Sleep-promoting neurons remodel their response properties to calibrate sleep drive with environmental demands.

Falling asleep at the wrong time can place an individual at risk of immediate physical harm. However, not sleeping degrades cognition and adaptive behavior. To understand how animals match sleep need with environmental demands, we used live-brain imaging to examine the physiological response properties of the dorsal fan-shaped body (dFB) following interventions that modify sleep (sleep deprivation, starvation, time-restricted feeding, memory consolidation) in Drosophila. We report that dFB neurons change their physiological response-properties to dopamine (DA) and allatostatin-A (AstA) in response to different types of waking. That is, dFB neurons are not simply passive components of a hard-wired circuit. Rather, the dFB neurons intrinsically regulate their response to the activity from upstream circuits. Finally, we show that the dFB appears to contain a memory trace of prior exposure to metabolic challenges induced by starvation or time-restricted feeding. Together, these data highlight that the sleep homeostat is plastic and suggests an underlying mechanism.

Introduction

The importance of sleep is highlighted by the observation that it is evolutionarily conserved despite directly competing with all motivated waking behaviors [1,2]. Not only does sleep compete with foraging, eating, and mating [3-6], for example, high sleep drive may be maladaptive in many circumstances since falling asleep could place the individual in danger of immediate physical harm [7,8]. On the other hand, sleep plays a critical role in learning and memory, supports adaptive behavior, and facilitates creative insight [9-12]. Together, these observations suggest that it will not be possible to fully understand sleep’s function without knowing how sleep circuits calibrate sleep drive with motivational states.

In recent years, a great deal of progress has been made dissecting circuits that regulate motivated behavior in flies [13-17]. Typically, the properties of a circuit are examined as a function of one internal state. For example, sleep circuits are evaluated after sleep loss [18-20], feeding in response to starvation or high dietary sugar [21,22], thirst following water deprivation [23], mating after social isolation [24], etc. However, since internal states can promote conflicting goal-directed behaviors, recent studies have begun to evaluate how circuits regulate competing activities such as feeding and sleep [5,25-29], thirst and hunger [23], sweet and bitter taste [30], hunger and mating [31], and thirst versus hunger relevant memory [23], for example.

A common theme that has emerged from these studies has been that a specific deprivation-state differentially activates a subset of peptidergic neurons that then modulate classic neurotransmitter systems, frequently dopamine (DA), to alter downstream circuits and thus motivated behavior [31-33]. For example, water deprivation preferentially activates a subset of peptidergic neurons that inhibit specific dopaminergic neurons to alter thirst-relevant memory [34]. In this model, the competition between state-specific peptidergic neurons will determine which motivated behavior will be expressed. An open question, however, is whether the neurons receiving peptidergic or dopaminergic input are constrained such that their responses are determined by upstream signals, or on the other hand, whether they can change their own response properties over time to influence a given outcome.

Sleep-promoting R23E10 neurons are well suited to investigate these relationships. R23E10 neurons are modulated both by dopaminergic neurons and allatostatin-A (AstA)-expressing circadian-neurons [19,23,25-27,35,36]. AstA-expressing neurons promote sleep by releasing glutamate onto R23E10 neurons [25]. AstA is functionally similar to galanin and has been implicated in both feeding and sleep [27,37]. Sleep-promoting R23E10 neurons comprise a sleep switch and are an integral part of the sleep homeostat [19,26]. The intrinsic properties of R23E10 neurons have been evaluated in the context of sleep loss where it seems they monitor redox processes as an indicator of energy metabolism [19,38,39]. Thus, R23E10 neurons may serve as command neurons of sorts that can integrate stimuli from competing internal states to gate information flow [40,41]. Both the input and the output connections to R23E10 are well documented [25,26,35,36]. Thus, the focus of this study is to evaluate sleep-promoting R23E10 neurons following challenges that modulate motivational states.

In this study, we identify independent sets of heterogeneous sleep-promoting neurons that change their responses to DA and AstA following challenges that influence sleep. In addition, we identify a wake-promoting effect of AstA on R23E10 neurons suggesting that the co-release of inhibitory AstA with excitatory glutamate may attenuate the overexcitement of R23E10 neurons during high sleep drive and allow animals to maintain wakefulness in dangerous or life-threatening conditions. Finally, we find that both time-restricted feeding and acute starvation enhance subsequent waking by remodeling the expression of DA receptors in sleep-promoting neurons. Together, these data provide new insights into the interaction between internal states and sleep regulation.

Results

Neurons projecting to the dorsal fan-shaped body are modulated by allatostatin

Fan-shaped body (dFB)-projecting R23E10 neurons are an important component of the sleep homeostat [19,38,39]. Sleep-promoting dFB neurons are believed to be inhibited by wake-promoting dopaminergic neurons and activated when glutamate is released from sleep-promoting AstA-expressing neurons [25,35,36]. Surprisingly, the role of AstA on R23E10 neurons has not been investigated. Thus, we used behavioral genetics and live-brain imaging to characterize the effects of AstA and DA on R23E10 neurons. The expression pattern of R23E10 neurons is shown in Fig 1A. Expressing the temperature-sensitive transient receptor potential cation channel (UAS-TrpA1) in R23E10 neurons and raising the temperature from 25°C to 31°C for 6 h increased sleep (Fig 1B and 1C). The parental controls do not show an increase in sleep during the exposure to 31°C (S1A Fig). Since many neuromodulators act through second messenger signaling cascades, we expressed the cyclic adenosine monophosphate (cAMP) sensor, UAS- Epac1-camps, in R23E10 neurons and used high-throughput live-brain imaging to monitor neuronal responses to bath applied DA (Fig 1D) [42,43]. As seen in Fig 1E, DA increases cAMP levels in dFB neurons as previously reported [36]. It is important to note that while R23E10>GFP identifies approximately 14 neurons/hemisphere, R23E10>UAS-Epac1-camps reliably labeled approximately 8 neurons/hemisphere (S1B Fig). Thus, we have confirmed that R23E10 neurons are sleep-promoting and respond to DA.

Fig 1

Sleep-promoting R23E10 neurons respond to AstA.

(A) Confirmation of R23E10-GAL4>UAS-GFP dFB projections (blue arrow). (B, C) R23E10-GAL4>UAS-dTrpA1 showed a significant increase in sleep at 31°C (red) compared to siblings maintained at 25°C (blue; n = 16/group). (D) Schematic of high-throughput Epac imaging. (E, F) Normalized FRET ratio in R23E10-GAL4>UAS-Epac1-camps in response to DA (3 × 10−5 M) and AstA (1e−6 M) (n = 3–7 cells/condition); (G) % change in FRET ratio in R23E10-GAL4>UAS-Epac1-camps following the application of neuroactive compounds (n = 6–21 cells/condition). (H) TTX (1e−7 M) does not prevent the response of R23E10>UAS-Epac1-camps to bath applied AstA (1e−6 M) (n = 3–6 cells/condition). Knocking down AstA-R1 attenuates the response of R23E10 neurons to AstA (red trace; n = 6). (I) Quantification of H (ANOVA F[,] = 15.4, p = 0.000183784; *p < 0.05, modified Bonferroni test). (J) Confocal stack of an AstA-GAL4>UAS-GFP fly brain stained with anti-GFP (J’) Anti-AstA (magenta) antibody. (J”) A merge image. (K) Single optical section (0.5 μm) of a R23E10-LexA>LexAop-GFP, AstA-GAL4>UAS-RFP fly brain stained with anti-GFP (K’) Anti-RFP (magenta). (K”) A merge image. (L) Application of 1 mM ATP induced a decrease in cAMP levels in R23E10 cells when the P2X2 receptor was expressed in AstA-GAL4-expressing neurons (blue trace), but not in control lines lacking the P2X2 receptor (red trace). (M) Quantification of L. Error bars represent standard error of the mean (SEM). Underlying data is in S1 Data and S1 Datasheet. AstA, allatostatin-A; cAMP, cyclic adenosine monophosphate; DA, dopamine; dFB, dorsal fan-shaped body; TTX, tetrodotoxin.

Given that R23E10 neurons are downstream of AstA-expressing neurons [25], we evaluated the response properties of R23E10 neurons to AstA and 14 other compounds that can influence motivational states [44]. As seen in Fig 1F and 1G, AstA reduced cAMP levels in R23E10 neurons. Previous studies indicate that AstA is an inhibitory peptide [45]. Individual traces for Fig 1G are plotted in S2A Fig. Consistent with the cAMP responses, knocking down a battery of neuropeptide receptors in R23E10 neurons using RNA interference (RNAi) did not substantially modify sleep (S2B Fig). To determine whether the effects of AstA on R23E10 neurons are direct, we incubated brains in tetrodotoxin (TTX), which inhibits the firing of action potentials. TTX does not prevent the AstA-mediated cAMP response of R23E10 cells (Fig 1H, green trace). To further confirm a direct role of AstA, we used RNAi to knock down the AstA-R1 receptor in R23E10 neurons. As seen in Fig 1H (red trace), the cAMP response triggered by application of AstA is attenuated. Quantification of these effects is shown in Fig 1I. Importantly, R23E10 neurons responded normally to DA when AstA-R1 levels are knocked-down indicating that the reduced response to AstA is not the result of a nonfunctional neuron (S2C Fig).

To better understand the relationship between AstA and R23E10 neurons, we used immunohistochemistry to evaluate the overlap between AstA and AstA-GAL4. As previously reported, AstA-GAL4 does not fully recapitulate the AstA expression pattern within the projections to the dFB (Fig 1J–1J”) [37]. Nonetheless, AstA-GAL4 is expressed near the dendritic fields of R23E10 suggesting a physical connection between the 2 group of neurons (Fig 1K–1K”) [25]. To evaluate functional connectivity of the AstA-GAL4, R23E10 circuit, we expressed the P2X2 activator [46] in AstA-GAL4 neurons while measuring cAMP with LexAop-Epac in R23E10-LexA neurons. As seen in Fig 1L and 1M, perfusion of ATP that activates P2X2, leads to a reduction of cAMP levels in R23E10 cells, mimicking the effect of AstA on R23E10 neurons reported above (Fig 1F). No changes in cAMP signaling were observed in parental controls perfused with ATP but that do not express P2X2 indicating the effects are specific to the activation of AstA-GAL4 neurons (Fig 1L, red trace). Thus, we show that the cAMP response of R23E10 neurons to AstA is similar to that seen when AstA-expressing cells are activated.

The sleep-promoting effects of AstA-expressing neurons have been shown to be due to the release of glutamate onto R23E10 neurons [25]. Because AstA is an inhibitory neuropeptide [44,45], we would predict the impact of AstA on R23E10 neurons would be wake promoting and that knocking down AstA-Rs would thus increase sleep. To test this hypothesis, we evaluated sleep after knocking down AstA-R1 or AstA-R2 in R23E10 neurons. As seen in Fig 2A and 2B and 2E and 2F, total sleep is increased when AstA-R1 or AstA-R2 is knocked down in R23E10 neurons, suggesting that the dFB is under consistent allatostatinergic inhibitory tone. Importantly, sleep is also more consolidated during the day (Fig 2C and 2G); sleep consolidation is also increased at night (S3A and S3B Fig). Furthermore, the latency to fall asleep at night is reduced (S3C and S3D Fig). The increase in sleep is not due to unhealthy or sick flies since waking activity is not reduced compared to parental controls (Fig 2D and 2H). To rule out off-target effects of the single RNAi line, we tested 3 additional, independent AstA-R1 RNAi lines and found that sleep is significantly increased in each line (S3E Fig). Given that increasing the activity of AstA neurons has been reported to increase sleep, but the R23E10>AstA-R data indicate that AstA provides a wake signal, we asked whether our results might be due to unknown environmental factors in our laboratory. To evaluate this possibility, we utilized 2 lines (AstA-GAL4 and R65D05-LexA) that express in AstA-expressing clock neurons and increase sleep [25]. As seen in S3F Fig, both AstA-GAL4-dTrpA1 and R65D05-LexA>LexAop-dTrpA1 lines substantially increased sleep consistent with previous reports [25]. Because AstA is an inhibitory neuropeptide, and similar mammalian receptors signals mainly through the Gi pathway [47], we hypothesized that the AstA receptors might be coupled to inhibitory G protein subunits. To test this hypothesis, we used RNAi to knockdown specific G protein subunits in sleep-promoting dFB neurons. As seen in Fig 2I, knocking down Goα47A had no effect on sleep, while knocking down Giα65A using 2 different RNAi lines significantly increased sleep. Furthermore, independently knocking down β and γ1 subunits in sleep-promoting R23E10 neurons also increased sleep (Fig 2I). Thus, these data suggest that the inhibition of dFB neurons can be achieved via coupling with inhibitory G proteins.

Fig 2

Knocking down AstA-R1 or AstA-R2 in R23E10 neurons increases sleep.

(A) Sleep in minutes per hour, (B) total sleep, and (C) daytime sleep-bout duration are increased when AstA-R1 is knocked down in R23E10 neurons compared with both UAS-Dcr2, R23E10-GAL4/+, and AstA-R1/+ parental controls (n = 16/condition, *p < 0.05, modified Bonferroni test). (D) Waking activity is not different between UAS-Dcr2, R23E10-GAL4/+> AstA-R1/+ experimental flies, and both UAS-Dcr2, R23E10-GAL4/+, and AstA-R1/+ controls (n = 16/condition, p > 0.05, modified Bonferroni test). (E) Sleep in minutes per hour, (F) total sleep, and (G) daytime sleep bout duration are increased when AstA-R2 is knocked down in R23E10 neurons compared with both UAS-Dcr2, R23E10-GAL4/+, and AstA-R2/+ parental controls (n = 16/condition, *p < 0.05, modified Bonferroni test). (H) Waking activity is significantly increased in UAS-Dcr2, R23E10-GAL4/+> AstA-R2/+ experimental flies compared to both UAS-Dcr2, R23E10-GAL4/+, and AstA-R2/+ controls (n = 16/condition, p < 0.05, modified Bonferroni test). (I) Total sleep is not changed when Goα47A levels are reduced in R23E10 neurons, but independently knocking down Giα65A, Gβ5, Gβ13F, Gβ76C, and Gγ1 in R23E10 cells increase total sleep compared with parental controls (n = 16/condition; *p < 0.05). Error bars represent SEM. Underlying data is in S2 Data and S1 Datasheet.

Sleep-promoting dFB neurons are diverse

dFB neurons have been hypothesized to gate different aspects of sleep such as locomotion, sensory thresholds, etc. [26,48]. This hypothesis suggests that the dFB could be comprised of independent sets of sleep-promoting neurons that each respond to distinct environmental challenges. If this were to be the case, then we should be able to identify novel dFB projecting GAL4 lines. We obtained 12 GAL4 lines from the Flylight collection that were selected to match the expression pattern of the original dFB sleep-promoting GAL4 lines (104y and C5) [10,49]. We activated these neurons by expressing UAS-TrpA1 as described above. Surprisingly, only 1 driver, R55B01 increased sleep similarly to R23E10 when compared to siblings maintained at 25°C (Figs 3A and 3B and S4A). The parental controls do not show an increase in sleep during the 6-h exposure to 31°C (S1 Fig). Similar results were obtained when expressing the sodium bacterial channel, UAS-NaChBac (S4B Fig). As seen in Fig 3C, R55B01 project to the dFB neuropil (blue arrow) and have cell bodies located in the same anatomical region in the brain as R23E10 neurons. To determine potential overlap between R23E10 and R55B01, we conducted co-labeling experiments by expressing RFP with GAL4/UAS and GFP with LexA/LexAop in the same fly. As seen in S5C–S5C” and S5D Fig R23E10-LexA is similar, but not identical, to the expression pattern of R23E10-GAL4. Importantly, R23E10-LexA and R55B01-GAL4 only share approximately 4 neurons in common (S5A–S5A” and S5B Fig). As with R23E10 neurons, R55B01>UAS-Epac1-camps labels fewer dFB projecting neurons (approximately 8 neurons) compared to R55B01>GFP. Similar to R23E10 neurons, R55B01>UAS-Epac1-camps show robust responses to AstA (Fig 3D). Moreover, AstA positive staining is co-localized with the GFP positive processes of R55B01 neurons (S5E–S5E” Fig). In contrast to R23E10 neurons, the response of R55B01 neurons to DA is either reduced or absent (Fig 3E). A box plot comparing the range of responses of R23E10 and R55B01 neurons to DA is shown in Fig 3F. Given the robust responses of R55B01 neurons to AstA, we evaluated sleep after knocking down AstA-R1 or AstA-R2. As seen in Fig 3G and 3H, sleep is increased upon knocking down either AstA-R1 or AstA-R2 in R55B01 neurons. Quantification of sleep architecture in R55B01>AstAR1 and R55B01>AstAR2 can be found in S6 Fig. Thus, we have identified an additional sleep-promoting driver that projects to the dFB, has little overlap with R23E10 neurons, and is less responsive to DA.

Fig 3

R55B01 neurons respond to AstA.

(A, B) R55B01-GAL4>UAS-dTrpA1 flies showed a significant increase in sleep at 31°C (red) compared to siblings maintained at 25°C (blue; n = 16/group). (C) Expression pattern of R55B01-GAL4>UAS-GFP (blue arrow indicates the location of dFB). (D) Normalized FRET ratio in R55B01>UAS-Epac1-camps in response to 1e−6 M AstA (n = 23 cells). (E) Normalized FRET ratio in R55B01>UAS-Epac1-camps to 3e−5 M DA (n = 10). (F) Box plot comparing the response properties of R23E10> UAS-Epac1-camps and R55B01> UAS-Epac1-camps neurons to 3e−5 M DA. The bottom and top of each box represents the first and third quartile, and the horizontal line dividing the box represents the median. The whiskers represent the minimum and maximum individual cell responses; n = 68 cells for R23E10 and n = 115 cells for R55B01. (G, H) Knocking down AstA-R1 or AstA-R2 in R55B01 neurons increases sleep. Sleep in minutes per hour in R55B01-GAL4>AstAR1 and R55B01-GAL4>AstAR2 (n = 16/condition). Error bars represent SEM. Underlying data is in S3 Data and S1 Datasheet. AstA, allatostatin-A; DA, dopamine; dFB, dorsal fan-shaped body.

R23E10 neurons differentially integrate sleep-relevant stimuli

Together with the literature, these data identify 2 major wake-promoting signals that impact R23E10 neurons, AstA and DA [35,36]. Although both signals are inhibitory, they derive from neuronal circuits with opposite functions; AstA is released from sleep-promoting neurons while DA is released from wake-promoting neurons [25,35,36]. Given these divergent roles, we hypothesize that AstA and DA will operate largely independently. To test this hypothesis, we used the live-brain imaging approach described above (Fig 1D) and evaluated the response of R23E10 neurons to either DA or allatostatin during conditions that alter sleep drive. Because R55B01 neurons respond less to DA, we focused solely on R23E10 neurons; R55B01 neurons will be studied later. We first evaluated the response properties of R23E10 neurons in unperturbed flies under conditions characterized by large changes in sleep time (e.g., ontogeny, gender, individual differences). We then evaluated the response properties of R23E10 neurons after experimental interventions that modulate sleep drive (e.g., sleep loss, starvation, training that induces long-term memory) [3,28,50-53] (Fig 4).

Fig 4

Sleep-promoting neurons integrate a variety of relevant stimuli.

(A, B) Response of R23E10 neurons in young (0–1 day old) and adult (6–8 day old) flies in response to AstA (10−6 M) (n = 13, 22 cells) or DA (3e−5 M) (n = 15, 14 cells). (C, D) Response of R23E10 neurons in spontaneously short-sleeping flies (<400 min/day) and normal sleeping siblings (600–900 min/day) (AstA, n = 18, 20 cells) and (DA, n = 12, 19 cells). (E, F) Response of R23E10 neurons following 12-h sleep deprivation compared to untreated siblings (AstA, n = 17, 17 cells) and (DA, n = 29, 26 cells). (G, H) Response of R23E10 neurons following 18 h starvation compared to untreated siblings; (AstA, n = 20, 14 cells) and (DA, n = 23, 15 cells). (I, J) Response of R23E10 neurons following social enrichment compared to isolated siblings; (AstA, n = 9, 5 cells) and (DA, n = 12, 8 cells). (K, L) Response of R23E10 neurons following spaced training compared to naïve controls (AstA, n = 4, 7 cells) and (DA, n = 25, 10 cells). (M) Sleep is increased in R23E10-GAL4>CaLexA flies following a spaced courtship training (n = 10 per group, p < 0.05). (N) Representative confocal stack of the brains of a naïve (top) and spaced trained (bottom) R23E10-GAL4>CaLexA flies stained with anti-GFP antibody. Flies were dissected on the day after courtship conditioning. (O) Quantification of GFP staining intensity in R23E10 neurons in naïve and spaced trained R23E10-GAL4>CaLexA flies (n = 107 neurons for naïve and 100 neurons for trained). Insets are quantification of the corresponding traces. Underlying data is in S4 Data and S1 Datasheet. AstA, allatostatin-A; DA, dopamine.

As with humans, sleep is highest in young flies and then stabilizes in early adulthood. Interestingly, R23E10 neurons are less responsive to DA in 0- to 1-day-old flies compared to 6- to 8-day-old mature adults; no changes were observed in response to AstA (Fig 4A and 4B). The response properties of R23E10 neurons to DA were similar in 6- to 8-day-old flies and 30- to 38-day-old flies S7 Fig. In contrast to age, we did not observe any changes in the response properties of R23E10 neurons in male and female flies despite large sexual dimorphisms in sleep behavior [52,54]. We have previously shown that individual differences in sleep time can be exploited to evaluate sleep regulation and function [55]. Importantly, our previously published data indicate that spontaneously short-sleeping flies appear to experience both high wake-drive and high sleep drive simultaneously [56]. With that in mind, we evaluated the response properties of R23E10 neurons in spontaneously short-sleeping R23E10>UAS-Epac1-camps (i.e., total sleep time less than 400 min) compared to normal sleeping siblings (800 to 960 min sleep). As seen in Fig 4C, R23E10 neurons appear to be under stronger inhibitory tone from AstA in spontaneous short sleepers compared to normal sleeping siblings. Although one would predict a stronger response of R23E10 neurons to the wake-promoting effects of DA, short-sleeping flies were less responsive to DA (Fig 4D). Future studies will be needed to determine whether incongruent responses of R23E10 neurons to AstA and DA reveal underlying deficits in sleep regulation.

We next asked whether distinct experimental interventions that modulate sleep drive would alter the physiological response properties of R23E10 neurons in similar or dissimilar ways. Both sleep deprivation and extended periods of starvation are followed by a compensatory increase in sleep [3,28,57]. To evaluate how sleep disruption would influence R23E10 neurons, flies were sleep deprived for 12 h or starved for 18 h and compared to untreated siblings. As seen in Fig 4E, cAMP responses to AstA are increased following sleep deprivation but were unchanged in response to DA (Fig 4F). Surprisingly, the response of R23E10 neurons to either AstA or DA were not changed while being starved (Fig 4G and 4H and see below). In contrast to sleep deprivation and starvation, social enrichment induces plasticity in specific neural circuits to increase sleep without exposing flies to sleep loss [51,58]. Changes in sleep following social enrichment have been mapped to pigment-dispersing factor (PDF)-expressing clock neurons [58]. Interestingly, AstA signaling is modulated by PDF [27]. Thus, to evaluate the impact of social rearing, we housed flies in a socially enriched environment (50 flies/vial) for 5 days and evaluated cAMP responses compared to isolated siblings. As seen in Fig 4I, R23E10 neurons of socially enriched or isolated flies show similar responses to AstA. However, DA responses of R23E10 cells are strongly reduced by social enrichment (Fig 4J). These data demonstrate that disparate behavioral manipulations that increase sleep drive in different ways induce independent physiological responses of R23E10 neurons to AstA and DA.

Sleep-dependent memory consolidation has been linked to both the ventral and the dFB [10,59]. Thus, we asked whether R23E10 neurons would modify their physiological responses to a training protocol that induces LTM [9]. As seen in Fig 4K, responses of R23E10 neurons to AstA are not different following courtship conditioning. However, training resulted in a dramatic reduction in the response of R23E10 neurons to DA (Fig 4L). Interestingly, the effect of training on DA responses could still be observed 24 h after the end of the training only returning to baseline 48 h later suggesting long-term plastic changes in R23E10 neurons (S8 Fig). To determine whether the changes in cAMP were due to nonspecific effects of courtship or to memory consolidation, we evaluated cAMP levels following a massed training protocol consisting of a single 3-h session that does not result in LTM formation [10]. As seen in S8 Fig, massed training did not alter the responses of R23E10 cells to DA. Furthermore, no changes in responses to DA were found when training was followed by 4 h of sleep deprivation S8 Fig. To further evaluate the role of the dFB in courtship memory, we used CaLexA (calcium-dependent nuclear import of LexA) to see if R23E10 neurons might show sustained activity following a spaced training protocol [60]. R23E10 flies expressing CaLexA were exposed to a training protocol consisting of 3 × 1 h individual pairings of a naïve male with a non-receptive female target separated by a rest period of 1 h. As seen in Fig 4M, sleep is increased following training compared with non-trained siblings consistent with previous reports [51,58]. R23E10 neurons show higher GFP signal in trained animals when assessed the following morning compared with their naïve counterparts indicating that sleep-promoting neurons are more active following courtship memory training (Fig 4N and 4O for quantification). Together, these data indicate that the response properties of R23E10 neurons display long-lasting changes following protocols that induce LTM.

Prior feeding experience alters the recruitment of DA receptors to modulate sleep

Initial studies indicated that the wake-promoting effects of DA on dFB neurons are mediated by Dopamine 1-like receptor 1 (Dop1R1) [35,36,61]. However, the role of Dop1R1 has been called into question [19]. We have recently shown that the constellation of receptors expressed on a neuron can be altered by starvation to include a receptor that is not typically present [3]. Specifically, our data suggest that the new receptor is recruited to amplify wake-promoting signals in clock neurons to allow animals to engage in adaptive waking behaviors. To determine whether the phenomenon of recruiting new wake-promoting receptors to a neuron will generalize to the dFB, we re-examined the role of Dop1R1 in R23E10 neurons. In addition, we also evaluated R55B01 neurons because these cells have a limited response to DA. We hypothesized that if Dop1R1 is needed to support waking following a metabolic challenge, knocking it down would manifest as an increase in sleep. Since 18 h of starvation did not change the response properties of R23E10 neurons (Fig 4G and 4H), we used time-restricted feeding to safely impose an alternate challenge of longer duration [62-64]. The time-restricted feeding protocol is shown in Fig 5A. Flies are only given access to food between 8 AM and 5 PM for a total of 7 days (restricted). After time-restricted feeding, flies are placed into Trikinetics tubes where they were allowed to eat ad lib while sleep is evaluated. Siblings that were maintained in vials with standard food available ad lib and flipped at the same times as their restricted counterparts served as treatment controls (Fig 5A). Consistent with previous reports [19,36], knockdown of Dop1R1 in R23E10 neurons did not alter sleep in flies that were able to feed ad lib (Fig 5B and 5D and 5E). Similarly, no changes in sleep were observed in untreated R55B01>Dop1R1 flies (Fig 5C–5E). However, both R23E10>Dop1R1 and R55B01>Dop1R1 flies displayed dramatic increases in sleep following 7 days of time-restricted feeding compared to their R23E10/+, R55B01/+, and Dop1R1 /+ parental controls (Fig 5F–5I). Changes in sleep during time-restricted feeding are shown in S9 Fig. These data are consistent with the hypothesis that, under certain circumstances, a new receptor can be recruited to amplify wake-promoting signals.

Fig 5

Time-restricted feeding increases sleep via Dop1R1.

(A) Protocol for time-restricted feeding; flies were given food during ZT 8 AM to 5 PM. (B, C) During ad lib feeding, sleep in R23E10>Dop1R1 and R55B01>Dop1R1 flies does not differ from R23E10/+, Dop1R1/+, or R55B01/+ parental controls (n = 13–16 flies/group ANOVA for Genotype F[2,42] = 2.8, p = 0.07 and ANOVA for Genotype F[2,41] = 1.9, p = 0.15 for R23E10 and R55B01, respectively). (D) Daytime sleep was not altered in either R23E10>Dop1R1 or R55B01>Dop1R1 flies compared to both parental controls (ANOVA for Genotype F[2,42] = 2.8, p = 0.02 and F[2,41] = 2.9, p = 0.06, *p < 0.05, modified Bonferroni test). (E) Nighttime sleep was not changed in either R23E10>Dop1R1 or R55B01>Dop1R1 flies compared to parental controls (ANOVA for Genotype F[2,42] = 0.9, p = 0.40 and ANOVA for Genotype F[2,41] = 0.79, p = 0.46, *p < 0.05, modified Bonferroni test). (F, G) Following time-restricted feeding, sleep is increased in R23E10>Dop1R1 and R55B01>Dop1R1 flies compared to R23E10/+, Dop1R1/+, or R55B01/+ parental controls (n = 13–16 flies/group ANOVA for Genotype F[2,40] = 68.8, p = 9.9eE-16 and ANOVA for Genotype F[2,35] = 38.1.8, p = 8.1eE-10 for R23E10 and R55B01, respectively). (H) Daytime sleep was significantly increased in R23E10>Dop1R1 and R55B01>Dop1R1 flies following time-restricted feeding (ANOVA for Genotype F[2,40] = 13.6, p = 2.96E-05 and F[2,35] = 30.5, p = 2.07E-08, *p < 0.05, modified Bonferroni test). (I) Nighttime sleep was not changed in either R23E10>Dop1R1 or R55B01>Dop1R1 flies compared to parental controls (ANOVA for Genotype F[2,40] = 11.41, p = 0. 0.0001 and ANOVA for Genotype F[2,35] = 8.9, p = 0.0007, *p < 0.05, modified Bonferroni test). Error bars represent SEM. Underlying data is in S5 Data and S1 Datasheet. ZT, zeitgeber time.

Time-restricted feeding is characterized by defined intervals of feeding and fasting. Thus, during time-restricted feeding, Dop1R1 may be recruited to R23E10 neurons to either support waking during starvation or to support waking when food is available. To distinguish between these possibilities, we monitored sleep in R23E10>Dop1R1 and R55B01>Dop1R1 flies and their parental controls during baseline, during 18 h of starvation, and for 3 days after being placed back onto food (Fig 6A and 6B). As above, no changes in baseline sleep were observed while knocking down Dop1R1 in either R23E10 or R55B01 neurons compared to parental controls (Fig 6A–6D). Moreover, R23E10>Dop1R1 and R55B01>Dop1R1 flies exhibited similar sleep patterns to parental controls during starvation (S10A Fig). However, during recovery following 18 h of starvation, an increase in sleep was observed in both R23E10>Dop1R1 and R55B01>Dop1R1 flies compared to their respective parental controls (Fig 6A and 6B). It is important to highlight that sleep in the experimental lines only diverged from their parental controls after several hours, or more, of recovery (Fig 6A and 6B arrows). Quantification of sleep during recovery day 1 and recovery day 2 is shown in S10B and S10C Fig. Sleep stabilized in both R23E10>Dop1R1 and R55B01>Dop1R1 flies on the third day of recovery and remained elevated for several days thereafter (Fig 6A, 6B and 6E). Thus, these data indicate that the Dop1R1 is recruited to R23E10 and R55B01 neurons to support waking behavior during recovery from starvation.

Fig 6

Starvation alters recovery sleep and the response properties of sleep-promoting neurons.

(A, B) Sleep profiles in R23E10>Dop1R1, R55B01>Dop1R1 flies, and their parental controls R23E10/+, Dop1R1/+, and R55B01/+ during baseline, 18 h of starvation and 3 days of recovery (n = 13–16 flies/group). (C, D) During baseline, no changes in daytime or nighttime sleep were observed in R23E10>Dop1R1 or R55B01>Dop1R1 flies compared to both parental controls (Daytime: ANOVA F[2,36] = 8.1, p = 0.001 and F[2,38] = 2.26, p = 0.11), (Nighttime: ANOVA F[2,36] = 1.5, p = 0.22 and ANOVA F[2,38] = 1.6, p = 0.20, *p < 0.05, modified Bonferroni test). (E) On recovery day 3, daytime sleep is increased in R23E10>Dop1R1 and R55B01>Dop1R1 flies compared to R23E10/+, Dop1R1/+, or R55B01/+ parental controls (ANOVA F[2,36] = 9.1, p = 0.0006 and F[2,38] = 12.1, p = 8.43E-05). (F) Nighttime sleep is not changed in R23E10>Dop1R1 and R55B01>Dop1R1 flies compared to R23E10/+, Dop1R1/+, or R55B01/+ parental controls (ANOVA F[2,36] = 3.9, p = 0.02 and F[2,38] = 0.09, p = 0.9 *p < 0.05, modified Bonferroni test). Error bars represent SEM. Underlying data is in S6 Data and S1 Datasheet.

To gain additional insight into the physiological impact of starvation, we used live-brain imaging to evaluate the responses of R23E10>UAS-Epac1-camps; UAS-Dop1R2 and R55B01>UAS-Epac1-camps; UAS-Dop1R2 to DA under baseline and on recovery day 2 from starvation. While both Dop1R1 and Dop1R2 couple to Gαs, Dop1R2 appears to also activate Gαq [65]. We hypothesized that knocking down Dop1R2 would reduce the response of R23E10 neurons to DA under baseline conditions and that the neurons would respond to DA following 2 days of recovery from starvation. R23E10>UAS-Dop1R2 flies displayed an increase in sleep under baseline conditions consistent with previous reports [19]. As seen in Fig 7A and 7B, R23E10>UAS-Epac1-camps; UAS-Dop1R2 neurons did not show a strong response to DA during baseline. However, during recovery from starvation R23E10>UAS- Epac1-camps; UAS-Dop1R2 neurons exhibited a significant increase in their response to DA suggesting a new receptor is added (Fig 7A and 7B). Next, we evaluated the effects of starvation on R55B01 neurons. As seen in Fig 7C and 7D, neither R55B01>UAS-Epac1-camps nor R55B01>UAS-Epac1-camps;UAS-Dop1R2 responded to DA under baseline conditions. Although much smaller than that observed for R23E10 neurons, R55B01>UAS-Epac1-camps;UAS-Dop1R2 neurons did respond modestly to DA after 2 days of recovery from starvation (Fig 7C and 7D). While we cannot exclude a role of other DA receptors (e.g., Dopamine/Ecdysteroid receptor) for the observed changes, when viewed with the sleep experiments shown above (Figs 5 and 6), these data suggest that during recovery from starvation the constellation of DA receptors in R23E10 and R55B01 changes and most likely includes the recruitment of Dop1R1. These data provide new insights into the mechanisms used by sleep circuits to link internal states and prior waking history with sleep need.

Fig 7

Starvation alters the response properties of sleep-promoting neurons.

(A) R23E10 cells expressing the cAMP sensor Epac1-camps respond to 3 × 10−5 M DA (black line; n = 22 neurons); the response is dramatically reduced in R23E10> UAS-Epac1-camps; Dop1R2 (white line; n = 25 neurons) and partially restored during recovery from 18 h of starvation (gray line; n = 29 neurons). (B) Quantification of a; F[2,73] = 7.7, p = 0.0009, *p < 0.05, modified Bonferroni test. (C) The response of R55B01> UAS-Epac1-camps (black; n = 13) and R55B01> UAS-Epac1-camps; Dop1R2 (white; n = 19) neurons to DA is very low; following recovery from starvation, the response of R55B01 neurons to DA increased (gray line; n = 20). (D) Quantification of c; F[2,49] = 9.9, p = 0.0002, *p < 0.05, modified Bonferroni test. Error bars represent SEM. Underlying data is in S7 Data and S1 Datasheet. cAMP, cyclic adenosine monophosphate; DA, dopamine.

Discussion

In this work, we ask whether the activity of sleep-promoting dFB neurons reflects the summation of their upstream inputs in a winner-take all strategy or if they can change their own response properties to influence how environmental demands alter behavioral state. Although one possibility does not preclude the other, our data indicate that both time-restricted feeding and approximately 18 h of starvation, alter the constellation of DA receptors expressed by R23E10 and R55B01 neurons. These results emphasize that the ability of upstream circuits to alter behavior will not only depend upon the strength of the incoming signals but also the recent historical context of R23E10 and R55B01 neurons themselves. In that regard, it important to note that while sleep competes with all motivated waking behavior, the proper regulation of motivated behaviors degrade in the absence of sleep [50,66-74]. Together, these observations suggest that it will not be possible to fully understand sleep’s function without knowing how sleep-promoting neurons alter their functional properties to prioritize conflicting motivational states.

The dFB is comprised of independent sets of sleep-promoting neurons

Previous studies indicate that the dFB can independently gate different aspects of sleep [26]. Moreover, the expression of the gap junction innexin6 in R23E10 neurons is required to gate sensory thresholds and sleep time [48]. Together, these data suggest the presence of independent sets of dFB neurons. Although we screened through a refined set of hand-selected GAL4 lines that project into the dFB in a manner similar to 104y-GAL4 and C5-GAL4 [10,49], we only identified 1 additional GAL4 line, R55B01, that reliably increased sleep when crossed with UAS-dTrpA1. Similar results were obtained when activating the neurons using the bacterial sodium channel UAS-NaChBac. This latter result is relevant given that sleep is modulated by temperature and that GABAergic neurons projecting onto dFB regulate temperature-dependent changes in sleep [75,76]. We also identified dFB projecting GAL4 lines that increased waking. However, the precise role that these GAL4 lines play in waking awaits additional inquiry. Nonetheless, these data suggest that sleep-promoting dFB neurons are limited in number. It should be noted that the extent of the overlap between R23E10 and R55B01 remains unclear as the R23E10-LexA driver does not fully recapitulate the R23E10-GAL4 expression patterns (S5 Fig). Although it may be possible to subdivide R23E10 and R55B01 neurons further, our data suggest that there are at least 2 independent sets of sleep-promoting neurons that have projections into the dFB (see below for additional discussion). In mice, sleep-promoting neurons are also diverse as assessed by their role in regulating behavior and their projection patterns [77].

Allatostatin inhibits sleep-promoting neurons

Although AstA-expressing neurons are believed to promote sleep by releasing glutamate onto R23E10 neurons [25], our data identify a wake-promoting effect of AstA on R23E10 neurons. Specifically, we find that knocking down either the AstA-R1 or AstA-R2 in R23E10 or R55B01 neurons using several independent RNAi lines results in a substantial increase in sleep. The observation that knocking down either the AstA-R1 or AstA-R2 in R23E10 neurons increases sleep, suggests that R23E10 neurons are under constant inhibitory tone from AstA and that both receptors play an important role in modulating sleep in untreated flies. In addition, we report that the application of 1 mM ATP to R23E10>UAS-Epac1-camps; AstA-GAL4>P2X2 neurons reduces cAMP levels similarly to that observed with application of AstA. Although our data appear to be in conflict with that of Ni and colleagues, who provide evidence that AstA-GAL4 provide an excitatory drive onto R23E10 neurons, we hypothesize that the co-release of inhibitory AstA with excitatory glutamate may attenuate the overexcitement of R23E10 neurons during high sleep drive and allow animals to maintain wakefulness in dangerous or life-threatening conditions.

In support of this hypothesis, a growing body of evidence indicates that sleep and wake regulating neurons co-release different transmitters and neuropeptides to better match sleep need with environmental demands [78]. Indeed, it has been suggested that antagonistic neurotransmitter co-release may both prevent excessive excitation of postsynaptic targets and increase the flexibility of neuronal networks [45,78,79]. For example, the co-release of GABA by histaminergic tuberomammillary nucleus is believed to prevent histamine-induced overexcitement of downstream circuits and thereby better support normal amounts of waking [80]. Similarly, the co-release of galanin by noradrenergic neurons is believed to prevent overexcitement of locus coeruleus [81]. Although the co-expression of AstA and glutamate has been established [25,82], our data reveal a new mechanism for how incompatible motivational drives can be regulated.

An alternate possibility to explain the wake-promoting effects of AstA on R23E10 and R55B01 neurons may be that R23E10 and R55B01 are, in fact, a heterogeneous sets of neurons that also include wake-promoting neurons and/or neurons whose primary role is to promote active sleep at the expense of deep sleep [83]. Active sleep is best observed following 24 h of sleep deprivation and is associated with dramatic reductions in the response to external sensory stimuli. If AstA were to inhibit a subset of R23E10 neurons that promote active sleep, one would predict an increase in deep sleep. The diversity in the responses to DA seen in both R23E10 and R55B01 neurons (Fig 3F) further supports the hypothesis that these GAL4 lines express in heterogeneous set of neurons with different functions; a previous study has found similar heterogeneity in R23E10 neurons [38]. Alternatively, it is possible that AstA does not inhibit the R23E10 neurons per se but rather alters temporal spike patterns to favor active sleep or deep sleep in a manner similar to that observed in clock neurons [84]. Indeed, AstA has been shown to alter the spike-timing precision of mechanoreceptor afferents in Carcinus maenas and Galanin, alters spontaneous spike firing in rodents [85,86]. These possibilities will be explored in future studies.

It is important to note that neuropeptides are notoriously pleiotropic and also work as neurohormones [44]. Thus, it is possible that additional sleep- and wake-promoting circuits downstream of AstA-GAL4 or 65D05-GAL4 will be found elsewhere in the brain. It is interesting to note that neurons expressing the mouse galanin, which is functionally similar to AstA, also regulate conflicting behaviors [77]. Specifically, galanin-expressing neurons that project to the tuberomammillary nucleus promote sleep while galanin-expressing neurons that project to the medial amygdala promote parental behaviors [77,87,88]. Thus, it will be important for future studies to discern how AstA and galanin circuits regulate competing activities in other circuits.

R23E10 neurons differentially encode arousal signals

Increased sleep drive may be maladaptive in many circumstances since falling asleep could place the individual in danger of immediate physical harm [7,8]. Increased sleep also competes with important waking behaviors such as foraging, eating, and mating [3,4]. However, sleep plays a critical role in learning and memory, supports adaptive behavior, and facilitates creative insight [9-12]. How do sleep-promoting neurons distinguish between competing drives? Our live-brain imaging data suggest the intriguing possibility that R23E10 neurons can distinguish between different types of waking and change their response properties accordingly. Specifically, R23E10 neurons selectively decrease their response to the wake-promoting effects of DA following conditions that promote plasticity. In contrast, R23E10 neurons increase their response to the wake-promoting effects of AstA during conditions when sleep drive is high but the expression of sleep might be dangerous (e.g., during sleep deprivation). Finally, R23E10 neurons appear to contain a memory trace of starvation as evidenced by the increase inhibitory tone conveyed by the recruitment of Dop1R1. It should be emphasized that we did not use TTX when evaluating the response properties of R23E10 neurons to different types of waking. As a consequence, it is possible that the difference in the intrinsic release of AstA and DA during these manipulations modified the response of dFB neurons. Such a possibility will be evaluated in future studies. Nonetheless, R23E10 neurons are plastic and can utilize AstA and DA in very different ways to favor specific behavioral outcomes.

As mentioned, increasing inhibitory signals onto sleep-promoting neurons during sleep deprivation may be an adaptive response that allows animals to stay awake despite increasing sleep drive. Indeed, lesioning galanin neurons in the preoptic area were found to reduce sleep rebound in zebrafish [89] and mice [90]. However, AstA is also reported to be a satiety signal that limits feeding [27,32,37,91]. The effects of AstA on feeding seem to be in conflict with the studies linking sleep deprivation to increased food intake [92-95]. Nonetheless, recent reports have identified complex interactions between hunger and satiety signals that may be nutrient specific and can be modified by Hebbian plasticity [32,34,40,44]. Thus, it is reasonable to expect that sleep deprivation must differentially activate neurons regulating a variety of motivated behaviors including satiety and sleep drive. Indeed, these data indicate that sleep deprivation may be harnessed as a tool to further explore how internal states impact neuronal circuits regulating conflicting goal-directed behaviors [30].

R23E10 neurons support long-term memory

Interestingly, the ventral fan-shaped body (vFB) promotes sleep and regulates the activity of Dopaminergic aSP13 neurons (DAN-aSP13s) to consolidate long-term memory as assessed using courtship conditioning [59]. Surprisingly, activating R23E10 neurons does not alter the activity of DAN-aSP13s neurons suggesting that the dFB may not be involved in courtship memory [59]. It should be noted, however, that R23E10 neurons regulate the activity of other sets of arousal promoting neurons including the MV1 Dopaminergic neurons and octopaminergic arousal neurons [25,96]. These data suggest that the dFB and vFB may play distinct roles in different memory assays. Nonetheless, our data clearly indicate that R23E10 neurons change their response properties following training that induces LTM but not massed training. Importantly, the changes in R23E10 neurons only return to baseline 48 h after training. It is possible that R23E10 neurons play an important, yet indirect, role in courtship memory. That is, activating R23E10 neurons strongly suppresses the response to external sensory stimuli [48,83]. Thus, the activity of R23E10 neurons may protect sleep by limiting the opportunity for external stimuli to wake the animal up. Under this scenario, dFB and vFB neurons would work in concert to carry out sleep functions.

In addition, we find that courtship conditioning increases the activity of R23E10 neurons as assessed by CaLexA. Consistent with this observation, spaced-training reduced the inhibitory effects of DA on R23E10 neurons. Importantly, massed-training, which only induced short-term memory, does not alter the response of R23E10 neurons to DA. A previous report has shown that the ectopic expression of Dop1R1 in dFB neurons reduces sleep and fails to rescue courtship conditioning memory in Dopamine transporter mutants (fmn) [36]. Together, these data emphasize that reducing Dopaminergic signaling to the dFB neurons is important for long-term memory following courtship conditioning.

Internal state remodels dopamine receptors in sleep-promoting neurons

The arousal promoting properties of DA projections to the dFB are well established [19,25,35,36,61]. However, which DA receptor is responsible for the increased waking remains controversial. Initial results, using the ectopic expression Dop1R1 and live-brain imaging implicated Dop1R1 [35,36]. Nonetheless, RNAi knockdown of Dop1R1 in dFB neurons did not alter sleep [35]. A subsequent study demonstrated that activation of Dop1R2 results in a transient hyperpolarization of dFB neurons followed by a lasting suppression of excitability lasting minutes; knocking down Dop1R2 but not Dop1R1 altered sleep [19]. Our data replicate previous RNAi studies that have failed to observe a change in sleep following the expression of Dop1R1 in dFB neurons in untreated flies. Indeed, no changes in FRET signal were observed upon the bath application of DA to untreated R23E10>UAS-Epac1-camps; Dop1R2 or R55B01>UAS-Epac1-camps; Dop1R2 flies further supporting the role of Dop1R2 in baseline sleep regulation.

However, in the context of time-restricted feeding and starvation, knocking down Dop1R1 in R23E10 or R55B01 neurons resulted in substantial increases in sleep. We hypothesize that both time-restricted feeding and starvation place energy-saving sleep drive in conflict with a metabolic signal that facilitates motivated waking behavior. Increased wake drive would allow the animals to be more alert and productive during their primary wake period. Under these circumstances, Dop1R1 would be recruited to provide additional inhibitory tone to sleep-promoting R23E10 and R55B01 neurons during the light period. Interestingly, starvation results in the recruitment of the pigment-dispersing factor receptor (Pdfr) into wake-promoting large ventrolateral neurons (lLNvs) to facilitate waking at the end of the biological day [3]. Whether other sleep- and wake-promoting circuits are regulated in a similar fashion is an open question that will require additional inquiry.

Conclusions

Given the important roles that DA and AstA play in regulating motivated behaviors, including sleep, we studied their impact on a subset of sleep-promoting neurons. Our data reveal that AstA is wake-promoting and likely serves to maintain waking during periods of high sleep drive. In addition, our data reveal that time-restricted feeding or 18 h of starvation recruits the Dop1R1 to sleep-promoting neurons to maintain wakefulness during the day. These results are consistent with our previous data showing that the wake-promoting lLNvs recruit the Pdfr following sleep loss to facilitate waking and that wing-cut reactivates a developmental sleep circuit [3,97]. The ability of sleep- and wake-promoting neurons to alter their own physiology, including the recruitment of a new receptor, provides important clues into sleep regulation and function.

Changes in sleep in parental controls following 6 h at 31°C.

(A) The % change from baseline sleep at 25°C seen following switch to 31°C between 9 AM and 3 PM. (B) Confocal image of R23E10>UAS-Epac1-camps. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

R23E10 neuropeptide RNAi screen for Daytime Sleep.

(A) Individual cAMP responses of R23E10 neurons expressing UAS-Epac1-camps shown as % change in FRET ratio during exposure to crustacean cardioactive peptide (CCAP), Drosophila myosuppressin (DMS), allatostatin C (astA C), proctolin, TPAEDFMRFamide, corticotropin-releasing factor-like diuretic hormone 44 (DH44), Tachykinin 1, Tachykinin 3, short neuropeptide F (sNPF), adipokinetic hormone (AKH), corazonin, and melatonin. (B) Daytime sleep in female 5-day-old flies expressing RNAi lines for the depicted neuropeptide receptors using R23E10-GAL4 and their parental controls (n = 14–16 flies/genotype). To be significant, the experimental lines must be significantly different from both the GAL4/+ (red line) and the UAS/+ (white bar) parental controls: FR: FMRFamide Receptor, DHRRR2: Diuretic hormone 44 receptor 2, DMSR1: Myosuppressin receptor 1, DHRRR1: Diuretic hormone 44 receptor 1, cchamideR: CCHamide-1 receptor, NepyR: RYamide receptor, Capar: Capability receptor, Pk2R1: Pyrokinin 2 receptor 1; CCKLR: Cholecystokinin-like receptor at 17D1A, DH31: Diuretic hormone 31,AstC-R1: Allatostatin C receptor 1, TkR99D: Tachykinin-like receptor at 99D, MsR2: Myosuppressin receptor 2, Lkr: Leucokinin receptor, CrzR: Corazonin receptor, Proc: Proctolin receptor. Red line is to facilitate comparisons with R23E10/+ parental control. (C) RNAi knockdown of the AstA-R1 receptor in R23E10 neurons did not affect the cAMP response to DA. Error bars represent SEM. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Expressing AstA-R1-RNAi or AstA-R2-RNAi in R23E10 neurons alters sleep architecture.

(A, B) Nighttime sleep bout duration is increased in UAS-Dcr2, R23E10-GAL4/+> AstA-R1RNAi/+, and UAS-Dcr2,R23E10-GAL4/+> AstA-R2RNAi/+ experimental flies compared with UAS-Dcr2, R23E10-GAL4/+, AstA-R1RNAi/+, and AstA-R2RNAi/+ parental controls (n = 16/condition, *p < 0.05, modified Bonferroni test). (C, D) Sleep latency is shortened in UAS-Dcr2, R23E10-GAL4/+> AstA-R1RNAi/+, and UAS-Dcr2,R23E10-GAL4/+> AstA-R2RNAi/+ experimental flies compared with UAS-Dcr2, R23E10-GAL4/+, AstA-R1RNAi/+, and AstA-R2RNAi/+ parental controls (n = 16/condition, *p < 0.05, modified Bonferroni test). (E) Total sleep is increased when levels of AstA-R1 is decreased in R23E10 sleep-promoting neurons using 3 additional independent RNAi lines (n = 16/condition, *p < 0.05, modified Bonferroni test). (F) Sleep is increased in AstA-GAL4>UASdTrpA1 and 65D05-LexA>LexAopdTrpA1 flies at 31°C compared with siblings maintained at 25°C; parental controls did not show an increase in sleep at 31°C (n = 14–16/condition and genotype, *p < 0.05, modified Bonferroni test). Error bars represent SEM. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Screen of Janelia-GAL4 lines that express in the dorsal fan-shaped body in a pattern similar to that observed with 104y-GAL4 and C5-GAL4.

(A) Despite similar anatomical profiles, most dorsal fan-shaped body drivers do not reliably impact sleep when expressing UAS-Transient receptor potential cation channel A1 and raising the temperature to 31°C (n = 14–16 flies/genotype). The data for R55B01>dTrpA are the same as in Fig 3A. (B) R23E10>UAS-NaChBac and R55B01>UAS-NaChBac sleep more than R23E10/+, UAS-NaChBac/+, and R55B01/+ parental controls. Error bars represent SEM. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Anatomy of R23E10 and R55B01 neurons.

(A) Representative confocal stack focusing on the area containing cell bodies of a R23E10-LexA>LexAop-GFP, R55B01-GAL4>UAS-RFP fly brain stained with anti-GFP antibody, (A’) anti-RFP antibody (magenta) and (A”) a merged image. Yellow arrows on the merge image indicate cells that express both GFP and RFP. (B) Quantification of the number of cells expressing only GFP, only RFP or both GFP and RFP. (C) Representative confocal stack focusing on the area containing the cell bodies of a R23E10-LexA>LexAop-GFP, R23E10-GAL4>UAS-RFP fly brain stained with anti-GFP antibody, (C’) anti-RFP antibody (magenta), and (C”) a merged image. (D) Quantification of the number of cells expressing only GFP, only RFP or both GFP and RFP. (E) Representative confocal stack of a using R55B01-GAL4>UAS-GFP brain stained with anti-GFP antibody, (E’) anti-AstA antibody (magenta), and (E”) a merged image. Error bars represent SEM. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Expressing AstAR1 or AstAR2 R55B01 neurons alters sleep architecture.

Sleep parameters in UAS-Dcr2, R55B01-GAL4/+> AstA-R1/+, UAS-Dcr2, R55B01-GAL4/+> AstA-R2/+ experimental flies and both UAS-Dcr2, R55B01-GAL4/+ and AstA-R1/+, AstA-R2/+ control flies for: (A, B) total sleep, (C, D) nighttime sleep bout duration, (E, F) nighttime sleep latency, (G, H) daytime sleep bout duration, and (I, J) counts/waking minute; (n = 16/condition, *p < 0.05, modified Bonferroni test). Error bars represent SEM. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Response of R23E10 neurons to Dopamine is not changed in older flies.

Response of R23E10 neurons in young (0–1 day old), adult (6–8 day old), and old (30–38 day old) flies in response to Dopamine (3e−5 M) (n = 15, 14, 6 cells). The data are expressed as amplitude % change; data for young and adult flies are taken from data in Fig 4B. Error bars represent SEM. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Space-trained flies vs. naïve controls.

Space-trained flies vs. naïve controls. The reduction of DA responses remained significant 24 h after the end of the training but not 48 h post-training. A Massed training courtship protocol that does not induce LTM had no significant effect on amplitude of DA response in R23E10 neurons; 4 h of sleep deprivation did not alter DA responses in R23E10 neurons (n = 12–25 cells, *p < 0.05). Error bars represent SEM. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Time-restricted feeding changes sleep-promoting neurons.

(A) Sleep profiles in R23E10>Dop1R1, flies and their parental controls R23E10/+, and Dop1R1/+ during baseline, 7 days of time-restricted feeding and 5 days of recovery (n = 24–26 flies/group); gray arrow indicates disruption in data collection. (B) Daytime sleep is increased following time-restricted feeding in R23E10>Dop1R1 compared to both parental controls. A Genotype (2) x Time (13) ANOVA revealed a Genotype X Time interaction: F[,] = 5.5, p = 9.99E-16; *p < 0.05, modified Bonferroni test. (C) Sleep profiles in R55B01>Dop1R1 flies and their parental controls R55B01/+, and Dop1R1/+ during baseline, 7 days of time-restricted feeding and 5 days of recovery (n = 24–30 flies/group) (Dop1R1-RNAi flies are the same as in A,B); gray arrow indicates disruption in data collection. (D) Daytime sleep is increased following time-restricted feeding in R23E10>Dop1R1 compared to both parental controls. A Genotype (2) x Time (13) ANOVA revealed a Genotype X Time interaction: ANOVA F[,] = 7.87, p = 9.99E-16; *p < 0.05, modified Bonferroni test. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Starvation alters recovery sleep.

(A) During starvation, sleep in R23E10>Dop1R1, R55B01>Dop1R1 flies is not consistently above or below R23E10/+, Dop1R1/+ or R55B01/+ parental controls (n = 13–16 flies/group; ANOVA F[2,36] = 2.2, p = 0.12 and ANOVA F[2,38] = 15.4, p = 1.2E-05, *p < 0.05, modified Bonferroni test). (B) Daytime sleep is not increased in R23E10>Dop1R1 or R55B01>Dop1R1 flies compared to both parental controls R23E10/+, Dop1R1/+, or R55B01/+ on recovery day 1 (ANOVA F[2,36] = 4.8, p = 0.02 and ANOVA F[2,38] = 0.5, p = 0.58, *p < 0.05, modified Bonferroni test). (C) Sleep is increased in R23E10>Dop1R1 and R55B01>Dop1R1 flies compared to R23E10/+, Dop1R1/+, or R55B01/+ parental controls on recovery day 2 (ANOVA for Genotype F[2,36] = 14.25, p = 2.75E-05 and ANOVA for Genotype F[2,38] = 8.35, p = 0.0009 for R23E10 and R55B01, respectively. Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Expressing Dop1R2-RNAi in R23E10 neurons increases sleep.

(A) Sleep is increased in R23E10>Dop1R2, flies compared to parental controls (n = 16 flies/group; 3 (genotype) X 24 (hour) repeated measures ANOVA reveals a significant interaction F[,1012] = 7.15, p = 0.1.4E-13). Underlying data is in S1 Datasheet.

(TIF)

Click here for additional data file.

Data underlying Fig 1.

(XLSX)

Click here for additional data file.

Data underlying Fig 2.

(XLSX)

Click here for additional data file.

Data underlying Fig 3.

(XLSX)

Click here for additional data file.

Data underlying Fig 4.

(XLSX)

Click here for additional data file.

Data underlying Fig 5.

(XLSX)

Click here for additional data file.

Data underlying Fig 6.

(XLSX)

Click here for additional data file.

Data underlying Fig 7.

(XLSX)

Click here for additional data file.

Data underlying S1–S11 Figs.

(XLSM)

Click here for additional data file.

Supplementary Material and Method.

(DOCX)

Click here for additional data file.

3 Dec 2021

Dear Dr Shaw,

Thank you for submitting your manuscript entitled "Sleep promoting neurons remodel their response properties to calibrate sleep drive with environmental demands" for consideration as a Research Article by PLOS Biology. I apologize for our delay in sending you an initial decision - we have had a bit of a hectic week catching up after the Thanksgiving Holiday.

Your manuscript has now been evaluated by the PLOS Biology editorial staff as well as by an academic editor with relevant expertise and I am writing to let you know that we would like to send your submission out for external peer review.

However, before we can send your manuscript to reviewers, we need you to complete your submission by providing the metadata that is required for full assessment. To this end, please login to Editorial Manager where you will find the paper in the 'Submissions Needing Revisions' folder on your homepage. Please click 'Revise Submission' from the Action Links and complete all additional questions in the submission questionnaire.

Once your full submission is complete, your paper will undergo a series of checks in preparation for peer review. Once your manuscript has passed the checks it will be sent out for review. To provide the metadata for your submission, please Login to Editorial Manager (https://www.editorialmanager.com/pbiology) within two working days, i.e. by Dec 07 2021 11:59PM.

If your manuscript has been previously reviewed at another journal, PLOS Biology is willing to work with those reviews in order to avoid re-starting the process. Submission of the previous reviews is entirely optional and our ability to use them effectively will depend on the willingness of the previous journal to confirm the content of the reports and share the reviewer identities. Please note that we reserve the right to invite additional reviewers if we consider that additional/independent reviewers are needed, although we aim to avoid this as far as possible. In our experience, working with previous reviews does save time.

If you would like to send previous reviewer reports to us, please email me at lsmith@plos.org to let me know, including the name of the previous journal and the manuscript ID the study was given, as well as attaching a point-by-point response to reviewers that details how you have or plan to address the reviewers' concerns.

During the process of completing your manuscript submission, you will be invited to opt-in to posting your pre-review manuscript as a bioRxiv preprint. Visit http://journals.plos.org/plosbiology/s/preprints for full details. If you consent to posting your current manuscript as a preprint, please upload a single Preprint PDF.

Given the disruptions resulting from the ongoing COVID-19 pandemic, please expect some delays in the editorial process. We apologize in advance for any inconvenience caused and will do our best to minimize impact as far as possible.

Feel free to email us at plosbiology@plos.org if you have any queries relating to your submission.

Kind regards,

Lucas

Lucas Smith

Associate Editor

PLOS Biology

lsmith@plos.org

18 Jan 2022

Dear Paul,

I am writing on behalf of my colleague Lucas Smith, who is currently on paternity leave. Thank you for submitting your manuscript "Sleep promoting neurons remodel their response properties to calibrate sleep drive with environmental demands." for consideration as a Research Article at PLOS Biology. Your manuscript has been evaluated by the PLOS Biology editors, by an Academic Editor with relevant expertise, and by three independent reviewers. Please accept our apologies for the delay in communicating the decision below to you.

In light of the reviews (below), we will not be able to accept the current version of the manuscript, but we would welcome re-submission of a much-revised version that takes into account the reviewers' comments. We cannot make any decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript is also likely to be sent for further evaluation by the reviewers.

We expect to receive your revised manuscript within 3 months.

Please email us (plosbiology@plos.org) if you have any questions or concerns, or would like to request an extension. At this stage, your manuscript remains formally under active consideration at our journal; please notify us by email if you do not intend to submit a revision so that we may end consideration of the manuscript at PLOS Biology.

**IMPORTANT - SUBMITTING YOUR REVISION**

Your revisions should address the specific points made by each reviewer. They all make (numerous) different points, but nothing seems to require an enormous amount of new experimental work (and often just text changes) - except for the DopR1 RNAi experiment.

Please submit the following files along with your revised manuscript:

1. A 'Response to Reviewers' file - this should detail your responses to the editorial requests, present a point-by-point response to all of the reviewers' comments, and indicate the changes made to the manuscript.

*NOTE: In your point by point response to the reviewers, please provide the full context of each review. Do not selectively quote paragraphs or sentences to reply to. The entire set of reviewer comments should be present in full and each specific point should be responded to individually, point by point.

You should also cite any additional relevant literature that has been published since the original submission and mention any additional citations in your response.

2. In addition to a clean copy of the manuscript, please also upload a 'track-changes' version of your manuscript that specifies the edits made. This should be uploaded as a "Related" file type.

*Re-submission Checklist*

When you are ready to resubmit your revised manuscript, please refer to this re-submission checklist: https://plos.io/Biology_Checklist

To submit a revised version of your manuscript, please go to https://www.editorialmanager.com/pbiology/ and log in as an Author. Click the link labelled 'Submissions Needing Revision' where you will find your submission record.

Please make sure to read the following important policies and guidelines while preparing your revision:

*Published Peer Review*

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. Please see here for more details:

https://blogs.plos.org/plos/2019/05/plos-journals-now-open-for-published-peer-review/

*PLOS Data Policy*

Please note that as a condition of publication PLOS' data policy (http://journals.plos.org/plosbiology/s/data-availability) requires that you make available all data used to draw the conclusions arrived at in your manuscript. If you have not already done so, you must include any data used in your manuscript either in appropriate repositories, within the body of the manuscript, or as supporting information (N.B. this includes any numerical values that were used to generate graphs, histograms etc.). For an example see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5

*Blot and Gel Data Policy*

We require the original, uncropped and minimally adjusted images supporting all blot and gel results reported in an article's figures or Supporting Information files. We will require these files before a manuscript can be accepted so please prepare them now, if you have not already uploaded them. Please carefully read our guidelines for how to prepare and upload this data: https://journals.plos.org/plosbiology/s/figures#loc-blot-and-gel-reporting-requirements

*Protocols deposition*

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Thank you again for your submission to our journal. We hope that our editorial process has been constructive thus far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Gabriel Gasque on behalf of

Lucas Smith

Associate Editor

PLOS Biology

lsmith@plos.org

*****************************************************

REVIEWS:

Reviewer #1: In this manuscript, Dissel and colleagues dissect the response of sleep-promoting neurons in Drosophila to allatostatin and dopamine, in both normal conditions and under various circumstances that modulate sleep need. They find in particular that dopamine responses are modulated by various starvation paradigms, which also modulate sleep and propose this is due to the upregulation of a specific dopamine receptor that is not normally required for dopamine sensitivity in these neurons under baseline conditions.

The paper is somewhat of a hodge-podge of observation about dFB sleep regulating neurons, spanning allatostatin inputs, a novel sleep-promoting dFB neuron, dopamine inputs, starvation, long-term memory, neural plasticity of receptor expression. This leads to a complicated set of loosely-connected results. However, I am favorably inclined towards this paper, with several interesting observations worthy of publication. I have a few small comments and one bigger comment.

To my mind, the most interesting finding is the sculpting of dopamine receptor subtypes in the dFB neurons and how this alters the integration of sleep responses to starvation. However, the conclusion is based on RNAi behavioral experiments against Dop1R1, showing an impact on sleep in some but not other starvation paradigms (Figure 5+6), combined with cAMP imaging experiments during starvation with RNAi against a different dopamine receptor Dop1R2 (Figure 7), showing that this has no effect on the cAMP signal to dopamine under conditions where starvation-induced sleep is dopamine dependent. The data is consistent with their interpretation, but showing that RNAi against Dop1R1 abrogates the effect would be the most direct demonstration, and I am not clear why this wasn't included.

Other specific comments:

1) the abstract says, "the Drosophila sleep homeostat (dfB)", but this is both confusing and misleading. dFB stands for "dorsal fan-shaped body neurons", not "sleep homeostat" as implied somehow here. More critically, calling these neurons the sleep homeostat is a bit of an overstatement-- the best evidence is that these are an important output arm of the sleep homeostat, but the evidence is out whether the whole sleep homeostatic mechanism is embodied by these neurons.

2) The P2X2 data showing a connection between AstA-neurons and R23E10 neurons could still be an indirect, or even a long-range, connection, given the slow timescale of the reporter. Does the recently published fly EM map provide any evidence of direct connections?

3) Knocking down AstAR2 in R55B01 neurons increases waking activity compared to the control strains. This could suggest an alternative interpretation of a specific sleep effect-- i.e. hyperlocomotion, that should at least be acknowledged in the main text when disucssing Figure S6.

4) line "we first evaluated the response properties or R23E10 neurons) should read "of"

5) In the Discussion, comments are made about the data being in conflict with Ni and colleagues, without elaboration. This point should be made clearer-- what was the Ni result specfically, how is your result different?

6) Discussion: Galanin is mentioned as the homolog of AstA multiple times, when once is enough. It is also a little inaccurate to call it the "mammalian" homolog, when it is in all vertebrates. It also confuses the connection between AstA and Galanin a little bit-- by traditional homology, these peptides look nothing alike. The claim is based on some similarity between the Galanin-receptor and the Allatostatin receptor, but other GPCRs also are similar-- are those orthologs, too? it is fine to claim functional similarities of these inhibitory neuropeptides but let's make the terms of this observation more explicit

7) Discussion: "lesioning Galanin neurons in the preoptic area reduces sleep rebound following sleep loss in mice". The impact of galanin loss on sleep rebound was shown first by Reichert 2019 in the Zebra Fish

8) Conclusion: "Our data reveal that the co-release of AstA is wake-promoting". This is overstated-- the co-release is presumeably the hypothesized one with glutamate, but this is not clearly and directly demonstrated in any experiment in this paper.

9) Figure 3A-- the data appears identical to the screen data presented in Figure S4. This should be acknowledged as such, as otherwise it implies the screen result was independently confirmed in a biological replicate, when it does not appear to have been so. It should be done as a separate confirmatory result from the screen in S4.

10) Figure 4 lacks stats, which are buried in the Figure S7, but these are critical for the interpretation of the curves. Couldn't a repeated measures stat capture this in the same panel? Or perhaps the graphs of the supplement can be made into smaller insets, especially for the key findings?

11) Figure 7D, the 55801/+>Epac/+ bar is small and slightly negative, but this doesn't quite match the data shown in Figure 7C, which has a short dip followed by data that are mostly positive. This disconnect makes me concerned about how the time series are being normalized and summed/averaged to generate the bar graphs that form the basis of the data analysis, here and in other figures measuring the cAMP levels. Would a time-series repeated measures stat serve better here?

12) In general, the standard for depicting data in figures nowadays is to show all the datapoints, not just a bar graph with SEM. That allows for a fuller representation of the data, highlights the Ns and variance involved, etc. I would highly recommend and strongly prefer such plots for all the bar graph data in the manuscript. e.g. Kempf et al., 2019. or Grubbs et al 2020 for a sleep-paper model systems in PloS Biology example.

Reviewer #2: The authors examined the role of allatostatin-A (AstA) on sleep-promoting R23B10/R55B01 positive dorsal fan-shaped body (dFB) neurons of Drosophila and found AstA decreased intracellular cyclic AMP (cAMP) level (Fig 1). Since knocking down of either AstA-R1 or AstA-R2 increased sleep (Fig 2, Fig 3GD), they proposed AstA promotes wakefulness by inhibiting dFB neurons, although the activation of AstA positive neurons by AstA-GAL4>dTRP-A1 promoted sleep (Fig S3C). They also examined the changes in the cAMP responses of R23B10/R55B01 positive dFB neurons to AstA and dopamine (DA) under different conditions, which affect sleep behaviors (Fig 4). Various changes were observed and apparently the responses to AstA and DA were independently and context-dependently controlled.

The authors also found that knocking down of Dop1R1 in R23B10/R55B01 positive dFB neurons, which did not affect sleep amount under ordinary condition, increased sleep under restricted feeding protocol (Fig 5). Knocking down of DopR1 in dFB also changed the sleep behavior during the recovery period from starvation (Fig 6). They showed the responses of dFB neurons with Dop1R2 knockdown to DA were changed from starvation to recovery (Fig 7).

With all these data, the authors concluded that dFB neurons changed their state, i.e. responsiveness to AstA and DA, which may be the basis for the distinction of the different type of sleep needs.

I enjoyed reading the manuscript and found it interesting and informative, although there are some inconsistency and insufficiency.

Major questions.

1. The biggest changes in sleep need and arousal threshold occur between day and night. I think it is necessary to examine the changes of dFB neurons at different time of the day. At least, the time of the experiments should be clearly stated.

2. Were the experiments of Fig 4 and Fig 7 performed with TTX? Otherwise, it is possible that the difference in the intrinsic release of AstA and DA modified the response of dFB neurons to them. This possibility should be clearly stated.

3. In addition to above, although this might be technically difficult, the basic intracellular cAMP level of dFB neurons before the application of AstA and DA may be different according to the context in Fig 4. This difference affects the magnitude of AstA and DA responses differently, since the direction of the response is reverse, i.e. decrease and increase. This should be also discussed.

4. The interpretation of Fig 6 is too speculative. Since there is no change in DA response after starvation in control flies as shown in Fig 4h, the mechanism would be different from Dop1R1 induction. The authors should demonstrate actually induction of Dop1R1 by immunohistochemistry or other methods. Fig 7 is also pretty incomplete. Since, there is no change in the DA response of the control fly after starvation, the DA response change in the flies during recovery period and that of the Dop1R2 RNAi flies after starvation should be included. In addition, since the magnitude of the effect of time restricted feeding in Fig 5 is very big, it is better to include the live imaging data of Dop1R1 knockdown before and after TRF.

Minor points

1. Please state the sex of the flies used clearly, although the authors wrote there were no big sexual dimorphism.

2. Please include the more discussion about Fig 4. For example, I do not understand why the response to DA did not change after sleep deprivation in Fig 4d.

3. The abstract should include Ast-A and DA which is helpful to the authors.

4. The phrase "Rather, the dFB neurons themselves can determine their response to the activity from upstream circuits" in the abstract does not make sense to me. I think the authors showed that dFB neurons changes their responsiveness from the previous inputs to them from other neurons, but not they sense the sleep need for themselves.

Reviewer #3: In this manuscript, Dissel et al examine examine how the dFB and input neurons are regulated in response to a number of different stresses. The manuscript focuses on two neuromodulations, AstA and DA, that impact R23E10 neurons that innervate the dFB. A major strength of this manuscript is that it provides cellular-level resolution of how sleep neurons are regulating depending on experience. This aspect of the paper is likely to generate broad interest because conveys the need to understand mechanisms regulating sleep in a number of different conditions. A number of experimental concerns are described below, some of which can be addressed through written clarification). There is a heavy reliance on ex vivo preparations, which may not reflect the brain function in an intact animal. In addition, the manuscript is written in a descriptive manner, and streamlining the writing and figures may help to convey the conceptual advances and increase the impact of the manuscript.

1. I find the logic laid out in the abstract suggesting the dFB determine responses to be confusing. The way it is written, it seems to confer cognitive capacity on a set of neurons . The data suggests these neurons are integrators of information of other areas, impacting the behavioral output. In the third paragraph of the introduction the logic is laid out, but it is not clear what the alternative is to being impacted by upstream regulators.

2. Why does Epac1-camps label fewer cells than GFP? Are the eight a different subset? Is the difference due to sensitivity/different levels of R23E10? It would be useful to show these neurons.

3. In 1i, Asta1-RNAi nearly abolishes FRET, but both AstARs impact sleep in Figure 2. Please explain and/or include the effects of AstA-R2-RNAi on Epac1 FRET.

4. If data were acquired where ATP is washed out in 1L this is important to show.

5. Fig 2I does not fully test the hypothesis that AstA signals through the identified G-proteins because there are likely many receptors using G-proteins in those cells. I suggest toning down thisstatement. Also, it's not clear what the alternative mode of action would be for AstaRs since they are G-protein coupled.

6. There is a large reliance on ex vivo imaging through. It is important to highlight the limitations of this or validate with an in vivo preparation. In addition, given the role of AstA in circadian regulation, emphasizing the timing of experiments throughout the manuscript would be helpful.

7. A weakness is the focus on AstA and DA, but there are likely additional contributors to the regulation of R23E10 neurons.

8. The TRF protocol used in figure 5 lists the restriction from 8AM-5PM. This should be in ZT, but it is also unclear why this timeframe was chosen, since it does not match the light dark cycle or any obvious component of the sleep cycle. I also question the idea that TRF results in substantial metabolic challenge. An alternative way of interpreting TRF is that misalignment is a metabolic challenge, and the metabolic stress is reduced during TRF. The robustness of the effects in 5F-I is very compelling. It would be informative to have data on sleep during the TRF protocol.

9. This may be semantic, but I think it's conceptually important to lay out whether Dop1R1 'supports waking behavior during recovery from starvation' or inhibits homeostatic recovery.

Minor:

1. In the abstract, the (dFB) abbreviation is confusing because it comes after the sleep homeostat, but is not an abbreviation for the homeostat.

2. The abstract discusses dFB, but the introduction describes R23E10 neurons, and then the results describe R23E10 ad projecting onto the dFB. Clearly laying out the circuit in the abstract/into would be helpful.

3. I find this sentence awkward. It would seem proper to thank them in the acknowledgments rather than describing the request in the results section. 'To expedite the identification of such lines, we asked Dr. Gerry Rubin and Dr. Arnim Jenett, for assistance [49]'

4. Check to make sure the Epac transgene is consistently written throughout the manuscript text and figures

5. In discussion, suggest changing the phrase 'historical context' to something more specific.

4 May 2022

Submitted filename: Response to Reviewers_May 2.docx

Click here for additional data file.

31 May 2022

Dear Dr Shaw,

Thank you for your patience while we considered your revised manuscript "Sleep promoting neurons remodel their response properties to calibrate sleep drive with environmental demands" for publication as a Research Article at PLOS Biology. Your revised study has been evaluated by the PLOS Biology editors, the Academic Editor and the original reviewers.

The reviews are appended below. As you will see, both Reviewers 2 and 3 are satisfied by the revision, however Reviewer 1 has noted that one of the key conclusions is still not adequately supported. While Reviewer 1 does acknowledge that the conclusions are currently written with the appropriate caveats, after discussion with the Academic Editor, we think it would be important to strengthen this conclusion with new data to more tightly link behavior and neurophysiology for a key point in the study, especially as the reagents and methodology are available.

Although we usually would not invite a second round of experimental revision, in this case we would like to give you the opportunity to address Reviewer 1’s comments and provide the requested Dop1R1 RNAi experiment. As you address the remaining reviewer comments, we also ask that you address the following editorial requests:

1 - Data request - Thank you for providing the underlying data for your figures. I could not find a corresponding excel file for a couple of figures or panels within these excel files (for example, I did not see the underlying data for Figure S2C or Figure S9). Can you please ensure that the supplementary data files contain underlying data for all relevant figures?

Please also ensure that figure legends in your manuscript include information on where the underlying data can be found, and ensure your supplemental data files have legends.

For more information on the PLOS Data policy, please see http://journals.plos.org/plosbiology/s/data-availability and this editorial: http://dx.doi.org/10.1371/journal.pbio.1001797

2. Species in abstract? - Please note that per journal policy, the model system/species studied should be clearly stated in the abstract of your manuscript.

3. Data not shown - Please note that per journal policy, we do not allow the mention of "data not shown", "personal communication", "manuscript in preparation" or other references to data that is not publicly available or contained within this manuscript. Please either remove mention of these data or provide figures presenting the results and the data underlying the figure(s).

4. Blurb - Please provide a blurb which (if accepted) will be included in our weekly and monthly Electronic Table of Contents, sent out to readers of PLOS Biology, and may be used to promote your article in social media. The blurb should be about 30-40 words long and is subject to editorial changes. It should, without exaggeration, entice people to read your manuscript. It should not be redundant with the title and should not contain acronyms or abbreviations.

Given the extent of revision needed, we cannot make a decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript may also be sent for further evaluation by a subset of the reviewers.

We expect to receive your revised manuscript within 3 months. Please email us (plosbiology@plos.org) if you have any questions or concerns, or would like to request an extension.

At this stage, your manuscript remains formally under active consideration at our journal; please notify us by email if you do not intend to submit a revision so that we may withdraw it.

**IMPORTANT - SUBMITTING YOUR REVISION**

Your revisions should address the specific points made by each reviewer. Please submit the following files along with your revised manuscript:

1. A 'Response to Reviewers' file - this should detail your responses to the editorial requests, present a point-by-point response to all of the reviewers' comments, and indicate the changes made to the manuscript.

*NOTE: In your point-by-point response to the reviewers, please provide the full context of each review. Do not selectively quote paragraphs or sentences to reply to. The entire set of reviewer comments should be present in full and each specific point should be responded to individually, point by point.

You should also cite any additional relevant literature that has been published since the original submission and mention any additional citations in your response.

2. In addition to a clean copy of the manuscript, please also upload a 'track-changes' version of your manuscript that specifies the edits made. This should be uploaded as a "Revised Article with Changes Highlighted" file type.

*Re-submission Checklist*

When you are ready to resubmit your revised manuscript, please refer to this re-submission checklist: https://plos.io/Biology_Checklist

To submit a revised version of your manuscript, please go to https://www.editorialmanager.com/pbiology/ and log in as an Author. Click the link labelled 'Submissions Needing Revision' where you will find your submission record.

Please make sure to read the following important policies and guidelines while preparing your revision:

*Published Peer Review*

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. Please see here for more details:

https://blogs.plos.org/plos/2019/05/plos-journals-now-open-for-published-peer-review/

*Protocols deposition*

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Thank you again for your submission to our journal. We hope that our editorial process has been constructive thus far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Luke

Lucas Smith, Ph.D.

Associate Editor

PLOS Biology

lsmith@plos.org

------------------------------------

REVIEWS:

Reviewer #1: The authors have mostly addressed my and the other reviewers' comments, but I would like to clarify one point. In response to my suggestion that "showing that RNAi against Dop1R1 abrogates the effect would be the most direct demonstration" they responded:

"An increase in sleep is observed after both Time Restricted Feeding (Figure 5) and a night of starvation (Figure 6) when the Dop1R1 is knocked down in R23E10 and R55B01 neurons. That is, both starvation paradigms yield similar outcomes.

R23E10 neurons express Dop1R2 under baseline conditions. As a consequence, knocking down the Dop1R1 will not prevent R23E10 neurons from responding to Dopamine. It is only by knocking down the

Dop1R2 and demonstrating that R23E10 neurons no longer respond to Dopamine under baseline conditions that we can evaluate the role of Dop1R1 after starvation."

My point was that they show in Figure 7D that the dopamine response comes back in recovery, even when Dop1R2 is absent. To complete the logical circuit, though, they should show also now that Dop1R1 RNAi inhibits the dopamine response in this recovery phase. Otherwise, they cannot formally rule out the possibility that this dopamine signal isn't derived from another source. I note they do concede in the text that, "While we cannot exclude a role of other Dopamine receptors (e.g. Dopamine/Ecdysteroid receptor) for the observed changes, when viewed with the sleep experiments shown above (Figures 5 and 6), these data suggest that during recovery from starvation the constellation of Dopamine receptors in R23E10 and R55B01 changes and most likely includes the recruitment of Dop1R1", which is perhaps good enough. However, I believe Reviewer 2 is making a similar set of points in point 4, "it is better to include the live imaging data of Dop1R1 knockdown before and after TRF"-- highlighting the importance of this point to support some of the most interesting findings of this paper.

Reviewer #2: I think the authors replied to all of my comments.

Reviewer #3: The authors have addressed all my outstanding concerns, and the manuscript has been significantly strengthened.

29 Jul 2022

Submitted filename: Response_Shaw.docx

Click here for additional data file.

2 Aug 2022

Dear Paul,

Thank you for submitting revised manuscript "Sleep promoting neurons remodel their response properties to calibrate sleep drive with environmental demands" for publication as a Research Article at PLOS Biology. This revised version of your manuscript has been evaluated by the PLOS Biology editors and the Academic Editor.

We were persuaded by the arguments you made in your response to the request from Reviewer 1 and overall, we are largely satisfied by the changes made in this revision. However, there is one important lingering issue which will need to be addressed before we can editorially accept your manuscript for publication. In our last decision letter we noted several instances where conclusions were supported by the statement "data not shown", which is against journal policy. In the revised manuscript, it seems that the statement "data not shown" was removed, but the data necessary to support those conclusions was not added. After discussion with the Academic Editor, we think that it would be important for your to add figures/reference existing data, as appropriate, to support those claims.

For example, please provide additional figures to support the following conclusions:

1) “Similar results were obtained when expressing the sodium bacterial channel, UAS-NaChBac”

2) “The response properties of R23E10 neurons to Dopamine were similar in 6-8 day old flies and 30-38 day old flies. In contrast to age, we did not observe any changes in the response properties of R23E10 neurons in male and female flies despite large sexual dimorphisms in sleep behavior [52, 54]”.

3) “Furthermore, no changes in responses to Dopamine were found when training was followed by 4 h of sleep deprivation”

4) “R23E10>UAS-Dop1R2RNAi flies displayed an increase in sleep under baseline conditions consistent with previous reports”

**Important: as you add these additional figures, please make sure to update your S1_Data file to include the underlying data for any new graphs.

We expect to receive your revised manuscript within two weeks, but do let us know if you need more time to address this request.

To submit your revision, please go to https://www.editorialmanager.com/pbiology/ and log in as an Author. Click the link labelled 'Submissions Needing Revision' to find your submission record. Your revised submission must include the following:

- a cover letter that should detail your responses to any editorial requests, if applicable, and whether changes have been made to the reference list

- a Response to Reviewers file that provides a detailed response to the reviewers' comments (if applicable)

- a track-changes file indicating any changes that you have made to the manuscript.

NOTE: If Supporting Information files are included with your article, note that these are not copyedited and will be published as they are submitted. Please ensure that these files are legible and of high quality (at least 300 dpi) in an easily accessible file format. For this reason, please be aware that any references listed in an SI file will not be indexed. For more information, see our Supporting Information guidelines:

https://journals.plos.org/plosbiology/s/supporting-information

*Published Peer Review History*

Please note that you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. Please see here for more details:

https://blogs.plos.org/plos/2019/05/plos-journals-now-open-for-published-peer-review/

*Press*

Should you, your institution's press office or the journal office choose to press release your paper, please ensure you have opted out of Early Article Posting on the submission form. We ask that you notify us as soon as possible if you or your institution is planning to press release the article.

*Protocols deposition*

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please do not hesitate to contact me should you have any questions.

Sincerely,

Luke

Lucas Smith, Ph.D.

Associate Editor,

lsmith@plos.org,

PLOS Biology

10 Aug 2022

Submitted filename: Response_Shaw.docx

Click here for additional data file.

16 Aug 2022

Dear Paul,

Thank you for the submission of your revised Research Article "Sleep promoting neurons remodel their response properties to calibrate sleep drive with environmental demands" for publication in PLOS Biology, and for addressing our last editorial requests. On behalf of my colleagues and the Academic Editor, Richard Benton, I am pleased to say that we can in principle accept your manuscript for publication, provided you address any remaining formatting and reporting issues. These will be detailed in an email you should receive within 2-3 business days from our colleagues in the journal operations team; no action is required from you until then. Please note that we will not be able to formally accept your manuscript and schedule it for publication until you have completed any requested changes.

Please take a minute to log into Editorial Manager at http://www.editorialmanager.com/pbiology/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production process.

PRESS

We frequently collaborate with press offices. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximise its impact. If the press office is planning to promote your findings, we would be grateful if they could coordinate with biologypress@plos.org. If you have previously opted in to the early version process, we ask that you notify us immediately of any press plans so that we may opt out on your behalf.

We also ask that you take this opportunity to read our Embargo Policy regarding the discussion, promotion and media coverage of work that is yet to be published by PLOS. As your manuscript is not yet published, it is bound by the conditions of our Embargo Policy. Please be aware that this policy is in place both to ensure that any press coverage of your article is fully substantiated and to provide a direct link between such coverage and the published work. For full details of our Embargo Policy, please visit http://www.plos.org/about/media-inquiries/embargo-policy/.

Thank you again for choosing PLOS Biology for publication and supporting Open Access publishing. We look forward to publishing your study.

Sincerely,

Luke

Lucas Smith, Ph.D., Ph.D.

Associate Editor

PLOS Biology

lsmith@plos.org