Regulation of Sleep by Neuropeptide Y-Like System in Drosophila melanogaster

Sleep is important for maintenance of normal physiology in animals. In mammals, neuropeptide Y (NPY), a homolog of Drosophila neuropeptide F (NPF), is involved in sleep regulation, with different effects in human and rat. However, the function of NPF on sleep in Drosophila melanogaster has not yet been described. In this study, we investigated the effects of NPF and its receptor-neuropeptide F receptor (NPFR1) on Drosophila sleep. Male flies over-expressing NPF or NPFR1 exhibited increased sleep during the nighttime. Further analysis demonstrated that sleep episode duration during nighttime was greatly increased and sleep latency was significantly reduced, indicating that NPF and NPFR1 promote sleep quality, and their action on sleep is not because of an impact of the NPF signal system on development. Moreover, the homeostatic regulation of flies after sleep deprivation was disrupted by altered NPF signaling, since sleep deprivation decreased transcription of NPF in control flies, and there were less sleep loss during sleep deprivation and less sleep gain after sleep deprivation in flies overexpressing NPF and NPFR1 than in control flies, suggesting that NPF system auto-regulation plays an important role in sleep homeostasis. However, these effects did not occur in females, suggesting a sex-dependent regulatory function in sleep for NPF and NPFR1. NPF in D1 brain neurons showed male-specific expression, providing the cellular locus for male-specific regulation of sleep by NPF and NPFR1. This study brings a new understanding into sleep studies of a sexually dimorphic regulatory mode in female and male flies.


Introduction
Sleep, consisting of a period of sustained quiescence that is associated with an increased arousal threshold, is a common phenomenon that widely exists in animals from vertebrates to invertebrates [1]. Flies share with mammals a similar sleep regulatory mechanism that involves two relatively independent processes: the circadian system that is responsible for consolidating sleep during the night and the homeostatic system that is responsible for wakefulness and sleep drive. Longer waking periods lead to longer and more intense sleep periods [2]. Sleep regulation in Drosophila melanogaster is anatomically located in the mushroom body, known to be involved in learning and memory [3], while approximately 150 clock neurons in the central nervous system are involved in setting circadian rhythms. These clock neurons are divided into three lateral neuron groups -dorsolateral neurons (LN d s), PDF-positive ventrolateral neurons (LN V s) and the fifth small ventrolateral neuron (5 th small LN v ) -and three dorsal neuron groups -dorsal neurons 1, 2, and 3 (DN1, DN2, DN3). The LN v s in the circadian neuronal system contribute to sleep regulation by promoting wakefulness controlled by the GABA receptor and PER protein [4,5,6]. Dorsal neurons function in modulating sleep suppression during starvation [7].
Neuropeptide Y (NPY), a 36-amino-acid peptide from the pancreatic polypeptide (PP) family, is one of the more abundant peptides in the central and peripheral nervous system in mammals [8]. NPY is involved in several physiological functions, such as food intake, hormonal release, circadian rhythms, cardiovascular disease, thermoregulation, stress response and anxiety [9,10,11,12]. In humans, NPY enhances sleep by prolonging the sleep period and reducing sleep latency and wakefulness [13]. In contrast, NPY in rats increases wakefulness and decreases sleep [14]. These findings indicate that NPY regulates sleep differently in different animal species, probably due to their diurnal or nocturnal difference. NPY acts through its trans-membrane receptor (a G-protein-coupled receptor) and through Gi/o signaling pathways (mediated by α or βγ subunits) to inhibit cAMP formation and intracellular Ca 2+ mobilization, and to modulate Ca 2+ and K + channels [15].
In invertebrates, neuropeptide F (NPF) and its receptor (NPFR1 is the only identified functional receptor for NPF in D. melanogaster) show conserved structure and function with NPY and NPY receptors in mammals [16]. NPF in D. melanogaster has been characterized for involvement in diverse behavioral responses, such as food intake, hypermobility, cooperative burrowing, alcohol sensitivity and locomotor rhythm [17,18,19,20]. However, regulation of sleep by NPF in invertebrates has not yet been characterized. In this study, we explored NPF function in sleep by over-expressing npf and npfr1 in brain-specific neurons.

Sleep analysis and statistics
Flies were placed in 65 mm × 5 mm glass tubes containing standard fly food at one end. They were acclimated in behavior tubes for at least 24 hr at 25°C in LD conditions, and then data were collected in LD for 4 days with the DAM System (Trikinetics, Waltham, MA) in 1-min bins. Sleep parameters were measured with Pysolo software (obtained from the website: http://www.pysolo.net) using averages over 4 days of LD [23]. Sleep deprivation was performed using a mechanical method to keep flies awake for 12 hours during the night after 3 days of baseline measurement [24]. Statistical analysis was performed in a SPSS program (SPSS, Chicago, IL, USA). Significance levels were determined by independent t-test, and * indicates p < 0.05, ** indicates p < 0.01, and *** indicates p < 0.001.
Mean sleep episode duration and sleep bout number are important for analysis of sleep quality [25]. Sleep latency -the time for flies to fall asleep once the lights are turned off or on during LD cycles -reflects the sleep pressure, which influences sleep-initiation neuronal firing [26]. Here, sleep latency was measured and analyzed from the time of lights off or lights on to the onset of the first sleep episode as previously reported [26].

Immunohistochemistry
NPF-expressing cells were identified through immunofluorescence methods in female and male brains as described previously [20]. The brain samples were viewed with Nikon ECLIPSE TE2000-E and Nikon D-ECLIPSE (Nikon, Tokyo, Japan) confocal microscopes. Confocal images were obtained at an optical section thickness of 1-2 µm, and NPF immunofluorescence was quantified by ImageJ (http:// rsb.info.nih.gov/ij/index.html) as previously described [20].

Quantitative real-time PCR for measurements of npf expression level
Npf expression level was analyzed as described in a previous report [20]. Npf amount was determined by the mean of three independent replicates. Significant differences were determined as mentioned above.

The efficiency of Gal4-UAS transgenic system for npf and npfr1 expression
The npf-gal4, UAS-npf, npfr1-gal4, UAS-npfr1, UAS-npf dsRNA i and UAS-NPFR dsRNAi lines have been widely used for regulation of npf and npfr1 expression [18,27]. Our experiments further demonstrated a one and a half -two-fold increase at the transcript level in npf and npfr1 over-expression lines, and a 50% decrease in npf and npfr1 knockdown (dsRNAs) lines ( Figure S1).

Sleep regulation by NPF and NPFR1 in D. melanogaster
To investigate the role of NPF and NPFR1 in sleep regulation of D. melanogaster, we first used the Gal4-UAS system to increase or decrease npf and npfr1 expression in NPF-and NPFR1-expressing cells in the brain, and detected the sleep amount both in females and males. Under 12 hr L: 12 hr D conditions (LD), male flies with npf over-expression slept more, with a significant increase in whole nighttime sleep (mean±SEM: 392±22 min for npf over-expression line and 221±18 and 292±23 min for UAS and Gal4 controls) (p < 0.001) but not daytime sleep (Figure 1), and males with overexpressed NPFR1 also increased sleep both during daytime (mean±SEM: 350±24 min for npfr1 over-expression line and 241±11 and 243±13 min for UAS and Gal4 control lines) (p < 0.05) and nighttime (mean±SEM: 506±19 min for npfr1 overexpression line and 311±15 and 333±16 min for UAS and Gal4 control lines) (p < 0.001) ( Figure 2). Furthermore, the sleep increase at nighttime was due to a significant increase in sleep episode duration in the males of overexpressing npf (mean ±SEM: 26.56±4. 16 Figure S2) and in females with both up-and down-regulated expression ( Figure S3) was not affected in comparison with sleep in the controls. These data indicate that the NPF neuronal system regulates sleep in a sex-dependent manner.
To exclude any impact of increased NPF signaling on embryonic and larval development, npfr1 was over-expressed only in adults. Larvae and newly emerged adults were reared at 18°C to express an active Gal80 protein that represses the activity of Gal4. Newly emerged flies were entrained for three LD cycles at 18 o C and then transferred to 30°C to block Gal80, in order to activate Gal4-driven gene expression. Results showed that over-expression of npfr1 only in adults can increase total sleep time, caused by a significant increase in sleep during the night ( Figure 4A & B). Further analysis showed that the increase in nighttime sleep was mainly derived from an increased sleep episode duration ( Figure 4C), consistent with the results above.

Influence of NPF and NPFR1 on sleep homeostasis
Homeostatic regulation is a key feature of sleep [28]. To investigate whether NPF and NPFR1 have a function in sleep homeostasis, we employed sleep deprivation experiments using a mechanical sleep deprivation method as previously mentioned [24]. One of the key features for sleep homeostatic regulation is that an obvious sleep rebound occurs after sleep deprivation ( Figure 5A). We measured the NPF transcript in flies both with and without sleep deprivation, and found that npf expression in the heads of adult flies was down-regulated after sleep deprivation for 12h ( Figure 5B

Npf levels in male and female flies during nighttime
Previous studies showed that the nocturnal phase of Drosophila courtship behavior is dependent on the male and not the female circadian clock [29], indicating there exists a sex-specific regulation pathway to balance sleep and mating behavior in male flies. Might NPF signaling in males participate in this pathway? Is NPF signaling decreased to suppress sleep for mating behavior in males when females are present or increased to promote sleep in males when females are not present? To explore these questions, we analyzed the npf expression levels in wild type males and females flies during nighttime with the presence of opposite sex or same sex couples. Results showed that the npf expression levels in female flies were not changed whether male flies were present or not ( Figure 6). In contrast, when females were present, male flies expressed a lower npf level at midnight (ZT18) that could promote increased activity for courtship behavior, and when females were not present males expressed a higher level of npf that could promote sleep ( Figure 6).

The neuronal localization of NPF
We found in this study that sleep regulation by NPF in flies is sex-dependent. What are the structures that mediate sexdependent regulation of sleep? Previous reports showed that NPF was expressed in two groups of dorsal neurons (D1 and D2), two groups of dorsomedial neurons (P1 and P2), two groups of dorsolateral neurons (L1-l and L1-s), one lateroventral group (L2), one subesophageal group (L2 and S) and the fan-shaped body (FB) [20,27,30]. Interestingly, NPF expression was male-specific in 2-4 D1 noncircadian neurons ( Figure 7A & B). Furthermore, daily expression of NPF in D1 neurons exhibited two daily peaks -one at ZT8 and another at ZT20 ( Figure 7C). The peak at ZT20 is consistent with the npf mRNA peak at ZT18 (Figure 6) in males like these in Figure  7C, which are isolated from females at this time.

Discussion
In this study, we analyzed the function of NPF and NPFR1 on the sleep phenotype in Drosophila and found that NPF and NPFR1 both promote sleep during the night in male flies, in a similar fashion to NPY's effect in humans. Either NPF or NPFR1 over-expression causes an increase in nighttime sleep by increasing average sleep episode duration and reducing sleep latency during the night. NPF and NPFR1 also play an important role in the homeostatic regulation of sleep. NPF and NPFR1 may regulate sleep in Drosophila because NPFR1 activated by NPF inhibits cAMP formation [15]. The activity of cAMP-dependent PKA has been reported to be involved in sleep regulation in Drosophila [3,31].
Sexually dimorphic behaviors are widely observed in invertebrate and vertebrate species. Locomotor activity in Drosophila was found to show sexual dimorphism [32,33]. However, sexually dimorphic regulation of sleep has rarely been reported. Sleep in the cyc 01 mutant fly is sexually dimorphic with reduced or absent locomotion in males and exaggerated locomotion in compensatory rebound after sleep deprivation in females [34]. Fujii et al. have found a male-specific function in courtship, indicating a male-specific signal circuit might function in this sexually dimorphic behavior [30]. The NPF signal system is shown to participate in the circuit in this paper. In Drosophila, sexual differentiation is ultimately dependent on a chromosomal signal that is different between males (XY) and females (XX), resulting in repression or activation of the key gene Sexlethal (Sxl), responsible for dosage compensation, somatic sexual development, oogenesis and sexually dimorphic neural development. The subordinate gene transformer (tra) is one of the downstream regulatory genes of Sxl that controls sex determination [35]. The expression of npf in Drosophila is regulated in both sexnonspecific and male-specific (ms) manners, and the ms-npf expression is controlled by the tra-dependent sexdetermination pathway [27], indicating NPF is regulated by the sexual differentiation pathway. The male-specific NPF brain neurons (D1) may provide the neuronal locus for male-specific regulation of sleep.
The circadian system is involved in fly sleep regulation, and the circadian pacemaker neurons participate in regulating sleep. Some NPF in fly brains is found in clock neurons including the LN v s, LN d s and DN1s, among which the LN v s  have been shown to promote wakefulness [4,5,6], while the dorsal circadian neurons modulate sleep suppression during starvation [7]. We found that NPF and NPFR1 promote sleep quality especially during the night by prolonging average sleep episode duration and reducing night sleep latency in males. While these effects are associated with D1-specific expression in males, it is possible that the clock-neuron specific expression in males also contributes to these effects, which are absent in females because the neural circuit is incomplete (i.e., clockneuron specific expression is present but D1 expression is absent.).
In humans, sleep is restorative and plays a crucial role in long-term memory storage and learning. Sleep disorders cause a series of health problems associated with cognitive deficiency, poor job performance and low productivity [36]. In Drosophila, aging and diet are associated with sleep change [37,38]. Starvation suppresses sleep and sleep deprivation is known to affect longevity [7,39]. Therefore, it is likely that the function of NPF involves regulation of interactions among feeding, sleep, copulation and even longevity. increase before lights on (n=32 for controls, and 34 for yw; npf-Gal4/UAS-npf). (B) The total sleep in npf-down-regulated females is not changed either (n=32 for controls, and 36 for yw; npf-Gal4/UAS-npf dsRNA ). (C) Total sleep in females with npfr1 over-expression is similar to that in control flies (n=54 for controls and 49 for yw; npfr1-Gal4/UAS-npfr1). (D) The total sleep in npfr1-down-regulated females is not changed (n=59 for controls, and 58 for yw; npfr1-Gal4/UAS-npfr1 dsRNA ). Linear graphs, error bars, white and black bars, and white and gray background are as indicated in Figure 1. (TIF)