- Email: [email protected]
Survival Under Heat Stress of Lines Selected for Fast and Slow Growth at Two Temperatures1
Survival Under Heat Stress of Lines Selected for Fast and Slow Growth at Two Temperatures1 B. B. BOHREN, J. C. ROGLER, and J. R. CARSON Department of Animal Sciences, Purdue University, West Lafayette, Indiana 47907 (Received for publication April 22, 1981) ABSTRACT Two replicates of four lines of White Leghorn chickens previously selected for fast and slow gain from 5 to 9 weeks of age in a hot (32.2 C) and cold (21.1 C) selection environment were grown from 5 to 9 weeks in the hot and cold temperature environments and then subjected to high temperature stress at 40.6 C. Four different experiments were run. The percent mortality was analyzed when approximately 50% of the birds had died. Birds reared in the cold environment were heavier and consistently had higher mortality than those reared in the hot environment when both were subjected to acute heat stress (40.6 C). The importance of acclimation to mild heat stress on response to acute heat stress was demonstrated by the mortality difference. No significant differences in survival under heat stress were found between lines selected in the hot and the cold selection environments. Lines selected for fast growth had significantly higher mortality rates under high temperature stress than lines selected for slow growth in replicate 1 but not in replicate 2. Estimates of the genetic correlation between weight gain from 5 to 9 weeks of age and percent mortality under heat stress after two generations of selection in replicate 2 was —.004, and after three generations of selection in replicate 1 was +.076. (Key words: heat stress, selection, growth rate, poultry) 1982 Poultry Science 61:1804-1808 INTRODUCTION The ability of an individual to withstand stress situations has been attributed to a number of physiological and physical factors. Reviews of studies on response to stress, and particularly stress due to high ambient temperatures, have been given by Siegel (1969), Freeman (1971), Edens and Siegel (1976), and Fox (1980). Body weight might be assumed to be associated with response to high temperature stress because of a smaller surface area relative to metabolic mass. Several workers have found breed differences in adult birds in resistance to various types of high temperature stress, with the smaller breeds such as Leghorns and Jungle fowl being more resistant to stress (see Hillerman and Wilson, 1955). But Fox (1951, 1980) has shown that heavier breeds such as White Plymouth Rocks and Rhode Island Reds were more resistant to high temperatures than White Leghorns if no water was available during the stress treatment. In growing birds, strains with heavier body weights generally show a higher mortality under
1 Journal Paper No. 8509, Purdue University Agricultural Experiment Station. This investigation was supported by NIH grant ES00310.
heat stress. Marks and Huston (1973) made this observation in quail selected for body weight at 4 weeks of age. Washburn et al. (1980) found selected broiler strains to be more affected by heat stress than an unselected and smaller control line. Wilson et al. (1975a) observed that a strain selected for a short survival time at high temperature by Wilson et al. (1966, 1975b) was larger at 6 weeks of age than a strain selected for a long survival time, but the difference in weight was not statistically significant. Bohren et al. (1981) selected two replicates of White Leghorns for four generations for fast and slow gain in body weight from 5 to 9 weeks of age in a hot (32.2 C) and a cold (21.1 C) environment and tested the four selected lines from each replicate in both environments each generation. The weight gains of the lines selected for fast and slow gains averaged over sex, replicates, and the environments of testing and selection differed by 64, 117, 121, and 159 g after the four generations of selection, respectively. The 9-week body weight differed between the fast and slow selected lines by 123, 200, 244, and 273 g after the four generations of selection, respectively. The hot growing environment from 5 to 9 weeks of age reduced the gain 104 g, about 20% of the gain observed in the cold rearing environment. The purpose of this experiment was to
1804
HEAT STRESS ON LINES SELECTED FOR GROWTH
determine if 1) selection for fast or slow growth, 2) selection in the hot or the cold environment, or 3) rearing from 5 to 9 weeks in the two environments would have any effect on the ability of the birds to survive under high temperature stress at 40.6 C. MATERIALS AND METHODS
Details of the management and experimental procedure in developing the two replicates of each line were presented by Bohren et al. (1981). The four lines used in this experiment were selected 1) for fast 5 to 9-week gain in the hot (32.2 C) environment, 2) for slow 5 to 9-week gain in the hot environment, 3) for fast 5 to 9-week gain in the normal (21.1 C) environment (hereafter referred to as the cold environment), and 4) for slow 5 to 9-week gain in the cold environment. After the gain in each line was measured in the hot and cold environments from 5 to 9 weeks of age, in some replicates and generations, samples of each line grown in each environment were subjected to high temperature stress at 40.6 C with a relative humidity of 50% and with both feed and water available. The effect of temperature stress on the birds from the selected lines and growing environments was measured as the percent mortality at variable periods after the initiation of stress. Except in Experiment 1, the comparisons were made when approximately 50% of all birds had died. Experiment 1. Males from replicate 2 after one generation of selection were used. When the birds were 9 weeks of age, the temperature in both the hot and cold chambers was raised to 40.6 C. Deaths occurred so rapidly that the temperature was returned to the original temperature after 6 hr. Mortality continued for 2 hr more, so the percent mortality was determined 8 hr after the initiation of the stress when 24.3% of the birds had died. There were 21 males observed for each selected line and growing environment. Experiment 2. Females of generation 2, replicate 1 were used for Experiment 2. When the birds were 9 weeks of age, the temperature in both the hot and cold chambers was raised to 37.8 C for 24 hr to provide some additional temperature acclimation to the birds reared in both environmental chambers. After 24 hr at 37.8 C, one pullet was chosen at random from each of 20 full-sib families in each line and growing environment. Five of the chosen
1805
individuals from each line and growing environment were randomly distributed into each of the four pens in the vertical column of a growing battery in the chamber previously assigned to the hot (32.2 C) environment. The temperature was then elevated to 40.6 C for the duration of the stress test. The percent mortality was analyzed after 17 hr of stress when 50.6% of the birds had died. Experiment 3. Females from generation 2, replicate 2, were used in Experiment 3. To increase the difference in possible temperature acclimation between the two rearing environments, at 9 weeks of age the temperature in the hot chamber was elevated to 37.8 C for 48 hr whereas the temperature in the cold chamber remained at 21.1 C. At the end of 48 hr, one progeny of each of 32 full-sib families in each line in the cold room was chosen at random and two of the 32 progeny of each line were randomly placed in each of 16 pens in the chamber previously assigned to the cold (21.1 C) environment. The same procedure for choosing progeny was followed in the hot room, and the resulting 32 birds were randomly distributed to 16 other pens in the cold chamber. The temperature was then elevated to 40.6 C for the duration of the stress test. The percent mortality was determined after 9.5 hr when 55.1% of the birds had died. Experiment 4. Females of generation 3, replicate 1 were used. At 9 weeks of age, the temperature in the hot room was increased to 37.8 C for 48 hr. After 48 hr, one pullet from each line in each growing environment was placed in each of 32 pens in the chamber previously assigned to the hot growing environment. Thus, each pen contained 8 pullets, one from each line grown in each environment. The percent mortality was calculated after 4 hr of exposure to the high temperature stress when 49.1% of the birds had died. The percentages were transformed to angles corresponding to the arcsin -J proportion after adjustment of all zero values to l/4n and all 100% values to 1 — l/4n as shown by Steel and Torrie (1980, page 236). Analysis of variance was applied to the transformed values based on the fixed model Y... = a + L. + E. + EL.. + e..... uk
^
i
J
U
(u)k
where fi represents the population mean, Lj is the effect of the i t h line (i = 1..4), Ej is the effect of the j growing environment (j =
1806
BOHREN ET AL.
1,2), and e,.., k is the random variation among individuals in the same line and reared from 5 to 9 weeks of age in the same environment. The random error was estimated for the transformed binomial data as the theoretical value 821/n, where n is the number of observations in each group (Steel and Torrie, 1980). RESULTS The mortality percentages observed for each line and rearing environment for the four experiments are shown in Table 1. The analyses of variance for the four experiments are shown in Table 2. In all experiments, the birds reared in the cold (21.1 C) environment had a significantly higher percent mortality than those reared in the hot (32.2 C) environment. Since the body weights and gains for the cold rearing environment were always greater, and the mortality percentages were higher than those for the hot rearing environment, the effects of body weight and rearing environment on percent mortality are completely confounded. An effort was made to differentiate between the effect of body weight and temperature preconditioning on the percent mortality. Short periods of higher temperatures were applied after the birds were 9 weeks of age. This short
interval treatment had little effect on body weight but would show the effect of additional temperature acclimation. For example, in Experiment 2, both chambers were increased to 37.8 C for 24 hr before applying the extreme stress at 40.6 C. In this case, 17 hr at 40.6 C were required to provide approximately 50% mortality. In Experiments 3 and 4, where only the hot chamber was increased to 37.8 C for 48 hr, the difference in mortality under extreme stress between birds grown in the hot and cold temperatures was increased. In addition, only 9.5 and 4 hr of extreme stress, respectively, were required to produce approximately 50% total mortality in these two experiments, with high frequencies of death among those grown in the cold temperature occurring in the first hour or two of stress. These experiments were also conducted to determine if differences in survival under heat stress existed among the four lines selected for fast and slow growth in the hot and cold environmental chambers. In Experiments 1 and 3 conducted on generations 1 and 3 of replicate 2, no significant differences among lines were observed. However, in Experiments 2 and 4 conducted on generations 2 and 3 of replicate 1 significant line effects were found. In these two experiments, the line effect mean squares were partitioned into orthogonal contrasts with
TABLE 1. Percent mortality under temperature stress at 40.6 C for four experiments.1 Lines Environment of selei:tion
Direction of selection
Growing environment Hot
Cold
X
Hot
Growing environmient Cold X
- Experiment 1 • 1 2 3 4
Hot Hot Cold Cold
Fast Slow Fast Slow
Mean
- Experiment 2 •
4.0 5.9 14.3 5.6
32.0 58.8 23.8 50.0
18.0 32.4 19.0 27.8
45.0 30.0 70.0 10.0
60.0 65.0 85.0 40.0
52.5 47.5 77.5 25.0
7.4
41.2
24.3
38.8
62.5
50.6
- Experiment 3 • 1 2 3 4 Mean
Hot Hot Cold Cold
Fast Slow Fast Slow
- Experiment 4 •
25.0 21.9 21.9 34.4
87.5 90.6 84.4 75.0
56.2 56.2 53.2 54.7
28.1 3.1 9.4 .8
98.2 75.0 96.9 81.2
63.2 39.0 53.2 41.0
25.8
84.4
55.1
10.4
87.8
49.1
1 The number of generations of selection for growth, replicate, number of observations in each cell, and the hours after initiation of stress are: Experiment 1, generation 1, replicate 2, 21 observations, 8 hr; Experiment 2, generation 2, replicate 1, 20 observations, 17 hr; Experiment 3, generation 2, replicate 2, 32 observations, 9.5 hr; Experiment 4, generation 3, replicate 1, 32 observations, 4 hr.
HEAT STRESS ON LINES SELECTED FOR GROWTH
1807
TABLE 2. Analyses of variance of percent mortality (transformed to angles in degrees) from temperature stress at 40.6 C
Source
df
Mean squares Experiment 1, Experiment 2, Generation 1, Replicate 2 Generation 2, Replicate 1 Mean squares
Between lines (L) Directions of selection (D) Environment of selection (ES) DX ES Between test environments (E) EX L Error (821/n)
Between lines (L) Directions of selection (D) Environment of selection (ES) DX ES Between test environments (E) EX L Error (821/n)
31.92
3 1 1 1 1 3
-
1183.17** 80.90 39.10
3
Experiment 3. Generation 2, Replicate 2 3.52 1 1 1
1 3
2687.55** 37.98 25.66
single_ degrees of freedom (Table 2). No significant difference in mortality was observed between the two environments in which selection occurred, but the lines selected for fast growth had significantly higher mortality in both experiments than the lines selected for slow growth. Therefore, a significant correlated response in percent mortality under heat stress from selection for 5 to 9-week gain occurred in replicate 1 but not in replicate 2. Genetic correlations between the weight gains from 5 to 9 weeks of age and the percent mortality under temperature stress were estimated after two generations of selection in replicate 2 and after three generations of selection in replicate 1 by use of equation 19.5 in Falconer (1960). These genetic correlation estimates were —.004 for replicate 2 and +.076 for replicate 1. The average estimate over both replicates was +.036. The interaction between the directions of selection and the environment in which selection occurred was significant in both Experiments 2 and 4. However, the interactions in the two experiments were quite different. In Experiment 2, the difference in mortality between the lines selected for fast and slow growth was greater in the cold selection environment, whereas in Experiment 4 the difference between the directions of selection was
369.53** 326.61** .21 781.21** 456.46** 20.80 41.05 Experiment 4, Generation 3, Replicate 1 240.57** 327.52** 17.91 376.29** 6126.70** 19.18 25.66
greatest in the hot selection environment. No reason was evident why this interaction should be different in the two experiments. DISCUSSION Interest centers on whether differences in body weight gains, either environmentally or genetically induced, are associated with differences in ability to survive under high temperature stress. Significantly larger weight gains and significantly lower survival were found for birds reared in the cold environment in all experiments. Because the birds compared in the two environments were genetically similar, one could conclude that a large negative environmental correlation exists between body weight and survival under heat stress. However, preconditioning at the two temperatures undoubtedly resulted in many physiological differences between the groups in addition to the weight difference. This is indicated by the effect of the 24 and 48-hr conditioning at 37.8 C, which would have little effect on body weight, but which had an effect on survival time under the extreme stress. Genetically determined weight differences due to selection for fast and slow gain from 5
1808
BOHREN ET AL.
t o 9 weeks gave different results for t h e t w o replicates. Under t e m p e r a t u r e stress, replicate 1 showed significant differences in m o r t a l i t y percentages (23.5%) b e t w e e n t h e lines selected in different directions, whereas replicate 2 showed n o significant differences in m o r tality percentages b e t w e e n t h e fast and slow lines. This result occurred in spite of t h e fact t h a t t h e differences b e t w e e n t h e fast and slow lines in gain and 9-week b o d y weight were nearly t h e same in t h e t w o replicates. Whereas differences in b o d y weight b e t w e e n t h e fast and slow growing lines appear t o be associated w i t h survival at high t e m p e r a t u r e , t h e correlated response in p e r c e n t m o r t a l i t y w h e n averaged over t h e t w o replicates (8.6%) suggests t h a t selection for fast and slow gain is n o t very effective in differentiating t h e fast and slow growing lines for survival ability u n d e r e x t r e m e stress. T h e differences b e t w e e n t h e t w o replicates could be d u e t o genetic drift of o t h e r factors m o r e i m p o r t a n t t h a n b o d y weight in producing differences b e t w e e n lines in survival at high t e m p e r a t u r e s . Selection for g r o w t h in t h e h o t and cold environments might be expected to select for genes increasing o r decreasing survival at higher t e m p e r a t u r e s , b u t only small and nonsignificant differences in survival percentages were f o u n d b e t w e e n t h e selection environments.
REFERENCES Bohren, B. B., J. R. Carson, and J. C. Rogler, 1981. Response to selection in two temperatures for fast and slow growth to nine weeks of age. Genetics 97:443-450. Edens, F. W., and H. S. Siegel, 1976. Modification of
corticosterone and glucose reponse by sympatholytic agents in young chickens during acute heat exposure. Poultry Sci. 55:1704-1712. Falconer, D. S., 1960. Introduction to Quantitative Genetics. The Ronald Press Co., New York, NY. Fox, T. W., 1951. Studies on heat tolerance in the domestic fowl. Poultry Sci. 40:477—483. Fox, T. W., 1980. The effect of thiouracil and thyroxine on resistance to heat shock. Poultry Sci. 59:2391-2396. Freeman, B. M., 1971. Stress and the domestic fowl: A physiological appraisal. World's Poultry Sci. J. 27:263-275. Hillerman, J. P., and W. O. Wilson, 1955. Acclimatization of adult chickens to environmental temperature changes. Am. J. Physiol. 180:591 — 595. Marks, H. L., and T. M. Huston, 1973. Response of selected quail lines to heat stress. Poultry Sci. 52:1668-1670. Siegel, H. S., 1969. Environmental stress and physiological compensating mechanisms in fowl: Temperature and respiratory regulation. Poultry Sci. 4 8 : 2 2 - 3 0 . Steel, R.G.D., and J. H. Torrie, 1980. Principles and Procedures of Statistics. McGraw-Hill Book Co., Inc., New York, NY. Washburn, K. W., R. Peavey, and G. M. Renwick, 1980. Relationship of strain variation and feed restriction to variation in blood pressure and reponse to heat stress. Poultry Sci. 59:2586— 2588. Wilson, H. R., A. E. Armas, I. J. Ross, R. W. Dorminey, and C. J. Wilcox, 1966. Familial differences of Single Comb White Leghorn chickens in tolerance to high ambient temperatures. Poultry Sci. 45:784-788. Wilson, H. R., D. W. MacLaury, T. H. Johnson, and C. D. Baird, 1975a. Relationship of heat tolerance and oxygen consumption in chickens. Poultry Sci. 54:299-300. Wilson, H. R., C. J. Wilcox, R. A. Voitle, C. D. Baird, and R. W. Dorminey, 1975b. Characteristics of White Leghorn chickens selected for heat tolerance. Poultry Sci. 54:126-130.
1 Journal Paper No. 8509, Purdue University Agricultural Experiment Station. This investigation was supported by NIH grant ES00310.
heat stress. Marks and Huston (1973) made this observation in quail selected for body weight at 4 weeks of age. Washburn et al. (1980) found selected broiler strains to be more affected by heat stress than an unselected and smaller control line. Wilson et al. (1975a) observed that a strain selected for a short survival time at high temperature by Wilson et al. (1966, 1975b) was larger at 6 weeks of age than a strain selected for a long survival time, but the difference in weight was not statistically significant. Bohren et al. (1981) selected two replicates of White Leghorns for four generations for fast and slow gain in body weight from 5 to 9 weeks of age in a hot (32.2 C) and a cold (21.1 C) environment and tested the four selected lines from each replicate in both environments each generation. The weight gains of the lines selected for fast and slow gains averaged over sex, replicates, and the environments of testing and selection differed by 64, 117, 121, and 159 g after the four generations of selection, respectively. The 9-week body weight differed between the fast and slow selected lines by 123, 200, 244, and 273 g after the four generations of selection, respectively. The hot growing environment from 5 to 9 weeks of age reduced the gain 104 g, about 20% of the gain observed in the cold rearing environment. The purpose of this experiment was to
1804
HEAT STRESS ON LINES SELECTED FOR GROWTH
determine if 1) selection for fast or slow growth, 2) selection in the hot or the cold environment, or 3) rearing from 5 to 9 weeks in the two environments would have any effect on the ability of the birds to survive under high temperature stress at 40.6 C. MATERIALS AND METHODS
Details of the management and experimental procedure in developing the two replicates of each line were presented by Bohren et al. (1981). The four lines used in this experiment were selected 1) for fast 5 to 9-week gain in the hot (32.2 C) environment, 2) for slow 5 to 9-week gain in the hot environment, 3) for fast 5 to 9-week gain in the normal (21.1 C) environment (hereafter referred to as the cold environment), and 4) for slow 5 to 9-week gain in the cold environment. After the gain in each line was measured in the hot and cold environments from 5 to 9 weeks of age, in some replicates and generations, samples of each line grown in each environment were subjected to high temperature stress at 40.6 C with a relative humidity of 50% and with both feed and water available. The effect of temperature stress on the birds from the selected lines and growing environments was measured as the percent mortality at variable periods after the initiation of stress. Except in Experiment 1, the comparisons were made when approximately 50% of all birds had died. Experiment 1. Males from replicate 2 after one generation of selection were used. When the birds were 9 weeks of age, the temperature in both the hot and cold chambers was raised to 40.6 C. Deaths occurred so rapidly that the temperature was returned to the original temperature after 6 hr. Mortality continued for 2 hr more, so the percent mortality was determined 8 hr after the initiation of the stress when 24.3% of the birds had died. There were 21 males observed for each selected line and growing environment. Experiment 2. Females of generation 2, replicate 1 were used for Experiment 2. When the birds were 9 weeks of age, the temperature in both the hot and cold chambers was raised to 37.8 C for 24 hr to provide some additional temperature acclimation to the birds reared in both environmental chambers. After 24 hr at 37.8 C, one pullet was chosen at random from each of 20 full-sib families in each line and growing environment. Five of the chosen
1805
individuals from each line and growing environment were randomly distributed into each of the four pens in the vertical column of a growing battery in the chamber previously assigned to the hot (32.2 C) environment. The temperature was then elevated to 40.6 C for the duration of the stress test. The percent mortality was analyzed after 17 hr of stress when 50.6% of the birds had died. Experiment 3. Females from generation 2, replicate 2, were used in Experiment 3. To increase the difference in possible temperature acclimation between the two rearing environments, at 9 weeks of age the temperature in the hot chamber was elevated to 37.8 C for 48 hr whereas the temperature in the cold chamber remained at 21.1 C. At the end of 48 hr, one progeny of each of 32 full-sib families in each line in the cold room was chosen at random and two of the 32 progeny of each line were randomly placed in each of 16 pens in the chamber previously assigned to the cold (21.1 C) environment. The same procedure for choosing progeny was followed in the hot room, and the resulting 32 birds were randomly distributed to 16 other pens in the cold chamber. The temperature was then elevated to 40.6 C for the duration of the stress test. The percent mortality was determined after 9.5 hr when 55.1% of the birds had died. Experiment 4. Females of generation 3, replicate 1 were used. At 9 weeks of age, the temperature in the hot room was increased to 37.8 C for 48 hr. After 48 hr, one pullet from each line in each growing environment was placed in each of 32 pens in the chamber previously assigned to the hot growing environment. Thus, each pen contained 8 pullets, one from each line grown in each environment. The percent mortality was calculated after 4 hr of exposure to the high temperature stress when 49.1% of the birds had died. The percentages were transformed to angles corresponding to the arcsin -J proportion after adjustment of all zero values to l/4n and all 100% values to 1 — l/4n as shown by Steel and Torrie (1980, page 236). Analysis of variance was applied to the transformed values based on the fixed model Y... = a + L. + E. + EL.. + e..... uk
^
i
J
U
(u)k
where fi represents the population mean, Lj is the effect of the i t h line (i = 1..4), Ej is the effect of the j growing environment (j =
1806
BOHREN ET AL.
1,2), and e,.., k is the random variation among individuals in the same line and reared from 5 to 9 weeks of age in the same environment. The random error was estimated for the transformed binomial data as the theoretical value 821/n, where n is the number of observations in each group (Steel and Torrie, 1980). RESULTS The mortality percentages observed for each line and rearing environment for the four experiments are shown in Table 1. The analyses of variance for the four experiments are shown in Table 2. In all experiments, the birds reared in the cold (21.1 C) environment had a significantly higher percent mortality than those reared in the hot (32.2 C) environment. Since the body weights and gains for the cold rearing environment were always greater, and the mortality percentages were higher than those for the hot rearing environment, the effects of body weight and rearing environment on percent mortality are completely confounded. An effort was made to differentiate between the effect of body weight and temperature preconditioning on the percent mortality. Short periods of higher temperatures were applied after the birds were 9 weeks of age. This short
interval treatment had little effect on body weight but would show the effect of additional temperature acclimation. For example, in Experiment 2, both chambers were increased to 37.8 C for 24 hr before applying the extreme stress at 40.6 C. In this case, 17 hr at 40.6 C were required to provide approximately 50% mortality. In Experiments 3 and 4, where only the hot chamber was increased to 37.8 C for 48 hr, the difference in mortality under extreme stress between birds grown in the hot and cold temperatures was increased. In addition, only 9.5 and 4 hr of extreme stress, respectively, were required to produce approximately 50% total mortality in these two experiments, with high frequencies of death among those grown in the cold temperature occurring in the first hour or two of stress. These experiments were also conducted to determine if differences in survival under heat stress existed among the four lines selected for fast and slow growth in the hot and cold environmental chambers. In Experiments 1 and 3 conducted on generations 1 and 3 of replicate 2, no significant differences among lines were observed. However, in Experiments 2 and 4 conducted on generations 2 and 3 of replicate 1 significant line effects were found. In these two experiments, the line effect mean squares were partitioned into orthogonal contrasts with
TABLE 1. Percent mortality under temperature stress at 40.6 C for four experiments.1 Lines Environment of selei:tion
Direction of selection
Growing environment Hot
Cold
X
Hot
Growing environmient Cold X
- Experiment 1 • 1 2 3 4
Hot Hot Cold Cold
Fast Slow Fast Slow
Mean
- Experiment 2 •
4.0 5.9 14.3 5.6
32.0 58.8 23.8 50.0
18.0 32.4 19.0 27.8
45.0 30.0 70.0 10.0
60.0 65.0 85.0 40.0
52.5 47.5 77.5 25.0
7.4
41.2
24.3
38.8
62.5
50.6
- Experiment 3 • 1 2 3 4 Mean
Hot Hot Cold Cold
Fast Slow Fast Slow
- Experiment 4 •
25.0 21.9 21.9 34.4
87.5 90.6 84.4 75.0
56.2 56.2 53.2 54.7
28.1 3.1 9.4 .8
98.2 75.0 96.9 81.2
63.2 39.0 53.2 41.0
25.8
84.4
55.1
10.4
87.8
49.1
1 The number of generations of selection for growth, replicate, number of observations in each cell, and the hours after initiation of stress are: Experiment 1, generation 1, replicate 2, 21 observations, 8 hr; Experiment 2, generation 2, replicate 1, 20 observations, 17 hr; Experiment 3, generation 2, replicate 2, 32 observations, 9.5 hr; Experiment 4, generation 3, replicate 1, 32 observations, 4 hr.
HEAT STRESS ON LINES SELECTED FOR GROWTH
1807
TABLE 2. Analyses of variance of percent mortality (transformed to angles in degrees) from temperature stress at 40.6 C
Source
df
Mean squares Experiment 1, Experiment 2, Generation 1, Replicate 2 Generation 2, Replicate 1 Mean squares
Between lines (L) Directions of selection (D) Environment of selection (ES) DX ES Between test environments (E) EX L Error (821/n)
Between lines (L) Directions of selection (D) Environment of selection (ES) DX ES Between test environments (E) EX L Error (821/n)
31.92
3 1 1 1 1 3
-
1183.17** 80.90 39.10
3
Experiment 3. Generation 2, Replicate 2 3.52 1 1 1
1 3
2687.55** 37.98 25.66
single_ degrees of freedom (Table 2). No significant difference in mortality was observed between the two environments in which selection occurred, but the lines selected for fast growth had significantly higher mortality in both experiments than the lines selected for slow growth. Therefore, a significant correlated response in percent mortality under heat stress from selection for 5 to 9-week gain occurred in replicate 1 but not in replicate 2. Genetic correlations between the weight gains from 5 to 9 weeks of age and the percent mortality under temperature stress were estimated after two generations of selection in replicate 2 and after three generations of selection in replicate 1 by use of equation 19.5 in Falconer (1960). These genetic correlation estimates were —.004 for replicate 2 and +.076 for replicate 1. The average estimate over both replicates was +.036. The interaction between the directions of selection and the environment in which selection occurred was significant in both Experiments 2 and 4. However, the interactions in the two experiments were quite different. In Experiment 2, the difference in mortality between the lines selected for fast and slow growth was greater in the cold selection environment, whereas in Experiment 4 the difference between the directions of selection was
369.53** 326.61** .21 781.21** 456.46** 20.80 41.05 Experiment 4, Generation 3, Replicate 1 240.57** 327.52** 17.91 376.29** 6126.70** 19.18 25.66
greatest in the hot selection environment. No reason was evident why this interaction should be different in the two experiments. DISCUSSION Interest centers on whether differences in body weight gains, either environmentally or genetically induced, are associated with differences in ability to survive under high temperature stress. Significantly larger weight gains and significantly lower survival were found for birds reared in the cold environment in all experiments. Because the birds compared in the two environments were genetically similar, one could conclude that a large negative environmental correlation exists between body weight and survival under heat stress. However, preconditioning at the two temperatures undoubtedly resulted in many physiological differences between the groups in addition to the weight difference. This is indicated by the effect of the 24 and 48-hr conditioning at 37.8 C, which would have little effect on body weight, but which had an effect on survival time under the extreme stress. Genetically determined weight differences due to selection for fast and slow gain from 5
1808
BOHREN ET AL.
t o 9 weeks gave different results for t h e t w o replicates. Under t e m p e r a t u r e stress, replicate 1 showed significant differences in m o r t a l i t y percentages (23.5%) b e t w e e n t h e lines selected in different directions, whereas replicate 2 showed n o significant differences in m o r tality percentages b e t w e e n t h e fast and slow lines. This result occurred in spite of t h e fact t h a t t h e differences b e t w e e n t h e fast and slow lines in gain and 9-week b o d y weight were nearly t h e same in t h e t w o replicates. Whereas differences in b o d y weight b e t w e e n t h e fast and slow growing lines appear t o be associated w i t h survival at high t e m p e r a t u r e , t h e correlated response in p e r c e n t m o r t a l i t y w h e n averaged over t h e t w o replicates (8.6%) suggests t h a t selection for fast and slow gain is n o t very effective in differentiating t h e fast and slow growing lines for survival ability u n d e r e x t r e m e stress. T h e differences b e t w e e n t h e t w o replicates could be d u e t o genetic drift of o t h e r factors m o r e i m p o r t a n t t h a n b o d y weight in producing differences b e t w e e n lines in survival at high t e m p e r a t u r e s . Selection for g r o w t h in t h e h o t and cold environments might be expected to select for genes increasing o r decreasing survival at higher t e m p e r a t u r e s , b u t only small and nonsignificant differences in survival percentages were f o u n d b e t w e e n t h e selection environments.
REFERENCES Bohren, B. B., J. R. Carson, and J. C. Rogler, 1981. Response to selection in two temperatures for fast and slow growth to nine weeks of age. Genetics 97:443-450. Edens, F. W., and H. S. Siegel, 1976. Modification of
corticosterone and glucose reponse by sympatholytic agents in young chickens during acute heat exposure. Poultry Sci. 55:1704-1712. Falconer, D. S., 1960. Introduction to Quantitative Genetics. The Ronald Press Co., New York, NY. Fox, T. W., 1951. Studies on heat tolerance in the domestic fowl. Poultry Sci. 40:477—483. Fox, T. W., 1980. The effect of thiouracil and thyroxine on resistance to heat shock. Poultry Sci. 59:2391-2396. Freeman, B. M., 1971. Stress and the domestic fowl: A physiological appraisal. World's Poultry Sci. J. 27:263-275. Hillerman, J. P., and W. O. Wilson, 1955. Acclimatization of adult chickens to environmental temperature changes. Am. J. Physiol. 180:591 — 595. Marks, H. L., and T. M. Huston, 1973. Response of selected quail lines to heat stress. Poultry Sci. 52:1668-1670. Siegel, H. S., 1969. Environmental stress and physiological compensating mechanisms in fowl: Temperature and respiratory regulation. Poultry Sci. 4 8 : 2 2 - 3 0 . Steel, R.G.D., and J. H. Torrie, 1980. Principles and Procedures of Statistics. McGraw-Hill Book Co., Inc., New York, NY. Washburn, K. W., R. Peavey, and G. M. Renwick, 1980. Relationship of strain variation and feed restriction to variation in blood pressure and reponse to heat stress. Poultry Sci. 59:2586— 2588. Wilson, H. R., A. E. Armas, I. J. Ross, R. W. Dorminey, and C. J. Wilcox, 1966. Familial differences of Single Comb White Leghorn chickens in tolerance to high ambient temperatures. Poultry Sci. 45:784-788. Wilson, H. R., D. W. MacLaury, T. H. Johnson, and C. D. Baird, 1975a. Relationship of heat tolerance and oxygen consumption in chickens. Poultry Sci. 54:299-300. Wilson, H. R., C. J. Wilcox, R. A. Voitle, C. D. Baird, and R. W. Dorminey, 1975b. Characteristics of White Leghorn chickens selected for heat tolerance. Poultry Sci. 54:126-130.