Any genetic improvement programs applied for livestock are based on two main approaches: selection and crossbreeding. By contrast to crossbreeding, intensive selection within a single population reduces genetic diversity and increases the inbreeding rate (Barczak et al. 2009). A definition for inbreeding is given by mating of individuals whose relatedness between them is greater than the average degree of relationship existing in population and capable changing genotypic frequencies on a population without modifying the gene frequencies (Lush, 1945). The rate of inbreeding needs to be limited to maintain diversity at an acceptable level, so that genetic variation will ensure that future animals can respond to changes in the environment and to selection. Without genetic variation, animals cannot adapt to these changes (Van Wyk et al. 2009). Heterozygosity and allelic diversities can be lost from small, closed, selected populations at a rapid rate. The loss of diversity and resulting increase in homozygosity might result in decreased productions and / or fitness of inbred animals (Lamberson and Thomas, 1994; Ercanbrack and Knight, 1991; Analla et al. 1998; Dario and Bufano, 2003). Furthermore, inbreeding depression in domestic animals can lead to a decrease in selection response and in potential genetic gains in economic traits. Measuring the effect of inbreeding on productive traits is important in order to estimate the magnitude of change associated with increases in inbreeding (Negussie et al. 2002; Barczak et al. 2009). The initial consequence of inbreeding is inbreeding depression, which reduces the performance of growth, production, health, fertility and survival traits (Fernandez and Toro, 1999). Furthermore, inbreeding depression in domestic animals can lead to a decrease in selection response and genetic gains potential on economic traits. The emergence of disorders due to recessive gene action might occur, as well. It is apparent that different breeds and populations, as well as different traits vary in their response to inbreeding. Some populations might show a very pronounced effect of increased inbreeding for a trait, whereas others might not demonstrate much of an effect (Negussie et al. 2002; Barczak et al. 2009). Many studies have reported the rate of inbreeding and inbreeding depression in sheep. Pedrosa et al. (2010) reported that average of inbreeding was 2.33% in Santa Inês sheep in Brazil. Van Wyk et al. (2009) and Selvaggi et al. (2010) reported that inbreeding rates was 16% in Elsenburg Dormer sheep and 8.1% in Leccese sheep, respectively. Akhtar et al. (2000) showed that inbreeding depression in Hissardale sheep in Pakistan was -0.093, -0.130 and -0.190 kg for BW6, BW9 and BW12, respectively. Dorostkar et al. (2012) found that inbreeding depression for body weight traits in Iranian Moghani sheep at birth, 3, 6, 9 and 12 months of age was 0.7, -0.291, -0.260, -0.180 and -0.410 kg, respectively, (per 1% increase in individual coefficient). Therefore, inbreeding is an important parameter to monitor and control in breeding programs. The aim of this study was to evaluate the effects of inbreeding on body weight at the ages of birth (BW0), 3 (BW3), 6 (BW6), 9 (BW9) and 12 months (BW12) in Iranian Shal sheep.
MATERIALS AND METHODS
Pedigree of 6692 animals from 90 sires and 1007 dams that were collected on Shal Breed Station in Ghazvin, during 1997 to 2013 years, were used to estimate inbreeding coefficients. In the pedigree, 16.40% of animals had unknown sire, 12.19% of animals had unknown dam and 12.13% both parents were unknown. Inbreeding coefficients for animals with unknown parents considered as zero. The modified algorithm of Colleau was used to estimate individuals inbreeding coefficient CFC software, (Sargolzaei et al. 2006). Description of pedigree is presented in Table 1. Details of used data for estimation of inbreeding depression are given in Table 2, where the number of records is shown after editing i.e. animals with weight > weight at the same month ± 2 SD are deleted. There were fewer records at 9 months than 6 months of age, possibly because the heavier animals at 8 or 9 months of age were sent to the market.
Table 1 Data description of the studied Shal sheep flock
Table 2 Number of observations, mean and standard deviation of traits
BW0: birth weight; BW3: body weight in 3 month of age; BW6: body weight in 6 month of age; BW9: body weight in 9 month of age and BW12: body weight in 12 month of age.
Data were analyzed by least squares analysis of variance using the general linear model (GLM) procedure of the SAS software package (SAS, 2004). The fixed effects were including: sex of lambs in two classes (male-female), type of birth in four classes (single, twins, triplets, quadruplet), age of the dam at lambing in seven classes (2 to 8 years old) and year of birth in 17 classes (1997 to 2013), respectively. Therefore, these effects were excluded from the final model. Moreover, the age of lambs was placed in the model as a covariate factor. By excluding or including various random effects, six univariate linear animal models were fitted for each trait. Direct additive genetic effect was presented in all models and only random effect in Model 1. Models 2 and 3 included maternal permanent environmental effect and maternal additive genetic effect, respectively. There was an additional effect [direct-maternal genetic covariance (σa,m)] in model 4 compared to model 3. Models 5 and 6 included both maternal effects and also with and without covariance between animal effects. Six univariate models were described as below:
y= Xb + Z1a + e Model 1
y= Xb + Z1a + Z3c + e Model 2
y= Xb + Z1a + Z2m + e Model 3
Cov (a,m)= 0
y= Xb + Z1a + Z2m + e Model 4
Cov (a,m)≠ 0
y= Xb + Z1a + Z2m + Z3c + e Model 5
Cov (a,m)= 0
y= Xb + Z1a + Z2m + Z3c + e Model 6
Cov (a,m)≠ 0
y: n×1 vector of observations in each considered trait.
b: vector of fixed effects with a significant effect on related traits. Overall, fixed effects were included: lamb’s sex (male and female, 2 classes), year of birth (1997 to 2013, 17 classes), birth type (single, twins, triplets, quadruplet, 4 classes) and dam age (2-8 years and older ewes, 7 classes), maternal permanent environmental effects, and residual effects, respectively.
a, m, c, and e: vectors of direct genetic effects, maternal genetic effects, maternal permanent environmental effects, and residual effects, respectively. It is assumed that these random effects are normally distributed with a mean of zero and variances AϬ2a, AϬ2m, IdϬc2 and IdϬe2, respectively. Also, Ϭa2, Ϭm2,Ϭc2 and Ϭe2 are direct additive genetic variance, maternal additive genetic variance, maternal permanent environmental variance, and residual variance, respectively. A is the additive numerator relationship matrix that is created using pedigree information. Id and In are identity matrices with dimensions equal to the number of dams and observations, respectively.
X, Z1, Z2 and Z3: design matrices (0 and 1) that are related to fixed effects, direct additive genetic effects, maternal additive genetic effects, and maternal permanent environmental effects to observations.
Log-likelihood ratio (Log L) tests were performed to determine significant random effects and consequently the most appropriate model for each considered traits. By inclusion of a random effect in the model, a significant increase was seen in the Log L compared to the reduced model (model without this effect). However, when the difference between the values of Log L was not greater than a critical value of χ2, the simplest model was considered to be the best model. Statistical significance for models set at 5% probability level. The best model for BW was the full model (Model 6) and for BW3, BW6, BW9 and BW12 was model 4.
RESULTS AND DISCUSSION
With the use of dense genomic marker data it is now possible to estimate inbreeding levels using the data alone, thus avoiding the problems of incomplete pedigree and also accounting for Mendelian segregation, but we do not have data to do this. Distribution of animals in different classes of inbreeding was shown in Table 3.
Table 3 Distribution of animals in different classes of inbreeding
F: inbreeding coefficients.
Based on the distribution of inbreeding coefficients, the animals were divided into 7 classes of inbreeding (F=0, 0<F≤5, 5<F≤10, 10<F≤15, 15<F≤20, 20<F≤25 and F>25). Inbreeding coefficients for the animals in the founder population (year 0) and the animals brought into the flock during the period under study were considered zero because their parents were unknown and there was no pedigree information. The results showed that inbreeding among the groups, most inbred animals (13.13%) of the animals with inbreeding coefficients zero to 5 percent that these results are confirmed low levels of inbreeding in the herd. In the herd, the only 0.20 percent of all animals, inbreeding coefficient greater than 25 percent and 75.84 percent of the population has an inbreeding coefficient is zero. Accordingly the maximum number of animals were considered as thefirst class of inbreeding (F=0) and minimum number of animals as seventh class for all studied weights. Descriptive statistics of inbreeding coefficients for whole population and inbred population are shown at Table 4. The mean of inbreeding coefficient in females and males were 1.40 and 1.58 %, respectively. Totally, 24.15% of animals were inbred with mean inbreeding coefficient of 6.28%. This illustrated that low mating of close relatives was occurred in this population. Inbreeding coefficient in this study was lower than other published results (Rzewuska et al. 2005; Norberg and Sorenson, 2007). This estimates were higher than studies on Baluchi sheep (Mehmannavaz et al. 2002), Moghani sheep (Dorostkar et al. 2012) and other Iranian sheep breeds. These low estimates were due to low accuracy of data recording on the station made low pedigree completeness. The highest inbreeding coefficient was 31.25% and most of inbred animals had inbreeding coefficients lower than 5%. Some animals of the studied population had presented high levels of inbreeding, reflecting the intensive use of few sires. Increasing trend for mean of inbreeding in whole animals, females and males by 17 years were shown at Figure 1.
Table 4 Descriptive statistics for inbreeding coefficients for the studied population of Shal sheep
Figure 1 Annual mean of inbreeding for all, female and male animals
The proportion of inbred animals increased from zero in 1997 to 3.84 in 2013. The proportion of inbred animals in 2013 may be a cause for concern, although the average level of inbreeding was still very low. The mean of inbreeding was zero in early years of studied period. The maximum inbreeding was observed in 2013 for male animals. The increased values of inbreeding in some years may be due to poor controlling on close relative mating and excessive using of some individuals as breeding rams. Mean of inbreeding in females was zero in 2006. The mean level of inbreeding was decreased in whole animals at 2006 and 2009. The reason of this decrease was probably because of ram admittance in herd and the prevention of closed mating in sheep by the breeder by not using very few sires, and using them fairly equally. In those years, the station began to perform synchronization of ovulation and some female and male animals were imported to the station. This decrease was observed in all females and males and it could say that breeders selected non-related animals for mating. In 2010, the mean of inbreeding level for all animals was 0.06%. This percentage was very low, but it illustrated that the mean of inbreeding had been increased compared to the base year (1997). Annual inbreeding rate for whole animals was 0.07% per year during 17 years of study. It was observed that the average inbreeding coefficient increased due to the reason that inbred males and female individuals belonging to the same population or flock are mated together. This estimate of inbreeding rate was less than 0.40, 1.00 and 1.53% reported by Huby et al. (2003), Norberg and Sorenson, (2007) and Van Wyk et al. (2009), respectively. Totally the inbreeding coefficient of 17 years was a non-significant and positive trend, so that in some years of decline but increase again, these fluctuations could be due to various factors such as the ram productive ewe percentage, rams herd displacement levels pedigree, the evolution of the parent changes in the number of sheep center and management methods different over the years. In this study, the average of inbreeding total sheep population Shal was born consistent throughout the year 1997 to 2013 was equal to 1.51 percent. Possibly resulting amount at causes due to lack of specific information regarding the number of parents and grandparents animals common, under-estimated. But inbreeding coefficients of inbred animals according to the number of these animals show in the population, the Sexual Intercourse of targeted largely and been controlled. Furthermore, the number of animals with high inbreeding coefficients in population indicates a lack of understanding in control mating close relatives in the population. The result of variance analysis showed that the year of birth had significant effects on all studied traits (P<0.01). Sex of lamb had significant effect on all traits (p<0.01). The significant effect of fixed factors in these characters could be assigned partly to the differences in the endocrine system of female and male lambs. Also, age of dam had significant effect on birth weight, BW3, BW6, BW9 and BW12 (P<0.05). Type of birth had a significant effect on weight changes in all traits (P<0.01). Single born lambs had higher body weights and pre-weaning growth rate than twins and triplets. Due to climate conditions, feedstuff availability and ewe nutrition, especially during late pregnancy in sheep, it is expected that the birth year affects growth traits. The effect of sex and type of birth can also be caused by differences in the endocrine system, possible loci related to growth on sex chromosome and competition between twins for uterine space, milk consumption and other maternal ability compared to single-born lambs. Single-born lambs were weighty than twins, which may be due to intense competition between twins; low milk production by ewe will not provide feed requirement of lambs and consequently they cannot express their potential capacity. It seems that increase in dam age had no effect on milk production and nursing of ewe of this breed. Nevertheless, there is a relationship between age of dam and BW because uterine environment will be better with increasing age. Regression coefficients per 1 % increase of inbreeding for birth weight, BW3, BW6, BW9 and BW12 were -0.001, -0.017, -0.005, -0.019 and -0.019 kg, respectively (Table 5).
Table 5 Inbreeding depression for studied traits per 1 percent increase in inbreeding coefficient
BW0: birth weight; BW3: body weight in 3 month of age; BW6: body weight in 6 month of age; BW9: body weight in 9 month of age and BW12: body weight in 12 month of age.
These regression coefficients show no significant inbreeding depression. These estimates for BW were higher than -0.0005 kg estimated for Baluchi sheep (Mehmannavaz et al. 2002), but, lower than those was reported by some other researchers like as Van Wyk et al. (2009) for Dormer sheep (-0.006 kg/1% inbreeding); Dorostkar et al. (2012) for Moghani sheep (-0.007 kg/1% inbreeding); Mandal et al. (2005) for Muzaffarnagari sheep (-0.01 kg/1% inbreeding). Also, individual regression coefficients for BW were estimated -0.0001, -0.00008 and -0.00009 kg per 1% inbreeding for Texel, Shropshire and Oxford Down, respectively (Norberg and Sorensen, 2007). Reason of variation in inbreeding coefficients could be due to differences among breeds in alleles segregating, amount of genetic variation in the base population, location, management, and diversity of the founders in the tested flock (MacKinnon, 2003). Dorostkar et al. (2012) reported that inbreeding coefficient was -0.007 kg/1% inbreeding; Mandal et al. (2005), Mehmannavaz et al. (2002) and Yavarifard et al. (2014) reported that inbreeding effect for Muzaffarnagari, Baluchi, and Mehraban breeds are -0.048, -0.026 and -0.014 kg per 1% inbreeding, respectively; current result of this paper for BW3 was lower than the mentioned reports. Inbreeding depression for BW6 and BW9 per 1% increase in inbreeding coefficient were -0.005, -0.019 kg, respectively. Estimation of inbreeding depression for BW6 per 1% inbreeding was -0.260 kg that reported by Dorostkar et al. (2012) in Moghani sheep and Akhtar et al. (2000) reported that -0.093 kg in Hissardale sheep; likewise, estimation of inbreeding depression for BW9 was -0.129 kg that reported by Mandal et al. (2005) in Mozafarnagari sheep, -0.180 kg by Dorostkar et al. (2012) in Moghani sheep and -0.130 kg by Akhtar et al. (2000) in Hissardale sheep that was lower than result of current study. The average of regression coefficient for body weight in 10 to 12 months age per 1% inbreeding was -0.112 kg (Mandal et al. 2005), -0.410 kg (Dorostkar et al. 2012) and -0.190 kg (Akhtar et al. 2000) in Mozafarnagari sheep, Moghani sheep and Hissardale sheep, respectively, which are in disagreement with the regression coefficients that found in this study (-0.019 kg % for 12 month weight). Inbreeding is generally associated with deterioration in growth in reproductive traits in small ruminants (Lamberson and Thomas, 1994; Wocac, 2003) and level of inbreeding may be an important factor for such effects to appear. Level of inbreeding was generally low (1.51%), mainly due to periodic introduction of unrelated ewes and rams which helped in controlling the rate of increase in the level of inbreeding. Although, some of the animals introduced may be relatives, but were assumed unrelated because of lack of pedigree recording at filed level from where such animals were purchased. This may be one of the factors that resulted in estimation of low level of inbreeding in the present flock. With the exception of few years, most of the animals were always inbred across different years but level of inbreeding was low. The inbreeding may accumulate quickly for a flock of this size due mainly to small effective population size as indicted by <a href="file:///D:/multimedia/seri%2024
Akhtar P., Ahmad Z., Mohiuddin G. and Abdullah M. (2000). Effect of inbreeding on different performance traits of Hissardale sheep in Pakistan. Pakistan Vet. J. 20(4), 169-172.
Analla M., Montilla J.M. and Serradilla J.M. (1998). Analyses of lamb weight and ewe litter size in various lines of Spanish Merino sheep. Small Rumin. Res. 29, 255-259.
Barczak E., Wolc A., Wojtowski J. and Slosarz P. (2009). Inbreeding and inbreeding depression on body weight in sheep. J. Anim. Feed. Sci. 18, 42-50.
Dario C. and Bufano G. (2003). Effect of inbreeding on milk production in Altamurana sheep breed. J. Anim. Prod. 55, 270-273.
Dorostkar M., Faraji Arough H., Shodja J., Rafat S.A., Rokouei M. and Esfandyari H. (2012). Inbreeding and inbreeding depression in Iranian Moghani sheep breed. J. Agric. Sci. Technol. 14, 249-556.
Ercanbrack S.K. and Knight A.D. (1991). Effects of Inbreeding on Reproduction and wool production of rambouilet, targhee and Columbia ewes. J. Anim. Sci. 69(12), 4734-4744.
Fernandez J. and Toro M.A. (1999). The use of mathematical programming to control inbreeding in selection schemes. J. Anim. Breed. Genet. 116, 447-466.
Huby M., Griffon L., Moureaux S., De Rochambeau H., Danchin Burge C. and Verrier E. (2003). Genetic variability of six french meat sheep breeds in relation to their genetic management. Genet. Sel. Evol. 35, 637-655.
Lamberson W.R. and Thomas D.L. (1994). Effects of Inbreeding in sheep: a review. Anim. Breed. Abstr. 52, 287-297.
Lush J.L. (1945). Animal Breeding Plans. IowaState College Press, Ames, Iowa, USA.
MacKinnon K.M. (2003). Analysis of Inbreeding in a Closed Population of Crossbred Sheep. MS Thesis. University of Blacksburg, Virginia, USA.
Mandal A., Pant K.P., Notter D.R., Rout P.K., Roy R., Sinha N.K. and Sharma N. (2005). Studies on inbreeding and its effects on growth and fleece traits of Muzaffarnagari sheep. Asian-Australas. J. Anim. Sci. 10, 1363-1367.
Mehmannavaz Y., Vaez Torshizi R., Salehi A. and Shorideh A. (2002).Inbreeding and its effect on production traits in Iranian Baluchi sheep. Pp. 263- 268 in Proc. 1st Iranian Conf. Genet. Breed. Appl. Livest. Poult. Aquat. Tehran, Iran.
Norberg E. and Sorensen A.C. (2007). Inbreeding trend and inbreeding depression in the Danish populations of Texel, Shropshire and Oxford Down. J. Anim. Sci. 85, 299-304.
Pedrosa V.B., Santana J.M.L., Oliveira P.S., Eler J.P. and Ferraz J.B.S. (2010). Population structure and inbreeding effects on growth traits of Santa Inês sheep in Brazil. Small Rumin. Res. 93, 135-139.
Rzewuska K., Klewiec J. and Martyniuk E. (2005). Inbred effect on reproduction and body weight in a closed flock of Booroola sheep. Anim. Sci. Pap. Rep. 23(4), 237-247.
Sargolzaei M., Iwaisaki H. and Colleau J.J. (2006). A tool for monitoring genetic diversity. Pp. 27-28 in Proc. 8th World Congr. Genet. Appl. Livest. Prod. Horizonte, Brazil.
SAS Institute. (2004). SAS®/STAT Software, Release 9.1. SAS Institute, Inc., Cary, NC. USA.
Selvaggi M., Dario C., Peretti V., Ciotola F., Carnicella D. and Dario M. (2010). Inbreeding depression in Leccese sheep. Small Rumin. Res. 89, 42-46.
Van Wyk J.B., Fair M.D. and Cloete S.W.P. (2009). Case study: the effect of inbreeding on the production and reproduction traits in the elsenburg Dormer sheep. Livest. Sci. 120(3), 218-224.