Genetic effects of aging on fitness and nonfitness traits in laying hens housed three per cage

Genetic Effects of Aging on Fitness and Nonfitness Traits in Laying Hens Housed Three per Cage M. C. Ledur,*,1,2 L.-E. Liljedahl,† I. McMillan,*,3 L. Asselstine,‡ and R. W. Fairfull‡ *Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario, N1G 2W1, Canada; †Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences, SE-750 07, Uppsala, Sweden; and ‡Centre for Food and Animal Research, Agriculture Canada, Ottawa, Ontario, K1A OC6 Canada The age-related changes in additive, Z-chromosome, and heterotic effects varied among strains, showing that strains differ in their genetic schemes in response to aging. Nonadditive, environmental, and phenotypic variances increased with age for all traits. Additive variance increased with age for EPF, EW, and AH. Z-chromosome variance increased with age for EW and AH. Heritabilities decreased with age, except for EPF and AH. On average, genetic variance increased with advancing age. Improvement in lifetime performance may be obtained by selecting birds at older ages. As the relative increase with age in additive variance was larger for egg production than for egg quality traits, selection for the latter could be performed at early stages.

(Key words: genetic effect, aging, egg production, egg quality, multiple-bird cage) 2003 Poultry Science 82:1223–1234

INTRODUCTION Although selection strategies applied to laying hens are normally based on individual records, commercial hens are kept during the production cycle with two or more birds per cage. Fairfull et al. (1983) reported that birds housed three per cage had lower egg production, lower mature body weight, and higher mortality than those housed one per cage. The later egg weights and early Haugh units were also affected by stocking rate. Increased colony size, reduction in floor and feeder space per bird, competition, and poor adaptability are causes for lower performance of birds housed in multiple cages (Champion and Zindel, 1968; Adams, 1974; Bell, 1981; Muir, 1985, 1996). Most economic traits evaluated in egg stock breeding programs deteriorate with advancing age,

2003 Poultry Science Assocation, Inc. Received for publication May 20, 2002. Accepted for publication October 29, 2002. 1 Sponsored by Conselho Nacional de Desenvolvimento Cientifico e Tecnologico, DF Brasilia, Brazil. 2 Present address: Embrapa Suinos e Aves, 89700-000—Concordia, SC, Brazil. 3 To whom correspondence should be addressed: imcmilla@ uoguelph.ca.

with the exception of egg weight (Liljedahl et al., 1984; Ledur et al., 2000a,b, 2002). The decline in egg production and some egg quality traits throughout the cycle might be faster when birds are kept three per cage than when they are housed one per cage, due to the factors described above. Therefore, expression of the genetic effects involved in those traits, as well as their variances with age, are expected to be influenced by the environment in different ways when hens are housed one or three per cage. Studies on the changes of genetic and environmental variation with age have been carried out only with layers housed one per cage. These studies have shown increased genetic and environmental variation in fitness and nonfitness traits with age (Liljedahl et al., 1984, 1999; Engstrom et al., 1992; Fairfull et al., 1999; Ledur et al., 2000a,b, 2002). Also, the pattern of age changes has varied among strains, showing genotypic differences in response to aging for different traits (Fairfull et al., 1999; Liljedahl et al., 1999; Ledur et al., 2000a,b, 2002).

Abbreviation Key: AH = albumen height; ASM = age at sexual maturity; CL = commercial line; EPF = egg number of the survivors; EPM = egg number including mortality and morbidity; EW = egg weight; SG = specific gravity.

1223

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

ABSTRACT Three White Leghorn strains, their twoway crosses, and two commercial lines were used to study the effects of aging on several parameters related to performance of fitness and nonfitness traits during the first laying cycle of hens housed three per cage. Egg number of the survivors (EPF) and egg number including mortality and morbidity (EPM) were divided into 12 periods of 28 d each, starting at age at sexual maturity. Egg weight (EW), specific gravity (SG), and albumen height (AH) were measured at 240, 350, and 450 d of age. Mean heterosis was significant over time, except for AH, increasing in magnitude with age for EPF, EPM, EW, and AH. Reciprocal effects were more important for egg quality than for egg production traits and were influenced by age.

1224

LEDUR ET AL.

MATERIALS AND METHODS

of three eggs was produced on 2 consecutive d), keeping only cages with three birds alive until the end of the evaluated period. Cages with less than 20% of potential production were removed from the analyses to account for morbidity. Number of eggs from the survivors (EPF) was a measurement of egg production free from influence of ASM, variable length of the records, mortality, and morbidity. The second measure of egg production was egg number from ASM but included all cages that had at least one egg. This number of eggs (EPM) included the effects of mortality and morbidity on egg production. It is similar to hen housed egg production but does not include the influence of ASM or variable length of records. For both measures of egg production, the first period was synchronized on ASM, so comparisons were based on the physiological age of the hens with the same length of records. For statistical analyses, the number of eggs from each cage was divided into 12 periods of 28 d each, beginning from ASM. From a total of 1,536 cages, 1,092 were used to analyze EPF and 1,535 for EPM.

Egg Quality Traits Egg weight (EW), egg specific gravity (SG), and albumen height (AH) for the three strains, six two-way crosses, and two commercial lines were measured during the first laying cycle at 240, 350, and 450 d of age. Eggs from approximately 420 hens from each genetic group were evaluated. The eggs were collected for 2 consecutive d, and the mean of the eggs from each cage was used as the observation for each period. A total of 1,519 observations were used to analyze EW, SG, and AH. Egg quality traits were measured as described by Ledur et al. (2002).

Birds and Housing

Statistical Analyses

Three parental strains from the industry, designated as 1, 2, and 3, were used in a factorial mating plan to produce nine different genetic groups: three strains and three pairs of reciprocal crosses. White Leghorn chicks were produced from pedigree matings using 16 sires and four dams per sire (64 dams) for each strain and cross. Two commercial lines (CL1 and CL2) were included in this experiment. At about 133 d the hens were housed three full-sibs per cage in a two-tier, stair step cage system in a randomized block design. Each genetic group (strain, cross, or commercial) was randomly assigned within each of the 16 cage rows representing complete blocks. Approximately 420 hens were housed for each genetic group. More details on animals and housing used in this experiment are described in Ledur et al. (2000a).

Repeated measures analyses from the general linear models procedure of SAS software (SAS Institute, 1989) were carried out using untransformed data, as these data fit the assumptions required for analyses of variance better than any of the transformations considered. A preliminary analysis was done to obtain general information about the populations used, such as average ASM, mortality, EPF, and EPM. The mean of each cage was used as the observation. For mortality, a categorical data analysis was carried out, and genetic group differences were tested with chi-squared statistics. A brief description of the methods applied is given here. For details of the statistical analyses used in the present study, such as description of the models, comparisons made, and estimation of heterosis and reciprocal effects, the reader should refer to Ledur et al. (2000a). Two models were used to analyze the data. One contained block and genetic group effects and the other had the strain additive genetic effect (A), strain Z-chromosome effect (Z), and heterosis (H), to explain differences between genetic groups. In the second model, the effects were estimated using information of the three strains and their two-way crosses, in a regression model. Variances

Egg Production Traits Egg production was recorded for each cage, 5 d/wk, from 138 to 576 d of age, and number of eggs in 7 d were estimated. Egg production was analyzed in two different ways. The first was by egg number from age at sexual maturity (ASM) (estimated as the age when a minimum

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

Heterosis has been shown to increase in magnitude with age not only in fitness traits but also for some nonfitness traits (Ledur et al., 2000a,b, 2002) when birds are kept one per cage. The magnitude of heterosis depends also on the environment (Barlow, 1981). Differences in heterosis when birds were housed one or three per cage were reported by Fairfull et al. (1983). Heterotic effects were higher with three birds per cage for most of the egg production traits but not for egg quality traits. In a review by Fairfull (1990), it was demonstrated that reciprocal effects were also influenced by environment. Differences between reciprocal crosses were greater with three birds per cage for the hen-day rate of egg production and viability than with one hen per cage. Because environment has an important effect on the expression of the genetic effects involved in egg production traits and the age-related changes of these effects and their variances have not been reported with hens housed in multiple-hen cages, it is of great importance to evaluate age changes in a simulated commercial environment to improve lifetime performance of layers. The objectives of the current study were to compare performance of pure strains, crosses, and commercial lines during the first laying cycle; to estimate mean heterosis and reciprocal effects and examine their trends with age; to estimate heritabilities, genetic effects (additive, Zchromosome, and heterosis), and their variances in each period of the first laying cycle; and to evaluate trends of these parameters with age in the first laying cycle, for fitness and nonfitness traits, when birds were housed three per cage.

1225

AGING ON FITNESS AND NONFITNESS TRAITS TABLE 1. Number of observations (n), estimated means of age at sexual maturity (ASM) (d) and egg production for both egg number of the survivors (EPF) and egg number including mortality and morbidity (EPM) populations per genetic group, as well as the observed mean mortality during the first laying cycle EPF population

Genetic group1

EPM population

ASM

EPF

n

ASM

EPM

Mortality (%)

1 2 3 Mean

113 100 93

165.2 163.9 167.5 165.5

269.7 282.3 262.5 271.6

135 142 141

165.4 164.4 167.6 165.8

259.0 263.7 244.1 255.6

7.1 12.7 13.1 11.0

12 21 13 31 23 32 Mean

113 106 93 95 91 87 106 95

292.1 283.9 290.8 288.4 279.4 282.2 286.4 286.2 274.3 280.6

141 140 140 141 140 140

CL1 CL2 Mean

157.0 156.7 161.7 157.5 161.7 160.1 159.0 160.1 151.0 155.8

157.3 156.6 161.5 157.8 161.8 160.2 159.2 160.8 150.7 155.8

284.0 275.7 267.8 269.5 256.3 261.9 269.2 272.0 255.3 263.7

6.9 8.8 13.3 13.3 16.2 16.2 12.5 9.0 13.8 11.4

160.2

281.1

160.4

264.5

11.9

Total/mean

1,092

138 138 1,535

CL1 = commercial line 1; CL2 = commercial line 2.

1

of the genetic effects were estimated according to Henderson’s Method 3 (Henderson, 1953), and heritabilities were calculated from variance components as follows:

losses due to mortality and morbidity, which were larger compared to those with one bird per cage (5.8 eggs in Ledur et al., 2000a). Average mortality was 11.9% during the first laying cycle. There were differences among genetic groups for mortality (P < 0.01). Strain 1 showed lower mortality than the other strains. The commercial lines also differed significantly for mortality (Table 1). However, when birds were housed one bird per cage (Ledur et al., 2000a), genetic groups did not show significant differences in mortality. The results indicate that genetic groups expressed their differences in mortality to a greater degree when subjected to a more stressful environment. Significant differences among genetic groups were found for ASM and egg production. As shown in Table 1, crosses reached ASM approximately 6.5 d earlier than pure strains of EPF and EPM populations. Differences

h2 = (σ2A + σ2Z)/σ2p. σ2p = σ2A + σ2Z + σ2H + σ2e.

RESULTS AND DISCUSSION Egg Production Traits Performance Means. As shown in Table 1, the averages of ASM were 160 d of age for EPF and EPM populations. The average EPF and EPM were 281.08 and 264.53 eggs, respectively. The difference of 16.55 eggs between EPF and EPM at the end of the cycle can be attributed to

TABLE 2. Reciprocal effects (±SE) of the different crosses for egg number of the survivors (EPF) and egg number including mortality and morbidity (EPM) in the evaluated periods1 and in the whole laying cycle 12 vs. 21 Periods 1 2 3 4 5 6 7 8 9 10 11 12 Tota 1

EPF 0.28 0.81 1.02 1.24 1.14 0.82 0.99 1.13 0.40 0.33 0.02 0.03 8.2

± ± ± ± ± ± ± ± ± ± ± ± ±

0.3 0.3** 0.3** 0.3** 0.4** 0.4* 0.4* 0.4** 0.4 0.5 0.5 0.5 3*

13 vs. 31 EPM

0.23 0.73 0.96 1.20 1.14 0.99 0.90 1.08 0.35 0.43 0.14 0.10 8.3

± ± ± ± ± ± ± ± ± ± ± ± ±

0.3 0.3* 0.4* 0.5** 0.5* 0.5 0.6 0.6 0.6 0.6 0.6 0.6 5

EPF 0.35 0.29 −0.16 −0.19 −0.02 0.18 0.30 0.31 0.35 0.33 0.54 0.12 2.4

± ± ± ± ± ± ± ± ± ± ± ± ±

23 vs. 32 EPM

0.3 0.3 0.3 0.4 0.4 0.4 0.4 0.5 0.5 0.5 0.5 0.5 4

0.21 −0.11 −0.64 −0.58 −0.27 −0.16 0.02 −0.19 −0.08 0.18 0.12 −0.17 −1.7

± ± ± ± ± ± ± ± ± ± ± ± ±

EPF 0.3 0.3 0.4 0.5 0.5 0.5 0.6 0.6 0.6 0.6 0.6 0.6 5

−0.01 0.34 0.10 0.04 0.06 −0.04 0.07 0.28 −0.73 −0.83 −0.71 −1.27 −2.7

First laying cycle was divided into 12 periods of 28 d each, from age at first egg. *P ≤ 005. **P ≤ 0.01.

± ± ± ± ± ± ± ± ± ± ± ± ±

EPM 0.4 0.3 0.3 0.4 0.4 0.4 0.5 0.5 0.5 0.5 0.5 0.5* 4

−0.36 −0.44 −0.47 −0.68 −0.49 −0.44 −0.18 −0.16 −0.62 −0.67 −0.47 −0.57 −5.6

± ± ± ± ± ± ± ± ± ± ± ± ±

0.3 0.3 0.4 0.5 0.5 0.5 0.6 0.6 0.6 0.6 0.6 0.6 5

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

n

1226

LEDUR ET AL.

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

FIGURE 1. Mean heterosis for egg production [egg number of the survivors (EPF) and egg number including mortality and morbidity (EPM)] of six strain crosses during the first laying cycle.

among strains, reciprocal crosses, commercials, and crosses vs. commercials were significant for ASM. Commercial lines reached sexual maturity approximately 3.5 d earlier than the crosses. Crosses produced more eggs (approximately 14.5 for EPF and 14 for EPM) than did pure strains. There were differences among strains for total EPF and EPM and also between commercial strains. Crosses produced more eggs than commercial strains, about 5.9 eggs on average for the EPF population. There was also a significant reciprocal effect for EPF, for which cross 12 laid eight eggs more than cross 21. Mean Heterosis and Reciprocal Effects. The mean heterosis expressed in absolute values was highly significant across the cycle for both traits and was also affected by the age of the hen. Mean heterosis increased curvilinearly with age in EPF (P < 0.01) and showed a curvilinear decrease (P < 0.06) with age in EPM (Figure 1). Perhaps this decline was due to a negative heterosis for viability when hens were housed three per cage. Fairfull (1990) reported that heterosis for viability is quite variable, ranging from −9 to 24%. An increase in heterosis for rate of lay from the early to the late period in the first laying cycle was found by Fairfull et al. (1987), with birds housed two and three per cage, which was similar to the results obtained for EPF. For EPF and EPM, the effects of ASM and the variable-length records were removed; a comparison between the two curves shows the effects of mortality and morbidity. Results from Ledur et al. (2000a) revealed that mortality and morbidity had an important positive effect on the expression of heterosis for egg production at the end of the cycle when birds were housed one per cage. However, the decline in mean heterosis for EPM with three birds per cage suggests a negative effect of mortality and morbidity on heterosis later in the cycle. When this result is contrasted to the one obtained by Ledur et al. (2000a) for EPM, it indicates the important effect of environment in the expression of mean heterosis with age.

FIGURE 2. A(I) = strain additive (a), Z(I) = Z-chromosome (b), and H(I, J) = heterotic (c) genetic effects estimated for egg number of survivors in each period of the first laying cycle of three strains and their two-way crosses.

The only significant reciprocal effect over time was shown by the cross involving strains 1 and 2 for EPF (Table 2). However, age had an important effect (P < 0.05) on this difference for EPF and EPM, each showing a quadratic pattern with age, increasing through period 4 and declining thereafter. Genetic Effects. The additive, Z-chromosome, and heterotic effects estimated for the three strains and their

AGING ON FITNESS AND NONFITNESS TRAITS

two-way crosses in 12 periods of the first laying cycle are presented in Figures 2 and 3 for EPF and EPM, respectively. The additive (A2 and A3) and Z-chromosome (Z2 and Z3) effects are expressed as a deviation from strain 1 effects (A1 and Z1).

For EPF, A2 was, on average, higher than A1 throughout the first laying cycle (P < 0.01), and this difference increased linearly with advancing age (P < 0.01; Figure 2a). The A3 was lower than A1 from periods 3 to 6. Differences between A3 and A1 followed a quadratic pattern (P < 0.01), increasing in periods 3 to 6 and decreasing after that. Although differences between Z2 and Z1 were significant across the cycle, they did not show any significant change with age (Figure 2b). The Z2 was lower than Z1 throughout the first laying cycle (P < 0.01). Heterotic effects for EPF were highly significant over time and showed different patterns with age (Figure 2c; Table 3). The H12 increased linearly with age (P < 0.01). A quadratic increase with age was shown by H13 (P < 0.01), and a quadratic decrease with age was shown by H23. Greater heterotic effects were shown by crosses involving strains 1 and 3. The H12 and H23 were quite similar from periods 2 to 8, but after that, H23 started to decrease, while H12 kept increasing, reaching a magnitude similar to H13. For EPM, A3 was significantly lower than A1 across the cycle, but this difference followed a quadratic pattern, increasing by periods 3 to 6 and decreasing after that (Figure 3). Heterotic effects were significant over time, with the exception of H23. The H12 followed a curvilinear pattern, increasing with age (Figure 3c; Table 3). The changes in H13 with age were not significant. However, H23 had a curvilinear decrease with advancing age. The H12 and H13 were quite similar, especially from periods 3 to 8. Results obtained by Ledur et al. (2000a), with the same traits evaluated in the current study and birds housed one per cage, showed increased divergence with age in additive, Z-chromosome, and heterotic effects among different genetic groups. However, the results reported here, with three birds per cage, showed that only differences in additive and heterotic effects, on average, increased with age for EPF, whereas for EPM only the heterotic effects showed increased divergence with age. The different patterns of age change of these effects among strains indicate different strategies of aging for distinct genetic groups. These findings are supported by results from the previous papers by Ledur et al. (2000a,b, 2002) with one bird per cage. A good example in the present study is the cross involving strains 2 and 3 (Figures 2 and 3), which had consistent heterosis for EPF and EPM until the middle of the cycle but very poor heterosis at the end of the cycle when compared with other crosses. For EPF and EPM, the additive and heterotic effects were most important to explain variation among genetic groups. Differences in Z-chromosome effects, on average, were important only for EPF in the first laying cycle. Differences in additive effects increased with age for EPF but decreased for EPM. Heterotic effects were on average higher when estimated for EPF than for EPM, with the exception of H12, which was quite similar in both traits (Table 3). The H23 was very similar in the beginning of the cycle for both traits, but after period 6 it declined faster in EPM.

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

FIGURE 3. A(I) = strain additive (a), Z(I) = Z-chromosome (b), and H(I, J) = heterotic (c) genetic effects estimated for egg number including mortality and morbidity in each period of the first laying cycle of three strains and their two-way crosses.

1227

1228

LEDUR ET AL. TABLE 3. Heterosis and percentage heterosis in parenthesis for egg number of the survivors (EPF) and egg number including mortality and morbidity (EPM) for different crosses1 in the evaluated periods2 and in the whole cycle H12 Periods 1 2 3 4 5 6 7 8 9 10 11 12 Mean

EPF −0.11 0.64 0.90 1.38 0.71 0.37 1.32 1.15 1.33 1.39 1.40 1.55 12.00

(−0.5) (2.5) (3.5) (5.8) (2.9) (1.5) (5.7) (5.2) (6.1) (6.5) (6.5) (7.6) (4.3)

H13 EPM 0.38 1.19 1.59 2.18 1.37 1.04 1.76 1.55 1.77 1.91 1.89 1.87 18.48

(1.7) (4.7) (6.4) (9.6) (6.0) (4.5) (8.1) (7.4) (8.7) (9.7) (9.6) (10.1) (7.1)

EPF 1.19 1.49 1.90 2.27 2.08 1.89 2.32 2.14 2.17 2.12 2.11 1.86 23.52

H23 EPM

(5.4) (5.9) (7.6) (10.0) (9.0) (8.3) (10.5) (10.1) (10.4) (10.3) (10.1) (9.4) (8.8)

1.52 1.79 1.66 1.84 1.38 1.27 1.58 1.31 1.34 1.26 1.30 0.88 17.12

(7.0) (7.3) (6.9) (8.5) (6.3) (5.8) (7.6) (6.6) (6.9) (6.6) (6.8) (4.8) (6.8)

EPF 0.93 0.76 1.12 1.50 0.96 0.89 1.08 0.89 0.59 0.19 −0.34 −0.15 8.40

(4.2) (3.0) (4.4) (6.6) (4.1) (3.8) (4.8) (4.1) (2.7) (0.9) (−1.5) (−0.7) (3.1)

EPM 0.93 0.69 1.04 1.42 0.76 0.65 0.66 0.45 0.10 −0.23 −0.58 −0.68 5.18

(4.3) (2.8) (4.3) (6.5) (3.5) (3.0) (3.2) (2.2) (0.5) (−1.2) (−3.0) (−3.7) (2.0)

2

In summary, strain 2 had a positive additive genetic effect during the first laying cycle for EPF and EPM, especially at the end of the cycle for EPF. However, strain 2 had a negative Z-chromosome effect throughout the cycle for EPF and should be avoided as a male line. This strain, when used as a male line, was expected to improve egg production at the end of the cycle when hens were housed one per cage (Ledur et al., 2000a). The cross of strains 2 and 3 gave poor heterotic effects at the end of the cycle. Use of strain 1 as a male line and strain 2 as a female line was the best choice to improve egg production late in the cycle, because in addition to the positive additive effect of strain 2, a high heterotic effect in egg production was also expressed by this cross at the end of the cycle. Variance Components. The variance component estimates in the different periods of the first laying cycle, using all genetic groups combined, are presented in Figures 4 and 5. For both traits, the Z-chromosome variance was very close to zero throughout the laying cycle. Although with differently shaped curves, the nonadditive, environmental, and phenotypic variances increased with age for EPF and EPM. The additive variance increased for EPF but decreased at the end of the cycle for EPM (Figure 4). Both residual and phenotypic variances increased with age following the same shape of curve in each of the traits (Figure 5). There was a decrease of these variances in the beginning of the cycle (from period 1 to 2) and an increase after that for EPF. However, residual and phenotypic variances for EPM did not show this decrease at the beginning of the cycle, and these variances increased much faster with age than those for EPF. Differences in patterns of variances with age between EPF and EPM reflect the effects of mortality and morbidity. The trends in variances obtained here with three birds per cage were generally similar to those reported in the literature with one bird per cage. According to Liljedahl

FIGURE 4. A(I) = strain additive, Z(I) = Z-chromosome, and H(I, J) = heterotic variances estimated for egg number of survivors (a) and egg number including mortality and morbidity (b) in each period of the first laying cycle of three strains and their two-way crosses.

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

Average heterosis for crosses, H12 = strains 1 and 2, H13 = 1 and 3, and H23 = 2 and 3. First laying cycle was divided into 12 periods of 28 d each, from age at first egg.

1

AGING ON FITNESS AND NONFITNESS TRAITS

1229

with age followed a pattern similar to those for the additive genetic variance. On average the heritabilities were low but higher for EPF than for EPM.

Egg Quality Traits

FIGURE 5. Residual and phenotypic variances estimated for egg number of survivors (a) and egg number including mortality and morbidity (b) in each period of the first laying cycle of three strains and their two-way crosses.

et al. (1984), genetic (additive, nonadditive, or sex-linked) and environmental variations are expressed to higher degrees as birds grow older. Increases in additive and environmental variations with age for egg number of survivors were also observed by Engstrom at al. (1992). When the measure of egg production involved mortality and morbidity (EPM), the additive variance increased by the middle of the cycle but declined thereafter with birds housed three per cage. However, with birds housed one bird per cage (Ledur et al., 2000a) the additive variance of EPM increased at the end of the cycle. Heritabilities. The changes in heritability estimates over time for EPF and EPM are shown in Figure 6. For both traits there was an increase in heritabilities in the beginning of the cycle, reaching the peak early and starting to decline between periods 4 and 7. The decline of heritability with advancing age appeared to be more rapid for EPM than for EPF. The trend in heritabilities

Performance Means. In the repeated measure analyses of variance for the egg quality traits, the genetic group effect was significant over time. Age was also significant and had a strong effect on genotypic performance. The mean performance of the egg quality traits declined with advancing age, with the exception of EW, which increased curvilinearly with age (Table 4). In general, SG decreased with advancing age, showing a quadratic response, steeper in decline from 350 to 450 d, whereas AH decreased linearly with advancing age. However, these age trends varied, depending on the genetic group. Differences among strains were significant over time for all egg quality traits and were influenced by age (Table 4). For EW these differences changed in a quadratic pattern, whereas for SG they increased linearly with age. For AH, differences between strain 1 and the other strains changed in a quadratic fashion, and differences between strains 2 and 3 changed linearly with advancing age. The commercial lines differed significantly across the cycle

FIGURE 7. Mean heterosis for the egg weight (EW), specific gravity (SG), and albumen height (AH) of six strain crosses during the first laying cycle.

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

FIGURE 6. Heritabilities of egg number of survivors (EPF) and egg number including mortality and morbidity (EPM) in each period of the first laying cycle of three strains and their two-way crosses.

1230

LEDUR ET AL. TABLE 4. Number of observations and estimated means1 of the egg quality traits by period and genetic group Specific gravity3

Egg weight (g)

Genetic group2

Albumen height (mm)

240 d

350 d

450 d

240 d

350 d

450 d

240 d

350 d

450 d

1 2 3 Mean

131 139 139

52.4 59.1 58.8 56.9

55.4 63.4 63.4 60.8

57.5 64.5 64.4 62.2

86.9 86.8 87.6 87.1

85.6 84.3 85.7 85.2

81.2 79.2 81.7 80.7

7.41 7.99 7.69 7.70

6.74 7.45 7.31 7.18

6.36 6.84 6.95 6.72

12 21 13 31 23 32 Mean

139 141 140 141 140 138

CL1 CL2 Mean

138 133

56.2 56.8 57.1 57.1 58.7 59.7 57.6 56.9 58.1 57.5

59.9 61.0 61.0 60.5 63.1 64.6 61.7 61.0 62.2 61.6

62.5 62.6 64.1 62.5 64.1 65.8 63.6 64.2 63.4 63.8

87.4 85.8 89.3 86.5 87.7 86.7 87.2 88.2 88.2 88.2

84.9 84.3 88.3 86.2 85.7 84.5 85.7 86.2 85.9 86.1

80.4 79.5 83.3 82.1 81.0 80.1 81.1 81.0 81.4 81.2

7.69 7.58 7.56 7.55 7.90 7.95 7.70 7.65 8.14 7.89

7.13 6.90 7.05 6.87 7.39 7.40 7.12 7.08 7.61 7.34

6.54 6.31 6.60 6.48 6.96 7.06 6.66 6.53 7.07 6.80

57.4

61.4

63.2

87.4

85.6

81.0

7.74

7.18

6.70

Mean/total

1,519

1

Standard errors for mean egg weight and specific gravity ranged from 0.2 to 0.3, and those for albumen height were from 0.05 to 0.06. 2 CL = commercial line. 3 (SG − 1) × 103, where SG = specific gravity.

only for AH. However, age had an important effect on these differences, not only on AH but also on EW, and showed a curvilinear pattern along the cycle. Mean Heterosis and Reciprocal Effects. The mean heterosis (absolute value) for EW was significant over time (P < 0.01) and increased linearly (P < 0.01) with age (Figure 7). For SG, mean heterosis was also significant across the cycle (P < 0.05); however, it did not change significantly with age. Although not significant on average across periods, mean heterosis for AH increased linearly in magnitude with age but with a negative sign. According to Fairfull and Gowe (1986), egg quality traits are characterized by little or no heterosis. Fairfull et al. (1987) found significant heterosis for EW but not for SG when birds were housed two and three per cage. An increase in mean heterosis with age for EW was also reported by Ledur et al. (2002) with one bird per cage. Current results have shown that heterosis contributes to the increase in EW and deterioration of AH late in the cycle. Differences between reciprocal crosses were important for EW and SG across periods. For AH, only differences between reciprocal crosses involving strains 1 and 2 were significant over time (Table 5). Age had a significant effect on these differences. For EW, differences between Reciprocals 12 and 21 followed a quadratic pattern across the

cycle. Differences between reciprocals involving Strains 1 and 3, and 2 and 3 changed in a linear fashion with age. For SG, differences between Reciprocals 12 and 21 changed in a quadratic fashion, while differences between reciprocal crosses involving Strains 1 and 3 changed linearly with age. For AH, differences between Reciprocals 13 and 31 changed in a quadratic fashion. Reciprocal differences for egg quality traits were also reported by Fairfull and Gowe (1986), who found these differences were low in magnitude, but more important for egg quality traits than heterosis. Similar results to those in the current study were found by Ledur et al. (2002) with one bird per cage, in which reciprocal effects for egg quality traits were also influenced by age. Age had different effects on different genetic groups, depending on the trait. In general, reciprocal effects seemed to increase in magnitude with age for EW and AH and decrease for SG (Table 5) with three birds per cage. Genetic Effects. For EW, additive genetic effects were highly significant across the cycle. Additive genetic effects of strains 2 and 3 were greater than the additive effect of strain 1 in all periods (Figure 8a). This finding implies that strains 2 and 3 had experienced more selection for EW than strain 1. The difference between A2 and A1 and between A3 and A1 changed significantly across the cycle,

TABLE 5. Reciprocal effects for the egg quality traits evaluated during the first laying cycle Reciprocal cross1 12 vs. 21 13 vs. 31 23 vs. 32 1

Specific gravity [(SG − 1) × 103]

Egg weight (g)

Albumen height (mm)

240 d

350 d

450 d

240 d

350 d

450 d

240 d

350 d

450 d

−0.6 ± 0.3* 0.0 ± 0.3 −1.0 ± 0.3**

−1.1 ± 0.4** 0.5 ± 0.4 −1.45 ± 0.4**

−0.1 ± 0.4 1.6 ± 0.4** −1.75 ± 0.4**

1.6 ± 0.3** 2.7 ± 0.3** 1.0 ± 0.3**

0.6 ± 0.4 2.1 ± 0.4** 1.2 ± 0.4**

0.9 ± 0.5* 1.2 ± 0.5** 1.0 ± 0.5*

0.11 ± 0.07 0.01 ± 0.07 −0.05 ± 0.07

0.24 ± 0.07** 0.18 ± 0.07** −0.01 ± 0.07

0.23 ± 0.08** 0.12 ± 0.08 −0.10 ± 0.08

ij vs. ji contrasts comparing cross of males from strain i and females from strain j with cross of males from strain j with females from strain i.

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

n

AGING ON FITNESS AND NONFITNESS TRAITS

1231

FIGURE 9. A(I) = strain additive (a), Z(I) = Z-chromosome (b), and H(I, J) = heterotic (c) genetic effects estimated for specific gravity, in each period of the first laying cycle of three strains and their twoway crosses.

increasing curvilinearly with advancing age. Similar results were found when birds were housed one per cage (Ledur et al., 2002). However, the only significant age change was for the difference between A3 and A1, which increased linearly with advancing age. Differences in Z-chromosome effects on EW were not significant on average during the first laying cycle. However, these differences were significantly influenced by age in a curvilinear fashion (Figure 8b). Difference between Z2 and Z1 increased from 350 to 450 d, indicating a negative sex-linked effect of strain 2 on EW at the end of the cycle. Evidence of important sex-linked effects on EW was also reported by Hagger (1985) through analysis of reciprocal effects. Heterotic effects on EW were highly significant across periods, with the exception of H23. Heterosis from crosses

of strains 2 and 3 was expected to be lower than heterosis involving crosses with strain 1 because those strains have greater additive effects on EW than strain 1. The heterotic effects changed with time in different ways (Figure 8c; Table 6). The H12 increased linearly with age (P < 0.01), and H13 increased at the end of the cycle, following a quadratic pattern (P < 0.01). No significant age change on H23 was observed during the first laying cycle. Similar results were obtained with one bird per cage (Ledur et al., 2002), except that H12 increased curvilinearly with age. For EW, strains 2 and 3 had greater additive genetic effects during the first laying cycle than strain 1 (Figure 8). When strain 3 was used as a male line in a cross with strain 2, the highest EW was produced in the first laying cycle, even though heterotic effects were poor when compared with those of the other crosses. Additive effects

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

FIGURE 8. A(I) = strain additive (a), Z(I) = Z-chromosome (b), and H(I, J) = heterotic (c) genetic effects estimated for egg weight, in each period of the first laying cycle of three strains and their two-way crosses.

1232

LEDUR ET AL. TABLE 6. Heterosis and percentage heterosis, in parenthesis, for each type of cross,1 in three different periods of the first laying cycle

Egg weight

Specific gravity

Albumen height

240 350 450 240 350 450 240 350 450

d d d d d d d d d

H12

H13

H23

0.8 (1.4) 1.1 (1.8) 1.6 (2.5) −0.3 (−0.3) −0.4 (−0.4) −0.3 (−0.3) −0.07 (−0.8) −0.08 (−1.1) −0.18 (−2.7)

1.5 (2.7) 1.4 (2.3) 2.4 (3.9) 0.7 (0.7) 1.6 (1.9) 1.3 (1.5) 0.01 (0.1) −0.07 (−0.9) −0.12 (−1.7)

0.3 (0.4) 0.5 (0.7) 0.5 (0.8) 0 (0) 0.1 (0.1) 0.1 (0.1) 0.09 (1.1) 0.02 (0.2) 0.12 (1.7)

Average heterosis for crosses, H12 = strains 1 and 2, H13 = 1 and 3, and H23 = 2 and 3.

1

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

played the major role in determining differences in EW among genetic groups. For SG, differences between A3 and A1 were highly significant across periods. The A3 was higher than A1 throughout the cycle, but this difference decreased linearly with age (Figure 9a). Although differences between A2 and A1 were not significant on average during the laying cycle, they were significantly influenced by age. At 240 d, A2 was higher than A1, but at 450 d the opposite happened, showing a linear trend with age. The Z-chromosome effects were highly significant for SG in the first laying cycle. The Z1 was higher than Z2 and Z3 across the cycle. These differences decreased linearly with age (Figure 9b). Heterotic effects on SG were important only for crosses with strains 1 and 3 (Figure 9c; Table 6). They were highly significant during the whole cycle and increased curvilinearly with age (P < 0.01). The other heterotic effects did not change significantly with age. For AH, additive genetic effects were highly significant across periods. The A2 and A3 were higher than A1 during the first laying cycle (Figure 10a). These differences changed with age in a curvilinear fashion (P < 0.01). The difference between A3 and A1 increased with advancing age, whereas the difference between A2 and A1 was maximum at 350 d. The Z-chromosome effects were also highly significant across the cycle. The Z1 was higher than Z2 and Z3, and this difference increased linearly with age for Z2 and in a curvilinear fashion for Z3 (Figure 10b). This result indicates that strain 1 should be preferred as a male line to improve AH late in the cycle. On average, the only significant heterotic effect across periods was H12, showing a negative effect on AH, which increased linearly with age (Figure 10c; Table 6). The H13 also changed linearly and had an increase in magnitude with advancing age, being negative late in the cycle. The H23 was positive and changed curvilinearly with age, decreasing at 350 d but increasing at 450 d. Strain 3 had more desirable genes when considering all egg quality traits analyzed in the present study. Strain 2 had also high additive genetic effects, with the exception of SG late in the cycle. The cross of strains 2 and 3 had positive heterotic effects for all traits during the evaluated period. Although these strains had negative Z-chromosome effects for SG and AH, the use of strain 2 as a male line seemed to result in a better equilibrium of the egg

FIGURE 10. A(I) = strain additive (a), Z(I) = Z-chromosome (b), and H(I, J) = heterotic (c) genetic effects estimated for albumen height, in each period of the first laying cycle of three strains and their twoway crosses.

1233

AGING ON FITNESS AND NONFITNESS TRAITS TABLE 7. Variance component estimates and heritabilities for egg quality traits in three different periods of the first laying cycle Egg weight Variance σ2A σ2Z σ2H σ2e σ2p 2 h

1

Specific gravity

Albumen height

240 d

350 d

450 d

240 d

350 d

450 d

240 d

350 d

450 d

13.06 0.04 0.68 7.18 20.96 0.63

19.64 0.03 0.62 9.46 29.75 0.66

19.11 0.21 1.82 12.41 33.55 0.58

2.83 1.81 0.13 7.49 12.26 0.38

1.88 0.99 0.93 11.87 15.67 0.18

2.43 0.42 0.46 14.60 17.91 0.16

0.11 0.00 0.00 0.34 0.45 0.24

0.24 0.01 0.00 0.35 0.60 0.42

0.16 0.01 0.02 0.41 0.60 0.28

1 2 σA = additive variance; σ2z = Z-chromosome variance; σ2H = heterotic variance; σ2e = environmental variance; σ2p = phenotypic variance; h2 = heritabilities.

typic variances increased with age for all egg quality traits during the first laying cycle. In general, heterosis was more important for egg production than for egg quality traits and normally increased in magnitude with advancing age. Mortality seemed to have a negative effect on heterosis for egg production when hens were housed three per cage. Reciprocal cross effects were more important in egg quality than egg production traits and were also influenced by age. Additive and heterotic effects were most important to explain variation in egg production traits among genetic groups, whereas for the egg quality traits, additive and Z-chromosome effects were most important. In general all these effects were affected by age, and the pattern of age changes showed clear differences among strains for all traits considered. This observation was also made by Ledur et al. (2000a,b) and Ledur et al. (2002) with one bird per cage for several traits. Strains differ in their response to aging, and the general increase of heterosis, additive, and sex-linked effects would be a successful strategy for genetic groups to maintain egg production and egg quality traits in advancing ages. Although heritabilities for almost all traits declined with age, these seemed to be due to the faster increase in environmental variance when compared with the increase in additive and Z-chromosome variances, which was also observed by Liljedahl et al. (1984). Genetic variance of egg production and egg quality traits increased with age when birds were housed three per cage. Zchromosome variance contributed to the increase of genetic variance of EW and AH. Improvement in lifetime performance may be obtained by selecting older birds. This possibly favors individuals with better DNA repair capacity or those who had more genes turned on or off as needed during the course of aging. However, selection at older ages for egg quality traits seems not to be as advantageous as for egg production traits. Egg production is a fitness trait and is expected to be much more affected by the environment than would nonfitness traits. Also, the relative increase in additive variation with age was larger for egg production than for egg quality traits. Therefore, selection for egg quality traits could be performed by evaluating the genetic stocks at the beginning of the cycle.

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

quality traits during the first laying cycle and was similar to or better than the commercial lines. With the exception of SG, the results showed, on average, increased divergence with age in additive, Z-chromosome, and heterotic effects among different genetic groups. The pattern of age changes in these genetic effects varied by genetic group and showed differences by genetic group in response to aging for all egg quality traits, including SG. Similar results, obtained for egg quality traits, were reported by Ledur et al. (2002) for hens housed one per cage. Most of the heterotic effects increased in magnitude with age and indicated that even for egg quality traits heterosis was expressed at higher levels later in life, although not always with a positive effect. Variance Components and Heritabilities. The variance component estimates and heritabilities for EW, SG, and AH in the different periods of the first laying cycle, using all genetic groups combined, are presented in Table 7. For EW, the additive, Z-chromosome, nonadditive, environmental, and phenotypic variances seemed to increase with age. For SG, the heterotic, environmental, and phenotypic variances increased with age, whereas the Zchromosome variance decreased along the first laying cycle. The additive variance decreased from 240 to 350 d but increased again at 450 d. Variances estimated for AH seemed to increase with age but with different patterns. The heritability estimates for EW seemed to decrease late in the cycle. This result was also shown by Liljedahl et al. (1984), Engstrom et al. (1986), and Grunder et al. (1989) within strains. The heritabilities of SG and AH were in the range of those summarized by Gowe and Fairfull (1995), decreasing with age for SG, but increasing from 240 to 350 d and decreasing again at 450 d for AH. Reduction in heritabilities with age were also found by Liljedahl et al. (1984) for almost all the traits studied and by Grunder et al. (1989) for SG estimated within strain. Besides the additive genetic effects, the sex-linked effects reported herein were also of great importance in explaining variation on egg quality traits among different genetic groups. The additive and Z-chromosome variances did not always increase with age, as was the case for SG. Their age trends varied, depending on the trait. However, the nonadditive, environmental, and pheno-

1234

LEDUR ET AL.

ACKNOWLEGMENTS The authors thank Shaver Poultry Breeding Farms (Cambridge, ON, Canada), in particular Al Kulenkamp, for providing the birds for this study and for financial support. Valuable comments on the written material were made by Bob Gowe (retired from Animal Research Centre, Ottawa, ON, Canada). Alan Grunder (retired from Animal Research Centre) also provided support to the operation of this project.

REFERENCES

Downloaded from http://ps.oxfordjournals.org/ at Bibliotheque de l'Universite Laval on July 6, 2015

Adams, A. W. 1974. Replacing versus not replacing laying hens removed from cages. Poult. Sci. 53:2092–2095. Barlow, R. 1981. Experimental evidence for interaction between heterosis and environment in animals. Anim. Breed. Abstr. 49:715–737. Bell, D. D. 1981. Cage selection and management. Feedstuffs 53(9):20–22. Champion, L. R., and H. C. Zindel. 1968. Performance of layers in single and multiple bird cages. Poult. Sci. 47:1130–1135. Engstrom, G., L. E. Liljedahl, M. Wilhelmson, and K. Johansson. 1992. The pattern of genetic and environmental variation in relation to ageing in laying hens. Genet. Sel. Evol. 24:265–275. Engstrom, G., C. Weyde, and L. E. Liljedahl. 1986. Genetic correlations and heritabilities for frequency of cracked eggs, egg number and egg weight in laying hens. Br. Poult. Sci. 27:55–61. Fairfull, R. W. 1990. Heterosis. Pages 913–933 in Poultry Breeding and Genetics. R. D. Crawford, ed. Elsevier Science, Amsterdam. Fairfull, R. W., and R. S. Gowe. 1986. Use of breed resources for poultry egg and meat production. Proceedings of the 3rd World Congress on Genetics Applied to Livestock Production 10:242–256, Lincoln, NE. Fairfull, R. W., R. S. Gowe, and J. A. B. Emsley. 1983. Diallel cross of six long-term selected Leghorn strains with emphasis on heterosis and reciprocal effects. Br. Poult. Sci. 24:133–158. Fairfull, R. W., R. S. Gowe, and J. Nagai. 1987. Dominance and epistasis in heterosis of White Leghorn strain crosses. Can. J. Anim. Sci. 67:663–680.

Fairfull, R. W., L. E. Liljedahl, and R. S. Gowe. 1999. Age-regulated expression of genetic and environmental variation in fitness traits. 2. Genetic effects and variances for viability among crosses from a factorial mating of six selected Leghorn strains. Can. J. Anim. Sci. 79:269–276. Gowe, R. S., and R. W. Fairfull. 1995. Breeding and genetics of egg laying chickens. Pages 435–455 in World Animal Science—C. Production System Approach. Poultry Production. P. Hunton, ed. Elsevier, Amsterdam. Grunder, A. A., R. M. G. Hamilton, R. W. Fairfull, and B. K. Thompson. 1989. Genetic parameters of egg shell quality traits and percentage of eggs remaining intact between oviposition and grading. Poult. Sci. 68:46–54. Hagger, C. 1985. Line and crossing effects in a diallel mating system with highly inbred lines of White Leghorn chickens. Theor. Appl. Genet. 70:555–560. Henderson, C. R., 1953. Estimation of variance and covariance components. Biometrics 9:226–252. Ledur, M. C., R. W. Fairfull, I. McMillan, and L. Asselstine. 2000a. Genetic effects of aging on egg production traits in the first laying cycle of White Leghorn strains and strain crosses. Poult. Sci. 79:296–304. Ledur, M. C., R. W. Fairfull, I. McMillan, R. S. Gowe, and L. Asselstine. 2000b. Genetic effects of ageing on fertility and hatchability in the first laying cycle of three White Leghorn strains and their two-way crosses. Br. Poult. Sci.41:552–561. Ledur, M. C., L. E. Liljedahl, R. W. Fairfull, I. McMillan, and L. Asselstine. 2002. Genetic effects of aging on egg quality traits in the first laying cycle of White Leghorn strains and strain crosses. Poult. Sci. 81:1439–1447. Liljedahl, L. E., R. W. Fairfull, and R. S. Gowe. 1999. Age-regulated expression of genetic and environmental variation in fitness traits. 1. Genetic effects and variances for egg production among crosses from a factorial mating of six selected Leghorn strains. Can. J. Anim. Sci. 79:253–267. Liljedahl, L. E., J. S. Gavora, R. W. Fairfull, and R. S. Gowe. 1984. Age changes in genetic and environmental variation in laying hens. Theor. Appl. Genet. 67:391–401. Muir, W. M. 1985. Relative efficiency of selection for performance of birds housed in colony cages based on performance in single bird cages. Poult. Sci. 64:2239–2247. Muir, W. M. 1996. Group selection for adaptation to multiplehen cages: selection program and direct responses. Poult. Sci. 75:447–458. SAS Institute. 1989. SAS/STAT User’s Guide. Version 6, 4th ed. SAS Institute Inc., Cary, NC.