Why do cells need sleep? The question sounds almost philosophical, but a team led by Zhijian Ji at UCSF has given it a concrete molecular answer. Using a custom-built biosensor they named GENTIS, the researchers discovered that individual cells accumulate protons during periods of sustained activity, and that this intracellular acidification directly triggers sleep behavior. The findings, published July 24 in Science Advances, offer the first direct link between cellular pH dynamics and sleep regulation.
The New Tool: GENTIS
To see what cells are doing when an animal sleeps, the team designed GENTIS (a modular sensor that reports nuclear translocation driven by ionic strength). The design is elegant: an acidic E5 helix (five copies of EVSALEK) fused to a basic K5 helix (five copies of KVSALKE), flanked by a nuclear localization signal (NLS) and a nuclear export signal (NES), with an N-terminal GFP for visualization.
At low ionic strength, electrostatic attraction between the oppositely charged helices keeps the NLS masked, so the NES dominates and the sensor stays in the cytoplasm. When ionic strength rises, the salt bridges break, exposing the NLS and triggering importin-dependent nuclear import. The entire translocation takes about 20 minutes and responds dose-dependently to osmotic stress (10% sorbitol, 250 mM NaCl). Importin RNAi against ima-3 and imb-1 abolished the translocation. A variant lacking the charged helices was constitutively nuclear, confirming the mechanism. Notably, 1,6-hexanediol, an inhibitor of liquid-liquid phase separation, had no effect.
Rhythmic Proton Pulses During Development
When Ji’s team expressed GENTIS under the rpl-28 promoter in C. elegans, they observed something striking: rhythmic nuclear translocation in intestinal cells precisely timed to the animal’s larval molts. Across the L1 through L4 larval stages, three distinct waves of GENTIS nuclear import appeared, and each wave coincided exactly with behavioral quiescence — the worm stopped pharyngeal pumping and arrested locomotion. The more GENTIS-positive nuclei a worm showed, the lower its pumping rate.
Something was causing ionic stress inside intestinal cells right when the worm entered a sleep-like state.
V-ATPase Disassembly Triggers Acidification
To find the source of the ionic signal, the team ran a systematic screen of RNAi and pharmacological agents. Only one compound robustly induced GENTIS translocation: FAc (4-(2-furyl)-3-buten-2-one), a known inhibitor of the C. elegans V-ATPase that targets the VHA-12 and VHA-13 subunits. The V-ATPase is a proton pump that acidifies intracellular compartments; inhibiting it apparently caused protons to leak into the cytoplasm.
Several lines of evidence confirmed this. FCCP, a protonophore that equilibrates protons across membranes, suppressed FAc-induced GENTIS translocation but did not block sorbitol-induced translocation — meaning FAc acts specifically through proton elevation, while sorbitol works through osmotic stress. A ratiometric pHluorin sensor confirmed that FAc caused intracellular acidification. And at the same doses (0.2 to 5 mg/mL), FAc induced reversible locomotion quiescence — the worms stopped moving.
Critically, the sleep-active neuron RIS showed calcium activation after FAc treatment, and FAc-induced sleep required the RIS neuron (aptf-1 mutants failed completely), the ALA neuron (ceh-14 mutants showed partial failure), the DMSR-1 receptor, and the intestinal H+/Na+ exchanger PBO-4.
Proton Buffering Suppresses Sleep
If proton accumulation drives sleep, then buffering those protons should reduce sleep. The team tested this with NH4Cl, a weak base that neutralizes intracellular acidity. The effect was dose-dependent: 0.5 mM NH4Cl suppressed sleep by 5% (p = 0.005), 5 mM by 29% (p = 0.004), and 50 mM by 44% (p = 0.0002). Importantly, NH4Cl did not block GENTIS translocation itself — it acted downstream of the acidification signal. During the natural L1-L2 molt, NH4Cl reduced locomotion quiescence but did not delay the onset of pumping quiescence, suggesting that proton signaling is specifically required for the behavioral component of sleep.
V-ATPase Downregulation in Real Time
To watch the V-ATPase itself during sleep, the team used CRISPR to tag VHA-13 with mScarlet. The results were unambiguous: after treatment with FAc, UV (0.1 J/m2), 20% sorbitol, or during natural molting (L1-L2 and L4-YA transitions), the apical membrane signal of mScarlet::VHA-13 progressively disappeared. The V-ATPase was being disassembled and removed from the membrane.
Chai Discovery’s modeling platform predicted that FAc binds near the ATP-binding pocket of VHA-13. Consistent with this, a gain-of-function mutant, vha-13(luc133), showed reduced FAc-induced sleep — a hyperactive pump resisted disassembly, and the worm slept less.
Why It Matters
This study provides the first direct molecular mechanism connecting cellular metabolism to sleep behavior. The sequence is clear: sustained cellular activity leads to V-ATPase disassembly, which allows protons to leak into the cytoplasm; the resulting acidification activates the RIS sleep-promoting neuron through a signaling cascade involving ALA, DMSR-1, and PBO-4; and the worm enters a quiescent state that persists until the proton load is cleared.
The finding suggests that sleep at the organismal level may originate from an ancient cellular need to manage proton balance. If the same mechanism operates in vertebrates, it could open entirely new avenues for sleep therapeutics — drugs that modulate V-ATPase activity or intracellular pH might one day treat insomnia or other sleep disorders without targeting neurotransmitter systems.
Limitations
The work was done entirely in C. elegans, and the sleep studied is developmentally timed sleep associated with molting, not adult homeostatic sleep. Whether V-ATPase disassembly and proton accumulation drive sleep in vertebrates — including humans — remains to be tested. The GENTIS sensor itself, while powerful, reports ionic strength rather than pH directly, requiring additional tools like pHluorin for confirmation.
Bottom Line
A modular biosensor called GENTIS revealed that intestinal cells in C. elegans accumulate protons during larval molts, and that this acidification — caused by disassembly of the V-ATPase proton pump — triggers sleep via the RIS neuron. Buffering protons with NH4Cl suppresses sleep, and blocking the pathway at any node (RIS, ALA, DMSR-1, PBO-4) abolishes it. The work establishes intracellular pH as a direct molecular driver of sleep behavior.
Source
Ji Z, et al. Cells sound a pH alarm: GENTIS sensor reveals proton accumulation drives sleep at the cellular level. Sci Adv. 2026 Jul 24;12(30):eaef3219. doi: 10.1126/sciadv.aef3219. PMID: 42497280.