GABAergic Neurons in the Nucleus Accumbens are Involved in the General Anesthesia Effect of Propofol

The mechanism underlying the hypnosis effect of propofol is still not fully understood. In essence, the nucleus accumbens (NAc) is crucial for regulating wakefulness and may be directly engaged in the principle of general anesthesia. However, the role of NAc in the process of propofol-induced anesthesia is still unknown. We used immunofluorescence, western blotting, and patch-clamp to access the activities of NAc GABAergic neurons during propofol anesthesia, and then we utilized chemogenetic and optogenetic methods to explore the role of NAc GABAergic neurons in regulating propofol-induced general anesthesia states. Moreover, we also conducted behavioral tests to analyze anesthetic induction and emergence. We found out that c-Fos expression was considerably dropped in NAc GABAergic neurons after propofol injection. Meanwhile, patch-clamp recording of brain slices showed that firing frequency induced by step currents in NAc GABAergic neurons significantly decreased after propofol perfusion. Notably, chemically selective stimulation of NAc GABAergic neurons during propofol anesthesia lowered propofol sensitivity, prolonged the induction of propofol anesthesia, and facilitated recovery; the inhibition of NAc GABAergic neurons exerted opposite effects. Furthermore, optogenetic activation of NAc GABAergic neurons promoted emergence whereas the result of optogenetic inhibition was the opposite. Our results demonstrate that NAc GABAergic neurons modulate propofol anesthesia induction and emergence.


Introduction
General anesthesia is a reversible condition of unconsciousness induced by anesthetic medications [1], which have been commonly used in clinical surgery for almost 200 years [2]. Particularly, the mechanism by which general anesthetic drugs unleash the general anesthesia effect and ulteriorly modify the state of consciousness remains unidentified [3]. When general anesthetic drugs are applied for operations, patients frequently have adverse reactions such as low blood pressure, respiratory depression, delayed waking, and even death [4]. As a result, delineating the identities of involved neurons and neural circuitry will advance our basic understanding and clinical applications of propofol to improve anesthetic safety and comfort.
Propofol, a common anesthetic agent in clinics [5], is being studied intensively to better understand the general anesthesia process. Numerous investigations have clarified that multiple brain biochemical targets and neuron projections participated in the concept [6]. Furthermore, propofol anesthesia is intimately linked to the sleep-arousal circuit [3,7], including GABAergic neurons of the tuberomammillary nucleus (TMN) [8], GABAergic neurons of the thalamic reticular nucleus (TRN) [9], coherent α oscillations engaging thalamus and cortex [10], and cholinergic and glutamatergic neurons in the basal forebrain (BF) [11]. The nucleus accumbens (NAc), located at the confluence of the basal nucleus and limbic system, is the major structure of the ventral striatum [12]. NAc GABAergic neurons also called Jing Yan, Bei-Ning Hang, Lin-Hui Ma and Jia-Tao Lin contributed equally to this work. 1 3 medium spiny neurons, constitute 96% of the neurons in the area [13]. Previous researches have found that it is closely linked to exercise, motivation, reward, and addiction [14,15], all of which are accomplished based on wakefulness. Recent studies have revealed especially that NAc is critical for sleep-wake regulation and participates in sleep rhythm generation [16] and sleep stability [6,17,18]. Luo et al. reported that optical activation of NAc D 1 R neurons sped up the transition from non-rapid eye movement sleep to wakefulness [19]. Likewise, inhibition of A 2A R neurons in the NAc lowered sleep duration [20]. According to related neuroimaging studies, the volume of NAc altered dramatically in individuals with disorders of consciousness (DOC) [21]. However, the precise nature of the NAc-related circuits in propofol anesthetic-induced unconsciousness has not been elucidated.
In the present study, immunofluorescence, western blotting, and patch-clamp were used to confirm the activities of NAc GABAergic neurons were altered during propofol anesthesia, and then chemogenetic and optogenetic modulation combined with EEG signals recording were used to investigate the necessity of GABAergic neurons of NAc in the change of consciousness during propofol anesthesia. Simultaneously, we also conducted behavioral tests to analyze behavioral induction and emergence. Afterward, we discussed the role of NAc GABAergic neurons played under general anesthesia. Our results provided several lines of evidence regarding NAc GABAergic neurons participating in anesthesia effect control.

Experimental Design
The experimental design of this study was shown as the schematic illustration of experimental procedure below.
a Animals in the control group and the propofol group (n = 6) were injected with intrapilid or propofol before immunofluorescence and western blotting. b Virus was microinjected into NAc of Vgat-Cre mice 28 days ago. For the in vitro electrophysiological recordings, Vgat-Cre mice brain slices (n = 15) were perfused with propofol, and pharmacological modulation of NAc GABAergic neurons was conducted. c For the chemogenetic modulation of the GABAergic neurons in NAc, Cre-dependent AAVs were injected into NAc of Vgat-Cre mice (n = 10). After the expression of the AAVs over the period of 28 days, anesthesia behavioral testing (n = 10) and anesthetic sensitivity testing (n = 10) were conducted. At the end of the experiment, mice were perfused and the fluorescent reporter was visualized in brain slices using an Olympus microscope (n = 10). d For the optogenetic modulation of the GABAergic neurons in NAc, Vgat-Cre mice (n = 8) were microinjected optical AAVs before 28 days. And then, fiber and EEG electrodes were implanted 21 days before behavioral tests. Equally to previous experiments, immunofluorescence was used to confirm the viral expression (n = 10).

Animals
Adult Vgat-cre mice (weighing 22-28g, age: 8-12 weeks) were provided by the Animal Center of Xuzhou Medical University. All mice were kept at a constant temperature of 24 ± 1°C and a 12-h light-dark cycle, with food and water available ad libitum.
To reduce the effect of the diurnal rhythm of sleep-wakefulness on experimental results, all experimental procedures were carried out during the daytime. Animal studies were authorized by the Institutional Animal Care and Use Committee of Xuzhou Medical University (IACUC no. 202205A249).

EEG Electrode and Fiber Implantation
For chemogenetic experiments, 3 weeks after AAV injections, EEG electrodes were attached directly above the mouse skull with dental cement to fix the electrodes [22].
For optogenetic experiments, optical fibers (diameter: 200 μm, length: 5 mm, NEWDOON, China) were chronically implanted into NAc 3 weeks after the injection of the virus and EEG electrodes were firmly placed in the animal skull.
The mice were approved to conduct behavioral experiments 1 to 2 weeks following EEG electrode and fiber implantation [23].

Anesthesia Behavioral Testing
Briefly, mice were acclimated in a cylindrical tank (30 cm in height, 25cm in diameter) with an EEG recording cable attached to their skulls for a 10-min adaption, and then the EEG recording started. EEGs were recorded via a sampling frequency of 1000 Hz and a bandpass set at 0-30 Hz using the Pinnacle EEG/EMG recording system (RWD Life Science, China).
For chemogenetic experiments, CNO (3 mg/kg) was administered intraperitoneally before behavioral tests. After 30 min of free exploration, mice were anesthetized (150 mg/kg, Propofol, China, ip) in the tank. When the limbs of mice were all facing up and could not be righted within 15 s twice as fast, it meant the righting reflex of mice disappeared (LORR). The duration from the onset of propofol injection to the disappearance of the righting reflex was defined as LORR time, also called time to induction. After then, the EEG was continually recorded for 20 min. When the mouse could turn independently from the supine posture but could not hold all four limbs facing up for 15 s twice as long, the righting reflex recovered (RORR). To assess the recovery time, the duration of the recovery was recorded as RORR time, also called time to emergence (n = 10).
For optogenetically behavioral testing, the mice were continuously optically stimulated (activation, 473 nm, 20 Hz, and a 10-ms duration; or inhibition, 594 nm, 20 Hz, and a 10-ms duration) every 60 s with 30-s intervals during the entire journey of the LORR time or RORR time, which was performed respectively with a 473 nm or 594 nm DPSS laser (NEWDOON, China). The final laser power was approximately 4 mW/mm −2 (n = 8).

Anesthetic Sensitivity Testing
The 50% effective dose (ED 50 ) of propofol administration could be used to verify the anesthetic sensitivity of the mice [11].
Chemogenetically, CNO (3 mg/kg) was given intraperitoneally in the experimental and control groups. After 30 min, propofol (50 mg/kg) was administered intraperitoneally to see if righting reflex had subsided. If the mice had reached LORR, placed them on the heating blanket until an emergence, and then administer propofol (25 mg/kg) as described above next time. If not, ceased the experiment and tried again with propofol (75 mg/kg) next time. The mice were subjected to repeated experiments over 5 days. Every time, the propofol injection dosage was increased or decreased by one. Besides, total propofol doses were 25 mg/kg, 50 mg/ kg, 75 mg/kg, 100 mg/kg, 125 mg/kg, 150 mg/kg, 175 mg/ kg, and 200 mg/kg (n = 10).
The percentage of LORR (Y) and the propofol (X) dose were fitted using nonlinear regression with GraphPad 8.0 to get the ED 50 for the dose-response curve.

Western Blotting
The expression of c-Fos in NAc was measured by western blotting (WB) analysis. The mice were pretreated for 1 h with intralipid/propofol (150 mg/kg) before having their brains extracted directly. After that, western blotting was conducted following the protocol described in the previous study [24]. Briefly, the proteins were transferred to PVDF membranes (Merck Millipore, ISEQ00010, USA) after separating the proteins with SDS-PAGE gels (20328ES50, YEASON). Next, blocking the membranes with BSA (3%, A8020, Solarbio) before the membranes were incubated with primary antibodies at 4°C overnight in a cold room. The antibodies used were as follows: c-Fos primary antibody (2250S, Cell Signaling Technology, 1:1000), β-actin primary antibody (AC004, ABclonal, 1:2000), and horseradish peroxidase-conjugated secondary antibodies (Beyotime, 1:2000). An ECL detection system (Beyotime) and the ImageJ software were used to quantify the protein bands (n = 6).

Immunofluorescence
Briefly, mice were perfused transcardially with 0.9% saline followed by 4% paraformaldehyde (PFA). After being postfixed in 4% paraformaldehyde for 6-8 h, the brains were placed in a 30% sucrose solution until sunk. The brain sections (30 μm) were obtained using a freezing microtome (CM1950, Leica Microsystems). To remove the OCT compound (SAKURA, USA), the sections were washed three times with PBS for 5 min each time. To inhibit nonspecific epitopes, the sections were submerged in 10% goat serum for 1 h at 37°C [23].
Image stacks on a laser scanning confocal microscope (Fluoview 1000, Olympus) were used to visualize the slices.
Pharmacological modulation of NAc GABAergic neurons: all control recordings were collected first before any anesthetic exposure. After control recordings, 10 μM propofol [26] was perfused for 3 min before collecting paired anesthetic recordings and continuously thereafter.
The collection of signals was under a MultiClamp 700B amplifier with a Digidata 1440B Data Acquisition System. Data were analyzed by using the software pClamp10.7 (Molecular Devices, Sunnyvale, USA) (n = 15).

Statistical Analysis
All of the data were presented as the mean±SEM. Comparisons between groups were shown with an unpaired t-test or two-way ANOVA followed by Tukey's post hoc test. GraphPad Prism 8.0 or MATLAB R2018a were used for statistical analysis. Statistical significance was defined as P < 0.05.

The Activity of NAc GABAergic Neurons was Suppressed by Propofol Anesthesia
Briefly, we injected the Vgat-cre mice with intralipid or propofol (Fig. 1a). Western blotting analysis demonstrated that the c-Fos expression in the propofol group was significantly lower than that in the control group (Fig. 1b, c). Following that, we used double immunofluorescence staining to assess the c-Fos expression of NAc GABAergic neurons (Fig. 1d). We discovered that the number of c-Fos + /GAD67 + in the control group was substantially greater than the number in the propofol group (Fig. 1e). In addition, electrophysiological recordings were performed on GABAergic neurons in the NAc (Fig. 1f). The results of whole-cell patch-clamp recording showed that, compared with the control group, the firing rate was decreased in the propofol group (Fig. 1h), and the resting potential was not significantly changed (Fig. 1g). Moreover, threshold current induced by step currents in NAc GABAergic neurons significantly increased in the propofol group (Fig. 1i). The considerable drop in the c-Fos expression and the increase of threshold current in NAc GABAergic neurons during anesthesia indicated that NAc GABAergic neurons are suppressed dramatically and its mechanism deserves further study.

Chemogenetic Activation of NAc GABAergic Neurons Diminished the Anesthesia Effect of Propofol
To explore the involvement of NAc GABAergic neurons during propofol anesthesia, we chemically stimulated NAc GABAergic neurons. To begin with, AAV-DIO-hM3Dq-mCherry or AAV-DIO-mCherry was injected into NAc precisely before behavioral tests (Fig. 2a, b). The hM3Dq group showed a significant rise in the c-Fos expression after CNO injection (Fig. 2c, d). During propofol anesthesia, we also recorded EEG signals for 2 min during induction and 10 min during emergence (Fig. 2e). Corresponding EEG power densities are also provided (Fig. 2f). After activation, the delta power percentage in the hM3Dq group was considerably lower than that in the mCherry group during induction (Fig. 2g, left). Seemingly, compared with the mCherry group, the delta power percentage of the hM3Dq group was lower during emergence (Fig. 2g, right).
To further illustrate these changes, the anesthesia time was recorded during the righting reflex. After activation, the duration of time to induction was substantially longer in the hM3Dq group compared with that in the mCherry group (Fig. 2h, left), whereas the duration of time to emergence was significantly shorter in the hM3Dq group compared with that in the mCherry group (Fig. 2h, right).
To identify the influence of NAc GABAergic neurons on propofol anesthesia sensitivity, different propofol doses were applied. Compared with the mCherry group, the doseresponse curve for LORR in the hM3Dq group was rightshifted, and the ED50 was significantly increased in the hM3Dq group (Fig. 2i).
These findings implied that chemically selective stimulation of NAc GABAergic neurons can prolong the induction of propofol anesthesia, facilitate propofol anesthesia recovery, and lower propofol sensitivity.

Chemogenetic Inhibition of NAc GABAergic Neurons Enhanced the Anesthesia Effect of Propofol
To further investigate the effects of inhibiting GABAergic neurons, AAV-DIO-hM4Di-mCherry or AAV-DIO-mCherry was used in the experiment and we next conducted experiments (Fig. 3a, b). With the injection of CNO, the decrease of c-Fos expression in the hM4Di group was obvious compared with that in the mCherry group (Fig. 3c, d). Representative EEG traces and the total power density of EEG were shown (Fig. 3e, f). According to the EEG signals collected, the delta power percentage in the hM4Di group was markedly larger than that in the mCherry group during LORR (Fig. 3g, left). During RORR, the delta power percentage in the hM4Di group appeared to be much higher than that in the mCherry group (Fig. 3g, right).   Additionally, the duration of time to induction was significantly shortened in the hM4Di group than that in the mCherry group after CNO injection (Fig. 3h,  left), whereas the time to the emergence was prolonged (Fig. 3h, right).
Different concentrations of propofol were given to validate the function of NAc GABAergic neurons in propofol anesthesia sensitivity. We found that the dose-response curve was left-shift and the ED 50 of propofol was lower in the hM4Di group (Fig. 3i).
These consequences suggested that the induction of propofol anesthesia can be promoted and the recovery can be prolonged through selective inhibition of NAc GABAergic neurons. Additionally, the sensitivity of propofol can be improved by selective inhibition of NAc GABAergic neurons.

Optogenetic Activation of NAc GABAergic Neurons Weakened the Anesthesia Effect of Propofol
To verify whether activation of NAc GABAergic neurons can affect the anesthesia effect during propofol anesthesia, we unilaterally injected the Cre-dependent AAV-DIO-hChR2-EYFP or AAV-DIO-EYFP into the NAc of Vgat-Cre mice (Fig. 4a, b). Immunolabeling findings indicated that the expression of c-Fos rose significantly in the ChR2 group (Fig. 4c, d) under the 20 Hz optical stimulation. Initially, we conducted behavioral tests with optogenetic stimulation (Fig. 4e). During the stimulation on LORR time, continuous stimulation of NAc GABAergic neurons induced a substantial drop in delta power (Fig. 4g, left) in the ChR2 group compared to the EYPF group, while changes were minor during RORR time (Fig. 4g, right). Subsequently, the time to induction in the ChR group was longer than that in the EYFP group (Fig. 5a, left). However, the time to emergence difference between the ChR2 and the EYFP groups was small (Fig. 5a, right).
We next gave common stimulation at 20 Hz during the RORR period (Fig. 4f). Analysis of EEG signals during emergence revealed that steady stimulation of NAc GABAergic neurons during RORR time induced a clear decrease in delta power in the ChR2 group compared with the EYPF group (Fig. 4h, right), whereas changes were small during LORR time (Fig. 4h, left). Afterward, behavioral analysis results showed that the time to induction in the ChR2 group did not differ from that in the EYFP group (Fig. 5b, left) after the stimulation. However, the time to emergence difference between the ChR2 and EYFP groups was significant (Fig. 5b, right).
These results showed that optogenetic activation of NAc GABAergic neurons was sufficient to cause cortical activity and behavioral emergence during general anesthesia.

Optogenetic Inhibition of NAc GABAergic Neurons Strengthened the Anesthesia Effect of Propofol
To inhibit the GABAergic neurons, optical fibers, and EEG electrodes were placed above NAc after the injection of AAV-DIO-eNpHR-mCherry or AAV-DIO-mCherry for 21 days (Fig. 5c). Under a 20-Hz optical stimulation, the eNpHR group showed a particular decrease in c-Fos expression (Fig. 5d-f). First and foremost, we stimulated induction for 2min optogenetically (Fig. 5g). Notably, during induction, optical stimulation on LORR revealed a significant increase in delta power of the eNpHR group than that in the mCherry group (Fig. 6a, left). However, there was no difference during RORR between the two groups (Fig. 6a, right). Additionally, the eNpHR group's LORR time was shorter than the mCherry group's after the stimulation (Fig. 6c, left). On the other hand, the RORR time in the eNpHR group and the mCherry group showed no change (Fig. 6c, right).
Furthermore, we optically stimulated the condition of emergence at 20 Hz (Fig. 5h). EEG data analysis revealed that continuous stimulation during RORR resulted in a considerable increase in delta power, but changes were small during LORR (Fig. 6b). Besides, behavioral results showed that after the stimulation, the period of RORR in the eNpHR-RORR group was much longer compared with the mCherry group (Fig. 6d,  right). However, the LORR time in the eNpHR-RORR group and the mCherry group did not differ (Fig. 6d, left).
Taken together, these results suggested that optogenetic inactivation of NAc GABAergic neurons could regulate the induction and emergence during general anesthesia.

Discussion
The purpose of this study was to explore whether NAc GABAergic neurons are involved in the loss of consciousness and recovery induced by propofol and the related mechanism.
Currently, the principle of the regulation of consciousness during general anesthesia has long been a mystery [7], and it has been ranked as one of the most pressing challenges to be addressed in the last century [27]. Intraoperative and postoperative adverse responses, such as intraoperative knowledge, postoperative delirium, convulsion, and sleeplessness, will occur if the level of anesthesia is too shallow or too deep [4]. Revealing the mechanism of general anesthetics controlling consciousness can assist to address the aforementioned unfavorable responses and then produce better anesthetics so that patients can navigate the perioperative phase more safely and easily. The mechanism of propofol-induced anesthesia has been investigated intensively. GABAergic neurons play a crucial function in general anesthesia [2,7,28]. Previous research on the role of NAc in general anesthesia has largely focused on the role of dopamine receptors [19,29] with little attention to the role of NAc GABAergic neurons in regulating consciousness. Our research fills in this gap by providing a theoretical foundation for the discovery of the principle of the general anesthetic principle and the exploration of associated neural circuits.
Here, through chemicalgenetic and optogenetic manipulation, coupled with EEG data and behavioral assessments, we found that GABAergic neurons in NAc play a key role in different phases of propofol anesthesia. During propofol anesthesia, we discovered that increasing the excitability of NAc GABAergic neurons lowered the sensitivity of mice to propofol and the depth of anesthesia, whereas inhibiting neurons increased the depth of anesthesia and propofol recovery time. Furthermore, mouse brain wave alterations can be caused by GABA neuron modulation, corroborating our findings. Taken together, our study explored the specific mechanism of NAc GABAergic neurons participating in propofol general anesthesia, which may provide a new strategy for the treatment of delayed emergency from propofol anesthesia.
c-Fos is an early immediate gene that is often used to reflect the activity of neurons [30]. We found that the expression of c-Fos in NAc GABAergic neurons decreased during propofol anesthesia. Besides, the threshold current was elevated in NAc GABAergic neurons after propofol perfusion by in vitro electrophysiology recordings. With the stimulation of step currents, the firing rate of NAc GABAergic neurons decreased dramatically. In addition, compared with the control group, when GABAergic neurons were chemically activated, the disappearance time of righting reflex was prolonged, the recovery time was shortened, and the doseresponse curve shifted to the right. The results were on the contrary during chemical inhibition. These results suggested that NAc GABAergic neurons are involved in the mechanism of propofol general anesthesia, and the sensitivity and the anesthesia effect of propofol may be affected by NAc GABAergic neurons. The activity of NAc is increased during the recovery from anesthesia and participates in the regulation of awakening. According to other studies, in the state of optogenetic activation, NAc D 1 R neurons can promote the awakening of mice [29]. Similarly, our results also suggested that the activation of NAc GABAergic neurons can also promote wakefulness in mice during propofol anesthesia. Therefore, NAc may affect the process of propofol anesthesia through the interaction of internal neuron groups. Among them, GABAergic neurons play an important role. But the internal mechanism of different subtypes of neurons and whether they antagonize or promote each other needs us to further explore.
When regulating GABAergic neurons, the sensitivity of mice to propofol also changed, which further proved the role of NAc GABA neurons in the anesthesia effect. In addition to the change in the recovery period of righting reflex, we found that during chemical activation or inhibition, the induction period of anesthesia also changed, which is different from the previous studies that NAc is only involved in sleep awakening, but has no effect on sleep [19], which indicates that although a large number of studies have shown that the nuclei of the sleep awakening circuits are involved in the principle of general anesthesia [31], the mechanism is not necessarily the same and requires a lot of research [10].
To further clarify the specific mechanism of NAc GABAergic neurons during anesthesia, yellow light stimulation was delivered. We found that the induction time was significantly prolonged, but the awakening time did not change. When the recovery phase was inhibited alone, there was no change in the brain wave during the induction period, but the proportion of delta waves in the recovery phase was significantly increased. The opposite results were obtained under blue light stimulation. This indicates that NAc GABAergic neurons participate in the whole process of propofol general anesthesia, which can not only affect the recovery after anesthesia but also prolong the induction period and then affect the general anesthesia induced by propofol. This result is similar to the view that NAc is not only involved in the awakening stage [32] but also affects loss of consciousness [33] when it comes to sevoflurane anesthesia. Therefore, intravenous and gas anesthetics may have a similar pathway to alter consciousness to some extent despite different molecular structures. Besides, the volume of NAc has changed during the development of the disease in patients with disturbance of consciousness [21,34], which is consistent with our findings on the importance of NAc. In the process of receiving sevoflurane and other inhalation anesthetics, the neuronal activity of NAc changed in varying degrees in the process of induction and recovery [35]. During propofol anesthesia, we also found that NAc GABAergic neurons were involved in the process of loss and recovery of consciousness. These results also suggested that the neural circuit of intravenous anesthesia may be similar to that of inhaled anesthetics while different from the transformation of consciousness from sleep to awakening.
The local distribution of different types of receptors [20,[36][37][38] or ion channels [39,40] may have an important influence on the process of loss of consciousness affected by these general anesthetics. At the molecular level, studies by Luo et al. have shown that inhalation anesthetics such as isoflurane may be involved in dopamine receptors and potassium channels [19]. The strong effects of intravenous anesthetics such as propofol and pentobarbital may be more closely related to GABA receptors [41], but the mechanism of these molecular effects is still unclear. There are a large number of dopamine receptors in NAc [42], which may be involved in the changes in the activity of GABAergic neurons [32]. Simultaneously, whether GABA receptors affect the activity of GABAergic neurons needs further exploration. Our research has the following important limitations. Firstly, we have observed that the activation of GABAergic neurons can reduce the anesthetic effect of propofol. It is still not clear whether the role of NAc GABAergic neurons in the loss of consciousness of intravenous anesthetics is universal. In the future, we need to use a variety of inhaled and intravenous anesthetics to fully understand the role of NAc in the principle of general anesthesia. Secondly, the neural mechanisms of propofol-induced anesthesia are various [7], and the molecular mechanism of regulating NAc GABAergic neurons in the brain needs to be further assessed to develop safer and more efficient intravenous anesthetics. Thirdly, propofol is an intravenous anesthetic agent and caudal vein injection has been used to evaluate the complex pharmacokinetics of anesthesia in recent years [9]. However, intraperitoneal injection of general anesthetics contributes to better regulation of the occurrence of the behavior and convenience for experimental observation [43,44]. Therefore, intraperitoneal injection of propofol is still widely used in rodents to observe the effects of anesthetics on behavior. In order to be more consistent with the pharmacokinetics of propofol, intravenous administration should also be explored in further experiments.

Conclusion
Our current results firstly demonstrate GABAergic neurons in NAc effectively modulate the general anesthesia effect of propofol and offer a potential brain site to target in future studies on propofol-induced alterations in consciousness. Therapeutic methods targeting NAc GABAergic neurons may offer the potential to further explore the etiologies of propofol anesthesia-related adverse effects and disorders of consciousness.