES and FS and high with PS with a probability of 95% of being at least 0.78 (value k; Table 3.4). Estimated genetic correlations of OR with FS Selection for OR and LS 40 and PS were negative (P = 1.00; Table 3.4), and nothing can be said about the sign of the estimated genetic correlation between OR and ES because it was imprecise. Genetic correlation between OR and LS was in agreement with values reported by other authors (Ruiz-Flores and Johnson, 2001; Rosendo et al., 2007 in pigs; Blasco et al., 1993a and Laborda et al., 2011a). Estimated values of the genetic correlation between OR and LS founded in literature was generally positive, but they were reported without or with high standard errors. Higher genetic correlations between LS and IE were obtained in other experiments in rabbits and pigs, possibly because the number of fetuses was measured at a later point of gestation (Blasco et al., 1993a in rabbits; Johnson et al., 1999 in pigs). The positive genetic correlations between LS and survival rates agree with estimates in the literature (Blasco et al., 1993; Argente et al., 1997; Laborda et al., 2012 in rabbits; Rosendo et al., 2007 in pigs). Besides, the genetic correlation between OR and IE was in accordance with the ones obtained in rabbits (0.58 by Laborda et al., 2012), pigs (0.44 by Johnson et al., 1999) and mice (0.81 by Clutter et al., 1990). Genetic correlations between all traits were estimated with low precision, especially for the genetic correlations between LS with both OR and ES, and OR with ES which have a very large interval of confidence. To obtain estimated genetic parameters with high precision, a large set of data would be needed. Nevertheless, the nature of this kind of experiments, which need techniques such as laparoscopy or slaughter the female to measure ovulation rate prevents from collecting a large number of data making the estimation of precise genetic correlations difficult. Although in this study, genetic parameters were estimated with limited database and low precision, they are within the range of the values reported in the literature. Features of the estimated marginal posterior distributions of the phenotypic correlations are summarized in Table 3.5. Phenotypic correlation between LS and OR was positive (P = 1.00, Table 3.5) but low. Estimated phenotypic correlation between LS and IE was high and positive (P = 1.00, Table 3.5). The posterior mean of phenotypic correlation Chapter 3 41 Table 3.4. Features of the marginal posterior distributions of the genetic correlation between the traits analyzed: litter size (LS), ovulation rate (OR), number of implanted embryos (IE), embryo survival (ES), fetal survival (FS) and prenatal survival (PS). Traits mean median HPD95% P k LS, OR 0.30 0.30 -0.12, 0.71 0.92a-0.05a LS, IE 0.66 0.68 0.34, 0.99 1.00a0.34ª LS, ES 0.54 0.59 -0.04, 0.95 0.94a-0.04a LS, FS 0.63 0.63 0.34, 0.96 1.00a0.34a LS, SP 0.85 0.86 0.77, 0.91 1.00a0.78a OR, IE 0.70 0.72 0.44, 0.93 1.00a0.46a OR, ES -0.09 -0.09 -0.53, 0.34 0.67b0.29b OR, FS -0.53 -0.50 -0.82, -0.27 1.00b-0.30b OR, PS -0.35 -0.35 -0.62, -0.09 0.99b-0.12b HPD95%: high posterior density interval at 95%; P: probability of the genetic correlation being greater than zero (superscript a), or less than zero (superscript b); k: limit for the interval a[k, +∞), b(-∞, k], having a probability of 95%. Table 3.5. Features of the marginal posterior distributions of the phenotypic correlation between the traits analyzed: litter size (LS), ovulation rate (OR), number of implanted embryos (IE), embryo survival (ES), fetal survival (FS) and prenatal survival (PS) Traits mean median HPD95% P k LS, OR 0.19 0.20 0.13, 0.26 1.00a0.14a LS, IE 0.72 0.72 0.70, 0.75 1.00a0.70ª LS, ES 0.67 0.68 0.64, 0.71 1.00a0.65a LS, FS 0.49 0.49 0.45, 0.54 1.00a0.45a LS, PS 0.88 0.88 0.87, 0.89 1.00a0.87a OR, IE 0.40 0.40 0.35, 0.45 1.00a0.36ª OR, ES -0.09 -0.09 -0.15, -0.04 1.00b-0.05b OR, FS -0.23 -0.23 -0.30, -0.17 1.00b-0.18b OR, PS -0.25 -0.25 -0.31, -0.20 1.00b-0.20b HPD95%: high posterior density interval at 95%; P: probability of the genetic correlation being greater than zero (superscript a), or less than zero (superscript b); k: limit for the interval a[k, +∞), b(-∞, k], having a probability of 95%. Selection for OR and LS 42 between OR and IE (Table 3.5) had similar magnitude and sign than the ones obtained in pigs and mice. Estimated phenotypic correlations between OR and survival rates were negative (P = 1.00, Table 3.5); however they were of low magnitude, especially the phenotypic correlation between OR and ES. Phenotypic correlations between LS and survival rates were positive (P = 1.00, Table 3.5), being moderate with ES and FS and high with PS. Positive correlations between LS and survival rates and negative correlations between OR and both FS and PS were in agreement with the estimates founded in the literature (Blasco et al., 1993a; Blasco et al., 1993b for a review in rabbits; Johnson et al., 1999; Rosendo et al., 2007 in pigs). 2. Response to selection In each period of selection, total responses to selection for all traits were estimated by the difference of line means between first and last generations. The estimated responses to selection for OR, LS and IE are shown in Figure 3.1. The correlated responses in ES, FS, and PS are shown in Figure 3.2. We can distinguish two periods of genetic responses. 2.1 Selection for ovulation rate After six generations of selection, OR increased in 1.36 ova, almost 1.5% per generation (0.22 ova/generation, Figure 3.1). In this experiment, selection for OR did not practically modify LS; correlated response in LS was 0.30 kits in 6 generations (0.05 kit/generation, Figure 3.1). Thus, only 22 % of the average increase in ovulation rate was realized as more kits at birth. Implanted embryos increased 1.11 embryos in 6 generations (0.18 embryos/ generation, Figure 3.1). Prenatal survival apparently showed a little decrease (0.013 in 6 generations, Figure 3.2). We did not observe any response in ES, but FS decreased consistently (0.038 in 6 generations, around 0.9 % per generation, Figure 2). Thus, this decrease in fetal survival seems to be responsible for the lack of correlated response observed in litter size. Our results are in agreement with estimated responses published by Laborda et al. (2011, 2012a, b) using data of 10 generations of selection for OR (line OR). Our estimated response of OR was similar to the ones reported in pigs by Leymaster and Christenson (2000) and in mice by Bradford (1969), but lower that those obtained in other studies (Cunningham et al., 1979; Rosendo et al., 2007 in pigs; Land and Chapter 3 43 Falconer, 1969 in mice). In these studies, the correlated response on litter size was close to zero, except the one observed by Cunningham et al. (1979), but it was estimated with a very high standard error (0.15 ± 0.13 pigs/ generation). In all cases, an increase in prenatal mortality was observed. As in this experiment, Freking et al. (2007) in pigs and Bradford (1969) in mice observed that post-implantation losses were the main cause for the uncorrelated response in litter size. Possible physiological causes for the lacking correlated response in litter size were already discussed with details by Laborda et al. (2011, 2012a, b). In conclusion, the results show that selection for ovulation rate could increase fetal mortality, whereas embryo mortality does not seem to have been modified. This fetal mortality has been the main cause for the lacking observed correlated response in litter size. Further studies are needed to explain the mechanisms that have increased fetal mortality in rabbits selected for high ovulation rate. 2.2 Two-stage selection for ovulation rate and litter size In the second period, seven generations of selection for ovulation rate and litter size have been performed. Ovulation rate continued increasing throughout the two-stage selection but with a lower rate than during the first period of selection, due to a decrease on the selection differential applied (Table 3.1). In fact, from generation 6, the proportion of females with extremely high ovulation rate increased 4.1% per generation in line OR against 0.9% in line OR_LS. Response in OR was estimated to be 1 ova, almost 0.9 % per generation (0.14 ova /generation, Figure 3.1). Direct response for LS was approximately 0.9 kits (0.13 kit /generation, Figure 3.1). Thus, around 93 % of the average increase in ovulation rate was realized as more kits at birth. The correlated response in IE was 1.14 embryos (Figure 3.1). Both embryonic and fetal survivals have been shown to contribute with the same amount in the increase observed in prenatal survival. A small positive change in ES and FS was observed (approximately 0.020 in 7 generations, Figure 3.2). Prenatal survival increased 0.077 in 7 generations, around 2 % per generation (Figure 3.2). The direct response in litter size was similar to the response estimated by Lamberson et al. (1991) after direct selection for LS during 8 generations in a line previously selected to increase ovulation rate. In pigs, after 8 generations of two-stage selection Ruiz-Flores and Johnson (2001) obtained greater direct responses in number of fully formed pigs and ovulation rate (0.33 ± 0.06 pig/generation and 0.26 ± Selection for OR and LS 44 0.07 ova/generation respectively). Their estimate of correlated response in prenatal survival was similar to the one observed in this study (0.078 in 7 generations). In the second period of selection, total number of kits born could be a good measurement of uterine capacity, since ovulation rate was high enough due to the direct selection applied during the first period. In populations selected for increasing ovulation rate, the total number of kits born is expected to represent uterine capacity more closely than in unselected populations (Lamberson et al., 1991, Johnson et al., 1999; RuizFlores et al., 2001). Thus, selection in the second period was performed to ameliorate uterine capacity and then indirectly prenatal survival. Observed changes in prenatal survival either happened during pre and post-implantation periods of gestation. In early stages of gestation, an improvement in the quality of oocytes (Torres, 1982 in rabbits; Koenig et al., 1986 in pigs) and lower variability of embryo development (Pope et al., 1988; Xi et al., 1990 in pigs) could explain the increase in embryo survival. To assess oocyte quality in both lines OR and OR_LS, one study was designed by Laborda et al. (2012c) to measure concentrations of ATP and glutathione (GSH), the main compound that protects the cell against the oxidative stress. Their results showed a difference in the concentration of GSH of 0.7 pmol /oocyte between the line OR and the line OR_LS that could indicate a higher number of mature oocytes in the line OR_LS. No difference between lines was found for ATP concentration. Both oocyte quality and variability of embryo development can affect embryonic and fetal survival. In later stages of gestation, an increase in prenatal survival could be associated with more uterine space and resources (Adams, 1960; Hafez, 1969), and more blood supply to the fetuses (Hafez, 1965; Duncan, 1969; Argente et al., 2003 in rabbits). Results from period one of selection show that, after six generations of selection for ovulation rate, ovulation rate responded to selection, but no correlated response on litter size was observed. Results from period two of selection show that, two-stage selection for ovulation rate and litter size would be effective in improving ovulation rate and litter size. Moreover, this increase has been due to reducing both pre and post-implantation mortalities. Chapter 3 45 Figure 3.1: Genetic trends for ovulation rate (OR), litter size (LS), and number of implanted embryos (IE) of line OR_LS. This line was selected for OR from generation 0 to 6, and for OR and LS from generation 7 to 13. Superscript a: mean of the estimated breeding value of the character at generation 6. Superscript b: mean of the estimated breeding value of the character at generation 13. Figure 3.2: Genetic trends for embryo survival (ES), fetal survival (FS) and prenatal survival (PS) of line OR_LS, this line was selected for OR from generation 0 to 6, and for OR and LS from generation 7 to 13. Superscript a: mean of the estimated breeding value of the character at generation 6. Superscript b: mean of the estimated breeding value of the character at generation 13. Selection for OR and LS 46 Implications Summarizing, the results obtained show that two-stage selection for ovulation rate and litter size could be more effective to increase both litter size and prenatal survival than either direct selection for litter size or ovulation rate. To support this hypothesis, results from this study will be compared with a control line which has been vitrified in generation six when the second period of selection was initiated. ACKNOWLEDGEMENTS This study was supported by the Comisión Interministerial de Ciencia y Tecnología CICYTAGL2005-07624-C03-01 CICYTAGL2008-05514-C02-01 and by funds from Generalitat Valenciana research programme (Prometeo 2009/125). Chapter 3 47 Literature cited Adams, C. E. 1960. Studies on prenatal mortality in the rabbit, Oryctolagus cuniculus: The amount and distribution of loss before and after implantation. J. Endocrinol. 19: 325-344. Argente, M. J., Santacreu, M. A., Climent, A., Bole,t G., and Blasco, A. 1997. Divergent selection for uterine capacity in rabbits. J. Anim. Sci. 75: 2350-2354. Argente, M. J., Santacreu, M. A., Climent, A., and Blasco, A. 2000. Genetic correlations between litter size and uterine capacity. 7th World Rabbit Congr. A: 333-338. Argente, M. J., Blasco, A., Ortega, J. A., Haley, C. S., and Visscher, P. M. 2003. Analyses for the presence of a major gene affecting uterine capacity in unilaterally ovariectomized rabbits. Genetics. 163: 1061-1068. Bennett, G. L., and Leymaster, K. A. 1989. Integration of ovulation rate, potential embryonic viability and uterine capacity into a model of litter size in swine. J. Anim. Sci. 67: 1230-1241. Bennett, G. L., and Leymaster, K. A. 1990. Genetic implications of a simulation model of litter size based on ovulation rate, potential embryonic viability and uterine capacity: I. Genetic theory. J. Anim. Sci. 68: 969-979. Blasco, A., Santacreu, M. A., Thompson, R., and Haley, C. S. 1993a. Estimates of genetic parameters for ovulation rate, prenatal survival and litter size in rabbits from an elliptical experiment. Livest. Prod. Sci. 34: 163-174. Blasco, A., Bidanel, J. P., Bolet, G., Haley, C., and Santacreu, M. A. 1993b. The genetics of prenatal survival of pigs and rabbits: a review. Livest. Prod. Sci. 37: 121. Blasco, A., Dando, P., Gogue, J., and Bidanel, J. P. 1996. Relationships between ovulation rate, prenatal survival and litter size in French Large White Pigs. J. Anim. Sci. 63: 143-148. Blasco, A., Ortega, J. A., Santacreu, M. A., and Climent, A. 2005. Divergent selection for uterine capacity in rabbits. I. Genetic parameters and response to selection. J. Anim. Sci. 83: 2297-2302. General Discussion 54 Two-stage selection was proposed as an alternative to the index selection, which would be less affected by the precision of the genetic parameters and the models used for data analysis. There is an experiment of two-stage selection in pigs, performed by RuizFlores and Johnson (2001) in which, response in litter size was approximately twice the response observed in experiments for litter size in pigs. These results were the basis for the two-stage selection performed during the second period of our experiment. Through the second period of selection, mean ovulation rate was high due to the direct selection applied during the first and second periods, and a subsequent number of potentially viable embryos exceeding uterine capacity was expected. Therefore, litter size could be a good measurement of uterine capacity. Thus, selection in the second period was performed to ameliorate indirectly uterine capacity (i.e prenatal survival). After seven generations of two-stage selection, a change in litter size was observed (0.13 kit/ generation) as a result of the increase of both ovulation rate and prenatal survival (0.14 ova and 0.01 per generation for OR and PS respectively). Thus, two-stage selection resulted in approximately 30% greater response in litter size than direct selection for litter size. It could be concluded from this experiment that the applied twostage selection procedure resulted in substantial changes in both ovulation rate and litter size at birth, with a subsequent reduction in prenatal mortality. Similar conclusion was obtained by Ruiz-Flores and Johnson (2001) in the two-stage pig experiment, though there was a greater direct response in litter size (0.33 ± 0.06 pig/generation). This higher response in litter size could be due to a higher response in ovulation rate (0.26 ± 0.07 ova per generation) since their estimate of correlated response in prenatal survival was the same as the one estimated in our study (0.01 per generation). Generally, estimates of heritability for ovulation rate were higher in pigs than in rabbits, but causes of this high estimates in pigs were not clear. Direct and correlated responses from this study were estimated by genetic trends in both periods of selection, as the difference of line means between first and last generations divided per generation number. This common method to estimate genetic response has one limit that it strongly depends on genetic parameters and the model used in the analysis. An alternative would be the use of a control population (Rochambeau et al., 1989; Baselga, 2004) (i.e the control population must be raised contemporaneously and under the same environment as the selected population). The control line has the Chapter 4 55 advantage of providing independent information of the model used in the data analysis. However, the main problems of using a control line are: the genetic drift that acts through generations and the undesired selection (usually for small size and closed populations), and the need of economic and experimental facilities. The use of cryopreserved control population can avoid disadvantages of maintaining control population without selection. Thus, in this experiment of selection, embryos from donor females belonging to 6th generation of line OR_LS (just when the two-stage selection period started) were vitrified and stored in liquid N2to produce the control population. Two-stage selection will continue until generation sixteen, i.e. the 10th generation of two-stage selection, and responses will be estimated by comparison with the cryopreserved control population. General Discussion 56 Literature cited Adams, C. E. 1960a. Prenatal mortality in the rabbit Oryctolagus cuniculus. J. Reprod. Fertil. 1: 36-44. Adams, C. E. 1960b. Studies on prenatal mortality in the rabbit, Oryctolagus cuniculus: The amount and distribution of loss before and after implantation. J. Endocrinol. 19: 325-344. Baselga, M. 2004. Genetic improvement of meat rabbits. Programmes and diffusion. In Proc. 8th World Rabbit Congress, 2004 September, Puebla, Mexico, 1-13. Bennett, G. L., and Leymaster, K. A. 1989. Integration of ovulation rate, potential embryonic viability and uterine capacity into a model of litter size in swine. J. Anim. Sci. 67: 1230-1241. Blasco, A., Santacreu, M. A., Thompson, R., and Haley, C. S. 1993. Estimates of genetic parameters for ovulation rate, prenatal survival and litter size in rabbits from an elliptical experiment. Livest. Prod. Sci. 34: 163-174. Bradford, G. E. 1969. Genetic control of ovulation rate and embryo survival in mice. I. Response to selection. Genetics. 61: 907-918. Cunningham, P. J., England, M. E., Young, L. D., and Zimmerman, D. R. 1979. Selection for ovulation rate in swine: Correlated response in litter size and weight. J. Anim. Sci. 48: 509-516. Falconer, D. S., and Mackay, T. F. C. 2001. Caracteres correlacionados. In: Introducción a la genética cuantitativa. Ed. Acribia S.A. Zaragoza, Spain. Pages 317-339. Geisert, R. D., and Schmitt, R. A. M. 2002. Early embryonic survival in the pig: Can it be improved? J. Anim. Sci. 80(E-Suppl.): E54-E65. Hafez, E.S.E. 1966. Effects of overcrowding in utero on implantation and fetal development in rabbit. J .Exp. Zool. 156: 269-88. Johnson, R. K., Nielsen, M. K., and Casey, D. S. 1999. Responses in ovulation rate, embryonal survival and litter traits in swine to 14 generations of selection to increase litter size. J. Anim. Sci. 77: 541-557. Chapter 4 57 Laborda, P. 2011. Selection for ovulation rate in rabbits. Universidad Politécnica de Valencia. Doctoral thesis. Laborda, P., Mocé, M. L., Santacreu, M. A., and Blasco, A. 2011. Selection for ovulation rate in rabbits: Genetic parameters, direct response, and correlated response on litter size. J. Anim. Sci. 89: 2981-2987. Laborda, P., Mocé, M. L., Blasco, A., and Santacreu, M. A. 2012. Selection for ovulation rate in rabbits: genetic parameters and correlated responses on survival rates. J. Anim. Sci. 90: 439-446. Lamberson, W. R., Johnson, R. K., Zimmerman, D. R., and Long, T. E. 1991. Direct response to selection for increased litter size, decreased age at puberty, or random selection following selection for ovulation rate in swine. J. Anim. Sci. 69: 31293143. Land, R. B., and Falconer, D. S. 1969. Genetic studies of ovulation rate in the mouse. Genet. Res. 13: 25-46. Leymaster, K. A., and Christenson, R. K. 2000. Direct and correlated responses to selection for ovulation rate or uterine capacity in swine. J. Anim. Sci. 78(Suppl. 1):68. (Abstr.). Rochambeau, H., De La Fuente, L. F., and Rouvier, R. 1989. Sélection sur la vitesse de croirssanse post-sevrage chez le lapain. Genet. Sel. Evol. 21: 527-546. Rosendo, A., Druet, T., Gogue, J., and Bidanel, J. P. 2007. Direct response to six generations of selection for ovulation rate or prenatal survival in Large White pigs. J. Anim. Sci. 85: 356-364. Ruiz-Flores, A., and Johnson, R. K. 2001. Direct and correlated responses to two-stage selection for ovulation rate and number of fully formed pigs at birth in swine. J. Anim. Sci. 79: 2286-2299. Santacreu, M. A. 2006. La supervivencia prenatal en la coneja reproductora. XXXI Symposium de cunicultura de ASESCU, Lorca, Murcia, Spain. Vol. I: 229-236. 58 Chapter 5 Conclusions C. Ziadi Instituto de Ciencia y Tecnología Animal, Universidad Politécnica de Valencia, 46071 Valencia. Spain. Chapter 5 59 1. The estimated heritabilities for all traits were low, with the exception of ovulation rate which had a moderate heritability (0.21). 2. The estimated genetic correlation between ovulation rate and litter size was positive but low. It was estimated with low precision, having a probability of 95% of being in the interval from -0.12 to 0.71. 3. The estimated genetic correlations between litter size and the survival rates were positive, being moderate with embryo survival and fetal survival and high with prenatal survival. 4. The estimated genetic correlations of ovulation rate with fetal survival and prenatal survival were negative. Nothing can be said about the sign of the estimated genetic correlation between ovulation rate and embryo survival. During the first period of selection: 5. Estimated response for ovulation rate was 0.22 per generation, but litter size did not respond to selection, due to an increasing in fetal mortality. During the two-stage selection: 6. Estimated response for litter size was 0.13 per generation, as a result of the increase in ovulation rate and prenatal survival. Estimated responses were 0.14 ova and 0.01 per generation for ovulation rate and prenatal survival, respectively. 7. The increase in prenatal survival occurred during both pre and post-implantation periods.