https://www.nature.com/articles/s41586-025-08734-4 Skip to main content Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript. Advertisement Advertisement Nature * View all journals * Search * Log in * Explore content * About the journal * Publish with us * Sign up for alerts * RSS feed 1. nature 2. articles 3. article Sleep pressure accumulates in a voltage-gated lipid peroxidation memory Download PDF Download PDF * Article * Open access * Published: 19 March 2025 Sleep pressure accumulates in a voltage-gated lipid peroxidation memory * H. Olof Rorsman ORCID: orcid.org/0000-0002-8263-2460^1^ na1, * Max A. Muller ORCID: orcid.org/0000-0002-9960-5020^2^ na1, * Patrick Z. Liu ORCID: orcid.org/0000-0001-9663-8233^1, * Laura Garmendia Sanchez ORCID: orcid.org/0009-0009-5933-3081^1, * Anissa Kempf^1^ nAff3, * Stefanie Gerbig^2, * Bernhard Spengler^2 & * ... * Gero Miesenbock ORCID: orcid.org/0000-0003-2940-3228^1 Show authors Nature (2025)Cite this article * Metrics details Subjects * Cellular neuroscience * Ion channels in the nervous system * Lipidomics * Sleep Abstract Voltage-gated potassium (K[V]) channels contain cytoplasmically exposed b-subunits^1,2,3,4,5 whose aldo-keto reductase activity^6,7,8 is required for the homeostatic regulation of sleep^9. Here we show that Hyperkinetic, the b-subunit of the K[V]1 channel Shaker in Drosophila^7, forms a dynamic lipid peroxidation memory. Information is stored in the oxidation state of Hyperkinetic's nicotinamide adenine dinucleotide phosphate (NADPH) cofactor, which changes when lipid-derived carbonyls^10,11,12,13, such as 4-oxo-2-nonenal or an endogenous analogue generated by illuminating a membrane-bound photosensitizer^9,14, abstract an electron pair. NADP^+ remains locked in the active site of K[V]b until membrane depolarization permits its release and replacement with NADPH. Sleep-inducing neurons^15,16,17 use this voltage-gated oxidoreductase cycle to encode their recent lipid peroxidation history in the collective binary states of their K[V]b subunits; this biochemical memory influences--and is erased by--spike discharges driving sleep. The presence of a lipid peroxidation sensor at the core of homeostatic sleep control^16,17 suggests that sleep protects neuronal membranes against oxidative damage. Indeed, brain phospholipids are depleted of vulnerable polyunsaturated fatty acyl chains after enforced waking, and slowing the removal of their carbonylic breakdown products increases the demand for sleep. Main The pore-forming a-subunits of voltage-gated potassium channels of the K[V]1 and K[V]4 families partner with non-membrane-integral b-subunits^1,2,3,4,5 whose sequences exhibit puzzling similarity with aldo-keto reductases^6,7--enzymes that reduce carbonyls to alcohols via the coupled oxidation of an NADPH cofactor. The isolated b-subunits show weak reductase activity towards a range of model aldehydes in vitro^18,19, relying on NADPH as the electron donor, but whether, on which native carbonyls, and to what end the assembled K [V] channel catalyses similar reactions in vivo is unknown. The exceptionally firm grip of K[V]b on its cofactor^8, which chokes catalysis, deepens the mystery of why an ion channel would be shackled to what appears to be a subpar enzyme. A hint at a possible answer has come from studies in Drosophila, where both the K[V]1 channel Shaker^20,21 and its b-subunit Hyperkinetic^7 are needed to sustain normal levels of sleep^22,23. The sleep-regulatory function of the channel complex has been mapped to a small number of sleep-control neurons whose axonal projections target the dorsal fan-shaped body in the central brain^15,17,24 (dFBNs). Sleep need is encoded in the electrical activity of these neurons^16, which fluctuates--in part^24--because Hyperkinetic modulates the inactivation kinetics of the Shaker current^9. During waking, electrons leaking from the saturated transport chains of the inner mitochondrial membrane produce superoxide and other reactive oxygen species (ROS), which convert the K[V]b pool to the NADP^ +-bound form^9,25. This prolongs the inactivation time constant of the associated potassium conductance^9,18,26,27, strengthens the repolarizing force that restores the resting membrane potential after each spike, and so enables dFBNs to fire at higher rates^9,28. Although the source (the mitochondrial electron transport chain) and the receiver (Hyperkinetic in complex with Shaker) of the sleep-promoting redox signal are known^9,25, the mode of communication between mitochondria and potassium channels remains undefined. K[V]b-bound NADPH is an unlikely direct target of ROS, not only because radical-induced hydrogen abstraction (which involves a single electron transfer) will not produce NADP^+ (which would require the loss of two electrons). As ROS spread from the inner mitochondrial membrane, they encounter many potential reaction partners before reaching Hyperkinetic at the cell surface. Among the most abundant and vulnerable ROS targets in the immediate vicinity of their site of origin are the polyunsaturated fatty acyl chains (PUFAs) of membrane lipids, whose peroxidation and subsequent fragmentation into carbonyls^10,11,12,13 can create chemical functionality fit for the active site of an aldo-keto reductase. In the crystal structure of the mammalian K[V]1.2-b2 channel complex, the substrate binding pocket is lined with hydrophobic residues and filled with unresolved electron density^4, as would be expected if a diverse group of lipid precursors disintegrated into a heterogeneous mix of apolar ligands. Recombinant K[V]b1 and K[V]b2 reduce synthetic analogues of lipid peroxidation products, such as 4-oxo-2-nonenal (4-ONE), 1-palmitoyl-2-oxovaleroyl-phosphatidylcholine or methylglyoxal, in vitro^18,19, but turnover is so slow that the effect on the concentrations of these molecules in vivo must be minimal. While K[V]b can therefore have no plausible role in the enzymatic clearance of toxic carbonyls, the very features that seem detrimental or baroque in a catalyst--the protein's stranglehold on NADP(H) and its linkage to a voltage-gated ion channel--could be essential if the assembly instead functioned as a biochemical memory cell (Fig. 1). Imagine that tight binding of NADP(H) causes the redox reaction to pause at the cofactor-exchange step. Each b-subunit then records a single exposure to an oxidizing substrate by flipping from the NADPH-bound to the NADP^+-bound form and stores this bit of information until NADP^+ is released and replaced by NADPH (Fig. 1a). The operational logic resembles that of a single-transistor dynamic random-access memory (DRAM) cell^29 (Fig. 1b): K[V]b corresponds to the storage capacitor of a DRAM cell; the oxidation state of NADP(H) plays the part of the electric charge on the capacitor; and the (low) basal reaction rate is equivalent to the leakage of charge from the capacitor, which gives the memory a finite lifetime that requires periodic refreshment^29. The analogy would be complete if, akin to the voltage across the transistor that gates access to the storage capacitor in a DRAM chip^29, the membrane potential across the voltage sensors of the a-subunit controlled the rate of cofactor exchange by the b-subunit (Fig. 1). Fig. 1: Information storage by K[V]b. figure 1 a, The bits 0 (left) and 1 (centre) are stored in the cofactor oxidation state of the K[V]b subunit. The memory is read out when the membrane potential across K[V]a depolarizes and K[V]b discharges NADP ^+ (right). b, The bits 0 (left) and 1 (centre) are stored in the electrical charge on the capacitor of a DRAM cell. The memory is read out when the voltage across the access transistor gate goes high and the capacitor discharges (right). Full size image Here we test several tenets of this model. We examine the lipids of rested and sleep-deprived brains for signs of oxidative damage; measure the effect on sleep of perturbing the clearance of peroxidized lipids; determine whether lipid peroxidation products influence the Shaker current of sleep-control neurons via the active site of Hyperkinetic; and analyse the interplay of voltage sensors and NADP(H) binding sites in the redox regulation of the channel. The results define an autoregulatory loop in which the K[V]1 channel population encodes the recent lipid peroxidation history of a neuron in the collective binary states of their b-subunits. This biochemical memory (which we equate to the accumulated sleep pressure) is read and erased during subsequent electrical activity, with the action potential frequency set by the fraction of K[V]b subunits previously loaded with NADP^+. A lipidomic fingerprint of sleep loss Because levels of oxidative stress may differ among tissues, brain regions or neuron types^9,30, we collected spatial maps of hundreds of lipids by means of high-resolution scanning microprobe matrix-assisted laser desorption/ionization mass spectrometry imaging (SMALDI-MSI). The lipid maps were acquired by scanning 10-um-thick cryosections of rested or sleep-deprived brains at a lateral resolution of 5 um x 5 um and overlaid on fluorescence images of dFBNs expressing R23E10-GAL4-driven^16 mCD8::GFP (Fig. 2a). Fig. 2: Sleep deprivation depletes brain phospholipids of polyunsaturated fatty acids. figure 2 a, Example fluorescence (top) and positive-ion SMALDI-MS images (bottom) of cryosections containing dFBNs marked with mCD8::GFP. The sections were cut from rested (left) or sleep-deprived brains (right). SMALDI-MS images show, from top to bottom, the spatial distributions of phosphatidylinositol 18:2/20:2 (m/z 887.5612, [M+Na] ^+), phosphatidylserine 18:3/20:5 (m/z 826.4618, [M+Na]^+), phosphatidylcholine 18:0/18:1 (m/z 788.6140, [M + H]^+), phosphatidylcholine 18:3/18:3 (m/z 778.5345, [M + H]^+), phosphatidylethanolamine 18:1/18:1 (m/z 744.5536, [M + H]^+) and phosphatidic acid 18:2/20:3 (m/z 723.4932, [M + H]^+). Scale bar, 200 mm. b, Hierarchical clustering of rested and sleep-deprived brains according to their glycerophospholipid profiles. Heat maps show the z-scored intensities of m/z signals differing with sleep history at an FDR-adjusted P < 0.05 (two-sided t-test). Lipids detected in MS^2 fragmentation experiments are annotated in green in the list of molecular assignments on the left. Each column represents a different cryosection (n = 9 per condition); sections of the same brain (n = 3 per condition) are grouped by grey bars on top. c, Volcano plot of sleep history-dependent changes in 380 m/z signals annotated as glycerophospholipids. Signals with more than twofold intensity changes and FDR-corrected P < 0.05 (two-sided t-test) are indicated in black. Numerical labels reference data points to lipid annotations in b. d, Features overrepresented in the subset of 51 differentially abundant lipids against the background set of all 380 glycerophospholipids. Asterisks indicate significant enrichment scores (FDR-corrected P < 0.05, Fisher's exact test). Because phosphatidylcholine and phosphatidylethanolamine lipids cannot be distinguished by exact mass alone, they are grouped as a single feature. LPC, lysophosphatidylcholine; LPE, lysophosphatidylethanolamine; LPI, lysophosphatidylinositol; PA, phosphatidic acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphatidylinositol; PS, phosphatidylserine; O-, alkyl ether linkage. Source Data Full size image Samples within each group had tightly correlated lipid profiles, but differences between groups--that is, between the rested and sleep-deprived states--were so stark that sleep histories could be accurately inferred from lipid composition alone; a single principal component captured 85% of the overall variance. Fifty-one out of 380 SMALDI-MSI signals annotated as glycerophospholipids and detected exclusively on tissue increased or decreased more than twofold after sleep loss, with a false discovery rate (FDR)-adjusted significance threshold of P < 0.05 and little, if any, spatial heterogeneity across the brain (Fig. 2a-c). The identities of 18 of these 51 differentially abundant phospholipids (35%) were confirmed by targeted MS^2 fragmentation after HPLC separation of a methyl tert -butyl ether extract of brain homogenates (Fig. 2a-c). In many cases these analyses also revealed the detailed fatty acid compositions of the parent species (Fig. 2b,c). Most lipids with high discriminatory power belonged to one of three classes, which form discernible blocks in the clustergram of Fig. 2b. The glycerophospholipids of rested brains carried inositol, serine, ethanolamine or choline head groups and were enriched in acyl chains with a combined median length of 37.5 carbons and a large degree of unsaturation; the number of double bonds averaged 5.0 +- 2.61 (mean +- s.d.) per lipid, with a median of 5 and a maximum of 12 (Fig. 2b-d). Phospholipids that were present at higher levels in sleep-deprived brains, by contrast, contained mostly choline and ethanolamine head groups, shorter acyl chains with a combined median length of 33.5 carbons, and many fewer double bonds than those in rested flies; the number of double bonds averaged 2.0 +- 2.03 (mean +- s.d.) per lipid, with a median of 2 (Fig. 2b-d). The third distinctive lipid class consisted of several species of phosphatidic acid, whose levels declined after sleep deprivation (Fig. 2b-d). Phosphatidic acid occupies a central position in the biosynthetic pathways of all glycerophospholipids^31,32 and promotes mitochondrial fusion when generated locally by a dedicated phospholipase D (mitoPLD)^33. Impaired mitoPLD activity in dFBNs causes sleep loss^25 . The lipidomic fingerprint of sleep-deprived brains indicates that their membranes are depleted of PUFAs, presumably as a consequence of oxidative damage, leaving behind a greater proportion of largely saturated phospholipids (Fig. 2d). The picture during rest is consistent with membrane repair via glycerophospholipid biosynthesis from phosphatidic acid precursors^31,32 and a reversal of the mitochondrial fragmentation that commonly accompanies periods of oxidative stress^34, including sleep deprivation^25. Lipid-derived carbonyls promote sleep The peroxidation of membrane lipids begins^11,12 with the abstraction of a bis-allylic hydrogen from a PUFA chain by a radical oxidant such as HOO* (the conjugate acid of \({{\rm{O}}}_{2}^{-}\)) or *OH. The resulting lipid radical reacts with O[2] to form a lipid peroxyl radical, which propagates the chain by abstracting a hydrogen from another PUFA, generating a new lipid radical and a lipid hydroperoxide^10,11,12,13. The reaction continues until two radicals combine in a termination step. The lipid hydroperoxides produced along the way undergo a series of rearrangements and scissions that give rise to a variety of short- and medium-chain carbonyl breakdown products^10,11,12,13, including the potential K[V]b substrate^18,19 4-ONE. Operating behind a primary bastion of enzymatic and non-enzymatic antioxidants^35, soluble short-chain dehydrogenases/reductases, such as carbonyl reductase 1 in mammals^36,37 and its functional homologue sniffer in Drosophila^38,39, form a second defensive ring against lipid peroxidation-derived carbonyls. We examined whether breaching and mending these secondary defences would recapitulate the well-documented effects on sleep of pro- and antioxidant manipulations^9,30,40. Indeed, hemizygous male carriers of the X-linked hypomorphic sniffer allele sni^1 showed increased sleep durations during the day and night (Fig. 3a-c and Extended Data Fig. 1a), owing to vastly extended, hyperconsolidated sleep episodes (Extended Data Fig. 1b,c), at an age before widespread neurodegeneration^38 produced locomotor deficits that could have been mistaken for sleep (Extended Data Fig. 1d). Sleep returned to or below wild-type levels when sni^1 mutants expressed a UAS-sni rescue transgene^38 (Fig. 3c and Extended Data Fig. 1a), and similarly when the alternative oxidase AOX, which shunts surplus electrons from ubiquinone to H[2]O, capped mitochondrial ROS production^9,41 (Fig. 3a,c), or when the putative carbonyl sensor Hyperkinetic was removed by RNA-mediated interference (RNAi), either pan-neuronally or in dFBNs of sni^1 mutant flies (Fig. 3b,c). These data place lipid peroxidation products downstream of mitochondrial respiration in the signalling chain that terminates on the Hyperkinetic pool of dFBNs to raise the pressure to sleep^9. Fig. 3: Lipid peroxidation products are intermediates in the signalling chain that couples mitochondrial electron transport to sleep. figure 3 a, The nSyb-GAL4- or R23E10-GAL4-driven expression of AOX in hemizygous sni^1 mutant males fully or partially restores wild-type sleep (two-way repeated-measures ANOVA with Holm-Sidak test; sample sizes in c). The sleep profiles of sni^1 mutants with pan-neuronal expression of AOX differ from those of sni^1 mutants (P < 0.0001) but not of wild-type flies (P = 0.0589), whereas the sleep profiles of sni^1 mutants with dFBN expression of AOX differ from those of both sni^1 mutants (P < 0.0001) and wild-type flies (P = 0.0007). b, nSyb-GAL4- or R23E10-GAL4-restricted interference with the expression of Hyperkinetic in hemizygous sni^1 mutant males partially or fully restores wild-type sleep (two-way repeated-measures ANOVA with Holm-Sidak test; sample sizes in c). The sleep profiles of sni^1 mutants with pan-neuronal expression of Hk^RNAi differ from those of both sni^1 mutants (P < 0.0001) and wild-type flies (P < 0.0001), whereas the sleep profiles of sni^1 mutants with dFBN expression of Hk^RNAi differ from those of sni^1 mutants (P < 0.0001) but not of wild-type flies (P = 0.1344). c, Sleep in hemizygous males carrying the sni^1 allele differs from wild-type (P < 0.0001; Kruskal-Wallis ANOVA with Dunn's test) but returns to or below control level if carriers also express sniffer (sni), AOX or Hk^RNAi pan-neuronally under the control of nSyb-GAL4 (sni: P = 0.1128; AOX: P > 0.9999; Hk^ RNAi: P = 0.0601) or in dFBNs under the control of R23E10-GAL4 (sni: P = 0.1151; AOX: P = 0.6694; Hk^RNAi: P > 0.9999). Note that the expression of the UAS-sni transgene appears leaky, as the sleep phenotype of sni^1 mutants is rescued in the absence of a GAL4 driver (P > 0.9999). Data are mean +- s.e.m.; n, number of flies; asterisks indicate significant differences (P < 0.05) from wild type in planned pairwise comparisons. For statistical details see Supplementary Table 1. Source Data Full size image A redox memory of lipid peroxidation To determine whether lipid peroxidation-derived carbonyls could alter the oxidation state of Hyperkinetic's cofactor, we obtained whole-cell voltage-clamp recordings from dFBNs and estimated the NADP ^+:NADPH ratio of the K[V]b population from the bi-exponential inactivation kinetics of the A-type current (I[A]) (Extended Data Fig. 2): a reduced cofactor increases, whereas an oxidized cofactor decreases, the rate of channel inactivation^9,18,26,27. If PUFA-derived carbonyls are endogenous electron acceptors at the active site of K[V]b, their ballooning levels in sni^1 mutants^38,39 should drive the Shaker-Hyperkinetic complex into the NADP^+-bound, slowly inactivating state. Increases in the fast and slow inactivation time constants (t[fast] and t[slow], respectively) of the A-type current relative to wild-type flies indicate that this was indeed the case (Fig. 4a,b). Fig. 4: Lipid peroxidation products alter the inactivation kinetics of I[A] via the active site of K[V]b. figure 4 a,b, The sni^1 allele increases the fast and slow inactivation time constants of I[A] in dFBNs of hemizygous carriers (turquoise) relative to wild-type males (grey) (b; t[fast]: P = 0.0060, two-sided t-test; t[slow]: P = 0.0253, two-sided Mann-Whitney test; examples of peak-normalized I[A] evoked by voltage steps to +30 mV in a). c,d, dFBNs expressing miniSOG were held at -80 mV, except during the voltage protocols required to measure I[A]. A 9-min exposure to blue light between the 0- and 10-min time points (d; blue) increases the fast and slow inactivation time constants of I[A] above their pre-illumination baselines (d; t[fast]: P = 0.0133; t[slow]: P = 0.0041; repeated-measures ANOVA; examples of peak-normalized I[A] evoked in the same dFBN by voltage steps to +30 mV in c). e,f, dFBNs were held at -80 mV, except during the voltage protocols required to measure I[A]. The inclusion of 50 uM 4-ONE in the intracellular solution (f) increases the fast and slow inactivation time constants of I[A] above the baselines recorded immediately after break-in (f; t [fast]: P = 0.0015; t[slow]: P = 0.0010; mixed-effects model; examples of peak-normalized I[A] evoked in the same dFBN by voltage steps to +30 mV in e). g, dFBNs were held at -80 mV, except during the voltage protocols required to measure I[A]. At 10 min after break-in, the inclusion of 50 uM 4-ONE, but not of 200 uM 4-HNE, in the intracellular solution increases the fast and slow inactivation time constants of I[A] from control to sleep-deprived levels, provided dFBNs express catalytically competent Hyperkinetic (t[fast]: P < 0.0001; t[slow]: P < 0.0001; Kruskal-Wallis ANOVA). Columns show population averages; dots represent individual cells; n, number of cells; asterisks indicate significant differences (P < 0.05) relative to the 0-min time point or control levels in planned pairwise comparisons by Holm-Sidak or Dunn's test. For statistical details see Supplementary Table 1. Source Data Full size image Plasma membrane-anchored miniSOG^14 allowed us to switch the cofactor acutely to the oxidized state^9 and follow its fate thereafter. The exposed chromophore of this light-oxygen-voltage-sensing (LOV) domain protein^14 transfers the energy of blue light efficiently to O[2], producing singlet oxygen (^1O[2]) which--presumably indirectly, via a burst of lipid peroxidation--converts the channel population to the NADP^+-bound form and induces sleep^9. The oxidation of the cofactor was detected as an increase in the fast and slow inactivation time constants after 9 min of blue light exposure, from initial mean values of 5.8 and 35 ms to final averages of 8.2 and 59 ms (Fig. 4c,d ). When the membrane potential was clamped at -80 mV, t[fast] and t [slow] stayed stably elevated for 20 min after the light-driven ^1O [2] generation stopped (Fig. 4c,d), consistent with a negligible rate of spontaneous NADP^+ exchange^8,18,27 that allows Hyperkinetic to retain a memory of an earlier encounter with an oxidizing substrate, even if that molecule is itself short-lived (estimated intracellular half-life^10 of lipid-derived carbonyls <4 s). In a direct test of the idea that PUFA-derived carbonyls are prominent among these substrates, we filled dFBNs through the patch pipette with the synthetic lipid peroxidation products 4-ONE or 4-hydroxynonenal (4-HNE)^10,12,13. Owing to their inherent reactivity and membrane-permeability, the equilibration of these carbonyls within the neuronal arbor was governed by complex reaction-diffusion kinetics that made their concentration profiles difficult to predict^ 13 and, in all likelihood, neither spatially uniform nor temporally stationary during the course of a recording. 4-ONE and 4-HNE are estimated (with large uncertainty) to be present in cells in the low to sub-micromolar range under basal conditions but reach millimolar concentrations during periods of oxidative stress^10. Although 4-HNE is viewed as a useful marker of lipid peroxidation because monoclonal antibodies can detect its protein adducts^42, mammalian K[V]b2 in vitro shows detectable catalytic activity only towards 4-ONE^19. If the substrate preferences of Drosophila Hyperkinetic were similar, 4-HNE could serve as an ideal control to distinguish effects due to the enzymatic conversion of reactive carbonyls from those potentially caused by indiscriminate protein modification^13. Comparisons of I[A] inactivation kinetics immediately after break-in and 10 min later revealed a clear slowing of the fast and slow time constants, with effect sizes similar to those after the miniSOG-driven photogeneration of ROS (Fig. 4e,f) or a night of mechanical sleep deprivation (Fig. 4g). Changes were seen only in dFBNs perfused with 50 uM 4-ONE; 200 uM 4-HNE, the addition of 0.15% methyl acetate vehicle to the intracellular solution, or the passage of time alone had no effect (Fig. 4g and Extended Data Fig. 3a). When the cells were held at -80 mV in 4-ONE for extended periods, the inactivation time constants completed much of their climbs to higher plateaux within the first 10 min and remained there for the rest of the recordings (Fig. 4f). Because each neuron in this experimental configuration was connected to a practically infinite reservoir of 4-ONE, however, the persistent slowing of inactivation could reflect continuous turnover of substrate rather than a lasting switch in the oxidation state of the cofactor; it can therefore not speak as unequivocally to the longevity of the redox memory as the enduring increase of t[fast] and t[slow] in miniSOG-expressing dFBNs after a finite light exposure can (Fig. 4d). Membrane resistances, membrane time constants, and the amplitude of the non-A-type potassium current remained approximately constant over the course of 30 min, but the magnitude of I[A] slowly declined (Extended Data Figs. 3a,b and 4). This trend is likely to reflect closed-state inactivation^43 rather than a gradual loss of voltage control over a portion of the channels before an increase in access resistance would have prompted us to terminate the recording: series resistances stayed within stable limits for 30 min, irrespective of the presence of 4-ONE or changes in command potential or the inactivation kinetics of I[A] (Extended Data Fig. 3c), but the steady-state half-inactivation voltages drifted towards more hyperpolarized potentials^43 (Extended Data Fig. 3d). Because the same slow rundown of I[A] was also observed in the absence of 4-ONE (Extended Data Fig. 3a), after miniSOG stimulation (Extended Data Fig. 4b), and in homozygous Hyperkinetic-null mutants (below), the effect cannot be explained by a direct irreversible 4-ONE hit on the b-subunit. For the most stringent proof that 4-ONE altered the Shaker current via its reduction at the active site of K[V]b (as opposed to an off-target modification on the channel or elsewhere), we expressed transgenes encoding catalytically active or dead Hyperkinetic^44 under R23E10-GAL4 control in dFBNs of Hyperkinetic-null mutant (Hk^1/ Hk^1) flies^9. Infiltrating the Shaker channel with a b-subunit devoid of oxidoreductase activity^18,27,44 (Hk(K289M)) rendered the fast and slow components of A-type inactivation resistant to 4-ONE, whereas the incorporation of functional K[V]b preserved the sensitivity of the channel (Fig. 4g and Extended Data Fig. 5). Impaired carbonyl clearance, the photogeneration of ROS, and synthetic 4-ONE exerted indistinguishable effects on I[A] in voltage-clamp recordings (Fig. 4a-f), but only carriage of the sni^1 mutation or miniSOG-mediated photooxidation also enhanced the spiking response of dFBNs to membrane depolarization (Fig. 5a,b). The delivery of 4-ONE through a patch electrode at the soma did not (Fig. 5c), in all likelihood because the diffusion time of 4-ONE to dFBN axons, which appear rich in Hyperkinetic but are connected to the cell body through a long, thin primary neurite (Fig. 5d), exceeded the brief intracellular half-life of the molecule^10,13. The release of endogenous lipid peroxidation products, by contrast, whether instigated by miniSOG or amplified by a lack of sniffer, was sufficiently decentralized to be felt also in remote parts of the neuron. The variable spread of externally supplied and internally generated carbonyls will matter little in measurements of voltage-gated potassium currents, which for space-clamp reasons are dominated by channels near the somatic recording site^45 (Fig. 5d-f), but come to the fore in recordings of action potentials if the spike initiation zone lies outside the diffusion distance of 4-ONE. Fig. 5: Lipid peroxidation products increase the excitability of dFBNs via axonal K[V]b. figure 5 a-c, Example voltage responses to current steps (left) and voltage-spike frequency functions (right; mean +- s.e.m.) of dFBNs. In each neuron, the size of the unitary current step was adjusted to produce a 5-mV deflection from a resting potential of -60 +- 5 mV. The sni^1 mutation steepens the voltage-spike frequency function of hemizygous carriers (turquoise, n = 11 cells) relative to wild-type males (grey, n = 10 cells) (a; genotype effect: P = 0.0003; current x genotype interaction: P < 0.0001; two-way repeated-measures ANOVA). Blue illumination for 9 min steepens the voltage-spike frequency function of dFBNs expressing miniSOG (blue, n = 6 cells) relative to controls kept in darkness (grey, n = 7 cells) (b; illumination effect: P = 0.0235; current x illumination interaction: P = 0.0008; two-way repeated-measures ANOVA). The inclusion of 50 uM 4-ONE in the intracellular solution (turquoise, n = 12 cells) does not steepen the voltage-spike frequency function relative to controls at the 10-min time point (grey, n = 10 cells) (c; 4-ONE effect: P = 0.9052; current x 4-ONE interaction: P = 0.7846; two-way repeated-measures ANOVA). d-f, Summed intensity projection of a stack of 22 confocal image planes (axial spacing 0.7973 um) through the fan-shaped body of a fly carrying the Hk^Flag allele (d) and single confocal image planes through the somatic regions of flies carrying the Hk^Flag allele (e) or an unmodified Hk locus (f). Specimens were stained with anti-Flag antibody (left); native R23E10-GAL4-driven mCD8::GFP fluorescence (yellow) is overlaid on the anti-Flag channel (turquoise) on the right. Scale bars, 50 mm. For statistical details see Supplementary Table 1. Source Data Full size image Voltage changes clear the redox memory The stability of cofactor binding suggests that each conversion of K [V]b to the NADP^+-bound state leaves an imprint lasting many minutes (Fig. 4c,d). We equate this imprint--or, more accurately, the imprint on the oxidation state of the Hyperkinetic pool of a dFBN as a whole--with a log of accumulated sleep pressure (Fig. 4g). As in a digital recording, the binary states of many elementary memory cells thus quantize a continuous variable, with a resolution determined by the number of single-bit units. Because sleep pressure is discharged via the electrical activity of dFBNs^16,17, action potentials should erase this memory by releasing NADP^+ and allowing its replacement with NADPH, whose concentration in the cytoplasm exceeds that of NADP ^+ by at least 40-fold^46. Such a mechanism would confirm a long-suspected quirk in the enzymatic cycle of K[V]b and offer a rationale for the association of the protein with a voltage-gated ion channel^8,27. We tested the prediction that cofactor exchange is voltage-controlled in both of our experimental configurations, using either the photogeneration of ROS by miniSOG (Fig. 6a,b and Extended Data Fig. 6a) or the inclusion of 50 uM 4-ONE in the intracellular solution (Fig. 6c,d and Extended Data Fig. 6b) to load K[V]b with NADP^+. Following the expected increases of the fast and slow inactivation time constants at 10 min after break-in, dFBNs were taken through simulated 20-min spike trains at 10 Hz under voltage clamp, with each 'action potential' consisting of a 3-ms somatic depolarization to +10 mV. Measurements of t[fast] and t[slow] after this sequence of voltage steps (that is, at 30 min after break-in) showed full reversals of the initial increases driven by miniSOG or 4-ONE (Fig. 6a-d). These reversals were themselves reversible: when dFBNs filled with 4-ONE were held at -80 mV for a further 10 min, the large surplus of 4-ONE in the patch pipette once again drove increases in both inactivation time constants (Fig. 6d), whereas a second 9-min light exposure accomplished the same for miniSOG-expressing cells (Fig. 6b). Fig. 6: Membrane depolarization clears the lipid peroxidation memory. figure 6 a,b, dFBNs expressing miniSOG were held at -80 mV in the intervals of 0-10 and 30-40 min (except during the voltage protocols required to measure I[A]) and repeatedly step-depolarized to +10 mV (3 ms, 10 Hz) between 10 and 30 min. Nine-minute exposures to blue light (between the 0- and 10-min and the 30- and 40-min time points) increase the fast and slow inactivation time constants of I[A] above their pre-illumination baselines (b; blue versus grey shading); a series of depolarization steps between 10 and 30 min reverses this increase (b; yellow shading; t[fast]: P < 0.0001; t[slow]: P = 0.0008; mixed-effects model; examples of peak-normalized I[A] evoked in the same dFBN by voltage steps to +30 mV in a). c,d, dFBNs were held at -80 mV in the intervals of 0-10 and 30-40 min (except during the voltage protocols required to measure I[A]) and repeatedly step-depolarized to +10 mV (3 ms, 10 Hz) between 10 and 30 min. The inclusion of 50 uM 4-ONE in the intracellular solution increases the fast and slow inactivation time constants of I[A] above the baselines recorded immediately after break-in (d; turquoise versus grey shading); a series of depolarization steps between 10 and 30 min counteracts this increase despite the continuous presence of 4-ONE (d ; yellow shading; t[fast]: P = 0.0053; t[slow]: P = 0.0012; mixed-effects model; examples of peak-normalized I[A] evoked in the same dFBN by voltage steps to +30 mV in c). Columns show population averages; dots represent individual cells; n, number of cells; asterisks indicate significant differences (P < 0.05) relative to the 0-min time point in planned pairwise comparisons by Holm-Sidak test. For statistical details see Supplementary Table 1. Source Data Full size image Occasionally, the reversal protocol pushed the inactivation time constants below their original baselines, suggesting that depolarization dissipated not only the oxidative strain applied by 4-ONE or miniSOG but also the internally sourced pressure already integrated by the channel complex before the experiment began. Consistent with this idea, dFBNs expressing catalytically inactive^18 ,27,44 Hk(K289M), which cannot form a redox memory (Fig. 4g and Extended Data Fig. 5a,b), often exhibit the fastest-inactivating A-type currents at baseline^9 and no modulation by 4-ONE or subsequent voltage changes (Extended Data Fig. 5a,b). The ability to remember exposures to lipid peroxidation products is an intrinsic property of K[V]1 channels, shared by neurons other than dFBNs (Extended Data Fig. 7a-e) and present in mammals, with broad--although not limitless^19--carbonyl selectivity. When HEK-293 cells coexpressing mouse K[V]1.4 and K[V]b2 were incubated in extracellular medium containing 12 mM methylglyoxal, a membrane-permeable dicarbonyl that serves as an established substrate ^19 for K[V]b2, the fast and slow inactivation time constants of the reconstituted A-type current rose and remained durably elevated for 20 min after the removal of methylglyoxal (Extended Data Fig. 7f-h). As in dFBNs, the memory of the carbonyl exposure was retained if the membrane containing the K[V]1.4-b2 complex was clamped at -80 mV but forgotten during a simulated 20-min spike train at 10 Hz (Extended Data Fig. 7i,j). Discussion Our experiments suggest that K[V]b subunits are voltage-gated memories used by neurons and other excitable cells to keep score of lipid peroxidation events. Information is stored in the oxidation state of a nicotinamide molecule bound so tightly that it should perhaps be considered a prosthetic group rather than a cofactor, even though two steps in a stop-and-go redox reaction cycle--hydride transfer and nicotinamide exchange--are used to move data to and from memory. Definitive proof that peroxidized lipids or their breakdown products are endogenous K[V]b substrates would require their co-purification with the native ion channel--a formidable challenge not only because of the expected molecular heterogeneity of these substrates^10,11,12,13, but also because their binding to K[V]b may be much looser than that of NADP(H); in contrast to the nucleotide binding cleft, which resembles a locked vice, the active site appears wide open in the crystal structure^8. Our experiments also suggest, but do not prove beyond doubt, that sleep loss causes widespread lipid peroxidation in the brain. Definitive proof would require a demonstration that peroxidation products accumulate, rather than that polyunsaturated phospholipids are depleted, as we have shown. Most previous attempts to measure lipid peroxidation after sleep loss have focused on a single end product, malondialdehyde^10, and yielded variable results^47,48,49,50 , perhaps because the picture seen through the lens of malondialdehyde is incomplete^42 or because the assays used for its detection report tissue oxidizability during analysis rather than pre-existing levels of peroxidized lipids. Our own attempts to quantify endogenous 4-ONE after sleep deprivation were thwarted by the short half-life^10,13 of the molecule in tissue: while SMALDI-MSI could easily detect 100 uM 4-ONE in isolation, the signal vanished when the same quantity of standard was spiked onto a brain section full of endogenous carbonyl-reactive nucleophiles^10,13 and enzymes^ 38 (Extended Data Fig. 8a). Trace amounts of 4-ONE captured by Girard's reagent during the derivatization of rested, but not sleep-deprived, brains must reflect the oxidation of the undepleted PUFA pools of these samples in vitro because the sni^1 mutation, which would have raised 4-ONE levels in vivo^38,39, caused no discernible increase at the time of measurement (Extended Data Fig. 8b). While our interpretation of sleep pressure as mitochondrially determined^9,25 lipid peroxidation history demands that sleep-control neurons are equipped to sense and respond to this history, integral redox sensors are a general feature of K[V]1 (and also some K[V]4) channels^1,2,3,4,5 in virtually all neurons and many other electrically excitable cells. What could be the purpose of b-subunits in this wider context? Redox control of electrical activity may protect non-renewable cells with high respiratory capacity and extensive membrane systems--such as those of the brain and heart--from oxidative damage if the electron supply to their mitochondria surpasses the demands of ATP synthesis^25. Depending on where this relief valve opens, the consequences may range from a few extraneous action potentials^28 (in order to re-balance energy consumption with mitochondrial electron flux) to the induction of sleep^9. Just as sodium spikes are universal information carriers filled with distinctive meaning by the different neurons that emit them, excitability control by K[V]b may be a general mechanism co-opted by dFBNs for the special purpose of regulating sleep. Methods Drosophila strains and culture Flies were reared on media of cornmeal (62.5 g l^-1), inactive yeast powder (25 g l^-1), agar (6.75 g l^-1), molasses (37.5 ml l^-1), propionic acid (4.2 ml l^-1), tegosept (1.4 g l^-1) and ethanol (7 ml l^-1) on a 12 h light:12 h dark cycle at 25 degC. All electrophysiological and lipidomic analyses (with the exception of studies of the effects of the sni^1 mutation) were performed on randomly selected female flies aged 2-6 days post eclosion. Experimental flies were heterozygous for all transgenes and homozygous for either a wild-type or mutant (Hk^1) Hyperkinetic allele^51,52, as stated. The R23E10-GAL4 driver^16,53 controlled the expression of the fluorescent label mCD8::GFP in dFBNs, along with an N-myristoylated covalent hexamer (myr-MS6T2) of the singlet oxygen generator miniSOG^54 or catalytically defective (Hk(K289M)) or functional versions of Hyperkinetic^44, as indicated. The Dh31-GAL4 line^55 targeted mCD8::GFP to neurons of the pars intercerebralis. Because sni is X-linked^38, it was most expedient to investigate its function in males. In behavioural experiments or 4-ONE analyses, hemizygous carriers of the sni^1 allele coexpressed UAS-sni^38, UAS-AOX^56 or UAS-Hk^RNAi (47805GD)^57 transgenes, either pan-neuronally^58 under the control of nSyb-GAL4 or in dFBNs^16,53 under the control of R23E10-GAL4, as noted. For electrophysiological recordings, dFBNs of hemizygous sni^1 mutants and wild-type males were marked with R23E10-GAL4-driven mCD8::GFP. A Hyperkinetic allele encoding an in-frame fusion to an N-terminal Flag epitope (Hk^Flag) was created through homology-dependent repair of a CRISPR-Cas9-generated double-strand break (WellGenetics). The Flag tag was inserted immediately after the initiating methionine of isoforms Hk-PK, Hk-PE, Hk-PL, and Hk-PM and connected to the remainder of the protein via a flexible linker (4x Gly-Gly-Ser). Sleep measurements and sleep deprivation Females or hemizygous sni^1 mutant males^38 aged 2-5 days were individually inserted into 65-mm glass tubes, loaded into Drosophila Activity Monitors (Trikinetics), and housed under 12 h light:12 h dark conditions. Flies were allowed to adapt to the monitors for a day, and the activity counts during the following two 24-h periods were averaged. Inactivity periods of >5 min were classified as sleep^ 59,60 (Sleep and Circadian Analysis MATLAB program^61). Immobile flies (<2 beam breaks per 24 h) were manually excluded. To deprive flies of sleep, a spring-loaded platform stacked with Trikinetics monitors was slowly tilted by an electric motor, released, and allowed to snap back to its original position^62. The mechanical cycles lasted 10 s and were repeated continuously for 12 h, beginning at zeitgeber time 12. SMALDI mass spectrometry imaging Dissected brains of rested and sleep-deprived flies were placed on PTFE-printed glass slides (Electron Microscopy Sciences), covered with ~3-5 ul gelatin (5% w/v in water), and snap frozen for shipping. For sectioning, dissected brains were thawed, suspended in 20 ul 5% gelatin, and transferred to a gelatin plateau created by removing the top half of a frozen block of 5% gelatin in a cryostat (Microm HM 525, ThermoFisher). After allowing the samples to refreeze during 10 min in the cryostat chamber, 10-um sections were cut and thaw-mounted onto glass slides. The sections were imaged in fluorescence (BX41, Olympus) and reflected light mode (VHX 5000, Keyence) and stored at -80 degC until further use. For SMALDI-MSI^63, the brain sections were thawed in a desiccator and spray-coated with 80 ul of a freshly prepared solution of 2,5-dihydroxybenzoic acid (DHB, Merck) using a SMALDIPrep ultrafine pneumatic spraying system (TransMIT GmbH). The DHB solution contained 60 mg of DHB in 999 ul acetone, 999 ul water, and 2 ul pure trifluoroacetic acid (TFA, Merck). In samples destined for 4-ONE analysis, a chemical derivatization step with Girard's reagent T (GirT, TCI Chemicals) preceded the application of the DHB matrix^64. The samples were spray-coated with 35 ul of a freshly prepared solution of 15 mg ml^-1 GirT in a 7:3 mixture of methanol and water containing 0.2% (v/v) TFA and incubated in a desiccator at room temperature for 2 h. Standards were prepared by applying 5-ul droplets of a tenfold dilution series of 4-ONE (Cayman Chemical) in methyl acetate, from 100 uM to 10 nM, onto blank glass slides or slides containing brain sections of rested flies. Standards underwent the same GirT-derivatization and matrix application steps as analytical samples. A home-built SMALDI-MS imaging ion source based on an AP-SMALDI^5 AF system (TransMIT GmbH) was coupled to an orbital trapping mass spectrometer (Q Exactive, ThermoFisher). Mass spectra were acquired at a mass resolution of 140,000 in positive-ion mode. A high voltage of 4 kV was applied to the sample holder. The standard pixel size of 5 um x 5 um in lipid analyses was increased to 25 um x 25 um for 4-ONE measurements to facilitate the detection of low-intensity signals. A single-ion-monitoring (SIM) experiment was performed first for 4-ONE, followed by a full MS scan. SMALDI-MS images were created in Mirion^65 (TransMIT GmbH) using a bin width of [?](m/z) = 0.004; the images were normalized to total ion charge^66. A digital mask created from a ubiquitous lipid signal was applied to the measurement area in order to exclude off-tissue pixels, and all images were stitched together in a single file to ensure uniform evaluation. An automatically generated list of all signals found in at least ten pixels in the stitched file was applied to the separate images to obtain the summed intensity of each signal. Signals were annotated in a bulk search against LIPID MAPS^67, allowing for [M + H]^+, [M+Na]^+, and [M + K]^+ adducts and selecting the most likely lipid(s) for each measured mass. All annotations with a mass deviation <5 ppm were exported for further validation in HPLC MS^2 fragmentation experiments. HPLC MS^2 fragmentation Approximately 1,300 rested and 1,300 sleep-deprived brains were collected in batches of 20-50 per session and snap frozen in plastic tubes. The frozen batches were combined in a glass Potter homogenizer, suspended in 50 ul ice-cold ammonium acetate (0.1% in water, Honeywell), manually homogenized, and transferred to a pre-cleaned Eppendorf tube. Lipids were extracted with 600 ul ice-cold methyl tert-butyl ether (MTBE, Sigma-Aldrich) and 150 ul methanol (VWR). After shaking the mixture for 1 h at 4 degC, 200 ul water (VWR) was added, the mixture was shaken for another 10 min, and the organic phase was collected after centrifugation for 5 min at 1,000g. The aqueous phase was re-extracted using an additional 400 ul MTBE, 120 ul methanol, and 100 ul water. The organic phases from both extraction steps were combined, and the solvent was evaporated under a stream of nitrogen for 30 min, leaving ~700 ug and ~800 ug of dry extract of rested and sleep-deprived samples, respectively. The extracts were stored at -80 degC until further use. An extraction blank was created by performing these steps without brain tissue. Lipid extracts were thawed, dissolved in 650 ul acetonitrile, 300 ul isopropanol, and 50 ul water (all VWR) in an ultrasonic bath, and separated on a C18 column (100 mm x 2.1 mm, 2.6 um particle size, 100 A pore size; Phenomenex) in an UltiMate 3000 Rapid Separation System (ThermoFisher) coupled to an orbital trapping mass spectrometer (Q Exactive HF-X, ThermoFisher) using a heated electrospray ionization source (HESI II, ThermoFisher). Data-dependent acquisition and MS^2 fragmentation experiments were based on the inclusion list obtained from SMALDI-MSI annotations, with [M + H] ^+, [M+Na] ^+, [M + K] ^+ and [M + NH[4]]^+ adducts in positive-ion mode. Since the ionization mechanisms of MALDI and electrospray MS differ, MS^2 fragmentation of lipid extracts was additionally performed in negative-ion mode, considering [M-H]^- and [M + CHO[2]]^- adducts, to increase the molecular coverage of SMALDI-MSI hits. Lipids were identified using LipidMatch^68. All MS^ 2-verified lipid annotations were validated by accurate mass and the detection of all fatty acids plus the head group. Only one annotation (PE 27:2) was based on accurate mass and head group alone. Electrophysiology Adult flies aged 2-6 days post eclosion were head-fixed to a custom mount using eicosane (Sigma). Cuticle, trachea, excess adipose tissue, and the perineural sheath were removed to create a small window, and the brain was continuously superfused with extracellular solution equilibrated with 95% O[2]-5% CO[2] and containing (in mM) 103 NaCl, 3 KCl, 5 TES, 8 trehalose, 10 glucose, 7 sucrose, 26 NaHCO [3], 1 NaH[2]PO[4], 1.5 CaCl[2,] 4 MgCl[2], pH 7.3, 275 mOsM. GFP-positive cells were visualized on a Zeiss Axioskop 2 FS mot microscope equipped with a 60x/1.0 NA water-immersion objective (LUMPLFLN60XW, Olympus) and a pE-300 white LED light source (CoolLED). Borosilicate glass electrodes (9-11 MO for dFBNs, 5-7 MO for neurons of the pars intercerebralis) were fabricated on a PC-10 micropipette puller (Narishige) or a DMZ Universal Electrode Puller (Zeitz) and filled with intracellular solution containing (in mM) 10 HEPES, 140 potassium aspartate, 1 KCl, 4 MgATP, 0.5 Na[3]GTP, 1 EGTA, pH 7.3, 265 mOsM. Where indicated, 50 uM 4-ONE or 200 uM 4-hydroxynonenal (4-HNE, Cayman Chemical) were added directly to the intracellular solution; in recordings from neurons of the pars intercerebralis, during which larger-diameter electrodes were used than in recordings from dFBNs, the 4-ONE concentration was lowered to 1 uM. Stock solutions of 4-ONE and 4-HNE were prepared in methyl acetate and ethanol, respectively; vehicle concentrations were not allowed to surpass 0.15% of the total volume after dilution. Recordings were obtained at room temperature with a MultiClamp 700B amplifier, lowpass-filtered at 10 kHz, and sampled at 20 or 50 kHz using Digidata 1440A or 1550B digitizers controlled through pCLAMP 10 or 11 (Molecular Devices). For photostimulation of miniSOG during whole-cell recordings^9, a 455-nm LED (Thorlabs M455L3) with a mounted collimator lens (Thorlabs ACP2520-A) and T-Cube LED Driver (Thorlabs) delivered 3.5-5 mW cm^-2 of optical power to the sample. Data were analysed using the NeuroMatic package^69 (http:// neuromatic.thinkrandom.com) in Igor Pro (WaveMetrics). Whole-cell capacitance compensation and bridge balance were used in voltage- and current-clamp recordings, respectively. Series resistances were monitored but not compensated and allowed to rise at most 20% above baseline--but never beyond 50 MO--during a recording. Uncompensated mean series resistances of ~40 MO in dFBNs (Extended Data Fig. 3c) caused predicted voltage errors of ~16 mV at typical I [A] amplitudes of ~400 pA (Extended Data Figs. 3a,b, 4 and 6). Input resistances were calculated from linear fits of the steady-state voltage changes elicited by 1-s steps of hyperpolarizing current (5-pA increments) from a pre-pulse potential of -60 +- 5 mV. Membrane time constants were estimated by fitting a single exponential to the voltage deflection caused by a hyperpolarizing 5-pA current step lasting 200 ms. Voltage-spike frequency functions were determined from voltage responses to a series of depolarizing current steps from a membrane potential of -60 +- 5 mV. To account for variations in input resistance within the dFBN population, the current required to produce a 5-mV hyperpolarizing voltage deflection from a pre-pulse potential of -60 +- 5 mV was used as a cell-specific unitary current step instead of a static 5-pA increment. Spikes were detected by finding minima in the time derivative of the membrane potential trace. Voltage-clamp experiments on dFBNs and neurons of the pars intercerebralis were performed in the presence of 1 uM tetrodotoxin (Tocris) and 200 uM cadmium to block sodium and calcium currents, respectively. Potassium currents were measured by stepping neurons from holding potentials of -10 or -110 mV for 400 ms to a series of test potentials spanning the range from -100 mV to +30 mV in 10-mV increments^9,24. Depolarizations from -110 mV produced the sum total of the cell's potassium currents (I[total], Extended Data Fig. 2a), whereas currents evoked by voltage steps from a holding potential of -10 mV lacked the I[A] (A-type or fast outward) component because voltage-gated potassium channels such as Shaker inactivated (Extended Data Fig. 2b). I[A] was calculated by subtracting this non-A-type component from I[total] (Extended Data Fig. 2c). To determine the fast and slow inactivation time constants^9, double-exponential functions were fit to the decaying phase of A-type currents elicited by 400-ms steps to +30 mV (Extended Data Fig. 2d). In cases where the fits of slow inactivation time constants were poorly constrained, only the fast inactivation time constants were included in the analysis. Spiking was simulated by 3-ms depolarizing pulses to +10 mV, repeated at 10 Hz for 20 min. Steady-state activation parameters were determined by applying depolarizing 400-ms voltage pulses from holding potentials of -10 or -110 mV; the pulses covered the range from -60 to +60 mV in steps of 10 mV. Linear leak currents were estimated from the slope of the current-voltage relationship at hyperpolarized potentials and subtracted. Steady-state inactivation parameters were obtained with the help of a two-pulse protocol, in which a 300-ms pre-pulse (-120 to +60 mV in 10-mV increments) was followed by a 400-ms test pulse to +30 mV; non-inactivating outward currents, measured from a pre-pulse potential of +10 mV, were subtracted. Peak A-type currents (I[A]) were normalized to the maximum current amplitude (I[max]) of the respective cell and plotted against the test or pre-pulse potentials (V). An estimated liquid junction potential^70 of 16.1 mV was subtracted post hoc. Curves were fit to the Boltzmann function \({I}_ {{\rm{A}}}/{I}_{\max }=1/\left(1+{{\rm{e}}}^{\frac{V-{V}_{0.5}}{k}}\ right)\) to determine the half-maximal activation and inactivation voltages (V[0.5]) and slope factors (k). HEK-293 cells (CRL-1573, American Type Culture Collection) were grown at 37 degC under 5% CO[2] in Dulbecco's modified Eagle's medium (DMEM) with 10% (v/v) fetal bovine serum and 100 U ml^-1 penicillin plus 100 ug ml^-1 streptomycin (ThermoFisher). The cells were neither externally authenticated nor routinely tested for mycoplasma contamination. Cells were transfected (Lipofectamine 3000, ThermoFisher) with a 1:1 mixture of CMV promoter-driven expression vectors encoding mouse K[V]1.4 and a bicistronic mouse K[V] b2-IRES2-EGFP cassette. A carbonyl-reactive residue^71 (Cys-13) in the N-terminal inactivation peptide of K[V]1.4 was mutated to serine. The growth medium was replaced during whole-cell recordings with extracellular solution containing (in mM) 10 HEPES, 140 NaCl, 5 KCl, 10 glucose, 2 CaCl[2], 1 MgCl[2], pH 7.4. Where indicated, HEK-293 cells were pre-incubated in extracellular solution supplemented with 12 mM methylglyoxal^19 for 1 h, followed by three washes with methylglyoxal-free solution, before data acquisition. GFP-positive cells were visually targeted with borosilicate glass electrodes (2-3 MO) filled with intracellular solution containing (in mM) 10 HEPES, 80 potassium aspartate, 60 KCl, 10 glucose, 2 MgATP, 1 MgCl [2], 5 EGTA, pH 7.3. Signals were acquired at room temperature with a MultiClamp 700B amplifier, lowpass-filtered at 10 kHz, and sampled at 20 kHz using a Digidata 1440 A digitizer controlled through pCLAMP 10 (Molecular Devices). Because untransfected HEK-293 cells lack voltage-gated conductances (Extended Data Fig. 7f), no channel blockers were present. To determine the fast and slow inactivation time constants, double-exponential functions were fit to the decaying phase of A-type currents elicited by 1-s steps to +30 mV. Spiking was simulated by 3-ms depolarizing pulses to +10 mV, repeated at 10 Hz for 20 min. Data were analysed using the NeuroMatic package^69 (http: //neuromatic.thinkrandom.com) in Igor Pro (WaveMetrics). Confocal imaging Dissected brains were fixed for 20 min in PBS with 4% (w/v) paraformaldehyde, washed 3 times for 20 min with 0.5% (v/v) Triton X-100 in PBS (PBST), and incubated sequentially at 4 degC in blocking solution (10% goat serum in PBST) overnight, with mouse monoclonal anti-Flag M2 antibodies (1:500, Sigma) in blocking solution for 2 days, and with goat anti-Mouse IgG Alexa Fluor 633 antibodies (1:500, ThermoFisher) for one day. The samples were washed 5 times with blocking solution before and after the addition of the secondary antibody, mounted in Vectashield, and imaged on a Leica TCS SP5 confocal microscope with an HCX IRAPO L 25x/0.95 water-immersion objective. Statistics and reproducibility With the exception of sleep measurements, no statistical methods were used to predetermine sample sizes. Flies of the indicated genotype, sex and age were selected randomly for analysis and assigned randomly to treatment groups if treatments were applied (for example, sleep deprivation). The investigators were not blinded to group allocation. SMALDI-MSI signal intensities were analysed in LipidSig^72 and MATLAB (The MathWorks). Global differences between normalized glycerophospholipid intensities in cryosections of rested and sleep-deprived brains were evaluated by multiple t-tests with FDR-adjusted P < 0.05, using the method of Benjamini-Hochberg. Statistical associations with sleep history of user-defined lipid features, such as the indicated double-bond equivalent ranges or phospholipid head groups, were computed by Fisher's exact test in LipidSig^72. Principal component and hierarchical cluster analyses were performed in MATLAB. The list of significantly different signals was exported and re-imported into Mirion to generate SMALDI-MS images for display. Behavioural and electrophysiological data were analysed in Prism 10 (GraphPad). All null hypothesis tests were two-sided. To control type I errors, P values were adjusted to achieve a joint a of 0.05 at each level in a hypothesis hierarchy; multiplicity adjusted P values are reported in cases of multiple comparisons at one level. Group means or their time courses were compared by paired t-test, one- or two-way repeated-measures ANOVA, or mixed-effects models in cases where a variable was not measured in all cells at all time points, as indicated in figure legends. Repeated-measures ANOVA and mixed-effect models used the Geisser-Greenhouse correction in all instances except the comparisons of >2 genotypes in Fig. 3a,b and Extended Data Fig. 1a and were followed by planned pairwise analyses with Holm-Sidak's multiple comparisons test. Where the assumption of normality was violated (as indicated by D'Agostino-Pearson test), group means were compared by Mann-Whitney test, Wilcoxon test, Kruskal-Wallis ANOVA or Friedman test, followed by Dunn's multiple comparisons test to evaluate planned pairwise differences. Test statistics, degrees of freedom, and exact P values are given in Supplementary Tables 1 and 2 . Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Data availability The SMALDI-MSI and LC-MS^2 datasets are accessible in METASPACE ( https://metaspace2020.eu/project/drosophila and https:// metaspace2020.eu/project/drosophila4ONE) and the MassIVE repository ( ftp://massive.ucsd.edu/v05/MSV000091767/), respectively. All other data generated and analysed during this study are included in the Source Data. References 1. Rehm, H. & Lazdunski, M. Purification and subunit structure of a putative K^+-channel protein identified by its binding properties for dendrotoxin I. Proc. Natl Acad. Sci. USA 85, 4919-4923 (1988). Article CAS PubMed Google Scholar 2. Parcej, D. N., Scott, V. E. & Dolly, J. O. Oligomeric properties of a-dendrotoxin-sensitive potassium ion channels purified from bovine brain. Biochemistry 31, 11084-11088 (1992). Article CAS PubMed MATH Google Scholar 3. Nakahira, K., Shi, G., Rhodes, K. J. & Trimmer, J. S. Selective interaction of voltage-gated K^+ channel b-subunits with a-subunits. J. Biol. Chem. 271, 7084-7089 (1996). Article CAS PubMed Google Scholar 4. Long, S. B., Campbell, E. B. & MacKinnon, R. Crystal structure of a mammalian voltage-dependent Shaker family K^+ channel. Science 309, 897-903 (2005). Article CAS PubMed MATH Google Scholar 5. Pongs, O. & Schwarz, J. R. Ancillary subunits associated with voltage-dependent K^+ channels. Physiol. Rev. 90, 755-796 (2010). Article CAS PubMed MATH Google Scholar 6. McCormack, T. & McCormack, K. Shaker K^+ channel b subunits belong to an NAD(P)H-dependent oxidoreductase superfamily. Cell 79, 1133-1135 (1994). Article CAS PubMed MATH Google Scholar 7. Chouinard, S. W., Wilson, G. F., Schlimgen, A. K. & Ganetzky, B. A potassium channel b subunit related to the aldo-keto reductase superfamily is encoded by the Drosophila Hyperkinetic locus. Proc. Natl Acad. Sci. USA 92, 6763-6767 (1995). Article CAS PubMed Google Scholar 8. Gulbis, J. M., Mann, S. & MacKinnon, R. Structure of a voltage-dependent K^+ channel b subunit. Cell 97, 943-952 (1999). Article CAS PubMed Google Scholar 9. Kempf, A., Song, S. M., Talbot, C. B. & Miesenbock, G. A potassium channel b-subunit couples mitochondrial electron transport to sleep. Nature 568, 230-234 (2019). Article CAS PubMed Google Scholar 10. Esterbauer, H., Schaur, R. J. & Zollner, H. Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic. Biol. Med. 11, 81-128 (1991). Article CAS PubMed Google Scholar 11. Yin, H., Xu, L. & Porter, N. A. Free radical lipid peroxidation: Mechanisms and analysis. Chem. Rev. 111, 5944-5972 (2011). Article CAS PubMed MATH Google Scholar 12. Ayala, A., Munoz, M. F. & Arguelles, S. Lipid peroxidation: Production, metabolism, and signaling mechanisms of malondialdehyde and 4-hydroxy-2-nonenal. Oxid. Med. Cell Longev. 2014, 360438 (2014). Article PubMed Google Scholar 13. Parvez, S., Long, M. J. C., Poganik, J. R. & Aye, Y. Redox signaling by reactive electrophiles and oxidants. Chem. Rev. 118, 8798-8888 (2018). Article CAS PubMed MATH Google Scholar 14. Shu, X. et al. A genetically encoded tag for correlated light and electron microscopy of intact cells, tissues, and organisms. PLoS Biol. 9, e1001041 (2011). Article CAS PubMed Google Scholar 15. Donlea, J. M., Thimgan, M. S., Suzuki, Y., Gottschalk, L. & Shaw, P. J. Inducing sleep by remote control facilitates memory consolidation in Drosophila. Science 332, 1571-1576 (2011). Article CAS PubMed Google Scholar 16. Donlea, J. M., Pimentel, D. & Miesenbock, G. Neuronal machinery of sleep homeostasis in Drosophila. Neuron 81, 860-872 (2014). Article CAS PubMed Google Scholar 17. Hasenhuetl, P. S. et al. A half-centre oscillator encodes sleep pressure. Preprint at BioRxiv https://doi.org/10.1101/ 2024.02.23.581780 (2024). 18. Weng, J., Cao, Y., Moss, N. & Zhou, M. Modulation of voltage-dependent Shaker family potassium channels by an aldo-keto reductase. J. Biol. Chem. 281, 15194-15200 (2006). Article CAS PubMed Google Scholar 19. Tipparaju, S. M., Barski, O. A., Srivastava, S. & Bhatnagar, A. Catalytic mechanism and substrate specificity of the b-subunit of the voltage-gated potassium channel. Biochemistry 47, 8840-8854 (2008). Article CAS PubMed MATH Google Scholar 20. Tempel, B. L., Papazian, D. M., Schwarz, T. L., Jan, Y. N. & Jan, L. Y. Sequence of a probable potassium channel component encoded at Shaker locus of Drosophila. Science 237, 770-775 (1987). Article CAS PubMed MATH Google Scholar 21. Iverson, L. E., Tanouye, M. A., Lester, H. A., Davidson, N. & Rudy, B. A-type potassium channels expressed from Shaker locus cDNA. Proc. Natl Acad. Sci. USA 85, 5723-5727 (1988). Article CAS PubMed Google Scholar 22. Cirelli, C. et al. Reduced sleep in Drosophila Shaker mutants. Nature 434, 1087-1092 (2005). Article CAS PubMed MATH Google Scholar 23. Bushey, D., Huber, R., Tononi, G. & Cirelli, C. Drosophila Hyperkinetic mutants have reduced sleep and impaired memory. J. Neurosci. 27, 5384-5393 (2007). Article CAS PubMed Google Scholar 24. Pimentel, D. et al. Operation of a homeostatic sleep switch. Nature 536, 333-337 (2016). Article CAS PubMed MATH Google Scholar 25. Sarnataro, R., Velasco, C. D., Monaco, N., Kempf, A. & Miesenbock, G. Mitochondrial origins of the pressure to sleep. Preprint at BioRxiv https://doi.org/10.1101/2024.02.23.581770 (2024). 26. Bahring, R. et al. Coupling of voltage-dependent potassium channel inactivation and oxidoreductase active site of Kvb subunits. J. Biol. Chem. 276, 22923-22929 (2001). Article PubMed MATH Google Scholar 27. Pan, Y., Weng, J., Cao, Y., Bhosle, R. C. & Zhou, M. Functional coupling between the Kv1.1 channel and aldoketoreductase Kvb1. J. Biol. Chem. 283, 8634-8642 (2008). Article CAS PubMed Google Scholar 28. Chintaluri, C. & Vogels, T. P. Metabolically regulated spiking could serve neuronal energy homeostasis and protect from reactive oxygen species. Proc. Natl Acad. Sci. USA 120, e2306525120 (2023). Article CAS PubMed Google Scholar 29. Dennard, R. H. How we made DRAM. Nat. Electron. 1, 372 (2018). Article Google Scholar 30. Vaccaro, A. et al. Sleep loss can cause death through accumulation of reactive oxygen species in the gut. Cell 181, 1307-1328.e15 (2020). Article CAS PubMed MATH Google Scholar 31. Kent, C. Eukaryotic phospholipid biosynthesis. Annu. Rev. Biochem. 64, 315-343 (1995). Article CAS PubMed Google Scholar 32. Vance, J. E. Phospholipid synthesis and transport in mammalian cells. Traffic 16, 1-18 (2015). Article CAS PubMed MATH Google Scholar 33. Choi, S. Y. et al. A common lipid links Mfn-mediated mitochondrial fusion and SNARE-regulated exocytosis. Nat. Cell Biol. 8, 1255-1262 (2006). Article CAS PubMed MATH Google Scholar 34. Jendrach, M., Mai, S., Pohl, S., Voth, M. & Bereiter-Hahn, J. Short- and long-term alterations of mitochondrial morphology, dynamics and mtDNA after transient oxidative stress. Mitochondrion 8, 293-304 (2008). Article CAS PubMed Google Scholar 35. Fridovich, I. Oxygen toxicity: a radical explanation. J. Exp. Biol. 201, 1203-1209 (1998). Article CAS PubMed MATH Google Scholar 36. Doorn, J. A., Maser, E., Blum, A., Claffey, D. J. & Petersen, D. R. Human carbonyl reductase catalyzes reduction of 4-oxonon-2-enal. Biochemistry 43, 13106-13114 (2004). Article CAS PubMed Google Scholar 37. Oppermann, U. Carbonyl reductases: the complex relationships of mammalian carbonyl- and quinone-reducing enzymes and their role in physiology. Annu. Rev. Pharmacol. Toxicol. 47, 293-322 (2007). Article CAS PubMed MATH Google Scholar 38. Botella, J. A. et al. The Drosophila carbonyl reductase sniffer prevents oxidative stress-induced neurodegeneration. Curr. Biol. 14, 782-786 (2004). Article CAS PubMed MATH Google Scholar 39. Martin, H. J. et al. The Drosophila carbonyl reductase sniffer is an efficient 4-oxonon-2-enal (4ONE) reductase. Chem. Biol. Interact. 191, 48-54 (2011). Article CAS PubMed Google Scholar 40. Hill, V. M. et al. A bidirectional relationship between sleep and oxidative stress in Drosophila. PLoS Biol. 16, e2005206 (2018). Article PubMed Google Scholar 41. Maxwell, D. P., Wang, Y. & McIntosh, L. The alternative oxidase lowers mitochondrial reactive oxygen production in plant cells. Proc. Natl Acad. Sci. USA 96, 8271-8276 (1999). Article CAS PubMed MATH Google Scholar 42. Murphy, M. P. et al. Guidelines for measuring reactive oxygen species and oxidative damage in cells and in vivo. Nat. Metab. 4, 651-662 (2022). Article PubMed MATH Google Scholar 43. Hattori, S., Murakami, F. & Song, W. J. Rundown of a transient potassium current is attributable to changes in channel voltage dependence. Synapse 48, 57-65 (2003). Article CAS PubMed MATH Google Scholar 44. Fogle, K. J. et al. CRYPTOCHROME-mediated phototransduction by modulation of the potassium ion channel b-subunit redox sensor. Proc. Natl Acad. Sci. USA 112, 2245-2250 (2015). Article CAS PubMed MATH Google Scholar 45. Bar-Yehuda, D. & Korngreen, A. Space-clamp problems when voltage clamping neurons expressing voltage-gated conductances. J. Neurophysiol. 99, 1127-1136 (2008). Article PubMed Google Scholar 46. Sallin, O. et al. Semisynthetic biosensors for mapping cellular concentrations of nicotinamide adenine dinucleotides. eLife 7, e32638 (2018). Article PubMed Google Scholar 47. Gopalakrishnan, A., Ji, L. L. & Cirelli, C. Sleep deprivation and cellular responses to oxidative stress. Sleep 27, 27-35 (2004). Article PubMed Google Scholar 48. Silva, R. H. et al. Role of hippocampal oxidative stress in memory deficits induced by sleep deprivation in mice. Neuropharmacology 46, 895-903 (2004). Article CAS PubMed MATH Google Scholar 49. Bellesi, M. et al. Sleep loss promotes astrocytic phagocytosis and microglial activation in mouse cerebral cortex. J. Neurosci. 37, 5263-5273 (2017). Article CAS PubMed MATH Google Scholar 50. Haynes, P. R. et al. A neuron-glia lipid metabolic cycle couples daily sleep to mitochondrial homeostasis. Nat. Neurosci. 27, 666-678 (2024). Article CAS PubMed MATH Google Scholar 51. Kaplan, W. D. & Trout, W. E. The behavior of four neurological mutants of Drosophila. Genetics 61, 399-409 (1969). Article CAS PubMed MATH Google Scholar 52. Stern, M. & Ganetzky, B. Altered synaptic transmission in Drosophila Hyperkinetic mutants. J. Neurogenet. 5, 215-228 (1989). Article CAS PubMed MATH Google Scholar 53. Jenett, A. et al. A GAL4-driver line resource for Drosophila neurobiology. Cell Rep. 2, 991-1001 (2012). Article CAS PubMed MATH Google Scholar 54. Ng, J. et al. Genetically targeted 3D visualisation of Drosophila neurons under electron microscopy and x-ray microscopy using miniSOG. Sci. Rep. 6, 38863 (2016). Article CAS PubMed Google Scholar 55. Tayler, T. D., Pacheco, D. A., Hergarden, A. C., Murthy, M. & Anderson, D. J. A neuropeptide circuit that coordinates sperm transfer and copulation duration in Drosophila. Proc. Natl Acad. Sci. USA 109, 20697-20702 (2012). Article CAS PubMed Google Scholar 56. Fernandez-Ayala, D. J. M. et al. Expression of the Ciona intestinalis alternative oxidase (AOX) in Drosophila complements defects in mitochondrial oxidative phosphorylation. Cell Metab. 9 , 449-460 (2009). Article CAS PubMed Google Scholar 57. Dietzl, G. et al. A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature 448, 151-156 (2007). Article CAS PubMed Google Scholar 58. Simpson, J. H. Rationally subdividing the fly nervous system with versatile expression reagents. J. Neurogenet. 30, 185-194 (2016). Article CAS PubMed MATH Google Scholar 59. Hendricks, J. C. et al. Rest in Drosophila is a sleep-like state. Neuron 25, 129-138 (2000). Article CAS PubMed MATH Google Scholar 60. Shaw, P. J., Cirelli, C., Greenspan, R. J. & Tononi, G. Correlates of sleep and waking in Drosophila melanogaster. Science 287, 1834-1837 (2000). Article CAS PubMed Google Scholar 61. Vecsey, C. G., Koochagian, C., Porter, M. T., Roman, G. & Sitaraman, D. Analysis of sleep and circadian rhythms from Drosophila activity-monitoring data using SCAMP. Cold Spring Harb. Protoc. 2024, pdb.prot108182 (2024). Article PubMed Google Scholar 62. Shaw, P. J., Tononi, G., Greenspan, R. J. & Robinson, D. F. Stress response genes protect against lethal effects of sleep deprivation in Drosophila. Nature 417, 287-291 (2002). Article CAS PubMed MATH Google Scholar 63. Spengler, B. & Hubert, M. Scanning microprobe matrix-assisted laser desorption ionization (SMALDI) mass spectrometry: Instrumentation for sub-micrometer resolved LDI and MALDI surface analysis. J. Am. Soc. Mass. Spectrom. 13, 735-748 (2002). Article CAS PubMed Google Scholar 64. Dreisbach, D., Heiles, S., Bhandari, D. R., Petschenka, G. & Spengler, B. Molecular networking and on-tissue chemical derivatization for enhanced identification and visualization of steroid glycosides by MALDI mass spectrometry imaging. Anal. Chem. 94, 15971-15979 (2022). Article CAS PubMed Google Scholar 65. Paschke, C. et al. Mirion--a software package for automatic processing of mass spectrometric images. J. Am. Soc. Mass. Spectrom. 24, 1296-1306 (2013). Article CAS PubMed MATH Google Scholar 66. Muller, M. A., Kompauer, M., Strupat, K., Heiles, S. & Spengler, B. Implementation of a high-repetition-rate laser in an AP-SMALDI MSI system for enhanced measurement performance. J. Am. Soc. Mass. Spectrom. 32, 465-472 (2021). Article PubMed Google Scholar 67. Sud, M. et al. LMSD: LIPID MAPS structure database. Nucleic Acids Res. 35, D527-D532 (2007). Article CAS PubMed Google Scholar 68. Koelmel, J. P. et al. LipidMatch: an automated workflow for rule-based lipid identification using untargeted high-resolution tandem mass spectrometry data. BMC Bioinformatics 18, 331 (2017). Article PubMed Google Scholar 69. Rothman, J. S. & Silver, R. A. NeuroMatic: an integrated open-source software toolkit for acquisition, analysis and simulation of electrophysiological data. Front. Neuroinformatics 12, 14 (2018). Article MATH Google Scholar 70. Neher, E. Correction for liquid junction potentials in patch clamp experiments. Methods Enzymol. 207, 123-131 (1992). Article CAS PubMed MATH Google Scholar 71. Ruppersberg, J. P. et al. Regulation of fast inactivation of cloned mammalian I[K](a) channels by cysteine oxidation. Nature 352, 711-714 (1991). Article CAS PubMed Google Scholar 72. Lin, W. J. et al. LipidSig: a web-based tool for lipidomic data analysis. Nucleic Acids Res. 49, W336-W345 (2021). Article PubMed Google Scholar Download references Acknowledgements The authors thank L. Ballenberger and C. Hartmann for help with dissections and D. Anderson, B. Dickson, B. Ganetzky, T. Holmes, H. Jacobs, J. Ng, G. Rubin, S. Schneuwly, J. Simpson, the Bloomington Stock Center and the Vienna Drosophila Resource Center for flies. This work was supported by grants from the European Research Council (832467) and the UK Medical Research Council (MR/V013238/1) to G.M., and from the German Research Foundation (Sp314/23-1, INST 162/500-1 FUGG) and the Hessian Ministry of Science and Education (LOEWE Center DRUID) to B.S.; H.O.R. and L.G.S. received doctoral training fellowships from Wellcome and La Caixa, respectively; M.A.M. was supported by a Kekule fellowship from the German Fonds der Chemischen Industrie; P.Z.L. was a Marshall Scholar; and A.K. held postdoctoral fellowships from the Swiss National Science Foundation and EMBO. Author information Author notes 1. Anissa Kempf Present address: Biozentrum, Universitat Basel, Basel, Switzerland 2. These authors contributed equally: H. Olof Rorsman, Max A. Muller Authors and Affiliations 1. Centre for Neural Circuits and Behaviour, University of Oxford, Oxford, UK H. Olof Rorsman, Patrick Z. Liu, Laura Garmendia Sanchez, Anissa Kempf & Gero Miesenbock 2. Institute of Inorganic and Analytical Chemistry, Justus-Liebig-Universitat, Giessen, Germany Max A. Muller, Stefanie Gerbig & Bernhard Spengler Authors 1. H. Olof Rorsman View author publications You can also search for this author inPubMed Google Scholar 2. Max A. Muller View author publications You can also search for this author inPubMed Google Scholar 3. Patrick Z. Liu View author publications You can also search for this author inPubMed Google Scholar 4. Laura Garmendia Sanchez View author publications You can also search for this author inPubMed Google Scholar 5. Anissa Kempf View author publications You can also search for this author inPubMed Google Scholar 6. Stefanie Gerbig View author publications You can also search for this author inPubMed Google Scholar 7. Bernhard Spengler View author publications You can also search for this author inPubMed Google Scholar 8. Gero Miesenbock View author publications You can also search for this author inPubMed Google Scholar Contributions H.O.R. performed all electrophysiological and behavioural experiments on flies and M.A.M. performed all lipidomic analyses, under the supervision of S.G. and B.S., on material prepared by L.G.S., H.O.R. and A.K. P.Z.L. characterized Hk^Flag flies and K[V] currents in HEK-293 cells. B.S. designed the SMALDI-MSI methodology and instrumentation. G.M. devised and directed the research and wrote the paper. Corresponding author Correspondence to Gero Miesenbock. Ethics declarations Competing interests M.A.M. and S.G. are employees of and B.S. is a consultant for TransMIT GmbH. The other authors declare no competing interests. Peer review Peer review information Nature thanks Leslie Griffith, Chun-Fang Wu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer review reports are available. Additional information Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Extended data figures and tables Extended Data Fig. 1 Sleep architecture and waking locomotor activity of sni^1 mutants. a, The nSyb-GAL4- or R23E10-GAL4-driven overexpression of sniffer (over)corrects the altered sleep profile of hemizygous sni^1 mutant males (P < 0.0001 for all pairwise comparisons, two-way repeated-measures ANOVA with Holm-Sidak test; sample sizes in b). b, The average sleep bout duration in hemizygous sni^1 mutant males differs from wild-type (P < 0.0001; Kruskal-Wallis ANOVA with Dunn's test) but returns to control level if carriers also express sniffer or AOX pan-neuronally under the control of nSyb-GAL4 (sni: P > 0.9999; AOX: P > 0.9999) or sniffer, AOX, or Hk^RNAi in dFBNs under the control of R23E10-GAL4 (sni: P > 0.9999; AOX: P = 0.1462; Hk^ RNAi: P > 0.9999). The average sleep bout durations of 6 sni^1 mutants exceeding 360 min are plotted at the top of the graph; mean and s.e.m. are based on the actual values. c, The number of sleep bouts in hemizygous sni^1 mutant males differs from wild-type (P < 0.0001; Kruskal-Wallis ANOVA with Dunn's test) but returns to control level if carriers also express sniffer or AOX pan-neuronally under the control of nSyb-GAL4 (sni: P > 0.9999; AOX: P > 0.9999) or sniffer, AOX, or Hk^RNAi in dFBNs under the control of R23E10-GAL4 (sni: P > 0.9999; AOX: P = 0.1492; Hk^RNAi: P > 0.9999). d, Hemizygous sni^1 mutant males show elevated waking locomotor activity relative to wild-type (P < 0.0001; Kruskal-Wallis ANOVA with Dunn's test). Data are means +- s.e.m.; n, number of flies; asterisks, significant differences (P < 0.05) from wild-type in planned pairwise comparisons. For statistical details see Supplementary Table 2. Source Data Extended Data Fig. 2 Measurement of the inactivation time constants of I[A]. a, Voltage steps from a holding potential of -110 mV (top) elicit the full complement of potassium currents in a dFBN (I[total], bottom). b , Stepping the same neuron from a holding potential of -10 mV (top) elicits potassium currents lacking the A-type component (I[non-A], bottom). c, Digital subtraction of I[non-A] (b, bottom) from I[total] (a, bottom) yields I[A]. Note the expanded timescale. d, Estimates of t[fast] and t[slow] are obtained from a double-exponential fit (red line) to the A-type current evoked by step depolarization to +30 mV. Extended Data Fig. 3 Inactivation kinetics and amplitudes of potassium currents, series resistances, and steady-state activation and inactivation curves of I[A] during the course of a 30-minute recording. a, dFBNs were held at -80 mV, except during the voltage protocols required to measure I[A]. In the absence of 4-ONE, the fast and slow inactivation time constants of I[A] (t[fast]: P = 0.2499; t[slow]: P = 0.5968; mixed-effects model), the amplitude of I[non-A] (P = 0.3527; mixed-effects model), input resistance (P = 0.6543; mixed-effects model), and membrane time constant (P = 0.5196; mixed-effects model) remain unchanged, but the amplitude of I[A] runs down during the course of the recording (P = 0.0004; mixed-effects model). Columns, population averages; dots, individual cells; n, number of cells; asterisks, significant differences (P < 0.05) relative to baseline in planned pairwise comparisons. b, c, dFBNs were held at -80 mV in the interval of 0-10 min (except during the voltage protocols required to measure I[A]) and repeatedly step-depolarized to +10 mV (3 ms, 10 Hz) between 10 and 30 min. The inclusion of 50 uM 4-ONE in the intracellular solution increases the fast and slow inactivation time constants of I[A] above the baselines recorded immediately after break-in (b, turquoise vs. grey shading); a series of depolarization steps between 10 and 30 min counteracts this increase despite the continuous presence of 4-ONE (b, yellow shading; t[fast]: P = 0.0054; t[slow]: P = 0.0014; mixed-effects model). The amplitude of I[A] runs down during the course of the recording (b, P = 0.0008; mixed-effects model); I[non-A] (b, P = 0.3120; mixed-effects model), input resistance (b, P = 0.4961; mixed-effects model), and membrane time constant (b, P = 0.2282; mixed-effects model) remain unchanged. Series resistance increases gradually (c, P = 0.0399; mixed-effects model) but remains within <20% of baseline and below 50 MO. Columns, population averages; dots, individual cells; n, number of cells; asterisks, significant differences (P < 0.05) relative to the 0-minute time point in planned pairwise comparisons by Holm-Sidak test. d, Steady-state activation and inactivation curves of I[A] in dFBNs immediately after break-in (0 min), after 10 min of dialysis with intracellular solution containing 50 uM 4-ONE (turquoise), and after a series of depolarization steps to +10 mV (3 ms, 10 Hz) between 10 and 30 min (yellow). Data are means +- s.e.m; solid lines, Boltzmann fits. The half-activation voltages and activation slope factors are identical at all time points (P = 0.5378, F test) but the half-inactivation voltages and inactivation slope factors differ (P < 0.0001, F test). For statistical details see Supplementary Table 2. Source Data Extended Data Fig. 4 Potassium current amplitudes and membrane properties of dFBNs in Fig. 4. a, dFBNs of hemizygous sni^1 mutant (turquoise) and wild-type males (grey) do not differ with respect to the amplitudes of I[A] (P = 0.4023, two-sided Mann-Whitney test) and I[non-A] (P = 0.6276, two-sided t-test), input resistance (P = 0.3014, two-sided t-test), and membrane time constant (P = 0.5267, two-sided Mann-Whitney test). b, dFBNs expressing miniSOG were held at -80 mV, except during the voltage protocols required to measure I[A], and exposed to blue light between the 0- and 10-minute time points (blue shading). The amplitude of I[A] runs down during the course of the recording (P = 0.0003; Friedman test); I[non-A] (P = 0.1116; Friedman test), input resistance (P = 0.4361; repeated-measures ANOVA), and membrane time constant (P = 0.3265; mixed-effects model) remain unchanged. c, dFBNs were held at -80 mV, except during the voltage protocols required to measure I[A], and dialyzed with 50 uM 4-ONE (turquoise shading). The amplitude of I[A] runs down during the course of the recording (P = 0.0024; mixed-effects model); I[non-A] (P = 0.2067; mixed-effects model), input resistance (P = 0.2942; mixed-effects model), and membrane time constant (P = 0.0783; mixed-effects model) remain unchanged. Columns, population averages; dots, individual cells; n, number of cells; asterisks, significant differences (P < 0.05) relative to the 0-minute time point in planned pairwise comparisons by Holm-Sidak or Dunn's test. For statistical details see Supplementary Table 2. Source Data Extended Data Fig. 5 Memory storage and erasure requires catalytically active K[V]b. a, b, dFBNs expressing a catalytically defective Hk(K289M) 'rescue' transgene in a homozygous Hk^1 mutant background. The cells were held at -80 mV between 0 and 10 min (except during the voltage protocols required to measure I[A]) and repeatedly step-depolarized to +10 mV (3 ms, 10 Hz) between 10 and 30 min. The inclusion of 50 uM 4-ONE in the intracellular solution fails to increase the fast and slow inactivation time constants of I[A] above the baselines recorded immediately after break-in (b, turquoise vs. grey shading); a series of depolarization steps between 10 and 30 min is similarly without effect (b, yellow shading; t[fast]: P = 0.6841, repeated-measures ANOVA; t[slow]: P = 0.7852, Friedman test; examples of peak-normalized I[A] evoked in the same dFBN by voltage steps to +30 mV in a). The amplitude of I[A] runs down during the course of the recording (b, P = 0.0087; repeated-measures ANOVA); I[non-A] (b, P = 0.6730; repeated-measures ANOVA), input resistance (b, P = 0.2615; repeated-measures ANOVA), and membrane time constant (b, P = 0.8143; repeated-measures ANOVA) remain unchanged. c, d, dFBNs expressing a catalytically competent Hk rescue transgene in a homozygous Hk^1 mutant background. The cells were held at -80 mV between 0 and 10 min (except during the voltage protocols required to measure I[A]) and repeatedly step-depolarized to +10 mV (3 ms, 10 Hz) between 10 and 30 min. The inclusion of 50 uM 4-ONE in the intracellular solution increases the fast and slow inactivation time constants of I[A] above the baselines recorded immediately after break-in (d, turquoise vs. grey shading); a series of depolarization steps between 10 and 30 min counteracts this increase despite the continuous presence of 4-ONE (d, yellow shading; t[fast]: P = 0.0020; t[slow]: P < 0.0001; Friedman test; examples of peak-normalized I[A] evoked in the same dFBN by voltage steps to +30 mV in c). The amplitude of I[A] runs down during the course of the recording (d, P < 0.0001; repeated-measures ANOVA); I[non-A] (d, P = 0.4334; repeated-measures ANOVA), input resistance (d, P = 0.5984; mixed-effects model), and membrane time constant (d, P = 0.9761; Friedman test) remain unchanged. Columns, population averages; dots, individual cells; n, number of cells; asterisks, significant differences (P < 0.05) relative to the 0-minute time point in planned pairwise comparisons by Holm-Sidak or Dunn's test. For statistical details see Supplementary Table 2. Source Data Extended Data Fig. 6 Potassium current amplitudes and membrane properties of dFBNs in Fig. 6. a, dFBNs expressing miniSOG were held at -80 mV in the intervals of 0-10 and 30-40 min (except during the voltage protocols required to measure I[A]) and repeatedly step-depolarized to +10 mV (3 ms, 10 Hz) between 10 and 30 min. Nine-minute exposures to blue light (between the 0- and 10-minute and the 30- and 40-minute time points) leave I [non-A] (P = 0.1673; mixed-effects model), input resistance (P = 0.0688; mixed-effects model), and membrane time constant (P = 3058; mixed-effects model) unchanged, but the amplitude of I[A] runs down during the course of the recording (P < 0.0001, mixed-effects model). b, dFBNs were held at -80 mV in the intervals of 0-10 and 30-40 min (except during the voltage protocols required to measure I[A]) and repeatedly step-depolarized to +10 mV (3 ms, 10 Hz) between 10 and 30 min. The cells were dialyzed with 50 uM 4-ONE in the intracellular solution (turquoise shading). The amplitude of I[A] runs down during the course of the recording (P < 0.0001; mixed-effects model); input resistance decreases after the series of depolarization steps (P = 0.0008; mixed-effects model); I[non-A] (P = 0.6240; mixed-effects model) and membrane time constant (P = 0.1258; mixed-effects model) remain unchanged. Columns, population averages; dots, individual cells; n, number of cells; asterisks, significant differences (P < 0.05) relative to the 0-minute time point in planned pairwise comparisons by Holm-Sidak test. For statistical details see Supplementary Table 2. Source Data Extended Data Fig. 7 Lipid peroxidation products alter the inactivation kinetics of I[A] in non-dFB neurons and cultured cells expressing mammalian K[V]1.4 and K[V]b2. a, Examples of peak-normalized transmembrane currents evoked by 1-s voltage pulses from -80 mV to +30 mV in a dFBN (grey) and a neuron of the pars intercerebralis (PI neuron) (black). A slowly activating outward current in the PI neuron interferes with an accurate measurement of t[slow]. b, c, PI neurons were held at -80 mV between 0 and 10 min (except during the voltage protocols required to measure I[A]) and repeatedly step-depolarized to +10 mV (3 ms, 10 Hz) between 10 and 30 min. The inclusion of 1 uM 4-ONE in the intracellular solution increases the fast inactivation time constant of I[A] above the baseline measured immediately after break-in (c, turquoise vs. grey shading); a series of depolarization steps between 10 and 30 min counteracts this increase despite the continuous presence of 4-ONE (c , yellow shading; P = 0.0009; Friedman test; examples of peak-normalized I[A] evoked in the same PI neuron by voltage steps to +30 mV in b). The amplitude of I[A] runs down during the course of the recording (c, P = 0.0075; repeated-measures ANOVA); I[non-A] (c, P = 0.1035; repeated-measures ANOVA), input resistance (c, P = 0.4532; mixed-effects model), and membrane time constant (c, P = 0.4861; Friedman test) remain unchanged. d, e, PI neurons were held at -80 mV between 0 and 10 min (except during the voltage protocols required to measure I[A]) and repeatedly step-depolarized to +10 mV (3 ms, 10 Hz) between 10 and 30 min. In the absence of 4-ONE, the fast inactivation time constant of I[A] (e, P = 0.3416; mixed-effects model; examples of peak-normalized I[A] evoked in the same PI neuron by voltage steps to +30 mV in d), the amplitude of I[non-A] (e, P = 0.3712; mixed-effects model), input resistance (e, P = 0.1304; mixed-effects model), and membrane time constant (e, P = 0.2109; mixed-effects model) remain unchanged, but the amplitude of I[A] runs down during the course of the recording (P = 0.0496; mixed-effects model). f, Examples of peak-normalized transmembrane currents evoked by 1-s voltage pulses from -80 mV to +30 mV in HEK-293 cells expressing mouse K[V]1.4 and K[V]b2 (grey), or in untransfected HEK-293 cells (black). g-j, HEK-293 cells expressing mouse K[V]1.4 and K[V]b2. A 1-h exposure to 12 mM methylglyoxal, followed by three washes with methylglyoxal-free solution, increases the fast and slow inactivation time constants of transmembrane currents relative to those of cells maintained in the absence of methylglyoxal (h, j, turquoise vs. grey shading; t[fast]: P < 0.0001; t[slow]: P < 0.0001; two-sided Mann-Whitney test). In cells held at -80 mV (except during the voltage protocols required to measure I[A]), the time constants remain stably elevated for 20 min (h, t[fast]: P = 0.4375; t[slow]: P = 0.1875; two-sided Wilcoxon test; examples of peak-normalized currents in g), but a series of depolarization steps (3 ms, 10 Hz, 20 min) to +10 mV reverses the increase (j, yellow shading; t[fast]: P = 0.0294, two-sided paired t-test; t[slow]: P = 0.0137, two-sided Wilcoxon test; examples of peak-normalized currents in i). The amplitude of I[A] runs down during the course of the recording (P = 0.0429, two-sided paired t-test). Columns, population averages; dots, individual cells; n, number of cells; asterisks, significant differences (P < 0.05) relative to the 0-minute time point by Holm-Sidak or Dunn's test. For statistical details see Supplementary Table 2. Source Data Extended Data Fig. 8 SMALDI-MSI analysis of 4-ONE. a, Mirror plot of mass spectra of 100 uM 4-ONE standard on a blank slide (top) or a brain cryosection (bottom). Spectra were acquired in single-ion-monitoring mode at the calculated m/z of the [4-ONE+GirT-H [2]O]^+ ion (268.2020); peaks with a mass deviation <5 ppm are labelled in green type. b, The intensity of the [4-ONE+GirT-H[2]O]^+ signal is decreased in cryosections of sleep-deprived brains (P = 0.0179, Kruskal-Wallis ANOVA) but not significantly altered in hemizygous sni^1 mutant males (P = 0.0560). Intensities on the left are normalized to a 100 uM 4-ONE standard on a blank slide; the scale is expanded on the right. Columns, population averages; dots, individual cryosections; n, number of cryosections; asterisks, significant differences (P < 0.05) relative to rested wild-type flies in planned pairwise comparisons by Dunn's test. For statistical details see Supplementary Table 2. Source Data Supplementary information Supplementary Tables Supplementary Tables 1 and 2. Reporting Summary Peer Review File Source data Source Data Figs. 2-6 and Source Data Extended Data Figs. 1 and 3-8 Rights and permissions Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Reprints and permissions About this article Check for updates. Verify currency and authenticity via CrossMark Cite this article Rorsman, H.O., Muller, M.A., Liu, P.Z. et al. Sleep pressure accumulates in a voltage-gated lipid peroxidation memory. Nature (2025). https://doi.org/10.1038/s41586-025-08734-4 Download citation * Received: 06 April 2023 * Accepted: 03 February 2025 * Published: 19 March 2025 * DOI: https://doi.org/10.1038/s41586-025-08734-4 Share this article Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative Download PDF Advertisement Advertisement Explore content * Research articles * News * Opinion * Research Analysis * Careers * Books & Culture * Podcasts * Videos * Current issue * Browse issues * Collections * Subjects * Follow us on Facebook * Follow us on Twitter * Sign up for alerts * RSS feed About the journal * Journal Staff * About the Editors * Journal Information * Our publishing models * Editorial Values Statement * Journal Metrics * Awards * Contact * Editorial policies * History of Nature * Send a news tip Publish with us * For Authors * For Referees * Language editing services * Submit manuscript Search Search articles by subject, keyword or author [ ] Show results from [All journals] Search Advanced search Quick links * Explore articles by subject * Find a job * Guide to authors * Editorial policies Nature (Nature) ISSN 1476-4687 (online) ISSN 0028-0836 (print) nature.com sitemap About Nature Portfolio * About us * Press releases * Press office * Contact us Discover content * Journals A-Z * Articles by subject * protocols.io * Nature Index Publishing policies * Nature portfolio policies * Open access Author & Researcher services * Reprints & permissions * Research data * Language editing * Scientific editing * Nature Masterclasses * Research Solutions Libraries & institutions * Librarian service & tools * Librarian portal * Open research * Recommend to library Advertising & partnerships * Advertising * Partnerships & Services * Media kits * Branded content Professional development * Nature Careers * Nature Conferences Regional websites * Nature Africa * Nature China * Nature India * Nature Italy * Nature Japan * Nature Middle East * Privacy Policy * Use of cookies * Your privacy choices/Manage cookies * Legal notice * Accessibility statement * Terms & Conditions * Your US state privacy rights Springer Nature (c) 2025 Springer Nature Limited Close Nature Briefing Sign up for the Nature Briefing newsletter -- what matters in science, free to your inbox daily. Email address [ ] Sign up [ ] I agree my information will be processed in accordance with the Nature and Springer Nature Limited Privacy Policy. Close Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing *