Overexpression of α-synuclein in Pigment Dispersing Factor neurons alters sleep–wake pattern by regulating lipid metabolism in Drosophila

Abstract

Parkinson’s disease (PD) is a complex neurodegenerative disorder, characterized by the aggregation of α-synuclein (α-syn). Current research increasingly indicates the prevalence of sleep–wake disorders in early-stage PD, although the underlying pathogenic mechanisms remain unclear. In this study, transgenic Drosophila models were utilized to observe excessive daytime sleepiness and impaired anticipation in flies overexpressing α-syn in pan-neurons and circadian clock neurons. Additionally, deficits in projection of Pigment Dispersing Factor (PDF) neuron terminals, which are involved in Drosophila sleep and circadian rhythm, were identified. An imbalance in lipid metabolism homeostasis was detected in the brains of α-syn overexpressing mutants. Ultimately, the inhibition of Sterol Regulatory Element-Binding Protein (SREBP) activity led to an improvement in the reduced daytime sleep duration phenotype. Our results suggest that lipid pathways play a role in sleep–wake disorders triggered by α-syn mutation and aggregation, thereby providing valuable insights into potential therapeutic avenues for disrupted sleep patterns associated with PD.

Graphical Abstract

Open in new tabDownload slide

The mutation and aggregation of α-synuclein (α-syn) are pivotal in synucleinopathies, a group of neurodegenerative diseases including Parkinson’s disease (PD) [1]. Encoded by the SNCA gene, α-syn undergoes structural alterations and misfolding in pathology, leading to its aggregation into inclusions that disrupt neurons and drive PD symptoms [2]. Sleep–wake disorders represent a non-motor symptom of PD, often preceding motor impairments and worsening as the disease progresses [3, 4]. PD-related sleep–wake disorders are primarily categorized as rapid eye movement sleep behavior disorder (RBD), excessive daytime sleepiness (EDS), insomnia, restless legs syndrome (RLS), obstructive sleep apnea (OSA), and circadian rhythm disorders [5]. Assessing early signs such as sleep disorders and α-syn levels in the brain contributes to understanding the core pathogenesis of PD, thus enabling the development of effective treatment methods.

Genome-wide association studies (GWAS) have identified SNCA, which encodes α-syn, as a major risk gene for sleep disorders [6]. Importantly, missense and point mutations of α-syn, such as the A30P, A53T, E46K, A53V, and V15A mutations, can aggravate the misfolding of α-syn and non-motor symptoms [7–11]. Clinical reports have indicated that the vast majority of PD patients carrying the SNCA^A53T mutation experience at least one type of sleep disorder [12]. A recent study revealed that SNCA^A53T mice, as young as 4 months old, exhibited symptoms of RBD without apparent motor impairments [13]. Recent research has illuminated the role of lipids in regulating sleep disorders [14]. Indeed, α-syn is a lipid-binding protein that is highly expressed in the human brain [15]. Overexpression of α-syn modifies the composition of phospholipids and fatty acids in neurons and impedes the formation of mitochondria-associated membranes (MAMs), thereby disrupting lipid metabolism [16]. Dysregulation of lipid metabolism also disrupts membrane fluidity and substance transport and promotes the aggregation of α-syn [17–19]. Moreover, many risk genes for PD are involved in lipid metabolism regulation, and clinical studies suggest that lipid-modulating drugs, such as statins, may have protective effects against PD [20]. For basic sleep research, studies using Drosophila melanogaster models have shown that PD risk genes, PTEN-induced putative kinase 1 (PINK1) and Glucocerebrosidase (GBA), alter the levels of phosphatidylserine and sphingolipids in neurons, leading to sleep disorders [21, 22]. These studies further support the notion that the use of PD fly models facilitates the understanding of the role of lipids in PD-related sleep disorders.

Similar to mammals, the sleep–wake pattern in Drosophila is primarily regulated by the circadian rhythm system and the sleep homeostasis system [23]. The Drosophila brain has approximately 150 circadian neurons, which transport various neuropeptides and exhibit different patterns of neuronal projections [24]. For instance, the small ventral lateral neurons (sLNvs) primarily induce daytime activity by releasing Pigment Dispersing Factor (PDF), thereby modulating locomotor rhythms and the sleep–wake pattern [25, 26]. Studies indicate that Drosophila with α-syn overexpression in dopaminergic neurons exhibit significant sleep–wake disorders, such as those seen in SNCA^A30P and SNCA^A53T mutations [27–29]. However, the potential mechanism through which α-syn mutations affect sleep remains unclear at present. In this study, we observed EDS and anticipation impairment and evaluated the impact of overexpressing α-syn on PDF neuron synaptic projections. Furthermore, an imbalance in lipid metabolism was identified in the brains of SNCA mutant flies, characterized by alterations in SREBP and related lipid gene levels. Importantly, targeted inhibition of SREBP expression by Betulin was found to alleviate sleep–wake patterns induced by α-syn. Together, this study utilized α-syn transgenic mutant fly models to investigate the role of lipid metabolism homeostasis in sleep disorders related to PD.

Methods

Fly lines

Male flies, ~10 days old, were used for all experiments. Flies were maintained in a 12-hour light:12-hour dark cycle at 24°C with equal population densities. The following fly stocks were used in this study: UAS-h[WT]αSyn (RRID:BDSC_8146), UAS-h[A30P]αSyn (RRID:BDSC_8147), UAS-h[A53T]αSyn (RRID:BDSC_8148), Elav-Gal4 (RRID:BDSC_8760), UAS-SREBP^WT (RRID:BDSC_8236), and UAS-SREBP^c.del (RRID:BDSC_8244) were obtained from the Bloomington Drosophila Stock Center (Bloomington, IN). The UAS-h[A53V]αSyn and UAS-h[E46K]αSyn were from the laboratory of Dr. Hongrui Meng (Institute of Neuroscience, Soochow University, China). The Timeless-Gal4, Pdf-Gal4, and Pdf-Gal4;UAS-CD8::GFP were from the laboratory of Dr. Yong Zhang (Cam-Su Genomic Resource Center, China) [30].

Sleep–wake behavioral assays

Locomotor activity and sleep behavior were recorded from individual males using the Drosophila Activity Monitoring (DAM) system (TriKinetics, Waltham, MA), as previously described [31]. The DAM monitor contains 32 channels, each connected to a small glass tube containing an individual fly, in which the movement of each fly can be monitored as they break the infrared beam that bisects the tube. Activity records were collected in 1-minute bins and analyzed using MATLAB. Sleep in Drosophila is defined as a bout of 5 or more minutes of inactivity [32]. The sleep parameters were calculated as follows: the amount of sleep, sleep duration, sleep bout number, and latency; morning anticipation, the single fly activity counts obtained in six 30-minute bins between Zeitgeber time (ZT) 22.5 and ZT24 (2 hours before lights on) divided by twelve 30-minute bins between ZT17.5 and ZT24 (6 hours before lights on), and evening anticipation was calculated with the same formula but taking as reference the lights off event [33]; and mean activity counts per minute while moving results from the activity when flies were awake (the number of recorded movements divided by the total time); individual males were recorded (16 flies per genotype) using the DAM system, and three independent experiments were performed for each genotype. For circadian experiments, flies were habituated for 5 days in a 12-hour light:12-hour dark condition before the light were turned off and activity was recorded for 5 days. Quantification of the circadian period, the power of the circadian rhythm, and the percentage of rhythmic flies are presented [31]. With data extracted using DAM fileScan113X, the sleep–wake patterns were analyzed using MATLAB, and circadian behavior was analyzed using FaasX [31].

Climbing assay

Twenty 10-day-old flies of each genotype were loaded into an empty plastic vial. After 5 minutes of acclimatization, each trial involved tapping the vial, then placing the vial down and allowing the flies to climb up. The number of flies that climbed and crossed the 12-cm mark in 15 seconds was recorded for each vial. The assay was repeated four times, and the data were expressed as an average of four trials per replicate. Similar steps were followed for 40-day-old flies.

Immunofluorescence and confocal microscopy

Brains at ZT2 and ZT14 from 10-day-old adult flies fixed by immersion in ice-cold 4% paraformaldehyde in phosphate-buffered saline (PBS) at room temperature for 40–50 minutes, dissected in chilled PBS (pH = 7.4), and then rinsed three times in PBS with 0.1% Triton X-100 (PBST) for 10 minutes each. Brains were first incubated overnight at 4°C with primary antibody after blocking for 90 minutes in PBST with 10% goat serum. They were then washed with PBST six times for 20 minutes each. Then the corresponding secondary antibody with AlexaFluor 555 or 647 (1:500, Invitrogen, CA, USA) was applied overnight at 4°C, and then washed with PBST six times for 20 minutes each at room temperature. The primary antibodies used were α-syn (AB138501, 1:500, Abcam, MA, USA), PDF (AB760350, 1:500, DSHB, IA, USA), and TH-488 (MAB319-AF488, 1:500, Millipore, MA, USA). Finally, the brains were mounted in Vectashield on a microscope slide, and images were acquired using a confocal microscope (ZEISS LSM 900, Germany). The sLNvs neuron clusters were analyzed as previously described [34]. The stacks of ventral sLNvs terminals were acquired using a 20× immersion lens with 5× digital zoom. Confocal images were exported to FIJI and analyzed using Simple Neurite Tracer Sholl Analysis.

Quantitative real-time PCR

The heads from 10-day-old male flies at a specific time were sampled and ground with Trizol Reagent (Ambion, TX, USA) according to the manufacturer’s protocol. These were then treated with chloroform to remove protein impurities, isopropyl alcohol to precipitate nucleic acids, 75% ethanol to wash, and Rnase-free water to dissolve the precipitate. The concentration of the nucleic acid mixture was measured, genomic DNA was removed, and the mRNA was reverse transcribed using RevertAid RT (Thermo Fisher Scientific, MA, USA). The quantitative real-time PCR assay was performed using an Applied Biosystems 7500 Real-Time PCR System (Thermo Fisher Scientific, MA, USA) and Taq Pro Universal SYBR qPCR Master Mix (Vazyme, China). Relative mRNA levels were normalized to endogenous rp49 expression, with assays conducted in triplicate, and analyzed using the 2-ΔCt method. The primers used are listed in Supplementary Table S1.

Western blotting

After treatment, total protein extracts were prepared from fly heads and homogenized in RIPA lysis buffer (150 mmol/L NaCl, 25 mmol/L Tris-HCl, pH 7.6, 1% sodium deoxycholate, 1% NP-40, and protease inhibitor from Roche). Proteins were isolated by SDS-PAGE (7.5%–12.5% gel) and transferred to a PVDF membrane (Millipore). The PVDF membrane was blocked with 5% milk powder and incubated for 1 hour at room temperature. The blots were incubated with the primary antibody at 4°C overnight. The primary antibodies used were SREBP (66875-1-Ig, 1:1000, Proteintech, China) and β-Tubulin (AB830032, 1:8000, Absin, China), which were incubated with the corresponding secondary antibody for 1 hour. The expected molecular weights of precursor SREBP (SREBP-p) and active cleaved SREBP (SREBP-c) are 130 and 70 kDa, respectively [35]. The proteins were visualized using ECL detection kits (Vazyme). The signal intensity of the bands was quantified using FIJI.

Drosophila Betulin administration

Betulin (HY-N0083, MCE, NJ, USA) was dissolved in DMSO. The solution was added to the Instant Drosophila Medium (FUNGENE, China) in water to achieve a final concentration of 1 mM. Male flies 5-day-old were placed in plastic vials containing food mixed with either the medicine or vehicle and fed for 48 hours [36]. Vehicle treatment consisted of adding solvent to the Instant Drosophila Medium.

Lipidomics and lipid analyses

Using a chilled centrifuge, we collected the heads of Timeless-Gal4, Tim>UAS-SNCA^A30P, and Tim>UAS-SNCA^A53T flies, totaling 100 mg of sample per group (approximately 1000 fly heads). Methanol (0.75 mL) was added to a 100 μL sample, which was placed into a glass tube with a Teflon-lined cap, and the tube was vortexed. MTBE (2.5 mL) was added, and the mixture was incubated for 1 hour at room temperature in a shaker. Phase separation was induced by adding 0.625 mL of MS-grade water. Upon 10 minutes of incubation at room temperature, the sample was centrifuged at 1000g for 10 minutes. The upper (organic) phase was collected, and the lower phase was re-extracted with 1 mL of the solvent mixture (MTBE/methanol/water [10:3:2.5, v/v/v]), and collecting the upper phase. Combined organic phases were dried and dissolved in 100 μL of isopropanol for storage, and then analyzed by LC-MS/MS. UHPLC-MS/MS analyses were performed using a Vanquish UHPLC system (Thermo Fisher, Germany) coupled with an Orbitrap Q Exactive™ HF mass spectrometer (Thermo Fisher, Germany) at Novogene Co., Ltd. (Beijing, China). Samples were injected onto a Thermo Accucore C30 column (150 × 2.1 mm, 2.6 μm) using a 20-minute linear gradient at a flow rate of 0.35 mL/min. The column temperature was set at 40°C. Mobile phase buffer A was acetonitrile/water (6/4) with 10 mM ammonium acetate and 0.1% formic acid, whereas buffer B was acetonitrile/isopropanol (1/9) with 10 mM ammonium acetate and 0.1% formic acid. The solvent gradient was set as follows: 30% B, initial; 30% B, 2 minutes; 43% B, 5 minutes; 55% B, 5.1 minutes; 70% B, 11 minutes; 99% B, 16 minutes; and 30% B, 18.1 minutes. Q ExactiveTM HF mass spectrometer was operated in positive [negative] polarity mode with sheath gas: 40 psi, sweep gas: 0 L/min, auxiliary gas rate: 10 L/min [7 L/min], spray voltage: 3.5 kV, capillary temperature: 320°C, heater temperature: 350°C, S-Lens RF level: 50, scan range: 114–1700 m/z, automatic gain control target: 3e6, normalized collision energy: 22, 24, and 28 eV [22 eV, 24 eV, 28 eV], injection time: 100 ms, isolation window: 1 m/z, automatic gain control target (MS2): 2e5, and dynamic exclusion: 6 seconds. The raw data files generated by UHPLC-MS/MS were processed using the LipidSearch to perform peak alignment, peak picking, and quantitation for each metabolite. The main parameters were set as follows: actual mass tolerance: 5 ppm; retention time tolerance: 0.05 minute; and signal/noise ratio: 3. After that, peak intensities were normalized to the total spectral intensity. Finally, qualitative and quantitative information about the substances in different samples was obtained. We applied univariate analysis (t-test) to calculate the statistical significance (p-value). The metabolites with Variable Importance in Projection (VIP) >1 and p-value <.05 and fold change (FC) ≥2 or FC ≤0.5 were differential metabolites. In the selection of differentially metabolites between the two groups, specific criteria were considered (criteria: Q/false discovery rate [FDR] < 0.1). Volcano plots were used to filter metabolites of interest based on Log2 (FC) and −log₁₀ (p-value) of metabolites. The correlation between differential metabolites were analyzed by cor () in R language (method = pearson). Statistical significance of the correlation between differential metabolites was calculated by cor.mtest () in R language. P-value <.05 was considered as statistically significant, and correlation plots were plotted by corrplot package in R language. Pathway enrichment can identify the key biochemical metabolic and signal transduction pathways associated with differential metabolites:

Where N represents the total number of metabolites involved in the Kyoto Encyclopedia of Genes and Genomes (KEGG) metabolic pathway, n is the number of differential metabolites, y is the number of metabolites annotated to a specific KEGG pathway, and x is the number of differential metabolites enriched in that KEGG pathway. If the ratio x/n > y/N is satisfied, the pathway is considered to be KEGG enriched. Using the hypergeometric test, p-values for pathway enrichment were obtained, with a threshold of p-value <.05.

Quantification and statistical analysis

All analyses were conducted using GraphPad Prism 10 (GraphPad Software, La Jolla, CA). Numerical data were analyzed using one-way analysis of variance (ANOVA), two-way ANOVA, or repeated measures ANOVA as appropriate. Results are presented as mean ± SEM. P <.05 was considered significant for all statistical tests. In experiments with multiple control groups, significance was determined by confirming that experimental lines differed significantly from all control lines.

Results

Overexpressing α-syn mutants altered sleep–wake pattern

In the initial phase of our study, the DAM system was utilized to evaluate the characteristics of sleep–wake pattern in 10-day-old transgenic flies expressing wild-type or mutant α-syn (SNCA^{WT/A30P/A53T/A53V/E46K}) under a pan-neuronal driver, Elav-Gal4. Compared to the control groups, flies with SNCA^A30P and SNCA^A53T mutations exhibited increased daytime sleep duration (Supplementary Figure S1A). In addition, flies with SNCA^A30P mutations displayed reduced nighttime sleep (Supplementary Figure S1B). Given that alterations in daytime sleep duration are intricately linked to the precise regulation of circadian neurons [37], the aggregation was further refined to specific neuronal clusters in the brain through Gal4-mediated overexpression of α-syn in circadian neuronal clusters. Sleep patterns in 10-day-old flies SNCA^A30P or SNCA^A53T were examined under the Timeless-Gal4 driver, which is expressed in all circadian cells [30] (Figure 1A). In comparison to the control groups, SNCA^A30P or SNCA^A53T flies exhibited increased daytime sleep duration, particularly in the latter half of the day (Figure 1, B and C). Additionally, no changes were observed in the amount of nighttime sleep driven by circadian neurons (Supplementary Figure S2A). Further analysis revealed that the primary reason for increased daytime sleepiness was an extension in the average length of sleep episodes (Figure 1D). No differences were found in the number of sleep episodes or sleep latency (Supplementary Figure S2, B and C). Subsequent analysis of circadian rhythms and activity revealed no significant effect of α-syn on locomotor behavior (Supplementary Figure S2, D–F). However, a notable reduction in activity levels was observed before lights on or off (Figure 1E), suggesting a potential impairment in the ability to perceive light–dark transitions in SNCA^A30P or SNCA^A53T flies. Wild-type flies utilize the endogenous circadian clock to anticipate the onset of dusk or dawn in a regular light–dark cycle, a phenomenon known as morning or evening anticipation [31]. Typically, flies become more active within the 3 hours preceding these events. Statistical analysis indicated that SNCA^A30P or SNCA^A53T flies demonstrated a reduced ability to predict light transitions in a 12-hour light:12-hour dark cycle under the Timeless-Gal4 driver (Figure 1, F and G) and Elav-Gal4 driver (Supplementary Figure S1, C and D). Activity during wakefulness was assessed to exclude the influence of motor disturbance on sleep monitoring (Figure 1H).

Furthermore, α-syn was observed to propagate from LNvs neurons (cell bodies and dendritic regions) to tyrosine hydroxylase-positive cells (TH) in 10-day-old SNCA^A30P or SNCA^A53T flies (Supplementary Figure S2G). These results are consistent with the inherent propagative nature of α-syn. Daytime sleepiness in flies may be closely related to synaptic connections between the dopaminergic nervous system and circadian neurons [38]. Next, the persistence of daytime sleepiness was noted in old SNCA^A30P or SNCA^A53T flies (40 days old) as they aged (Supplementary Figure S3, A–F). Due to the decreased activity level of old SNCA flies in DAM detection and climbing assay, their increased sleep phenotype might be indistinguishable from the locomotor deficit (Supplementary Figure S3, G and H). Thus, these results further indicated that sleep disorders precede the onset of PD’s motor dysfunction.

Impact of α-syn on PDF neuronal projections in Drosophila

Considering the close association between morning anticipation and PDF circadian neurons, which play a pivotal role in regulating the behavioral activity rhythms [39], we refined the circadian neuronal clusters and investigated the sleep–wake patterns by SNCA^A30P or SNCA^A53T mutants under the Pdf-Gal4 driver (Figure 2A). Comparative analysis revealed that both total and average daytime sleep duration were significantly increased in SNCA mutant flies, primarily in the latter half of the day (Figure 2, C and D). Other sleep characteristics were similar to those of flies driven by Timeless-Gal4, with no significant differences observed (Supplementary Figure S3, A–C). Under the Pdf-Gal4 driver, anticipatory behaviors in SNCA^A30P or SNCA^A53T flies were impaired (Figure 2, E and F and Supplementary Figure S3D). We found no impact of activity changes on sleep monitoring in these flies (Supplementary Figure S3E).

To detect the effect of α-syn on PDF neurons, immunostaining on LNvs neurons using PDF antibody was performed. Firstly, based on anatomical analysis, no significant loss in the PDF neuron number was found (Supplementary Figure S5). Then, utilizing the previously described analysis method for sLNvs terminal projections [40], we examined alterations in the terminal projections of PDF neurons (Figure 2, G and H). The overexpression of α-syn in PDF neurons reduced the complexity of dendritic branches along the axonal projections. Compared to the control group, the decreased axonal crossings resulting from α-syn overexpression led to a more compact overall morphology, almost eliminating the remodeling of dorsal projections (Figure 2I). Significant morphological contraction of axons was also observed at ZT14. Hence, it was suggested that α-syn mutants affect the pattern of PDF neuronal projections.

Overexpressing α-syn mutants regulated lipid homeostasis and phospholipid metabolism

The involvement of lipid pathways in regulating sleep processes has been well demonstrated in prior studies [41, 42]. To elucidate the impact of SNCA mutants on sleep, lipidomics of overexpression in fly brains was investigated, identifying targets associated with remodeling PDF neuronal projections. Lipid composition of the brain was assessed using LC-MS/MS, revealing distinct alterations in multiple types of lipid metabolites in SNCA^A30P or SNCA^A53T mutants (Figure 3, A and B). In comparsion to the control, α-syn overexpression resulted in changes in phosphatidylcholine (PC), phosphatidylethanolamine (PE), cardiolipin (CL), and acylcarnitine (AcCa) (Figure 3, C and D and Supplementary Table S2). KEGG pathway analysis emphasized dysregulation in glycerophospholipid metabolism, arachidonic acid metabolism, glycine, serine, and threonine metabolism, alpha-linolenic acid metabolism, ubiquinone and other terpenoid-quinone biosynthesis, and oxidative phosphorylation pathways (Figure 3, E and F and Supplementary Table S3). Following Q/FDR value correction, the selected lipid metabolites exhibited significant variations, among which PE (20:0CHO/18:2) demonstrated notable alterations post-correction (Supplementary Table S3). An effect of α-syn on phospholipid metabolism was suggested by these results.

To obtain initial insights into transcriptional metabolic pathways affected by α-syn overexpression, we quantified mRNA expression of key metabolic genes in the whole brain of SNCA^A30P or SNCA^A53T flies (Figure 3G). We observed an upregulation in the transcription levels of sterol regulatory element-binding protein (SREBP), a regulatory factor involved in lipid synthesis. Next, the transcription levels of downstream lipid synthesis genes regulated by SREBP were further investigated. It was found that acetyl-CoA carboxylase (ACC), acetyl coenzyme A synthase (AcCoAS), and malic enzyme (Men) exhibited varying degrees of upregulation, while no significant changes were detected in the mRNA levels of fatty acid synthase (FASN1). The primary function of SREBP is to regulate the transcription of lipid genes by entering the cell nucleus following proteolytic cleavage from the endoplasmic reticulum (ER). The protein levels of precursor (SREBP-p) and active cleaved SREBP (SREBP-c) were found to be increased in the brains of SNCA^A30P or SNCA^A53T flies compared to the control group (Figure 3, H–J). Betulin, a small molecule that specifically inhibits SREBP maturation, reduces cholesterol and fatty acid biosynthesis [43, 44]. After 48 hours of administration, Betulin effectively decreased the expression of SREBP-p and SREBP-c in the brains of flies (Figure 3, H–J). Moreover, we identified a downregulation in the levels of regulatory factors associated with PE synthesis (Figure 3G), including phosphatidylethanolamine cytidylyltransferase (ECT) and phosphatidylserine decarboxylase (PSD1). These findings suggest that α-syn may impact both the phosphatidylserine decarboxylase pathway in MAMs and the Kennedy pathway to lipid synthesis.

SREBP activity regulates the sleep–wake pattern and PDF neuron terminal projections

A recent study proposed the involvement of the SREBP-Men signaling pathway in regulating nighttime sleep duration in flies [36]. Therefore, our focus was on investigating the impact of SREBP on sleep in SNCA^A30P or SNCA^A53T flies. Feeding the SREBP inhibitor Betulin resulted in a reduction in daytime sleep duration compared to the mutant group, primarily reflected in the shortened mean sleep length (Figure 4, A and B and Supplementary Figure S6, A and B). Significant restoration of evening anticipation was observed in the treated group, although morning anticipation impairment remained (Figure 4C and Supplementary Figure S6, C and D). Activity analysis indicated a possible relationship between increased daytime activity in the treated group and the restoration of light-off anticipation behavior (Supplementary Figure S6E). PDF neuron immunostaining revealed varying degrees of structural recovery in neuron terminal projections following Betulin treatment, regardless of whether PDF expression levels were high (ZT2) or low (ZT14) (Figure 4, D–F). The inhibition of SREBP potentially enhanced the functional role of arousal molecules. These results indicated that pharmacological inhibition of SREBP restored the sleep–wake patterns and PDF neuron terminal projections.

Additionally, we assessed the impact of increased SREBP activity on daytime sleep behavior in Drosophila. Neuronal overexpression of full-length SREBP (SREBP^WT) driven by Timeless-Gal4 replicated the daytime sleepiness observed with α-syn overexpression, primarily in the latter half of the day (Supplementary Figure S7, A–C). Similarly, the same phenomenon was observed under the Pdf-Gal4 driver. Constitutive overexpression of the active nuclear form of SREBP (SREBP^c.del) also increased daytime sleep in flies (Supplementary Figure S7, D–F). Significant evening anticipation dysfunction occurred only in flies with neuronal overexpression of full-length SREBP, whether driven by Timeless-Gal4 or Pdf-Gal4 (Supplementary Figure S7, G and H). These results suggested that alterations in SREBP-related pathways are closely related to the sleep–wake patterns in Drosophila.

Discussion

Our results demonstrate that Drosophila carrying the SNCA^A30P or SNCA^A53T mutation under the control of circadian neurons exhibit daytime sleepiness similar to that observed in PD patients (Figure 1, A–D). Simultaneously, there was a significant reduction in rhythm-related anticipation observed both in the morning and evening (Figure 1, E–H), consistent with previous findings obtained in flies overexpressing α-syn [27]. The activity levels of SNCA^A53T mice exhibit significant changes before lights off, similar to the phenomenon observed in flies in response to light changes [45]. Although the specific mechanisms involved differ, this suggests that α-syn impacts the sensitivity of model organisms to light recognition. Given recent findings indicating that EDS serves as a clinical marker for PD conversion [46], the SNCA mutant fly model can be used to investigate the potential mechanisms underlying the association between EDS and PD. Anatomical and pathological correlation analyses have shown that α-syn pathology exists in brain regions, such as the locus coeruleus and raphe nuclei, which are involved in sleep regulation in PD patients [47]. While these sleep phenotypes have been reported, the reasons for α-syn-induced alterations in sleep patterns remain unclear. Compared to the sleep phenotype induced by Timeless-Gal4, which drives the expression of non-rhythm-related cells such as glial cells, Pdf-GAL4 specifically drives the expression of rhythm-related cells, including sLNvs neurons (Figure 2, A–F). Daytime sleep in flies is intricately connected to the sLNvs neurons in the brain [37]. It was found that α-syn altered the projections of sLNvs neuronal terminals in the fly brain, which are crucial for the neuroplasticity and neurotransmission of sleep–wake patterns [48]. Given that the axonal terminals of PDF neurons exhibited thinner and fewer projections at ZT14 compared to the control group after α-syn overexpression (Figure 2, H and I), which goes beyond the typical 24-hour morphological oscillation, it is speculated that α-syn overexpression may have caused neurodegenerative impairment or permanent oversimplification. Although the plasticity of sLNvs terminals is noteworthy and shows significant variability, studies have demonstrated that the oversimplification of these terminals does not mechanistically lead to changes in general anticipation or circadian rhythms [49]. However, α-syn pathology is generally concentrated in specific regions of the nervous system. Unlike the loss of dopaminergic neurons, no significant damage or degeneration was observed in serotonergic cells overexpressing α-syn [50] (Supplementary Figure S2G). Recent studies have shown that TH neurons establish connections with PDF neurons and other subgroups of circadian neurons through G protein-coupled receptors (GPCRs) [38]. Aggregation of α-syn potentially disrupts the signaling between circadian neurons and non-circadian neurons. Meanwhile, α-syn affects organelle and neuronal function by altering the actin filament network, which is involved in neuronal axon growth and morphology maintenance [51, 52]. Additionally, our research shows that young SNCA flies exhibit no differences in cell body or circadian function (Supplementary Figure S5). However, different SNCA mutations in older flies lead to varying degrees of circadian rhythm shifts as they age [27]. Thus, further investigation is needed to explore the impact of α-syn mutations and aggregation on the structural plasticity of circadian neurons.

Lipid synthesis and breakdown also influence neuronal structural remodeling. While the impact of α-syn on lipids varies across different mutations, our results indicate abnormal lipid metabolism in the brains of SNCA^A30P and SNCA^A53T mutant flies, characterized primarily by changes in phospholipid levels (Figure 3, A–F). Feeding flies with modafinil (a wake-promoting drug) revealed an elevation in the relative ratio of PE to PC in the brain, highlighting the significance of these lipid classes in regulating sleep–wake patterns [53]. The overexpression of A53T α-syn in HEK cells subjected to fatty acid overload showed increased triacylglycerol and lipid droplet levels, disrupting neuronal lipid metabolism [54]. Currently, information on lipid changes in animal models of PD is limited. Significant differences in lysophosphatidylcholine (LPC) levels have been observed in the substantia nigra of 6-hydroxydopamine (6-OHDA) mice [55]. Interestingly, LPC is also a significant marker in the metabolomics of patients with daytime sleepiness, playing a role in inflammatory signaling [56]. α-syn has been shown to play a role in various lipid metabolic pathways, significantly impacting its toxicity [57, 58]. The dysregulation of arachidonic acid and other polyunsaturated fatty acid metabolic pathways suggests that α-syn overexpression affects neuronal membrane fluidity and permeability. Additionally, a downregulation in the expression of PSD1 and ECT genes involved in PE synthesis was observed (Figure 3G). Overexpression of α-syn mutants reduces the interaction of MAMs, hindering PE conversion and further disrupting vesicular transport and organelle function [59]. The downregulation of PE levels restricts the formation of LC3-PE autophagosomes, thereby affecting autophagic function [60]. Autophagic dysfunction has been observed in patients with EDS, accompanied by decreased levels of LC3 protein [61]. From a technical standpoint, future analyses of lipids may be conducted based on chain length, saturation level, or lipid class to investigate their distinct biological functions [62]. Hence, it is hypothesized that research on phospholipid metabolism could provide new insights into the treatment of sleep–wake disorders in PD.

SREBP is an important gene involved in fatty acid synthesis and is considered a genetic risk factor for sporadic PD [63]. Elevated transcriptional and protein levels of SREBP were found in the brains of SNCA mutant flies (Figure 4, G–J). Similarly, overexpression of SREBP driven by circadian neurons also resulted in daytime sleepiness in flies (Supplementary Figure S7), consistent with previous findings. A recent GWAS found a close association between SREBP and sleep disorders, such as EDS [64, 65]. However, it should be noted that unlike what was observed after α-syn overexpression in vitro, where there was a reduction in SREBP-1 activation and an increase in SREBP-2 activation, this discrepancy may be attributed to only one SREBP protein being present in flies [66]. Next, through targeted inhibition of SREBP by Betulin, the sleep–wake cycle and PDF neuronal terminals were partially restored (Figure 4, A–C), although the extent of neuronal recovery remained limited (Figure 4, D–F). By intervening in phospholipid and fatty acid biosynthesis, Betulin partially reversed the contraction of PDF neuron terminals. In PINK1 and Parkin mutant flies, feeding phosphatidylserine rescued vesicle transport within neurons and restored sleep [67]. Recent research has shown that SREBP-2 influences neuronal lipid homeostasis, especially in metabolically vulnerable neurons [68]. SREBP and the kinase involved in PE synthesis are involved in dendritic morphogenesis during neural development [69, 70]. These findings suggest that SREBP not only influences fatty acid synthesis but also maintains the integrity of neuronal structures by regulating phospholipids [71–73]. Additionally, the precursor protein of SREBP is primarily synthesized in the ER, and abnormal expression of ER kinases indirectly influences SREBP-associated lipid synthesis [74]. Current research indicates that the upregulation of protein kinase R (PKR)-like ER kinase (PERK) gene promotes the expression of PDF in Drosophila, highlighting the essential role of the ER in balancing sleep–wake patterns [75]. Exploring the functional roles of lipids and their associated regulatory factors may serve as one of the preventive measures for PD-related sleep disorders.

A limitation of this study is the Drosophila models. The current assessment relies solely on sleep monitoring systems to evaluate sleep architecture, which cannot analyze sleep phases akin to human sleep, including rapid eye movement (REM) and non-rapid eye movement (NREM) phases. Moreover, none of the manipulations used were induced in adults; α-syn overexpression was constitutive throughout development and might have influenced neuronal development. Therefore, the observed phenotypes could result from developmental differences. In fact, lipid metabolism is crucial for the development and maintenance of LNv neurons, which continually increase lipid intake during development to manage prolonged elevated input activity [76].

In summary, our study revealed that the overexpression of α-syn mutants in Drosophila impacts the projections of axonal terminals of PDF circadian neurons, disrupts lipid balance in the brain, and eventually leads to EDS and anticipation disorders in flies. Furthermore, we demonstrated that α-syn mutants influenced sleep–wake patterns through the SREBP lipid pathway, and that sleepiness was alleviated with an inhibitor of SREBP. Future investigations will delve deeper into elucidating the functional role of lipids in neurodegenerative disease progression, thereby contributing to the development of innovative therapies for PD.

Supplementary Material

Supplementary material is available at SLEEP online.

Acknowledgments

The authors would like to thank Hongrui Meng laboratories for helpful comments and Novogene company for the lipidomics and lipid analyses.

Funding

This work was supported by grants from the Suzhou Science, Education and Health Project (KJXW202387), the STI2030-Major Projects (2021ZD0203400), National Natural Science Foundation of China (32271206), Science and Technology Innovation Project of Xiongan New Area (2023XAGG0073), Jiangsu Provincial Medical Key Discipline (ZDXK202217), Suzhou Key Laboratory (SZS2023015), and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

Disclosure Statement

Financial disclosure: None. Nonfinancial disclosure: None.

Data Availability

All data are available upon request.

References