Ventral pallidal GABAergic neurons control wakefulness associated with motivation through the ventral tegmental pathway

The ventral pallidum (VP) regulates motivation, drug addiction, and several behaviors that rely on heightened arousal. However, the role and underlying neural circuits of the VP in the control of wakefulness remain poorly understood. In the present study, we sought to elucidate the specific role of VP GABAergic neurons in controlling sleep–wake behaviors in mice. Fiber photometry revealed that the population activity of VP GABAergic neurons was increased during physiological transitions from non-rapid eye movement (non-REM, NREM) sleep to either wakefulness or REM sleep. Moreover, chemogenetic and optogenetic manipulations were leveraged to investigate a potential causal role of VP GABAergic neurons in initiating and/or maintaining arousal. In vivo optogenetic stimulation of VP GABAergic neurons innervating the ventral tegmental area (VTA) strongly promoted arousal via disinhibition of VTA dopaminergic neurons. Functional in vitro mapping revealed that VP GABAergic neurons, in principle, inhibited VTA GABAergic neurons but also inhibited VTA dopaminergic neurons. In addition, optogenetic stimulation of terminals of VP GABAergic neurons revealed that they promoted arousal by innervating the lateral hypothalamus, but not the mediodorsal thalamus or lateral habenula. The increased wakefulness chemogenetically evoked by VP GABAergic neuronal activation was completely abolished by pretreatment with dopaminergic D1 and D2/D3 receptor antagonists. Furthermore, activation of VP GABAergic neurons increased exploration time in both the open-field and light–dark box tests but did not modulate depression-like behaviors or food intake. Finally, chemogenetic inhibition of VP GABAergic neurons decreased arousal. Taken together, our findings indicate that VP GABAergic neurons are essential for arousal related to motivation.


Introduction
The ventral pallidum (VP) is a major component of the basal ganglia, which plays a key role in regulating locomotion, learning, and motivation, all of which rely on heightened arousal [1][2][3][4]. Functional disorders of the VP are associated with neuropsychiatric disorders [5], including depression [6] and drug addiction [7][8][9], which are often accompanied by sleep disturbances [5,10], suggesting that the VP may play an important role in the regulation of sleep and wakefulness. However, whether the VP is involved in sleep-wake regulation remains poorly understood.
The VP receives dense inputs from the nucleus accumbens (NAc), which is mostly composed of GABAergic dopamine D 1 receptor (D 1 R)-positive neurons in the direct pathway and D 2 R-positive neurons in the indirect pathway [11]. Our previous studies have shown that NAc D 1 R neurons and NAc D 2 R neurons promote arousal and sleep, respectively [12]. Both NAc D 1 R neurons and NAc D 2 R neurons send dense inputs to the VP; however, the precise role of the VP in sleep/wake regulation has not yet been elucidated. Blockade of GABA A receptors in the VP via local administration of bicuculline, which activates VP neurons, greatly increases locomotion [13][14][15]. Other evidence shows that the VP promotes cocaine selfadministration [16], which is accompanied by heightened arousal. Thus, we hypothesized that the VP represents a principal wake-promoting region.
The VP is a highly heterogeneous structure that is composed principally of GABAergic projection neurons [4]. Anatomical and functional studies have shown that VP GABAergic neurons regulate motivation, depressionlike, and feeding behaviors via their projections to the mediodorsal thalamus (MD) [17], ventral tegmental area (VTA) [18,19], lateral habenula (LHb) [6], and lateral hypothalamus (LH) [20]. However, it is unclear whether the VP also controls arousal through these neural circuits. Anatomical tracings and electrophysiological recordings have shown that VP GABAergic neurons send dense inputs to the VTA, and form connections with VTA neurons [21]. It is noteworthy that the VP drives positive reinforcement through the VTA [21], the latter of which is composed of wake-promoting dopaminergic and glutamatergic neurons, as well as arousal-reducing GABAergic neurons [22,23]. Thus, the VTA represents a candidate region that receives direct axonal connections from VP GABAergic neurons. However, the putative circuit mechanism of the VP-VTA in sleep-wake regulation remains to be elucidated.
In the present study, we sought to uncover the specific role of VP GABAergic neurons in controlling sleep-wake behaviors. Fiber photometry unveiled arousal-correlated neuronal activity in VP GABAergic neurons across the spontaneous sleep-wake cycle. By using chemogenetic and optogenetic manipulations combined with polysomnographic recordings, we demonstrated that VP GABAergic neurons were sufficient to initiate and maintain behavioral arousal. In vivo and in vitro optogenetic-mediated circuit mapping revealed that VP GABAergic neurons regulated arousal through disinhibition of VTA dopaminergic neurons. Pharmacological manipulations further demonstrated that dopaminergic signaling was necessary for VP GABAergic neurons in promoting arousal. Importantly, we found that consistent with increased wakefulness, activation of VP GABAergic neurons concomitantly increased motivation, as demonstrated by longer exploration times. Finally, chemogenetic inhibition or lesioning of VP GABAergic neurons demonstrated that VP GABAergic neuronal activity was necessary for maintaining arousal. Collectively, our results provide several lines of evidence regarding VP GABAergic neurons and their circuit mechanisms in promoting arousal and motivation.

Population activity of VP GABAergic neurons is increased during non-rapid eye movement (NREM) sleep-to-wake transitions
To investigate the physiological real-time activity of VP GABAergic neurons across spontaneous sleep-wake cycles in mice, we recorded calcium activity of VP GABAergic neurons via fiber photometry [24,25]. The AAV 2 -Ef1α-DIO-GCaMP6f construct was unilaterally injected into the VP of Vgat-Cre mice, and an optical fiber was implanted into the VP. GCaMP6f signals and electroencephalograms (EEGs)/ electromyograms (EMGs) were simultaneously recorded in freely moving mice within their home cages (Fig. 1a). We used immunofluorescent staining of substance P to label the boundary of the VP [4] (Supplementary Fig. 1a). GCaMP6fexpressing cells were located in the dorsolateral VP (Fig. 1b), and 95% of these cells overlapped with anti-GABA immunostaining (Fig. 1c). Interestingly, we observed that changes in the population activity of VP GABAergic neurons were highly and consistently associated with sleep-wake-stage transitions (Fig. 1d, e). In particular, VP GABAergic neurons displayed lower GCaMP6f activity during non-rapid eye movement (non-REM, NREM) sleep (i.e., high delta power density, increased EEG amplitude and decreased EMG activity), whereas they exhibited moderate activity during wakefulness (i.e., low delta power density, decreased EEG amplitude and increased EMG activity) and even higher activity during REM sleep (i.e., high theta power density, decreased EEG amplitude and no EMG activity) (Fig. 1d, e). Notably, a dramatic increase in the GCaMP6f activity of VP GABAergic neurons was initiated at~4 s before NREM-towake and 2 s before NREM-to-REM transitions, whereas decreased GCaMP6f activity occurred before wake-to-NREM transitions or REM-to-wake transitions (Fig. 1f). Although the average calcium activity of VP GABAergic neurons decreased with prolonged time spent in wakefulness, the mean ΔF/F during wakefulness was higher than that during NREM sleep, indicating a potential role of VP GABAergic neurons in maintaining arousal. These findings demonstrate that the population activity of VP GABAergic neurons was associated with sleep-wake transitions. Higher activity of VP GABAergic neurons was appeared in transitions from NREM sleep to wakefulness, suggesting a physiological role of VP GABAergic neurons in the regulation of wakefulness.
Chemogenetic or optogenetic activation of VP GABAergic neurons is sufficient to induce and maintain wakefulness Having identified increased activity of VP GABAergic neurons during transitions from NREM sleep to wakefulness, we next interrogated the causal role of VP GABAergic neurons in regulating arousal. Chemogenetic and optogenetic approaches were used to manipulate the activity of VP GABAergic neurons during simultaneous EEG/EMG recordings. The AAV 10 -hSyn-DIO-hM3Dq-mCherry construct was bilaterally injected into the VP of Vgat-Cre mice (Fig. 2a). Immunochemical staining showed that mCherry was precisely expressed in the VP (Fig. 2b, Supplementary Fig. 1b-e). In vitro electrophysiological experiments showed that hM3Dq+ GABAergic neurons were activated after clozapine-n-oxide (CNO) application (Fig. 2c). C-Fos staining confirmed the activation of mCherry+ neurons by CNO injection ( Supplementary  Fig. 2a-c). Importantly, chemogenetic activation of VP GABAergic neurons significantly increased wakefulness and concomitantly decreased both NREM and REM sleep during the 5 h following CNO (1 mg/kg) administration at 09:00 (light on at 07:00), as compared with these parameters in the vehicle control ( Fig. 2d-f). CNO administration induced a decrease in low frequency, high-amplitude EEG activity together with an increase in EMG activity (Fig. 2d), suggesting that chemogenetic activation of VP GABAergic neurons induced behavioral arousal. Administration of CNO at 1 mg/kg resulted in a 78% increase in wakefulness and a 38% reduction in NREM sleep during the 5-h post-injection period (Fig. 2e, f). In contrast, there was no significant difference in the EEG power density at each stage (Fig. 2g), further indicating that activation of VP GABAergic neurons induced physiological arousal.
Next, we used optogenetics to precisely activate VP GABAergic neurons and to clarify whether activation of VP GABAergic neurons could initiate wakefulness. The AAV 10 -hSyn-DIO-ChR2-mCherry construct was injected into the bilateral VP with optical fibers targeting these neurons in Vgat-Cre mice ( Supplementary Fig. 3a, b). Optogenetic stimulation increased c-Fos expression in mCherry+ neurons ( Supplementary Fig. 3c). Acute optogenetic activation of VP GABAergic neurons awakened mice from NREM sleep, with a significant increase in EMG activity after EEG desynchronization ( Supplementary  Fig. 3d). Short latencies for sleep-to-wake transitions were observed during blue-light pulses at frequencies from 5-30 Hz ( Supplementary Fig. 3e). Optogenetic stimulation induced wakefulness during periods of NREM sleep in all trials (35 trails) in VP-ChR2 mice, versus only 4 out of 40 trials in VP-mCherry control mice ( Supplementary Fig. 3f). Importantly, chronic optogenetic stimulation (5-ms pulses at 20 Hz, with 10-s on/20-s off for 120 cycles) of VP GABAergic neurons for 1 h consistently maintained longterm arousal in VP-ChR2 mice ( Supplementary Fig. 3g, h). Furthermore, the time spent in arousal was increased by 2.1fold, with a 97% decrease in NREM sleep. Taken together, these findings indicate that chemogenetic or optogenetic activation of VP GABAergic neurons was sufficient to initiate and maintain behavioral arousal.

VP GABAergic neurons regulate arousal through the VTA pathway
To uncover the circuit mechanism of VP GABAergic neurons in regulating arousal, we first performed anterograde tracing of VP GABAergic neurons ( Supplementary Fig. 2d-l). AAV-DIO-mCherry was injected into the VP, after which abundant mCherry terminals were found in the VTA (Supplementary Fig. 2e-h), which is consistent with previous findings that the VP innervates VTA neurons [6,21]. Surprisingly, after chemogenetic activation of VP GABAergic neurons, we found a robust increase of c-Fos expression in VTA tyrosine hydroxylase-positive (TH+) neurons ( Supplementary  Fig. 2m, p), which play an important dopaminergic role in the regulation of arousal and motivation-related behaviors [22]. Next, we employed an optogenetic approach to precisely stimulate VP GABA -VTA projections in vivo and in vitro. The AAV 10 -hSyn-DIO-ChR2-mCherry construct was injected into the VP with bilateral optical fibers targeting axonal terminals in the VTA of Vgat-Cre mice (Fig. 3a). Blue-light stimulation induced intense c-Fos expression in the VTA (Fig. 3b), principally in TH+ neurons ( Supplementary Fig. 4a, b), suggesting that optogenetic stimulation of VP GABA -VTA projections in vivo activated VTA dopaminergic neurons. Optogenetic stimulation of the VTA at 20 Hz immediately awakened mice from NREM sleep, inducing decreased EEG delta power/amplitude, while concomitantly increasing EMG activities (Fig. 3c), suggesting that stimulation of VP GABA -VTA projections induced behavioral arousal. Short latencies for sleep-to-wake transitions were observed during blue-light pulses onto the VTA at frequencies ranging from 5-30 Hz (Fig. 3d). In total, we performed 33 trials of optogenetic stimulation in VP-ChR2-VTA mice and found that mice awakened from NREM sleep in 30 trials, versus only 2 of 31 trials in VP-mCherry-VTA control mice (Fig. 3e). In addition, chronic stimulation of VP GABA -VTA projections for 1 h also dramatically increased long-term arousal (Fig. 3f). These data reveal that the activation of VP GABAergic projections to the VTA was sufficient to promote arousal.

VP GABAergic neurons form monosynaptic connections with both VTA dopaminergic and GABAergic neurons
VP GABAergic neurons send inhibitory inputs to the VTA [6,26]; however, chemogenetic activation of VP GABAergic neurons and optogenetic activation of VP GABA -  Supplementary Fig. 5b). Although the latencies and amplitudes of light-induced inhibitory postsynaptic currents (IPSCs) in TH+ and TH− cells were not significantly different from one another, the latencies of IPSCs from 6/21 connected TH+ cells were longer than 5 ms, indicating that VP GABAergic neurons may also indirectly innervate these TH+ cells, while the latencies of all connected TH− cells were less than 5 ms, suggesting that the latter were all monosynaptic connections ( Supplementary Fig. 5c, d). Single-cell PCR showed that the majority of GFP-negative cells (non-GABAergic cells, 6 out of 7 cells) expressed TH RNA, whereas only one of these cells expressed vesicle glutamate transporter 2 (Vglut2) RNA, indicating that most GFP-negative cells were TH+ dopaminergic neurons ( Supplementary Fig. 5e). Under the cell-attached mode, optogenetic stimulation of axonal terminals from VP GABAergic neurons in the VTA inhibited the firing of both TH+ cells (Supplementary Fig. 5f-i) and TH− cells ( Supplementary Fig. 5j-m) via light-evoked IPSCs, which were blocked by the GABA A receptor antagonist, gabazine (SR-95531) ( Supplementary Fig. 5i, m). In addition, we found that three TH+ cells, but no TH− cells, were inhibited by VP GABAergic inputs through GABA B receptors ( Supplementary Fig. 6a-e). Taken together, our c-Fos and electrophysiological results demonstrate that VP GABAergic neurons principally targeted VTA TH− cells, resulting in disinhibition of TH+ cells, which putatively underlies control of arousal by VP GABAergic neurons.

VP GABAergic neurons regulate arousal partially through the LH pathway
Our anterograde tracing data showed that VP GABAergic neurons projected to the LH (Supplementary Fig. 2i-l), the latter of which plays an important role in sleep-wake regulation [27]. Chemogenetic activation of VP GABAergic neurons increased c-Fos expression in LH orexinergic neurons ( Supplementary Fig. 2n, q) but not in melaninconcentrating hormone (MCH) neurons [28] (Supplementary Fig. 2o, r), indicating that VP GABAergic neurons may also regulate arousal through the LH pathway. Thus, we used an optogenetic approach to stimulate VP GABA -LH projections in vivo and in vitro ( Supplementary Fig. 7a). Dense mCherry-positive axonal terminals of VP GABAergic neurons were found in the LH ( Supplementary Fig. 7b), and optogenetic stimulation induced GABA A R-mediated IPSCs ( Supplementary Fig. 7c-e). Subsequently, we stimulated VP GABA -LH projections in vivo ( Supplementary  Fig. 7f). Interestingly, optogenetic stimulation of VP GABA -LH projections also increased c-Fos expression in the LH (Supplementary Fig. 7g-h), indicating that several subtypes of LH neurons were disinhibited. In vivo optogenetic stimulation of VP GABA -LH projections also induced behavioral arousal ( Supplementary Fig. 7i). Short latencies for sleep-to-wake transitions were observed during blue-light stimulation in the LH at frequencies ranging from 5-20 Hz ( Supplementary Fig. 7j). In addition, chronic stimulation of VP GABA -LH projections for 1 h induced a 1.7-fold increase in wakefulness ( Supplementary Fig. 7k, l). These data suggest that the LH was also an important downstream nucleus for VP GABAergic neurons in promoting arousal.
The MD and LHb are also innervated by VP GABAergic neurons. Hence, we also stimulated VP GABA -MD (Supplementary Fig. 8a-d) and VP GABA -LHb (Supplementary Fig. 8e-h) projections in vivo. However, optogenetic stimulation of these pathways did not alter sleep-wake transitions, suggesting that the MD and LHb were not involved in VP GABAergic neurons promoting arousal.

Dopaminergic release is necessary for VP GABAergic neurons in regulating arousal
Our results showed that VP GABAergic neurons promoted arousal through the VTA within the ventral midbrain. To further investigate the role of dopamine in these effects, we investigated whether systemic blocking of D 1 and D 2 receptors would abolish wakefulness induced by chemogenetic activation of VP GABAergic neurons (Fig. 4a). We administered pretreatments of the dopaminergic D 1 receptor antagonist, SCH23390 (0.03 mg/kg), and the dopaminergic D 2 /D 3 receptor antagonist, raclopride (2 mg/kg), based on previous studies [29,30], in Vgat-Cre mice that expressed The VP was infected by the hM3Dq-mCherry virus. Scale bar = 200 µm. c Bath application of CNO at 5 µM increased firings in an hM3Dq-mCherry neuron. d Typical examples of a relative EEG power heatmap and EEG/EMG traces after vehicle or CNO (1 mg/kg) injection at 9:00 in an hM3Dq Vgat-Cre mouse. e The total time spent in NREM sleep and wakefulness for 5 h after administration of vehicle or CNO. f Time-course of wakefulness, NREM sleep, and REM sleep in Vgat-Cre mice expressing hM3Dq after administration of vehicle or CNO. g EEG power density for 5 h after vehicle or CNO administration. Data shown are the mean ± SEM (n = 9; *P < 0.05, **P < 0.01, using repeated-measures ANOVA followed by paired t-test).
hM3Dq in the VP. Chemogenetic activation of VP GABAergic neurons significantly increased wakefulness; however, this increased wakefulness was completely abolished by pretreatments with SCH23390 and raclopride (Fig. 4b, c). Treatment with SCH23390 and raclopride but without CNO did not increase arousal or NREM sleep in the light phase. These results further demonstrate that activation of VTA TH+ dopaminergic neurons plays a crucial role in mediating arousal controlled by VP GABAergic neurons.

Activation of VP GABAergic neurons increases motivation
The VP has been reported to be involved in numerous motivation-related behaviors, including motor function, learning/memory, and reward [4]. In order to explore whether activation of VP GABAergic neurons promotes motivation-related arousal, we tested a series of behaviors after chemogenetic activation of VP GABAergic neurons. Chemogenetic activation of VP GABAergic neurons increased locomotion and the time spent in the central area in an open-field test (OFT) (Fig. 4d), indicating increased exploratory behavior. In contrast, there were no changes in behavior in the elevated plus-maze (EPM) test or the tailsuspension test (Fig. 4e, f), indicating that activation of VP GABAergic neurons did not change anxiety-like or depression-like behaviors. Importantly, activation of VP GABAergic neurons increased the number of entries into the light box during the light-dark box test, further indicating that activation of VP GABAergic neurons specifically promoted exploratory behavior (Fig. 4g). In addition, food intake within 3 h and 12 h after CNO administration was not significantly altered, indicating that activation of VP GABAergic neurons did not directly affect food intake (Fig. 4h). Taken together, increased arousal mediated by activation of VP GABAergic neurons was related to motivation-related exploratory behaviors.

VP GABAergic neurons are essential for arousal
In order to determine whether the activity of VP GABAergic neurons is essential for sustaining arousal, we specifically injected the AAV 10 -hSyn-DIO-hM4Di-mCherry construct into the VP of Vgat-Cre mice (Fig. 5a). CNO administration inhibited VP GABAergic neurons, with a decrease in c-Fos expression within mCherry+ cells (Fig. 5b, c). Strikingly, the total duration of wakefulness was significantly decreased by 30% during 3 h after CNO administration at 21:00 (light off at 19:00), concomitant with a 1.1-fold increase in NREM sleep and a 65% increase in REM sleep, as compared with these parameters in vehicle controls (Fig. 5d-f). In addition, there was no significant difference in the power density of NREM sleep after CNO administration (Fig. 5g), indicating that inhibition of VP GABAergic neurons induced physiological sleep.
In addition, lesioning of VP GABAergic neurons decreased arousal in the dark phase ( Supplementary Fig. 9). A combination of AAV 8 -hSyn-DIO-Caspase3 and AAV 8 -hSyn-DIO-eYFP constructs was bilaterally injected into the VP (Supplementary Fig. 9a), and ablation of GABAergic neurons was confirmed by anti-GABA immunofluorescent staining ( Supplementary Fig. 9b). Consistent with our inhibition results, lesioning of VP GABAergic neurons decreased the time spent in wakefulness of the first 4 h in the dark phase ( Supplementary Fig. 9c, d), as compared with that of GFP controls, whereas the EEG power density during NREM sleep was not altered ( Supplementary  Fig. 9e). Collectively, these results demonstrate that inhibition or lesioning of VP GABAergic neurons decreased the amount of wakefulness.

Discussion
The VP is an important component of the ventral basal ganglia and regulates numerous motivation-related behaviors-including reward, motor function, and learning/ memory-that rely on heightened arousal [2,4,31,32]. In addition, the VP is connected to several sleep-wake regulatory nuclei, as it receives NAc projections and projects to both the VTA and LH [4,33]. In the present study, we report that VP GABAergic neurons regulated behavioral arousal. A physiological correlation of VP GABAergic neurons and arousal was indicated by increased population activity during wakefulness. Furthermore, bidirectional chemogenetic and optogenetic manipulations demonstrated that VP GABAergic neurons were vital for promoting arousal. Moreover, we elucidated that VP GABAergic neurons regulated arousal through the VTA pathway, in which dopaminergic signaling was found to be necessary for arousal. Taken together, our findings identify a novel region in sleep-wake regulation and bridge several gaps in the role of the ventral basal ganglia in sleep-wake regulation associated with motivation ( Supplementary Fig. 10).
Recently, numerous studies have discovered a crucial role of the mesolimbic dopaminergic system in sleep-wake regulation [22,23,29]. VTA dopaminergic neurons and Fig. 3 Optogenetic stimulation of the VP GABA -VTA pathway induces arousal. a Sagittal diagram for in vivo optogenetic stimulation of the VP GABA -VTA pathway in Vgat-Cre mice. b Optogenetic stimulation of axonal terminals of VP GABAergic neurons in the VTA induced dense c-Fos expression in TH+ neurons. c Optogenetic activation of the VP GABA -VTA pathway induced an immediate transition from NREM sleep to wakefulness. The EEG amplitude was decreased, with a decreasing delta power and increasing EMG activity, after blue-light stimulation in ChR2 mice, but not in mCherry control mice. d The latency from NREM sleep to wakefulness decreased with increased stimulation frequency. e Sleep stages after VTA blue-light stimulation in ChR2-mCherry mice or mCherry control mice. f Durations of wakefulness, NREM sleep, and REM sleep during 1-h optogenetic stimulation of the VP GABA -VTA pathway during the light period (n = 4. **P < 0.01, using repeated-measures ANOVA, followed by paired t-test).
glutamatergic neurons promote arousal [22,23], while GABAergic VTA neurons decrease arousal [34]. Furthermore, VP-to-VTA projections regulate several motivationrelated behaviors [21]. In the current study, we report that VP GABAergic neurons functionally connect to VTA dopaminergic, glutamatergic, and GABAergic neurons. In vivo stimulation of VP GABAergic neurons increased VTA dopaminergic neuronal activities, and stimulation of VP GABA -VTA projections induced arousal. Further analysis of this circuit mechanism showed that VP GABAergic neurons predominantly inhibited VTA GABAergic neurons, the latter of which inhibit wake-promoting dopaminergic and glutamatergic neurons. Although VP GABAergic neurons formed some direct inhibitory connections with VTA wake-promoting dopaminergic and glutamatergic neurons and decreased the firing of these neurons, the general effect in vivo was a disinhibition of these neurons and induction of arousal. Our previous study demonstrated that NAc D 1 R neurons promote arousal through disinhibition of VTA dopaminergic neurons, but that NAc D 1 R neurons exerted sparse inhibitory connections onto VTA dopaminergic neurons [25]. Moreover, in the present study, pretreatment with dopaminergic antagonists completely blocked the wakefulness induced by the activation of VP GABAergic neurons, further demonstrating that activation of VTA dopaminergic neurons plays a crucial role in mediating arousal via VP GABAergic neurons.
In the present study, we also found that VP GABAergic neurons regulated arousal through the LH pathway. Several subtypes of neurons in the LH regulate sleep-wake behaviors [27]; LH orexinergic neurons are necessary for maintaining arousal, while several subtypes of GABAergic neurons promote sleep [35]. In our present study, in vivo stimulation of VP GABAergic neurons increased the activity of LH orexinergic neurons but not MCH neurons. We speculate that activation of VP GABAergic neurons may inhibit LH inhibitory interneurons and disinhibit wakepromoting neurons (i.e., orexinergic neurons) [36], which would explain why we found increased c-Fos expression in LH orexinergic neurons following activation of VP GABAergic neurons. Indeed, a subset of LH GABAergic neurons send inputs to the sleep-promoting ventrolateral preoptic area and regulate wakefulness [37], whereas another subset sends inputs to other wake-promoting regions, including the tuberomammillary nucleus, ventral periaqueductal gray, and locus coeruleus, all of which may promote sleep. Nevertheless, certain subtypes of inhibitory interneurons may also receive VP GABAergic projections and regulate arousal. Collectively, our findings suggest that the VP GABA -LH circuit involved in controlling arousal may be partially mediated through activating LH wakepromoting orexinergic neurons.
Interestingly, a previous study has shown that chemogenetic activation of basal forebrain (BF) GABAergic neurons also promotes arousal [38]. Although the BF and VP are located nearby to one another, they have distinct projections and functions. By synthesizing our present findings with those of previous studies, we suspect that GABAergic neurons in both the VP and BF are important in regulating arousal, which may each contribute to different higher-order brain functions. In our present study, we also found that the population activity of VP GABAergic neurons decreased following prolonged wakefulness but was still higher than that during NREM sleep, suggesting that VP GABAergic neurons may play a role in maintaining wakefulness. Interestingly, the calcium activity of VP GABAergic neurons was also increased during NREM-to-REM transitions, but chemogenetic or optogenetic activation of these neurons only promoted arousal, which is consistent with previous findings on the function of NAc D 1 R neurons [25]. We suspect that several neurons in wakepromoting nuclei are activated during both wakefulness and REM sleep, but not NREM sleep.
The major input to the VP originates from the NAc. Our previous data have shown that NAc D 1 R neurons regulate wakefulness, whereas NAc D 2 R/A 2A R neurons projecting to the VP regulate sleep [12,25]. However, although both NAc D 1 R and D 2 R neurons are GABAergic neurons, they exhibit opposing functions. Due to less than 3% colocalization of D 1 R-and D 2 R-expressing fibers in the VP [39], we hypothesize that NAc D 1 R and D 2 R neurons each project to distinct targets in the VP. Indeed, a recent study showed that VP GABAergic neurons receive equal inputs from NAc D 1 R and D 2 R neurons [26]. However, NAc D 2 R neurons may specifically target VP wake-promoting neurons (e.g., wake-promoting GABAergic projection neurons), whereas NAc D 1 R neurons may specifically target VP GABAergic interneurons. Our present results have revealed that VP GABAergic neurons are wake-promoting,  . Data shown are the mean ± SEM (*P < 0.05, **P < 0.01, compared with vehicle control, two-way ANOVA followed by Tukey post-hoc test, paired or unpaired t test). but there are still some GABAergic interneurons in the VP that may receive NAc D 1 R neurons inputs, which may decrease arousal. Further studies should clarify how NAc D 1 R/D 2 R neurons target specific VP neuronal subtypes and elucidate the differential roles of VP glutamatergic and cholinergic neurons in sleep-wake regulation.
The VP plays an active role in integrating information in the ventral basal ganglia, which is a key component in the brain for generating motivation-related behaviors, including locomotion, reward, learning/memory, addiction, and aversion. Inhibition of NAc D 2 R axonal terminals to the VP is sufficient to enhance motivation [40], which is consistent with another previous finding that inhibition of NAc D 2 R terminals in the VP promotes behavioral arousal [12]. An optogenetic study showed that activity in VP GABAergic neurons projecting to the VTA drives positive reinforcement [21], indicating that VP-VTA GABAergic projection neurons promote motivation-related behavior via influencing VTA dopaminergic neurons. Our present results support these findings that arousal induced by VP GABAergic neurons projecting to the VTA contributed to motivation-related behavior. In contrast, the VP also sends inputs to the LH, which plays a key role in feeding behaviors [41], but activation of VP GABAergic neurons in our present study did not significantly influence food intake, indicating that non-feeding-related subtypes of LH neurons may have been preferentially activated. Taken together, previous studies and our present findings clearly elucidate a NAc-VP-VTA/LH circuit in modulating arousal, which also significantly influences motivation-related behaviors; hence, this elucidated circuit constitutes a crucial link between arousal and motivation.
VP-mediated encoding of and contributions to response vigor are specific to the ability of incentive cues to invigorate reward-seeking behaviors, upon which reward delivery is contingent [42]. A recent study showed that stimulating VP GABAergic neurons facilitated cocaineseeking and that cued seeking increased the number of calcium-signaling events in VP GABAergic neurons [35]. These findings strongly support our present conclusion that increasing the activity of VP GABAergic neurons is necessary for controlling and maintaining wakefulness, which may provide a foundation for motivation-related and drug-seeking behaviors. Several VP-implicated neuropsychiatric disorders induce sleep-wake alterations, indicating that the NAc-VP-VTA/LH circuit may represent a potential therapeutic target for sleep disorders, especially those related to neuropsychiatric disorders. For example, reward-specific firing in the VP is present in a greater proportion and arises soon following reward delivery [43], which may explain aberrant arousal in reward situations. As the VP regulates motivation and cognition, adaptations in this system may contribute to the mood and mnemonic disruptions that can accompany Parkinson's disease [44].
In conclusion, we found a key role of VP neurons in integrating information from the ventral basal ganglia and in promoting arousal and motivation-related behaviors through a ventral midbrain pathway.

Chemicals and drug administrations
CNO (C4759, LKT, USA) at 1 mg/kg was intraperitoneally (i.p.) injected at 09:00 or 21:00. SCH23390 (Sigma, USA) at 0.03 mg/kg and raclopride (Sigma, USA) at 2 mg/kg were freshly prepared and i.p. administered at 15 min before CNO administration. All of the drugs were injected at a volume of 10 ml/kg of body weight.

EEG/EMG electrode-implantation surgery
Briefly, mice were anesthetized under 1.5% isoflurane in oxygen at a 0.8-LPM flow rate. Two stainless-steel screws, which served as EEG electrodes, were implanted into the skull above the right cortex (coordinates: anteroposterior [AP]: ±1.0 mm, mediolateral [ML] + 1.5 mm), and two EMG electrodes were implanted in the dorsal neck musculature to monitor muscular activity, as previously described [46,47]. All EEG electrodes were fixed to the skull via dental cement.

EEG recordings and analysis
EEG recordings and analysis were performed as previously described [48][49][50]. Cortical EEG and neck EMG signals were amplified and filtered (Biotex, Kyoto, Japan. EEG, 0.5-30 Hz; EMG, . Each frequency band in the EEG was calculated via SleepSign software using the fast Fourier transformation (FFT) method. For FFT analysis, Fourier transformation was used to calculate the power variable (uV 2 ), and absolute power spectra of the EEG data were computed every 4 s over a 0-25-Hz window with 0.25-Hz resolution. Absolute power spectra were transferred into relative changes by taking a synchronization value as 100%. For sleep-stage analysis, EEGs and EMGs were automatically classified using 4-s epochs for wakefulness, REM sleep, and NREM sleep according to standard criteria.

Fiber photometry
Fiber-photometry experiments were performed in both light and dark periods as previously described [24,25]. Photometry data were exported to MATLAB Mat files from Spike2 for further analysis. We derived the value of the photometry signal (ΔF/F) by calculating (F-F 0 )/F 0 , where F 0 is the median fluorescent signal. The averaged ΔF/F was calculated during all times of sleep-wake states. For analyzing state transitions, we determined each state transition and aligned ΔF/F in a ± 50-s window around each point that was calculated. The average peak of the ΔF/F was selected and compared for different sleep stages.

Optogenetic stimulation during polygraphic recordings
For in vivo light stimulation, light-pulse trains were generated via a laser stimulator (SEN-7103, Nihon Kohden, Japan) and output through an isolator (ss-102J, Nihon Kohden, Japan). A rotating optical joint (FRJ_FC-FC, Doric Lenses, Canada) was used to relieve torque and was attached to the external end of the optical fiber. For acute photostimulation, each stimulation epoch was applied 20 s after identifying a stable NREM or REM sleep event by real-time online EEG/EMG analysis. Light-pulse trains (5-ms duration each) were programmed and conducted during the light period, when mice are inactive. The cut-off line for stage transitions was 60 s after the laser was turned on. For chronic photostimulation, programmed light-pulse trains (5-ms pulses at 20 Hz, with 10-s on/ 20-s off for 120 cycles) were used from 09:00 to 10:00. EEG/EMG recordings during the same period on the previous day served as a baseline control. Power intensities of blue or yellow light at the tip of the optical fiber were calibrated to emit 3-7 mW [51].

Elevated plus-maze test (EPM)
Each mouse was placed in the center of the intersecting arms, facing an open arm, and was allowed to explore the apparatus for 5 min. The time spent in each arm was recorded by an experimenter who was blinded to the treatment [52]. Anxiety-like behavior was determined by measuring the time spent in the open arms.

Open-field test (OFT)
Mice were gently placed in the center of the field, and movement was recorded for 5 min with a video-tracking system. The time spent in the center of the arena (defined as a 20 × 20 cm zone in the center of the apparatus) was measured [52].

Light-dark box test
Each mouse was released in the center of the light compartment and was allowed to explore the arena for 5 min. The number of entries and time spent in the white compartment were recorded [53].

Tail-suspension test
Mice were suspended by the tail from a metal rod using adhesive tape. The rod was fixed 45 cm above the ground. The test session was recorded for 5 min, and the immobility time was determined by an observer who was blinded to the treatment conditions [52].

Feeding-behavior assays
Mice were individually housed in home cages with a mild food restriction by removing chow food during the final 8 h before lights were turned off (19:00). All of the tests were performed using a counter-balanced within-subjects design. Mice received i.p. injections of either vehicle or CNO (1 mg/kg) at 19:00, 15 min before access to~5 g of fresh standard chow. Food intake was measured at 3 and 12 h after CNO injection [54].

Immunohistochemistry
Immunohistochemistry was performed as described previously [25,52,55]. Mice were deeply anesthetized at 1.5 h after CNO administration or 0.5 h after light stimulation and were then perfused intracardially with phosphate-buffered saline (PBS) followed by 4% paraformaldehyde (PFA). After each brain sank in a 30% sucrose solution, coronal slices (30 μm) were cut into four series via a microtome (CM1950, Leica, Germany).

Statistical analysis
Data are expressed as the mean ± standard error of the mean (SEM). Statistical significance was assessed using two-tailed paired Student's t tests to compare the total sleep amounts and behavioral results between the two groups. One-way, two-way, or repeated-measures analyses of variance (ANOVAs) were used to compare sleep amounts, followed by pairwise comparisons via Tukey post-hoc tests. EEG power spectra were compared using one-way ANOVAs followed by Fisher's PLSD tests or via non-paired, two-tailed Student's t tests. Two sets of frequencies were analyzed by chi-square tests. A two-tailed P value <0.05 was considered to be statistically significant. All of the data were analyzed using Prism 8.0 software.

Data availability
The data that support the findings of this study are available from the corresponding author upon request.

Code availability
Data analyses were conducted in Matlab using scripts available from the corresponding author upon reasonable request.

Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/.