Animals

This research endeavor obtained approval from the Animal Care and Use Committees of Zunyi Medical University and adhered to the guidelines outlined in the Guide for the care and use of laboratory animals in China (No. 14924, 2001). The VGAT-IRES-Cre mouse strains were acquired from the Jackson Laboratory (JAX Mice and Services), while the C57BL/6J strain was procured from Changsha Tianqin Technology Co., Ltd (Changsha, China). Mice were housed in standard facilities following a 12 h light-dark cycle, with continuous access to food and water. To mitigate potential confounding influences of circadian rhythm, all behavioral assessments and EEG experiments were conducted between 10:00 a.m. and 4:00 p.m.

AAV Vectors and Drugs

The viruses utilized in this study were sourced from Brain-VTA (China), encompassing Adeno-associated viruses (AAVs) expressing GCaMP (rAAV-Eflα-DIO-Gcamp6s-WPRE-pA, PT-0091), optogenetic tools (rAAV-Eflα-DIO-hChR2-mCherry-WPRE-pA, PT-0002; rAAV-Eflα-DIO-eNpHR-mCherry-WPRE-pA, PT-0007), mCherry (rAAV-Eflα-DIO-mCherry-WPRE-pA, PT-0013), and diphtheria toxin A (DTA) (rAAV-Eflα-DIO-DTA-WPRE-pA, PT-0775). After the injection of the neurotoxic virus, no behavioral tests beyond anesthetic behavioral assessments were conducted. Isoflurane was procured from RWD Life Science (China), while pentobarbital and lidocaine were obtained from Chaohui Pharmaceutical (Shanghai, China). The rabbit anti-GABA antibody was manufactured by Sigma Corporation (USA), and CY3 Donkey anti-rabbit IgG was sourced from Abcam Corporation (USA).

Stereotaxic Surgery

Mice were anesthetized with 1.4% isoflurane, and local anesthesia was administered via subcutaneous injection of lidocaine (1%). Subsequently, the mice were positioned on a stereotaxic apparatus (RWD Life Science, Shenzhen, China). Utilizing a microsyringe pump, AAV viruses were bilaterally injected into the Zona Incerta (ZI) at coordinates of anterior-posterior [AP]: − 1.50 mm, medial-lateral [ML]: ±0.95 mm, and dorsal-ventral [DV]: − 4.50 mm through a glass micropipette (150 nL/side; infusion rate: 30 nL/min). The micropipette was retracted at least 10 min post-viral injection. For in-vivo fiber photometry recordings and optogenetic investigations, optical fibers were unilaterally implanted above the ZI at coordinates of AP: − 1.50 mm, ML: ± 0.95 mm, DV: − 4.20 mm, and anchored with three cranial screws and dental cement. Following the implantation procedure, mice were granted a minimum of 3 weeks for post-surgical recovery before the commencement of experiments.

In-vivo Fiber Photometry Recordings

In our prior investigations, a multi-channel fiber photometry system from ThinkerTech Nanjing Bioscience (Nanjing, China) was employed. The fluorescence signals emanating from GCaMP were captured utilizing multifunction data acquisition software provided by Thinker Tech Nanjing Bioscience Inc. Optical transmission between the commutator and the implanted fiber was facilitated through an optical fiber manufactured by Newton (China). Mice were initially situated within an acrylic glass chamber coupled to an isoflurane vaporizer originating from RWD Life Science (Shenzhen, China). An anesthesia monitor (Vamos; Dräger Company, Germany) was integrated with the chamber to oversee isoflurane levels. A baseline period of 5 min was observed prior to anesthesia induction. Subsequently, mice were anesthetized with 1.4% isoflurane, and the instances of loss of righting reflex (LORR) and recovery of righting reflex (RORR) were recorded, with data collection terminating 5 min post-RORR. Continuous delivery of gas for 20 min post-LORR ensured equilibration of isoflurane concentration in the brain. MATLAB 2019a (MathWorks, Cambridge, US) was utilized for fiber photometry data analysis. The values of fluorescence change (ΔF/F) were calculated via the formula: (F - F0)/F0, where F denoted the test fluorescence signal and F0 represented the mean and standard deviation of the baseline signal [14].

Behavioral Testing

The loss of righting reflex (LORR) serves as an indicator of isoflurane anesthesia induction time, while the recovery of righting reflex (RORR) time is commonly utilized as a standardized measure for assessing the emergence time from general anesthesia. To achieve selective depletion of GABAergic neurons within the Zona Incerta (ZI), VGAT -IRES-Cre mice underwent bilateral injections of rAAV-Eflα-DIO-DTA-WPRE-pA into the ZI using aseptic procedures. As depicted in Fig. 1B, the mice were acclimated in an anesthesia chamber for 5 min. Subsequently, 1.4% isoflurane along with oxygen (O2) at a rate of 1 L/min was administered into the chamber, inducing isoflurane-related unconsciousness in mice. The period from the initiation of isoflurane treatment to LORR onset was defined as the LORR duration. Post this, anesthetic maintenance was sustained for 20 min to ensure brain isoflurane concentration equilibrium. Following removal from the chamber, mice were allowed to recover from anesthesia, with the interval from the conclusion of isoflurane infusion to RORR denoting the RORR duration. Throughout the experiment, isoflurane levels within the anesthesia chamber were monitored using an anesthesia monitor (Vamos; Dräger Company, Germany). An electric blanket equipped with a rectal temperature probe was positioned at the base of the anesthesia chamber to uphold the mice’s body temperature at 37 ℃. For optogenetic investigations, rAAV-EF1a-DIO-ChR2-mCherry/rAAV-EF1a-DIO-NpHR-mCherry/rAAV-EF1a-DIO-mCherry were administered into the ZI of VGAT -IRES-Cre mice, with an optic fiber implanted into the ZI region. Optical power was fine-tuned to 5mW using an optical power meter (PM100D, Thorlabs). Light pulses of 473 nm with a width of 150 ms at 2 Hz and 589 nm with a width of 20 ms at 0.1 Hz, as outlined in prior research [15], were employed during the LORR and RORR phases, respectively. Post experimentation, mice were euthanized, and immunofluorescence assessments were conducted to confirm virus expression and specific transfection.

Fig. 1

Phase-dependent calcium alterations in ZI GABAergic neurons during isoflurane anaesthesia. A Top: Schematic diagram of fiber photometry recording during isoflurane anesthesia in freely moving mice. Bottom: Timeline for quantifying the LORR and RORR of isoflurane anesthesia. (PMT, Photomultiplier tube is a kind of detection device that can convert weak light signals into electronic signals and amplify them). B Schematic of calcium signal recording model into the ZI of a VGAT -Cre mouse. C Expression of GCaMP6f in the ZI of a VGAT -Cre mouse. Viral expression (GCaMP6s, green) in the ZI and colabeling with GABA neurons (GABA immunofluorescence, red). D Individual transitions from wakefulness to isoflurane-induced LORR with color-coded fluorescent intensities (LORR were represented the point of 0). E Average responses from the state transitions during the process of induction expressed as the mean (red) ± SEM (shaded). (F) The fluorescence calcium signals significantly reduced after isoflurane-induced unconsciousness (The baseline (wake: − 200 to − 100 s) vs. anaesthesia period (100 to 200 s), P = 0.0475; n = 8), the paired Student’s t-tests). G Individual transitions from isoflurane anesthesia state to arousal with color-coded fluorescent intensities (RORR were represented the point of 0). H Average responses from the state transitions during the process of recovery expressed as the mean (red) ± SEM (shaded). I The fluorescence calcium signals ascended after RORR (The baseline (anaesthesia: − 200 to − 150 s) vs. early emergence period (0 to 100 s); P = 0.0055; The baseline: (anaesthesia: − 200 to − 150s) vs. emergence period (100 to 200 s), P = 0.0026; the paired Student’s t-tests; n = 8, *P < 0.05, **P < 0.01)

Histological Localization of Cannula Position and Immunohistochemistry

All mice were euthanized under deep anesthesia induced by 2% isoflurane and lidocaine (2%). Subsequent to achieving deep anesthesia, transcranial infusion of 200 mL of 0.01 M PBS followed by 50 mL of 4% PFA was conducted. Post infusion, the brains were promptly extracted for histological examination. The brains were coronally sectioned into 30 µm frozen sections using a cryostat (CM1950; Leica, Germany) to pinpoint the injection virus localization in accordance with the Paxinos & Franklin mouse brain atlas. GABAergic neurons within the Zona Incerta (ZI) were subject to immunohistochemical staining with anti-GABA antibody generated in rabbit (A2052, Sigma, USA). For lesion experiments, the areas of GABA fluorescence in the ZI were quantified blindly by evaluating positively immunostained neurons in ImageJ (USA). GABA-positive neurons were tallied within a 0.5 mm x 0.5 mm region. The proportions of GABA fluorescence area were determined across three adjacent sections (each spaced 90 μm apart) of the brain, with the mean count per section serving as the data representation. All imaging was performed via the Olympus BX63 virtual microscopy system.

EEG Recording and Analysis

EEG acquisition was synchronized with the behavioral testing procedures. The EEG signals were acquired and filtered within the frequency range of 0.1 to 300 Hz. Relative powers in specific frequency bands (δ: 1–4 Hz, θ: 4–8 Hz, α: 8–12 Hz, β: 12–25 Hz, and γ: 25–60 Hz) were calculated by averaging the signal power across each band’s frequency spectrum and then normalizing it by the total power within the 1–60 Hz range, consistent with methodologies detailed in prior studies [2, 16].Spectrograms were generated through bandpass filtration spanning from 1 to 60 Hz, utilizing multitaper techniques implemented in MATLAB 2016a (R2016a; MathWorks).An initial 5-minute EEG signal recording session was conducted prior to induction. Subsequently, continuous EEG signal recording was maintained throughout the lesion experiments, covering the stages of induction, maintenance, and recovery. Following this, EEG signals were continuously captured starting from 5 min after the mice regained consciousness from isoflurane anesthesia. Power spectrum analysis, as illustrated in Fig. 2C, was carried out on data obtained during the induction and recovery phases under isoflurane anesthesia within the framework of optogenetic experiments. The burst suppression rate (BSR) analysis during the recovery phase was performed using MATLAB 2016a to reduce redundancy and enhance technical terminology clarity.

Fig. 2

Lesion of ZI GABAergic neurons delayed arousal from isoflurane anesthesia. A Schematic showing bilateral injection of Cre-independent AAVs into the ZI. B Image showing GABA staining from a mouse with specific ZI lesion using AAV-CAG-DIO-DTA (NS represents normal saline). C Proportions of GABA-fluorescent area in ZI, indicating that the lesion group animals were selectively ablated ZI GABAergic neurons (P = 0.0001 by independent-samples t-test; n = 6 per group). D Protocol for behavioral and electroencephalogram (EEG) recording of induction time and emergence time. Spectrograms of EEG power during the isoflurane anaesthesia period in the control group. E BSR at recovery process of isoflurane in control group and lesion group (P = 0.0001 by independent-samples t-test; n = 6 per group). F The induction time in control group and lesion group during the maintenance of 1.4% isoflurane anesthesia. G Lesion of ZI GABAergic neurons displaying no significant change in the power ratios of the EEG. Spectrograms of EEG power during the isoflurane anaesthesia period in lesion group (H) and the control group (K). I Lesion of ZI GABAergic neurons peolonged the recovery time from 1.4% isoflurane anesthesia. (J) During the recovery time, lesion of ZI GABAergic neurons significantly altered the power ratios of the EEG. (δ band: lesion group vs. control group; P = 0.0003 by independent-samples t-test;β band: lesion group vs. control group; P = 0.018 by independent-samples t-test, γ band: lesion group vs. control group; P = 0.01 by independent-samples t-test; n = 6 per group)

Data Analysis

Statistical analyses were carried out using the GraphPad Prism software package (version 6.0; GraphPad Software Inc., San Diego, CA, USA). Normality tests were applied to all datasets. Paired Student’s t-tests were utilized to assess differences in calcium signals pre- and post-events, LORR and RORR durations, and changes in burst suppression rate (BSR) within groups (ChR2-light-on vs. ChR2-light-off or NpHR-light-on vs. NpHR-light-off). Independent-samples t-tests were employed to compare proportions of GABA-fluorescent area, LORR and RORR times, as well as EEG power between lesion and sham groups. Additionally, independent-samples t-tests were conducted to compare LORR and RORR times and alterations in EEG power bands in optogenetic experiments between mcherry-light-on vs. ChR2-light-on or mcherry-light-on vs. NpHR-light-on groups.Data were represented as mean ± standard deviation (SD) or mean ± standard error of the mean (SEM). Statistical significance was established at p < 0.05 for all analyses.