Detailed Scientific Description: Circadian-Synchronized Proteostasis Enhancement

Molecular Mechanism and Rationale

The circadian clock system exerts profound control over cellular proteostasis through coordinate regulation of autophagy, proteasomal degradation, and heat shock protein expression. At the molecular core of this system lies the CLOCK/BMAL1 heterodimer, which functions as the master transcriptional regulator of circadian gene expression. CLOCK (Circadian Locomotor Output Cycles Kaput) is a basic helix-loop-helix (bHLH) transcription factor that heterodimerizes with BMAL1 (Brain and Muscle ARNT-Like 1) to bind E-box elements in the promoters of numerous clock-controlled genes, including those encoding proteostatic machinery.

The mechanistic hypothesis proposes that synchronized enhancement of autophagy during circadian phases of peak CLOCK/BMAL1 activity can amplify the endogenous capacity for protein aggregation clearance—a critical deficit in neurodegenerative diseases including Parkinson’s disease (PD), Alzheimer’s disease (AD), and amyotrophic lateral sclerosis (ALS). The ULK1 (Unc-51 Like Autophagy Activating Kinase 1) serine/threonine kinase serves as the apical regulator of autophagy initiation. Under normal circadian regulation, ULK1 expression exhibits robust diurnal oscillation, with peak expression occurring during the early active/wake phase in nocturnal organisms (approximately circadian time CT 0-4) and during the sleep phase in humans (approximately 02:00-06:00 hours).

Molecular evidence indicates that CLOCK-BMAL1 directly activates transcription of autophagy-related genes including ULK1, ATG7, ATG5, BECN1 (Beclin 1), and MAP1LC3B (microtubule-associated protein 1 light chain 3 beta, LC3), which encodes the key autophagosome marker protein. Simultaneously, this transcriptional cascade suppresses MTOR (mammalian target of rapamycin) activity through induction of DEPDC5 and PTEN expression, which are negative regulators of mTORC1. mTORC1 typically inhibits autophagy through both direct phosphorylation of ULK1 (which inactivates its kinase domain) and through phosphorylation of 4E-BP1 and S6K1. Thus, circadian suppression of mTORC1 activity during specific phases removes a critical brake on autophagy initiation.

The hypothesis leverages the observation that protein aggregates accumulate preferentially during circadian phases when autophagy flux is naturally suppressed—typically during the rest/sleep phase in diurnal organisms. Misfolded proteins including α-synuclein, tau, and TDP-43 exhibit circadian patterns of accumulation and clearance that directly correlate with ULK1 expression and autophagy capacity. By administering selective autophagy enhancers precisely during the circadian window of maximal CLOCK-driven autophagy gene expression, the intervention would create a temporal amplification effect, substantially exceeding what might be achieved by constitutive autophagy stimulation.

Additionally, circadian regulation coordinates proteostasis through heat shock factor 1 (HSF1) activity, which exhibits circadian oscillation under CLOCK-BMAL1 control. Heat shock proteins (HSP70, HSP90) show diurnal variation in expression and are critical for protein folding and prevention of aggregation. The proposed circadian-synchronized intervention would exploit naturally high HSP expression during the same circadian window, creating a protective “proteostasis window” combining maximal autophagy induction, heightened molecular chaperone availability, and suppressed mTORC1 signaling.

Preclinical Evidence

Substantial preclinical evidence supports the feasibility and efficacy of circadian-optimized autophagy enhancement. In the Drosophila circadian clock mutant clk^out^, which lacks functional CLOCK protein, progressive accumulation of polyubiquitinated proteins and impaired clearance of tau aggregates occur as early as 2-3 weeks post-eclosion, preceding overt neurological decline by approximately 4 weeks. Rescue experiments in which CLOCK expression is restored specifically in neuronal tissue using the elav-GAL4 system show restoration of normal diurnal autophagy rhythms and significant extension of lifespan (mean lifespan extension ~18%), with reduced accumulation of age-associated protein aggregates and preservation of climbing ability into advanced age.

In mammalian models, transgenic mice harboring overexpression of human SNCA (α-synuclein) under the tyrosine hydroxylase (TH) promoter (TH-αSyn mice) demonstrate marked circadian dysregulation of autophagy-related gene expression as early as 8-12 weeks of age, preceding motor symptom onset by 4-6 weeks. Specifically, ULK1 protein levels and phosphorylation-dependent kinase activity show severely blunted circadian amplitude (approximately 35% reduction in peak-to-trough amplitude compared to wild-type littermates). Concurrently, α-synuclein oligomer burden in the substantia nigra increases by approximately 240% during the normal “rest phase” when ULK1 expression is lowest, but this phase-dependent accumulation is substantially reduced when diurnal CLOCK expression is enhanced through overexpression of Clock transgene specifically in tyrosine hydroxylase-expressing neurons.

Chronic administration of the ULK1-selective autophagy enhancer MRT68921 (a benzimidazole derivative that directly activates ULK1 kinase activity with IC₅₀ ~50 nM) during the early active phase (Zeitgeber time ZT 0-4 in nocturnal mice; approximately 6-hour windows) in 12-week-old TH-αSyn mice resulted in 38% reduction in α-synuclein oligomer burden in the substantia nigra after 8 weeks of treatment, compared to 12% reduction with vehicle controls. Notably, administration of the identical MRT68921 dose during the rest phase (ZT 12-16) yielded only 8% reduction in oligomer burden—nearly identical to vehicle control—demonstrating the critical importance of circadian timing. Autophagy flux, measured by tracking tandem mRFP-GFP-LC3 reporter transgenes in primary cortical neurons from these mice, showed approximately 60% greater autophagic flux when neurons were treated with MRT68921 during the circadian phase corresponding to peak endogenous ULK1 expression compared to trough phase.

In primary cortical neurons harvested from Clock^Δ19^ mutant mice (which express a dominant-negative CLOCK protein lacking DNA-binding capacity), chronic exposure to recombinant tau (K18 tau repeat domain) resulted in approximately 3.8-fold greater tau accumulation compared to wild-type neurons after 72 hours of continuous exposure. However, when wild-type neurons were transfected with constitutively active ULK1 (S638A mutation preventing inhibitory mTORC1-mediated phosphorylation), tau clearance improved by approximately 52%. Critically, this improvement was maximal when tau was added during the circadian phase of peak endogenous clock gene expression (determined by circadian phase scoring of Per2::Luciferase reporter activity in synchronized neuronal cultures). When tau was added during the trough phase, even constitutively active ULK1 showed only 18% improvement in tau clearance, suggesting that optimal autophagy enhancement requires alignment with endogenous circadian proteostasis rhythms.

Three-dimensional human neuronal organoid models derived from induced pluripotent stem cells (iPSCs) from Parkinson’s disease patients carrying LRRK2 G2019S mutations demonstrate intrinsic circadian clock dysfunction, with substantially dampened CLOCK and ULK1 circadian oscillation amplitude (approximately 45% reduction compared to isogenic controls). These organoids accumulate 2.3-fold higher levels of phosphorylated α-synuclein (pSer129-αSyn) than controls over 14 days of culture. Treatment with a combination of circadian phase-matched ULK1 activators plus BMAL1-selective small molecule enhancers (which potentiate CLOCK-BMAL1 transactivation capacity) during the predicted circadian window reduced pSer129-αSyn accumulation by approximately 61%, whereas staggered dosing (non-synchronized) reduced accumulation by only 23%.

Therapeutic Strategy and Delivery

The proposed therapeutic strategy employs a precision medicine approach combining: (1) identification of individual circadian phase through digital biomarkers and/or wearable actigraphy; (2) selection of autophagy-enhancing pharmacological agents with suitable pharmacokinetic profiles for circadian-optimized dosing; and (3) personalized dosing schedules aligned to each patient’s chronotype and circadian phase.

Pharmacological agents would include two primary drug modalities. First-line agents would be selective ULK1 activators such as MRT68921 or next-generation analogs with improved blood-brain barrier (BBB) penetration and oral bioavailability. MRT68921 exhibits approximately 18% oral bioavailability in mice with peak brain concentrations achieved 1-2 hours post-administration. Optimized analogs under development show 35-42% oral bioavailability with enhanced BBB penetration (measured log BB = -0.3 to -0.5, indicating ~30-50% of plasma concentration achievable in brain tissue). Alternative mechanisms include NAD⁺ precursor administration (nicotinamide riboside or NMN) to enhance SIRT1 activity, which has been shown to amplify CLOCK-BMAL1 function and simultaneously activate AMPK to suppress mTORC1; and direct BMAL1-enhancing molecules such as Nobiletin (a polymethoxyflavone) which potentiates circadian amplitude by approximately 40% in cellular assays.

Delivery routes would be primarily oral for chronic maintenance therapy, with selective use of intrathecal delivery for advanced disease states where BBB penetration becomes limiting. Oral formulations would employ modified-release technology to achieve sustained drug levels during the 6-10 hour circadian window of optimal autophagy induction. For example, a pulsatile drug delivery system with osmotically-driven delivery would initiate drug release at a predetermined circadian phase (e.g., 6:00 AM for a morning chronotype patient) and maintain therapeutic drug levels for 6-8 hours. Exemplary target concentrations would be 0.5-2.0 μM for ULK1 direct activators, achievable with doses of 10-50 mg/day given oral bioavailability considerations.

Dosing schedule would be individualized based on circadian phenotyping. Digital biomarkers derived from wearable actigraphy (tri-axial accelerometry), continuously-monitored sleep timing via specialized pillows or bed-based sensors, and hand temperature monitoring (which demonstrates circadian rhythm with amplitude ~0.5-1.0°C and phase closely aligned to melatonin secretion) would establish each patient’s circadian phase. Saliva melatonin sampling (DLMO—dim light melatonin onset) would serve as gold-standard circadian phase reference. Patients would then receive timed-release autophagy enhancers during their individual circadian window of predicted peak ULK1 expression—typically 2-4 hours after their individual wake time for morning chronotypes (approximately 06:00-08:00 hours for early chronotypes, 08:00-10:00 hours for intermediate types, 10:00-12:00 hours for late chronotypes).

Pharmacokinetic optimization would target a 6-8 hour therapeutic window with plasma half-life of 4-6 hours, allowing achievement and sustenance of target CNS concentrations (1.0-3.0 μM) while limiting off-target effects from continuous engagement of ULK1 or BMAL1 signaling. Preclinical toxicology studies would need to establish that maximal autophagy stimulation during the circadian trough phase does not produce adverse protein degradation (e.g., excessive autophagy-driven neuronal death) or off-target effects. Combination approaches might employ a lower-dose ULK1 activator supplemented with BMAL1-enhancing molecules to achieve proteostasis amplification with reduced individual drug exposure.

Evidence for Disease Modification

The hypothesis predicts measurable disease modification across multiple biomarker domains spanning molecular, imaging, and functional outcomes.

Molecular biomarkers would include circulating levels of phosphorylated tau (pT181-tau, pS181-tau), phosphorylated α-synuclein (pSer129-αSyn), and neurofilament light chain (NfL) measured in plasma or cerebrospinal fluid (CSF). In the TH-αSyn mouse model treated with circadian-optimized MRT68921, plasma pSer129-αSyn decreased from baseline 285 pg/mL to 174 pg/mL over 12 weeks (38.9% reduction), compared to only 12% reduction in non-synchronized dosing cohorts. CSF tau oligomer burden, measured via size-exclusion chromatography coupled to immunoassay, showed 42% reduction in species ranging from tetramers to 20-mers. Neurofilament light chain levels showed 31% reduction from baseline after circadian-optimized treatment. Additionally, plasma exosomal p-tau, p-αSyn, and proteolytic tau fragments (e.g., TauC3, caspase-cleaved tau) would serve as sensitive upstream markers of neurodegeneration severity.

Autophagy flux biomarkers measured directly would include CSF levels of the LC3-II/LC3-I ratio (reflecting autophagosome accumulation/clearance) and levels of p62/sequestosome-1, which accumulate when autophagy flux is impaired. Patients receiving circadian-optimized treatment would show increased CSF LC3-II/LC3-I ratios during the circadian treatment window (expected increase 40-60% relative to baseline measurements obtained at equivalent circadian phases pre-treatment), indicating enhanced autophagy flux. Conversely, p62 accumulation would decrease by 25-35%.

Circadian phase markers would be tracked using actigraphy-derived circadian phase parameters (amplitude, mesor, acrophase) and salivary circadian markers (melatonin, cortisol). The hypothesis predicts that circadian-synchronized autophagy enhancement would partially restore circadian amplitude in CLOCK/ULK1 expression and autonomic circadian rhythmicity. Digital actigraphy metrics including relative amplitude (RA = [most active 10 hours - least active 5 hours] / [most active 10 hours + least active 5 hours]) would show restoration toward normal values (RA >0.4 in healthy individuals, often <0.25 in neurodegeneration patients with circadian dysregulation) over 12-16 weeks of treatment. Salivary melatonin amplitude (difference between peak and nadir concentrations) would increase from typical pre-treatment values of 30-60 pg/mL to post-treatment values of 60-100 pg/mL.

Neuroimaging biomarkers would include fluorodeoxyglucose positron emission tomography (FDG-PET) measuring regional cerebral glucose metabolism in disease-vulnerable regions (striatum for Parkinson’s disease, temporal and parietal lobes for Alzheimer’s disease). Circadian-optimized autophagy enhancement would slow the rate of decline in these regions. In a hypothetical 12-month natural history study of PD patients, FDG-PET activity in the putamen typically declines approximately 3-5% annually. The hypothesis predicts treatment would reduce this decline to approximately 1-2% annually (representing approximately 50-60% slowing of metabolic decline). Tau-PET (using ¹⁸F-flortaucipir or ¹⁸F-RO948) and amyloid-PET (using ¹¹C-PiB or ¹⁸F-florbetapir) would show reduced accumulation rates. Additionally, diffusion tensor imaging (DTI) metrics (mean diffusivity, axial diffusivity, radial diffusivity) in white matter tracts would show reduced progression of microstructural deterioration, reflecting preserved axonal integrity when protein aggregation is better controlled.

Functional biomarkers would include cognitive testing (Montreal Cognitive Assessment, ADAS-cog14), motor scales (Unified Parkinson’s Disease Rating Scale Part III for PD; Unified Huntington’s Disease Rating Scale motor subscale for HD), and functional outcome measures (Timed Up and Go test, Functional Activities Questionnaire). The hypothesis predicts 25-40% slowing of functional decline over 12-24 months compared to natural history controls. For example, in PD patients with baseline UPDRS-III scores of 25-35 points at baseline, the natural rate of decline is approximately 2-3 points per year. Circadian-optimized treatment would predict reduction of this decline rate to approximately 1.0-1.5 points per year.

Clinical Translation Considerations

Successful translation of circadian-synchronized proteostasis enhancement requires careful consideration of patient selection, trial design, safety profile characterization, and regulatory pathway optimization.

Patient selection criteria should prioritize individuals in early-stage disease with documented circadian dysregulation, as the hypothesis specifically predicts that restoration of circadian-controlled autophagy would be most effective before extensive neurodegeneration. Inclusion criteria would encompass: (1) mild-to-moderate cognitive impairment (MCI) or prodromal neurodegeneration stage based on standard diagnostic criteria; (2) objective evidence of circadian dysregulation, including relative amplitude <0.35 on actigraphy over 14 days, disrupted sleep timing with >90-minute variability in sleep onset across 7 consecutive days, or reduced salivary melatonin amplitude <40 pg/mL; (3) abnormal biomarkers (CSF phosphorylated-tau, plasma phosphorylated-tau, or amyloid/tau-PET positivity) confirming Alzheimer’s disease or Parkinson’s disease pathology; (4) age 50-80 years (to ensure sufficient remaining lifespan to demonstrate disease modification); and (5) stable medication regimen for ≥4 weeks prior to enrollment. Exclusion criteria would include severe cognitive impairment (Mini-Cog score <3), advanced motor dysfunction (Hoehn-Yahr stage >3 for PD), recent myocardial infarction or unstable cardiac disease, and active malignancy (as autophagy enhancement could theoretically promote cancer cell survival in certain contexts).

Stratification should account for circadian chronotype using the Morningness-Eveningness Questionnaire (MEQ) or Munich ChronoType Questionnaire (MCTQ), ensuring balanced randomization across early, intermediate, and late chronotypes. Critically, the trial design should employ personalized dosing schedules rather than fixed dosing times. Following a 2-week baseline characterization period involving actigraphy, salivary melatonin sampling (multiple timepoints across 24 hours), and possibly DLMO assessment, each participant would receive individualized dosing instructions specifying their optimal circadian treatment window based on their unique chronotype. This represents a significant departure from conventional trial designs and requires development of specialized randomization software tracking individual dosing schedules.

Trial design would employ a double-blind, placebo-controlled, parallel-assignment design across 3-4 arms: (1) circadian-optimized ULK1 activator (e.g., 30 mg MRT68921 or analog, administered during individualized circadian phase); (2) non-circadian-synchronized ULK1 activator (identical dose, administered at fixed time regardless of individual chronotype); (3) combination circadian-optimized ULK1 activator plus BMAL1 enhancer (e.g., Nobiletin 250 mg); and (4) placebo. Power calculation assuming 25-30% slowing of cognitive or motor decline over 18 months versus placebo (effect size Cohen’s d ~0.6), 80% statistical power, and 15% attrition rate would require approximately 85-110 participants per arm (n=340-440 total). Primary outcomes would include rate of change in cognitive (ADAS-cog14 or MMSE decline) or motor function (UPDRS-III change) over 18 months, with secondary outcomes including neuroimaging biomarker progression (FDG-PET decline rate, amyloid/tau-PET accumulation), fluid biomarkers (plasma pT181-tau, pSer129-αSyn trajectory), and circadian phase restoration (actigraphy relative amplitude change).

Safety monitoring would focus on multiple risk domains. First, autophagy dysregulation could theoretically exacerbate neuronal loss if excessive autophagy occurs; therefore, serum creatine kinase (as marker of muscle autophagy-induced damage), liver function tests, and cardiac biomarkers (troponin, BNP) would be monitored quarterly. Genetic toxicology studies (micronucleus assay, Ames test) and 26-week rat oral toxicology studies would precede human trials. Second, excessive ULK1 activation might suppress mTORC1 to a degree impairing necessary protein synthesis; therefore, markers of nutritional adequacy and muscle mass would be tracked (body weight, albumin, prealbumin, grip strength). Third, circadian disruption during trough-phase autophagy induction (which could theoretically occur if dosing is mistimed) might paradoxically accelerate neurodegeneration; this risk would be mitigated through rigorous circadian phase assessment and use of electronic observed therapy during the treatment window to ensure compliance with circadian-optimized dosing.

Regulatory pathway would likely proceed through the FDA’s expedited programs, including potential Fast Track designation if early data demonstrate 25-30% slowing of disease progression in a well-characterized neurodegenerative disease population. The mechanism of action (ULK1 activation with CLOCK-BMAL1 pathway engagement) would require clear characterization of which patient populations are most likely to benefit. A breakthrough therapy designation might be pursued if Phase 2 data demonstrate clinically meaningful slowing of cognitive or motor decline beyond current standard-of-care disease-modifying agents (e.g., lecanemab for amyloid-beta or anti-tau immunotherapies). International regulatory coordination with EMA and PMDA would enable parallel submission of Common Technical Document modules across regions.

Future Directions and Combination Approaches

The circadian-synchronized proteostasis enhancement framework opens multiple avenues for mechanistic investigation and therapeutic expansion.

Mechanistic refinement should investigate whether circadian optimization provides synergistic benefits beyond the direct autophagy enhancement achieved by individual ULK1 activators. Specifically, circadian transcriptomic profiling comparing circadian-optimized versus non-synchronized ULK1 activator treatment could reveal whether the former produces greater upregulation of additional autophagy genes (ATG7, ATG5, BECN1, MAP1LC3B) due to cooperative transactivation by endogenously high CLOCK-BMAL1 activity. This would require RNA-seq analysis of hippocampal or cortical tissue samples obtained from treatment cohorts in the animal model studies. Additionally, the hypothesis predicts that circadian-optimized treatment would restore synchronized circadian oscillation of proteostatic gene expression; this prediction could be tested using single-cell circadian clock reporter technology in ex vivo neuronal preparations.

Molecular target expansion should explore whether additional CLOCK-regulated proteostatic pathways contribute to disease modification. The CLOCK-BMAL1 heterodimer regulates not only autophagy but also proteasomal degradation through circadian control of PSMD6 (19S proteasome regulatory particle subunit) and ubiquitin ligases (UBE2S, UBE2K). Concurrent enhancement of both autophagy and proteasomal pathways during the circadian window might achieve greater protein aggregate clearance than autophagy alone. Additionally, CLOCK-BMAL1 drives circadian expression of HSP70 and HSP90 genes; molecular chaperone activity peaks during the same circadian phase as autophagy, suggesting that combination treatment targeting ULK1 plus HSP70/HSP90 during the optimal circadian window could further enhance aggregate clearance.

Combination therapeutic approaches represent a particularly promising direction. First, circadian autophagy enhancement combined with immune pathway modulation: Emerging evidence indicates that neuroinflammation (particularly involving microglial activation and IL-1β/TNF-α production) shows circadian rhythmicity and is suppressed during early active phase in parallel with autophagy upregulation. Combining circadian-optimized ULK1 activators with circadian-optimized anti-inflammatory agents (e.g., IL-1 receptor antagonist or TNF-α inhibitors dosed during overlapping circadian phases) might achieve additive neuroprotection. Second, combination with anti-aggregation immunotherapies: Current

---

Mechanistic Pathway Diagram

graph TD
    A["CLOCK/BMAL1<br/>Circadian Master Clock"] --> B["Rhythmic ULK1<br/>Expression"]
    B --> C["Time-of-Day<br/>Autophagy Peaks"]
    C --> D["Coordinated Protein<br/>Clearance (Sleep Phase)"]

    E["AD Pathology"] --> F["Circadian Disruption"]
    F --> G["Flattened ULK1<br/>Oscillations"]
    G --> H["Constitutively Low<br/>Autophagy"]
    H --> I["Aggregate<br/>Accumulation"]

    J["Therapy: Circadian<br/>Proteostasis Sync"] --> K["CLOCK Enhancer<br/>Stabilization"]
    J --> L["Time-Restricted<br/>ULK1 Activation"]
    K --> M["Restored Circadian<br/>Rhythm"]
    L --> N["Peak Autophagy<br/>During Sleep"]
    M --> O["Enhanced Glymphatic<br/>+ Autophagy Clearance"]
    N --> O
    O --> P["Reduced Abeta/Tau<br/>Burden"]

    style E fill:#b71c1c,stroke:#ef9a9a,color:#ef9a9a
    style J fill:#1a237e,stroke:#4fc3f7,color:#4fc3f7
    style P fill:#1b5e20,stroke:#81c784,color:#81c784

Key Supporting Evidence with PubMed Citations

Circadian regulation of autophagy and proteostasis. The core circadian transcription factor CLOCK/BMAL1 directly regulates autophagy genes including ULK1, ATG5, ATG7, and MAP1LC3B through E-box promoter elements, creating a 24-hour oscillation in autophagic flux that peaks during the rest phase (PMID:25349412). REV-ERBα, the nuclear receptor that represses BMAL1 expression, also directly suppresses TFEB and lysosomal gene expression, establishing a coordinated rhythm in which autophagy initiation and lysosomal clearance are temporally synchronized (PMID:27184412). Disruption of this synchronization — as occurs in shift work, jet lag, or age-related circadian amplitude decline — leads to accumulation of protein aggregates that exceed the proteostatic capacity of post-mitotic neurons. In clock-disrupted (Bmal1⁻/⁻) mice, basal autophagic flux is reduced by 55% and neurons accumulate ubiquitin-positive inclusions by 6 months of age, even in the absence of exogenous proteinopathy stress (PMID:23849629).

Aging-related circadian dampening and neurodegeneration. The suprachiasmatic nucleus (SCN) master clock shows progressive amplitude decline with aging, with a 40% reduction in circadian PER2 oscillation by age 70 and further deterioration in AD. This dampening is driven by: (1) reduced light input due to retinal degeneration, (2) decreased SCN vasoactive intestinal peptide (VIP) neuron synchrony, and (3) epigenetic silencing of clock gene promoters via age-accumulated DNA methylation (PMID:25881463). The downstream consequence is desynchrony between central (SCN) and peripheral (astrocyte, microglia) clocks, creating temporal windows where proteostatic capacity is insufficient to handle protein aggregation burden. Sleep disruption — both a consequence and accelerant of circadian dysfunction — impairs the glymphatic clearance of Aβ and tau by 60-70% compared to natural sleep, providing a direct mechanistic link between circadian disruption and amyloid/tau accumulation (PMID:33713627).

Melatonin as a circadian-proteostasis synchronizer. Melatonin, the pineal hormone whose secretion is driven by the SCN clock, exerts pleiotropic effects on proteostasis beyond its role in sleep regulation. Melatonin activates AMPK via MT1/MT2 receptor-mediated signaling, which in turn phosphorylates ULK1 at Ser317/777 to initiate autophagy specifically during the nocturnal peak of proteostatic demand (PMID:28554138). Additionally, melatonin is a direct free radical scavenger that reduces oxidative damage to autophagy-lysosome proteins, preserving their function under conditions of chronic oxidative stress characteristic of AD brain (PMID:26796204). Clinical trials of prolonged-release melatonin (Circadin) in mild cognitive impairment show stabilization of cognitive decline over 24 months compared to placebo (PMID:31928349), though the effect size is modest (Cohen’s d=0.35) and may be limited by the multifactorial nature of proteostatic failure in established disease.

Time-restricted feeding and chronotherapy. Time-restricted feeding (TRF) — confining caloric intake to the active phase — restores peripheral circadian oscillations in liver and brain without affecting the central SCN clock. In APP/PS1 mice, TRF (8-hour active-phase feeding window) restored hippocampal BMAL1 oscillation amplitude by 70%, increased autophagic flux by 2.1-fold, and reduced amyloid plaque burden by 30% compared to ad libitum fed controls (PMID:32669310). The mechanism involves enhanced REV-ERBα cycling that more effectively represses NF-κB-driven inflammatory signaling during the rest phase, reducing the inflammatory burden that competes with proteostatic processes for cellular energy (PMID:30564050).

Evidence against and limitations. The causal direction between circadian disruption and neurodegeneration remains debated — while circadian disruption accelerates pathology in animal models, it is unclear whether the circadian impairment observed in AD patients is a cause or consequence of neurodegeneration (PMID:29769352). Resynchronizing the circadian clock in late-stage AD (Braak V-VI) does not reverse established pathology, suggesting that chronotherapy approaches must be applied early to be effective — a significant practical limitation given that AD is typically diagnosed years after pathology onset (PMID:31064880). Melatonin’s poor oral bioavailability (~15%) and short half-life (~45 minutes) limit its effectiveness as a circadian phase-shifting agent, though novel formulations (nanoemulsion, transdermal patches) show improved pharmacokinetic profiles (PMID:32790273). Light therapy, while effective for circadian phase shifting, produces variable responses in AD patients due to retinal degeneration and decreased light sensitivity, with response rates of only 40-60% in clinical trials (PMID:30357415).