Activation of the Lateral Habenula-Ventral Tegmental Area Neural Circuit Contributes to Postoperative Cognitive Dysfunction in Mice

Abstract

Postoperative cognitive dysfunction (POCD) is common and is associated with poor outcome. Neural circuit involvement in POCD is unknown. Lateral habenula (LHb) that regulates coping and depression-like behaviors after aversive stimuli is activated by surgery in the previous study. Here, surgery activated LHb and ventral tegmental area (VTA) are presented. VTA is known to receive projections from LHb and project to the prefrontal cortex and hippocampus. Direct chemogenetic inhibition of LHb or damaging LHb attenuates surgery-induced learning and memory impairment, N-methyl-d-aspartate (NMDA) receptor activation, endoplasmic reticulum stress, inflammatory responses and cell injury in the VTA, and activation of rostromedial tegmental nucleus, an intermediate station to connect LHb with VTA. LHb inhibition preserves dendritic spine density in the prefrontal cortex and hippocampus. Retrograde inhibition of LHb via its projections to VTA attenuated surgery-induced learning and memory dysfunction is observed. Retrograde activation of LHb induced learning and memory dysfunction is observed. Inhibition of NMDA receptors, dopamine synthesis, and endoplasmic reticulum stress in the VTA reduced surgery-induced learning and memory impairment, inflammatory responses, and cell injury are observed. These results suggest that surgery activates the LHb-VTA neural circuit, which contributes to POCD and neuropathological changes in the brain. These novel findings represent initial evidence for neural circuit involvement in surgery effects.

Keywords: chemogenetic inhibition; endoplasmic reticulum stress; lateral habenula; postoperative cognitive dysfunction; ventral tegmental area.

Figure 1

Surgery impaired learning and memory in mice. a,b) Performance in the open field test. c) Performance in the novel object recognition test. d) Performance in the training sessions of Barnes maze test. e) Performance in the memory phase of Barnes maze test. f) Performance in the context‐ and tone‐related fear conditioning test. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 15). Results were analyzed by one‐way or two‐way repeated measures ANOVA (panel d) and t‐test (all other panels). *P < 0.05 for the comparison, **P < 0.01 for the comparison, ^P < 0.05 compared with the corresponding data on day 1.

Figure 2

Activation of the LHb‐VTA circuit of mice with surgery and effectiveness of inhibition of this activation by a DREADDs approach. a) Left panel: Representative immunofluorescent images of c‐Fos staining (green) at 3 h (S3h) after surgery. Scale bar: 10 µm. Right panel: Quantitative data of the number of c‐Fos positive cells. b,c) Expression of c‐Fos in the LHb and VTA, respectively, at 3 h (S3h), 24 h (S24h), 48 h (S48h), or 72 h (S72h) after operation. d,e) Number of c‐Fos positive cells in the LHb and VTA, respectively, at 3 h after surgery. Left panel: Representative immunofluorescent images of c‐Fos staining. Scale bar: 100 µm. Right panel: quantitative data. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 5 for panels (a–c) = 6 for panels (d,e). Results were analyzed by t‐test (panel a) and one‐way ANOVA followed with Tukey test (all other panels). * P < 0.05 for the comparison, ** P < 0.01 for the comparison.

Figure 3

DREADDs‐based inhibition of the LHb‐VTA circuit attenuated learning and memory decline in mice with surgery. a,b) Performance in the open field test. c) Performance in the novel object recognition test. d) Performance in the training sessions of Barnes maze test. e) Performance in the memory phase of Barnes maze test. f) Performance in the context‐ and tone‐related fear conditioning test. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 15). Results were analyzed by one‐way and two‐way repeated measures ANOVA (panel (d)) and one‐way ANOVA followed with Tukey test (all other panels). *P < 0.05 for the comparison, ** P < 0.01 for the comparison, ^P < 0.05 compared with the corresponding data on day 1.

Figure 4

Retrograde inhibition of LHb neurons via the projections from LHb to VTA attenuated surgery‐induced learning and memory impairment. a) Top: schematic presentation of viral injections. Bottom: a) representative image of hM4Di‐mCherry expression in the LHb. Scale bar: 500 µm. b) Representative images of hM4Di‐transduced neurons and c‐Fos expression after surgery with intraperitoneal injection of saline (top panels, scale bar: 100 µm) and magnified images of a selected area shown in the top right image (scale bar: 10 µm) in the LHb. c,d) Performance in the open field test. e) Performance in the novel object recognition test. f) Performance in the training sessions of Barnes maze test. g) Performance in the memory phase of Barnes maze test. h) Performance in the context‐ and tone‐related fear conditioning test. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 15). Results were analyzed by one‐way and two‐way repeated measures ANOVA (panel (f)) and one‐way ANOVA followed with Tukey test (all other panels). *P < 0.05 for the comparison, **P < 0.01 for the comparison, ^P < 0.05 compared with the corresponding data on day 1.

Figure 5

Chemical lesion of the LHb attenuated cognitive dysfunction in mice with surgery. a) Representative images of Nissl staining showing structural damage in the LHb. Scale bar: 100 µm. b) Left panel: Representative immunofluorescent images of c‐Fos in the VTA at 3 h after surgery. Scale bar: 100 µm. Right panel: Quantitative results of c‐Fos positive cells. c,d) Performance in the open field test. e) Performance in the novel object recognition test. f) Performance in the training sessions of Barnes maze test. g) Performance in the memory phase of Barnes maze test. h) Performance in the context‐ and tone‐related fear conditioning test. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 6 for panel (b), = 15 for all other panels). Results were analyzed by one‐way and two‐way repeated measures ANOVA (panel (f)) and one‐way ANOVA followed with Tukey test (all other panels). *P < 0.05 for the comparison, **P < 0.01 for the comparison, ^P < 0.05 compared with the corresponding data on day 1. Hip: hippocampus; IBO: ibotenic acid; MHb: medial habenular nucleus; PV: paraventricular thalamic nucleus; TH: tyrosine hydrolase.

Figure 6

Chemogenetic activation of LHb neurons via the projections from LHb to VTA impaired learning and memory. a) Top: schematic presentation of the viral injections. Bottom: representative images of hM3Dq‐mCherry expression in the LHb. Scale bar: 500 µm). b,c) Performance in the open field test. d) Performance in the novel object recognition test. e) Performance in the training sessions of Barnes maze test. f) Performance in the memory phase of Barnes maze. g) Performance in the context‐ and tone‐related fear conditioning test. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 15). Results were analyzed by one‐way and two‐way repeated measures ANOVA (panel (e)) and one‐way ANOVA followed with Tukey test (all other panels). *P < 0.05 for the comparison, **P < 0.01 for the comparison, ^P < 0.05 compared with the corresponding data on day 1.

Figure 7

Inhibition of the LHb‐VTA circuit reduced NMDA receptor activation and endoplasmic reticulum stress in mice with surgery. a) Expression of p‐NR1 in the VTA at 24 h (S24h), 48 h (S48h), or 72 h (S72h) after surgery. Left panel: Representative images of Western blots. Right panel: Quantitative results. b,c) Expression of p‐NR1, XBP1s and CHOP in the VTA at 24 h after surgery. Left panel: Representative images of Western blots. Right panel: Quantitative results. d) Concentrations of IL‐1β and IL‐6 in the VTA at 24 h after surgery. e) Expression of caspase‐3 and cleaved caspase‐3 in the VTA at 24 h after surgery. Top panel: Representative images of Western blots. Bottom panel: Quantitative results. f) Number of TUNEL positive cells in the VTA at 24 h after surgery. Left panel: Representative images of TUNEL staining. Scale bar: 100 µm. Right panel: Quantitative data. g) Number of dopaminergic neurons in the VTA at 20 days after surgery. Left panel: Representative images of staining for tyrosine hydrolase (TH). Scale bar: 400 µm (left column) and 100 µm (right column). Right panel: Quantitative data of TH‐positive cells. h) Dendritic spine density in the prefrontal cortex (PFC) and hippocampus (HIP) at 20 days after surgery. Left panel: Representative images of Golgi staining. Scale bar: 5 µm. Right panel: Quantitative data. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 5 for panel (a), = 6 for all other panels). Results were analyzed by one‐way ANOVA followed with Tukey test. * P < 0.05 for the comparison, ** P < 0.01 for the comparison.

Figure 8

NMDA receptor inhibition by MK‐801 attenuated endoplasmic reticulum stress and neuropathological changes in mice with surgery. a,b) Expression of p‐NR1, XBP1s and CHOP in the VTA at 24 h after surgery. Left panel: Representative images of Western blots. Right panel: Quantitative results. c) Concentrations of IL‐1β and IL‐6 in the VTA at 24 h after surgery. d) Expression of caspase‐3 and cleaved caspase‐3 in the VTA at 24 h after surgery. Left panel: Representative images of Western blots. Right panel: Quantitative results. e) Number of TUNEL positive cells in the VTA at 24 h after surgery. Left panel: Representative images of TUNEL staining. Scale bar: 100 µm. Right panel: Quantitative data. f) Number of dopaminergic neurons in the VTA at 20 days after surgery. Left panel: Representative images of staining for tyrosine hydrolase (TH). Scale bar: 400 µm (left column) and 100 µm (right column). Right panel: Quantitative data of TH positive cells. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 6). Results were analyzed by one‐way ANOVA followed with Tukey test. * P < 0.05 for the comparison, ** P < 0.01 for the comparison.

Figure 9

NMDA receptor inhibition by MK‐801 attenuated postoperative cognitive decline in mice. a,b) Performance in the open field test. c) Performance in the novel object recognition test. d) Performance in the training sessions of Barnes maze test. e) Performance in the memory phase of Barnes maze test. f) Performance in the context‐ and tone‐related fear conditioning test. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 15). Results were analyzed by one‐way and two‐way repeated measures ANOVA (panel (d)) and one‐way ANOVA followed with Tukey test (all other panels). *P < 0.05 for the comparison, **P < 0.01 for the comparison, ^P < 0.05 compared with the corresponding data on day 1.

Figure 10

Tyrosine hydroxylase inhibition by MIT mitigated learning and memory decline in mice with surgery. a,b) Performance in the open field test. c) Performance in the novel object recognition. d) Performance in the training sessions of Barnes maze. e) Performance in the memory phase of Barnes maze. f) Performance in the context‐ and tone‐related fear conditioning test. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 15). Results were analyzed by one‐way and two‐way repeated measures ANOVA (panel (d)) and one‐way ANOVA followed with Tukey test (all other panels). *P < 0.05 for the comparison, **P < 0.01 for the comparison, ^P < 0.05 compared with the corresponding data on day 1. MIT: 3‐Iodo‐L‐tyrosine.

Figure 11

Endoplasmic reticulum stress inhibition by TUDCA reduced neuropathological changes in mice with surgery. a) Expression of XBP1s and CHOP in the VTA at 24 h after surgery. Left panel: Representative images of Western blots. Right panel: Quantitative results. b) Concentrations of IL‐1β and IL‐6 in the VTA at 24 h after surgery. c) Expression of caspase‐3 and cleaved caspase‐3 in the VTA at 24 h after surgery. Left panel: Representative images of Western blots. Right panel: Quantitative results. d) Number of TUNEL positive cells in the VTA at 24 h after surgery. Left panel: Representative images of TUNEL staining. Scale bar: 100 µm. Right panel: Quantitative data. e) Number of dopaminergic neurons in the VTA at 20 days after surgery. Left panel: Representative images of staining for tyrosine hydrolase (TH). Scale bar: 400 µm (left column) and 100 µm (right column). Right panel: Quantitative data of TH positive cells. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 6). Results were analyzed by one‐way ANOVA followed with Tukey test. ** P < 0.01 for the comparison.

Figure 12

Endoplasmic reticulum stress inhibition by TUDCA attenuated learning and memory decline in mice with surgery. a,b) Performance in the open field test. c) Performance in the novel object recognition test. d) Performance in the training sessions of Barnes maze test. e) Performance in the memory phase of Barnes maze test. f) Performance in the context‐ and tone‐related fear conditioning test. Data are presented as mean ± S.D. with the presentation of data of each individual animal (n = 15). Results were analyzed by one‐way and two‐way repeated measures ANOVA (panel d) and one‐way ANOVA followed with Tukey test (all other panels). **P < 0.01 for the comparison, ^P < 0.05 compared with the corresponding data on day 1.

Figure 13

Diagram of possible neural circuits and molecules for postoperative cognitive dysfunction. PFC: prefrontal cortex, HIP: hippocampus, LHb: Lateral habenula, CHOP: CCAAT‐enhancer‐binding protein homologous protein, DA: Dopamine, ER: endoplasmic reticulum, NMDAR: N‐methyl‐d‐aspartate receptors, POCD: postoperative cognitive dysfunction, VTA: ventral tegmental area, XBP1s: X‐box binding protein 1.