The different roles of glucocorticoids in the hippocampus and hypothalamus in chronic stress-induced HPA axis hyperactivity

Abstract

Hypothalamus-pituitary-adrenal (HPA) hyperactivity is observed in many patients suffering from depression and the mechanism underling the dysfunction of HPA axis is not well understood. Chronic stress has a causal relationship with the hyperactivity of HPA axis. Stress induces the over-synthesis of glucocorticoids, which will arrive at all the body containing the brain. It is still complicated whether glucocorticoids account for chronic stress-induced HPA axis hyperactivity and in which part of the brain the glucocorticoids account for chronic stress-induced HPA axis hyperactivity. Here, we demonstrated that glucocorticoids were indispensable and sufficient for chronic stress-induced hyperactivity of HPA axis. Although acute glucocorticoids elevation in the hippocampus and hypothalamus exerted a negative regulation of HPA axis, we found that chronic glucocorticoids elevation in the hippocampus but not in the hypothalamus accounted for chronic stress-induced hyperactivity of HPA axis. Chronic glucocorticoids exposure in the hypothalamus still exerted a negative regulation of HPA axis activity. More importantly, we found mineralocorticoid receptor (MR) - neuronal nitric oxide synthesis enzyme (nNOS) - nitric oxide (NO) pathway mediated the different roles of glucocorticoids in the hippocampus and hypothalamus in regulating HPA axis activity. This study suggests that the glucocorticoids in the hippocampus play an important role in the development of HPA axis hyperactivity and the glucocorticoids in the hypothalamus can't induce hyperactivity of HPA axis, revealing new insights into understanding the mechanism of depression.

Figure 1. Chronic stress induced-depression requires glucocorticoids.

(A) Plasma CORT levels in the mice exposed to CMS with or without metyrapone (s.c., 100 mg/kg/d) for 28 days (control n = 7, CMS n = 6, CMS+ metyrapone n = 6). (B–E) Metyrapone abolished CMS-induced behavioral modifications. Immobility time in the TST (B, control n = 12, CMS n = 12, CMS+ metyrapone n = 16) and FST (C, control n = 12, CMS n = 12, CMS+ metyrapone n = 16), sucrose preference (D, control n = 12, CMS n = 12, CMS+ metyrapone n = 15), and locomotor activity (E, control n = 12, CMS n = 12, CMS+ metyrapone n = 16) of the mice exposed to CMS with or without metyrapone (s.c., 100 mg/kg/d) for 28 days. (F) Western blotting showing CRF levels in the hypothalamus of adult mice exposed to CMS with or without metyrapone (n = 4 for each group), arrow heads indicate the location of nearest band in the ladder. Error bars denote SEM, * p<0.05, ** p<0.01, *** p<0.001, compared to control; # p<0.05, ## p<0.01, ### p<0.001, compared to CMS; one-way ANOVA.

Figure 2. High concentration of glucocorticoids is sufficient to induce depressive behaviors and hyperactivity of HPA axis.

(A) Immobility time in the TST after 28 days CORT treatment (40 mg/kg/d, s.c.) (n = 14 for each group). (B) Immobility time in the FST 28 days CORT treatment (40 mg/kg/d, s.c.) (n = 14 for each group). (C) Sucrose preference of mice 28 days CORT treatment (40 mg/kg/d, s.c.) (n = 14 for each group). (D) Locomotor activity of the mice 28 days CORT treatment (40 mg/kg/d, s.c.) (n = 14 for each group). (E) Representative western blotting of CRF and GAPDH in the hypothalamus of mice treated with CORT (40 mg/kg/d s.c.) or DMSO for 28 d (n = 5 for each group) and (F) the statistical data of the western blotting experiment. Arrow heads indicate the location of nearest band in the ladder. (G) Immobility time in the TST after 28 days treatment with DMSO (20 µl/g/d, s.c.) (n = 11 for each group). (H) Immobility time in the FST after 28 days treatment with DMSO (20 µl/g/d, s.c.) (n = 11 for each group). (I) Sucrose preference after 28 days treatment with DMSO (20 µl/g/d, s.c.) (n = 11 for each group). (J) Increased weight of mice after 28 days treatment with DMSO (20 µl/g/d, s.c.) (n = 11 for each group). (K) Locomotor activity of the mice after 28 days treatment with DMSO (20 µl/g/d, s.c.) (n = 11 for each group). Error bars denote SEM, *p<0.05, two-tailed Student's t test.

Figure 3. Selective infusion of CORT into the hypothalamus increases the level of CORT in the hypothalamus persistently.

(A) Representative photo indicated the target of the infusion (the left PVN region of the hypothalamus). 3V: third ventricle. (B) The concentration of CORT in the whole hypothalamus 24 h after infusion of CORT. n = 3 for each group. (C) The concentration of CORT in the whole hypothalamus 28 d after infusion of CORT. n = 3 for each group. Error bars denote SEM, *** p<0.001 compared to control group, two-tailed Student's t test in (B), one-way ANOVA in (C).

Figure 4. High concentration of glucocorticoids in the hypothalamus exerts negative regulation of HPA axis.

Depressive-like behaviors test were measured 28 d after infusion of CORT (10 µM, 1 µl) into each PVN: (A) Immobility time in the TST (n = 15 for each group). (B) Immobility time in the FST (n = 15 for each group). (C) Sucrose preference of mice (n = 15 for each group). (D) Locomotor activity of the mice within 5 min (n = 15 for each group). (E) Representative western blotting of CRF and GAPDH in the hypothalamus (left) and the statistical data of the western blotting experiment (right) (n = 4 for each group). Representative western blotting of CRF and GAPDH in the hypothalamus 28days after the infusion of 10 µM CORT (F) or 0.1 µM CORT into the PVN (H) and their statistical data (G, I), n = 4 for each group. Arrow heads indicate the location of nearest band in the ladder. (J) The concentration of CORT in the plasma 28 d after infusion. n = 4 for each group, the 10 µM CORT group and the 0.1 µM CORT were analyzed separately. Error bars denote SEM, *p<0.05, ***p<0.001, two-tailed Student's t test except for one-way ANOVA in (E).

Figure 5. High concentration of glucocorticoids in the hippocampus plays opposite roles in acute and chronic phase in regulating HPA axis activity.

(A) Western blotting showing CRF protein expression in the hypothalamus 2 hours after the infusion of CORT (10 µM, 2 µl) into the DG regions. n = 3 for each group. Arrow heads indicate the location of nearest band in the ladder. (B) The CORT level in the plasma 2 hours after the infusion of CORT (10 µM, 2 µl) into the DG regions. n = 4 for each group. (C) RT-PCR showing CRF mRNA in the hypothalamus 28 days after the infusion of CORT (10 µM, 2 µl) into the DG regions. n = 3 for each group. (D) The CORT level in the plasma 28 days after the infusion of CORT (10 µM, 2 µl) into the DG regions. n = 6 for each group. (E) Representative imaging of CRF-positive cells in the PVN of the hypothalamus 28 days after the infusion of CORT (10 µM, 2 µl) into the DG regions. The result was repeated in 3 mice for each group (data not shown). Immobility time in the TST (F, DMSO n = 12, CORT n = 16) and FST (G, DMSO n = 12, CORT = 16), sucrose preference (H, DMSO n = 12, CORT = 15) 28 days after the infusion of CORT (10 µM, 2 µl) into the DG regions. Error bars denote SEM, *P<0.05, **P<0.01, two-tailed Student's t test.

Figure 6. MR-nNOS pathway mediates the different roles of glucocorticoids in the hippocampus and hypothalamus in regulating HPA axis activity.

(A) Hippocampal and hypothalamic CORT concentrations of adult mice under normal condition (control) or exposed to 28 days CMS. CMS led to a similar elevation of CORT concentration both in the hippocampus (n = 5 for each group) and hypothalamus (n = 4 for each group). (B) Representative photos of MR and GR immunofluorescence in the hippocampus and hypothalamus. The immunofluorescence was repeated at least in 3 mice. (C) Western blotting showing GR, MR protein expression in the hippocampus and hypothalamus under basic condition or under acute stressful condition. n = 3 for each group. (D) Western blotting showing CRF protein expression in the hypothalamus 28 days after microinjection of CORT with or without spironolactone, aldosterone, mifepristone into the DG. n = 4 for each group. (E–G) Western blotting showing nNOS, MR, GR, and nitrotyosine expression in the hippocampus 7 days (E) or 28 days (F) after infusion of CORT with or without spironolactone, aldosterone, mifepristone into the hippocampus. 7days: n = 4 for each group; 28days: n = 3 for each group. (H) Western blotting showing nNOS, MR, GR, and nitrotyrosine protein expression in the hypothalamus 7 days or 28 days after infusion of CORT into the hypothalamus. n = 3 for each group. Arrow heads indicate the location of nearest band in the ladder. Error bars denote SEM, *P<0.05, **P<0.01, ***P<0.001, compared to DMSO; ^#P<0.05, ^##P<0.05, compared to CORT. One-way ANOVA in (D, E, F, G), two-way ANOVA in (C, H).

Figure 7. NO mediates the different roles of glucocorticoids in hippocampus and hypothalamus in regulating HPA axis activity.

Representative imaging of nNOS-positive cells in the hippocampus (left column) and in the PVN regions of the hypothalamus (right column) after 28 days CORT (40 mg/kg, s.c.) or DMSO treatment (A) and the number of nNOS-positive cells (B), n = 4 in DMSO group, n = 5 in CORT group. (C) The concentration of NO in the hippocampus or hypothalamus after 28 days CORT treatment (40 mg/kg, s.c.), n = 4 in DMSO group, n = 5 in CORT group. (D) Western blotting showing CRF levels in the hypothalamus at day 28 of CORT treatment with or without CPTIO or ODQ, n = 4 for each group. CORT/DMSO represented administration of CORT via s.c. plus infusion of DMSO into bilateral DG of the hippocampus, and so on. (E) Western blotting showing CRF levels in the hypothalamus 28 days after infusion of DETA/NONOate, n = 4 for each group. Infusion of DETA/NONOate into the hippocampus increased CRF expression in the hypothalamus (left). Infusion of DETA/NONOate into the PVN regions of hypothalamus did not change the CRF expression in the hypothalamus (right). Arrow heads indicate the location of nearest band in the ladder. (F) Representative imaging of CRF-positive cells (red) and DAPI-labeled cells (blue) in the hypothalamus 28 days after infusion of DETA/NONOate into the hippocampus (upper row) or hypothalamus (bottom row). Note that the CRF signal in the PVN region of the hypothalamus in mice received DETA/NONOate infusion into the hippocampus was stronger. All imaging represent 4 individual mice. Error bars denote SEM, *P<0.05, **P<0.01, ***P<0.001 compared to control group, ^#<0.05, ^##P<0.01, compared to CORT/DMSO group, two-tailed Student's t test in (B, C, E), one-way ANOVA in (D).

Figure 8. A model showing glucocorticoids plays different roles in the hippocampus and hypothalamus in modulating HPA axis activity under stressful state.

(A) An overview of the design to compare the roles of glucocorticoids in the hippocampus and hypothalamus in HPA axis hyperactivity and depressive-like behaviors. (B) A model describing the different roles of glucocorticoids in the hippocampus and hypothalamus in depression. The Acute stress stimulates glucocorticoids synthesis and releasing in adrenal cortex. Under acute stressful state, when glucocorticoids arriving in the hippocampus or hypothalamus, glucocorticoids exerts negative regulation of the synthesis of CRF in PVN neurons by GR. Under chronic stressful state, glucocorticoids in the hippocampus impair the negative feedback modulation of the synthesis of CRF in PVN neurons in the hypothalamus by disrupting hippocampal GR. However, glucocorticoids in the hypothalamus still exert negative regulation on the synthesis of CRF in PVN neurons. Glucocorticoids in the hippocampus disrupt GR function through MR-nNOS-NO pathway. (−) exerting negative regulation of HPA axis activity; (+) exerting positive regulation of HPA axis activity.