|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
COMPARATIVE AND EVOLUTIONARY PHYSIOLOGY
UR83-Unité de Recherches Avicoles, Institut National de la Recherche Agronomique, Centre de Tours-Nouzilly, 37380 Nouzilly, France
Submitted 31 January 2007 ; accepted in final form 4 July 2007
| ABSTRACT |
|---|
|
|
|---|
hypothalamic-pituitary-adrenal axis; corticosterone; stress
Differences in HPA axis reactivity following restraint (27, 32, 44) have previously been reported in two divergent genotypes of Japanese quail selected for long (LTI) or short (STI) duration of tonic immobility (TI). TI is an unlearned catatonic state, and this behavioral response has been shown to be positively correlated with other measurements of fear (31, 40). Several behavioral tests conducted in LTI and STI quail have led to the conclusion that quail that show long duration of the TI response are more fearful than quail that show a short duration of the TI response (17). Higher CORT responses to restraint previously reported in STI quail than in LTI quail were unexpected if CORT levels are considered to be indicators of the stress response. Nevertheless, similar results have been shown in humans, with anxious subjects exhibiting lower neuroendocrine activation during acute stress compared with nonanxious subjects (29). Although these quail genotypes have been widely characterized for their behavioral responses to stress, little is known about their physiological responses to stress. Thus, in view of the possible interaction (described above) between behavioral and physiological responses to stress, it seemed important to identify the origin of the genetic variations in HPA axis reactivity to stress in LTI and STI quail to explain the earlier results.
The influence of genetic factors upon variations in HPA axis activity has been suggested in humans by twin and family studies (28, 34, 37, 39, 42). Moreover, wide variations in HPA axis activity have been described between inbred strains of mice (18, 30) and rats (20, 48, 49) and between different breeds of pigs (10, 15, 25). Bird strains have been selected for divergent adrenal responses to ACTH (14), immobilization (50), cold (3), and social stress (21). Various mechanisms have been suggested to explain the genetic differences in HPA axis activity in these different models. The causes of variations in HPA axis activity may thus take place in the central structures and involve mechanisms at different organizational levels of the HPA axis (hypothalamus, pituitary, adrenal glands) or through different external and internal modulatory mechanisms such as feedback.
Several experimental findings have suggested that the functional differences in HPA axis reported between Lewis and Fisher 344 rat strains (4, 13, 57, 62) may have a central origin, with major differences in hypothalamic CRF neuron function (57), whereas others have reported changes at the pituitary (1) or adrenal cortex levels (20). Moncek et al. (41) recently proposed the hypothesis that the variations in reactivity of the HPA axis in Lewis and Fisher rats probably originates from pituitary and adrenal levels. On the other hand, comparing five inbred rat strains, Gomez et al. (20) showed that the lowest CORT response to immobilization stress observed in Lewis and Wistar-Kyoto rats did not result from lower ACTH secretion, which suggests that the adrenal cortex is hyporeactive in these strains. Similarly, others (10, 11, 63) have demonstrated that an adrenal origin is the cause of high cortisol responses in pigs. Furthermore, selection of Japanese quail for high serum CORT levels in response to immobilization resulted in different steroidogenic properties of adrenocortical cells in vitro (7).
The aim of the study reported here was to identify the origin of the differences in HPA axis responsiveness observed between LTI and STI genotypes. In view of the mounting scientific evidence described above, showing the involvement of adrenal and/or the pituitary levels in the genetic variations in HPA axis reactivity for several models, we hypothesized that LTI and STI genotypes of quail might differ in their adrenal and/or pituitary function. Injection of 1-24 ACTH has been reported to stimulate active CORT release from the adrenal glands in birds (2, 23, 35, 36, 43). Pharmacological challenge using various doses of 1-24 ACTH has provided a valuable approach to investigating adrenal function in LTI and STI genotypes. Various experimental findings have shown that CRF is more potent than AVP in CORT release in several mammals [rats (19, 46, 59); ewes (33)]. On the other hand, contradictory results have been reported concerning the respective effects of these two peptides in birds, with CRF being sometimes more active (8, 61) and sometimes less active (38, 58). To further investigate their respective effects on pituitary activation in LTI and STI genotypes, pharmacological studies were undertaken using CRF or AVT alone or both peptides together.
| MATERIALS AND METHODS |
|---|
|
|
|---|
The corresponding partial regions of mRNA coding for CRF, AVT, and proopiomelanocortin hormone (POMC) were amplified by RT-PCR to characterize the nucleotide acid sequence of the active CRF, AVT, and ACTH peptide regions. Hypothalamic and pituitary total RNA were isolated using the RNeasy Mini purification kit (Qiagen). RNA concentrations were measured by absorption at 260 nm. Total RNA samples (2.5 µg) were reverse transcribed using 200 units Superscript II RNase H– RT (Invitrogen) and oligo(dT) primers (Sigma Aldrich Chimie).
The sequence of the forward CRF primer (5'-TCGGAGGAGCCGCCCATCTCC-3') was based on the consensus sequence derived from alignment of known CRF nucleotide acid sequences of the human (GenBank accession no. NM_000756), rat (GenBank accession no. X03036), pig (GenBank accession no. AF440229), sheep (GenBank accession no. J00803), and fish (GenBank accession no. AF098629). The reverse CRF primer sequence (5'-TCACTTCCCGATGATTTCCA-3') was based on the known CRF nucleotide acid sequence of the chicken (GenBank accession no. AJ621492). The sequences of the forward (5'-CTGCCCCTCTCCTTCCTCT-3') and reverse (5'-TCAGAGGCCAGGCTGCTT-3') AVT primers were based on the known AVT nucleotide acid sequence of the chicken (GenBank accession no. X55130). The sequence of the forward (5'-TTCCGCTGGGGCAAGCCGGT-3') and reverse [5'-TCACT(C/G)(G/A)CCCTTCTTGT(A/G)(G/C)G-3'] POMC primers were based on the consensus sequence derived from alignment of known POMC nucleotide acid sequences of the chicken (GenBank accession no. NM_001031098), human (GenBank accession no. NM_000939), bovine (GenBank accession no. NM_174151), and rat (GenBank accession no. AF510391). The CRF, AVT, and POMC primers were provided by Sigma Aldrich Chimie.
The PCR cycle consisted of a denaturation step (95°C, 30 s), an annealing step (CRF 55°C, 30 s; AVT 60°C, 30 s), and an elongation step (72°C, 60 s) performed with Taq polymerase (Invitrogen-LifeTechnologies). After 40 cycles, the final elongation step was prolonged for 10 min. PCR fragments were analyzed by horizontal agarose electrophoresis and visualized via ethidium bromide fluorescence.
Cloning and Sequencing
The PCR fragment was ligated using a TA-cloning vector (pCR 2.1-TOPO TA Cloning kit; Invitrogen-LifeTechnologies). TOP10 chemically competent Escherichia coli cells were transformed with the resulting ligation mixture, and recombinant bacterial colonies were selected by blue/white screening. Recombinant plasmid DNA was purified with the NucleoSpin Plasmid DNA Purification kit (Macherey-Nagel) and sequenced on both strands using the BigDye Terminator version 3.1 Cycle Sequencing Kit (Applied Biosystems). The nucleotide sequences of the clones corresponding to partial quail CRF, AVT, and POMC cDNA were deposited in the GenBank/EMBL database.
Animals and Rearing Conditions
Japanese quail (Coturnix japonica) from the 36th generation of two divergent genotypes selected for STI or LTI were used in the study (40). Quail were identified by wing banding on the day of hatching. Quail were exposed to continuous light until 3 wk of age and then to a 16:8-h light-dark (16L/8D; light on at 0600) rhythm. The quail from the different genotype and age groups were reared in different collective battery cages. Food and water were provided ad libitum. The caretaker checked the quail daily in the morning (from 0830) and refilled the feeders whenever necessary. No care was provided on the day of the experiment. The birds were treated according to the European Communities Council Directive of November 24, 1986 (86/609/EEC). All procedures described here fully comply with French legislation on research involving animal subjects. All experiments were carried out with due regard to legislation governing the ethical treatment of animals, and investigators were certificated by the French governmental authority for carrying out these experiments (No. 06255).
In Vivo Experimental Procedures
Three successive experiments were performed on sexually mature quail at the age of 6 wk. Groups of quail (n = 4–6/group) of the same genotype and sex were randomly constituted 1 wk before experimentation and placed in small collective battery cages. Experiments were performed between 0830 and 1730 since it has been previously reported that basal CORT levels remain stable throughout the day in LTI and STI quail (26). The experiments consisted of physical restraint in a crush cage or pharmacological challenge. An average of seven to eight quail was used for each experimental point. Blood samples were collected from each quail directly in a tube containing EDTA (2 mg/ml blood) following death by decapitation after submission to the different treatments. All samples were temporarily stored on ice. Following centrifugation at 2,000 g for 15 min at 4°C, plasma samples were separated and stored deep frozen at –20°C until measurement of CORT using a specific RIA (16).
A group of quail bled immediately after capture and transferred from their home cage to the test room was included in each trial to assess basal CORT concentrations in the specific conditions of each experiment.
Treatments
Restraint tests. After transfer to the test room, quail were placed in a "crush-cage" in the form of a wooden box (15 cm length x 5 cm width x 10 cm height) closed at the top by a netting cover. Quail were restrained for a period of 10 min and then bled immediately. This restraint test differed from the selection test, which consisted of placing each quail on its back in a U-shaped cradle, hand-restraining it for 10 s to induce TI, and then releasing it.
Pharmacological challenge. After transfer to the test room, quail were weighed individually to adjust the dose injected to body weight. Quail were injected in the pectoralis major muscle with mammalian 1-24 ACTH (1 mg = 100 IU; Immediate Synacthen; Novartis), mature human/rat CRF peptide (Phoenix Pharmaceuticals), or AVT (Phoenix Pharmaceuticals) at different doses diluted in saline solution (0.9% NaCl wt/vol). Quail were placed back in their home cages immediately after the injection for various durations (ranging between 10 and 120 min) and bled thereafter. Control quail received a representative volume of the vehicle (0.9% NaCl; 400–700 µl volume). Dose-response and time-course relationships were investigated.
In Vivo Experiments
Experiment 1: 1-24 ACTH dose-response and time-course studies.
ACTH DOSE RESPONSE.
CORT dose responses to 1-24 ACTH were analyzed in two separate experiments depending on whether we were testing adrenal sensitivity or maximum CORT adrenal response capacity. Adrenal sensitivity was measured in quail injected with 1-24 ACTH at increasing doses of 0, 1.25, 2.5, 5, and 10 µg/kg body wt and bled 10 min postinjection. The maximum CORT adrenal response capacity was measured in quail injected with 1-24 ACTH at increasing doses of 0, 10, 25, 100, and 250 µg/kg body wt and bled 20 min after injection.
ACTH TIME COURSE.
Quail were injected with doses of 2.5, 10, or 100 µg/kg body wt of 1-24 ACTH or with a representative volume of the vehicle. Quail were then placed back in their home cages and bled 5 (only for the dose of 10 µg/kg body wt), 10, 20, 40, 60, or 120 min after injection.
Experiment 2: CRF dose-response and time-course studies. Quail were injected with CRF at doses of 10 or 100 µg/kg body wt. The quail were returned to their home cages immediately after the injection for 15- or 30-min periods and bled thereafter.
Experiment 3. To compare the potency of CRF and AVT peptides to activate the pituitary-adrenal axis and to investigate their respective possible involvement in the CORT response to stress, both peptides were included in a single comparative study. Quail were thus either submitted to a restraint period of 15 min or injected with 100 µg/kg body wt of 1-24 ACTH, 100 µg/kg body wt CRF, 10 or 50 µg/kg body wt AVT, or injected with a combination of 100 µg/kg body wt CRF and 50 µg/kg body wt AVT. Quail were placed back in their home cages for a 15-min period and bled thereafter.
Statistical Analysis
CORT values were subjected to multifactorial ANOVA to assess the effects of genotype, sex, treatment, and their interactions using the Statview IV program (Abacus Concept). Normality of CORT findings and equality of the variances of the different groups were checked before performing ANOVA. Whenever specific factor and interaction effects reached significance (P < 0.05), post hoc tests were performed using the Fisher test (protected least significant difference). CORT values were expressed as means ± SE, and the level of significance was P < 0.05 unless otherwise stated.
| RESULTS |
|---|
|
|
|---|
A 132-bp CRF fragment (GenBank accession no. AY786511), a 472-bp AVT fragment (GenBank accession no. AY786510), and a 342-bp POMC (GenBank accession no. AY786509) were obtained by RT-PCR and were consistent with the predicted lengths. They corresponded to amino acids 125–167 of chicken pro-CRF, 6–161 of chicken pro-AVT, and 138–250 of chicken POMC. The degrees of similarity of the partial quail pro-CRF, pro-AVT, and POMC cDNA sequences with the corresponding chicken DNA sequences were 99% (131/132), 94% (443/472), and 93% (321/345), respectively. Comparison of the partial predicted amino acid sequences of quail CRF, AVT, and POMC with the chicken, human, rat, pig, bovine, sheep, and fish homologs are presented in Fig. 1. The quail CRF active peptide region (i.e., 41 amino acids) presented full identity with chicken, human, pig, and rat sequences, whereas it differed by 7 and 8 amino acids out of 41 from the bovine and ovine sequences, respectively (Fig. 1A). The quail AVT active peptide region (i.e., 9 amino acids) presented full identity with the chicken AVT peptide but differed by one amino acid from the human, bovine, and rat AVP active peptide region (Fig. 1B). Comparison of the partial predicted amino acid sequences of POMC in quail with homologs in other species indicated that the region encoding the 7-39 ACTH peptide was identical to the chicken region and differed by seven amino acids from the human region and by nine from bovine and rat regions (Fig. 1C).
|
Experiment 1.
DOSE RESPONSE.
Measurement of CORT responses 10 min after injection with doses of 1-24 ACTH ranging between 1.25 to 10 µg/kg body wt indicated that comparable dose-response relationships between CORT concentrations and ACTH doses were found in both genotypes (dose effect, P < 0.0001; genotype effect, P = 0.1; Fig. 2A). On the other hand, a significant sex effect (P = 0.05) was observed on this dose-response relationship. Injection of the 2.5 µg/kg body wt dose led to a significant increase (P = 0.002) in plasma CORT concentration (from 4.3 ± 0.8 to 13.8 ± 1.9 ng/ml in males and from 6.5 ± 1.8 to 12.7 ± 2 ng/ml in females).
|
TIME COURSE.
Injection of saline solution had no significant effect (P = 0.9) on CORT levels (Fig. 3A), whereas injection of 1-24 ACTH at doses of 2.5, 10, or 100 µg/kg body wt induced significant increases (P < 0.0001) in CORT levels (Fig. 3, B–D) that did not differ significantly (P > 0.23) between genotypes. Time-course responses differed according to the dose. A dose of 2.5 µg/kg body wt resulted in maximum CORT levels of 9.7 ± 1.2 ng/ml within 10 min postinjection. CORT concentrations measured 20 min postinjection (4.1 ± 0.7 ng/ml) were no longer significantly different (P = 0.08) from basal levels (2.3 ± 0.5 ng/ml) in either genotype. Maximum mean CORT levels (28.8 ± 1.8 ng/ml) for STI and LTI quail of both sexes measured in response to the injection of a 10 µg/kg body wt dose were also reached within 10 min but remained at high levels for periods lasting between 5 to 20 min postinjection and returned to basal levels (3.4 ± 1.2 ng/ml) from 40 min (3.6 ± 0.7 ng/ml) onward. Injection of 1-24 ACTH at a dose of 100 µg/kg body wt resulted in increases in CORT levels that remained high between 10 and 60 min postinjection in STI and LTI quail of both sexes. CORT concentrations measured 120 min after injection were similar to basal values.
|
|
| DISCUSSION |
|---|
|
|
|---|
Investigation of adrenal function by injection of physiological doses of 1-24 ACTH ranging between 1.25 to 10 µg/kg body wt in this study (i.e., doses inducing increases in CORT levels in the range of physical treatment or stress) suggests that adrenal sensitivity is comparable in LTI and STI genotypes of both sexes. As reported in other bird species (2, 35), the duration of CORT response following the injection of 1-24 ACTH and time intervals to reach maximum CORT levels were found to be dependent upon dose, and again the kinetics were comparable for both genotypes. Thus CORT responses were already at maximum levels within 5 min postinjection for doses up to 10 µg/kg body wt. Maintenance of CORT concentrations at similar levels between 20 and 60 min postinjection of 100 µg/kg body wt of 1-24 ACTH in three of the four groups of quail suggests that maximum CORT levels were reached in this interval or that CORT release was stable throughout this period. Longer time intervals to reach the maximum response have been reported in other studies (2, 35), but this difference in timing might be due to the exact nature of the 1-24 ACTH used. Indeed, different time-course responses have been reported in broilers using two different forms of 1-24 ACTH (immediate and delayed synacthen; see Ref. 22). On the other hand, the higher CORT levels measured in STI quail than in LTI quail in response to the injection of pharmacological doses of 1-24 ACTH (i.e., over 10 µg/kg body wt) indicated that maximum adrenal CORT response capacity was slightly, but significantly, higher in STI than in LTI quail. We can therefore conclude that genetic selection for divergent duration of TI has somehow affected the maximum adrenal CORT response capacity. Results from previous studies have similarly indicated that the adrenal level can be a major area of genetic variability in HPA axis reactivity in various species such as broilers (7), rats (20), and pigs (10, 11, 63). However, differences in adrenal responsiveness between LTI and STI genotypes were only observed following the injection of very high doses of 1-24 ACTH, which induced greater increases in CORT levels than physical stress (i.e., restraint in a crush cage) (27) and which thus could be considered as supraphysiological doses. On the other hand, CORT responses in the range of those measured following the restraint test were observed after injection of much lower doses of 1-24 ACTH (<10 µg/kg body wt) and led to comparable CORT responses for the two genotypes. It is therefore unlikely that the difference in adrenal responsiveness observed can be the cause of the difference in CORT responses between LTI and STI genotypes resulting from restraint stress.
Possible involvement at the pituitary level in the genotype difference in CORT response to restraint was investigated by injecting AVT and/or CRF. It has been shown in previous studies that AVT is active in stimulating CORT release in birds, and is even more potent than CRF in ducks (8), pigeons (61), and European starlings (45). Similarly, in this study, high doses of AVT induced CORT responses of comparable amplitude to those induced in response to high doses of 1-24 ACTH in both genotypes, whereas CORT levels induced by injection of CRF were of much lower amplitude. These results indicate that AVT is very active and much more potent than CRF in stimulating the pituitary-adrenal axis in LTI and STI genotypes. On the other hand, results from other studies have indicated that central administration of CRF is more effective than AVT in inducing CORT release in the plasma of chickens (38, 58), and it has been reported that CRF is a more potent secretagogue than AVP in several mammals [rats (19, 46, 59); ewes (33)]. However, the effects of CRF in birds were investigated in several earlier studies using the ovine CRF peptide (8, 61), but subsequent characterization of the chicken CRF sequence (60) showed that it differs from the ovine sequence by 7 out of 41 amino acids. Given these differences in peptide sequence, it could be hypothesized that the weaker potency of CRF in inducing ACTH release observed in previous studies compared with AVT was due to the fact that the natural CRF was different from the ovine CRF. Although the individual differences in amino acids are predominantly conservative, it cannot be excluded that these differences in amino acids may result in differences in the tertiary structure of the peptide, and thus impair its biological activity. This hypothesis could not be valid in the present study since we used a commercially available rat/human CRF whose structure was shown in this study to be homologous to the quail CRF. On the other hand, because it was administrated peripherally, we cannot exclude the possibility that the dose of 100 µg/kg body wt of CRF was not sufficient to induce the maximum pituitary response. However, we used a very high dose compared with the doses reported in the literature. It therefore appears very likely that AVT is a greater ACTH secretagogue in LTI and STI quail.
Similar CORT levels in LTI and STI quail in response to AVT, and above stress-induced levels, suggest that the difference between genotypes might not be because of a difference in the ability of the pituitary gland to secrete ACTH. These results also suggest that there is probably no difference in receptor capacity for AVT between the genotypes. However, it remains to be explored whether the pituitary gland's sensitivity to AVT is comparable or differs between the two genotypes to demonstrate absence of involvement of the AVT pathway in the genotype differences in CORT response.
Interestingly, higher CORT responses were measured in STI quail compared with LTI quail following CRF injection at the dose of 100 µg/kg body wt. We have no ready argument to explain why lower CORT levels were induced in response to CRF injection in experiment 3. Moreover, the higher or similar levels induced by CRF (experiment 2) compared with those induced by restraint stress suggest that activation of the pituitary gland by the CRF pathway can be sufficient to induce the CORT response observed following restraint stress, with the possible exception of male STI quail. We therefore cannot exclude the possibility that a difference in pituitary responsiveness to CRF might be involved in the differential HPA axis reactivity to stress observed between LTI and STI genotypes. Pituitary gland release of ACTH through CRF activation could be the rate-limiting step regulating CORT secretion in response to acute stress because of limits in receptor capacity for CRF.
A synergic effect between AVT and CRF on CORT release has previously been reported in ducks (8). Surprisingly, although both peptides had a secretagogue effect on CORT, a synergic effect was not observed in the present study. We hypothesized that this was because of the fact that the doses of AVT used already induced the maximum responses. Interestingly, the CORT levels reached in response to combined injection of CRF and AVT peptides were comparable for the two genotypes.
Other mechanisms, not yet investigated in LTI and STI quail, might be involved in the CORT differences between the two genotypes detected after restraint stress. A regulation mechanism may occur above the pituitary gland, at the level of the hypothalamus, and could result in differences in secretion of AVT and/or CRF (47) between the LTI and STI genotypes. This may reflect differences in cognitive processes or differences in negative CORT feedback. It would also be interesting to investigate CORT-binding globulin activity and CORT metabolism, known to be involved in regulating CORT levels (6, 12, 56), which have not been studied to date in LTI and STI quail.
In conclusion, genetic selection for short duration of TI was found to be associated with a greater maximum CORT adrenal response capacity in quail, whereas adrenal sensitivity was comparable in both genotypes. The present results strongly suggest that the differences in HPA axis reactivity to acute stress between LTI and STI genotypes do not involve functional differences at the level of the adrenal glands. We also showed that AVT was much more potent than CRF in stimulating ACTH and CORT release in LTI and STI quail and that the ability of the pituitary gland to secrete ACTH was similar between the two genotypes. The AVT pathway may not be involved in the genotype difference, but pituitary sensitivity in response to AVT remains to be investigated further. On the other hand, differences in the pituitary response to CRF might be involved in the differences in HPA axis reactivity to stress observed between LTI and STI genotypes. However, the respective involvement of CRF and AVT in the HPA axis response to acute stress requires further investigation in quail before firm conclusions can be reached.
| GRANTS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
Present address of D. Hazard: Laboratoire PsyNuGen (Psychoneuroimmunologie, Nutrition, Génétique) INRA UMR1286, CNRS UMR5226, Université Bordeaux 2, Institut François Magendie, 146 rue Léo Saignat 33077 Bordeaux, France.
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |