Prenatal stress modulates HPA axis homeostasis of offspring through dentate TERT independently of glucocorticoids receptor

Abstract

In response to stressful events, the hypothalamic-pituitary-adrenal (HPA) axis is activated, and consequently glucocorticoids are released by the adrenal gland into the blood circulation. A large body of research has illustrated that excessive glucocorticoids in the hippocampus exerts negative feedback regulation of the HPA axis through glucocorticoid receptor (GR), which is critical for the homeostasis of the HPA axis. Maternal prenatal stress causes dysfunction of the HPA axis feedback mechanism in their offspring in adulthood. Here we report that telomerase reverse transcriptase (TERT) gene knockout causes hyperactivity of the HPA axis without hippocampal GR deficiency. We found that the level of TERT in the dentate gyrus (DG) of the hippocampus during the developmental stage determines the responses of the HPA axis to stressful events in adulthood through modulating the excitability of the dentate granular cells (DGCs) rather than the expression of GR. Our study also suggests that the prenatal high level of glucocorticoids exposure-induced hypomethylation at Chr13:73764526 in the first exon of mouse Tert gene accounted for TERT deficiency in the DG and HPA axis abnormality in the adult offspring. This study reveals a novel GR-independent mechanism underlying prenatal stress-associated HPA axis impairment, providing a new angle for understanding the mechanisms for maintaining HPA axis homeostasis.

Fig. 1. Hyperactivity of the HPA axis in Tert knockout mice without GR and MR deficiency.

a RT-PCR showing the levels of CRF and GAPDH mRNA in the hypothalamus of Tert^−/− and WT mice. n = 4. Student’s t test. b Western blot showing the levels of CRF and GAPDH protein in the hypothalamus of Tert^−/− and WT mice. n = 5. Student’s t test. c, d Experimental design and representative images showing GFP⁺ cells in the PVN of Crf-Cre; Tert^+/+, Crf-Cre; Tert^+/−, or Crf-Cre; Tert^−/− mice. n = 3. e Immunofluorescence showing the number of CRF⁺ cells in the PVN of Tert^−/− and WT mice. n = 4. Student’s t test. f The concentrations of CORT in the plasma of Tert^−/− and WT mice treatment with DEX or vehicle. n = 5. Two-way ANOVA. g The concentrations of ACTH in the plasma of Tert^−/− and WT mice treatment with DEX or vehicle. n = 5. Two-way ANOVA. h, i RT-PCR showing the levels of GR, MR, and GAPDH mRNA in the HP, PC, HT, and AG of Tert^−/− and WT mice. n = 5. j, k Western blot showing the levels of GR, MR, and GAPDH protein in the HP, PC, HT, and AG of Tert^−/− and WT mice. n = 5. HP Hippocampus, PC Prefrontal cortex, HT Hypothalamus, AG Amygdala. *P < 0.05, **P < 0.01, and ***P < 0.001, ns indicates no significant difference. Error bars indicate s.e.m.

Fig. 2. The level of dentate TERT determines the HPA axis activity without relevance to GR and MR.

Western blot showing the levels of CRF and GAPDH in the hypothalamus (a) and the concentration of CORT in the plasma (b), of 5-week-old Tert^−/− and WT mice received microinjection of 1 μl of RV-TERT-EGFP or RV-EGFP into the bilateral DGs. Two months after virus injection, the samples for western blot were prepared and the concentration of CORT in the plasma was measured. n = 4–8. One-way ANOVA. The concentration of CORT (c) and ACTH (d) in the plasma, and the number of CRF cells (e) in the PVN of 5-week-old Tert^−/− mice received injection of 1 μl of RV-TERT-EGFP or RV-EGFP into the bilateral DGs. n = 5. One-way ANOVA. f In another cohort of 5-week-old Tert^−/− and WT mice, 1 μl of RV-TERT-EGFP or RV-EGFP was injected into the DGs. Two months after injection, mice were exposed to 1 h-restrain stress and the plasma were collected for CORT measurement at 0.5, 1, and 5 hours after the restrain stress exposure. n = 5–6. Two-way ANOVA. g Western blot showing the level of GR, MR, and GAPDH in the DG. The samples were prepared from mice in a. One-way ANOVA. h The concentration of CORT in the plasma of Tert^−/− and WT mice at different time points after sexual behavior. *P < 0.05, **P < 0.01, and ***P < 0.001, ns indicates no significant difference. Error bars indicate s.e.m.

Fig. 3. Telomerase catalytic activity is involved in the modulation of the HPA axis activity.

The number of c-FOS⁺ cells in the PVN (a), the expression of CRF and GAPDH in the hypothalamus (b), and the concentration of CORT in the plasma (c), in C57BL/6 mice 2 months after injection of 1 μl of shTERT or shNT into the bilateral DGs. n = 4–6. Student’s t test. The number of c-FOS⁺ cells in the PVN (d), the expression of CRF and GAPDH in the hypothalamus (e), and the concentration of CORT in the plasma (f), in mice 2 months after injection of 1 μl of LV-TERT-EGFP or LV-EGFP into the bilateral DGs. n = 5. Student’s t test. The expression of GR in the DG 2 months after shTERT (g) or LV-TERT-EGFP infection (h). n = 4–6. Student’s t test. The number of c-FOS⁺ cells in the PVN (i) and the concentration of CORT in the plasma (j) in mice 2 months after injection of 1 μl of LV-TERTΔ-EGFP or LV-EGFP into the bilateral DGs. n = 6. Student’s t test. The number of c-FOS⁺ cells in the PVN (k), the expression of CRF and GAPDH in the hypothalamus (l), the concentration of CORT in the plasma (m), and the expression of GR in the DG (n), in mice 2 months after injection of 1 μl of RV-TERT-EGFP or RV-EGFP into the bilateral DGs. n = 5. Student’s t test. *P < 0.05, **P < 0.01, and ***P < 0.001, ns indicates no significant difference. Error bars indicate s.e.m.

Fig. 4. Hypo-excitability of DGCs following TERT deficiency impairs negative feedback regulation of the HPA axis.

a Representative neuronal firing of DGCs of Tert^−/− and WT mice measured by whole-cell current clamp recording. The same results were observed in 15 individual DGCs from 3 mice in each group. Student’s t test. Experimental design (b) and analytical data showing the number of c-FOS⁺ cells in the DG (c) and the concentration of CORT in the plasma (d) of Tert^−/− and WT mice received microinjection of AAV-mCherry or AAV-hM3Dq-mCherry into the DGs. n = 7–8. Two-way ANOVA. Schematic experimental design and experimental paradigm for g–h. Analytical data showing the number of c-FOS⁺ cells in the DG (g) and the concentration of CORT in the plasma (h). *P < 0.05, **P < 0.01, ***P < 0.001, ns indicates no significant difference. Error bars indicate s.e.m.

Fig. 5. Prenatal stress-induced dentate TERT deficiency leads to dysfunctional HPA axis in adult offspring.

a The mRNA levels of TERT, GR, and GAPDH in the hippocampus of newborn mice experienced prenatal stress. n = 3-4. One-way ANOVA. b The mRNA levels of TERT, GR, and GAPDH in the DG of 3-month-old offspring experienced prenatal stress. n = 4–5. One-way ANOVA. c The concentration of CORT in the plasma of 3-month-old offspring experienced prenatal stress. n = 3–6. One-way ANOVA. d CpG sites in the promotor and exons of mouse Tert gene. e Methylation levels of Chr13:73764379 and Chr13:73764526 in the DG of 3-month-old offspring experienced prenatal stress. n = 5. One-way ANOVA, Kruskal-Wallis test. f Luciferase experiments were performed in 293T cells transfected with control plasmid pTK, plasmid pTERT, or plasmid pTERT379mut. n = 4. Two-way ANOVA. Schematic design of the construction of AD-TERT526mut-GFP, and the TERT mRNA level (g) and activity (h) in the cultured hippocampal NSCs cells infected with AD-GFP, AD-TERT-GFP, or AD-TERT526mut-GFP. n = 4–6. One-way ANOVA. i The methylation level of Chr13:73764526 and the expression of TERT in the DG after selectively editing of DNA methylation of Chr13:73764379. A volume of 1 μl of LV-Tert526-SV40-DNMT3A or LV-Scramble-SV40-DNMT3A was microinjected into the DG of 5-week mice experienced prenatal stress. Two months later, the tests were performed. n = 5. One-way ANOVA. The mRNA level of TERT, GR, MR, and GAPDH in the DG (j, k) and the concentration of CORT in the plasma (l) of 3-month-old offspring infused LV-TERT-EGFP or LV-EGFP into the DG of 5 weeks old mice experienced prenatal stress. n = 4. One-way ANOVA. m, n Representative photos and analytical data showing the number of c-FOS⁺ cells in the DG of Tert^−/− and WT mice with or without prenatal stress. Acute stress (1-hour restraining) was performed 30 minutes before collection of the plasma and perfusion with 4% PFA. n = 5. Two-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001, ns indicates no significant difference. Error bars indicate s.e.m.