GABAB receptor activation contributes to post-surgery cognitive impairments in mice by inducing hippocampal BDNF hypermethylation

Abstract

Background: Clinical observations have shown that brain-derived neurotrophic factor (BDNF) expression is reduced in patients with impaired cognitive function after anaesthesia and surgery. Abnormal epigenetic changes may contribute to this condition. The role of the GABAergic system, which is crucial in anaesthetic mechanisms and cognitive function, requires further investigation to determine whether it affects BDNF epigenetic regulation and cognitive outcomes.

Materials and methods: Experiments employed a mouse model perioperative neurocognitive disorder (PND) that comprised laparotomy under isoflurane inhalation anaesthesia. Mice were subjected to behavioural assessments, including the Y-maze and novel object recognition tests for cognitive function and molecular analyses (quantitative polymerase chain reaction, qPCR) and methylation-specific PCR to assess BDNF expression and epigenetic modifications. GABA_B receptor deactivation was induced by a selective antagonist and changes in BDNF expression and gene methylation were measured.

Results: Aged mice in the PND model had reduced BDNF expression in hippocampal neurons. BDNF protein reduction was linked to increased methylation of the Bdnf gene, initiated by GABA_B receptor activation and hyperphosphorylation of downstream signalling proteins. Furthermore, BDNF overexpression experiments demonstrated a partial reversal of cognitive impairment in mice, supporting a causal role for BDNF methylation in PND.

Conclusions: This study emphasizes the vital interaction between the GABAergic system and the epigenetic control of Bdnf, shedding light on the mechanisms underlying anaesthesia-related cognitive impairment. Our findings suggest that inhibition of GABA_B receptor activation to reverse BDNF hypermethylation may represent a potential therapeutic strategy for alleviating PND.

Keywords: GABAergic; anaesthesia and surgery; brain-derived neurotrophic factor; epigenetic methylation; perioperative neurocognitive disorder.

Figure 1.

Laparotomy induced cognitive impairment. (a) Experimental design. (b) Results of the entries (left) and duration (right) in novel arm in Y-maze tests (n = 10). (c) Experimental design. (d) Results of discrimination index in novel object recognition (NOR) tests (n = 10). (e) The Z-score results accumulated according to results of Y-maze and NOR tests (n = 10). (f and g) The mobility tests results of mice in Y-maze (f) and open field (g) (n = 10). ns: no significant, *p < .05, **p < .01.

Figure 2.

BDNF depression in dCA1 after the laparotomy was highly relevant to the cognitive impairment. (a) Western blot tests results of BDNF expressed in dCA1 of hippocampus (n = 4). (b) The expression changes of BDNF mRNA in the dCA1 (n = 4). (c) The BDNF expressed in dCA1 was positively correlated to the Z-score. (d) The fEPSP in dCA1 by stimulated the Schaffer in LTP tests (n = 5). (e) The schematic of the virus microinjection. (f–h) The behavioural tests results in Y-maze (f) and NOR (g) (n = 10). And the Z-score according to the Y-maze and NOR results (h) (n = 10). (i) The fEPSP in dCA1 after Schaffer collateral stimulation was retested after BDNF overexpression in dCA1 (n = 6). ns: no significant, **p < .01.

Figure 3.

Laparotomy induced hypermethylation of Bdnf in dCA1. (a) The heatmap of the methylation status in CpG islands of Bdnf (n = 5). (b–g) The relative methylation changes at each CpG site in first to sixth predicted CpG islands (n = 5). *p < .05, **p < .01.

Figure 4.

Administration of a DNMTs antagonist reversed the cognition impairment induced by laparotomy. (a) The chromatin immunoprecipitation (ChIP) tests for the combination of Bdnf and DNMT3a (left) and DNMT3b (right) (n = 4). (b) The schematic of DNMTs antagonist zebularine microinjection into dCA1 (n = 4). (c–d) The BDNF (c) and mRNA (d) expression changes in dCA1 (n = 4). (e,f) The behavioural tests results in Y-maze tests (e) and NOR tests (f) (n = 10). (g) The Z-score results according to the Y-maze and NOR tests (n = 10). Veh: vehicle, Zeb: zebularine, ns: no significant, *p < .05, **p < .01.

Figure 5.

GABAergic activity related to Bdnf hypermethylation induced cognitive impairment after laparotomy. (a) The experimental design for the inhalation without laparotomy. (b–d) The results of behavioural tests in Y-maze (b) and NOR (c) (n = 10). And Z-score according to the behavioural results (d) (n = 10). (e) The BDNF expression in dCA1 in inhalation and control groups (n = 4). (f) The GAD67 expression changes in dCA1 in inhalation (ISO) and surgery (A&S) groups (n = 4). (g) The schematic of virus microinjection into dCA1 for c-Fos labelling. (h) The c-Fos expression in GABAergic neurons in dCA1 (n = 4). (i) The schematic of virus microinjection into dCA1 for GABAergic conditional knockout (CKO). (j) The relative methylation changes in dCA1 after GABAergic CKO (n = 4). (k) The BDNF expression changes in dCA1 after GABAergic CKO (n = 4). ISO: isoflurane, A&S: anaesthesia and surgery, ns: no significant, *p < .05, **p < .01.

Figure 6.

GABAergic activation induced GABA_B receptor hyperphosphorylation, which led to Bdnf hypermethylation after laparotomy. (a) The expression changes of GABA_A and GABA_B receptors in phosphorylation status in dCA1 (n = 4). (b) The schematic diagram of bicuculline microinjection into dCA1. (c) The BDNF changes in dCA1 after the vehicle (Veh) or bicuculline (Bic) administration in surgery mice (n = 4). (d) The relative changes of Bdnf methylation in dCA1 after GABA_A receptor antagonist bicuculline was microinjected into dCA1 of surgery mice (n = 4). (e) The schematic diagram of GABA_B receptor antagonist saclofen microinjection into dCA1. (f) The BDNF changes in dCA1 after the vehicle (Veh) or saclofen (Sac) administration in surgery mice (n = 4). (g) Microinjection of saclofen into dCA1 of surgery mice influenced the Bdnf methylation changes (n = 4). Bic: bicuculline, Sac: saclofen, Veh: vehicle. ns: no significant, *p < .05, **p < .01.