Neurotensin in the extended amygdala maintains wakefulness in novel environments

Significance

When animals encounter new surroundings, they usually stay awake and alert—an adaptive response that likely increases their chances of survival. But how the brain connects the experience of novelty to this heightened state of wakefulness has remained a mystery. In this study, we found a specific pathway in the extended amygdala, involving corticotropin-releasing factor (CRF)-expressing neurons that use the signaling molecule neurotensin, which promotes sustained wakefulness in novel environments. By selectively activating or silencing this pathway, we showed that it is required for sustaining alertness in unfamiliar environments. These findings uncover a fundamental brain mechanism that translates environmental uncertainty into adaptive states of alertness.

Abstract

Animals remain awake in unfamiliar environments to assess potential safety threats, a process involving changes in neuronal activity within sleep–wake regulatory brain regions. However, the specific circuits and neurotransmitters involved remain poorly understood. Here, we show that neurotensin (NTS) peptides in corticotropin-releasing factor (CRF) neurons of the lateral part of the interstitial nucleus of the posterior limb of the anterior commissure (IPACL) play a key role in maintaining wakefulness in response to environmental changes. Activation of IPACL^CRF neurons increased wakefulness, whereas their inhibition or deletion of NTS reduced wakefulness in novel environments. These neurons are activated in response to exposure to a novel environment and project primarily to the substantia nigra pars reticulata (SNr) and release NTS, which modulates wakefulness. These findings suggest that NTS signaling from IPACL^CRF neurons to the SNr is essential for sustaining wakefulness in unfamiliar or changing environments.

Sleep is a fundamental physiological process, and sleep disruption has been linked to various conditions, including psychiatric disorders, diabetes, and disturbances in the reproductive cycle (1–3). In addition to its basic biological importance, the temporal regulation of sleep and wakefulness serves as a crucial mechanism for environmental adaptation. For example, by modulating diurnal and nocturnal behaviors, animals reduce the risk of predation by adjusting their sleep patterns according to the time of day (4). In critical situations, such as the presence of predators or the sudden unavailability of food, animals can override the need for sleep to prioritize immediate survival. In unfamiliar environments, staying awake enables animals to evaluate potential threats and ensure the safety of their surroundings. A similar phenomenon occurs in humans, commonly referred to as the “first night effect”(5), wherein individuals experience increased wakefulness during their first night in an unfamiliar setting. This heightened vigilance typically diminishes by the second night (6).

Various brain regions and neuronal populations involved in sleep regulation have been reported (7). For instance, orexin neurons in the lateral hypothalamus, dopaminergic neurons in the ventral tegmental area (VTA) and substantia nigra pars reticulata (SNr), and corticotropin-releasing factor (CRF) neurons in the paraventricular nucleus (PVN) are known to promote wakefulness (8–11). Conversely, GABAergic neurons in the preoptic area, parafacial zone, the VTA, ventrolateral periaqueductal gray, ventral zona incerta, basal forebrain, and amygdala play key roles in sleep regulation (12–20). These complex neural circuits coordinate the regulation of sleep and wakefulness.

Several brain regions are reported to be activated in response to novel environments (21–24). Furthermore, exposure to such environments elevates blood corticosterone levels and increases arousal (25, 26). This response is mediated by activation of the hypothalamic–pituitary–adrenal (HPA) axis, driven by CRF neurons, which are key players in the stress response (27). CRF neurons are thought to play a critical role in maintaining arousal (26). These neurons are not only located in the PVN but also in the bed nucleus of the stria terminalis (BNST) and the amygdala (28–31). Although PVN^CRF neurons are known to contribute to wakefulness (10, 32, 33), the role of CRF neurons in other regions in regulating wakefulness, particularly during exposure to novel environments, remains unclear.

In this research, we uncover that neurotensin (NTS) peptides in CRF neurons in the lateral part of the interstitial nucleus of the posterior limb of the anterior commissure (IPACL^CRF neurons) projecting to the SNr, contribute to the maintenance of wakefulness during environmental novelty.

Results

We exposed wild-type (WT) mice to either a novel or familiar environment at zeitgeber time (ZT) 4 (ZT0 corresponds to lights on) and recorded electroencephalography (EEG) and electromyography (EMG) (Fig. 1A). The latency to non–rapid eye movement (NREM) sleep was significantly longer in mice exposed to the novel environment than in those exposed to the familiar environment. Cumulative wakefulness was also increased during exposure to the novel environment (Fig. 1B). These findings suggest that specific neuronal mechanisms contribute to the maintenance of wakefulness in response to environmental novelty. Given that CRF neurons are involved in the stress response, we hypothesized that CRF neurons play a role in regulating sleep and wakefulness under such conditions.

Fig. 1.

CRF neurons in the IPACL are active in a novel environment. (A) Schematic illustration of the experimental procedure. (B) Representative EEG and EMG traces recorded in familiar and novel environments are shown on the Left. The red line in each panel indicates the time point when the mice were moved. The latency to NREM sleep following exposure to a familiar or novel environment is shown in the middle (*P < 0.05 vs. familiar; Student’s t test). Cumulative wakefulness after exposure is shown on the Right (*P < 0.05 vs. familiar; two-way repeated-measures ANOVA). (C) Schematic illustration of the experimental procedure. (D) Representative fluorescence images of different brain regions are shown on the Left. CRF and c-Fos are labeled in green and magenta, respectively. White arrowheads indicate c-Fos-positive CRF neurons. (Scale bar, 100 µm.) The percentage of c-Fos-positive CRF neurons in each brain region following exposure to familiar and novel environments is shown on the Right. Data are presented as mean ± SD. Statistical details are provided in SI Appendix, Table S3.

To investigate how exposure to a novel environment affects CRF neuronal activity, we collected mouse brains 1.5 h after exposure to either a novel or familiar environment (Fig. 1C) and stained them for c-Fos, a marker of neuronal activation. We observed a significant increase in c-Fos-positive CRF neurons in the PVN and the IPACL, while activity decreased in the posterior amygdala (central amygdala, CeA). No significant changes were observed in the BNST (Fig. 1D). Based on previous studies identifying a role for PVN^CRF neurons in regulating sleep and wakefulness (10, 34), we focused subsequent analysis on the lPACL and CeA.

The IPACL is located in the anterior part of the amygdala and contains CRF neurons known to drive locomotor activity and avoidance behavior (35, 36). In addition, c-Fos expression in the IPACL increases in response to both innate and learned disgust reactions (37). The CeA has also been implicated in the regulation of sleep and wakefulness (17). To investigate the roles of the IPACL and CeA in wakefulness in response to environmental novelty, we employed a chemogenetic approach to inhibit neuronal activity in CRF neurons within these regions during exposure to a novel environment (Fig. 2A). We injected an adeno-associated virus (AAV) carrying Cre-dependent hM4Di-mCherry (AAV-CMV-flex-hM4Di-mCherry) into the IPACL and CeA of CRF-iCre mice. Due to the anatomical proximity of these regions, hM4Di-mCherry expression was observed in both regions (SI Appendix, Fig. S1A). Inhibition of these neurons in combination with clozapine-N-oxide (CNO) injection significantly reduced the latency to NREM sleep and decreased cumulative wakefulness during exposure to a novel environment at ZT4 (Fig. 2B). To further clarify the functional roles of these neurons, we activated CRF neurons in the IPACL and CeA using hM3Dq in a familiar environment (home cage) at ZT4. Activation of these neurons led to an increase in wakefulness compared to saline controls (Fig. 2 C and D and SI Appendix, Figs. S1 and S2). Chemogenetic inhibition of CRF neurons in the IPACL and CeA did not affect vigilance states at ZT4 but significantly reduced wakefulness at ZT12 in a familiar environment (SI Appendix, Figs. S1 and S2). Activation of these neurons did not significantly impact anxiety-related behavior (SI Appendix, Fig. S3). These results suggest that CRF neurons in the IPACL and/or CeA contribute to maintaining wakefulness in novel environments during the day.

Fig. 2.

Effects of chemogenetic and optogenetic manipulation of CRF neurons in the IPACL on sleep and wakefulness. (A) Schematic of the experimental procedure and representative fluorescence images. (Scale bar: 500 µm.) (B) Representative EEG and EMG traces in a novel environment after saline or CNO injection are shown on the Left. Red lines indicate the time of injection. The Middle panel shows the latency to NREM sleep after saline or CNO injection (*P < 0.05 vs. saline; paired t test). The Right panel shows cumulative wakefulness (*P < 0.05 vs. saline; two-way repeated-measures ANOVA). (C) Schematic of the experimental procedure and representative fluorescence images. (Scale bar: 500 µm.) (D) Time spent in wakefulness, the number of wakefulness bouts, and average bout duration during the 3 h following saline or CNO injection (*P < 0.05 vs. saline; Student’s t test). (E) Schematic of the optogenetic stimulation protocol and representative fluorescence images. Light stimulation (10 Hz) was applied during ZT4–6. (Scale bar: 1,000 µm.) (F) Time spent in wakefulness, number of wakefulness bouts, and average bout duration during ZT4–6. “Off” represents the data without light stimulation (*P < 0.05 vs. Off; two-way repeated-measures ANOVA). Data are presented as mean ± SD. Details of the statistical analysis are provided in SI Appendix, Table S3.

Given that the number of c-Fos-positive cells in CRF neurons of the CeA was reduced while that of the IPACL was elevated during exposure to a novel environment (Fig. 1D), we hypothesized that activation of IPACL^CRF neurons is critical for maintaining wakefulness in such contexts. To test this, we used optogenetics to selectively activate IPACL^CRF (Fig. 2E) or CeA ^CRF neurons (SI Appendix, Figs. S4 and S5). Cre-dependent AAVs expressing channelrhodopsin-2 (ChR2) fused to an enhanced yellow fluorescent protein (AAV-CMV-flex-ChR2-EYFP) or enhanced green fluorescent protein (EGFP) (AAV-CMV-flex-EGFP) were injected into the IPACL or CeA, and optical fibers were implanted to enable region-specific light stimulation. Optogenetic activation of IPACL^CRF neurons significantly increased the duration and total amount of wakefulness time compared to EGFP controls (Fig. 2 E and F and SI Appendix, Fig. S6). In contrast, activation of CeA^CRF neurons did not alter wakefulness (SI Appendix, Fig. S4). These results suggest that, unlike CeA^CRF neurons, IPACL^CRF neurons are critical for maintaining wakefulness.

To investigate the neuronal dynamics of CRF neurons in the IPACL during exposure to a novel environment, we performed in vivo single-cell calcium imaging. An AAV encoding the fluorescent calcium indicator, GCaMP6f, was injected into the IPACL of CRF-iCre mice, followed by the implantation of a gradient refractive index (GRIN) lens (Fig. 3A). Mice were placed in their home cage (Familiar-1) for 30 min, then transferred to a novel environment (Novel) for 60 min, and subsequently returned to the home cage (Familiar-2) for an additional 30 min (Fig. 3B). Calcium signals from individual neurons were recorded using a microendoscope (Fig. 3A). Approximately 10% of IPACL^CRF neurons were activated in the novel environment (Fig. 3 C and D and SI Appendix, Fig. S8), consistent with the c-Fos immunostaining results (Fig. 1D). To assess whether handling affected calcium dynamics, we compared calcium activity during the 60 s before and after tail lifting. During the Familiar-1 period, no significant changes were observed in the activity of novel environment-responsive neurons, suggesting that handling alone does not substantially influence their activity (SI Appendix, Fig. S7). We further analyzed the temporal dynamics of calcium activity during exposure to the novel environment. Activity peaked approximately 15 min after placement in the novel setting and gradually declined after about 40 min (Fig. 3E). In contrast, similar transient increases in calcium activity were not observed following handling in the home cage, indicating that the observed activity changes are specifically driven by environmental novelty (Fig. 3F). To exclude the possibility that increased locomotion in the novel environment accounted for the observed neuronal activity increases, we analyzed locomotor speed alongside neuronal activity throughout the entire session (SI Appendix, Figs. S8 and S9). Time-aligned traces of locomotion speed and calcium activity from 25 novel-responsive neurons revealed that neural activation did not consistently follow changes in movement. Furthermore, correlation analysis between locomotion speed and neuronal activity across these neurons showed that only two cells exhibited significant correlations—one positive and one negative—indicating that, overall, neuronal activity in novel-responsive neurons was independent of locomotor behavior. These results suggest that a subset of CRF neurons in the IPACL is selectively activated in response to exposure to a novel environment.

Fig. 3.

Calcium dynamics of CRF neurons in the IPACL during exposure to a novel environment. (A) Schematic of the imaging experiment, showing a representative coronal image of GCaMP6f expression and GRIN lens placement, along with a sample cell map. (Scale bar: 500 µm.) (B) Schematic diagram of the environmental transition protocol. (C) Pie chart showing the proportion of responsive neurons under each environmental condition (n = 255 neurons from N = 4 mice). (D) Heatmap of 30-s binned AUC values illustrating activity across neurons throughout the experiment. The color bar indicates environmental response types, as shown in (B). White dashed lines denote handling time points. (E and F) Temporal changes in calcium activity in neurons responsive to the novel environment: (E) during the Novel period (*P < 0.05, **P < 0.01, ***P < 0.001 vs. Familiar-1 period; Friedman test with Dunn’s post hoc test) and (F) during the Familiar-1 period (vs. 0 to 15 min of Familiar-1 period). Data are shown as medians (center lines), interquartile ranges (boxes), and maximum and minimum values (whiskers).

The amygdala, including the CeA and IPACL, consists of several types of neurons (38). To determine which neurotransmitters are expressed in CRF neurons within the IPACL, we first performed in situ hybridization. Our results revealed that nearly all IPACL^CRF neurons are GABAergic, with approximately 50% coexpressing dynorphin (Dyn) and 10% coexpressing NTS (39) (Fig. 4A). We also confirmed the gene expression profile of CeA^CRF neurons, which was consistent with previous reports (SI Appendix, Fig. S10) (40).

Fig. 4.

Neurotensin in CRF neurons in the IPACL maintains wakefulness in a novel environment. (A) Representative images of fluorescent in situ hybridization (FISH) in the IPACL, showing expression of CRF neurons alongside Vgat, Pdyn, and Nts. (Scale bar: 100 µm.) Quantification of the percentage of Vgat-, Pdyn-, or Nts-positive cells among CRF neurons in the IPACL is shown. (B) Schematic drawing of the experimental design and a representative fluorescence image. AAV encoding sgRNAs targeting Crf, Nts, Pdyn, Vgat, or a control (lacZ), along with Cre-inducible hM3Dq were injected into CRF-iCre; LSL-Cas9 mice. (Scale bar: 500 µm.) Saline or CNO was administered intraperitoneally at ZT4. (C) Time spent in wakefulness, number of wakefulness bouts, and duration of wakefulness during the 3 h following saline or CNO injection are shown (*P < 0.05 vs. saline or lacZ; two-way repeated-measures ANOVA followed by post hoc Tukey–Kramer test). (D) Schematic drawing of the experimental procedure. Mice with Nts deletion (Nts-cKO) and control (lacZ) in CRF neurons in the IPACL were exposed to a novel environment at ZT4. (E) Representative EEG and EMG traces from control (Ctrl) and Nts-cKO mice in a novel environment (Left). Latency to NREM sleep following exposure to the novel environment is shown (middle) (*P < 0.05 vs. control; Student’s t test). Cumulative wakefulness after exposure to the novel environment is shown on the Right. Data are presented as mean ± SD. Details of the statistical analyses are provided in SI Appendix, Table S3.

To identify which neurotransmitters in IPACL^CRF neurons are critical for maintaining wakefulness during exposure to novel environments, we employed in vivo genome editing using the CRISPR/Cas9 system. We injected AAV expressing a U6 promoter-driven guide RNA (gRNA) and Cre-dependent hM3Dq-mCherry (AAV-U6-gRNA-CAG-flex-hM3Dq-mCherry) into the IPACL of CRF-iCre/LSL-SpCas9-EYFP mice. We used gRNAs targeting Crf, Vgat, Pdyn, Nts, and LacZ (as a control) to delete these genes in IPACL^CRF neurons (Fig. 4B). The results of the gene deletion were confirmed by immunostaining and/or a plasmid-based expression system (SI Appendix, Figs. S11 and S12). Three weeks after the AAV injection, we administered saline or CNO to activate CRF neurons and recorded EEG and EMG in a familiar environment. Notably, Nts-conditional knockout (cKO) mice (lacking Nts in IPACL^CRF neurons) showed no differences in wakefulness duration, the number of wakefulness bouts, or bout duration between saline and CNO- injected groups (Fig. 4C and SI Appendix, Fig. S13). In contrast, Vgat-cKO and Pdyn-cKO mice exhibited increased wakefulness in the CNO-injected group compared to the saline group but showed decreased wakefulness compared to LacZ controls following CNO injection (Fig. 4C and SI Appendix, Fig. S13). We also confirmed that there was no significant difference in daily sleep–wake patterns between Nts-cKO and LacZ control mice (SI Appendix, Fig. S14). These results suggest that NTS in IPACL^CRF neurons is essential for maintaining wakefulness in response to novel environmental exposure.

Next, we investigated the role of NTS in IPACL ^CRF neurons in wakefulness induced by a novel environment (Fig. 4D). Nts-cKO mice were transferred from a familiar to a novel environment, and EEG and EMG were recorded. As expected, Nts-cKO mice exhibited a shorter latency to NREM sleep in the novel environment compared to LacZ control mice (Fig. 4E). These findings suggest that NTS in IPACL^CRF neurons is critical for maintaining wakefulness during exposure to novel environments.

To identify the downstream brain regions regulated by IPACL^CRF neurons that contribute to wakefulness during exposure to novel environments, we injected AAVs encoding Cre-dependent tdTomato and synaptophysin-EGFP (a presynaptic marker) into the IPACL of CRF-iCre mice. Analysis of nerve terminal distribution revealed that these neurons primarily projected to the SNr (Fig. 5A). In contrast, downstream mapping of CeA^CRF neurons revealed projections to the BNST, dorsal raphe nucleus, lateral parabrachial nucleus, and other regions (SI Appendix, Fig. S15A). Notably, IPACL^CRF neurons did not project to the CeA, nor did CeA^CRF neurons project to the IPACL (Fig. 5A and SI Appendix, Fig. S15A). To further validate projections from the IPACL to the SNr, we injected red RetroBeads into the SNr. Retrograde labeling revealed red fluorescence in the IPACL, confirming direct projections from IPACL^CRF neurons to the SNr (Fig. 5B). These results suggest that IPACL^CRF neurons innervate the SNr, highlighting a key pathway involved in wakefulness regulation during novel environment exposure.

Fig. 5.

Neurotensin release from the IPACL to the SNr is critical for the maintenance of wakefulness. (A) Schematic representation of the experimental design and fluorescence images. Mixed AAVs were injected into the IPACL of CRF-iCre mice. Red and green colors indicate the axons and presynapses of CRF neurons in the IPACL. Scale bars are 1,000 µm for the full image and 100 µm for the magnified view. (B) Schematic drawing of the experimental design and representative fluorescence images. Retrobeads were injected in the SNr. Scale bars represent 1,000 µm and 200 µm in full and magnified views, respectively. (C) Schematic drawing of the experimental design. hM3Dq-mCherry was expressed in CRF neurons in the IPACL of CRF-iCre mice, and mixed AAVs were injected into the SNr. Saline or CNO was administered via intraperitoneal injection at ZT4. (D) Time spent in wakefulness, number of wakefulness bouts, and duration of wakefulness during the 3 h after saline or CNO injection are shown (*P < 0.05 vs. saline or lacZ; two-way repeated-measures ANOVA followed by post hoc Tukey–Kramer test). (E) Schematic drawing of the experimental design. Mice lacking the neurotensin receptor (Ntsr1) in the SNr were placed in a novel environment at ZT4. (F) Representative EEG and EMG traces of control (ctrl) or Ntsr1-cKO mice in a novel environment are shown on the Left. The latency to NREM sleep following exposure to the novel environment is shown in the Middle (*P < 0.05 vs. familiar; Student’s t test). Cumulative wakefulness after exposure to the novel environment is depicted on the right (*P < 0.05 vs. saline; two-way repeated-measures ANOVA). Data are presented as mean ± SD. Detailed statistical analysis is provided in SI Appendix, Table S3.

To investigate the functional connectivity between IPACL^CRF neurons and the SNr, we used the CRISPR/Cas9 system (Fig. 5C). As NTS receptor 1 (NtsR1) is expressed in the SNr (SI Appendix, Fig. S15B), we generated SNr-specific Ntsr1 conditional knockout (Ntsr1-cKO) mice by injecting AAV gRNAs targeting Ntsr1 gene (Fig. 5D). In addition, we injected AAVs encoding Cre-dependent hM3Dq-mCherry into the IPACL of CRF-iCre mice. In control mice (LacZ gRNA), activation of IPACL^CRF neurons increased wakefulness, consistent with previous results (Figs. 1 and 5). However, in Ntsr1-cKO mice, chemogenetic activation of these neurons failed to increase wakefulness (Fig. 5), suggesting that NTS release from IPACL^CRF neurons to the SNr is involved in wakefulness. To further assess the role of NTS signaling between IPACL^CRF neurons and the SNr during novel environment exposure, we placed Ntsr1-cKO and control mice in a novel environment (Fig. 5 D and E). Compared to controls, Ntsr1-cKO mice exhibited shorter latency to NREM sleep, reduced cumulative wakefulness, and decreased duration of wakefulness per bout (Fig. 5 E and F and SI Appendix, Fig. S16). These findings indicate that NTS release from IPACL^CRF neurons to the SNr is essential for the maintenance of wakefulness in response to novel environments.

Discussion

In this research, by combining optogenetic and chemogenetic manipulations, calcium imaging, CRISPR/Cas9-mediated genome editing, and neuronal tracing approaches, we demonstrate that NTS in IPACL^CRF neurons plays a crucial role in maintaining wakefulness in novel environments. Activation of IPACL^CRF neurons significantly increased wakefulness, while inhibition of these neurons during exposure to a novel environment shortened the latency to NREM sleep. Notably, deletion of the Nts gene in IPACL^CRF neurons also resulted in a shortened latency to NREM sleep under novel environmental conditions. Furthermore, we identified that IPACL^CRF neurons predominantly project to the SNr, and deletion of the Ntsr1 gene in the SNr likewise resulted in a shortened latency to NREM sleep in a novel environment. Overall, this study uncovers a unique function of NTS and its neural circuitry in the maintenance of arousal and highlights the IPACL^CRF-SNr pathway as a critical mechanism for sustaining wakefulness during exposure to novel environments (Fig. 6).

Fig. 6.

Hypothetical neuronal circuit between the IPACL and SNr for the maintenance of wakefulness in novel environments In this model, when animals are exposed to novel environments, CRF-expressing neurons in the IPACL become activated and primarily project to the SNr. These neurons release neurotensin peptides, which are critical for sustaining wakefulness in response to environmental novelty. This circuit highlights the role of NTS release in maintaining arousal, enabling animals to assess and respond to new and potentially threatening stimuli.

The IPACL, CeA, and BNST are anatomically contiguous nuclei within the central extended amygdala, all of which express similar neuropeptides, including NTS. However, recent studies suggest that these subnuclei serve distinct functions in emotional behavior and sleep–wake regulation (35, 41–44). The CeA has been implicated in regulating rapid eye movement (REM) sleep (45). Furthermore, Ma et al. recently reported that NTS-expressing neurons in the CeA are active during NREM sleep, and their activation promotes it. Moreover, CRISPR/Cas9-mediated deletion of Nts or Vgat in the CeA resulted in increased NREM sleep (17). These findings indicate that Nts and Vgat in the CeA are involved in regulating NREM sleep. In contrast, our results demonstrated that NTS within CRF neurons in the IPACL is critical for the maintenance of wakefulness in novel environments. CRF neurons in the CeA and IPACL do not have reciprocal projections and innervate different downstream targets (Fig. 5), leading to distinct regulation of sleep and wakefulness by these neural circuits.

Anatomically, the interstitial nucleus of the posterior limb of the anterior commissure (IPAC) is defined as a subregion of the central extended amygdala, which also includes the BNST and CeA (41). Recent studies have shown that, like other subnuclei of the central extended amygdala, the IPACL is involved in diverse functions related to emotional regulation. For instance, anterior IPACL^CRF neurons have been shown to promote locomotor activity and avoidance behaviors. These neurons project to the SNr and release CRF peptides that act on CRF receptor type 1 on axon terminals originating from neurons in the external globus pallidus (35). Consistent with these findings, our study also showed that the activation of IPACL^CRF neurons increased wakefulness (Fig. 1). However, CRF peptides in this circuitry may not be involved in the maintenance of wakefulness in novel environments, as deletion of the Crf gene in the IPACL did not affect wakefulness when these neurons were activated (Fig. 5). Furthermore, NTS neurons in the IPAC have been shown to promote hedonic eating (46). These neurons are activated by palatable food even when animals are sated, and activation of these neurons enhances hedonic feeding, which could potentially relate to the regulation of wakefulness. However, the NTS neurons in the IPAC examined in that study likely represent a population distinct from the Crf-expressing NTS neurons in the IPACL that we focus on. Although NTS neurons in the anterior IPAC contribute to hedonic feeding (46), NTS neurons in the posterior IPAC (designated here as the IPACL), identified in this study, are associated with the maintenance of wakefulness in novel environments. Furthermore, it has been shown that both innate and learned disgust responses are linked to increased neuronal activity in the IPACL, which may also contribute to increased wakefulness (37). The IPACL may, therefore, contain multiple subpopulations, each regulating different behavioral and physiological functions. Further studies will be required to elucidate additional roles of the IPACL in behavior and physiology.

Our findings revealed that IPACL^CRF neurons predominantly innervate the SNr, and the release of NTS peptide from these neurons likely influences the activity of Ntsr1-expressing neurons in the SNr. In the brain, NTS is known to modulate dopaminergic activity, particularly in regions such as the VTA and SNr, both of which are involved in reward processing and motor control (47). Ntsr1 expression is observed in the SNr (SI Appendix, Fig. S15), and these receptors are known to couple with multiple signaling pathways, including Gs, Gq/11, Gi/o, and G12/13 (48). Furthermore, the inactivation of GABAergic neurons in the SNr (such as parvalbumin and Gad2 neurons) has been reported to increase locomotor activity and reduce sleep (49). Our data also indicated that GABA and Dyn in the IPACL^CRF neurons have a limited contribution to the maintenance of wakefulness (Fig. 5). Given that the receptors for these neurotransmitters are primarily inhibitory, it is likely that the release of NTS, GABA, and Dyn from IPACL^CRF neurons modulates the activity of GABAergic neurons in the SNr, ultimately influencing the maintenance of wakefulness during exposure to novel environments.

CRF peptides are known to mediate physiological stress responses, including activation of the HPA axis and the sympathetic nervous system, primarily due to the characteristics of CRF neurons in the PVN (50). Notably, activation of CRF neurons in the PVN has been shown to increase wakefulness (10). Consistent with this study, we also observed an increase in c-Fos-positive cells in the PVN in response to exposure to a novel environment (Fig. 1D). Therefore, CRF neurons in both the PVN and IPACL are essential for maintaining wakefulness in novel environments.

In conclusion, our findings demonstrate that NTS peptides in CRF neurons within the IPACL, which innervate the SNr, play a crucial role in maintaining wakefulness during exposure to novel environments. When exposed to unfamiliar surroundings, animals must remain alert to respond to unexpected threats, utilizing this neuronal circuit to effectively regulate their sleep and wake states.

Methods

Animal Usage.

All experimental procedures were conducted in accordance with the guidelines of the Institutional Animal Care and Use Committees at the Research Institute of Environmental Medicine, Nagoya University (approval numbers: R240003, R240004). Every effort was made to reduce animal suffering and minimize the number of animals used in the experiments.

Animals.

All male mice (3 to 8 mo) used in the experiments were of a C57BL/6J background and housed under a 12-h light–dark cycle (lights on from 08:00 to 20:00). The room temperature was maintained at 23 ± 2 °C, with humidity levels between 40% and 60%. Light intensity in the housing area ranged from 200 to 400 lx. Food and water were available ad libitum. The following mouse lines were used: Crh^{tm2(icre)Ksak} (CRF-iCre) (51), FVB.129(B6)-Gt(ROSA)26Sor^{tm1.1(CAG-cas9*,-EGFP)Fezh}/J (LSL-SpCas9-EGFP) (52), and Slc32a1^tm2(cre)Lowl/J (53) (Vgat-IRES-Cre) (54) mice. To eliminate human bias during behavioral and sleep recordings, all experiments were conducted using blinded testing procedures.

Viruses.

AAV-CAG-DIO-GCaMP6f-WPRE was produced and purified using the baculovirus/Sf9 expression system as previously described (55). Other AAVs were produced using a triple-transfection, helper-free method and purified as previously described (56, 57). Titers of recombinant AAV vectors were determined by qPCR.

Stereotaxic Injections.

Mice were anesthetized using 2% isoflurane (095–06573, FUJIFILM Wako Pure Chemical Corporation) or a combination of three anesthetic agents (3MIX) (58) and secured in a stereotaxic frame (Model 940, David Kopf Instruments, USA). After opening the scalp, a small hole was drilled in the skull above the injection site. A glass pipette (GC150-10; Harvard Apparatus), pulled using a puller (P-97, Sutter Instrument), was used to inject AAVs or drugs through a Nanoject III (Drummond Scientific Company) or a microinjector (BJ-110, BEX CO., LTD, Japan). The details of the injected materials in all experiments are summarized in SI Appendix, Table S1.

EEG/EMG Surgery and Vigilance Analysis.

Male mice were anesthetized with 2% isoflurane or 3MIX and implanted with three EEG electrodes (U-1430–01, Wilco) on the skull and two EMG electrodes (AS633, Cooner wire) in the rhomboid muscle at 10 wk of age, as previously described (59). To manage postoperative pain and inflammation, carprofen (RimadylR, Zoetis) was administered at a dose of 20 mg/kg (subcutaneous injection) on the day of surgery and the following day. After surgery, mice were allowed to recover for at least 7 d. Subsequently, a cable with a slip ring (Kissei Comtec Co., Ltd) was attached to each mouse in its cage for 7 d before starting EEG and EMG recordings.

Sleep/Wake Recordings.

EEG and EMG signals were filtered at 1.5 to 30 Hz for EEG and 15 to 300 Hz for EMG and amplified using an amplifier (AB-610J, Nihon Kohden). The digital sampling rate was set at 128 Hz. Animal behavior was monitored using a charged-coupled device (CCD) video camera (SPK-E700CHP1, Keiyo Techno, Japan) along with an infrared activity sensor (Kissei Comtec). All EEG and EMG data were recorded using VitalRecorder software (Kissei Comtec).

Vigilance State Analysis.

EEG analysis was performed using the Fast Fourier Transform. Power spectra were calculated for the 0 to 10 Hz range with 1 Hz resolution, focusing on the δ (1 ≤ δ < 6 Hz) and θ (6 ≤ θ < 11 Hz) frequency bands. EEG and EMG data were automatically screened in 4-s epochs by SleepSign software (Kissei Comtec) and classified as previously described (59). Wakefulness was defined by high EMG amplitude or locomotor activity with low EEG amplitude, NREM sleep by low EMG amplitude and high EEG δ power, and REM sleep by low EMG amplitude with >50% θ activity. Vigilance states were automatically classified using SleepSign and manually corrected where necessary. A change in the vigilance state was confirmed when three consecutive epochs exhibited distinct EEG and EMG characteristics as defined above.

Optogenetics Surgery.

After the AAV injection, mice were secured in a stereotaxic frame, and EEG and EMG electrodes were implanted. A hole was drilled in the skull above the IPACL (anteroposterior [AP]: −0.6, mediolateral [ML]: ±2.6, dorsoventral [DV]: −4.1) or CeA (AP: −1.4, ML: ±2.9, DV: −4.0). Optical fibers (400-µm diameter core and 0.39 numerical aperture [NA]; Kyocera Corporation, Japan) equipped with a copper ferrule (1.25 mm in diameter) were implanted. To prevent light leakage, the optic fibers and EEG screws were secured with dental cement (Repairsin, GC) mixed with bamboo charcoal powder. The optic fiber was connected to a light-emitting diode (LED) light source (470 nm, 5 mW at the tip of the fiber). Activation of CRF neurons was achieved using blue light at a frequency of 10 Hz with 4 ms pulse durations, controlled by a pulse generator (SEN 3401; Nihon Kohden, Tokyo, Japan) connected to the LED driver.

Calcium Imaging Surgery.

Mouse surgeries were performed as previously described (60). More than 1 wk after the injection of AAV (AAV-CAG-DIO-GCaMP6f-WPRE), a customized GRIN lens probe (length: 4.81 mm, diameter: 0.6 mm, NA: 0.5; Inscopix, CA) was implanted above the right IPACL (AP: −0.65, ML: +2.65, DV: −4.4). The implanted lens probe was secured to the skull using UV-curable optical adhesive (NOA-81; Norland Products, NJ). The skull was then covered with Super-Bond (C&B Kit; Sun Medical, Shiga, Japan) and dental cement (Repairsin). A stainless-steel bar was attached for head fixation. Once sufficient GCaMP6f signals were confirmed, a baseplate (Inscopix) was secured using blue light-curing resin (Flow-It ALC; Pentron, Tokyo, Japan). Post hoc histological analyses were conducted after imaging to verify GCaMP6f expression and the placement of the implanted lens.

Single-Cell In Vivo Calcium Imaging.

More than 2 d after baseplate mounting, mice were prepared for imaging. At least 1 d before the experiment, a miniature integrated microscope system (nVista HD 2.0; Inscopix) was attached to each mouse, and mice were acclimated to the test box for 1 h. On the day of the experiment, the nVista system was reattached, and mice were allowed to acclimate for 30 min before the start of calcium imaging. Images were acquired using Inscopix Data Acquisition Software (ver. 2.0.4; Inscopix) at 8 frames per second, with LED power set to 12% and a gain of 3.5. Simultaneously, mouse behavior was recorded using a video camera triggered by the nVista system. Recordings were conducted in three sessions: 30 min in the home cage (Familiar-1), 60 min in a novel environment (Novel), and a final 30 min back in the home cage (Familiar-2). During Familiar-1, the mouse was briefly picked up and returned to the cage at the 15-min mark during the first home cage period (Fig. 2B).

Imaging Data Analysis.

Imaging data were preprocessed using Inscopix Data Processing Software (IDPS ver. 1.9.2; Inscopix) as follows: spatial down-sampling by a factor of 2, spatial band-pass filtering with low and high cut-offs of 0.005 and 0.5, respectively, and motion corrected. Following preprocessing, calcium transients from individual neurons were extracted using constrained non-negative matrix factorization for microendoscopic data (CNMFe) (61), with the following parameters: trace output units = df over noise, minimum pixel correlation = 0.8, minimum peak-to-noise ratio = 8, merging threshold = 0.7, ring size factor = 1.4, and closing kernel size = 1. All extracted traces were manually inspected using the IDPS interface. Traces reflecting signals from multiple neurons or noncellular sources were excluded. Subsequent data processing and analysis were performed using custom-written MATLAB scripts (MATLAB R2023a; Mathworks, MA). Traces were smoothed using the robust locally weighted scatterplot smoothing (RLOWESS) model to reduce background fluctuations caused by prolonged recordings and then z-scored. From the obtained z-score values, the distributions of signal and noise were estimated for each neuron using a Gaussian mixture model, and a threshold was set at the 99.99% cumulative probability of the noise. Signal values below the threshold were considered as noise and set to zero. The area under the curve (AUC) was then calculated for each 2-s bin. To identify substantially responsive neurons in the novel environment, we performed 5,000 bootstrap resampling iterations on 2-s binned data from the three periods (Familiar-1, Novel, and Familiar-2), excluding 20-s before and after the mouse was lifted by the tail. Distributions were estimated using kernel density estimation, and neurons were identified as significantly activated if their activity distributions overlapped with other conditions at a probability of ≤2.5%.

Behavioral Assessments.

Locomotor activity during in vivo imaging: Video recordings of mouse behavior during the in vivo calcium imaging session were acquired from above using a camera synchronized with the nVista system. To quantify locomotion, three body points—the nose, the neck (base of the head), and the base of the tail—were tracked using DeepLabCut (ver. 2.3.5) (62). Locomotion speed was calculated based on frame-by-frame displacements of the neck point, and the resulting speed values were smoothed using a 1-s moving average window.

Elevated Plus Maze (EPM) test: The EPM apparatus consisted of two open and two closed arms (25 × 5 cm each) extending from a central platform (5 × 5 cm). The closed arms had walls 20 cm high. The apparatus was elevated 40 cm above the floor, and ambient lighting was maintained at 6 lx. Mice were placed on the central platform, and their behavior was recorded for 15 min using a CCD camera. Data were analyzed using LimeLight software (Actimetrics, Wilmette).

Open Field Test (OF): The OF test was conducted in a square chamber (45 cm W × 30 cm D × 30 cm H) under 20 lx lighting. The field was virtually divided into two zones: a central area (13 × 10 cm) and a peripheral area. Mice were recorded for 30 min using a CCD camera, and the footage was analyzed automatically with Ethovision software.

Light-Dark Box Test (LD Box): The LD box consisted of a dark compartment (15 cm W × 45 cm D × 30 cm H) and a light compartment (30 cm W × 45 cm D), illuminated at 20 lx. The entire apparatus was elevated 60 cm above the floor. Behavior was recorded for 15 min using a CCD camera and analyzed automatically with Ethovision software.

Fluorescence In Situ RNA Hybridization by RNAscope.

Mice were deeply anesthetized with 3% isoflurane and perfused sequentially with saline, followed by 10% formalin (066–03847, FUJIFILM Wako). After perfusion, brains were carefully removed and postfixed in 10% formalin overnight at 4 °C, then transferred to a 30% sucrose solution in phosphate-buffered saline (PBS) for at least 2 d. The brains were then frozen in embedding solution (4583, Sakura Finetek Japan, Japan) and stored at −80 °C.

Frozen brains were sectioned into 20 µm slices using a cryostat (CM3050-S, Leica Microsystems K.K., Japan) and mounted onto glass slides. Slides were processed using the RNAscope Multiplex Fluorescent v2 Assay (#323100, Advanced Cell Diagnostics, Hayward, CA) according to the manufacturer’s standard protocol. Briefly, slides were treated with hydrogen peroxide at room temperature for 10 min, followed by boiling in target retrieval reagent at 98 to 102 °C for 5 min, and protease digestion (Protease III) at 40 °C in a HybEZ hybridization oven (Advanced Cell Diagnostics) for 30 min.

Subsequently, the slides were hybridized with target probes (CRF: 316091, Vgat: 319191-C4, Nts: 420441-C2, pDyn: 318771-C3) at 40 °C for 2 h and washed twice with wash buffer (WB) at room temperature for 2 min each. Sections were sequentially incubated with amplifier (AMP)1 (30 min), AMP2 (30 min), and AMP3 (15 min) buffers, with WB washes between steps. For fluorescent detection, RNA probes were conjugated to Opal 520, Cy3, or Cy5 using the TSA Plus Fluorescence system (Perkin Elmer, Waltham, MA). Finally, slices were mounted with ProLong Gold Antifade Mountant (Thermo Fisher Scientific). One-quarter of the brain slices from three mice were randomly selected for analysis.

Validation of the CRISPR/Cas9 Method: Cut-On System Using Human Embryonic Kidney (HEK) Cells.

To validate the efficiency of the CRISPR/Cas9 gene-editing system, we employed a “cut-on” strategy in HEK cells. This method involves introducing targeted double-strand breaks into a plasmid using CRISPR/Cas9, followed by activation of a reporter system that indicates successful gene editing. Specifically, a 40-bp gRNA-targeted sequence was inserted between the ATG start codon and the EYFP coding sequence, resulting in a frameshift that prevented EYFP expression. Upon introduction of the CRISPR/Cas9 system with the appropriate gRNA, targeted deletions restore the correct reading frame, allowing for EYFP protein production (SI Appendix, Fig. S12A). The gRNA sequence and cut-on system details are provided in SI Appendix, Table S2.

Electrophysiology Recording.

After 3 to 4 wk of stereotaxic AAV injection, mice were deeply anesthetized by using isoflurane and decapitated. Brains were rapidly isolated and chilled in ice-cold bubbled (95% O2 and 5% CO2) cutting solution (mM: 110 K-Gluconate, 15 KCl, 0.05 EGTA, 5 HEPES, 26.2 NaHCO3, 25 Glucose, 3.3 MgCl2, 0.0015 3-((+/−)-2-Carboxypiperazin-4-yl)-Propyl-1-Phosphonic acid (CPP)). Coronal brain sections of 300 µm thickness were made by using a vibratome (VT-1200S, Leica). The slices were incubated in a bubbled (95% O2 and 5% CO2) bath solution (mM: 124 NaCl, 3 KCl, 2 MgCl2, 2 CaCl2, 1.23 NaH2PO4.2H2O, 26 NaHCO3, 25 Glucose) at 35 °C for 30 min following another 30 min incubation at room temperature in the same solution. After the incubation period, the brain slices were placed in a recording chamber (RC-26G, Warner Instruments, Hamden, CT) which was perfused with 95% O2 and 5% CO2 bubbled bath solution at a rate of 1.5 mL/min using a peristaltic pump (Miniplus3, Gilson, USA). An infrared camera (C3077-78, Hamamatsu Photonics, Hamamatsu, Japan) was installed in an upright fluorescence microscope (BX51WI, Olympus, Tokyo, Japan) together with an electron-multiplying charge-coupled device camera (Evolve 512 delta, Photometrics, Tucsaon, AZ) and a monitor visualized both images. Glass micropipettes were prepared from borosilicate glass capillaries (GC150-10, Harvard Apparatus, Cambridge, MA) using a horizontal puller (P1000 Sutter Instrument, Novato, CA) maintaining a pipette resistance of 4 to 8 MΩ. During recordings from neurons expressing ACR2-mCherry, patch pipettes were filled with a K-gluconate-based internal solution composed of (in mM): 138 K-gluconate, 10 HEPES, 8 NaCl, 0.2 EGTA, 2 Mg-ATP, and 0.5 Na2-GTP, adjusted to a pH of 7.3 with KOH. The osmolality of the solution was maintained between 285 and 290 mOsm. After confirming the expression, the pipette was positioned toward the cell. Manual positive pressure was applied to the patch pipette, which was released when the tip made contact with the cell membrane, forming a giga seal. Subsequently, the patch membrane was ruptured by briefly applying strong suction to create a whole-cell configuration. During the recordings, the electrophysiological properties of the cells were continuously monitored using an Axopatch 200B amplifier (Axon Instruments, Molecular Devices, Sunnyvale, CA). The output signals were low-pass filtered at 5 kHz and digitized at a sampling rate of 10 kHz. Patch clamp data were recorded through an analog-to-digital (AD) converter (Digidata 1550A, Molecular Devices) with pClamp 10.7 software (Molecular Devices). Blue light stimulation (475 ± 17.5 nm, 9.25 mW/mm²) was generated by a light source that utilized a light-emitting diode (Spectra Light Engine, Lumencor, USA) and directed to the microscope stage via a 1.0 cm diameter optical fiber. For chemogenetic activation and inhibition, CNO (30 μM, Enzo Life Sciences, Farmingdale, NY) was dissolved in aCSF and applied for 2 min.

Immunohistochemistry.

Mouse brains were isolated as described in the in-situ RNA hybridization section above. Coronal brain sections (40 µm thick) were prepared and placed in PBS containing 0.25% Triton X-100 (35501–15, Nacalai Tesque, Japan) and 1% bovine serum albumin (A7905-500G, MilliporeSigma), referred to as PBS-BX. Sections were washed three times for 10 min each. The brain sections were then incubated overnight at 4 °C in PBS-BX containing one of the following primary antibodies: rabbit anti-c-Fos antibody (1:1,000, ab222699, Abcam), rabbit anti-NTS antibody (1:1,000, AB5496, MilliporeSigma), guinea pig anti-prodynorphin antibody (1:1,000, AB5519, MilliporeSigma), rabbit anti-CRF antibody (1:1,000, HAC-HM04-01RBP90, Gunma University), or rabbit anti-VGAT (1:1,000, SAB2700790, SIGMA). After primary antibody incubation, sections were washed three times for 10 min each in PBS-BX and subsequently incubated for 2 h at room temperature and/or overnight at 4 °C with one of the following secondary antibodies: CF 594 donkey anti-rabbit immunoglobulin G (IgG) (1:1,000, 20015, Biotium, Inc.), CF 647 donkey anti-rabbit IgG (1:1,000, 20047, Biotium, Inc.), CF 647 donkey anti-guinea pig IgG (1:1,000, 20047, Biotium, Inc.), or CF 594 donkey anti-guinea pig IgG (1:1,000, 20152, Biotium, Inc.) in PBS-BX. Following secondary antibody incubation, sections were again washed three times for 10 min each in PBS-BX and then stained with 4’,6-diamidino-2-phenylindole for nuclear visualization. Finally, sections were mounted, and coverslips were secured using 50% glycerol in PBS. For CRF peptide immunostaining, colchicine (3 μL, 10 μg/μL) was administered into the lateral ventricles of mice, and brains were fixed 36 to 48 h later.

Acquisition of Fluorescence Images.

Fluorescence images were acquired using either a fluorescent microscope (BZ-X710, KEYENCE) or a confocal microscope (LSM710, Carl Zeiss). Exposure times for the BZ-X710 and gain settings for the LSM710 were individually optimized for each sample to ensure clear and accurate imaging.

Statistical Analyses.

Statistical analyses were conducted using OriginPro 2019 (OriginLab, Northampton, MA), Prism (9.5.0, GraphPad), or Microsoft Excel. All statistical information is in SI Appendix, Table S3.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

Study data are included in the article and/or SI Appendix.