Brainwide dopamine dynamics across sleep-wake transitions

Article activity feed

1. 

   eLife

   May 11, 2026

   eLife Assessment

   This study provides important insights regarding the temporal dynamics of dopamine across sleep/wake transitions in several brain areas. Using multi-site fiber photometry combined with EEG/EMG recordings, the study revealed heterogenous dynamics across both cortical and several subcortical areas. Although the evidence for these observations is solid, evidence for the proposed mechanisms driving DA dynamics is incomplete. Overall, the study may have a substantial impact on several fields working on the neurobiology of DA signaling.

   Read the original source
2. 

   eLife

   May 11, 2026

   Reviewer #1 (Public review):

   Summary:

   In this manuscript, Chen, Tu, and Lu focused on how brain-wide dopamine release dynamically changes during sleep/wake state transitions. Using multi-site fiber photometry to monitor DA release, alongside simultaneous EEG and EMG recordings, the authors show distinct DA dynamics during transitions from NREM to WAKE, REM to WAKE, WAKE to NREM, and NREM to REM. Next, they analyze temporal coordination between regions using cross-correlation analysis. Finally, chemogenetic activation of VTA or DRN but not SNc dopamine neurons is shown to promote wakefulness.

   Strengths:

   The manuscript addresses an interesting question: how brainwide dopamine activity evolves across sleep/wake transitions. The combination of multi-site DA recordings with simultaneous EEG/EMG monitoring is technically sophisticated. The …

   Reviewer #1 (Public review):

   Summary:

   In this manuscript, Chen, Tu, and Lu focused on how brain-wide dopamine release dynamically changes during sleep/wake state transitions. Using multi-site fiber photometry to monitor DA release, alongside simultaneous EEG and EMG recordings, the authors show distinct DA dynamics during transitions from NREM to WAKE, REM to WAKE, WAKE to NREM, and NREM to REM. Next, they analyze temporal coordination between regions using cross-correlation analysis. Finally, chemogenetic activation of VTA or DRN but not SNc dopamine neurons is shown to promote wakefulness.

   Strengths:

   The manuscript addresses an interesting question: how brainwide dopamine activity evolves across sleep/wake transitions. The combination of multi-site DA recordings with simultaneous EEG/EMG monitoring is technically sophisticated. The experimental logic is generally clear, and the dataset is rich. The result has several interesting observations.

   Weaknesses:

   The authors used the GRAB-DA2m sensor to monitor dopamine release. Although DA2m exhibits higher affinity for dopamine compared to NE (around 15-fold difference in EC50 in HEK cell assays), it is still possible that NE contributes to the recorded signals, particularly during sleep/wake transitions when locus coeruleus activity is strongly modulated. Given the widespread and state-dependent dynamics of NE, this potentially needs to be addressed.

   Similarly, the chemogenetic experiments rely on CNO to activate hM3Dq-expressing dopamine neurons. However, it is well established that CNO can be converted to clozapine in rodents, and clozapine itself is known to influence sleep/wake. Although the authors included non-hM3Dq-expressing mice as controls, the potential confounding effects of clozapine on sleep regulation remain a concern.

   Midbrain dopamine neurons exhibit both tonic and phasic firing patterns. In Figure 1, most reported dopamine transitions appear relatively slow. However, some faster, phasic-like components are observable. For example, in NAc-L during REM-to-WAKE transitions, there are 2 phasic-like decreases between −20 and 0 s. The authors used laser-evoked stimulation experiments in the VTA and DRN and showed that 2 s versus 10 s stimulation produces distinct dopamine kinetics, suggesting that different firing patterns generate distinct DA dynamics. Moreover, the temporal profiles vary not only across regions but also across transitions within the same region. For example, in CeA, the NREM-to-WAKE transition shows a relatively rapid decrease, whereas REM-to-WAKE displays a much slower decline. Similarly, some regions (e.g., NAc-L NREM-to-WAKE, DRN REM-to-WAKE) show faster changes, while others (e.g., mPFC WAKE-to-NREM, VTA NREM-to-WAKE) show slower kinetics. These observations argue against a simple region-specific explanation and instead suggest that distinct firing modes may differentially contribute depending on transition type.

   While cross-correlation analysis provides insight into the temporal coordination of DA signals across regions, several limitations should be considered. Sleep/wake transitions are inherently non-stationary events, whereas cross-correlation assumes relatively stable signal properties within the analysis window. This mismatch may bias lag estimates and obscure transient lead-lag relationships. Moreover, the temporal resolution of fiber photometry and the kinetics of genetically encoded DA sensors limit the precision with which timing relationships can be interpreted, particularly for sub-second lags.

   In the Introduction, the authors state that they aim to address 'which dopaminergic populations causally drive these patterns.' However, the chemogenetic approach used operates on a relatively slow timescale: CNO-induced activation takes 15-30 minutes to produce effects, and the induced changes are long-lasting. In contrast, the dopamine transitions described in Figure 1 occur on a much faster timescale compared to CNO manipulation. Thus, while chemogenetic activation demonstrates that stimulating VTA or DRN dopamine neurons promotes wakefulness, it does not directly establish that these populations causally drive the rapid transition-related DA dynamics observed in the photometry recordings.

   Read the original source
3. 

   eLife

   May 11, 2026

   Reviewer #2 (Public review):

   In "Brainwide dopamine dynamics across sleep-wake transitions", Chen et al. provide a thorough description of how dopamine dynamics fluctuate across sleep-wake transitions and in transitions between sleep states. To achieve this, the authors used multi-channel fiber photometry and a genetically encoded fluorescent dopamine reporter to simultaneously measure dopamine dynamics in 8 brain regions. They also used EEG measurements to precisely quantify and time transitions between sleep states and wakefulness. Finally, the authors used channelrhodopsin to examine dopamine dynamics following subregion stimulation and chemogenetics to test the causal relationship between activation of distinct dopamine neuron populations and their effects on sleep state.

   The conclusions made by the authors in this study are modest …

   Reviewer #2 (Public review):

   In "Brainwide dopamine dynamics across sleep-wake transitions", Chen et al. provide a thorough description of how dopamine dynamics fluctuate across sleep-wake transitions and in transitions between sleep states. To achieve this, the authors used multi-channel fiber photometry and a genetically encoded fluorescent dopamine reporter to simultaneously measure dopamine dynamics in 8 brain regions. They also used EEG measurements to precisely quantify and time transitions between sleep states and wakefulness. Finally, the authors used channelrhodopsin to examine dopamine dynamics following subregion stimulation and chemogenetics to test the causal relationship between activation of distinct dopamine neuron populations and their effects on sleep state.

   The conclusions made by the authors in this study are modest and appropriate given the largely observational nature of the principal findings. The use of optogenetics to probe regional dopamine signaling following activation of distinct nuclei is interesting, but not entirely novel and constrained in interpretability. Similarly, the chemogenetics experiment largely confirms previous studies, which the authors correctly cited in the text.

   The principal findings of this study are based on strong methodological and analytical methods. Implanting 8 optical fibers in a single mouse, along with EEG/EMG electrodes, is technically challenging, providing valuable, simultaneous measurements of dopamine fluctuations across the brain. This enables the strong correlational and time-locked analyses performed by the authors in Figure 2. What's more, the use of EEG/EMG electrodes provides time-locked descriptions of sleep states, enabling precise comparisons between the dopamine signal and sleep state transitions.

   The paper has some weaknesses that the authors could address. The analyses in Figure 1 could be strengthened to show how dopamine changes during transitions between specific sleep states. The injection sites for channelrhodopsin and chemogenetic viruses could be validated to strengthen the interpretation of those results. Also, a stronger justification for the experiments conducted in Figure 3 could be provided, as they seem unrelated to the present study.

   Overall, this study has strong descriptive power, convincingly showing how dopamine fluctuates across sleep states. Some of the other aspects of the paper, however, are somewhat limited in novelty and interpretation.

   Read the original source
4. 

   eLife

   May 11, 2026

   Author response:

   Public Reviews:

   Reviewer #1 (Public review):

   Summary:

   In this manuscript, Chen, Tu, and Lu focused on how brain-wide dopamine release dynamically changes during sleep/wake state transitions. Using multi-site fiber photometry to monitor DA release, alongside simultaneous EEG and EMG recordings, the authors show distinct DA dynamics during transitions from NREM to WAKE, REM to WAKE, WAKE to NREM, and NREM to REM. Next, they analyze temporal coordination between regions using cross-correlation analysis. Finally, chemogenetic activation of VTA or DRN but not SNc dopamine neurons is shown to promote wakefulness.

   Strengths:

   The manuscript addresses an interesting question: how brainwide dopamine activity evolves across sleep/wake transitions. The combination of multi-site DA recordings with simultaneous EEG/EMG monitoring is …

   Author response:

   Public Reviews:

   Reviewer #1 (Public review):

   Summary:

   In this manuscript, Chen, Tu, and Lu focused on how brain-wide dopamine release dynamically changes during sleep/wake state transitions. Using multi-site fiber photometry to monitor DA release, alongside simultaneous EEG and EMG recordings, the authors show distinct DA dynamics during transitions from NREM to WAKE, REM to WAKE, WAKE to NREM, and NREM to REM. Next, they analyze temporal coordination between regions using cross-correlation analysis. Finally, chemogenetic activation of VTA or DRN but not SNc dopamine neurons is shown to promote wakefulness.

   Strengths:

   The manuscript addresses an interesting question: how brainwide dopamine activity evolves across sleep/wake transitions. The combination of multi-site DA recordings with simultaneous EEG/EMG monitoring is technically sophisticated. The experimental logic is generally clear, and the dataset is rich. The result has several interesting observations.

   Weaknesses:

   The authors used the GRAB-DA2m sensor to monitor dopamine release. Although DA2m exhibits higher affinity for dopamine compared to NE (around 15-fold difference in EC50 in HEK cell assays), it is still possible that NE contributes to the recorded signals, particularly during sleep/wake transitions when locus coeruleus activity is strongly modulated. Given the widespread and state-dependent dynamics of NE, this potentially needs to be addressed.

   We thank the reviewer for raising this important methodological consideration. While we acknowledge that a minor contribution from norepinephrine (NE) to the DA2m signal cannot be categorically excluded, several convergent lines of evidence give us confidence that the signals we recorded primarily reflect dopamine release.

   First, DA2m has substantially lower affinity for NE compared to dopamine. The reported EC₅₀ for NE is ~1200 nM [1], which is ~15-fold higher than for dopamine. In contrast, extracellular NE levels in the prefrontal cortex are typically in the low nanomolar range (generally <5 nM under basal conditions) [2,3]. Because physiological NE concentrations are orders of magnitude below the sensor’s EC₅₀ threshold, NE is highly unlikely to drive significant DA2m activation in vivo.

   Second, our optogenetic experiments provide direct functional validation. The targeted stimulation of midbrain dopaminergic neurons elicited robust DA2m signal responses across both cortical and subcortical brain areas. This confirms that the sensor reliably captures evoked dopamine release within our specific experimental paradigm.

   Finally, the spontaneous DA2m signal dynamics we observed across sleep-wake states functionally diverge from previously reported patterns of cortical NE release [4]. For example, in Figure 1C, our DA2m recordings in the mPFC revealed high activity during wakefulness, alongside pronounced, sharp changes during NREM-to-WAKE transitions. In contrast, prior study [4] show that NE exhibits comparatively mild fluctuations during wakefulness and transitions between NREM. This temporal and kinetic divergence further supports that our recorded signals isolate region-specific dopaminergic dynamics rather than generalized NE arousal activity.

   Taken together, these physiological, functional, and kinetic distinctions indicate that while a negligible contribution from NE cannot be entirely ruled out, it is highly unlikely to account for a substantial portion of the DA2m signals observed during sleep-wake transitions in our study.

   Similarly, the chemogenetic experiments rely on CNO to activate hM3Dq-expressing dopamine neurons. However, it is well established that CNO can be converted to clozapine in rodents, and clozapine itself is known to influence sleep/wake. Although the authors included non-hM3Dq-expressing mice as controls, the potential confounding effects of clozapine on sleep regulation remain a concern.

   We appreciate the reviewer raising this important point regarding the metabolism of CNO. We are aware of the evidence suggesting that CNO can undergo back-metabolism to clozapine in rodents, which could potentially exert independent effects on sleep-wake architecture. To mitigate this concern, we strictly employed several experimental safeguards:

   (A) Non-hM3Dq Control Group: As noted by the reviewer, we included a cohort of mice that did not express the hM3Dq receptor but received the same dosage of CNO (1 mg/kg). In these animals, we observed no significant alterations in sleep-wake states compared to saline baseline (Figure S3), suggesting that at this dosage, any clozapine produced was below the threshold for behavioral modulation of sleep.

   (B) Dosage Selection: We utilized a relatively low dose of CNO (1 mg/kg), which is widely reported in the literature to minimize the accumulation of clozapine to levels that would interfere with EEG-defined sleep states in rodents [5]. Furthermore, studies have demonstrated that while higher doses of CNO (e.g., 5–10 mg/kg) can produce clozapinelike effects on sleep architecture, lower doses around 1 mg/kg do not yield significant alterations in cortical EEG power distribution or sleep-wake amounts in control animals [6,7].

   Midbrain dopamine neurons exhibit both tonic and phasic firing patterns. In Figure 1, most reported dopamine transitions appear relatively slow. However, some faster, phasic-like components are observable. For example, in NAc-L during REM-to-WAKE transitions, there are 2 phasic-like decreases between −20 and 0 s. The authors used laser-evoked stimulation experiments in the VTA and DRN and showed that 2 s versus 10 s stimulation produces distinct dopamine kinetics, suggesting that different firing patterns generate distinct DA dynamics. Moreover, the temporal profiles vary not only across regions but also across transitions within the same region. For example, in CeA, the NREM-to-WAKE transition shows a relatively rapid decrease, whereas REM-to-WAKE displays a much slower decline. Similarly, some regions (e.g., NAc-L NREM-to-WAKE, DRN REM-toWAKE) show faster changes, while others (e.g., mPFC WAKE-to-NREM, VTA NREM-toWAKE) show slower kinetics. These observations argue against a simple region-specific explanation and instead suggest that distinct firing modes may differentially contribute depending on transition type.

   We thank the reviewer for this insightful comment. We agree that midbrain dopamine neurons exhibit both tonic and phasic action-potential firing patterns. As summarized by Grace et al., dopamine neurons recorded using in vivo electrophysiology can display a slow, irregular, single-spike “tonic” firing pattern, typically around 2–10 Hz, as well as burst-like “phasic” firing patterns [8].

   However, our recordings were performed using GRAB-DA2m fiber photometry. Therefore, our measurements reflect extracellular dopamine dynamics in the recorded target regions rather than the action-potential firing patterns of midbrain dopamine neurons. GRABDA2m has subsecond sensor kinetics and is suitable for detecting extracellular dopamine transients occurring over hundreds of milliseconds to seconds, as well as slower dynamics occurring over seconds to tens of seconds [1], which matches the timescale of the sleep–wake transition-related dynamics observed in previous studies [9,10]. Nevertheless, GRAB-DA2m fiber photometry in our study does not directly resolve dopamine neuron spike timing or distinguish tonic from phasic firing modes. Accordingly, we interpret our signals as extracellular dopamine concentration dynamics rather than as direct measurements of tonic or phasic neuronal firing.

   Therefore, the transition-aligned dopamine signals shown in Figure 1 should be interpreted as dopamine dynamics occurring over seconds-to-tens-of-seconds around sleep–wake transitions, rather than as dopamine neuron firing patterns. In addition, these traces represent GRAB-DA2m signals averaged across sessions and mice within a ±30 s window centered on each sleep/wake transition. Thus, they do not necessarily represent individual dopamine transient patterns on single transitions. We also acknowledge the reviewer’s observation that faster phasic-like components are visible in some traces, including the decreases in the NAc-L preceding REM-to-WAKE transitions. Direct electrophysiological recordings of dopamine neuron firing during sleep–wake transitions would be useful in future studies to determine how tonic and phasic firing modes contribute to the observed dopamine dynamics.

   In the laser-evoked stimulation experiments shown in Figure 3, we thank the reviewer for the thoughtful interpretation. The results indicate that different stimulation durations can produce distinct dopamine release dynamics in downstream projection regions. Moreover, prolonged optogenetic stimulation was associated with more sustained dopamine responses, suggesting that the temporal profile of extracellular dopamine dynamics depends, at least in part, on the duration and region of dopaminergic input [1]. We also agree with the reviewer that the temporal profiles of the GRAB-DA2m signals vary not only across regions, but also across sleep/wake transitions within the same region. For example, in CeA, the NREM-to-WAKE transition shows a relatively rapid dopamine decrease, whereas the REM-to-WAKE transition displays a slower decline.

   Similarly, faster dopamine changes are observed in some region/transition combinations, such as NAc-L during NREM-to-WAKE and DRN during REM-to-WAKE, whereas slower kinetics are observed in others, such as mPFC during WAKE-to-NREM and VTA during NREM-to-WAKE. Together, these effects reflect both region-specific mechanisms and transition-dependent differences in dopaminergic activity.

   While cross-correlation analysis provides insight into the temporal coordination of DA signals across regions, several limitations should be considered. Sleep/wake transitions are inherently non-stationary events, whereas cross-correlation assumes relatively stable signal properties within the analysis window. This mismatch may bias lag estimates and obscure transient lead-lag relationships. Moreover, the temporal resolution of fiber photometry and the kinetics of genetically encoded DA sensors limit the precision with which timing relationships can be interpreted, particularly for sub-second lags.

   We thank the reviewer for raising these important considerations. The temporal relationships between regional dopamine signals were assessed using cross-covariance analysis. We agree that cross-covariance analysis has limitations when applied to sleep/wake transitions, because these transitions are inherently non-stationary events. Although cross-covariance centers the signals by subtracting their means and is therefore less sensitive to baseline offsets than raw cross-correlation, it still summarizes the lagdependent covariance between two signals over the selected analysis window. Therefore, the inferred lag should be interpreted as a transition-level measure of temporal coordination rather than a precise estimate of instantaneous lead–lag timing.

   To minimize the influence of brief or unstable state fluctuations, we only included transitions in which both the preceding and following sleep/wake epochs lasted at least 30 s, and excluded epochs shorter than 30 s [4]. This criterion helped ensure that the analyzed events represented well-defined transitions between sustained behavioral states rather than transient or fragmented episodes. Although dopamine signals may still change dynamically within the transition window, and the temporal resolution of fiber photometry and the kinetics of genetically encoded GRAB-DA2m sensors limit the precision with which fine-scale timing relationships can be interpreted, dopamine signals were relatively stable within each behavioral state, as shown in Fig. 1B and reported previously [1,9,10] Thus, we believe that cross-covariance analysis provides useful information about the temporal coordination of dopamine dynamics across regions.

   In the Introduction, the authors state that they aim to address 'which dopaminergic populations causally drive these patterns.' However, the chemogenetic approach used operates on a relatively slow timescale: CNO-induced activation takes 15-30 minutes to produce effects, and the induced changes are long-lasting. In contrast, the dopamine transitions described in Figure 1 occur on a much faster timescale compared to CNO manipulation. Thus, while chemogenetic activation demonstrates that stimulating VTA or DRN dopamine neurons promotes wakefulness, it does not directly establish that these populations causally drive the rapid transition-related DA dynamics observed in the photometry recordings.

   We thank the reviewer for this thoughtful comment. We agree that chemogenetic manipulation operates on a much slower timescale than the rapid dopamine transients observed during sleep–wake transitions, and therefore does not directly recapitulate these fast dynamics. In particular, CNO-induced activation unfolds over minutes and produces sustained changes in neuronal activity, whereas the DA signals we report fluctuate on a sub-second to second timescale. Our intention with the chemogenetic experiments was not to mimic the precise temporal profile of endogenous DA signals, but rather to test whether increasing the activity of specific dopaminergic populations is sufficient to influence behavioral state.

   In this context, our results show that activation of VTA or DRN dopaminergic neurons robustly promotes wakefulness, supporting a causal role for these populations in sleep– wake regulation at the circuit level. However, we agree that these data do not by themselves establish that these neurons directly generate the rapid transition-related DA dynamics observed in the photometry recordings.

   Reviewer #2 (Public review):

   In "Brainwide dopamine dynamics across sleep-wake transitions", Chen et al. provide a thorough description of how dopamine dynamics fluctuate across sleep-wake transitions and in transitions between sleep states. To achieve this, the authors used multi-channel fiber photometry and a genetically encoded fluorescent dopamine reporter to simultaneously measure dopamine dynamics in 8 brain regions. They also used EEG measurements to precisely quantify and time transitions between sleep states and wakefulness. Finally, the authors used channelrhodopsin to examine dopamine dynamics following subregion stimulation and chemogenetics to test the causal relationship between activation of distinct dopamine neuron populations and their effects on sleep state.

   The conclusions made by the authors in this study are modest and appropriate given the largely observational nature of the principal findings. The use of optogenetics to probe regional dopamine signaling following activation of distinct nuclei is interesting, but not entirely novel and constrained in interpretability. Similarly, the chemogenetics experiment largely confirms previous studies, which the authors correctly cited in the text.

   The principal findings of this study are based on strong methodological and analytical methods. Implanting 8 optical fibers in a single mouse, along with EEG/EMG electrodes, is technically challenging, providing valuable, simultaneous measurements of dopamine fluctuations across the brain. This enables the strong correlational and time-locked analyses performed by the authors in Figure 2. What's more, the use of EEG/EMG electrodes provides time-locked descriptions of sleep states, enabling precise comparisons between the dopamine signal and sleep state transitions.

   The paper has some weaknesses that the authors could address. The analyses in Figure 1 could be strengthened to show how dopamine changes during transitions between specific sleep states. The injection sites for channelrhodopsin and chemogenetic viruses could be validated to strengthen the interpretation of those results. Also, a stronger justification for the experiments conducted in Figure 3 could be provided, as they seem unrelated to the present study.

   Overall, this study has strong descriptive power, convincingly showing how dopamine fluctuates across sleep states. Some of the other aspects of the paper, however, are somewhat limited in novelty and interpretation.

   The analyses in Figure 1 could be strengthened to show how dopamine changes during transitions between specific sleep states.

   We appreciate the reviewer’s thoughtful suggestion. We agree that the directionality and kinetics of dopamine changes during sleep/wake transitions may provide important information beyond state-level dopamine quantification.

   In this study, mice were recorded for 4–5 h during each sleep session. Across the recording period, mice frequently transitioned from NREM to WAKE, WAKE to NREM, NREM to REM, and REM to WAKE. Transitions from WAKE to REM were rarely observed and therefore were not included in the transition analysis. Accordingly, we focused our analysis on the four major transition types: NREM-to-WAKE, WAKE-to-NREM, NREM-toREM, and REM-to-WAKE [4,9,11].

   For each transition type, dopamine dynamics were analyzed separately by aligning the zscored GRAB-DA2m signal to the transition onset and averaging across all epochs of the same transition type. To minimize the influence of brief or unstable state fluctuations, we excluded transitions in which either the preceding or following sleep/wake epoch lasted less than 30 s. The resulting transition-triggered dopamine traces were then averaged across sessions and mice for each transition type independently.

   Thus, the transition analysis preserves the directionality of state changes rather than pooling all sleep/wake transitions together. Because dopamine signals differ across behavioral states, transitions between neighboring states produce distinct temporal profiles when aligned to the transition point [4,9-11]. For example, REM-to-WAKE transitions may show a rapid increase in dopamine in the mPFC, whereas WAKE-to-NREM or NREM-to-REM transitions may show slower and more modest decreases. These transition - specific kinetics may reflect distinct underlying mechanisms, including changes in dopamine neuron firing or local terminal modulation.

   The injection sites for channelrhodopsin and chemogenetic viruses could be validated to strengthen the interpretation of those results.

   We agree with the reviewer that precise histological validation is essential for the correct interpretation of our optogenetic and chemogenetic findings.

   Regarding the chemogenetic experiments, as noted, we provide examples of virus expression in the VTA, DRN, and SNc in Figure 4. By demonstrating the consistency and restriction of our targeting across the entire cohort (VTA, SNc, and DRN), we confirmed that our observed sleep effects were regionally specific. Our data only included mice with accurate targeting and no substantial virus "leakage" into adjacent nuclei.

   We thank the reviewer for this insightful observation regarding the regional dopamine (DA) responses following SNc stimulation. While the SNc is traditionally associated with the dorsal striatum (DLS), several studies have demonstrated that SNc dopaminergic neurons also project to the nucleus accumbens, particularly the lateral shell [12,13]. Furthermore, recent work characterizing the functional heterogeneity of midbrain DA neurons suggests that SNc subpopulations can drive significant DA release in ventral striatal subregions [14]. We appreciate the reviewer’s caution regarding potential off-target effects. While our histological criteria for validation post recordings were stringent, we acknowledge that in any midbrain manipulation, the close anatomical proximity of the VTA and SNc makes it technically challenging to guarantee zero involvement of neighboring VTA neurons. However, by using mice with the most restricted virus expression and fibers targeting, we have minimized this potential confound as much as is technically feasible with current viral and optogenetic methods.

   Also, a stronger justification for the experiments conducted in Figure 3 could be provided, as they seem unrelated to the present study.

   We thank the reviewer for this comment. The experiments in Figure 3 were designed to systematically map the sources of dopaminergic inputs to key brain regions examined in this study [15], including the mPFC, DLS, NAc, and CeA. Establishing these input–output relationships is important for interpreting the photometry signals observed during sleep– wake transitions.

   Specifically, we found that optogenetic activation of VTA dopaminergic neurons elicits DA responses in all four regions, whereas activation of DRN dopaminergic neurons induces responses in the mPFC, DLS, and CeA, and activation of SNc dopaminergic neurons induces responses in the mPFC, NAc, and DLS. These results reveal partially overlapping but distinct projection patterns across dopaminergic populations.

   Taken together, these data provide a circuit-level framework suggesting that VTA, SNc, and DRN dopaminergic neurons may contribute differentially and with distinct weights to the DA signals observed in these regions during sleep wake transitions.

   Overall, this study has strong descriptive power, convincingly showing how dopamine fluctuates across sleep states. Some of the other aspects of the paper, however, are somewhat limited in novelty and interpretation.

   We appreciate the reviewer’s assessment that our study convincingly demonstrates how dopamine fluctuates across sleep states. We agree that the primary contribution of this work is descriptive and foundational. At the same time, we respectfully emphasize that rigorous, comprehensive descriptive studies are essential, particularly when addressing phenomena that have not been systematically characterized. Prior to this work, dopamine dynamics during natural sleep–wake transitions had not been measured simultaneously across multiple brain regions.

   Our multi-site photometry approach advances the field in several important ways. Technically, the combination of simultaneous eight-region fiber photometry with EEG/EMG recordings represents a substantial methodological advance, enabling brainwide, network-level analysis of dopamine dynamics during natural state transitions. This approach reveals emergent features—such as temporal coordination and inter-regional lead–lag relationships—that cannot be captured using single-site recordings. Moreover, integrating brain-wide measurements with region-specific manipulations allows circuitlevel insights that would not be accessible from either approach alone.

   Conceptually, our findings revealed the region, sleep/wake transition type -specific and bidirectional dopamine dynamics, instead of the prevailing view of dopamine as a uniform arousal signal: dopamine decreases in certain limbic regions, such as the central amygdala and nucleus accumbens lateral shell, during arousal transitions, while increasing in cortical and other striatal regions. These results refine simplified models of dopaminergic regulation of arousal. In addition, our data reveal differential circuit contributions, with the VTA and DRN—but not the SNc—promoting wakefulness, highlighting functional specialization within the dopamine system.

   We acknowledge that some aspects of our study, including the optogenetic mapping and chemogenetic experiments, build on established methodologies and in part confirm prior findings. However, these experiments also provide several new insights. First, whereas individual dopamine sources have often been studied in isolation, our systematic comparison across VTA, SNc, and DRN using consistent methods reveals distinct brainwide functional contributions that were not previously established. Second, our optogenetic mapping does not simply recapitulate known projection patterns, but instead uncovers quantitative differences in dopamine release kinetics and magnitude across source–target pairs, which inform the heterogeneity of the transition dynamics. Finally, our findings provide a crucial anatomical and temporal framework for future research on the specific mechanisms driving these dynamics and their precise functional consequences.

   References:

   (1) Sun, F. et al. Next-generation GRAB sensors for monitoring dopaminergic activity in vivo. Nat Methods 17, 1156-1166, doi:10.1038/s41592-020-00981-9 (2020).

   (2) Ihalainen, J. A., Riekkinen, P., Jr. & Feenstra, M. G. Comparison of dopamine and noradrenaline release in mouse prefrontal cortex, striatum and hippocampus using microdialysis. Neurosci Lett 277, 71-74, doi:10.1016/s0304-3940(99)00840-x (1999).

   (3) Berridge, C. W. & Abercrombie, E. D. Relationship between locus coeruleus discharge rates and rates of norepinephrine release within neocortex as assessed by in vivo microdialysis. Neuroscience 93, 1263-1270, doi:10.1016/s0306-4522(99)00276-6 (1999).

   (4) Silverman, D. et al. Activation of locus coeruleus noradrenergic neurons rapidly drives homeostatic sleep pressure. Sci Adv 11, eadq0651, doi:10.1126/sciadv.adq0651 (2025).

   (5) Anaclet, C. et al. The GABAergic parafacial zone is a medullary slow wave sleeppromoting center (vol 17, pg 1217, 2014). Nat Neurosci 17, 1841-1841, doi:DOI 10.1038/nn1214-1841d (2014).

   (6) Ma, C. Y. et al. Microglia regulate sleep through calcium-dependent modulation of norepinephrine transmission. Nat Neurosci 27, 249-258, doi:10.1038/s41593-02301548-5 (2024).

   (7) Traut, J. et al. Effects of clozapine-N-oxide and compound 21 on sleep in laboratory mice. Elife 12, doi:10.7554/eLife.84740 (2023).

   (8) Grace, A. A., Floresco, S. B., Goto, Y. & Lodge, D. J. Regulation of firing of dopaminergic neurons and control of goal-directed behaviors. Trends Neurosci 30, 220-227, doi:10.1016/j.tins.2007.03.003 (2007).

   (9) Darmohray, D. et al. Brainstem circuit for sickness-induced sleep. Sci Adv 11, doi:ARTN eady024510.1126/sciadv.ady0245 (2025).

   (10) Hasegawa, E. et al. Rapid eye movement sleep is initiated by basolateral amygdala dopamine signaling in mice. Science 375, 994-+, doi:10.1126/science.abl6618 (2022).

   (11) Ding, X. et al. Neuroendocrine circuit for sleep-dependent growth hormone release. Cell 188, 4968-4979 e4912, doi:10.1016/j.cell.2025.05.039 (2025).

   (12) Poulin, J. F. et al. Mapping projections of molecularly defined dopamine neuron subtypes using intersectional genetic approaches. Nat Neurosci 21, 1260-1271, doi:10.1038/s41593-018-0203-4 (2018).

   (13) Lerner, T. N. et al. Intact-Brain Analyses Reveal Distinct Information Carried by SNc Dopamine Subcircuits. Cell 162, 635-647, doi:10.1016/j.cell.2015.07.014 (2015).

   (14) Azcorra, M. et al. Unique functional responses differentially map onto genetic subtypes of dopamine neurons. Nat Neurosci 26, 1762-1774, doi:10.1038/s41593023-01401-9 (2023).

   (15) Eban-Rothschild, A., Rothschild, G., Giardino, W. J., Jones, J. R. & de Lecea, L. VTA dopaminergic neurons regulate ethologically relevant sleep-wake behaviors. Nat Neurosci 19, 1356-1366, doi:10.1038/nn.4377 (2016).

   Read the original source
5. 

   Version published to 10.64898/2026.02.13.705774 on bioRxiv

   Feb 16, 2026