Hypothesis

Mid‑afternoon darkness activates a NAD+‑dependent SIRT1/CREB cascade that upregulates BDNF and restores hippocampal spine density eroded by chronic dim‑light exposure.

Mechanistic rationale

Prolonged exposure to dim light (<50 lux) reduces hippocampal BDNF and dendritic spine density, impairing spatial learning 1. Subsequent bright light and darkness periods fully reverse these deficits, indicating that darkness itself drives restorative processes 2. The restorative effect coincides with upregulation of the SIRT1/CREB/synapsin axis 3. We propose that the darkness‑triggered rise in intracellular NAD+, resulting from lowered neuronal metabolic rate during low‑photon conditions, activates SIRT1 deacetylase activity. Activated SIRT1 deacetylates CREB at Ser133, enhancing its transcriptional activity on the Bdnf promoter and on synapsin‑I. Increased BDNF then promotes dendritic spine formation and synaptic plasticity, completing the repair cycle.

Novel prediction

If the NAD+‑SIRT1 step is essential, then a 30‑minute dark interval administered at 14:00 each day for 14 days will:

• Elevate serum BDNF levels by ≥15 % relative to baseline.
• Improve performance on a virtual Morris water maze task (reduced escape latency by ≥20 %).
• These changes will be absent when participants receive the same darkness protocol while taking the SIRT1 inhibitor EX527 (5 mg orally) or when the dark interval is replaced by dim light (<30 lux).

Experimental design (human)

• Participants: 60 healthy adults aged 18‑35, randomized to three groups (n=20 each): Darkness + placebo, Darkness + EX527, Dim‑light control.
• Procedure: Daily 14:00‑14:30 session under either <5 lux darkness or ~30 lux dim light (matched for lack of visual cues). EX527 or placebo taken 30 min before the session.
• Outcomes: Serum BDNF measured via ELISA at 0, 7, and 14 days; cognitive testing with a computerized radial‑arm maze analogue; actigraphy to confirm sleep timing unchanged.
• Analysis: Mixed‑effects model testing group × time interaction. A significant increase in BDNF and cognitive gain only in the Darkness + placebo group would support the hypothesis; lack of effect in the Darkness + EX527 group would falsify the NAD+‑SIRT1 dependence.

Falsifiability

The hypothesis is falsifiable if:

• Darkness exposure fails to raise BDNF or improve spatial performance despite protocol adherence.
• SIRT1 blockade does not attenuate the BDNF or cognitive gains.
• Dim‑light control produces equal or greater BDNF elevation than darkness.

Such outcomes would indicate that either the proposed metabolic‑SIRT1 link is incorrect or that other pathways mediate darkness‑induced neuroplasticity.