- Email: [email protected]
Intensity and duration of corticosterone response to stressful situations in Japanese quail divergently selected for tonic immobility
Available online at www.sciencedirect.com
General and Comparative Endocrinology 155 (2008) 288–297 www.elsevier.com/locate/ygcen
Intensity and duration of corticosterone response to stressful situations in Japanese quail divergently selected for tonic immobility D. Hazard a, M. Couty a, S. Richard b, D. Gue´mene´ a
a,*
UR83-Unite´ de Recherches Avicoles, Institut National de la Recherche Agronomique, Centre de Tours-Nouzilly, 37380 Nouzilly, France b UMR85 INRA-CNRS-Universite´ de Tours-Haras Nationaux, Physiologie de la Reproduction et des Comportements, Centre INRA de Tours-Nouzilly, 37380 Nouzilly, France Received 6 December 2006; revised 26 April 2007; accepted 7 May 2007 Available online 13 May 2007
Abstract Two genotypes of Japanese quail have been divergently selected since the 1980s for long (LTI) or short (STI) duration of tonic immobility [Mills, A.D., Faure, J.M., 1991. Divergent selection for duration of tonic immobility and social reinstatement behavior in Japanese quail (Coturnix coturnix japonica) chicks. J. Comp. Psychol. 105(1), 25–38.], an unlearnt catatonic state characteristic of a behavioral fear response ([Jones, R.B., 1986. The tonic immobility reaction of the domestic fowl: a review. World’s Poult. Sci. J. 42(1), 82–97.]; [Mills, A.D., Faure, J.M., 1991. Divergent selection for duration of tonic immobility and social reinstatement behavior in Japanese quail (Coturnix coturnix japonica) chicks. J. Comp. Psychol. 105(1), 25–38.]). The results of several behavioral tests conducted in LTI and STI quail have led to the conclusion that LTI quail are more fearful than STI quail [Faure, J.M., Mills, A.D., 1998. Improving the adaptability of animals by selection. In: Grandin, T. (Eds.), Genetics and the behavior of domestic animals. Academic Press, San Diego, pp. 235–264.]). However, few studies to date have focused on the Hypothalamic-Pituitary-Adrenal (HPA) axis response to stressful situations in LTI and STI quail, although the HPA axis is involved in fear responses [Siegel, H.S., 1971. Adrenals, Stress and the Environment. World’s Poult. Sci. J. 27, 327–349.]. The corticosterone (CORT) response to various putatively stressful situations was therefore assessed in LTI and STI genotypes of quail in order to investigate their HPA axis reactivity to stress. Repeated induction of TI or 1 min manual restraint induced significant and comparable increases in CORT levels in both genotypes as a TI response. On the other hand, higher CORT responses were found in STI than in LTI quail when the manual restraint period lasted for 2 min or after restraint in a crush cage. Maximum CORT responses and genotype differences were maintained throughout the latter test even when it lasted for 120 min. Investigation of the CORT response to a single TI episode showed that CORT levels at the end of TI were negatively correlated with TI duration. Other experimental contexts consisting of isolation in a familiar or novel environment or the presentation of a novel object induced slight but significant and comparable increases in CORT response in both genotypes, whereas change of cagemates did not. In conclusion, the present findings indicate that differences in HPA axis response are observed between LTI and STI genotypes when quail are submitted to intense stressors, resulting in a high and prolonged CORT response. By contrast, plasma corticosterone concentrations do not differ between STI and LTI quail in response to stressful situations of lower intensity, which evoke responses limited in amplitude and duration. Genetic selection for divergent duration of TI has thus affected the HPA axis response to stress, and LTI and STI quail constitute an interesting model to investigate genetic variability of HPA axis activity in birds. More specifically, these genotypes of quail could be used to investigate the occurrence of functional differences at different levels of the HPA axis in order to explain the present findings. 2007 Elsevier Inc. All rights reserved. Keywords: HPA axis; Corticosterone; Stress; Quail; Tonic immobility
1. Introduction *
Corresponding author. Fax: +33 2 47 42 77 78. E-mail address: [email protected] (D. Gue´mene´).
0016-6480/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.ygcen.2007.05.009
Several genetic selection programs have been set up for different fear–related behavioral traits in mammals (rats,
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
Bignami, 1965; mice, DeFries et al., 1978) and in birds (chickens (Faure and Folmer, 1975); pheasants (Boyer et al., 1973); quail (Mills and Faure, 1991). The results of these selection programs indicate that animals from divergent genotypes have different behavioral and physiological responses to fearful situations. Moreover, studies in rodents have reported coherent sets of behavioral and physiological responses which are consistent over time and across situations, better known as coping styles (Koolhaas et al., 1999). Thus, rats may be classified as proactive or reactive, depending on the way they respond to stressors. Similar classifications have been established in other species, notably in pigs (Hessing et al., 1994), hens (Korte et al., 1997) and great tits (Groothuis and Carere, 2005). Some animals react in an active fashion (fight or flight) while others are more likely to show inhibited behavior during the acute fear response (Koolhaas et al., 1999). The latter cease all ongoing activities and immediately freeze when a source of danger is detected, thereby reducing the likelihood of detection and attack from a predator. In such situations, they may show a tonic immobility (TI) response if caught (feigning death) (Jones, 1986; Gentle et al., 1989). TI has therefore been described as an unlearnt catatonic state which is thought to be the final stage in a chain of anti-predator behavior patterns (Jones, 1986; Mills and Faure, 1991). Two genotypes of Japanese quail have been divergently selected since the 1980s for long (LTI) or short (STI) duration of tonic immobility (Mills and Faure, 1991). Several behavioral tests conducted in LTI and STI quail demonstrated that quail which show a long duration of TI response are more fearful than quail which show a short duration of TI response, and that duration of the TI reaction is a reliable indicator of underlying fearfulness (Jones and Mills, 1983; Mills and Faure, 1986; Jones et al., 1991; Faure and Mills, 1998). LTI and STI quail have therefore been widely used as a biological model to investigate fearrelated behavioral responses in birds, but these quail also constitute an appropriate model to study the putative relationship between fearfulness and Hypothalamic–Pituitary– Adrenal (HPA) axis reactivity to stress. Indeed, elicitation of instant and adequate fear responses requires activation of the HPA axis and in the final stage release of corticosteroids (corticosterone in birds) from the adrenal glands (Siegel, 1971, 1980; Harvey and Hall, 1990), which are required to re-establish homeostasis via feedback mechanisms and to facilitate behavioral adaptation. Moreover, corticosteroids have been shown to influence the consolidation and potentiation of fear as well as the facilitation or extinction of avoidance in mammals (Schulkin et al., 1994, 1998). A line of Japanese quail divergently selected for their high corticosterone (CORT) response to restraint (Satterlee and Johnson, 1988) has been shown to exhibit the longest duration of TI (Jones et al., 1992; Satterlee et al., 1993). On the other hand, STI quail have been reported to exhibit higher CORT levels than LTI quail in response to restraint
289
stress (Jones et al., 1994; Re´mignon et al., 1998; Hazard et al., 2005a). However, only partial and rudimentary investigations into the CORT response to potentially stressful situations have been performed to date in LTI and STI quail (Launay, 1993; Jones et al., 1994; Faure et al., 1996; Re´mignon et al., 1998; Hazard et al., 2005a) and the results have been inconclusive. This is partly due to the lack of consistency in the experimental designs used (generation number, age, sex, rearing conditions, sampling interval. . .). This can be a very important issue, since CORT levels are very sensitive to any variation in physiological and environmental factors (Hazard et al., 2005a,b). It is therefore impossible from previous studies to reach firm conclusions about differences in the CORT response to stress between LTI and STI genotypes. The aim of the present study was to asses CORT responses in LTI and STI genotypes of quail submitted to a number of putatively stressful situations under careful and standardized experimental conditions in order to characterize HPA axis reactivity to stress in these genotypes and to explore further the possible effects of genetic selection for a fear-related behavioral response on HPA axis reactivity to stress. Duration of TI has been reported to be significantly prolonged when circulating CORT concentrations are increased by administration of CORT in domestic fowl (Jones et al., 1988). Moreover, greater TI responses have been reported to be associated with high endogenous levels of CORT in mammals (Carli et al., 1979; Kalin et al., 1998). Therefore, we first investigated the putative relationship between TI responses (i.e. the selection parameter) and CORT responses in LTI and STI genotypes. Then, CORT response patterns to various physical and social stressors were investigated in LTI and STI quail. The stressors consisted of restraint in a crush cage, presentation of a novel object, isolation in a familiar or a novel environment and change of cagemates. These situations were chosen because they have previously been shown to induce CORT release in birds and mammals (Seggie and Brown, 1975; Launay, 1993; Buijs et al., 1997; Ruis et al., 2001; Carobrez et al., 2002; Hazard et al., 2005a). Moreover, it was anticipated that studying the responses to isolation and placing in a novel environment might provide greater understanding of responses to restraint, since both of these stressors are components of the restraint test. 2. Materials and methods 2.1. Animals and rearing conditions Six-week-old Japanese quail (Coturnix japonica) from the 36th generation of two genotypes divergently selected for short (STI) or long (LTI) duration of tonic immobility (TI) were used in the study (Mills and Faure, 1991). Eight successive hatches were necessary to provide the quail used in the present study and the quail used for each specific test were from a single hatch. Quail were identified by wing-banding on the day of hatching. Quail were exposed to continuous light until 3 weeks of age, then to a 16 h light/8 h dark (16L/8D) (light on 6.00 am) rhythm. The quail from the different genotypes and experimental groups were reared in different collective battery cages. The caretaker checked the quail daily in the
290
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
morning (from 08.30 h) and refilled the feeders whenever necessary (at least twice a week). No care was provided on the day of the experiment. Food and water were provided ad libitum.
2.2. Experimental procedures Groups of 4–6 quail of the same genotype and sex were randomly constituted and placed in wire rearing battery cages one week before the experiments. At the age of 6 weeks, quail were submitted to one of 9 different stressors, depending upon the test being conducted. The stressors consisted of a tonic immobility test (Experiment 1a, 1b), repeated induction of TI (Experiment 1b), manual restraint (Experiment 1b), restraint in a crush cage for a short or long duration (Experiment 2), presentation of a novel object (Experiment 3a), isolation in a familiar (Experiment 3b) or novel environment (Experiment 3c), or change of cagemates (Experiment 3d). Quail from both genotypes and sexes were included in each experiment, except in Experiment 1a for which only LTI quail were used since the duration of TI in STI quail is too short to study the kinetics of the CORT response in this genotype. Following each treatment, blood was collected from each quail directly after decapitation and temporarily stored on ice in a tube containing EDTA (2 mg/ml blood). Following centrifugation at 2000g for 15 min at 4 C, plasma was separated and stored at 20C until measurement of CORT levels using a specific radio immunoassay (Etches, 1976). An additional group of non-treated quail, which were only captured and transferred to the test room for blood sampling, was included as a control in each experiment in order to measure basal corticosterone concentrations. This brief capture and transfer did not affect CORT levels (data not shown). Sacrifice by decapitation was used in this study since we have previously shown this sampling procedure affects basal CORT levels least compared to venipuncture, cardiac or jugular puncture (Hazard et al., 2004). Using this procedure, we were able to observe even very small amplitude responses. Serial bleedings were not considered since several studies have reported that serial bleedings increase CORT levels (Harvey et al., 1980; Johnson and van Tienhoven, 1981; Beuving and Vonder, 1986). Moreover, in order to avoid an unexpected effect on CORT levels of repeatedly taking a quail from the cage, all quail from one cage were bled at the same time.
2.3. Specific treatments 2.3.1. Experiment 1a: Tonic immobility Quail of the LTI genotype were tested individually for TI response after transfer to a testing room adjacent to the rearing room (ntotal = 91; ntotal refers to the number of quail used in each experiment and in the statistical analysis: 91 in the present experiment). Each quail was placed on its back in a U-shaped cradle, hand-restrained for 10 s to induce tonic immobility and then released; this procedure mimics that used for the selection program (Mills and Faure, 1991). When induction was unsuccessful (i.e. if TI lasted less than 10 s), the experimenter immediately reattempted TI induction (maximum of 5 attempts). When TI lasted at least 10 s, TI duration (i.e., the time until the quail righted itself) was recorded. In this first experiment, the test was terminated if the quail was still in TI after 20 min. Quail were bled either immediately after TI ended naturally or immediately after the experimenter terminated the TI. 2.3.2. Experiment 1b: Tonic immobility, repeated induction, manual restraint Tonic immobility. Quail from both genotypes were included in this second experiment and TI was induced as described above, except that after taking into account the results obtained in Experiment 1a, TI was limited to 5 min duration. Quail were bled immediately after TI ended naturally or following termination (i.e. after 5 min) (ntotal = 39). Quail failing to exhibit TI after 5 inductions were nevertheless bled and included in the group of quail used to measure CORT response after repeated induction (i.e. 5·). Moreover, in order to characterize CORT response to TI more fully and to investigate the effects of the TI induction procedure, CORT responses were measured following several TI induction procedures and manual restraint.
Repeated inductions. Quail were subjected to the TI induction procedure as described above. However, instead of being released at the end of the 10 s induction period, quail were manually righted and the procedure was repeated 5 times (ntotal = 28). Manual restraint. Under this treatment, the TI procedure was also mimicked but each quail was manually maintained on its back for 1 or 2 min (ntotal = 44). 2.3.3. Experiment 2: Restraint test in crush cage This experiment consisted of two independent trials. For each trial, quail were transferred to the test room and placed individually in a ‘‘crush-cage’’ in the form of a wooden box (15 cm length · 5 cm width x 10 cm height) closed at the top by a netting cover. Quail have only very restricted movement in such a crush cage. In the first trial, quail were restrained for 10 min and then either immediately bled or placed back in their home cages for 10, 30, 50 or 110 min and then bled (ntotal = 100). Quail were thus sacrificed 10, 20, 40, 60 or 120 min after the beginning of the restraint period. In the second trial, quail were restrained in the crush cage for 10, 20, 40, 60, or 120 min and then bled immediately (ntotal = 186). 2.3.4. Experiment 3a: Presentation of a novel object Each experimental battery cage initially contained a group of 4 quail. First, a control quail was captured and bled immediately after introducing a novel object into the battery cage. The novel object was in the form of a red balloon of 12 cm of diameter. The 3 remaining quail were then bled concomitantly 5, 10 or 60 min after introduction of the balloon (ntotal = 168). 2.3.5. Experiment 3b: Isolation in a familiar environment After capture in its home battery cage, each quail was placed alone in another battery cage of similar size and shape (35.5 cm length · 22 cm width · 16.5 cm height) in the same room for 5, 10 or 60 min (ntotal = 124). Visual communication between the isolated quail and the other quail in the room was prevented, but acoustic and olfactory communication was still possible. 2.3.6. Experiment 3c: Isolation in a novel environment The procedure used was comparable to that described above except that the cages where the quail were placed for the test consisted of large empty wooden boxes (65 cm length · 41 cm width · 28 cm height) closed at the top by a netting cover and located in an adjacent room (ntotal = 128). The use of a cage differing totally in structure and size from the home cages increased the probability of measuring CORT responses specific to the novel environment. Moreover, a larger cage than the home cage was used to test the effects of a novel environment since a cage smaller than the home cage had already been used for the restraint test and it was intended to avoid misinterpretation through observing an unexpected effect similar to the restraint. 2.3.7. Experiment 3d: Change of cagemates The procedure consisted of exchanging two quail out of 4 from one battery cage with two others of the same sex but from the opposite genotype. Quail of the LTI genotype had never previously met quail of the STI genotype. All quail from each of the newly formed groups were sacrificed at the same time, either 5, 10 or 60 min after transfer (ntotal = 124). In order to analyze the respective effects of social stress and/or transfer, data for quail which were submitted to both transfer and exchange or only exchange were first analyzed separately. Pairs of quail not transferred and not mixed with new counterparts constituted control groups.
2.4. Statistical analysis For each experiment, CORT values were subjected to multifactorial ANOVA to assess the effects of the different specific experimental factors and their interactions using the Statview V program (Abacus Concept Inc. Berkeley, USA). The factors tested were treatment/sampling time, genotype (except for Experiment 1a) and sex, and their respective interactions. Whenever ANOVA reached significance (P < 0.05), post-hoc tests were performed using the Fisher test (PLSD). Correlation analyses
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297 between CORT concentrations and duration of TI were performed. CORT values are expressed as means ± standard error and the level of significance was P < 0.05 unless otherwise stated.
3. Results 3.1. Experiment 1a: Tonic immobility TI was induced in 76 out of 83 quail after only 1 attempt, and a maximum of 2 inductions was necessary to induce TI in the other quail. Minimum TI duration was 31 s, whereas it was terminated intentionally by the experimenter in the 5 quail remaining 20 min in TI. Durations of TI were regularly distributed throughout the 20 min period. Correlation analyses calculated between CORT levels and duration of TI revealed that there was a significant negative correlation (r = 0.45, P < 0.0001) between these 2 variables (Fig. 1A). The data were assigned to 5 classes corresponding to successive periods of 5 min duration depending on TI duration: control, ]0–5], ]5–10], ]10–15], ]15–20]. ANOVA of CORT data with class and sex as independent variables showed that there was a significant effect of TI duration and no significant sex effect. Indeed, quail for which TI lasted between 31 s and 5 min exhibited significantly higher CORT levels (5.0 ± 0.4 ng/ml) than basal CORT levels (1.5 ± 0.3 ng/ml) (Fig. 1B). CORT levels in quail spending more than 5 min in TI were intermediate, mean values ranging from 3.0 ± 0.3 ng/ml to 2.3 ± 0.6 ng/ml for the periods ]5–10] and ]15–20], respectively, but the difference from basal CORT values no longer reached significance.
291
sexes (Fig. 2A and B). Genotype and time factors, and their interaction, significantly affected CORT responses whereas the sex factor did not (Fig. 2A). Placing in the crush-cage for 10 min induced a significant increase in CORT levels in both genotypes. The differences in CORT concentrations between undisturbed and restrained birds appeared to be higher in quail of the STI genotype, since 6- and 3-fold increases were measured for STI and LTI quail, respectively. When the quail were placed back in their home battery cages following the 10 min restraint period, CORT levels progressively decreased to reach levels that did not differ from basal levels within 10 and 30 min for LTI and STI quail, respectively. When the restraint period was prolonged, CORT levels remained stable at maximum levels for both genotypes (Fig. 2B) and were significantly lower in LTI than in STI quail throughout the restraint period (i.e. 10–120 min). 3.4. Experiment 3a: Presentation of a novel object Introduction of a novel object into the home cage induced a significant increase in CORT levels in female LTI quail and in both sexes of the STI quail, but not in LTI males. A significant sex effect was observed but there was no genotype effect (Fig. 3A). Maximum CORT levels were measured 5 min after presentation of the balloon and the highest CORT levels reached a maximum concentration of approximately 4.0 ng/ml. CORT concentrations did not differ from basal values 10 min after introduction of the balloon. 3.5. Experiment 3b: Isolation in a familiar environment
3.2. Experiment 1b: Tonic immobility, repeated induction and manual restraint Duration of TI ranged between 10 and 134 s in STI quail while the minimum duration in LTI quail was 25 s, and 50% of LTI quail were still in TI after 5 min (Fig. 1C). ANOVA of CORT levels in response to TI, repeated induction or manual restraint revealed no significant sex effect (Fig. 1D). TI was associated with significant and comparable increases in CORT levels in the two genotypes, and the average CORT level calculated for both genotypes was 6.8 ± 0.6 ng/ml (Fig. 1D). Five repetitions of the 10 s induction procedure induced similar increases in CORT levels (5.1 ± 0.6 ng/ml) in both genotypes (Fig. 1D). Similarly, 1 min manual restraint induced significant and comparable increases in CORT levels in both genotypes (6.0 ± 0.7 ng/ml) (Fig. 1D). By contrast, 2 min manual restraint induced a significantly greater increase in CORT levels in STI quail (11.0 ± 2.0 ng/ml) than in LTI quail (6.3 ± 0.7 ng/ml) (Fig. 1D).
Isolation of quail in a familiar environment induced significant increases in CORT levels (P = 0.05) (Fig. 3B). A significant sex effect (P = 0.0001) upon CORT levels was found whereas the genotype effect was not significant (P = 0.9). Maximum CORT levels of approximately 3.0 ng/ml were measured after 5 min of isolation in STI males, whereas a delay of 10 min was observed for the other 3 experimental groups. 3.6. Experiment 3c: Isolation in a novel environment Social isolation in a novel environment induced a significant increase in CORT levels (P < 0.001) whereas neither the genotype (P = 0.78) nor the sex (P = 0.2) effects were significant (Fig. 3C). Maximum CORT levels which remained lower than 4.0 g/ml were measured at 5 min after beginning isolation. After 10 min of isolation in the novel environment, CORT concentrations were comparable to basal values.
3.3. Experiment 2: Restraint test in a crush cage
3.7. Experiment 3d: Change of cagemates
The basal CORT levels measured for control quail were low and did not differ significantly between genotypes or
Neither the effect of exchange (P = 0.15), either for the quail that were submitted to both transfer and exchange
292
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
A
CORT
15 (ng/ml) r = - 0.45 P < 0.0001
10
CORT (ng/ml)
B
ANOVA
10
Treatment P=0.0002 Sex P=0.85
b 5
5
a
a
a
a
C
0
TI (sec)
5
TI 20 (min)
10
15
D
CORT
0 C ]0 -5 ]5 ] -1 ]10 0] -1 ]1 5] 520 ]
0
*
15 (ng/ml)
300
200
10
c
b b b
ANOVA Genotype Sex Treatment Gen. x Sex Gen. x Treat.
P=0.3 P=0.24 P=0.001 P=0.1 P=0.06
b
b 100
TI (min)
b
b
5
LTI a
a
STI
C Ix 5 T M I R M 1 R 2
LT I ST I
C Ix 5 M TI R M 1 R 2
0
0
Fig. 1. Changes in corticosterone concentrations over time (CORT, ng/ml plasma) in LTI quail following tonic immobility (A and B) (7 6 n 6 39) (ntotal = 91) Duration of tonic immobility in LTI and STI quail (C) (nLTI = 13, nSTI = 26). Corticosterone concentrations (ng/ml plasma) in LTI and STI quail following 5 induction procedures (I ·5) with immobilization over 10 s for each, tonic immobility period (TI), 1 min (MR 1) and 2 min (MR 2) of manual restraint (D) (10 6 n 626). Control samples (C) were collected from undisturbed quail. (means ± SE) a, b & c = means for a given genotype with common superscripts were not significantly different from each another (P > 0.05) *significantly different between genotypes (P < 0.05).
or only to exchange, nor the effect of genotype (P = 0.15) significantly affected CORT levels (Fig. 3D). The significant sex effect (P < .0001) observed resulted from higher basal CORT levels in male than in female quail. 4. Discussion First HPA axis activity was characterized during the course of the TI response in LTI and STI genotypes, TI corresponding to the selection parameter used (Mills and Faure, 1991). Interestingly, changes in CORT levels after the TI test did not differ significantly between the two genotypes, although (as expected) TI behavior lasted longer in LTI than in STI quail. Furthermore, maximum CORT levels were measured in LTI quail showing the shortest duration of TI response, while CORT responses in LTI quail which remained naturally for less than 5 min in TI and in LTI quail for which the TI response was terminated 5 min after onset were comparable. Although we cannot exclude the possibility that this initial increase in CORT levels at the onset of the TI response played a critical role
in the duration of this behavioral response, this is not evident from the present findings. Indeed, in such a case CORT levels at the onset of the TI response in quail expressing longer TI responses should be higher than in those expressing shorter duration responses. However, while TI responses lasted longer in LTI than in STI quail, CORT responses at the beginning of the TI test (i.e. following TI induction) were similar in both genotypes. Moreover, a negative correlation was found between CORT levels and duration of the TI response in LTI quail. Thus, although CORT concentrations in quail spending more than 5 min in TI were intermediate between basal and maximum concentrations, our results suggest that maintenance in a TI state does not require maintenance of high CORT concentrations, and the TI state, which is a behavioral fear response, does not maintain high CORT levels. Moreover, return of CORT levels to basal values is not a prerequisite for the quail to right itself. Indeed, quail showing the shortest TI responses righted themselves whereas they had higher than basal CORT levels. The findings of the present study thus suggest that there is no relationship between
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
A
293
CORT (ng/ml)
30 ANOVA
c
25
Genotype P<0.004 Sex P=0.61 Time P<0.0001 Time x Gen. P=0.0004
20 b
15
a
a
LTI ab
ab
10 5
b
a
a
STI ab
a a
0 0 10 20 40 60 120
B
0 10 20 40 60 120
Time (min)
CORT (ng/ml)
30
ANOVA
25
b
b b
20 15
b ab
10 5 0
b
b b
Genotype Sex Time Interactions
P<0.0001 P=0.12 P=0.001 P>0.4
b
ab
a
a
0 10 20 40 60 120
Time 0 10 20 40 60 120 (min)
Fig. 2. Changes in corticosterone concentrations over time (CORT, ng/ml plasma) in LTI and STI quail following restraint period in a crush cage of fixed duration of 10 min. (A) (5 6 n 6 11) or of variable durations lasting 10, 20, 40, 60 or 120 min (B) (12 6 n 6 20) (means ± SE) control samples (time 0) were collected from undisturbed quail. The restraint period is shown by horizontal shaded bars. a, b & c = means for a given genotype with common superscripts were not significantly different from each another (P > 0.05).
duration of TI response and CORT levels. We can also hypothesize from the present findings that the initial increase in CORT associated with the TI response results from the process of manual restraint performed during the TI induction procedure. This hypothesis is further supported by the results that we obtained after manual restraint mimicking TI induction. Indeed, similar CORT responses were measured in both genotypes of quail following several TI induction procedures, manual restraint on the back for 1 min and following a short TI response. The hypothesis that the increase in CORT concentrations at the beginning of TI is due to the manual restraint during the induction procedure is also supported by similar increases in CORT levels reported in laying hens in response to manual restraint (Korte et al., 1997). Our present findings contrast somewhat with previously reported results indicating positive relationships between fearinduced behavior and circulating levels of glucocorticoids in different avian and mammalian species (Carli et al.,
1979; Kalin et al., 1998). Greater freezing and TI responses in rabbits (Carli et al., 1979) and rhesus monkeys (Kalin et al., 1998) were associated with high endogenous levels of CORT. By contrast, differences in TI duration between LTI and STI genotypes in the present study were not associated with different basal CORT levels. Moreover, in quail selected for divergent CORT responses to a brief period of immobilization (Satterlee and Johnson, 1988), quail with higher HPA axis reactivity also exhibited longer TI responses (Jones et al., 1992), whereas the opposite was found in LTI and STI genotypes (Hazard et al., 2005a). In addition, the tonic immobility reaction has been shown to be prolonged following CORT infusion in chicken (Jones et al., 1988), and exogenous administration of CORT potentiated fear-induced freezing behavior in rats (Coordimas et al., 1994). Although it would be important to investigate the effects of exogenous administration of CORT on TI duration in LTI and STI quail to firmly conclude on the relationship between CORT levels and TI
294
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
A 5
CORT (ng/ml)
LTI Fem.
Male
STI
b
4
b
3
b
b
a
a a
ANOVA
ab ab
2 1
Male
Fem.
a a
a
a
a a
a
Genotype P=0.5 Sex P=0.0001 Time P<0.0001 Time x Gen. P=0.03
0
B
5 4 ANOVA
3
Genotype Sex Time Interactions
2 1
P=0.9 P=0.0001 P=0.05 NS
0
C
5 4 ANOVA
3
Genotype Sex Time Interactions
2 1
P=0.78 P=0.2 P<0.001 NS
0
D
5 4 ANOVA
3
Genotype Sex Time Interaction
2 1 0 0 5 10 60 0 5 10 60
0 5 10 60
0 5 10 60
P=0.15 P<0.0001 P=0.15 NS
Time (min)
Fig. 3. Changes in corticosterone concentrations over time (CORT, ng/ml plasma) in LTI and STI quail following the presentation of a novel object (balloon) (A), isolation in a familiar environment (B) or a novel environment (C) and change of cagemates (D) for durations of 5, 10 or 60 min (6 6 n 6 12). (Means ± SE) Control samples (time 0) were collected from undisturbed quail. a & b = means for a given genotype with common superscripts were not significantly different from each another (P > 0.05).
response, it is likely that the duration of TI does not depend on CORT levels in these genotypes of quail. The present study provides new information about the time-course of the CORT response, especially in the case of long lasting acute stress, since only the short term effects of restraint lasting 5 (Jones et al., 1994), 10 (Hazard et al., 2005a) or 15 min (Re´mignon et al., 1998) have previously been investigated. The present findings showed that differences in CORT responses between the two genotypes appeared very quickly, since significantly higher CORT levels were observed in the STI than in LTI genotype after 2 min of manual restraint. The CORT response reached its
maximum before 10 min in both genotypes, since CORT levels did not increase further after this interval in the restraint situation. This result is consistent with CORT responses reported in high and low feather pecking genotypes of laying hens, in which the difference in the increases of CORT levels between the two genotypes was maximum between 1 and 4 min (Korte et al., 1997). Interestingly, increases in CORT levels were observed in response to all the experimental stressors tested, with the exception of social stress induced by cagemate exchange. However, LTI and STI quail did not differ in their CORT responses to social isolation or presentation of a novel
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
object, and the increases observed were of much lower amplitude than those induced in the restraint test. In addition, CORT levels induced in response to isolation or the presentation of a novel object did not remain stable throughout testing but returned quickly to basal values, while CORT concentrations induced in response to restraint in the crush cage remained at maximum levels in both genotypes throughout the 120 min period. It appears from the present study that differences in HPA axis responsiveness between the 2 genotypes only prevailed following manual restraint and restraint in a crush cage. The latter treatment appeared to be the most stressful situation according to the intensity and duration of the CORT responses and thus required high CORT synthesis and release from the adrenal glands. LTI quail have been reported to be more fearful than STI quail (Jones et al., 1991). Therefore, since the HPA axis is involved in the response to fearful stimuli and corticosterone is reported to be a valuable indicator of the stress response (Siegel, 1980; Harvey and Hall, 1990), we would have expected LTI quail to have higher CORT responses to fearful stimuli than STI quail. Indeed, using a reverse strategy consisting of selecting quail on physiological responses (i.e., on high or low CORT responses to a restraint), Satterlee and Johnson (1988) showed that the high stress line exhibited the highest CORT response and the longest TI response (Jones et al., 1992). However, the differences observed indicate that the opposite situation occurred in the present study, as in a previous study (Hazard et al., 2005a).Whatever the underlying mechanisms, the apparently conflicting findings (i.e., higher CORT responses to restraint associated with shorter TI responses) in our genotypes and the reverse situation in others could be viewed as evidence that TI reactions and adrenocortical responses are not mutually coselected in LTI and STI quail. This hypothesis is further supported by the fact that no relationship between CORT concentration after restraint and TI was found in a second generation cross (F2) between the LTI and STI genotypes (Mignon-Grasteau et al., 2003). We could hypothesize that differences in CORT response to restraint between genotypes involved differences in CORT synthesis capacity. However, differences in CORT synthesis capacity do not seem to be involved in the genotype difference, since we have previously reported comparable CORT responses to 1–24 ACTH injection in LTI and STI quail (Hazard et al., 2005a). This hypothesis remains to be investigated more fully before reaching a firm conclusion regarding adrenal involvement in the genotype difference in CORT responses to restraint. Moreover, differences in functioning of the pituitary and/or hypothalamus might also be involved in the differences in CORT response to stress between LTI and STI quail and thus remain to be investigated. Measurement of circulating ACTH could provide information about pituitary activity in STI and LTI quail. Moreover, greater understanding of the differential HPA responsiveness in STI and LTI quail would also benefit from investigation into the cortico-
295
steroid-binding globulin levels in these two genotypes of quail. Indeed, corticosteroid-binding globulins influence the bioavailability of CORT to target cells, and similar plasma concentrations of CORT in STI and LTI quail may mask different levels of free CORT (Deviche et al., 2001). Nevertheless, the divergent results between CORT and TI responses are not necessarily inconsistent, since it is possible that CORT responses are not related to the degree of fearfulness but are somehow a reliable indicator of the physiological response developed by birds to cope with a specific stressful situation. Indeed, CORT produced in response to stress can be used to provide energy to support the physical response (Rees et al., 1983; Joseph and Ramachandran, 1992). For instance, an increase in CORT levels associated with physical activity has been reported in ducks (Rees et al., 1985). Thus, it cannot be excluded that the differences in CORT response to restraint might result from differences in behavior being expressed during the test, since STI quail have previously been reported to struggle frequently during 10 min restraint in a crush cage, whereas this behavior was virtually not expressed by quail of the LTI genotype (Jones et al., 1994). However, such a relationship between CORT response and fear behavior is not straightforward. Indeed, proactive animals, which exhibit active behavior in response to various stressful situations, are traditionally characterized by weak HPA axis reactivity (Koolhaas et al., 1999). Finally, we can also hypothesize that differences in CORT response between genotypes may involve different psychobiological processes of perception, resulting in limited HPA axis activation in the LTI genotype and/or hyper-activation of the HPA axis in STI quail. Indeed, the higher HPA axis response to restraint than to all other stressors in STI quail suggests that quail of this genotype may perceive this stress differently whereas this seemed not to be the case in LTI quail since their HPA axis response was fairly similar whatever the stressor. However, this study does not provide sufficient evidence to support this hypothesis and further investigations into HPA axis function at different levels will be necessary to understand the relationships between animals’ perceptions of a stressful situation, their behavioral responses and their endocrine responses. In conclusion, HPA axis responsiveness was comparable in the two genotypes when the CORT response was limited in both intensity and duration, i.e. in response to TI, presentation of a novel object, and isolation in a familiar or novel environment, suggesting that such situations are weakly stressful. In contrast, genotype differences were observed when CORT concentrations reached higher levels in STI quail in response to manual restraint and restraint in the crush cage. In the latter experimental situations, high CORT levels remained throughout when exposure to the stimulus was prolonged. Clearly the selection program for divergent TI responses has not only had specific effects on TI duration, but has also affected the maximum CORT
296
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
response to restraint stress in an unexpected way. Thus, since increases in CORT levels towards maximum levels and maintenance at such concentrations have negative effects on physiological functions and various production parameters such as growth and meat quality (Chen et al., 1991; Re´mignon et al., 1998), the use of a behavioral parameter such as TI in a selection program should not be implemented before the consequences have been more fully characterized. Acknowledgments We thank Dr. A.D. Mills and Dr. J.M. Faure who set up and managed the selection program that provided the quail used in this study. We also thank D. Raine for a valuable contribution to improving the quality of the English manuscript. The authors also thank the many people who contributed to this study, especially J.-M. Brigant and J.-M. Hervouet for expert technical assistance. D. Hazard was supported by grants from the Institut National de la Recherche Agronomique and the Conseil Re´gional de la Re´gion Centre for completion of a Ph.D. References Beuving, G., Vonder, G.M., 1986. Comparison of the adrenal sensitivity to ACTH of laying hens with immobilization and plasma baseline levels of corticosterone. Gen. Comp. Endocrinol. 62 (3), 353–358. Bignami, G., 1965. Selection for high rates and low rates of avoidance conditioning in the rat. Anim. Behav. 13 (2), 221–227. Boyer, J.P., Melin, J.M., Bourdens, P., 1973. Activity test in young pheasants. Ann. Genet. Sel. Anim. 5, 417–418. Buijs, R.M., Wortel, J., Van Heerikhuize, J.J., Kalsbeek, A., 1997. Novel environment induced inhibition of corticosterone secretion: physiological evidence for a suprachiasmatic nucleus mediated neuronal hypothalamo-adrenal cortex pathway. Brain Res. 758 (1–2), 229–236. Carli, G., Farabollini, F., di Prisco, C.L., 1979. Plasma corticosterone and its relation to susceptibility to animal hypnosis in rabbits. Neurosci. Lett. 11 (3), 271–274. Carobrez, S.G., Gasparotto, O.C., Buwalda, B., Bohus, B., 2002. Longterm consequences of social stress on corticosterone and IL-1[beta] levels in endotoxin-challenged rats. Physiol. Behav. 76 (1), 99–105. Chen, M.T., Lin, S.S., Lin, L.C., 1991. Effect of stresses before slaughter on changes to the physiological, biochemical and physical characteristics of duck muscle. Br. Poult. Sci. 32 (5), 997–1004. Coordimas, K.P., LeDoux, J.E., Gold, P.W., Schulkin, J., 1994. Corticosterone potentiation of learned fear. In: de Kloert, E.R., Azmatia, E.C., Longfield, P.W. (Eds.), Brain Corticosteroid Receptors: Studies of the Mechanism, Function and Neurotoxicity of Corticosteroid Action. New York Academy of Sciences, New York, pp. 392–393. DeFries, J.C., Gervais, M.C., Thomas, E.A., 1978. Response to 30 generations of selection for open-field activity in laboratory mice. Behav. Genet. 8 (1), 3–13. Deviche, P., Breuner, C., Orchinik, M., 2001. Testosterone, corticosterone, and photoperiod interact to regulate plasma levels of binding globulin and free steroid hormone in Dark-eyed Juncos, Junco hyemalis. Gen. Comp. Endocrinol. 122 (1), 67–77. Etches, R.J., 1976. A radioimmunoassay for corticosterone and its application to the measurement of stress in poultry. Steroids 28 (6), 763–773. Faure, J.M., Folmer, J.C., 1975. Etude ge´ne´tique de l’activite´ pre´coce en open-field du jeune poussin. Ann. Genet. Sel. Anim. 7, 123–132.
Faure, J.M., Mills, A.D., 1998. Improving the adaptability of animals by selection. In: Grandin, T. (Ed.), Genetics and the behavior of domestic animals. Academic Press, San Diego, pp. 235–264. Faure, J.M., Mills, A.D., Launay, F., Francois, N., Williams, J., Gue´mene´, D., Re´mignon, H., 1996. Se´lection pour le comportement: l’exemple de la caille. Compte rendu de l’Acade´mie d’Agriculture de France 82, 61–70. Gentle, M.J., Jones, R.B., Woolley, S.C., 1989. Physiological changes during tonic immobility in Gallus gallus var domesticus. Physiol. Behav. 46 (5), 843–847. Groothuis, T.G., Carere, C., 2005. Avian personalities: characterization and epigenesis. Neurosci. Biobehav. Rev. 29 (1), 137–150. Harvey, S., Hall, T.R., 1990. Hormones and stress in birds: activation of the hypothalamo-pituitary-adrenal axis. Prog. Comp. Endocrinol. 342, 453–460. Harvey, S., Merry, B.J., Phillips, J.G., 1980. Influence of stress on the secretion of corticosterone in the duck (Anas platyrhynchos). J. Endocrinol. 87 (1), 161–171. Hazard, D., Faure, J. M., Gue´mene´, D., 2004. Effects of sampling sites, photoperiod, and age on adrenal responsiveness in Japanese quail. Eighth International Symposium on Avian Endocrinology. Scottsdale, Arizona, USA, p 75. Hazard, D., Couty, M., Faure, J.M., Gue´mene´, D., 2005a. Relationship between Hypothalamic-Pituitary-Adrenal Axis Responsiveness and Age, Sexual Maturity Status and Sex in Japanese Quail Selected for Long or Short Duration of Tonic Immobility. Poult. Sci. 84 (12), 1913–1919. Hazard, D., Couty, M., Faure, J.M., Gue´mene´, D., 2005b. Daily and photoperiod variations of Hypothalamic-Pituitary-Adrenal Axis Responsiveness in Japanese Quail Selected for Short or Long Tonic Immobility. Poult. Sci. 84, 1920–1925. Hessing, M.J., Hagelso, A.M., Schouten, W.G., Wiepkema, P.R., van Beek, J.A., 1994. Individual behavioral and physiological strategies in pigs. Physiol. Behav. 55 (1), 39–46. Johnson, A.L., van Tienhoven, A., 1981. Plasma concentrations and corticosterone relative to photoperiod, oviposition, and ovulation in the domestic hen. Gen. Comp. Endocrinol. 43 (1), 10–16. Jones, R.B., 1986. The tonic immobility reaction of the domestic fowl: a review. World’s Poult. Sci. J. 42 (1), 82–97. Jones, R.B., Mills, A.D., 1983. Estimation of fear in two lines of the domestic chick: correlation between various methods. Behav. Process. 8, 243–253. Jones, R.B., Beuving, G., Blokhuis, H.J., 1988. Tonic immobility and heterophil/lymphocyte responses of the domestic fowl to corticosterone infusion. Physiol. Behav. 42 (3), 249–253. Jones, R.B., Mills, A.D., Faure, J.M., 1991. Genetic and experiential manipulation of fear-related behavior in Japanese quail chicks (Coturnix coturnix japonica). J. Comp. Psychol. 105 (1), 15–24. Jones, R.B., Satterlee, D.G., Ryder, F.H., 1992. Fear and distress in Japanese quail chicks of two lines genetically selected for low or high adrenocortical response to immobilization stress. Horm. Behav. 26 (3), 385–393. Jones, R.B., Mills, A.D., Faure, J.M., Williams, J.B., 1994. Restraint, fear, and distress in Japanese quail genetically selected for long or short tonic immobility reactions. Physiol. Behav. 56 (3), 529–534. Joseph, J., Ramachandran, A.V., 1992. Alterations in carbohydrate metabolism by exogenous dexamethasone and corticosterone in post-hatched White Leghorn chicks. Br. Poult. Sci. 33 (5), 1085– 1093. Kalin, N.H., Shelton, S.E., Rickman, M., Davidson, R.J., 1998. Individual Differences in Freezing and Cortisol in Infant and Mother Rhesus Monkeys. Behav. Neurosci. 112 (1), 251–254. Koolhaas, J.M., Korte, S.M., De Boer, S.F., Van Der Vegt, B.J., Van Reenen, C.G., Hopster, H., De Jong, I.C., Ruis, M.A.W., Blokhuis, H.J., 1999. Coping styles in animals: current status in behavior and stress-physiology. Neurosci. Biobehav. Rev. 23 (7), 925–935. Korte, S.M., Beuving, G., Ruesink, W., Blokhuis, H.J., 1997. Plasma catecholamine and corticosterone levels during manual restraint in
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297 chicks from a high and low feather pecking line of laying Hens. Physiol. Behav. 62 (3), 437–441. Launay, F., 1993. Conse´quences comportementales et physiologiques de se´lection pour l’e´motivite´ et l’attraction sociale chez la caille japonaise (Coturnix japonica). Ph.D. Thesis, Universite´ de Rennes 1 (FRA). Mignon-Grasteau, S., Roussot, O., Delaby, C., Faure, J.M., Mills, A., Leterrier, C., Gue´mene´, D., Constantin, P., Mills, M., Lepape, G., Beaumont, C., 2003. Factorial correspondence analysis of fear-related behaviour traits in Japanese quail. Behav. Process. 61 (1–2), 69–75. Mills, A.D., Faure, J.M., 1986. The estimation of fear in domestic quail: correlations between various methods and measures. Biol. Behav. 11, 235–243. Mills, A.D., Faure, J.M., 1991. Divergent selection for duration of tonic immobility and social reinstatement behavior in Japanese quail (Coturnix coturnix japonica) chicks. J. Comp. Psychol. 105 (1), 25–38. Rees, A., Harvey, S., Phillips, J.G., 1983. Habituation of the corticosterone response of ducks (Anas platyrhynchos) to daily treadmill exercise. Gen. Comp. Endocrinol. 49 (3), 485–489. Rees, A., Harvey, S., Phillips, J.G., 1985. Transitory corticosterone responses of ducks (Anas platyrhynchos) to exercise. Gen. Comp. Endocrinol. 59 (1), 100–104. Re´mignon, H., Mills, A.D., Gue´mene´, D., Desrosiers, V., Garreau-Mills, M., Marche, M., Marche, G., 1998. Meat quality traits and muscle characteristics in high or low fear lines of Japanese quails (Coturnix japonica) subjected to acute stress. Br. Poult. Sci. 39 (3), 372–378.
297
Ruis, M.A., te Brake, J.H., Engel, B., Buist, W.G., Blokhuis, H.J., Koolhaas, J.M., 2001. Adaptation to social isolation. Acute and longterm stress responses of growing gilts with different coping characteristics. Physiol. Behav. 73 (4), 541–551. Satterlee, D.G., Johnson, W.A., 1988. Selection of Japanese quail for contrasting blood corticosterone response to immobilization. Poult. Sci. 67 (1), 25–32. Satterlee, D.G., Jones, R.B., Ryder, F.H., 1993. Short-latency stressor effects on tonic immobility fear reactions of Japanese quail divergently selected for adrenocortical responsiveness to immobilization. Poult. Sci. 72 (6), 1132–1136. Schulkin, J., McEwen, B.S., Gold, P.W., 1994. Allostasis, amygdala, and anticipatory angst. Neurosci. Biobehav. Rev. 18 (3), 385–396. Schulkin, J., Gold, P.W., McEwen, B.S., 1998. Induction of corticotropinreleasing-hormone gene expression by glucocorticoids: implication for understanding the states of fear and anxiety and allostatic load. Psychoneuroendocrinology 23 (3), 219–243. Seggie, J.A., Brown, G.M., 1975. Stress response patterns of plasma corticosterone, prolactin, and growth hormone in the rat, following handling or exposure to novel environment. Can. J. Physiol. Pharmacol. 53 (4), 629–637. Siegel, H.S., 1971. Adrenals, stress and the environment. World’s Poult. Sci. J. 27, 327–349. Siegel, H.S., 1980. Physiological stress in birds. Bioscience 30, 529–534.
General and Comparative Endocrinology 155 (2008) 288–297 www.elsevier.com/locate/ygcen
Intensity and duration of corticosterone response to stressful situations in Japanese quail divergently selected for tonic immobility D. Hazard a, M. Couty a, S. Richard b, D. Gue´mene´ a
a,*
UR83-Unite´ de Recherches Avicoles, Institut National de la Recherche Agronomique, Centre de Tours-Nouzilly, 37380 Nouzilly, France b UMR85 INRA-CNRS-Universite´ de Tours-Haras Nationaux, Physiologie de la Reproduction et des Comportements, Centre INRA de Tours-Nouzilly, 37380 Nouzilly, France Received 6 December 2006; revised 26 April 2007; accepted 7 May 2007 Available online 13 May 2007
Abstract Two genotypes of Japanese quail have been divergently selected since the 1980s for long (LTI) or short (STI) duration of tonic immobility [Mills, A.D., Faure, J.M., 1991. Divergent selection for duration of tonic immobility and social reinstatement behavior in Japanese quail (Coturnix coturnix japonica) chicks. J. Comp. Psychol. 105(1), 25–38.], an unlearnt catatonic state characteristic of a behavioral fear response ([Jones, R.B., 1986. The tonic immobility reaction of the domestic fowl: a review. World’s Poult. Sci. J. 42(1), 82–97.]; [Mills, A.D., Faure, J.M., 1991. Divergent selection for duration of tonic immobility and social reinstatement behavior in Japanese quail (Coturnix coturnix japonica) chicks. J. Comp. Psychol. 105(1), 25–38.]). The results of several behavioral tests conducted in LTI and STI quail have led to the conclusion that LTI quail are more fearful than STI quail [Faure, J.M., Mills, A.D., 1998. Improving the adaptability of animals by selection. In: Grandin, T. (Eds.), Genetics and the behavior of domestic animals. Academic Press, San Diego, pp. 235–264.]). However, few studies to date have focused on the Hypothalamic-Pituitary-Adrenal (HPA) axis response to stressful situations in LTI and STI quail, although the HPA axis is involved in fear responses [Siegel, H.S., 1971. Adrenals, Stress and the Environment. World’s Poult. Sci. J. 27, 327–349.]. The corticosterone (CORT) response to various putatively stressful situations was therefore assessed in LTI and STI genotypes of quail in order to investigate their HPA axis reactivity to stress. Repeated induction of TI or 1 min manual restraint induced significant and comparable increases in CORT levels in both genotypes as a TI response. On the other hand, higher CORT responses were found in STI than in LTI quail when the manual restraint period lasted for 2 min or after restraint in a crush cage. Maximum CORT responses and genotype differences were maintained throughout the latter test even when it lasted for 120 min. Investigation of the CORT response to a single TI episode showed that CORT levels at the end of TI were negatively correlated with TI duration. Other experimental contexts consisting of isolation in a familiar or novel environment or the presentation of a novel object induced slight but significant and comparable increases in CORT response in both genotypes, whereas change of cagemates did not. In conclusion, the present findings indicate that differences in HPA axis response are observed between LTI and STI genotypes when quail are submitted to intense stressors, resulting in a high and prolonged CORT response. By contrast, plasma corticosterone concentrations do not differ between STI and LTI quail in response to stressful situations of lower intensity, which evoke responses limited in amplitude and duration. Genetic selection for divergent duration of TI has thus affected the HPA axis response to stress, and LTI and STI quail constitute an interesting model to investigate genetic variability of HPA axis activity in birds. More specifically, these genotypes of quail could be used to investigate the occurrence of functional differences at different levels of the HPA axis in order to explain the present findings. 2007 Elsevier Inc. All rights reserved. Keywords: HPA axis; Corticosterone; Stress; Quail; Tonic immobility
1. Introduction *
Corresponding author. Fax: +33 2 47 42 77 78. E-mail address: [email protected] (D. Gue´mene´).
0016-6480/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.ygcen.2007.05.009
Several genetic selection programs have been set up for different fear–related behavioral traits in mammals (rats,
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
Bignami, 1965; mice, DeFries et al., 1978) and in birds (chickens (Faure and Folmer, 1975); pheasants (Boyer et al., 1973); quail (Mills and Faure, 1991). The results of these selection programs indicate that animals from divergent genotypes have different behavioral and physiological responses to fearful situations. Moreover, studies in rodents have reported coherent sets of behavioral and physiological responses which are consistent over time and across situations, better known as coping styles (Koolhaas et al., 1999). Thus, rats may be classified as proactive or reactive, depending on the way they respond to stressors. Similar classifications have been established in other species, notably in pigs (Hessing et al., 1994), hens (Korte et al., 1997) and great tits (Groothuis and Carere, 2005). Some animals react in an active fashion (fight or flight) while others are more likely to show inhibited behavior during the acute fear response (Koolhaas et al., 1999). The latter cease all ongoing activities and immediately freeze when a source of danger is detected, thereby reducing the likelihood of detection and attack from a predator. In such situations, they may show a tonic immobility (TI) response if caught (feigning death) (Jones, 1986; Gentle et al., 1989). TI has therefore been described as an unlearnt catatonic state which is thought to be the final stage in a chain of anti-predator behavior patterns (Jones, 1986; Mills and Faure, 1991). Two genotypes of Japanese quail have been divergently selected since the 1980s for long (LTI) or short (STI) duration of tonic immobility (Mills and Faure, 1991). Several behavioral tests conducted in LTI and STI quail demonstrated that quail which show a long duration of TI response are more fearful than quail which show a short duration of TI response, and that duration of the TI reaction is a reliable indicator of underlying fearfulness (Jones and Mills, 1983; Mills and Faure, 1986; Jones et al., 1991; Faure and Mills, 1998). LTI and STI quail have therefore been widely used as a biological model to investigate fearrelated behavioral responses in birds, but these quail also constitute an appropriate model to study the putative relationship between fearfulness and Hypothalamic–Pituitary– Adrenal (HPA) axis reactivity to stress. Indeed, elicitation of instant and adequate fear responses requires activation of the HPA axis and in the final stage release of corticosteroids (corticosterone in birds) from the adrenal glands (Siegel, 1971, 1980; Harvey and Hall, 1990), which are required to re-establish homeostasis via feedback mechanisms and to facilitate behavioral adaptation. Moreover, corticosteroids have been shown to influence the consolidation and potentiation of fear as well as the facilitation or extinction of avoidance in mammals (Schulkin et al., 1994, 1998). A line of Japanese quail divergently selected for their high corticosterone (CORT) response to restraint (Satterlee and Johnson, 1988) has been shown to exhibit the longest duration of TI (Jones et al., 1992; Satterlee et al., 1993). On the other hand, STI quail have been reported to exhibit higher CORT levels than LTI quail in response to restraint
289
stress (Jones et al., 1994; Re´mignon et al., 1998; Hazard et al., 2005a). However, only partial and rudimentary investigations into the CORT response to potentially stressful situations have been performed to date in LTI and STI quail (Launay, 1993; Jones et al., 1994; Faure et al., 1996; Re´mignon et al., 1998; Hazard et al., 2005a) and the results have been inconclusive. This is partly due to the lack of consistency in the experimental designs used (generation number, age, sex, rearing conditions, sampling interval. . .). This can be a very important issue, since CORT levels are very sensitive to any variation in physiological and environmental factors (Hazard et al., 2005a,b). It is therefore impossible from previous studies to reach firm conclusions about differences in the CORT response to stress between LTI and STI genotypes. The aim of the present study was to asses CORT responses in LTI and STI genotypes of quail submitted to a number of putatively stressful situations under careful and standardized experimental conditions in order to characterize HPA axis reactivity to stress in these genotypes and to explore further the possible effects of genetic selection for a fear-related behavioral response on HPA axis reactivity to stress. Duration of TI has been reported to be significantly prolonged when circulating CORT concentrations are increased by administration of CORT in domestic fowl (Jones et al., 1988). Moreover, greater TI responses have been reported to be associated with high endogenous levels of CORT in mammals (Carli et al., 1979; Kalin et al., 1998). Therefore, we first investigated the putative relationship between TI responses (i.e. the selection parameter) and CORT responses in LTI and STI genotypes. Then, CORT response patterns to various physical and social stressors were investigated in LTI and STI quail. The stressors consisted of restraint in a crush cage, presentation of a novel object, isolation in a familiar or a novel environment and change of cagemates. These situations were chosen because they have previously been shown to induce CORT release in birds and mammals (Seggie and Brown, 1975; Launay, 1993; Buijs et al., 1997; Ruis et al., 2001; Carobrez et al., 2002; Hazard et al., 2005a). Moreover, it was anticipated that studying the responses to isolation and placing in a novel environment might provide greater understanding of responses to restraint, since both of these stressors are components of the restraint test. 2. Materials and methods 2.1. Animals and rearing conditions Six-week-old Japanese quail (Coturnix japonica) from the 36th generation of two genotypes divergently selected for short (STI) or long (LTI) duration of tonic immobility (TI) were used in the study (Mills and Faure, 1991). Eight successive hatches were necessary to provide the quail used in the present study and the quail used for each specific test were from a single hatch. Quail were identified by wing-banding on the day of hatching. Quail were exposed to continuous light until 3 weeks of age, then to a 16 h light/8 h dark (16L/8D) (light on 6.00 am) rhythm. The quail from the different genotypes and experimental groups were reared in different collective battery cages. The caretaker checked the quail daily in the
290
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
morning (from 08.30 h) and refilled the feeders whenever necessary (at least twice a week). No care was provided on the day of the experiment. Food and water were provided ad libitum.
2.2. Experimental procedures Groups of 4–6 quail of the same genotype and sex were randomly constituted and placed in wire rearing battery cages one week before the experiments. At the age of 6 weeks, quail were submitted to one of 9 different stressors, depending upon the test being conducted. The stressors consisted of a tonic immobility test (Experiment 1a, 1b), repeated induction of TI (Experiment 1b), manual restraint (Experiment 1b), restraint in a crush cage for a short or long duration (Experiment 2), presentation of a novel object (Experiment 3a), isolation in a familiar (Experiment 3b) or novel environment (Experiment 3c), or change of cagemates (Experiment 3d). Quail from both genotypes and sexes were included in each experiment, except in Experiment 1a for which only LTI quail were used since the duration of TI in STI quail is too short to study the kinetics of the CORT response in this genotype. Following each treatment, blood was collected from each quail directly after decapitation and temporarily stored on ice in a tube containing EDTA (2 mg/ml blood). Following centrifugation at 2000g for 15 min at 4 C, plasma was separated and stored at 20C until measurement of CORT levels using a specific radio immunoassay (Etches, 1976). An additional group of non-treated quail, which were only captured and transferred to the test room for blood sampling, was included as a control in each experiment in order to measure basal corticosterone concentrations. This brief capture and transfer did not affect CORT levels (data not shown). Sacrifice by decapitation was used in this study since we have previously shown this sampling procedure affects basal CORT levels least compared to venipuncture, cardiac or jugular puncture (Hazard et al., 2004). Using this procedure, we were able to observe even very small amplitude responses. Serial bleedings were not considered since several studies have reported that serial bleedings increase CORT levels (Harvey et al., 1980; Johnson and van Tienhoven, 1981; Beuving and Vonder, 1986). Moreover, in order to avoid an unexpected effect on CORT levels of repeatedly taking a quail from the cage, all quail from one cage were bled at the same time.
2.3. Specific treatments 2.3.1. Experiment 1a: Tonic immobility Quail of the LTI genotype were tested individually for TI response after transfer to a testing room adjacent to the rearing room (ntotal = 91; ntotal refers to the number of quail used in each experiment and in the statistical analysis: 91 in the present experiment). Each quail was placed on its back in a U-shaped cradle, hand-restrained for 10 s to induce tonic immobility and then released; this procedure mimics that used for the selection program (Mills and Faure, 1991). When induction was unsuccessful (i.e. if TI lasted less than 10 s), the experimenter immediately reattempted TI induction (maximum of 5 attempts). When TI lasted at least 10 s, TI duration (i.e., the time until the quail righted itself) was recorded. In this first experiment, the test was terminated if the quail was still in TI after 20 min. Quail were bled either immediately after TI ended naturally or immediately after the experimenter terminated the TI. 2.3.2. Experiment 1b: Tonic immobility, repeated induction, manual restraint Tonic immobility. Quail from both genotypes were included in this second experiment and TI was induced as described above, except that after taking into account the results obtained in Experiment 1a, TI was limited to 5 min duration. Quail were bled immediately after TI ended naturally or following termination (i.e. after 5 min) (ntotal = 39). Quail failing to exhibit TI after 5 inductions were nevertheless bled and included in the group of quail used to measure CORT response after repeated induction (i.e. 5·). Moreover, in order to characterize CORT response to TI more fully and to investigate the effects of the TI induction procedure, CORT responses were measured following several TI induction procedures and manual restraint.
Repeated inductions. Quail were subjected to the TI induction procedure as described above. However, instead of being released at the end of the 10 s induction period, quail were manually righted and the procedure was repeated 5 times (ntotal = 28). Manual restraint. Under this treatment, the TI procedure was also mimicked but each quail was manually maintained on its back for 1 or 2 min (ntotal = 44). 2.3.3. Experiment 2: Restraint test in crush cage This experiment consisted of two independent trials. For each trial, quail were transferred to the test room and placed individually in a ‘‘crush-cage’’ in the form of a wooden box (15 cm length · 5 cm width x 10 cm height) closed at the top by a netting cover. Quail have only very restricted movement in such a crush cage. In the first trial, quail were restrained for 10 min and then either immediately bled or placed back in their home cages for 10, 30, 50 or 110 min and then bled (ntotal = 100). Quail were thus sacrificed 10, 20, 40, 60 or 120 min after the beginning of the restraint period. In the second trial, quail were restrained in the crush cage for 10, 20, 40, 60, or 120 min and then bled immediately (ntotal = 186). 2.3.4. Experiment 3a: Presentation of a novel object Each experimental battery cage initially contained a group of 4 quail. First, a control quail was captured and bled immediately after introducing a novel object into the battery cage. The novel object was in the form of a red balloon of 12 cm of diameter. The 3 remaining quail were then bled concomitantly 5, 10 or 60 min after introduction of the balloon (ntotal = 168). 2.3.5. Experiment 3b: Isolation in a familiar environment After capture in its home battery cage, each quail was placed alone in another battery cage of similar size and shape (35.5 cm length · 22 cm width · 16.5 cm height) in the same room for 5, 10 or 60 min (ntotal = 124). Visual communication between the isolated quail and the other quail in the room was prevented, but acoustic and olfactory communication was still possible. 2.3.6. Experiment 3c: Isolation in a novel environment The procedure used was comparable to that described above except that the cages where the quail were placed for the test consisted of large empty wooden boxes (65 cm length · 41 cm width · 28 cm height) closed at the top by a netting cover and located in an adjacent room (ntotal = 128). The use of a cage differing totally in structure and size from the home cages increased the probability of measuring CORT responses specific to the novel environment. Moreover, a larger cage than the home cage was used to test the effects of a novel environment since a cage smaller than the home cage had already been used for the restraint test and it was intended to avoid misinterpretation through observing an unexpected effect similar to the restraint. 2.3.7. Experiment 3d: Change of cagemates The procedure consisted of exchanging two quail out of 4 from one battery cage with two others of the same sex but from the opposite genotype. Quail of the LTI genotype had never previously met quail of the STI genotype. All quail from each of the newly formed groups were sacrificed at the same time, either 5, 10 or 60 min after transfer (ntotal = 124). In order to analyze the respective effects of social stress and/or transfer, data for quail which were submitted to both transfer and exchange or only exchange were first analyzed separately. Pairs of quail not transferred and not mixed with new counterparts constituted control groups.
2.4. Statistical analysis For each experiment, CORT values were subjected to multifactorial ANOVA to assess the effects of the different specific experimental factors and their interactions using the Statview V program (Abacus Concept Inc. Berkeley, USA). The factors tested were treatment/sampling time, genotype (except for Experiment 1a) and sex, and their respective interactions. Whenever ANOVA reached significance (P < 0.05), post-hoc tests were performed using the Fisher test (PLSD). Correlation analyses
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297 between CORT concentrations and duration of TI were performed. CORT values are expressed as means ± standard error and the level of significance was P < 0.05 unless otherwise stated.
3. Results 3.1. Experiment 1a: Tonic immobility TI was induced in 76 out of 83 quail after only 1 attempt, and a maximum of 2 inductions was necessary to induce TI in the other quail. Minimum TI duration was 31 s, whereas it was terminated intentionally by the experimenter in the 5 quail remaining 20 min in TI. Durations of TI were regularly distributed throughout the 20 min period. Correlation analyses calculated between CORT levels and duration of TI revealed that there was a significant negative correlation (r = 0.45, P < 0.0001) between these 2 variables (Fig. 1A). The data were assigned to 5 classes corresponding to successive periods of 5 min duration depending on TI duration: control, ]0–5], ]5–10], ]10–15], ]15–20]. ANOVA of CORT data with class and sex as independent variables showed that there was a significant effect of TI duration and no significant sex effect. Indeed, quail for which TI lasted between 31 s and 5 min exhibited significantly higher CORT levels (5.0 ± 0.4 ng/ml) than basal CORT levels (1.5 ± 0.3 ng/ml) (Fig. 1B). CORT levels in quail spending more than 5 min in TI were intermediate, mean values ranging from 3.0 ± 0.3 ng/ml to 2.3 ± 0.6 ng/ml for the periods ]5–10] and ]15–20], respectively, but the difference from basal CORT values no longer reached significance.
291
sexes (Fig. 2A and B). Genotype and time factors, and their interaction, significantly affected CORT responses whereas the sex factor did not (Fig. 2A). Placing in the crush-cage for 10 min induced a significant increase in CORT levels in both genotypes. The differences in CORT concentrations between undisturbed and restrained birds appeared to be higher in quail of the STI genotype, since 6- and 3-fold increases were measured for STI and LTI quail, respectively. When the quail were placed back in their home battery cages following the 10 min restraint period, CORT levels progressively decreased to reach levels that did not differ from basal levels within 10 and 30 min for LTI and STI quail, respectively. When the restraint period was prolonged, CORT levels remained stable at maximum levels for both genotypes (Fig. 2B) and were significantly lower in LTI than in STI quail throughout the restraint period (i.e. 10–120 min). 3.4. Experiment 3a: Presentation of a novel object Introduction of a novel object into the home cage induced a significant increase in CORT levels in female LTI quail and in both sexes of the STI quail, but not in LTI males. A significant sex effect was observed but there was no genotype effect (Fig. 3A). Maximum CORT levels were measured 5 min after presentation of the balloon and the highest CORT levels reached a maximum concentration of approximately 4.0 ng/ml. CORT concentrations did not differ from basal values 10 min after introduction of the balloon. 3.5. Experiment 3b: Isolation in a familiar environment
3.2. Experiment 1b: Tonic immobility, repeated induction and manual restraint Duration of TI ranged between 10 and 134 s in STI quail while the minimum duration in LTI quail was 25 s, and 50% of LTI quail were still in TI after 5 min (Fig. 1C). ANOVA of CORT levels in response to TI, repeated induction or manual restraint revealed no significant sex effect (Fig. 1D). TI was associated with significant and comparable increases in CORT levels in the two genotypes, and the average CORT level calculated for both genotypes was 6.8 ± 0.6 ng/ml (Fig. 1D). Five repetitions of the 10 s induction procedure induced similar increases in CORT levels (5.1 ± 0.6 ng/ml) in both genotypes (Fig. 1D). Similarly, 1 min manual restraint induced significant and comparable increases in CORT levels in both genotypes (6.0 ± 0.7 ng/ml) (Fig. 1D). By contrast, 2 min manual restraint induced a significantly greater increase in CORT levels in STI quail (11.0 ± 2.0 ng/ml) than in LTI quail (6.3 ± 0.7 ng/ml) (Fig. 1D).
Isolation of quail in a familiar environment induced significant increases in CORT levels (P = 0.05) (Fig. 3B). A significant sex effect (P = 0.0001) upon CORT levels was found whereas the genotype effect was not significant (P = 0.9). Maximum CORT levels of approximately 3.0 ng/ml were measured after 5 min of isolation in STI males, whereas a delay of 10 min was observed for the other 3 experimental groups. 3.6. Experiment 3c: Isolation in a novel environment Social isolation in a novel environment induced a significant increase in CORT levels (P < 0.001) whereas neither the genotype (P = 0.78) nor the sex (P = 0.2) effects were significant (Fig. 3C). Maximum CORT levels which remained lower than 4.0 g/ml were measured at 5 min after beginning isolation. After 10 min of isolation in the novel environment, CORT concentrations were comparable to basal values.
3.3. Experiment 2: Restraint test in a crush cage
3.7. Experiment 3d: Change of cagemates
The basal CORT levels measured for control quail were low and did not differ significantly between genotypes or
Neither the effect of exchange (P = 0.15), either for the quail that were submitted to both transfer and exchange
292
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
A
CORT
15 (ng/ml) r = - 0.45 P < 0.0001
10
CORT (ng/ml)
B
ANOVA
10
Treatment P=0.0002 Sex P=0.85
b 5
5
a
a
a
a
C
0
TI (sec)
5
TI 20 (min)
10
15
D
CORT
0 C ]0 -5 ]5 ] -1 ]10 0] -1 ]1 5] 520 ]
0
*
15 (ng/ml)
300
200
10
c
b b b
ANOVA Genotype Sex Treatment Gen. x Sex Gen. x Treat.
P=0.3 P=0.24 P=0.001 P=0.1 P=0.06
b
b 100
TI (min)
b
b
5
LTI a
a
STI
C Ix 5 T M I R M 1 R 2
LT I ST I
C Ix 5 M TI R M 1 R 2
0
0
Fig. 1. Changes in corticosterone concentrations over time (CORT, ng/ml plasma) in LTI quail following tonic immobility (A and B) (7 6 n 6 39) (ntotal = 91) Duration of tonic immobility in LTI and STI quail (C) (nLTI = 13, nSTI = 26). Corticosterone concentrations (ng/ml plasma) in LTI and STI quail following 5 induction procedures (I ·5) with immobilization over 10 s for each, tonic immobility period (TI), 1 min (MR 1) and 2 min (MR 2) of manual restraint (D) (10 6 n 626). Control samples (C) were collected from undisturbed quail. (means ± SE) a, b & c = means for a given genotype with common superscripts were not significantly different from each another (P > 0.05) *significantly different between genotypes (P < 0.05).
or only to exchange, nor the effect of genotype (P = 0.15) significantly affected CORT levels (Fig. 3D). The significant sex effect (P < .0001) observed resulted from higher basal CORT levels in male than in female quail. 4. Discussion First HPA axis activity was characterized during the course of the TI response in LTI and STI genotypes, TI corresponding to the selection parameter used (Mills and Faure, 1991). Interestingly, changes in CORT levels after the TI test did not differ significantly between the two genotypes, although (as expected) TI behavior lasted longer in LTI than in STI quail. Furthermore, maximum CORT levels were measured in LTI quail showing the shortest duration of TI response, while CORT responses in LTI quail which remained naturally for less than 5 min in TI and in LTI quail for which the TI response was terminated 5 min after onset were comparable. Although we cannot exclude the possibility that this initial increase in CORT levels at the onset of the TI response played a critical role
in the duration of this behavioral response, this is not evident from the present findings. Indeed, in such a case CORT levels at the onset of the TI response in quail expressing longer TI responses should be higher than in those expressing shorter duration responses. However, while TI responses lasted longer in LTI than in STI quail, CORT responses at the beginning of the TI test (i.e. following TI induction) were similar in both genotypes. Moreover, a negative correlation was found between CORT levels and duration of the TI response in LTI quail. Thus, although CORT concentrations in quail spending more than 5 min in TI were intermediate between basal and maximum concentrations, our results suggest that maintenance in a TI state does not require maintenance of high CORT concentrations, and the TI state, which is a behavioral fear response, does not maintain high CORT levels. Moreover, return of CORT levels to basal values is not a prerequisite for the quail to right itself. Indeed, quail showing the shortest TI responses righted themselves whereas they had higher than basal CORT levels. The findings of the present study thus suggest that there is no relationship between
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
A
293
CORT (ng/ml)
30 ANOVA
c
25
Genotype P<0.004 Sex P=0.61 Time P<0.0001 Time x Gen. P=0.0004
20 b
15
a
a
LTI ab
ab
10 5
b
a
a
STI ab
a a
0 0 10 20 40 60 120
B
0 10 20 40 60 120
Time (min)
CORT (ng/ml)
30
ANOVA
25
b
b b
20 15
b ab
10 5 0
b
b b
Genotype Sex Time Interactions
P<0.0001 P=0.12 P=0.001 P>0.4
b
ab
a
a
0 10 20 40 60 120
Time 0 10 20 40 60 120 (min)
Fig. 2. Changes in corticosterone concentrations over time (CORT, ng/ml plasma) in LTI and STI quail following restraint period in a crush cage of fixed duration of 10 min. (A) (5 6 n 6 11) or of variable durations lasting 10, 20, 40, 60 or 120 min (B) (12 6 n 6 20) (means ± SE) control samples (time 0) were collected from undisturbed quail. The restraint period is shown by horizontal shaded bars. a, b & c = means for a given genotype with common superscripts were not significantly different from each another (P > 0.05).
duration of TI response and CORT levels. We can also hypothesize from the present findings that the initial increase in CORT associated with the TI response results from the process of manual restraint performed during the TI induction procedure. This hypothesis is further supported by the results that we obtained after manual restraint mimicking TI induction. Indeed, similar CORT responses were measured in both genotypes of quail following several TI induction procedures, manual restraint on the back for 1 min and following a short TI response. The hypothesis that the increase in CORT concentrations at the beginning of TI is due to the manual restraint during the induction procedure is also supported by similar increases in CORT levels reported in laying hens in response to manual restraint (Korte et al., 1997). Our present findings contrast somewhat with previously reported results indicating positive relationships between fearinduced behavior and circulating levels of glucocorticoids in different avian and mammalian species (Carli et al.,
1979; Kalin et al., 1998). Greater freezing and TI responses in rabbits (Carli et al., 1979) and rhesus monkeys (Kalin et al., 1998) were associated with high endogenous levels of CORT. By contrast, differences in TI duration between LTI and STI genotypes in the present study were not associated with different basal CORT levels. Moreover, in quail selected for divergent CORT responses to a brief period of immobilization (Satterlee and Johnson, 1988), quail with higher HPA axis reactivity also exhibited longer TI responses (Jones et al., 1992), whereas the opposite was found in LTI and STI genotypes (Hazard et al., 2005a). In addition, the tonic immobility reaction has been shown to be prolonged following CORT infusion in chicken (Jones et al., 1988), and exogenous administration of CORT potentiated fear-induced freezing behavior in rats (Coordimas et al., 1994). Although it would be important to investigate the effects of exogenous administration of CORT on TI duration in LTI and STI quail to firmly conclude on the relationship between CORT levels and TI
294
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
A 5
CORT (ng/ml)
LTI Fem.
Male
STI
b
4
b
3
b
b
a
a a
ANOVA
ab ab
2 1
Male
Fem.
a a
a
a
a a
a
Genotype P=0.5 Sex P=0.0001 Time P<0.0001 Time x Gen. P=0.03
0
B
5 4 ANOVA
3
Genotype Sex Time Interactions
2 1
P=0.9 P=0.0001 P=0.05 NS
0
C
5 4 ANOVA
3
Genotype Sex Time Interactions
2 1
P=0.78 P=0.2 P<0.001 NS
0
D
5 4 ANOVA
3
Genotype Sex Time Interaction
2 1 0 0 5 10 60 0 5 10 60
0 5 10 60
0 5 10 60
P=0.15 P<0.0001 P=0.15 NS
Time (min)
Fig. 3. Changes in corticosterone concentrations over time (CORT, ng/ml plasma) in LTI and STI quail following the presentation of a novel object (balloon) (A), isolation in a familiar environment (B) or a novel environment (C) and change of cagemates (D) for durations of 5, 10 or 60 min (6 6 n 6 12). (Means ± SE) Control samples (time 0) were collected from undisturbed quail. a & b = means for a given genotype with common superscripts were not significantly different from each another (P > 0.05).
response, it is likely that the duration of TI does not depend on CORT levels in these genotypes of quail. The present study provides new information about the time-course of the CORT response, especially in the case of long lasting acute stress, since only the short term effects of restraint lasting 5 (Jones et al., 1994), 10 (Hazard et al., 2005a) or 15 min (Re´mignon et al., 1998) have previously been investigated. The present findings showed that differences in CORT responses between the two genotypes appeared very quickly, since significantly higher CORT levels were observed in the STI than in LTI genotype after 2 min of manual restraint. The CORT response reached its
maximum before 10 min in both genotypes, since CORT levels did not increase further after this interval in the restraint situation. This result is consistent with CORT responses reported in high and low feather pecking genotypes of laying hens, in which the difference in the increases of CORT levels between the two genotypes was maximum between 1 and 4 min (Korte et al., 1997). Interestingly, increases in CORT levels were observed in response to all the experimental stressors tested, with the exception of social stress induced by cagemate exchange. However, LTI and STI quail did not differ in their CORT responses to social isolation or presentation of a novel
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
object, and the increases observed were of much lower amplitude than those induced in the restraint test. In addition, CORT levels induced in response to isolation or the presentation of a novel object did not remain stable throughout testing but returned quickly to basal values, while CORT concentrations induced in response to restraint in the crush cage remained at maximum levels in both genotypes throughout the 120 min period. It appears from the present study that differences in HPA axis responsiveness between the 2 genotypes only prevailed following manual restraint and restraint in a crush cage. The latter treatment appeared to be the most stressful situation according to the intensity and duration of the CORT responses and thus required high CORT synthesis and release from the adrenal glands. LTI quail have been reported to be more fearful than STI quail (Jones et al., 1991). Therefore, since the HPA axis is involved in the response to fearful stimuli and corticosterone is reported to be a valuable indicator of the stress response (Siegel, 1980; Harvey and Hall, 1990), we would have expected LTI quail to have higher CORT responses to fearful stimuli than STI quail. Indeed, using a reverse strategy consisting of selecting quail on physiological responses (i.e., on high or low CORT responses to a restraint), Satterlee and Johnson (1988) showed that the high stress line exhibited the highest CORT response and the longest TI response (Jones et al., 1992). However, the differences observed indicate that the opposite situation occurred in the present study, as in a previous study (Hazard et al., 2005a).Whatever the underlying mechanisms, the apparently conflicting findings (i.e., higher CORT responses to restraint associated with shorter TI responses) in our genotypes and the reverse situation in others could be viewed as evidence that TI reactions and adrenocortical responses are not mutually coselected in LTI and STI quail. This hypothesis is further supported by the fact that no relationship between CORT concentration after restraint and TI was found in a second generation cross (F2) between the LTI and STI genotypes (Mignon-Grasteau et al., 2003). We could hypothesize that differences in CORT response to restraint between genotypes involved differences in CORT synthesis capacity. However, differences in CORT synthesis capacity do not seem to be involved in the genotype difference, since we have previously reported comparable CORT responses to 1–24 ACTH injection in LTI and STI quail (Hazard et al., 2005a). This hypothesis remains to be investigated more fully before reaching a firm conclusion regarding adrenal involvement in the genotype difference in CORT responses to restraint. Moreover, differences in functioning of the pituitary and/or hypothalamus might also be involved in the differences in CORT response to stress between LTI and STI quail and thus remain to be investigated. Measurement of circulating ACTH could provide information about pituitary activity in STI and LTI quail. Moreover, greater understanding of the differential HPA responsiveness in STI and LTI quail would also benefit from investigation into the cortico-
295
steroid-binding globulin levels in these two genotypes of quail. Indeed, corticosteroid-binding globulins influence the bioavailability of CORT to target cells, and similar plasma concentrations of CORT in STI and LTI quail may mask different levels of free CORT (Deviche et al., 2001). Nevertheless, the divergent results between CORT and TI responses are not necessarily inconsistent, since it is possible that CORT responses are not related to the degree of fearfulness but are somehow a reliable indicator of the physiological response developed by birds to cope with a specific stressful situation. Indeed, CORT produced in response to stress can be used to provide energy to support the physical response (Rees et al., 1983; Joseph and Ramachandran, 1992). For instance, an increase in CORT levels associated with physical activity has been reported in ducks (Rees et al., 1985). Thus, it cannot be excluded that the differences in CORT response to restraint might result from differences in behavior being expressed during the test, since STI quail have previously been reported to struggle frequently during 10 min restraint in a crush cage, whereas this behavior was virtually not expressed by quail of the LTI genotype (Jones et al., 1994). However, such a relationship between CORT response and fear behavior is not straightforward. Indeed, proactive animals, which exhibit active behavior in response to various stressful situations, are traditionally characterized by weak HPA axis reactivity (Koolhaas et al., 1999). Finally, we can also hypothesize that differences in CORT response between genotypes may involve different psychobiological processes of perception, resulting in limited HPA axis activation in the LTI genotype and/or hyper-activation of the HPA axis in STI quail. Indeed, the higher HPA axis response to restraint than to all other stressors in STI quail suggests that quail of this genotype may perceive this stress differently whereas this seemed not to be the case in LTI quail since their HPA axis response was fairly similar whatever the stressor. However, this study does not provide sufficient evidence to support this hypothesis and further investigations into HPA axis function at different levels will be necessary to understand the relationships between animals’ perceptions of a stressful situation, their behavioral responses and their endocrine responses. In conclusion, HPA axis responsiveness was comparable in the two genotypes when the CORT response was limited in both intensity and duration, i.e. in response to TI, presentation of a novel object, and isolation in a familiar or novel environment, suggesting that such situations are weakly stressful. In contrast, genotype differences were observed when CORT concentrations reached higher levels in STI quail in response to manual restraint and restraint in the crush cage. In the latter experimental situations, high CORT levels remained throughout when exposure to the stimulus was prolonged. Clearly the selection program for divergent TI responses has not only had specific effects on TI duration, but has also affected the maximum CORT
296
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297
response to restraint stress in an unexpected way. Thus, since increases in CORT levels towards maximum levels and maintenance at such concentrations have negative effects on physiological functions and various production parameters such as growth and meat quality (Chen et al., 1991; Re´mignon et al., 1998), the use of a behavioral parameter such as TI in a selection program should not be implemented before the consequences have been more fully characterized. Acknowledgments We thank Dr. A.D. Mills and Dr. J.M. Faure who set up and managed the selection program that provided the quail used in this study. We also thank D. Raine for a valuable contribution to improving the quality of the English manuscript. The authors also thank the many people who contributed to this study, especially J.-M. Brigant and J.-M. Hervouet for expert technical assistance. D. Hazard was supported by grants from the Institut National de la Recherche Agronomique and the Conseil Re´gional de la Re´gion Centre for completion of a Ph.D. References Beuving, G., Vonder, G.M., 1986. Comparison of the adrenal sensitivity to ACTH of laying hens with immobilization and plasma baseline levels of corticosterone. Gen. Comp. Endocrinol. 62 (3), 353–358. Bignami, G., 1965. Selection for high rates and low rates of avoidance conditioning in the rat. Anim. Behav. 13 (2), 221–227. Boyer, J.P., Melin, J.M., Bourdens, P., 1973. Activity test in young pheasants. Ann. Genet. Sel. Anim. 5, 417–418. Buijs, R.M., Wortel, J., Van Heerikhuize, J.J., Kalsbeek, A., 1997. Novel environment induced inhibition of corticosterone secretion: physiological evidence for a suprachiasmatic nucleus mediated neuronal hypothalamo-adrenal cortex pathway. Brain Res. 758 (1–2), 229–236. Carli, G., Farabollini, F., di Prisco, C.L., 1979. Plasma corticosterone and its relation to susceptibility to animal hypnosis in rabbits. Neurosci. Lett. 11 (3), 271–274. Carobrez, S.G., Gasparotto, O.C., Buwalda, B., Bohus, B., 2002. Longterm consequences of social stress on corticosterone and IL-1[beta] levels in endotoxin-challenged rats. Physiol. Behav. 76 (1), 99–105. Chen, M.T., Lin, S.S., Lin, L.C., 1991. Effect of stresses before slaughter on changes to the physiological, biochemical and physical characteristics of duck muscle. Br. Poult. Sci. 32 (5), 997–1004. Coordimas, K.P., LeDoux, J.E., Gold, P.W., Schulkin, J., 1994. Corticosterone potentiation of learned fear. In: de Kloert, E.R., Azmatia, E.C., Longfield, P.W. (Eds.), Brain Corticosteroid Receptors: Studies of the Mechanism, Function and Neurotoxicity of Corticosteroid Action. New York Academy of Sciences, New York, pp. 392–393. DeFries, J.C., Gervais, M.C., Thomas, E.A., 1978. Response to 30 generations of selection for open-field activity in laboratory mice. Behav. Genet. 8 (1), 3–13. Deviche, P., Breuner, C., Orchinik, M., 2001. Testosterone, corticosterone, and photoperiod interact to regulate plasma levels of binding globulin and free steroid hormone in Dark-eyed Juncos, Junco hyemalis. Gen. Comp. Endocrinol. 122 (1), 67–77. Etches, R.J., 1976. A radioimmunoassay for corticosterone and its application to the measurement of stress in poultry. Steroids 28 (6), 763–773. Faure, J.M., Folmer, J.C., 1975. Etude ge´ne´tique de l’activite´ pre´coce en open-field du jeune poussin. Ann. Genet. Sel. Anim. 7, 123–132.
Faure, J.M., Mills, A.D., 1998. Improving the adaptability of animals by selection. In: Grandin, T. (Ed.), Genetics and the behavior of domestic animals. Academic Press, San Diego, pp. 235–264. Faure, J.M., Mills, A.D., Launay, F., Francois, N., Williams, J., Gue´mene´, D., Re´mignon, H., 1996. Se´lection pour le comportement: l’exemple de la caille. Compte rendu de l’Acade´mie d’Agriculture de France 82, 61–70. Gentle, M.J., Jones, R.B., Woolley, S.C., 1989. Physiological changes during tonic immobility in Gallus gallus var domesticus. Physiol. Behav. 46 (5), 843–847. Groothuis, T.G., Carere, C., 2005. Avian personalities: characterization and epigenesis. Neurosci. Biobehav. Rev. 29 (1), 137–150. Harvey, S., Hall, T.R., 1990. Hormones and stress in birds: activation of the hypothalamo-pituitary-adrenal axis. Prog. Comp. Endocrinol. 342, 453–460. Harvey, S., Merry, B.J., Phillips, J.G., 1980. Influence of stress on the secretion of corticosterone in the duck (Anas platyrhynchos). J. Endocrinol. 87 (1), 161–171. Hazard, D., Faure, J. M., Gue´mene´, D., 2004. Effects of sampling sites, photoperiod, and age on adrenal responsiveness in Japanese quail. Eighth International Symposium on Avian Endocrinology. Scottsdale, Arizona, USA, p 75. Hazard, D., Couty, M., Faure, J.M., Gue´mene´, D., 2005a. Relationship between Hypothalamic-Pituitary-Adrenal Axis Responsiveness and Age, Sexual Maturity Status and Sex in Japanese Quail Selected for Long or Short Duration of Tonic Immobility. Poult. Sci. 84 (12), 1913–1919. Hazard, D., Couty, M., Faure, J.M., Gue´mene´, D., 2005b. Daily and photoperiod variations of Hypothalamic-Pituitary-Adrenal Axis Responsiveness in Japanese Quail Selected for Short or Long Tonic Immobility. Poult. Sci. 84, 1920–1925. Hessing, M.J., Hagelso, A.M., Schouten, W.G., Wiepkema, P.R., van Beek, J.A., 1994. Individual behavioral and physiological strategies in pigs. Physiol. Behav. 55 (1), 39–46. Johnson, A.L., van Tienhoven, A., 1981. Plasma concentrations and corticosterone relative to photoperiod, oviposition, and ovulation in the domestic hen. Gen. Comp. Endocrinol. 43 (1), 10–16. Jones, R.B., 1986. The tonic immobility reaction of the domestic fowl: a review. World’s Poult. Sci. J. 42 (1), 82–97. Jones, R.B., Mills, A.D., 1983. Estimation of fear in two lines of the domestic chick: correlation between various methods. Behav. Process. 8, 243–253. Jones, R.B., Beuving, G., Blokhuis, H.J., 1988. Tonic immobility and heterophil/lymphocyte responses of the domestic fowl to corticosterone infusion. Physiol. Behav. 42 (3), 249–253. Jones, R.B., Mills, A.D., Faure, J.M., 1991. Genetic and experiential manipulation of fear-related behavior in Japanese quail chicks (Coturnix coturnix japonica). J. Comp. Psychol. 105 (1), 15–24. Jones, R.B., Satterlee, D.G., Ryder, F.H., 1992. Fear and distress in Japanese quail chicks of two lines genetically selected for low or high adrenocortical response to immobilization stress. Horm. Behav. 26 (3), 385–393. Jones, R.B., Mills, A.D., Faure, J.M., Williams, J.B., 1994. Restraint, fear, and distress in Japanese quail genetically selected for long or short tonic immobility reactions. Physiol. Behav. 56 (3), 529–534. Joseph, J., Ramachandran, A.V., 1992. Alterations in carbohydrate metabolism by exogenous dexamethasone and corticosterone in post-hatched White Leghorn chicks. Br. Poult. Sci. 33 (5), 1085– 1093. Kalin, N.H., Shelton, S.E., Rickman, M., Davidson, R.J., 1998. Individual Differences in Freezing and Cortisol in Infant and Mother Rhesus Monkeys. Behav. Neurosci. 112 (1), 251–254. Koolhaas, J.M., Korte, S.M., De Boer, S.F., Van Der Vegt, B.J., Van Reenen, C.G., Hopster, H., De Jong, I.C., Ruis, M.A.W., Blokhuis, H.J., 1999. Coping styles in animals: current status in behavior and stress-physiology. Neurosci. Biobehav. Rev. 23 (7), 925–935. Korte, S.M., Beuving, G., Ruesink, W., Blokhuis, H.J., 1997. Plasma catecholamine and corticosterone levels during manual restraint in
D. Hazard et al. / General and Comparative Endocrinology 155 (2008) 288–297 chicks from a high and low feather pecking line of laying Hens. Physiol. Behav. 62 (3), 437–441. Launay, F., 1993. Conse´quences comportementales et physiologiques de se´lection pour l’e´motivite´ et l’attraction sociale chez la caille japonaise (Coturnix japonica). Ph.D. Thesis, Universite´ de Rennes 1 (FRA). Mignon-Grasteau, S., Roussot, O., Delaby, C., Faure, J.M., Mills, A., Leterrier, C., Gue´mene´, D., Constantin, P., Mills, M., Lepape, G., Beaumont, C., 2003. Factorial correspondence analysis of fear-related behaviour traits in Japanese quail. Behav. Process. 61 (1–2), 69–75. Mills, A.D., Faure, J.M., 1986. The estimation of fear in domestic quail: correlations between various methods and measures. Biol. Behav. 11, 235–243. Mills, A.D., Faure, J.M., 1991. Divergent selection for duration of tonic immobility and social reinstatement behavior in Japanese quail (Coturnix coturnix japonica) chicks. J. Comp. Psychol. 105 (1), 25–38. Rees, A., Harvey, S., Phillips, J.G., 1983. Habituation of the corticosterone response of ducks (Anas platyrhynchos) to daily treadmill exercise. Gen. Comp. Endocrinol. 49 (3), 485–489. Rees, A., Harvey, S., Phillips, J.G., 1985. Transitory corticosterone responses of ducks (Anas platyrhynchos) to exercise. Gen. Comp. Endocrinol. 59 (1), 100–104. Re´mignon, H., Mills, A.D., Gue´mene´, D., Desrosiers, V., Garreau-Mills, M., Marche, M., Marche, G., 1998. Meat quality traits and muscle characteristics in high or low fear lines of Japanese quails (Coturnix japonica) subjected to acute stress. Br. Poult. Sci. 39 (3), 372–378.
297
Ruis, M.A., te Brake, J.H., Engel, B., Buist, W.G., Blokhuis, H.J., Koolhaas, J.M., 2001. Adaptation to social isolation. Acute and longterm stress responses of growing gilts with different coping characteristics. Physiol. Behav. 73 (4), 541–551. Satterlee, D.G., Johnson, W.A., 1988. Selection of Japanese quail for contrasting blood corticosterone response to immobilization. Poult. Sci. 67 (1), 25–32. Satterlee, D.G., Jones, R.B., Ryder, F.H., 1993. Short-latency stressor effects on tonic immobility fear reactions of Japanese quail divergently selected for adrenocortical responsiveness to immobilization. Poult. Sci. 72 (6), 1132–1136. Schulkin, J., McEwen, B.S., Gold, P.W., 1994. Allostasis, amygdala, and anticipatory angst. Neurosci. Biobehav. Rev. 18 (3), 385–396. Schulkin, J., Gold, P.W., McEwen, B.S., 1998. Induction of corticotropinreleasing-hormone gene expression by glucocorticoids: implication for understanding the states of fear and anxiety and allostatic load. Psychoneuroendocrinology 23 (3), 219–243. Seggie, J.A., Brown, G.M., 1975. Stress response patterns of plasma corticosterone, prolactin, and growth hormone in the rat, following handling or exposure to novel environment. Can. J. Physiol. Pharmacol. 53 (4), 629–637. Siegel, H.S., 1971. Adrenals, stress and the environment. World’s Poult. Sci. J. 27, 327–349. Siegel, H.S., 1980. Physiological stress in birds. Bioscience 30, 529–534.