| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Centre for Integrative Physiology (S.M., M.J.S., J.A.R.) School of Biomedical and Clinical Laboratory Sciences, College of Medicine and Veterinary Medicine, University of Edinburgh, Edinburgh, Scotland EH8 9XD, United Kingdom; and Department of Pharmacology (D.M.), University of Texas Health Science Center at San Antonio, San Antonio, Texas 78229-3900
Address all correspondence and requests for reprints to: Prof. John A Russell, Centre for Integrative Physiology, School of Biomedical and Clinical Laboratory Sciences, College of Medicine and Veterinary Medicine, Hugh Robson Building, George Square, Edinburgh, Scotland EH8 9XD, United Kingdom. E-mail: j.a.russell{at}ed.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In vitro, measuring ACTH secretion from acutely dispersed anterior pituitary cells showed increased corticotroph sensitivity in pregnancy to CRH and to CRH augmentation by vasopressin, attributable to increased intracellular cAMP action. Hence, in late pregnancy, reduced anterior pituitary CRHR1 or V1b receptor expression did not impair corticotroph responses to CRH or vasopressin. Rather, diminished secretagogue secretion in vivo accounts for reduced action of stress levels of exogenous CRH or vasopressin alone; the late pregnancy attenuated ACTH secretory response to swim stress is deduced to be due to reduced vasopressin release by parvocellular paraventricular nuclei neurones.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although CRH is the major stimulator of ACTH secretion in rats, its action is potentiated by vasopressin (20, 21, 22, 23, 24), coproduced in varying amounts in the parvocellular PVN neurones (25, 26, 27). Vasopressin acts on corticotrophs via V1b receptors, activating the protein kinase C pathway, which interacts with the cAMP/protein kinase A pathway activated by binding of CRH to CRH1 receptors (14, 28, 29).
Change in the production of vasopressin by parvocellular PVN neurones is considered important in several states of altered activity of the HPA axis (30, 31, 32, 33, 34, 35, 36, 37), as is the responsiveness of corticotrophs to vasopressin through regulation of V1b (V3) receptor expression (33, 38, 39). In lactation, when HPA axis responses to stressors are also reduced (7, 40, 41), vasopressin is proposed to be more important as an ACTH secretagogue (9, 36, 42). However, in lactation, there is a sustained large increase in basal daily ACTH secretion, and reduced corticosterone secretion (43). In contrast, in pregnancy, basal activity of the maternal HPA axis shows a progressive attenuation of the circadian increase in ACTH secretion, so that daily ACTH output is minimal close to the end of pregnancy (43). Corticosterone secretion decreases in midpregnancy, but plasma concentration increases toward the end of pregnancy, to just-above-nonpregnant levels (43), though there may be a substantial fetal contribution (44), so more is secreted than in lactation. Reduced basal expression of both CRH and vasopressin mRNAs in the parvocellular PVN in pregnancy indicates that reduced production of both peptides may underlie the reduced daily ACTH secretion in pregnancy (8). The relative roles of CRH and vasopressin in regulating stress responses of the HPA axis in pregnancy are, so far, unexplored.
We have examined here the hypothesis that reduced vasopressin production or action is important in the reduced ACTH secretory response to stress in pregnancy. We measured corticotroph responsiveness to vasopressin and CRH in vivo and in vitro, and we assessed the roles of endogenous CRH and vasopressin with antalarmin, a nonpeptide CRHR1 antagonist (45), and dP(Tyr(Me)2),Arg-NH29)AVP, a V1a/b receptor antagonist (46, 83). The results lead us to conclude that vasopressin secretion by parvocellular PVN neurones is suppressed in pregnancy, and this can account for the reduced ACTH stress responsiveness.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Jugular cannulation.
Pregnant and virgin rats were implanted with a jugular cannula for blood sampling, 5–7 d before the experiment, under halothane:nitrous oxide anesthesia. A SILASTIC brand (Dow Corning Corp., Midland, MI) catheter (inside diameter, 0.5 mm; outside diameter, 1 mm), containing sterile heparinized saline (20 U/ml heparin, 0.9% saline), was inserted into the right jugular vein so that the tip lay within the right atrium. The cannula was secured with suture thread, exteriorized at the back of the neck, plugged, and held by a strip of adhesive tape sutured to the skin. At 0800 h on the day of experiment (d 21 of pregnancy), the jugular cannula was flushed and connected to a syringe filled with sterile heparinized saline. All procedures were performed in accord with accepted standards of humane animal care, UK Home Office requirements, and under National Institutes of Health (NIH) guidelines after review and ethical approval by the Institutional Animal Care and Use Committee of the University of Texas Health Science Center at San Antonio.
In vivo analysis of ACTH secretion
Effect of CRHR1 antagonist, antalarmin, on stress-induced ACTH secretion in vivo.
The CRHR1 antagonist antalarmin [20 mg/kg (47)] or vehicle (1 ml/kg, 10% ethanol and 10% cremaphor EL in ddH20) was injected ip between 0830–0900 h. Antalarmin was dissolved in vehicle at 60 C. Two basal blood samples (0.4 ml) were collected 60 min and 90 min after ip injection into a tube containing 15 µl 5%-EDTA/100 µl blood and withdrawn blood replaced with 0.9% saline. All rats were then forced to swim for 90 sec in deep water (40 cm in a plastic bucket 60 cm high and 40 cm in diameter) at 19 C, and further blood samples were taken 5, 15, and 70 min after the swim stress. Blood samples were cooled on ice and centrifuged at 12,000 x g for 5 min, and plasma was separated and stored at –70 C until assay for ACTH.
Antalarmin and ACTH secretory response to exogenous CRH.
Further groups of virgin and d-21 pregnant rats, cannulated for jugular blood sampling as above, were given antalarmin or vehicle as above, immediately after a basal blood sample was withdrawn. Two further blood samples were taken 60 and 90 min after antalarmin, then CRH (200 ng/kg) was injected iv, and further blood samples were collected 5, 15, and 70 min after CRH. This protocol allowed evaluation of the effects of antalarmin on CRH stimulation of ACTH secretion on the same time-course as in the study on antalarmin and swim-stress. Blood plasma samples were collected and stored as above.
Effect of exogenous vasopressin on ACTH secretion.
Two basal blood samples (0.4 ml) were taken 1 and 1.5 h after flushing of the jugular cannula at 0800 h. Immediately after the second basal sample, each rat received either vasopressin [1.7 µg/kg; after (42)] or saline vehicle (500 µl/kg) iv, and further blood samples were taken after 5, 15, 30, and 60 min. The aim was to achieve effective concentrations of vasopressin in hypothalamo-hypophysial portal blood to stimulate ACTH secretion, although the systemic blood concentration is supraphysiological for systemic actions of vasopressin. Blood plasma samples were collected and stored as above.
Effect of V1a/b receptor antagonist and stress-stimulated ACTH secretion.
Two basal blood samples (0.4 ml) were taken between 0830–0900 h (60–90 min after cannula flushing); the vasopressin receptor antagonist [dP(Tyr(Me)2,Arg-NH29)AVP] (10 µg/kg) or vehicle (500 µl/kg) was injected iv [after (37)]. The antagonist is V1-selective and is the most potent of such compounds (46, 83). A blood sample was taken 15 min after antagonist injection, and immediately all rats were forced to swim for 90 sec in deep water (19 C), with further blood samples taken 5, 15, and 60 min after the swim. To test the effectiveness of the antagonist, a second injection of vasopressin receptor antagonist (10 µg/kg) or saline vehicle (500 µl/kg) was given immediately after the 60-min blood sample. Fifteen minutes after injection of the V1a/b antagonist, a blood sample was taken, and vasopressin (1.7 µg/kg) was given to all rats, with a further blood sample taken 10 min later. Blood plasma samples were collected and stored as above.
Combined actions of exogenous CRH and vasopressin on ACTH secretion in vivo.
In this study, each rat was given an iv injection of a combination of CRH (200 ng/kg) and vasopressin (1.7 µg/kg) after the basal blood samples, and further blood samples were collected as in the vasopressin study above.
In situ hybridization analysis for CRHR1, V1bR, and POMC mRNAs
Preparation of radiolabeled probes for in situ hybridization.
Antisense and sense riboprobes for rat CRHR1 were generated using a Hind III or EcoRI linearized pGEM4Z plasmid containing a 606-bp fragment of rat CRHR1 [directed to exonic sequences encoding amino acids 161–362 (48, 49); a gift from Dr. S. J. Lolait, Bristol, UK], respectively. Rat vasopressin receptor V1b riboprobes were generated using Hind III or EcoRI linearized pBluescript II KS (±) plasmid containing a 464-bp fragment of vasopressin receptor V1b (corresponding to a region in the 5' untranslated region immediately upstream of the putative start site; probe I in (50); a gift from Dr. S. J. Lolait) to generate the antisense or sense DNA templates, respectively. Radiolabeled riboprobes were generated by RNA polymerase transcription reactions with the respective T3, T7, or SP6 polymerase, 1 µg of each of the linearized templates with 35S-UTP (40 mCi/1 ml, PerkinElmer Life and Analytical Sciences, Beaconsfield, UK) according to the manufacturer’s protocol (Promega, Madison, WI). The riboprobes were purified through Sephadex G-50 columns (Amersham Biosciences UK Ltd., Chalfont St. Giles, UK). An oligonucleotide probe complementary to rat POMC mRNA bases 711–756 (51) was end labeled with 35S-deoxy-ATP using terminal transferase as described by the manufacturer (Roche Molecular Biochemicals, Mannheim, Germany) and purified with a QIA quick nucleotide removal column (QIAGEN, Hilden, Germany).
Tissue processing for in situ hybridization.
Pituitaries were rapidly removed from rats killed by decapitation and snap-frozen on dry ice; coronal 14-µm-thick cryostat sections were freeze-mounted onto gelatin or poly-l-lysine-coated slides (VWR International Ltd., Lutterworth, UK) and stored at –70 C until used. Sections were fixed with 4% (wt/vol) paraformaldehyde in 0.1 M PBS (20 mM NaH2PO4, 80 mM Na2HPO4, pH 7.4) for 10 min (pH 7.2, 5 min with formaldehyde for sections used with POMC oligonucleotide probe) and washed twice for 5 min in 1x PBS. Sections were treated with 0.25% acetic anhydride in 0.1 M triethanolamine (pH 8.0) for 10 min, washed in 1x PBS for 5 min (ddH2O for 2 min for POMC probe), dehydrated through an ethanol series (70, 80, and 95%, 2 min each), and air-dried. For oligonucleotide probes, a 5-min chloroform wash and 2-min ethanol series (100%, 95%) was also included after the dehydration step. Sections were prehybridized in 0.3 M NaCl, 5 mM Tris-HCl (pH 7.5), 0.5 mM EDTA (pH 7.5), 0.5x Denhardt’s solution, 0.25 mg/ml sheared single-stranded salmon sperm DNA, 0.05 mg/ml yeast tRNA, and 50% vol/vol deionized formamide for 2 h at 50 C. Radiolabeled riboprobes (1x 106 cpm/section) or oligonucleotide probe (0.1x 106 cpm/section) were incubated overnight at 50 C or 37 C, respectively. The riboprobe hybridization solution comprised 50% vol/vol deionized formamide, 0.01 M dithiothreitol, radiolabeled probe (10 x 106 cpm/ml) in hybridization buffer (0.6 M NaCl; 10 mM Tris-HCl, pH 7.6; 1 mM EDTA, pH 7.5; 1x Denhardt’s solution; 0.1 mg/ml denatured salmon sperm DNA; 0.1 mg/ml yeast tRNA; and 10% dextran sulfate). The oligonucleotide hybridization solution comprised 50% vol/vol deionized formamide, 0.02 M dithiothreitol, radiolabeled probe (7 x 106 cpm/ml) in hybridization buffer (0.6 M NaCl; 5 mM Tris-HCl, pH 7.5; 0.5 mM EDTA, pH 7.5; 0.5x Denhardt’s solution; 0.05 mg/ml denatured salmon sperm DNA; 0.05 mg/ml yeast tRNA; 0.05 mg/ml yeast total RNA; 0.05 mg/ml poly(A); and 1.25% dextran sulfate). For riboprobes, sections were washed 3 x 5 min in 2x saline sodium citrate (SSC) at room temperature, incubated with 30 µg/ml ribonuclease A at 37 C for 1 h, washed for 30 min in 2x SSC at room temperature, followed by two washes with 0.1x SSC at 60 C for 90 min. Sections were dehydrated in a series of 50, 70, and 90% ethanol in 0.3 M ammonium acetate for 2 min each, air-dried, and exposed to Hyperfilm-ßmax (Amersham Biosciences) for 15 d. For the POMC oligonucleotide probe, sections were washed in four changes of 1x SSC at 45 C (15 min each), then twice in 1x SSC at room temperature for 30 min, rinsed in ddH2O, air-dried, and exposed to Hyperfilm-ß max for 24 d.
Semiquantitative analysis of autoradiographs.
Analysis of developed autoradiographic film was performed using a computer-based image analysis system and software (NIH Image v1.62). Integrated film grain density (area of film grain per unit area of tissue profile) was measured in duplicate 0.53 x 0.53-mm boxes in either lobe of each pituitary section. For each animal (n = 8–17), measurements were made on six replicate sections. Background measurements were made on areas adjacent to the pituitary and subtracted; measurements over a 14C polymer strip (Amersham Biosciences) confirmed that tissue measurements were on the dynamic part of the radioactivity/film grain density curve.
In vitro release assays and RIAs.
Collection of hypothalamus and median eminence for CRH assay.
Rats were decapitated with a guillotine between 0900–1000 h and the brain and pituitary removed. The median eminence, with the attached pituitary stalk, and a tissue block containing the hypothalamus were snap-frozen on dry ice and sonicated in 0.5 M acetic acid and 0.1 M HCl. CRH content in the lysate was determined by RIA as previously described (52), with intra- and interassay coefficients of variation of 5% and 10%, respectively. CRH contents per median eminence or hypothalamus, alone or combined, were calculated.
Isolation of acutely dissociated rat anterior pituitary cells.
Acutely dissociated anterior pituitary cells were isolated essentially as previously described (53). Briefly, anterior pituitary glands from each experimental group were pooled (4–8 rats per group), finely chopped, and trypsinized in DMEM [Invitrogen, Paisley, Scotland, UK; supplemented with 25 mM HEPES (pH7.4) and 0.25% wt/vol BSA, DMEM-BSA] containing 0.25 mg/ml trypsin (Worthington Biochemical Corp., Lakewood, NJ) and 10 mg/ml deoxyribonuclease I for 25 min at 37 C. Trypsinization was terminated by resuspending cells in DMEM-BSA containing 0.5 mg/ml Lima Bean Trypsin inhibitor (Sigma-Aldrich, Poole, UK) and 100 kallikrein units/ml aprotinin (Sigma-Aldrich) before filtering through a 100-µm nylon mesh. Cells were centrifuged (150 x g for 5 min) and resuspended in 10 ml DMEM-BSA with gentle rotation for 2 h at 37 C before a final wash with DMEM-BSA. Cell viability, as assessed by Trypan Blue exclusion, was more than 95%. For ACTH secretion studies, 1 x 105 cells (4–6 replicates per experiment) were incubated in a total vol of 400 µl containing the respective concentrations of ACTH secretagogues for 1 h at 37 C. Total ACTH content in 1 x 105 cells in the absence of secretagogues was measured after freeze-thawing. For vasopressin (V1a/b) receptor antagonist studies, cells were preincubated for 5 min with the mixed V1a and V1b receptor antagonist [dP(Tyr(Me)2,Arg-NH29)AVP] before exposure to secretagogue. Secretion was stopped by placing the tubes on ice for 15 min, cells were pelleted at 150 x g, and supernatant was stored at –70 C before RIA.
ACTH RIAs.
For in vitro secretion studies, supernatants were assayed for immunoreactive ACTH using a double-antibody precipitation RIA essentially as previously described (52). Duplicate aliquots of supernatant were incubated overnight in RIA buffer (0.05 M sodium phosphate buffer, pH 7.4; 0.1% BSA; 0.1% Triton X-100; 3% polyethylene glycol 6000; 2.5 mM EDTA; and 100 kallikrein inhibitor units aprotinin/ml) containing 12,000–15,000 cpm/tube 125I-ACTH (Phoenix Pharmaceuticals Inc., Belmont, CA) and a 1:100,000 dilution of a sheep polyclonal ACTH antiserum [anti-[corticotropin-(2–16)-peptide] IgG (54), a kind gift from Prof. P. J. Lowry, University of Reading]. Samples were then incubated with donkey antisheep IgG (1:25 dilution, Scottish Antibody Production Unit, Lanarkshire, UK) and nonimmune rabbit serum (1:400 dilution, Scottish Antibody Production Unit) for 3 h at 4 C and bound radioactivity precipitated with 6% PEG-6000 followed by centrifugation at 1950 x g for 25 min at 4 C. Assay sensitivity was 13 pg/ml, with intraassay coefficient of variation less than 10%. ACTH in rat plasma was assayed by RIA using a kit (which uses a porcine antiserum recognizing ACTH amino acids 5–18; MP Biomedicals, Asse-Relegem, Belgium). Assay sensitivity was 10 pmol, with intraassay coefficient of variation less than 10%.
Drugs and chemicals
CRH and vasopressin were from Bachem AG, Switzerland; 8-(4-chlorophenylthio)-cAMP (8-CPT-cAMP) a cell-permeable cAMP analog, was from Biolog Life Science Institute, Bremen, Germany. The CRH antagonist, antalarmin, was a kind gift from Dr. George P. Chrousos, NIH. The V1a/b antagonist [dP(Tyr(Me)2,Arg-NH29)AVP] was a generous gift of Prof. M. Manning, Medical College of Ohio, Toledo. All other reagents were of the highest analytical grade from Sigma-Aldrich or BDH-Merck unless indicated otherwise.
Statistical analysis
All data are means ± SEM (n = number of experiments/rats). The CRH and vasopressin receptor expression data were analyzed by Student’s t test. The POMC expression data were analyzed by one-way ANOVA, with post hoc Newman-Keuls tests. All data from blood sampling were analyzed by repeated-measures two-way ANOVA, with post hoc Newman-Keuls tests. Data from in vitro experiments were analyzed with two-way ANOVA and post hoc Newman-Keuls tests. To pool the results from in vitro ACTH release assays in dispersed pituitary cells (see Fig. 5B
), the ratios of ACTH release relative to basal were first calculated for each treatment by group in each experiment. Then the differences in these values between the pregnant and virgin groups were calculated as ratios (relative to virgin) for each experiment. Finally, means of these values (pregnant: virgin ratios within an experiment) were then calculated across experiments and analyzed using a Kruskal-Wallis one-way ANOVA on ranks, followed by Mann-Whitney rank-sum tests, with a value of 1 as the null hypothesis. In all analyses, P < 0.05 was regarded as significant.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). There were no significant differences for CRH content in the rest of the hypothalamus in both pregnant groups (d 10 and d 21) compared with virgins (t test, P = 0.33 and 0.17, respectively). Median eminence CRH content, expressed as a percentage of that in the whole hypothalamus, representing the proportion of total CRH readily available for secretion, was reduced significantly in the d-21 pregnant group (by 11%; t test, P < 0.05) but not in the d-10 group.
|
, A and B). Strong expression in the pars intermedia persisted throughout pregnancy and was too dense for quantification. Assay of total ACTH peptide content in acutely dispersed anterior pituitary cells showed no difference in ACTH contents between virgin and either d-10 or d-21 pregnant rats (Fig. 1C).
|
) in accordance with other studies (55). No difference in expression between 21-d pregnant and virgin rats was found (Fig. 2C). V1b receptor mRNA expression was also detected in the anterior pituitary from both virgin and 21-d pregnant rats, but with no significant expression in pars intermedia (Fig. 2B), also in accordance with other studies (49). Quantification of silver grain density showed that V1b receptor mRNA expression was decreased by 19% in the anterior pituitary from d-21 pregnant rats compared with virgins (Fig. 2D; P < 0.05, t test).
|
). Two-way ANOVA for repeated measures showed a significant effect of swimming at 5 and 15 min (F(4, 80) = 110.34, P < 0.0001), of group (F(3, 20) = 7.68, P < 0.01), and of time X group interaction (F(12, 80) = 12.62, P < 0.0001). After swim stress, plasma ACTH concentrations were significantly increased in vehicle- and antalarmin-treated virgin and pregnant rats at 5 and 15 min, with peak values at 5 min. Plasma ACTH concentrations returned to close-to-basal levels by 70 min post swim stress, though levels tended to be greater in the pregnant rats at this time point. The plasma ACTH response 5 min after swim stress was less in vehicle-treated pregnant rats than in vehicle-treated virgins [increase from basal in pregnant rats was 53% of the increase in virgin rats; Fig. 3; repeated-measures ANOVA (RM ANOVA) P < 0.05]. ACTH release in response to swim stress was attenuated by antalarmin pretreatment in both virgin and pregnant rats compared with the respective vehicle-treated groups (Fig. 3A) [antalarmin inhibited swim-stimulated ACTH release (measured at 5 min) by 57% and 53% in virgin and pregnant rats, respectively; Fig. 3, A and B; RM ANOVA, P < 0.05]. There was no difference in plasma ACTH concentrations at 70 min after swim stress between the respective vehicle- and antalarmin-treated groups (Fig. 3A).
|
). After CRH, two-way ANOVA for repeated measures showed a significant main effect of treatment (F(1, 28) = 41.56, P < 0.0001), as well as a group X time interaction (F(5, 40) = 11.56, P < 0.0001). There was a significant effect of CRH in all groups at 5 min (P < 0.05), with greater plasma ACTH concentration in vehicle-treated virgins than in vehicle-treated pregnant rats [P < 0.05; confirming a previous study (7)]; antalarmin reduced the ACTH response to CRH similarly in virgin and pregnant rats (P < 0.05). At 15 min after CRH, plasma ACTH concentration remained increased in the vehicle-treated virgin and pregnant rats (P < 0.05 vs. basal), and the increase was greater in the virgin rats (P < 0.05); in the antalarmin-treated rats, there was no longer a significant ACTH response to CRH at 15 min. There were no significant effects of antalarmin or CRH at 70 min after CRH. Increments in plasma ACTH concentration from basal (after antalarmin or vehicle, before CRH) showed a similar pattern (increments at 5 min: virgin/vehicle, +431.0 ± 51.7 pg/ml; pregnant/vehicle, +208.0 ± 21.5 pg/ml; virgin/antalarmin, +110.5 ± 18.7 pg/ml; pregnant/antalarmin, +96.1 ± 16.2 pg/ml); two-way ANOVA showed significant effect of group (P < 0.05) and treatment (P < 0.05); antalarmin reduced the increment in plasma ACTH concentration after CRH in virgin rats by 74% and in pregnant rats by 54%.
ACTH secretory response to exogenous vasopressin
This experiment tested whether reduced action of vasopressin on ACTH secretion in pregnant rats could explain the reduced stress response in pregnancy. Two-way ANOVA for repeated measures showed a significant effect of time (F(5, 105) = 77.69, P < 0.0001) and of group (F(3, 21) = 45.26, P < 0.0001). Plasma ACTH concentration was increased at 5 min and 15 min after iv vasopressin (1.7 µg/kg) injection in both virgin and pregnant rats, with the peak response at 5 min (Fig. 4A; RM ANOVA: P < 0.0001). The ACTH response was significantly less in pregnant rats (56% of the virgin value at 5 min). Vehicle injection was without effect. Vasopressin-treated rats lay prostrate for several minutes after injection, perhaps a consequence of vasopressor actions.
|
For both virgin and pregnant rats, ACTH secretion from dispersed anterior pituitary cells in vitro was dose-dependently related to CRH concentration (Fig. 5A; P < 0.05, two-way ANOVA). There was a significant main effect of treatment (F(8, 68) = 34.64, P < 0.0001), as well as a group X treatment interaction (F(8, 68) = 4.23, P < 0.001). Vasopressin alone had no significant effect on ACTH secretion in vitro in either pregnant or virgin rats (0.01–10 nmol/liter; the effect of 2 nmol/liter is shown in Fig. 5A). In contrast, 0.1 nmol/liter CRH plus 2 nmol/liter vasopressin, and the cell-permeable cAMP analog, 8-CPT-cAMP, were more effective in stimulating ACTH secretion by anterior pituitary cells from 21-d pregnant rats than from virgins (Fig. 5B; *, P < 0.05, Kruskal-Wallis one-way ANOVA on ranks, followed by Mann-Whitney rank-sum tests).
In vivo, combination of CRH (200 ng/kg, iv, as above) and vasopressin (1.7 µg/kg, iv, as above) significantly increased plasma ACTH concentrations at 5 min and 15 min after injection, with the peak response at 5 min [Fig. 6; RM ANOVA: a significant effect of time (F(5, 25) = 42.50, P < 0.0001) but not of pregnancy (F(1, 5) = 0.48, P = 0.84)]. In striking contrast to the responses to CRH or vasopressin alone (Figs. 3 and 4), there were no differences in ACTH concentrations between pregnant and virgin rats at any time point in response to CRH and vasopressin given together (Fig. 6).
|
; P < 0.05, RM ANOVA). The V1a/b antagonist significantly reduced the ACTH response to swimming stress at 5min (by 40%; P < 0.01, RM ANOVA), but not 15 min, in virgin rats. However, the V1a/b antagonist did not reduce the ACTH secretory response to swim stress in pregnant rats at any time point. By 60 min post stress, plasma ACTH concentrations had returned to near basal, except in the pregnant-vehicle group, and were not significantly different among groups. To test the effectiveness of the V1a/b antagonist on vasopressin actions on ACTH secretion, the same dose of the V1a/b antagonist or vehicle was given iv to the treatment or vehicle group, respectively, at 60 min after stress, and vasopressin (1.7 µg/kg) was given to all groups 15 min later (Fig. 7C). Vasopressin increased plasma ACTH concentration in both virgin and pregnant vehicle-treated groups; but as above, vasopressin caused a greater ACTH response in the vehicle-treated virgin rats compared with the vehicle-treated pregnant rats. The V1a/b antagonist largely prevented the ACTH response to exogenous vasopressin in both virgin and pregnant groups (by 89% and 91%, respectively). The V1a/b receptor antagonist at 10 µmol/liter, but not at 100 nmol/liter, blocked the augmentation by vasopressin (2 nmol/liter) of CRH (0.1 nmol/liter)-stimulated ACTH secretion from acutely dispersed anterior pituitary cells (data not shown). Whereas vehicle-treated rats injected with vasopressin rapidly adopted a prostrate posture, lasting for several minutes, this behavioral response to vasopressin was completely prevented by prior injection of the V1a/b antagonist.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CRH in the hypothalamus
CRH from the hypothalamus is the major regulator of ACTH secretion in rodents (21, 56, 57), but there was only a modest reduction in CRH content in the median eminence in pregnancy. Because CRH mRNA expression in pPVN neurones is also reduced (8), this indicates a quasi balanced reduction in basal synthesis and secretion of CRH, consistent with the loss of the circadian rise in ACTH secretion toward the end of pregnancy (43). The marked reduction in ACTH secretion in response to stress in pregnancy seems unlikely to be due to unavailability of CRH for secretion but may be a result of reduced excitation of CRH pPVN neurones (9, 58) with enhanced slow central glucocorticoid negative feedback in pregnancy (8).
Anterior pituitary POMC mRNA and ACTH
The apparently paradoxical decrease in POMC mRNA expression, but not in anterior pituitary ACTH content, at the end of pregnancy is readily explained by the reduced daily secretion of ACTH, due to loss of the circadian increase in secretion (43). Thus, the reduced POMC mRNA expression can still produce enough ACTH for the reduced daily secretion, without reducing ACTH stores. Reduced POMC mRNA expression may result from the reduced daily stimulation by CRH, as deduced above, because CRH increases POMC mRNA expression (16, 17, 18, 19). Alternatively, because basal POMC mRNA expression is evidently not reduced by ablation of the hypothalamic CRH output via the median eminence, reduced POMC mRNA expression may result from increased feedback inhibition by glucocorticoids (19, 59). Corticosterone production is modestly increased toward the end of pregnancy (43), despite the reduced daily ACTH secretion, because sensitivity of the adrenal cortex to ACTH is increased by estrogen action (60), and the fetuses also produce corticosterone (44). Nonetheless, the attenuated ACTH secretory response to stress near the end of pregnancy is not a result of reduced availability of stored ACTH.
Anterior pituitary CRHR1 mRNA expression
The present finding that anterior pituitary CRHR1 mRNA expression is not reduced in pregnancy indicates that CRHR down-regulation, previously reported (7), is not a result of reduced gene expression. Posttranslational control of CRHR1 expression has been deduced previously from several lines of evidence (61). Expression of CRHR1 in corticotrophs is negatively regulated by CRH (61, 62), and vasopressin potentiates inhibition of CRHR1 expression by CRH (61). By contrast, vasopressin increases CRHR1 mRNA expression in vivo (61), although it decreases expression in vitro (63). Reduced corticotroph CRHR1 expression in pregnancy, as indicated by ligand binding studies (7, 42), is not likely to be due to negative regulation by CRH or vasopressin, because the present and previous studies indicate reduced basal secretion of these secretagogues (8). The reported increase in corticosterone secretion toward the end of pregnancy may be responsible for reduced CRHR expression, as glucocorticoid decreases CRH binding in the anterior pituitary (19, 64). Although CRHR1 mRNA is negatively regulated, acutely, by glucocorticoid (19, 65), this may not apply for basal levels of glucocorticoid (19, 59), consistent with lack of a reduction in CRHR1 mRNA expression in pregnancy in the present study. The reduced anterior pituitary POMC mRNA, but not CRHR1 mRNA, expression in pregnancy is in accord with the higher threshold for glucocorticoid inhibition of CRHR1 mRNA expression (19).
V1b mRNA in the anterior pituitary
The decreased V1b mRNA expression in the anterior pituitary, found in the present study, may underlie the reported reduced V1b receptor binding (42). However, the relationship between V1b receptor mRNA and protein expression is not simple: translation is an important site of regulation (66), and this was not investigated here. It should be noted that our measurements were not confined to corticotrophs and that, unlike the CRHR1, the V1b receptor is expressed in some other anterior pituitary cells, as yet unidentified (49, 55). By contrast with down-regulation of CRHR1 expression, V1b receptor expression is increased in conditions of HPA axis activation (67). Glucocorticoid increases V1b receptor mRNA expression but decreases V1b receptor binding (67, 68, 69). Consequently, the decrease in V1b receptor mRNA expression in pregnancy found here is not explained by any increase in glucocorticoid secretion in pregnancy. Moreover, vasopressin stimulation of ACTH secretion has been reported to be resistant to negative feedback by glucocorticoid (70); and although glucocorticoid reduces inositol phosphate stimulation of ACTH secretion (28), it increases the coupling efficiency of V1b receptors to inositol phosphate formation (69). The role of vasopressin in regulating V1b receptor expression is not entirely clear, though there is evidence for both negative (50) and positive regulation (67, 68). Reduced expression of vasopressin mRNA in pPVN neurones in pregnancy (8) suggests reduced production of vasopressin. Thus, reduced vasopressin release by pPVN neurones into the hypothalamo-hypophysial portal system under basal conditions in pregnancy may lead to the decrease in anterior pituitary V1b receptor protein (42) and mRNA (this study) expression in pregnancy.
CRHR1 antagonist (antalarmin) and ACTH secretion
To test the involvement of CRH in the response to swim stress in pregnancy, we used a CRHR1 antagonist, antalarmin (45). We confirmed that antalarmin antagonized stimulation of in vivo ACTH secretion by exogenous CRH, and we found that antalarmin reduced the normalized ACTH response to swimming similarly in virgin and 21-d pregnant rats. The dose of antalarmin and the time-course used here were based on a previous pharmacokinetic study (71) and reports of the half-life of effects on conditioned fear behavior and ACTH and corticosterone secretion (72). Previous studies have focused on behavioral effects of this antagonist (71, 72), as it accesses the brain to act on central CRH1 receptors (73). The present report clearly shows a reduced ACTH response to swim stress after antalarmin; others have shown reduced stress-induced ACTH secretion with only low-intensity footshocks (72), or have shown no effects on the ACTH response to immobilization stress after chronic antalarmin treatment (74). However, because central CRH, acting via CRH1 receptors, has been implicated positively in the HPA stress response (75, 76), though also negatively (77), we cannot exclude a contribution from antagonism of centrally released CRH action by antalarmin in the reduced ACTH secretory responses to stress. However, antalarmin strongly reduced the ACTH response to exogenous CRH in both virgin and pregnant rats, and to extents similar to those by which it reduced the ACTH responses to forced swimming (Fig. 3). Thus, its effects on stress responses in the present study can reasonably be attributed to antagonism of CRH actions on corticotrophs. At the dose used in this study, anatalarmin was not as effective as astressin B (a peptide CRHR1 antagonist) or NBI 30775 (a nonpeptide CRHR1 antagonist) tested against in vivo CRH stimulation of ACTH secretion, but it was as effective as NBI 30775 tested against lipopolysaccharide-stimulated ACTH secretion in vivo (78). The similar effects of antalarmin in reducing the normalized ACTH response to swim stress in the virgin and pregnant rats indicate that the reduced ACTH stress response in pregnant rats is not attributable to lack of stress-induced CRH secretion. Nonetheless, we confirmed that the effectiveness of exogenous CRH in stimulating ACTH secretion is reduced in pregnancy (7). This has been explained as a result of reduced CRHR1 availability (7) and, as discussed above, is not a result of depletion of the ACTH store in the corticotrophs. However, V1b receptor expression in the anterior pituitary (discussed above), and vasopressin mRNA expression in the pPVN (8) are reduced in pregnancy. Consequently, CRH secreted in response to stress, or exogenous CRH, may be less effective because of, respectively, reduced stress-stimulated or basal secretion or action of vasopressin.
Vasopressin and ACTH secretion
The simplest conclusion from the present finding that, under basal conditions, exogenous vasopressin was less effective in stimulating ACTH secretion in pregnant rats is that this reflects down-regulation of V1b receptors in corticotrophs. Our further experiments led to rejection of this explanation, because vasopressin interacted with CRH to stimulate ACTH secretion in pregnant rats as effectively as in virgin rats. Thus, the alternative possibility is that, because exogenous vasopressin is less effective when basal secretion of CRH is reduced (21), this could be so in pregnancy. Similarly, reduced actions of exogenous CRH in pregnancy could be a consequence of reduced vasopressin secretion into portal blood. This scenario was supported by our finding that the ACTH secretory responses in pregnant and virgin rats to coadministration of CRH and vasopressin were indistinguishable from each other. The possibility that the combined in vivo treatment with CRH and vasopressin was equally effective in pregnant and virgin rats because of a right shift in the dose-response curves for CRH or vasopressin alone in pregnancy was eliminated by the in vitro experiments. The present findings contrast with the greater effectiveness of exogenous vasopressin in stimulating ACTH secretion in lactating rats (42).
In vitro studies
Under in vitro conditions, in which we could control the ambient concentrations of secretagogues, there were greater ACTH secretory responses of dispersed corticotrophs from late pregnant rats to CRH, and a greater (
60%) augmentation of the effects of CRH by vasopressin, compared with virgin controls. An explanation for these greater effects of CRH and vasopressin on ACTH secretion could be changes in postreceptor mechanisms in the corticotrophs. CRH action on CRH1 receptors stimulates cAMP generation, which activates protein kinase A and leads to increased intracellular [Ca2+] via L- and P-type channels, which triggers exocytosis and ACTH secretion (14, 79). However, CRH stimulates the generation of less cAMP by pituitary segments from pregnant rats than from virgin rats (7); similarly, cAMP generation stimulated by 1 nmol/liter CRH was 33% less in acutely dispersed anterior pituitary cells from pregnant vs. virgin rats (Johnstone, H. E., J. A. Russell, and M. J. Shipston, unpublished). The present finding that 8-CPT-cAMP was more effective in stimulating ACTH secretion by anterior pituitary cells from pregnant than virgin rats indicates that the reduced stimulation by CRH of cAMP generation may be compensated by a greater effectiveness of cAMP. This could also explain the greater augmentation by vasopressin of CRH stimulation of ACTH secretion, as vasopressin binding to V1b receptors activates protein kinase C and exerts a synergistic effect on CRH-stimulated cAMP production (29, 80). Also, as noted above, increased coupling efficiency of the V1b receptors to inositol phosphate production, as a result of increased glucocorticoid action (69), may explain the greater augmentation of CRH action by vasopressin in pregnancy. There was an apparent discrepancy between the in vivo and in vitro results, in that in vivo combined administration of CRH and vasopressin had similar effects on ACTH secretion in pregnant and virgin rats, but in vitro the combined effects of vasopressin and CRH were greater on anterior pituitary cells from pregnant rats. This is likely to be simply a result of the use of large doses of CRH and vasopressin in vivo, to answer the question of whether ACTH secretion in pregnant rats could be driven to similar levels as in virgin rats by the secretagogues given together, perhaps resulting in near-maximal responses, whereas in vitro graded concentrations were tested to evaluate interactions.
V1a/b antagonist and ACTH secretion
In contrast with antalarmin, the V1a/b antagonist reduced the secretion of ACTH in response to swim stress in virgin, but not in pregnant, rats. This indicates that vasopressin does not contribute significantly to the ACTH stress response in pregnant rats, and is in contrast with models of chronic stress, such as experimental arthritis, in which vasopressin is the major stimulator of ACTH secretion (33). Relevant to the present study is the recent finding that, in transgenic mice with targeted inactivation of the V1b receptor gene, the ACTH response to forced swimming is substantially reduced compared with normal controls (81). This indicates that vasopressin is normally important in driving ACTH secretion in response to the forced swimming stressor.
We confirmed that the antagonist blocks the augmenting action of vasopressin on CRH-stimulated ACTH secretion by corticotrophs in vitro, and the action of exogenous vasopressin on ACTH secretion in vivo in both virgin and pregnant rats. However, this antagonist is also effective at V1a receptors that mediate vasopressin actions on blood vessels (46), and we cannot exclude indirect actions of vasopressin or the antagonist on ACTH secretion through cardiovascular changes. However, systemic secretion of vasopressin is not increased with swim stress (82). Interestingly, whereas antalarmin reduced the ACTH response at both 5 and 15 min after swim stress, the V1a/b antagonist reduced the ACTH response only at 5 min (only in virgin rats). These findings suggest that, in virgin, but not in pregnant rats, the role of vasopressin in the forced swimming paradigm is to promote the rapid secretion of ACTH. The differential action of the V1a/b antagonist, but not the CRHR1 antagonist, on the ACTH stress response in pregnant and virgin rats, together with the similar effects of combined exogenous CRH and vasopressin, leads to the conclusion that reduced secretion of vasopressin by parvocellular PVN neurones underlies the attenuated ACTH stress response in pregnancy. Reduced basal secretion of vasopressin and CRH may account, respectively, for reduced anterior pituitary V1b mRNA and POMC mRNA expression in pregnancy. These changes do not lead to reduced responsiveness to CRH and vasopressin because of increased effectiveness of postreceptor signaling, and an undiminished ACTH store.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Present address for S.M.: Department of Pharmacology, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, Texas 78229-3900.
First Published Online December 9, 2004
Abbreviations: 8-CPT-cAMP, 8-(4-chlorophenylthio)-cAMP; CRHR, CRH receptor; HPA, hypothalamo-hypophysial-adrenal; POMC, proopiomelanocortin; PVN, paraventricular nuclei; RM ANOVA, repeat-measures ANOVA; SSC, saline sodium citrate.
Received October 18, 2004.
Accepted for publication December 1, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
