- Email: [email protected]
Effect of long-term divergent selection on growth characteristics in Japanese quail
Effect of Long-Term Divergent Selection on Growth Characteristics in Japanese Quail1 S. E. Aggrey,2 G. A. Ankra-Badu, and H. L. Marks Poultry Genetics and Biotechnology Laboratory, Department of Poultry Science, University of Georgia, Athens, Georgia 30602-2772 females. The dynamics of the growth curve parameters indicate that selection for decreased 4-wk BW shifted the growth curve for females as well as altering the trajectories of growth in both sexes. However, selection for increased 4-wk BW only resulted in altering the trajectory of growth in the males. Long-term selection resulted in asymmetry of response in the low and high lines. In addition, different genes may respond differently to the same selection pressure in opposite directions. The use of the 4-parameter Richards model to analyze growth data from such an experiment provided a better understanding of how selection can alter the rate and trajectory of growth to affect the genetically determined growth potential of quail. Consequently selection for increased or decreased 4-wk BW affected the sexes differently.
(Key words: divergent selection, quail, asymmetric response, growth rate, growth trajectory) 2003 Poultry Science 82:538–542
a growth curve, has had limited use in poultry (Knizetova et al., 1991a,b; Hyankova et al., 2001; Aggrey, 2002). The logistic and Gompertz models have fixed growth trajectories with points of inflection at about 50 and 37% of the asymptote, respectively (Ricklefs, 1968). These parameter models are special cases of the more flexible Richards model, which has a variable point of inflection specified by the shape parameter, m (Richards, 1959). When the m value has a value of 2.0 or 1.0, the Richards model is equivalent to the logistic or Gompertz models, respectively. There is a general agreement in literature about the effect of selection for weight or growth on increased fatness (Eisen, 1976; Leclercq et al., 1989; Mignon-Grasteau et al., 2000); however, the effect on selection on altering the shape of the growth curve or growth trajectory has been conflicting. Results from long-term selection experiment in mice indicated that selection for size or growth rate had little effect on shape of the growth curve (Timon and Eisen, 1969; Eisen, 1976). Anthony et al. (1996) also reported that the shape of the growth curve was not affected by short-term selection in Japanese quail at any age. Hyankova et al. (2001), on the contrary, reported alterations in growth curve shape in quail when selected for relative growth in the short-term.
INTRODUCTION The application of quantitative genetics and biometrical methods to long-term selection programs have been well documented (Siegel, 1962; Marks, 1971, 1978; Anthony et al., 1991; Hyankova et al., 2001). Consequently, the growth patterns of experimental lines selected for growth have been studied (Marks, 1978; Anthony et al., 1991; Akbas and Oguz, 1998; Mignon-Grasteau et al., 2000). Although growth curves have been studied for a variety of reasons, changes in the shape of the growth curve resulting from selection for BW have received only limited investigation. Modeling growth data to study inherent differences in growth patterns have not been fully utilized because most poultry growth data have been fitted with the logistic and Gompertz equations, which assume a fixed growth shape. The generalized Richards model, which also assesses the shape of
2003 Poultry Science Association, Inc. Received for publication August 16, 2002. Accepted for publication December 12, 2002. 1 Supported by state and Hatch funds allocated to the Georgia Agricultural Experimental Stations of the University of Georgia. 2 To whom correspondence should be addressed: saggrey@arches. uga.edu.
538
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015
ABSTRACT The current study was undertaken to examine the effect of long-term selection for 4-wk BW on growth characteristics in divergent lines of Japanese quail and their control. Growth rate was significantly higher in males than females in all the lines. There was a significant increase in growth rate of the females selected for increased 4-wk BW over the control females, as well as a significant decline in growth rate of males selected for decreased 4-wk BW compared to their control counterparts. It appeared that selection for increased 4-wk BW did not alter the rate of growth in the males compared to the control males; however, in the females, selection for increased 4-wk BW resulted in an increase in growth rate. On the other hand, selection for decreased 4-wk BW resulted in a decline in growth rate in males but not in the
GROWTH PARAMETERS AND DIVERGENT SELECTION
MATERIALS AND METHODS Growth data were collected from Japanese quail lines selected for increased (high) or decreased (low) 4-wk BW for 30 generations (Darden and Marks, 1988) and their randombred control. Each line was propagated by a single hatch of approximately 200 quail using 30 paired matings. Quail from the control line were intermingled with the divergently selected lines during the rearing period. Quail chicks were hatched, wing-banded, and placed in quail battery brooders. All quail had ad libitum access to a 28% CP and 2,947 kcal ME/kg of diet and water. Hatch weights were collected on all birds, as well as BW each week until 8 wk of age. The birds were sexed at 6 wk of age according to their plumage color pattern.
Growth Model To estimate the expected BW at a specific age, a 4parameter growth curve modified from Richards (1959) growth function (Sugden et al., 1981) was fitted to the BW data collected. The growth curve was of the following form Wt = WA[1 − (1 − m)exp[−K (t − ti)/mm/(1−m)]]1/(1−m) where Wt is the weight of bird at time t, WA is the asymptotic (mature) BW, K is the maximum relative growth (per week), ti is the age at maximum rate of growth (week), and m is a shape parameter, with the property that m1/(1−m) is relative weight at ti. Individual growth curve parameters for the Richards curve were estimated for each bird using PROC NLIN (Marquart algorithm) (SAS Institute Inc., 1996). Differences among lines and sex within lines for the Richards model parameters were tested using PROC GLM (SAS Institute Inc., 1996).
RESULTS AND DISCUSSION Means and standard deviation for BW at various ages of both sexes for each line are presented in Table 1. As expected BW of the high and low lines were significantly different from that of the control line. Body weights of both sexes were similar from hatch until 4 wk of age after which the females become heavier than the males. The Richards growth model fitted to data on sex within lines are shown in Figure 1. The coefficients for the parameters of the fitted curve are presented in Table 2. The growth patterns of both sexes within lines were not identical; therefore, data were analyzed separately for each sex. Studies have shown that Japanese quail exhibit sexual dimorphism (Aggrey and Cheng, 1994; Du Preez and Sales, 1997; Hyankova et al., 2001). In chickens, Carte and Siegel (1970) eliminated sexual dimorphism in response to selection for growth by adjusting for scaling effects. Fitting the growth model to log-transformed data only reduced the residual variance but did not eliminate sexual dimorphism. Growth rate was higher in males (P < 0.05) than females in all the lines. Sex differences in growth rate has also been reported in European quail (Coturnix coturnix) (Du Preez and Sales, 1997) and Japanese quail (Hyankova et al., 2001). There was no difference between the rate of growth of the males of the high and the control lines and between females of the low and control lines. On the other hand, there was a significant increase in growth rate of the high line over the control line for females, as well as a significant decline in growth rate of males of the low line compared to their control counterpart. It appears that selection for increased 4-wk BW did not alter the rate of growth in the males compared to the control males; however, in the females, selection for increased 4-wk BW resulted in an increase in growth rate. Conversely, selection for decreased 4-wk BW resulted in a decline in growth rate in males but not in the females. The response of rate of growth to divergent selection for 4-wk BW appeared to be dependent on sex and direction of the selection pressure. The age at maximum growth occurred at approximately the same age in both the control and the high lines regardless of sex (Table 2). A similar result was reported by Anthony et al. (1986). However, this contradicts Marks (1978) and Hyankova et al. (2001) whose Japanese quail selection experiments observed a push back of the age of maximum growth in the upward selected lines as compared to the control lines. For the low lines, the age at maximum growth was about 5 d later, suggesting a prolonged growing period. In the present study, selection for decreased 4-wk BW resulted in a shift of the growth curve. Popular growth models, such as the Gompertz and the logistic, have fixed growth shapes with inflection points at 37 and 50% of the asymptote, respectively. These models are special cases of the Richards model, which has a variable point of inflection defined by the shape or growth trajectory parameter, m. When the
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015
Brisbin et al. (1987) suggested that the shape of a growth curve has a greater propensity to change in response to environmental changes than the asymptotic weight or growth rate and may be used to study the effects of environmental stress on growth. This inherently implies that growth models with fixed shapes may not contribute to the understanding of the effects of factors like dietary and environmental changes on growth. Aggrey (2002) further suggested that shape of the growth curve may reflect the architecture of body composition and could therefore be used to manipulate the desired body composition at a given age. Therefore, selection for changes of the shape of a growth curve may be a useful tool for improving the efficiency of lean meat production. The objective of the present study was to investigate the effect of long-term divergent selection for 4-wk BW on growth characteristics of Japanese quail and compare these characteristics with the nonselected randombred control base population using the generalized growth model.
539
540
AGGREY ET AL. TABLE 1. Mean and standard deviation for BW at different ages in three experimental Japanese quail lines1 BW Low Age (wk)
Male (n = 96)
Control Female (n = 105)
Male (n = 102)
High Female (n = 81)
Male (n = 84)
Female (n = 83)
g 0 1 2 3 4 5 6 7 8
4.74 11.04 20.99 35.31 48.19 60.75 67.86 72.09 73.41
± ± ± ± ± ± ± ± ±
0.46 2.27 4.59 6.75 8.19 8.35 7.51 5.57 5.15
4.74 11.64 21.67 36.34 49.82 62.96 71.94 81.91 87.19
± ± ± ± ± ± ± ± ±
0.46 2.39 5.07 7.19 8.15 8.11 8.39 9.62 9.18
5.41 18.92 39.91 64.07 84.87 96.13 100.39 101.98 104.06
± ± ± ± ± ± ± ± ±
0.51 2.78 4.57 5.83 6.44 7.10 6.16 6.10 6.05
5.43 19.06 40.23 64.66 87.14 101.94 116.59 127.82 130.80
± ± ± ± ± ± ± ± ±
0.47 3.28 5.79 7.07 7.54 7.94 10.69 10.78 9.58
7.4 33.91 79.58 132.71 171.37 195.72 205.07 208.39 210.85
± ± ± ± ± ± ± ± ±
0.74 4.14 7.16 10.15 11.08 13.21 12.15 13.71 14.57
7.38 33.63 81.56 138.29 179.18 205.56 219.13 233.89 240.94
± ± ± ± ± ± ± ± ±
0.67 3.44 8.37 11.09 10.99 12.21 12.36 15.08 14.55
1 Control = unselected randombred; Low = selected for decreased 4-wk BW; High = selected for increased 4wk BW.
250
Body weight (g)
200 150 100 50 0 0
2
4
6
8
Age (wk) Control: Male High: Female
Control: Female Low: Male
High: Male Low: Female
FIGURE 1. Growth curve of Japanese quail lines divergently selected for 4-wk body weight and their controls.
selection criterion only resulted in altering the trajectory of growth in the males. Alteration in the shape of the curve should be due to the degree of genetic flexibility of the shape parameter and the pleiotropic relationship between the shape parameter and the selection criterion. As postulated by Brisbin et al. (1986), growth shape is likely to be related to the relative timing and functioning of intrinsic biological processes that relate to the assimilation of food and its synthesis into new body tissues. It would therefore be possible to genetically alter the shape of the curve, which would consequently alter body composition. Aggrey (2002) reported a correlation between the shape parameter and abdominal fat in chickens. Brisbin et al. (1986) further suggested that the shape parameter has a greater propensity to change in response to environmental changes than the asymptotic BW or growth rate and can be used to assess effect of nongenetic changes as well. It should therefore not be assumed that the growth of a given species may be characterized by fixed shape curve under all genetic and environmental conditions and that even under uniform conditions, a single curve shape may not characterize both sexes. Data from Table 1 and Figure 1 show that the response to divergent selection for 4-wk BW was asymmetrical. The asymmetrical response to selection was probably due to the duration of the selection experiment, because the same experimental lines exhibited symmetrical response through 11 generations of selection (Darden and Marks, 1988). A 12-generation divergent selection experiment reported by Anthony et al. (1986) also showed symmetry in response to 4-wk BW. After 30 generations of selection, the 4-wk BW means of the lines averaged over sexes were 49 g (low), 86 g (control), and 175 g (high). The absolute difference between the high and the control line was more than twice of that between the low and the control. The absolute difference of the divergent lines from the control is even more magnified in the asymptotic BW. Asymptotic BW increased by about 100 and 70%, respectively, in the male and female
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015
growth curve shape value is 2.0 or 1.0, the Richards model is equivalent to the logistic or Gompertz models, respectively. Females had higher (P < 0.05) m values than their male counterparts with lines. The shape value for the females for all the lines were close to one. A shape value of one from the Richards model indicate that the females followed the Gompertz model, and selection for either increased or decreased 4-wk BW did not change the shape of the growth curve in females. However, for the males, the shape value increased in the low line and decreased in the high line when compared to the control males. The control males had a growth curve shape between the logistic and the Gompertz models. The selection for decreased 4-wk BW altered the shape of the curve to resemble more of the Gompertz curve, whereas selection for increased 4-wk BW modified the shape towards the logistic curve. The dynamics of the growth curve parameters indicate that selection for decreased 4-wk BW shifted the growth curve in the females as well as altering the trajectories of growth in both sexes. However, in the high line, the
541
GROWTH PARAMETERS AND DIVERGENT SELECTION TABLE 2. Estimates of Richards growth curve parameters in three experimental Japanese quail lines Low Growth parameters
Male
Asymptotic weight, A (g) Shape parameter, m Rate of growth, k (g/wk) Age at maximum growth, t (wk)
75.92 1.81 0.73 3.11
± ± ± ±
0.89a 0.17a 0.06a 0.09a
Control Female
101.41 1.09 0.39 3.22
± ± ± ±
3.28bc 0.14b 0.04b 0.12a
Male 104.42 1.67 0.89 2.34
± ± ± ±
0.45bc 0.09c 0.04c 0.05b
1
High Female
144.01 0.93 0.43 2.44
± ± ± ±
2.31d 0.09b 0.03b 0.09b
Male 212.82 1.42 0.85 2.23
± ± ± ±
0.10e 0.08d 0.03c 0.05b
Female 246.21 0.98 0.60 2.11
± ± ± ±
1.61f 0.06b 0.02d 0.06b
Growth parameters with no common superscript are significantly different (P ≤ 0.05). Control = unselected randombred; Low = selected for decreased 4-wk BW; High = selected for increased 4-wk BW.
a–f 1
ACKNOWLEDGMENT The authors thank Cheryl Pearson Gresham, Department of Poultry Science, University of Georgia, Athens, GA, for her technical assistance and collection of data.
REFERENCES Aggrey, S. E. 2002. Comparison of three non-linear and spline regression models for describing chicken growth curves. Poult. Sci. 81:1782–1788.
Aggrey, S. E., and K. M. Cheng. 1994. Animal model analysis of genetic (co)variance of growth traits in Japanese quail. Poult. Sci. 73:1822–1828. Akbas, Y., and I. Oguz. 1998. Growth curve parameters of line of Japanese quail (Coturnix coturnix japonica), unselected and selected for four-week body weight. Arch. Geflugelkd. 62:104–109. Anthony, N. B., K. E. Nestor, and H. L. Marks. 1996. Shortterm selection for four-week body weight in Japanese quail. Poult. Sci. 71:1985–1993. Anthony, N. B., D. A. Emmerson, K. E. Nestor, W. L. Bacon, P. B. Siegel, and E. A. Dunnington. 1991. Comparison of growth curves of weight selected populations of turkeys, quail and chickens. Poult. Sci. 70:13–19. Anthony, N. B., K. E. Nestor, and W. L. Bacon. 1986. Growth curves of Japanese quail as modified by divergent selection for 4-week body weight. Poult. Sci. 65:1825–1833. Barbato, G. F. 1992. Divergent selection for exponential growth rate at fourteen or forty-two days of age. 1. Early responses. Poult. Sci. 71:1985–1993. Brisbin, I. L., Jr., C. T. Collins, G. C. White, and D. A. McCallum. 1987. A new paradigm for the analysis and interpretation of growth data: The shape of things to come. Auk 104:552–554. Brisbin, I. L., Jr., G. C. White, P. B. Bush, and L. A. Mayack. 1986. Sigmoid growth analysis of wood ducks: The effect of sex, dietary protein and cadmium on parameters of the Richards model. Growth 50:41–50. Carte, I. F., and P. B. Siegel. 1970. Scaling effects and the inheritance of juvenile body weight in chickens. Can. J. Genet. Cytol. 12:724–727. Darden J. R., and H. L. Marks. 1988. Divergent selection for growth in Japanese quail under split and complete nutritional environments. 1. Genetic and correlated responses to selection. Poult. Sci. 67:519–529. Du Preez, J. J., and J. Sales. 1997. Growth rate of different sexes in European quail (Coturnix coturnix). Br. Poult. Sci. 38:314–315. Eisen, E. J. 1976. Results of growth curve analyses in mice and rats. J. Anim. Sci. 42:1108–1023. Hyankova, L., H. Knizetova, L. Dedkova, and J. Hort, 2001. Divergent selection for shape of growth curve in Japanese quail. 1. Responses in growth parameters and food conversion. Br. Poult. Sci. 42:583–589. Knizetova, H., J. Hyanek, B. Knize, and J. Roubicek. 1991a. Analysis of growth curves in fowl. I. Chickens. Br. Poult. Sci. 32:1027–1038. Knizetova, H., J. Hyanek, B. Knize, and H. Prochazkova. 1991b. Analysis of growth curves in fowl. II. Ducks. Br. Poult. Sci. 32:1039–1053. Leclercq, B., G. Guy, and F. Rudeaux. 1989. Growth characteristics and lipid distribution in two lines of chickens selected for low or high abdominal fat. Genet. Sel. Evol. 21:69–80. Marks, H. L. 1971. Selection for four-week body weight in Japanese quail under two nutritional environments. Poult. Sci. 50:931–937.
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015
high lines but declined by only about 25% in the low lines. A similar observation was also made by Barbato (1992). Body weight at the same age in long-term divergently selected lines may be considered as different traits. Genes may respond differently to the same selection pressure under different internal physiological environment, and the forces of natural selection may act differently on the divergent lines as well. It appeared that changes in the asymptotic weights were a result of changes growth rate, shape parameter, and hatch weight during the divergent selection for 4wk BW. Selection for increased 4-wk BW has resulted in a correlated increase in hatch weight for both sexes. For the high lines, hatch weight coupled with increased rate of growth contributed to affect the asymptotic BW in females, whereas in the males, it was increased hatch weight and altered growth trajectory, which compounded to influence asymptotic BW. Selection for decreased 4-wk BW also resulted in a correlated reduction in hatch weight; however, in the low lines, the change in the female shape parameter was marginal, and the reduction in the growth rate was not significant to cause a substantial deviation from the control. Reduced hatch weight and concurrent decline in growth rate and alteration of growth trajectory played roles in determining the asymptotic BW in the male low line. The use of a general growth model (Richards) to analyze growth data from such an experiment provided a better understanding of how selection altered the rate and trajectory of growth to affect the genetically determined growth potential of quail. Consequently, selection for increased or decreased 4-wk BW affected the sexes differently. Long-term divergent selection may cause asymmetry in response as a result of possibly different genes responding differently to the selection pressure.
542
AGGREY ET AL.
Marks, H. L. 1978. Growth curve changes associated with longterm selection for body weight in Japanese quail. Growth 42:129–140. Mignon-Grasteau, S., M. Piles, L. Varona, H. de Rochambeau, J. P. Poivey, A. Blasco, and C. Beaumont. 2000. Genetic analysis of growth curve parameters for male and female chickens resulting from selection on shape of curve. J. Anim. Sci. 78:2515–2524. Richards, F. J. 1959. A flexible growth function for empirical use. J. Exp. Botany 10:290–300. Ricklefs, R. E. 1968. Patterns of growth in birds. Ibis 110:419–451.
SAS Institute. 1996. SAS User’s Guide. Version 6.12. SAS Institute Inc., Cary, NC. Siegel, P. B. 1962. Selection for body weight at eight weeks of age 1. Short term responses and heritabilities. Poult. Sci. 41:954–962. Sugden, L. G., E. A. Driver, and M. C. S. Kingsley. 1981. Growth and energy consumption by captive mallards. Can. J. Zool. 59:1567–1570. Timon, V. M., and E. J. Eisen. 1969. Comparison of growth curves of mice selected and unselected for postweaing gain. Theor. Appl. Genet. 39:345–351.
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015
(Key words: divergent selection, quail, asymmetric response, growth rate, growth trajectory) 2003 Poultry Science 82:538–542
a growth curve, has had limited use in poultry (Knizetova et al., 1991a,b; Hyankova et al., 2001; Aggrey, 2002). The logistic and Gompertz models have fixed growth trajectories with points of inflection at about 50 and 37% of the asymptote, respectively (Ricklefs, 1968). These parameter models are special cases of the more flexible Richards model, which has a variable point of inflection specified by the shape parameter, m (Richards, 1959). When the m value has a value of 2.0 or 1.0, the Richards model is equivalent to the logistic or Gompertz models, respectively. There is a general agreement in literature about the effect of selection for weight or growth on increased fatness (Eisen, 1976; Leclercq et al., 1989; Mignon-Grasteau et al., 2000); however, the effect on selection on altering the shape of the growth curve or growth trajectory has been conflicting. Results from long-term selection experiment in mice indicated that selection for size or growth rate had little effect on shape of the growth curve (Timon and Eisen, 1969; Eisen, 1976). Anthony et al. (1996) also reported that the shape of the growth curve was not affected by short-term selection in Japanese quail at any age. Hyankova et al. (2001), on the contrary, reported alterations in growth curve shape in quail when selected for relative growth in the short-term.
INTRODUCTION The application of quantitative genetics and biometrical methods to long-term selection programs have been well documented (Siegel, 1962; Marks, 1971, 1978; Anthony et al., 1991; Hyankova et al., 2001). Consequently, the growth patterns of experimental lines selected for growth have been studied (Marks, 1978; Anthony et al., 1991; Akbas and Oguz, 1998; Mignon-Grasteau et al., 2000). Although growth curves have been studied for a variety of reasons, changes in the shape of the growth curve resulting from selection for BW have received only limited investigation. Modeling growth data to study inherent differences in growth patterns have not been fully utilized because most poultry growth data have been fitted with the logistic and Gompertz equations, which assume a fixed growth shape. The generalized Richards model, which also assesses the shape of
2003 Poultry Science Association, Inc. Received for publication August 16, 2002. Accepted for publication December 12, 2002. 1 Supported by state and Hatch funds allocated to the Georgia Agricultural Experimental Stations of the University of Georgia. 2 To whom correspondence should be addressed: saggrey@arches. uga.edu.
538
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015
ABSTRACT The current study was undertaken to examine the effect of long-term selection for 4-wk BW on growth characteristics in divergent lines of Japanese quail and their control. Growth rate was significantly higher in males than females in all the lines. There was a significant increase in growth rate of the females selected for increased 4-wk BW over the control females, as well as a significant decline in growth rate of males selected for decreased 4-wk BW compared to their control counterparts. It appeared that selection for increased 4-wk BW did not alter the rate of growth in the males compared to the control males; however, in the females, selection for increased 4-wk BW resulted in an increase in growth rate. On the other hand, selection for decreased 4-wk BW resulted in a decline in growth rate in males but not in the
GROWTH PARAMETERS AND DIVERGENT SELECTION
MATERIALS AND METHODS Growth data were collected from Japanese quail lines selected for increased (high) or decreased (low) 4-wk BW for 30 generations (Darden and Marks, 1988) and their randombred control. Each line was propagated by a single hatch of approximately 200 quail using 30 paired matings. Quail from the control line were intermingled with the divergently selected lines during the rearing period. Quail chicks were hatched, wing-banded, and placed in quail battery brooders. All quail had ad libitum access to a 28% CP and 2,947 kcal ME/kg of diet and water. Hatch weights were collected on all birds, as well as BW each week until 8 wk of age. The birds were sexed at 6 wk of age according to their plumage color pattern.
Growth Model To estimate the expected BW at a specific age, a 4parameter growth curve modified from Richards (1959) growth function (Sugden et al., 1981) was fitted to the BW data collected. The growth curve was of the following form Wt = WA[1 − (1 − m)exp[−K (t − ti)/mm/(1−m)]]1/(1−m) where Wt is the weight of bird at time t, WA is the asymptotic (mature) BW, K is the maximum relative growth (per week), ti is the age at maximum rate of growth (week), and m is a shape parameter, with the property that m1/(1−m) is relative weight at ti. Individual growth curve parameters for the Richards curve were estimated for each bird using PROC NLIN (Marquart algorithm) (SAS Institute Inc., 1996). Differences among lines and sex within lines for the Richards model parameters were tested using PROC GLM (SAS Institute Inc., 1996).
RESULTS AND DISCUSSION Means and standard deviation for BW at various ages of both sexes for each line are presented in Table 1. As expected BW of the high and low lines were significantly different from that of the control line. Body weights of both sexes were similar from hatch until 4 wk of age after which the females become heavier than the males. The Richards growth model fitted to data on sex within lines are shown in Figure 1. The coefficients for the parameters of the fitted curve are presented in Table 2. The growth patterns of both sexes within lines were not identical; therefore, data were analyzed separately for each sex. Studies have shown that Japanese quail exhibit sexual dimorphism (Aggrey and Cheng, 1994; Du Preez and Sales, 1997; Hyankova et al., 2001). In chickens, Carte and Siegel (1970) eliminated sexual dimorphism in response to selection for growth by adjusting for scaling effects. Fitting the growth model to log-transformed data only reduced the residual variance but did not eliminate sexual dimorphism. Growth rate was higher in males (P < 0.05) than females in all the lines. Sex differences in growth rate has also been reported in European quail (Coturnix coturnix) (Du Preez and Sales, 1997) and Japanese quail (Hyankova et al., 2001). There was no difference between the rate of growth of the males of the high and the control lines and between females of the low and control lines. On the other hand, there was a significant increase in growth rate of the high line over the control line for females, as well as a significant decline in growth rate of males of the low line compared to their control counterpart. It appears that selection for increased 4-wk BW did not alter the rate of growth in the males compared to the control males; however, in the females, selection for increased 4-wk BW resulted in an increase in growth rate. Conversely, selection for decreased 4-wk BW resulted in a decline in growth rate in males but not in the females. The response of rate of growth to divergent selection for 4-wk BW appeared to be dependent on sex and direction of the selection pressure. The age at maximum growth occurred at approximately the same age in both the control and the high lines regardless of sex (Table 2). A similar result was reported by Anthony et al. (1986). However, this contradicts Marks (1978) and Hyankova et al. (2001) whose Japanese quail selection experiments observed a push back of the age of maximum growth in the upward selected lines as compared to the control lines. For the low lines, the age at maximum growth was about 5 d later, suggesting a prolonged growing period. In the present study, selection for decreased 4-wk BW resulted in a shift of the growth curve. Popular growth models, such as the Gompertz and the logistic, have fixed growth shapes with inflection points at 37 and 50% of the asymptote, respectively. These models are special cases of the Richards model, which has a variable point of inflection defined by the shape or growth trajectory parameter, m. When the
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015
Brisbin et al. (1987) suggested that the shape of a growth curve has a greater propensity to change in response to environmental changes than the asymptotic weight or growth rate and may be used to study the effects of environmental stress on growth. This inherently implies that growth models with fixed shapes may not contribute to the understanding of the effects of factors like dietary and environmental changes on growth. Aggrey (2002) further suggested that shape of the growth curve may reflect the architecture of body composition and could therefore be used to manipulate the desired body composition at a given age. Therefore, selection for changes of the shape of a growth curve may be a useful tool for improving the efficiency of lean meat production. The objective of the present study was to investigate the effect of long-term divergent selection for 4-wk BW on growth characteristics of Japanese quail and compare these characteristics with the nonselected randombred control base population using the generalized growth model.
539
540
AGGREY ET AL. TABLE 1. Mean and standard deviation for BW at different ages in three experimental Japanese quail lines1 BW Low Age (wk)
Male (n = 96)
Control Female (n = 105)
Male (n = 102)
High Female (n = 81)
Male (n = 84)
Female (n = 83)
g 0 1 2 3 4 5 6 7 8
4.74 11.04 20.99 35.31 48.19 60.75 67.86 72.09 73.41
± ± ± ± ± ± ± ± ±
0.46 2.27 4.59 6.75 8.19 8.35 7.51 5.57 5.15
4.74 11.64 21.67 36.34 49.82 62.96 71.94 81.91 87.19
± ± ± ± ± ± ± ± ±
0.46 2.39 5.07 7.19 8.15 8.11 8.39 9.62 9.18
5.41 18.92 39.91 64.07 84.87 96.13 100.39 101.98 104.06
± ± ± ± ± ± ± ± ±
0.51 2.78 4.57 5.83 6.44 7.10 6.16 6.10 6.05
5.43 19.06 40.23 64.66 87.14 101.94 116.59 127.82 130.80
± ± ± ± ± ± ± ± ±
0.47 3.28 5.79 7.07 7.54 7.94 10.69 10.78 9.58
7.4 33.91 79.58 132.71 171.37 195.72 205.07 208.39 210.85
± ± ± ± ± ± ± ± ±
0.74 4.14 7.16 10.15 11.08 13.21 12.15 13.71 14.57
7.38 33.63 81.56 138.29 179.18 205.56 219.13 233.89 240.94
± ± ± ± ± ± ± ± ±
0.67 3.44 8.37 11.09 10.99 12.21 12.36 15.08 14.55
1 Control = unselected randombred; Low = selected for decreased 4-wk BW; High = selected for increased 4wk BW.
250
Body weight (g)
200 150 100 50 0 0
2
4
6
8
Age (wk) Control: Male High: Female
Control: Female Low: Male
High: Male Low: Female
FIGURE 1. Growth curve of Japanese quail lines divergently selected for 4-wk body weight and their controls.
selection criterion only resulted in altering the trajectory of growth in the males. Alteration in the shape of the curve should be due to the degree of genetic flexibility of the shape parameter and the pleiotropic relationship between the shape parameter and the selection criterion. As postulated by Brisbin et al. (1986), growth shape is likely to be related to the relative timing and functioning of intrinsic biological processes that relate to the assimilation of food and its synthesis into new body tissues. It would therefore be possible to genetically alter the shape of the curve, which would consequently alter body composition. Aggrey (2002) reported a correlation between the shape parameter and abdominal fat in chickens. Brisbin et al. (1986) further suggested that the shape parameter has a greater propensity to change in response to environmental changes than the asymptotic BW or growth rate and can be used to assess effect of nongenetic changes as well. It should therefore not be assumed that the growth of a given species may be characterized by fixed shape curve under all genetic and environmental conditions and that even under uniform conditions, a single curve shape may not characterize both sexes. Data from Table 1 and Figure 1 show that the response to divergent selection for 4-wk BW was asymmetrical. The asymmetrical response to selection was probably due to the duration of the selection experiment, because the same experimental lines exhibited symmetrical response through 11 generations of selection (Darden and Marks, 1988). A 12-generation divergent selection experiment reported by Anthony et al. (1986) also showed symmetry in response to 4-wk BW. After 30 generations of selection, the 4-wk BW means of the lines averaged over sexes were 49 g (low), 86 g (control), and 175 g (high). The absolute difference between the high and the control line was more than twice of that between the low and the control. The absolute difference of the divergent lines from the control is even more magnified in the asymptotic BW. Asymptotic BW increased by about 100 and 70%, respectively, in the male and female
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015
growth curve shape value is 2.0 or 1.0, the Richards model is equivalent to the logistic or Gompertz models, respectively. Females had higher (P < 0.05) m values than their male counterparts with lines. The shape value for the females for all the lines were close to one. A shape value of one from the Richards model indicate that the females followed the Gompertz model, and selection for either increased or decreased 4-wk BW did not change the shape of the growth curve in females. However, for the males, the shape value increased in the low line and decreased in the high line when compared to the control males. The control males had a growth curve shape between the logistic and the Gompertz models. The selection for decreased 4-wk BW altered the shape of the curve to resemble more of the Gompertz curve, whereas selection for increased 4-wk BW modified the shape towards the logistic curve. The dynamics of the growth curve parameters indicate that selection for decreased 4-wk BW shifted the growth curve in the females as well as altering the trajectories of growth in both sexes. However, in the high line, the
541
GROWTH PARAMETERS AND DIVERGENT SELECTION TABLE 2. Estimates of Richards growth curve parameters in three experimental Japanese quail lines Low Growth parameters
Male
Asymptotic weight, A (g) Shape parameter, m Rate of growth, k (g/wk) Age at maximum growth, t (wk)
75.92 1.81 0.73 3.11
± ± ± ±
0.89a 0.17a 0.06a 0.09a
Control Female
101.41 1.09 0.39 3.22
± ± ± ±
3.28bc 0.14b 0.04b 0.12a
Male 104.42 1.67 0.89 2.34
± ± ± ±
0.45bc 0.09c 0.04c 0.05b
1
High Female
144.01 0.93 0.43 2.44
± ± ± ±
2.31d 0.09b 0.03b 0.09b
Male 212.82 1.42 0.85 2.23
± ± ± ±
0.10e 0.08d 0.03c 0.05b
Female 246.21 0.98 0.60 2.11
± ± ± ±
1.61f 0.06b 0.02d 0.06b
Growth parameters with no common superscript are significantly different (P ≤ 0.05). Control = unselected randombred; Low = selected for decreased 4-wk BW; High = selected for increased 4-wk BW.
a–f 1
ACKNOWLEDGMENT The authors thank Cheryl Pearson Gresham, Department of Poultry Science, University of Georgia, Athens, GA, for her technical assistance and collection of data.
REFERENCES Aggrey, S. E. 2002. Comparison of three non-linear and spline regression models for describing chicken growth curves. Poult. Sci. 81:1782–1788.
Aggrey, S. E., and K. M. Cheng. 1994. Animal model analysis of genetic (co)variance of growth traits in Japanese quail. Poult. Sci. 73:1822–1828. Akbas, Y., and I. Oguz. 1998. Growth curve parameters of line of Japanese quail (Coturnix coturnix japonica), unselected and selected for four-week body weight. Arch. Geflugelkd. 62:104–109. Anthony, N. B., K. E. Nestor, and H. L. Marks. 1996. Shortterm selection for four-week body weight in Japanese quail. Poult. Sci. 71:1985–1993. Anthony, N. B., D. A. Emmerson, K. E. Nestor, W. L. Bacon, P. B. Siegel, and E. A. Dunnington. 1991. Comparison of growth curves of weight selected populations of turkeys, quail and chickens. Poult. Sci. 70:13–19. Anthony, N. B., K. E. Nestor, and W. L. Bacon. 1986. Growth curves of Japanese quail as modified by divergent selection for 4-week body weight. Poult. Sci. 65:1825–1833. Barbato, G. F. 1992. Divergent selection for exponential growth rate at fourteen or forty-two days of age. 1. Early responses. Poult. Sci. 71:1985–1993. Brisbin, I. L., Jr., C. T. Collins, G. C. White, and D. A. McCallum. 1987. A new paradigm for the analysis and interpretation of growth data: The shape of things to come. Auk 104:552–554. Brisbin, I. L., Jr., G. C. White, P. B. Bush, and L. A. Mayack. 1986. Sigmoid growth analysis of wood ducks: The effect of sex, dietary protein and cadmium on parameters of the Richards model. Growth 50:41–50. Carte, I. F., and P. B. Siegel. 1970. Scaling effects and the inheritance of juvenile body weight in chickens. Can. J. Genet. Cytol. 12:724–727. Darden J. R., and H. L. Marks. 1988. Divergent selection for growth in Japanese quail under split and complete nutritional environments. 1. Genetic and correlated responses to selection. Poult. Sci. 67:519–529. Du Preez, J. J., and J. Sales. 1997. Growth rate of different sexes in European quail (Coturnix coturnix). Br. Poult. Sci. 38:314–315. Eisen, E. J. 1976. Results of growth curve analyses in mice and rats. J. Anim. Sci. 42:1108–1023. Hyankova, L., H. Knizetova, L. Dedkova, and J. Hort, 2001. Divergent selection for shape of growth curve in Japanese quail. 1. Responses in growth parameters and food conversion. Br. Poult. Sci. 42:583–589. Knizetova, H., J. Hyanek, B. Knize, and J. Roubicek. 1991a. Analysis of growth curves in fowl. I. Chickens. Br. Poult. Sci. 32:1027–1038. Knizetova, H., J. Hyanek, B. Knize, and H. Prochazkova. 1991b. Analysis of growth curves in fowl. II. Ducks. Br. Poult. Sci. 32:1039–1053. Leclercq, B., G. Guy, and F. Rudeaux. 1989. Growth characteristics and lipid distribution in two lines of chickens selected for low or high abdominal fat. Genet. Sel. Evol. 21:69–80. Marks, H. L. 1971. Selection for four-week body weight in Japanese quail under two nutritional environments. Poult. Sci. 50:931–937.
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015
high lines but declined by only about 25% in the low lines. A similar observation was also made by Barbato (1992). Body weight at the same age in long-term divergently selected lines may be considered as different traits. Genes may respond differently to the same selection pressure under different internal physiological environment, and the forces of natural selection may act differently on the divergent lines as well. It appeared that changes in the asymptotic weights were a result of changes growth rate, shape parameter, and hatch weight during the divergent selection for 4wk BW. Selection for increased 4-wk BW has resulted in a correlated increase in hatch weight for both sexes. For the high lines, hatch weight coupled with increased rate of growth contributed to affect the asymptotic BW in females, whereas in the males, it was increased hatch weight and altered growth trajectory, which compounded to influence asymptotic BW. Selection for decreased 4-wk BW also resulted in a correlated reduction in hatch weight; however, in the low lines, the change in the female shape parameter was marginal, and the reduction in the growth rate was not significant to cause a substantial deviation from the control. Reduced hatch weight and concurrent decline in growth rate and alteration of growth trajectory played roles in determining the asymptotic BW in the male low line. The use of a general growth model (Richards) to analyze growth data from such an experiment provided a better understanding of how selection altered the rate and trajectory of growth to affect the genetically determined growth potential of quail. Consequently, selection for increased or decreased 4-wk BW affected the sexes differently. Long-term divergent selection may cause asymmetry in response as a result of possibly different genes responding differently to the selection pressure.
542
AGGREY ET AL.
Marks, H. L. 1978. Growth curve changes associated with longterm selection for body weight in Japanese quail. Growth 42:129–140. Mignon-Grasteau, S., M. Piles, L. Varona, H. de Rochambeau, J. P. Poivey, A. Blasco, and C. Beaumont. 2000. Genetic analysis of growth curve parameters for male and female chickens resulting from selection on shape of curve. J. Anim. Sci. 78:2515–2524. Richards, F. J. 1959. A flexible growth function for empirical use. J. Exp. Botany 10:290–300. Ricklefs, R. E. 1968. Patterns of growth in birds. Ibis 110:419–451.
SAS Institute. 1996. SAS User’s Guide. Version 6.12. SAS Institute Inc., Cary, NC. Siegel, P. B. 1962. Selection for body weight at eight weeks of age 1. Short term responses and heritabilities. Poult. Sci. 41:954–962. Sugden, L. G., E. A. Driver, and M. C. S. Kingsley. 1981. Growth and energy consumption by captive mallards. Can. J. Zool. 59:1567–1570. Timon, V. M., and E. J. Eisen. 1969. Comparison of growth curves of mice selected and unselected for postweaing gain. Theor. Appl. Genet. 39:345–351.
Downloaded from http://ps.oxfordjournals.org/ at New York University on May 22, 2015