Nucleus accumbens controls wakefulness by a subpopulation of neurons expressing dopamine D1 receptors

Nucleus accumbens (NAc) is involved in behaviors that depend on heightened wakefulness, but its impact on arousal remains unclear. Here, we demonstrate that NAc dopamine D1 receptor (D1R)-expressing neurons are essential for behavioral arousal. Using in vivo fiber photometry in mice, we find arousal-dependent increases in population activity of NAc D1R neurons. Optogenetic activation of NAc D1R neurons induces immediate transitions from non-rapid eye movement sleep to wakefulness, and chemogenetic stimulation prolongs arousal, with decreased food intake. Patch-clamp, tracing, immunohistochemistry, and electron microscopy reveal that NAc D1R neurons project to the midbrain and lateral hypothalamus, and might disinhibit midbrain dopamine neurons and lateral hypothalamus orexin neurons. Photoactivation of terminals in the midbrain and lateral hypothalamus is sufficient to induce wakefulness. Silencing of NAc D1R neurons suppresses arousal, with increased nest-building behaviors. Collectively, our data indicate that NAc D1R neuron circuits are essential for the induction and maintenance of wakefulness.

This is a solid manuscript by Luo, Li, Wang et al. Utilizing state of the art tools the authors show that acute activation of nucleus accumbens NAc D1Rs promote transitions to wakefulness (they initiate it), while chronic opto activation of D1R neurons promotes prolonged wakefulness. Circuit mapping shows that these NAc D1R MSNs preferentially target non-DA neurons (probably GABA) in the VTA/SN, however they find no direct input from D1R MSNs to SN DA neurons, very sparse input to VTA DA neurons. Overall this is a well executed series of studies. The overall novelty of their work was diminished somewhat by a similar paper recently published by the Delecea group, however, it is still an important addition to literature. My suggestions and concerns are detailed below. Thoughts: -Firstly, as mentioned above, the subject is largely lacking in novelty aside from the D1R output portion of the results (they have essentially just focused on the post-synaptic aspect of the previous DeLecea study). However, the cell-type specific circuit mapping of D1R NAc MSNs-VTA neurons is valuable in elucidating the connectivity mediating the behavioral effect observed.
-The focus on D1R MSNs in the NAc should really be supplemented by complementary studies in D2R MSNs. It doesn't seem appropriate to just refer to other studies for this. -Overall, the study would be strengthened by focusing on the NAc-VTA/SN component of the story and leaving out the LH. The LH part seems like it was tossed in and thus seems incomplete. o Not to mention that the LH data is not in agreement with previous studies→ another reason why they need to either perform detailed circuit analysis or just leave it out.
Minor notes: o Fig 1: • For FP experiments, they used 3 mice but instead of averaging the responses within mice and reporting that, they report all data points together for each condition→ this doesn't seem like how this should be done o On the top of page 8, it is unclear what recordings were done here. It should be stated more clearly in the text even if it is also in the figure. o Why did they use male and female mice for the ephys if no female mice were used behaviorally? It's not the biggest deal but it definitely isn't ideal. o The laser power they reported was 20-30 mW out of the fiber tip for opto. That is very high! I would find it unlikely that there was no damage at that power lever. o What kind of AAVs were used (serotypes)? Which promoters were used? Were they FLEX or DIO? There is not a lot of info about the viruses.
Reviewer #2 (Remarks to the Author): In their manuscript, Luo et al. used up to date methodologies such as optogenetic, chemogenetic and calcium imaging to elegantly show that nucleus accumbens GABAergic neurons expressing dopamine D1 receptors play a crucial role in wakefulness. Overall, the demonstration is really impressive and convincing. It is also nicely described and illustrated. Below, I made some remarks and requests on the work.
Line 103-106, the authors wrote that the NAc D1R neurons start to increase their firing before wake and REM sleep at the transitions with NREM. Indeed, it is visible in Fig. 1f. It would be of interest to give a mean time value here to figure out how much the neurons anticipate states changes.
The authors state line 115 that intense Fos expression was induced in hM3Dq expressing neurons. The illustration Fig. 2C showing such expression is too small to give an idea of the number and distribution of Fos+ neurons. It would be nice to enlarge the photo or show an enlarge photo as a supplementary figure. A drawing showing the distribution of Fos with a quantification would be perfect. Again line 163 for optogenetic, same comment on the Fos staining with a very small photography. Line 242: are instead of were 262-263: I don't get how the authors conclude that inputs from NAc D1R neurons to nigrostriatal and mesolimbic neurons are different. Indeed, NAc D1R neurons do not project to nigrostriatal neurons. Further, the projection of NAc on VTA neurons is not destined to DA neurons. Line 265-284: stimulation of the ChR2 fibers in the LH and the midbrain induced waking similarly. For the LH, the author further found that only 37% of the LH neurons were inhibited on slices. I have two question there, is it known whether stimulating passing fibers expressing ChR2 give rise to no effect? A Fos staining might help to answer such question. My second question would be on the types of inhibited neurons in the LH. Indeed, it should be GABAergic neurons inhibiting Wake-active neurons. The authors did look at the types of neurons responding? Either in term of electrophysiological or neuroanatomical characteristics? Line 318: The authors cite a publication in press showing that NAc D2R-expressing neurons strongly promote sleep. They should develop the results obtained in this study. This is indeed important in order to get the whole picture of the role of the NAc neurons in sleep. 330-331: rephrasing of the sentence is necessary 334: on the firing of DA VTA neurons, the article of Dahan et al. (2007) is missing. 340-342: repetitions again in the sentence Line 343-344: the information on the neuronal target of adenosine in the Nac as published by : Discussion on the LH is focusing on GABAergic neurons involved in waking. However, this is relatively a new concept whereas there are many publications showing the presence of GABAergic neurons involved in sleep including a very recent one in Nature about ZI neurons. These publications need to be discussed. Missing are also publications demonstrating that a projection from the NAc to LH neurons exists. This is important since many of the fibers there are just passing in the MFB. Line 362 They state that the NAc-VP pathway is also important for sleep without developing the arguments. Again, this is important to figure out which projection does what. Overall, the discussion needs to provide a clearer picture on the role of the different NAc neurons and projections. They should also discuss the role of interconnections between the two populations as reported in their supplementary Fig. 5. A drawing would be particularly helpful in that matter.
Reviewer #3 (Remarks to the Author): The study is intensive and extensive using various cutting edge techniques, and the results appeared to be reliable. However, it was difficult to understand the functional importance of the work.
I understand from the data presented that NAc D1R neuron circuits are essential for the induction and maintenance of wakefulness, but the manuscript does not address anything about how NAc D1R neuron circuits are physiologically and pathophysiology important. The abstract is merely a list of the findings and does not address functional importance of the NAc D1R neuron circuits. Most of the experiments, except in vivo fiber photometry and electron microscopy experiments, are likely to be supraphysiological and do not provide much information regarding how these NAc D1R neuron circuits coordinate with other systems to regulate sleep and wakefulness.
Both D1R and D2R neuron exist in the NAc, and D1R and D2R-expressing neurons play complementary and sometimes opposing roles in higher brain functions. A recent research reported that direct stimulation of the NAc A2AR/D2R-expressing neurons induced remarkable non-rapid eye movement (non-REM, NREM) sleep.
These taken together with the fact that the endogenous molecule mediating the NAc D2R and D1R mediating effects is DA, it is important to study how these two DA receptive systems are regulated to control physiological sleep/wake and study how disruptions of theses systems affect sleep and other behavioral states.
The authors concluded that NAc D1R-expressing neurons are essential in controlling wakefulness and are involved in physiological arousal via the LH and midbrain circuits, suggesting that the NAc should be considered as a potential target area for therapy in neuropsychiatric disorders with sleep-wake alterations.
This conclusion does not hold much in terms of functional significance and no data presented suggested that the NAc should be considered as a potential target area for therapy in neuropsychiatric disorders with sleep-wake alterations.

Reviewer #1
This is a solid manuscript by Luo, Li, Wang et al. Utilizing state of the art tools the authors show that acute activation of nucleus accumbens NAc D1Rs promote transitions to wakefulness (they initiate it), while chronic opto activation of D1R neurons promotes prolonged wakefulness. Circuit mapping shows that these NAc D1R MSNs preferentially target non-DA neurons (probably GABA) in the VTA/SN, however they find no direct input from D1R MSNs to SN DA neurons, very sparse input to VTA DA neurons. Overall this is a well executed series of studies. The overall novelty of their work was diminished somewhat by a similar paper recently published by the Delecea group, however, it is still an important addition to literature.
My suggestions and concerns are detailed below.
Thoughts: -Firstly, as mentioned above, the subject is largely lacking in novelty aside from the D1R output portion of the results (they have essentially just focused on the post-synaptic aspect of the previous DeLecea study). However, the cell-type specific circuit mapping of D1R NAc MSNs-VTA neurons is valuable in elucidating the connectivity mediating the behavioral effect observed.

Response:
We thank the comments. As the reviewer mentioned, De Lecea found that the NAc is an important downstream nucleus for the VTA in mediating wakefulness.
In our study, we provided a series of direct evidence showing that NAc D 1 R-neurons control wakefulness. With the newly obtained data, we also revealed that two distinct populations of D 1 R and D 2 R neurons in the NAc play the opposite role in sleep-wake control. Besides, neuronal circuits mapping data showed that two important downstream targets (midbrain and LH) for NAc D 1 R neurons in arousal control.
-The focus on D1R MSNs in the NAc should really be supplemented by complementary studies in D2R MSNs. It doesn't seem appropriate to just refer to other studies for this.

Response:
As requested, we carried out a complementary study in D 2 R neurons by using D 2 R-Cre mice and chemogenetic methods to activate or inhibit NAc  -Overall, the study would be strengthened by focusing on the NAc-VTA/SN component of the story and leaving out the LH. The LH part seems like it was tossed in and thus seems incomplete.
o Not to mention that the LH data is not in agreement with previous studies→ another reason why they need to either perform detailed circuit analysis or just leave it out.

Response:
To better understand the NAc D1R -LH circuit in sleep-wake regulation, we added new data to address these questions by in vitro electrophysiological experiment and c-Fos staining, please see supplementary figure 5 (page 51, lines 1136-1150), and texts in pages 10-11, lines 285-301.
To test whether NAc D 1 R neurons directly inhibit wake-promoting LH GABA neurons (Herrera et al., 2015;Venner et al., 2016), we crossed D 1 R-Cre and GAD67-GFP mice to generate a new mouse D 1 R-Cre::GAD67-GFP. o Why did they use male and female mice for the ephys if no female mice were used behaviorally? It's not the biggest deal but it definitely isn't ideal.

Response:
We used male mice for behavioral tests to avoid the influence of estrous cycles that might cause data variation. In vitro electrophysiological study was employed to explore the connections between NAc neurons infected with ChR2 and downstream neurons. We did not find significant differences in path-clamp data of different genders, so that we used both genders in patch-clamp study.
o The laser power they reported was 20-30 mW out of the fiber tip for opto. That is very high! I would find it unlikely that there was no damage at that power lever.

Response:
Thanks for the comments. In our in vivo optogenetic study, we reported that light density was 20-30 mW/mm 2 , not 20-30 mW.
Actually, the laser power from the tip of fiber (200-µm-diameter) was lower than 3 mW. The optic fiber illuminated area was estimated as a small spot (~380 μm diameter; 0.11 mm 2 area), and this area was used to calculate the light density. Thus, the light density was estimated to be lower than 25.5 mW/mm 2 , and we reported that light density was 20-30 mW/mm 2 . This parameter was used based on previous publications Yuan et al., 2017).
o What kind of AAVs were used (serotypes)? Which promoters were used? Were they FLEX or DIO? There is not a lot of info about the viruses.
As requested, we added the information in the main text (   Line 265-284: stimulation of the ChR2 fibers in the LH and the midbrain induced waking similarly. For the LH, the author further found that only 37% of the LH neurons were inhibited on slices. I have two question there, is it known whether stimulating passing fibers expressing ChR2 give rise to no effect? A Fos staining might help to answer such question.

Response:
We thank the important comments. In brain slices, we found that projection  Fig. 5a).

We found that NAc D 1 R neurons innervated both GFP-positive and
GFP-negative neurons in the LH, showing no distinct preference (67% vs. 47%; Supplementary Fig. 5b, d-g). Meanwhile, light-evoked IPSCs latency was less than 5 ms ( Supplementary Fig. 5c), indicating the direct connection. Because

GAD67 is a marker for GABA, our results suggested that NAc D 1 R neurons would target both GABA neurons expressing GAD67 and non-GAD67 neurons
including GAD65 GABAergic neurons and non-GABAergic neurons. Fig. 5h-i). Furthermore, photostimulation of LH terminals from NAc D 1 R neurons induced wakefulness.

Activation of NAc D 1 R neurons by CNO induced robust c-Fos expression in the LH orexin positive neurons (Supplementary
Taken together, these data indicated that NAc D1R →LH circuit was involved in arousal control partially through disinhibiting LH wake-promoting orexin neurons.
Line 318: The authors cite a publication in press showing that NAc D2R-expressing neurons strongly promote sleep. They should develop the results obtained in this study. This is indeed important in order to get the whole picture of the role of the NAc neurons in sleep.

Response:
As requested, we carried out a complementary study in D 2 R neurons by using

Response:
As suggested, we added this paper in the revised manuscript.

Response:
We modified the sentence to "Our data support

Response:
As suggested, we added this paper in the revised manuscript. sleep. They also conducted a retrograde tracing study and confirmed a monosynaptic projection from the NAc to ZI Lhx6+ neurons. We examined IPSCs in ZI neurons following optogenetic stimulation of NAc D 1 R neurons terminals in brain slices.
Indeed, we observed sparse ChR2 positive fibers around lhx6+ neurons in the ZI (the image below), but we did not detect functional connection (n = 7 cells from 2 mice, data not shown), suggesting that NAc D1R →ZI pathway might not be crucial for arousal regulation of NAc D 1 R neurons.

Several publications demonstrate that anatomical and functional
projections from the NAc to LH neurons exist. (Heimer et al., 1991;Zahm and Heimer, 1993).

1) Anatomical observations using anterograde and retrograde tracing methods have shown that NAc projection neurons send axons to the LH
2) Pharmaco-behavioral findings support the proposed NAc to LH circuit: increased feeding following NAcSh inhibition is prevented by concomitant infusion of a GABA A R agonist into the LH (Maldonado-Irizarry et al., 1995;Urstadt et al., 2013).

3) In vivo single-unit recording revealed that photostimulation of NAcSh
fibers modulated neuronal activity in the LH (Prado et al., 2016).

4)
Using optogenetic assisted mapping approach, scientists found that NAc projection neurons inhibited LH neurons (Larson et al., 2015;. Together, these data suggest that NAc neurons really send projections to the LH neurons. We cited related publications in the text of revised manuscript (Line 271).
Line 362 They state that the NAc-VP pathway is also important for sleep without developing the arguments. Again, this is important to figure out which projection does what.

Response:
We thank the important comments. Although both NAc D 1 R and D 2 R-expressing neurons innervate the VP, quantitative analysis revealed that less than 3% colocalization of D 1 -and D 2 -expressing fibers in the VP, strongly indicating separate D 1 -and D 2 -neurons projections to VP . In our study, we think that activation of NAc D1R →VP pathway promotes arousal, whereas activation of NAc A2AR →VP pathway induced sleep that has been confirmed by Oishi . Overall, the discussion needs to provide a clearer picture on the role of the different NAc neurons and projections. They should also discuss the role of interconnections between the two populations as reported in their supplementary Fig. 5. A drawing would be particularly helpful in that matter.

Response:
As suggested, we added a drawing to clarify our hypothesis (Please see Reviewer #3 (Remarks to the Author): The study is intensive and extensive using various cutting edge techniques, and the results appeared to be reliable. However, it was difficult to understand the functional importance of the work.  , because adenosine A 2A Rs are co-expressed with dopamine D 2 Rs in the same neurons in the NAc (Lazarus et al., 2011).
In addition to the DA afferents, NAc neurons receive glutamatergic inputs from cortex and subcortex regions, as well as inhibitory inputs from local interneurons and collateral inputs from other medium spiny neurons (Floresco, 2015;. Thus, the endogenous molecule mediating the effects of NAc D 2 R and D 1 R neurons is not only just DA, but also glutamate and GABA.
Besides, adenosine is an obvious candidate for activating NAc D 2 R positive neurons because excitatory adenosine A 2A R were colocalized with D 2 Rs.
Therefore, the NAc has the capability to integrate locomotion with motivational behavior through dopaminergic inputs, contextual content from the hippocampus, and emotional information from the amygdala.
In our study, we found that NAc D 1 R-and D 2 R neurons play the opposite role in sleep-wake regulation, while other distinct roles in reward, sensitization and feeding behaviors are also reported by previous studies (Bock et al., 2013;Lobo et al., 2010;Smith et al., 2013). A balance between these two subpopulations in the NAc is likely necessary for sleep-wake behavior and other behavioral states.
The authors concluded that NAc D1R-expressing neurons are essential in controlling wakefulness and are involved in physiological arousal via the LH and midbrain circuits, suggesting that the NAc should be considered as a potential target area for therapy in neuropsychiatric disorders with sleep-wake alterations.
This conclusion does not hold much in terms of functional significance and no data presented suggested that the NAc should be considered as a potential target area for therapy in neuropsychiatric disorders with sleep-wake alterations.

Response:
Many studies have shown that NAc dysfunction causes numerous neurological disorders including depression, anxiety disorders, addictive behaviors and compulsive disorders (Lobo et al., 2012;Lobo and Nestler, 2011;Salgado and Kaplitt, 2015;Vetrivelan et al., 2010;Zarrindast and Khakpai, 2015). These neurological disorders have been reported to associate with sleep-wake alterations (Abbott and Videnovic, 2016;O'Sullivan et al., 2008;Shulman et al., 2001). Our study demonstrated that the NAc is important in regulating sleep and wakefulness, and inhibitory NAc projections to the VTA/SNc and the LH, which are shown to be involved in neuropsychiatric behaviors, such as motivation, reward and compulsive feeding (Bocklisch et al., 2013;Larson et al., 2015;.
Our results uncover the distinct roles of NAc D 1 R and D 2 R neurons in sleep-wake regulation, and NAc D 1 R neurons modulate physiological and functional wakefulness mainly through the VTA/SNc and the LH pathways.
Considering the importance of the NAc-VTA/SNc and -LH pathways in physiological and pathophysiological functions, we thought that the NAc may be considered as a potential target area for therapy in neuropsychiatric disorders with sleep-wake alterations.

The specific comments of reviewers and editors are addressed with the point-by-point responses as follows:
Reviewer #1 (Remarks to the Author): The manuscript has been adequately revised. I have nothing further.
Reviewer #2 (Remarks to the Author): The authors well answered to my requests.
Response: Thank you very much for the comments.
I have only one comment concerning the drawing of the pathways involved.
I believe this is a very nice addition to the paper.
I believe that a few information should be added to it.
-It should be added that adenosine excites the D2-expressing neurons in the NAc.  . NAc D 2 R/A 2A R to VP pathway has been considered to be involved in sleep control , whereas NAc D 2 R/A 2A R to LH and VTA pathways are not important for sleep induction of NAc D 2 R/A 2A R neurons . However, the role of NAc D 1 R to VP pathway in sleep-wake regulation is needed to uncover. Activation of NAc D Gs/olf proteins and stimulate the activity of adenylate cyclase, followed by PKA activation via cAMP accumulation. By contrast, D 2 Rs and A 1 Rs are associated with Gi/o proteins to inhibit cAMP production Nagai et al., 2016). In vitro and in vivo electrophysiological studies indicate that D 1 R activation modulates the intrinsic excitability of NAc neurons, and usually increases neuron activities, whereas D 2 R activation often attenuates spike activities (Hopf et al., 2003;Perez et al., 2006;West and Grace, 2002 Conversely, how come DA inhibit the D1 wake inducing neurons?

Response:
The D 1 R is coupled to adenylate cyclase through G s/olf and activates protein kinase A (PKA), whereas the D 2 R inhibits adenylate cyclase through G i/o (Herve et al., 1993;Nagai et al., 2016). In vitro and in vivo electrophysiological studies indicate that D 1 R activation modulates the intrinsic excitability of NAc neurons, and usually increases neuron activities, whereas D 2 R activation often attenuates spike activities (Hopf et al., 2003;Perez et al., 2006;West and Grace, 2002). Thus, DA would differentially modulate neuronal activity of NAc two subpopulations through its action on excitatory D 1 Rs or inhibitory D 2 Rs, both of which are involved in wakefulness.
We added above discussion into the legend of supplementary figure 8.
-You also show only one pathway from the D2 NAc neurons to the VP, there are no additional projections?
It would be nice to also discuss these points in the text or the legend of the figure.
Response: Anatomical studies show that NAc D 2 R/A 2A R neurons mainly project to the VP, and send sparse fibers to the LH and VTA (Francis et al., 2015;Zhang et al., 2013). However, optogenetic-assisted circuit mapping studies find that light-evoked IPSCs in VP cells, but not in VTA neurons . A weak functional connection between NAc D 2 R/A 2A R neurons and LH cells is detected in vitro . In vivo optical stimulation of NAc A 2A R neurons terminals in the VP, but not the LH and the VTA, produces NREM sleep. Thus, only NAc D 2 R/A 2A R to VP pathway is important in sleep induction.
As requested, we added these major points to the legend of supplementary figure   8.