47
Growth. Development & Aging, 2003, 67, 47-54
Dynamics of Relative Growth Rate in Japanese Quail Lines Divergently Selected for Growth and Their Control I
Samuel
E.
Aggrey2
Poultry Genetics and Bioteclmology Laboratory Department of Poultry Science The University of Georgia Athens, GA 30602-2772, USA
2
Corresponding Author Dr. Sammy Aggrey Department of Poultry Science The University of Georgia Athens, GA 30602, USA Phone: 706-542-1354 Fax: 706-542-1827 E-mail:
[email protected]
ABSTRACT: This study was undertaken to test the hypothesis that relative growth is the same in Japanese quail divergently selected for 4-week body weight for 30 generations, and their control. Relative growth rate is the increase in body weight per unit of body weight per unit of time, and it represents the efficiency of an animal as a producer of new body mass material. The maximum relative growth rate ofthe divergent and control lines were observed dur ing the first week. Selection for high 4-week body weight has resulted in higher relative growth rate for the first two weeks compared to quail from the low line, but by week 3, quail from the low line had higher relative growth rate compared to quail from the high line. Asymptotic body weight increased by about 100 and 70%, respectively in the male and fe male high lines and declined by about 25% in the ISupported by State and Hatch funds allocated to the Georgia Agricultural Experimental Stations of the Uni versity of Georgia
low lines. Absolute growth rate is thought to be re lated to the cumulative growth already achieved whereas relative growth rate is related to the amount of growth remaining. However, selection on absolute weight basis can elicit a response in rela tive growth rate in the early part ofthe developmen tal period which possibly pre-determines the asymp totic body weight. Since relative growth rate depends on the amount of growth remaining, the rate of decline in relative growth is then set accord ingly. Furthermore, it is thought that different sets of genes may operate between the developmental pe riod and maturation period. KEY WORDS: Relative growth rate, divergent se lection, Japanese quail, growth models INTRODUCTION Growth in biological terms is related to changes in size and shape. The bases for growth are the pro
48
cesses of hyperplasia, hypertrophy and cell differen tiation. However, these processes can be affected by the environment, including nutrition and random events, causing growth to fluctuate and thus mak ing the study of growth at a single age unattractive. Mathematical models have been used extensively to fit growth curves. Growth patterns generated by such mathematical models smooth these deviations using previous and future weights to predict a spe cific age-weight point (DeNise and Brinks, 1985). Growth functions also allow for the study of growth rate and the estimation of changes in the shape of the curve during selection (Fitzhugh, 1976; Knizeto va et al., 1991; Maruyama et al., 1998; Hyankova et al., 2001). Non-linear models have been used to de scribe growth curves in chickens. The logistic growth function has been applied (Sang,1962; Grossman and Bohren, 1982; Grossman et al., 1985) to chicken data although its symmetrical form does not correspond to the growth pattern of chickens (Knizetova et al., 1991). The Laird form (Laird et al., 1965) of the Gompertz model (Gompertz, 1925) has been the model of choice for poultry data (Tzeng and Becker, 1981; Anthony et at., 1991; Barbato, 1991 and Mignon-Grasteau, et al., 1999) because of its overall fit and the biological meaning ascribed to the model parameters (Ricklefs, 1985). Absolute growth rate measures cumulative growth already achieved per unit of time, and rela tive growth rate measures remaining growth per unit of time. Relative growth rate (RGR) is defined as the ratio of the rate of change in body weight to the achieved growth at a given age (Laird, 1965). It represents the efficiency of an animal as a producer of new body mass material. Body weight measured at each age is the accumulation of weight over all previous ages, and may not readily reveal age-spe cific variation in growth rate, and Laird (1965) has suggested that age-specific variation should be as sessed with relative growth rate rather than abso lute growth rate. Divergent selection for growth has been studied using absolute cumulative growth data (Darden and Marks, 1988; Anthony et at., 1991; Hyankova et al., 2001). Selection for high 8-week body weight in chickens (Marks, 1979, 1980) and high 4-week body
AGGREY
weight in Japanese quail (Marks, 1978) resulted in marked increases in relative growth only during the first 2 weeks, and not the later part of the growth cycle. Growth studies using cumulative growth data do not assess the efficiency of synthesizing new body material. Therefore dynamics of growth as influ enced by genetic and environmental changes may be better studied with RGR than with absolute growth rate. Relative growth rate assesses both ab solute growth and the relative efficiency to synthe size new body mass material. The current study ex amines the degree to which divergent selection affects the efficiency of growth and test the hypothe sis that relative growth is the same in Japanese quail divergently selected for 4-week body weight for 30 generations, and their control. MATERIALS AND METHODS Growth data were collected from Japanese quail lines divergently selected for 4-week body weight for 30 generations (Darden and Marks, 1988) and their control. The lines hereafter are referred to as high, low and control. Quail chicks were hatched in a single hatch, wing-banded, and placed in quail battery brooders. The quail lines were raised sepa rately. All quail had ad libitum access to a 28% crude protein and 2, 947 kcal ME/kg diet and water. Hatch weights were collected on all birds, as well as weekly body weight until 8 weeks of age. Records from quail that died before 8 weeks of age were dis carded. One hundred and seven-two, 132 and 126 growth records from low, high and control lines, re spectively were used in the analysis. The number of individuals of each sex within each line is given in Table 1. Gompertz-Laird model: The Laird form of the Gompertz equation (Laird et at., 1965) was used to estimate relative growth. The following equation describes the Gompertz-Laird growth model for fit ting absolute weights,
Wt = Wo exp[(UK)(1 - exp-Kt)]
(1)
where Wt is the weight ofthe bird at time t, Wo is the initial body (hatch) weight, L is the instanta neous growth rate (per week), K is the rate of expo
RELATIVE GROWTH IN JAPANESE QUAIL
49
nential decay of the initial specific growth rate of L (per week). The asymptotic body weight, W A is de rived as:
(2)
WA= Wo exp(LlK)
From equation (1) the instantaneous growth rate is given as:
(3)
dWt = L exp(-Kt) x Wt dt
and the relative growth rate (RGR) as defined by Laird (1965) is given as: RGR
= dW t x ~
stitute Inc., 1996). Japanese quail exhibit sexual di morphism in growth (Aggrey and Cheng, 1994) therefore the results were grouped by sex within line. Pearson correlations among the estimated pa rameters were also determined using PROC CORR (SAS Institute Inc., 1996). The differences in RGR between lines within sex from at each week from 1 to 8 weeks of age were tested using PROC GLM (SAS Institute Inc., 1996). Differences between means at each age among the high, control and low quail lines for each sex were tested using the TUKEY option of PROC GLM (SAS Institute Inc., 1996).
(4)
== L exp(-Kt)
Wt
RESULTS AND DISCUSSION
The RGR of Equation (1) can be re-written as:
Figure 1 illustrates the relative growth curve for males (a) and females (b) for the quail lines diver gently selected for 4-week body weight and their control. The relative mean growth rates for the three quail lines are presented in Table 1. Maxi mum relative growth for all the lines were observed in the first week. For both males and females, the relative growth rates of high and low line quail de-
dt
(5) Relative growth rate parameters (L and K) from the Gompertz-Laird function in Equation (4), and WA derived with Equation (1) for the three Japanese quail lines, were estimated for each bird using PROC NLIN (Marquart algorithm) (SAS In
TABLE 1
Relative growth rate (g g-l week-I) of Japanese quail lines divergently selected
for 4-week body weight and their control.
Male
Female
Age (wks)
Low N=72
Control N=48
High N=54
Low N=100
1
1.21±O.45 a
2.41±O.43 b
3.54±O.62c
1.46±O.45 3
2.51±O.55 b
3.59±O.48c
2
O.90±O.23 a
1.16±O.17b
1.42±O.35c
O.86±O.23 a
1.13±O.18b
1.44±O.46c
3
O.71±O.12a
O.61±O.09 b
O.67±O.06c
O.70±O.15 a
O.62±O.10b
O.70±O.07 a
4
O.37±O.07 a
O.34±O.06 b
O.29±O.05c
O.38±O.O8 3
O.35±O.07 b
O.30±O.06 c
5
O.28±O.06 a
O.14±O.06 b
O.15±O.04b
O.27±O.O73
O.17±O.05 b
O.15±O.03 c
6
O.14±O.06 a
O.08±O.05 b
O.05±O.02c
O.15±O.05 a
O.14±O.05 a
O.07±O.03 b
7
O.O8±O.O5 3
O.03±O.02 b
O.03±O.02 b
O.14±O.O5 3
O.10±O.05 b
O.07±O.03c
8
O.03±O.02 a
O.OhO.01 b
O.Ol±O.01 b
O.07±O.05 a
O.03±O.02 b
O.03±O.02 b
Control N=78
High N=78
Means within rows for males and females with different superscripts are significantly different (P < 0.05).
50
AGGREY
4
C1) .....m
a::: ..... ~~
-
..s::::
o~ (!)~ C1)~
3
L..
2
>
1
+=i
(a)
m C1)
a:::
0 0
2
1
3
5
4
7
6
8
Age (weeks) • Control
-~
..,.. Low
High
(b)
\ \
1
3
2
5
4
6
7
8
Age (weeks) • Control
-~
High
..,.. Low
Figure 1. Relative growth rates of divergently selected Japanese quail Jines and their controll. (a) Males (b) Females lControf=Unselected; Low=Selected for low 4 week body weight; High=Selected for high 4 week body weight
RELATIVE GROWTH IN JAPANESE QUAIL
viate about equally from the controls at week 1. There are significant line differences (P
