Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 17 (2), e2006409
eCollection

A Single Pair of Leucokinin Neurons Are Modulated by Feeding State and Regulate Sleep-Metabolism Interactions

Affiliations

A Single Pair of Leucokinin Neurons Are Modulated by Feeding State and Regulate Sleep-Metabolism Interactions

Maria E Yurgel et al. PLoS Biol.

Abstract

Dysregulation of sleep and feeding has widespread health consequences. Despite extensive epidemiological evidence for interactions between sleep and metabolic function, little is known about the neural or molecular basis underlying the integration of these processes. D. melanogaster potently suppress sleep in response to starvation, and powerful genetic tools allow for mechanistic investigation of sleep-metabolism interactions. We have previously identified neurons expressing the neuropeptide leucokinin (Lk) as being required for starvation-mediated changes in sleep. Here, we demonstrate an essential role for Lk neuropeptide in metabolic regulation of sleep. The activity of Lk neurons is modulated by feeding, with reduced activity in response to glucose and increased activity under starvation conditions. Both genetic silencing and laser-mediated microablation localize Lk-dependent sleep regulation to a single pair of Lk neurons within the Lateral Horn (LHLK neurons). A targeted screen identified a role for 5' adenosine monophosphate-activated protein kinase (AMPK) in starvation-modulated changes in sleep. Knockdown of AMPK in Lk neurons suppresses sleep and increases LHLK neuron activity in fed flies, phenocopying the starvation state. Further, we find a requirement for the Lk receptor in the insulin-producing cells (IPCs), suggesting LHLK-IPC connectivity is critical for sleep regulation under starved conditions. Taken together, these findings localize feeding-state-dependent regulation of sleep to a single pair of neurons within the fruit fly brain and provide a system for investigating the cellular basis of sleep-metabolism interactions.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Lk is required for metabolic regulation of sleep.
(A, B) IHC using antibody against Lk. White arrows indicate LHLK cell bodies, and yellow arrows indicate SELK cell bodies. Lk staining in flies expressing Lk-RNAi results in loss of Lk protein level in LHLK neurons (B) compared to UAS-Lk-RNAi/+ control (A). The brains were counterstained with nc82 (bruchpilot, gray). Scale bar = 100 μm. Color-coded images reflect fluorescence intensity, where minimum = 0 and max = 4,025. Scale bar for color-coded images = 10 μm. (C) Sleep is significantly reduced in starved UAS-Lk-RNAi/+ (n ≥ 20, p < 0.0001) and dcr2,Lk-GAL4/+ (n ≥ 25, p < 0.0001) control flies, while no significant differences are observed in Lk-GAL4>UAS-dcr2,Lk-RNAi flies (n = 13, p > 0.99). Two-way ANOVA (F [2, 141] = 10.87). White bars in column graphs represent amount of sleep during the day (ZT 0–12), while colored bars represent night sleep (ZT 12–24). (D) Sleep profile representative of (C). Flies are placed in food tubes during day 1 (fed, gray), then transferred to agar during day 2 (starved, blue). White/black bars represent lights on and off, respectively. (E, F) Lk staining in Lkc275 mutants reveals reduced protein levels compared to w1118 control (E). (G) Sleep is significantly reduced in starved w1118 controls (n = 63, p < 0.0001) and flies harboring one copy of Lkc275 (n = 68, p < 0.0001), while no significant differences are observed in Lkc275 mutants (n = 58, p = 0.90). Two-way ANOVA (F [2, 372] = 53.57). (H) Sleep profile representative of data in (G). (I, J) Lk protein levels are reduced in Lk−/−(GAL4) flies compared to w1118 control (J). (K) Lk−/− (GAL4) (n ≥ 47; p = 0.31) mutant flies and Lk+/−(GAL4) (n = 68, p = 0.35) fail to suppress sleep in response to starvation, while w1118 control flies suppress sleep (n = 75, p < 0.0001). Two-way ANOVA (F [2, 376] = 57.03). (L) Sleep profile representative of data in (K). All columns represent mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001. Underlying data can be found in S1 Data. ANOVA, analysis of variance; dcr2, dicer-2; GAL4, galactose-responsive transcription factor; IHC, immunohistochemistry; LHLK, Lateral Horn leucokinin; Lk, leucokinin; max, maximum; nc82, neuropil marker; RNAi, RNA interference; SELK, subesophageal ganglion leucokinin; UAS, upstream activation sequence; ZT, Zeitgeber time.
Fig 2
Fig 2. LHLK neurons are required for the metabolic regulation of sleep.
(A) Whole-brain and ventral nerve cord confocal reconstruction of Lk-GAL4>CD8::GFP. GFP-expressing neurons (green). Immunostaining for anti-Lk (red) reveals colocalization of LHLK and SELK neurons (yellow). The brain and ventral nerve cord are counterstained with nc82 (magenta). Scale bar = 100 μm. (B) GFP expression in the ventral nerve cord (ABLK neurons) of flies carrying Lk-GAL4>CD8::GFP is blocked by the tsh-GAL80 transgene. (C) Blocking synaptic release in Lk neurons with TNT impairs starvation-induced sleep suppression (n = 29, p = 0.60), while impTNT controls suppress sleep (n ≥ 27, p = 0.0002). No differences were observed between genotypes during the fed state (p = 0.06). Two-way ANOVA (F [1, 109] = 4.88). (D) Starvation-induced sleep suppression is absent in tsh-GAL80;Lk-GAL4 flies (n = 41, p = 0.12), while controls expressing impTNT suppress sleep (n = 33, p < 0.001). Sleep duration while fed does not differ significantly between tsh-GAL80;Lk-GAL4>TNT and impTNT flies (p = 0.21). Two-way ANOVA (F [1, 144] = 22.53). (E) TNT expression in Apt-GAL4 neurons abolishes starvation-induced sleep suppression (n ≥ 18, p = 0.48) and shows a significant increase in sleep during the fed state (p = 0.01), compared to impTNT that suppresses sleep (n ≥ 34, p < 0.0001). Two-way ANOVA (F [1, 103] = 18.22). (F) Expression of Lk-RNAi in Apt-GAL4 neurons (Apt-GAL4>UAS-dcr2,Lk-RNAi, n ≥ 12) blocks starvation-induced sleep suppression (p = 0.877), while UAS-Lk-RNAi/+ (n = 79, p < 0.0001) and dcr2,Apt-GAL4/+ (n = 13, p < 0.0001) controls suppress sleep. Two-way ANOVA (F [2, 201] = 12.44). (G) Expression of UAS-Lk under control of Apt-GAL4 in the Lkc275 mutant background restores starvation-induced sleep suppression (n = 15, p = 0.002) compared to Lkc275 flies and mutant control flies UAS-Lk/+;Lkc275 (n = 16; p = 0.98) or Apt-GAL4/+;Lkc275 (n = 16, p = 0.88). Control flies, Apt-GAL4/+;Lkc275/+ (n ≥ 16, p < 0.0001) or UAS-Lk/+;Lkc275/+ (n = 21, p < 0.0001), suppress sleep in response to starvation. Two-way ANOVA (F [4, 159] = 8.13). All columns are mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001. Underlying data can be found in S1 Data. ANOVA, analysis of variance; Apt, Apterous; CD8::GFP; membrane-tethered GFP, LK-GAL4>CD8:GFP;tshGAL80; dcr2, dicer-2; GAL4, galactose-responsive transcription factor; GFP, green fluorescent protein; impTNT, inactive variant of tetanus toxin; LHLK, Lateral Horn leucokinin; Lk, leucokinin; nc82, neuropil marker; RNAi, RNA interference; TNT, tetanus toxin; tsh, teashirt; UAS, upstream activation sequence.
Fig 3
Fig 3. Laser-induced microablation of LHLK neurons.
(A) Diagram representative of targeted multiphoton ablation. Third-instar larvae expressing UAS-mCD8::GFP in Lk neurons are placed dorsally onto a microscope slide, and neurons are ablated. Following ablation, larvae are placed in vials containing food and allowed to grow. Sleep in food vials and on agar is measured 5–7 days posteclosion in the DAMS. IHC is performed to verify ablated neurons. (B) Expression pattern of Lk-GAL4 during third-instar larval stage visualized with mCD8::GFP (green). The CNS was counterstained with nc82 (magenta). Scale bar = 100 μm. (C) Representative images of neuron pre- (top, white arrow) and postablation (bottom, red arrow). Scale bar = 10 μm. (D) Flies with bilateral laser ablation of LHLK neurons fail to suppress sleep when starved (n = 12, p = 0.09, t = 1.18), while ablation of a pair of ALK neurons (n = 9, p = 0.007, t = 3.04) and nonablated controls (n = 14, p = 0.01, t = 2.73) suppresses sleep in response to starvation. Unpaired t test. (E) Sleep profile representative of (D). Flies are placed in food tubes during day 1 (fed, gray), then transferred to agar during day 2 (starved, blue). White/black bars represent lights on and off, respectively. (F) Representative images of GFP-expressing Lk neurons post-LHLK (right) and ALK ablation (middle) and treated but nonablated controls (left). Red arrows indicate ablated neurons. White and yellow arrows indicate intact LHLK neurons and ALK neurons, respectively. Scale bar = 100 μm. All columns are mean ± SEM; *p < 0.05; **p < 0.01; ***p <0.001. Underlying data can be found in S1 Data. ALK, anterior leucokinin; CNS, central nervous system; DAMS, Drosophila Activity Monitor System; GAL4, galactose-responsive transcription factor; GFP, green fluorescent protein; IHC, immunohistochemistry; IR, infrared light; LHLK, Lateral Horn leucokinin; Lk, leucokinin; mCD8, membrane-tethered; nc82, neuropil marker; UAS, upstream activation sequence; ZT, Zeitgeber time.
Fig 4
Fig 4. LHLK neurons have increased activity during the starved state.
(A) Diagram of ex vivo Ca2+ imaging. Fed or 24-hr–starved adult female flies were dissected and placed dorsally onto chamber. UAS-Gerry fluorescence is recorded for 120 seconds with a Ti-Inverted Confocal microscope using a 20× air objective. (B) Average ratio of GCaMP/mCherry is increased in LHLK neurons during the starved state compared to the fed state ex vivo (n ≥ 8, p = 0.006, t = 3.154). Unpaired t test. (C) No significant differences in GCaMP/mCherry were detected in SELK neurons during the fed or starved state ex vivo (n ≥ 7, p = 0.95, t = 0.06). Unpaired t test. (D) Ex vivo application of 400 mM 2DG and 200 mM glucose to fed fly brains (n = 12) increases the GCaMP/mCherry fluorescence ratio in LHLK neurons compared to control (artificial hemolymph solution alone, n = 11, p = 0.01) and gluc application (200 mM, n = 12, p < 0.0001). Gluc application alone reduced the GCaMP/mCherry ratio compared to hemolymph-like solution control (p = 0.03). One-way ANOVA (F [2, 32] = 17.10). (E) Diagram of in vivo Ca2+ imaging. A portion of the head cuticle of a fed or 24-hr–starved adult female fly was removed, and GCaMP6m/mCherry fluorescence was recorded for 120 seconds. Fluorescence intensity scale represents the ratio range of GCaMP/mCherry ranging from 2 (max) to 0 (min). (F) GCaMP/mCherry ratio is increased in LHLK neurons during starvation (n = 30) compared to fed (n ≥ 19, p = 0.0036) and 3-hr re-fed controls (n = 29, p = 0.047). One-way ANOVA (F [2, 75] = 6.2). Scale bar = 10 μm. Error bars for GCamMP6m/mCherry ratio during the fed versus starved state indicate SEM; *p < 0.05; **p < 0.01; ***p < 0.001. Underlying data can be found in S1 Data. ANOVA, analysis of variance; GCaMP6m, GFP-calmodulin and M13 peptide sequence; GFP, green fluorescent protein; gluc, glucose; LHLK, Lateral Horn leucokinin; max, maximum; min, minimum; SELK, subesophageal ganglion leucokinin; UAS, upstream activation sequence; UAS-Gerry, GCaMP6m-mCherry; 2DG, 2 deoxy-glucose.
Fig 5
Fig 5. AMPKα in LHLK neurons is required for starvation-induced sleep suppression.
(A) Knockdown of AMPKα in Lk neurons (Lk-GAL4, n = 32, p = 0.14) abolishes starvation-induced sleep suppression, while control flies Lk-GAL4/+ (n = 27, p < 0.0001) and AMPKα-RNAi/+ (n ≥ 58, p < 0.0001) suppress sleep. During the fed state, sleep is significantly reduced in Lk-GAL4>UAS-AMPKα-RNAi compared to controls Lk-GAL4/+ (p < 0.0001) and AMPKα-RNAi/+ (p < 0.0001). Two-way ANOVA (F [2, 230] = 22.70). (B) Sleep profile representative of (A). Flies are placed in food tubes during day 1 (fed, gray), then transferred to agar during day 2 (starved, blue). White/black bars represent lights on and off, respectively. (C) Flies expressing UAS-AMPKα-RNAi in brain Lk neurons (tsh-GAL80;Lk-GAL4) fail to suppress sleep in response to starvation (n ≥ 56, p = 0.16) compared to control flies that suppress sleep (tsh-GAL80;Lk-GAL4/+, n ≥ 45, p < 0.0001) and AMPKα-RNAi/+ (n ≥ 58, p < 0.0001). Sleep while fed is significantly reduced in tsh-GAL80;Lk-GAL4>UAS-AMPKα-RNAi compared to tsh-GAL80;Lk-GAL4/+ (p = 0.0004) and AMPKα-RNAi/+ (p < 0.0001). Two-way ANOVA (F [2, 364] = 23.12). (D) Sleep profile representative of (C). (E) In vivo Ca2+ imaging during the fed state shows an increase in the average ratio of GCaMP/mCherry in LHLK neurons expressing UAS-AMPKα-RNAi;UAS-Gerry (gray and white, n = 10) compared to flies harboring Gerry alone (gray; n = 8, p = 0.01, t = 2.86). Unpaired t test. (F) During the starved state, no significant differences in LHLK Ca2+ activity are observed in Lk-GAL4>UAS-AMPKα-RNAi;UAS-Gerry flies (blue and white, n = 8) compared to flies harboring Gerry alone (blue, n = 8, p = 0.44, t = 0.78), Unpaired t test. Fluorescence intensity scale represents the ratio range of GCaMP6m/mCherry ranging from 2 (max) to 0 (minimum). Scale bar = 10 μm. Error bars for GCamp6m/mCherry ratio during the fed versus starved state indicate SEM; *p < 0.05; **p < 0.01; ***p < 0.001. Underlying data can be found in S1 Data. AMPK, 5′ adenosine monophosphate-activated protein kinase; ANOVA, analysis of variance; GAL4, galactose-responsive transcription factor; GCaMP6m, GFP-calmodulin and M13 peptide sequence; GFP, green fluorescent protein; LHLK, Lateral Horn leucokinin; Lk, leucokinin; max, maximum; RNAi, RNA interference; tsh, teashirt; UAS, upstream activation sequence; UAS-Gerry, GCaMP6m-mCherry; ZT, Zeitgeber time.
Fig 6
Fig 6. Lk targets the IPCs to promote wakefulness during starvation.
Expression pattern for Lkr−/−>UAS-mCD8::GFP (A), R65C07-GAL4 (C), R67D01-GAL4 (F), and Dilp2-GAL4 (I). The brains were counterstained with nc82 (magenta). Scale bar = 100 μm. (B) Lkr−/−(GAL4) mutants fail to suppress sleep in response to starvation (n ≥ 147, p = 0.09) compared to control w1118 flies (n = 155, p < 0.0001), Lkr+/− (n ≥ 134, p < 0.0001), and UAS-Lkr/+ (n = 26, p < 0.0001). Expression of UAS-Lkr in neurons labeled by Lkr−/−(GAL4) (Lkr−/−(GAL4); UAS-Lkr) restores starvation-induced suppression (n = 28, p = 0.0003). There were no significant differences during the fed state between control (UAS-Lk/+) and rescue flies (p = 0.10) or Lk−/− (p = 0.24). Two-way ANOVA (F [4, 965] = 13.05). (D) Blocking synaptic release in Lkr-expressing neurons that label the dFSB (R65C07-GAL4>UAS-TNT) does not affect starvation-induced sleep suppression (n = 48, p < 0.0001), similar to impTNT controls (n = 50, p < 0.0001). Sleep while fed is reduced in R65C07-GAL4>TNT flies compared to control (p = 0.0025). Two-way ANOVA (F [1, 109] = 11.17). (E) No impairments in starvation-induced sleep suppression were observed when knocking down Lkr in Lkr-expressing neurons (R65C07-GAL4>UAS-dcr2,Lkr-RNAi, n ≥ 45, p < 0.0001). Control UAS-Lkr-RNAi/+ (n = 27, p < 0.0001) and R65C07-GAL4/+ (n = 32, p < 0.0001) suppress sleep in response to starvation. Two-way ANOVA (Fcolumn [1, 205] = 130.8). (G) Expression of TNT in R67D01-GAL4, which labels the IPCs, impairs starvation-induced sleep suppression (n = 31, p = 0.47), while impTNT control flies suppress sleep (n = 48, p < 0.0001). Two-way ANOVA (F [1, 154] = 37.02). (H) Starvation-induced sleep suppression is absent with Lkr knockdown in Lkr-expressing neurons (R67D01, n = 41, p = 0.53). Control UAS-Lkr-RNAi/+ (n = 25, p < 0.0001) and dcr2,R67D01-GAL4 (n ≥ 53, p < 0.0001) suppress sleep in response to starvation. Two-way ANOVA (F [2, 232] = 13.86). (J) Knocking down Lkr in the IPCs (Dilp2-GAL4>UAS-dcr2,Lkr-RNAi, n ≥ 30, p = 0.67) results in flies that fail to suppress sleep in response to starvation, while dcr2,Dilp2-GAL4/+ control flies suppress sleep (n = 25, p = 0.0016). Two-way ANOVA (F [1, 108] = 4.1). (K) Sleep profile representative of (J). Flies are placed in food tubes during day 1 (fed, gray), then transferred to agar during day 2 (starved, blue). White/black bars represent lights on and off, respectively. All columns are mean ± SEM; *p < 0.05; **p < 0.01; ***p < 0.001. Underlying data can be found in S1 Data. ANOVA, analysis of variance; CD8::GFP, membrane-tethered GFP; LK-GAL4>CD8:GFP;tshGAL80; dcr2, dicer-2; dFSB, dorsal fan-shaped body; Dilp2, Drosophila insulin-like peptide 2; GAL4, galactose-responsive transcription factor; GFP, green fluorescent protein; impTNT, inactive variant of tetanus toxin; IPC, insulin-producing cell; Lk, leucokinin; Lkr, leucokinin receptor; nc82, neuropil marker; RNAi, RNA interference; TNT, tetanus toxin; tsh, teashirt; UAS, upstream activation sequence; ZT, Zeitgeber time.

Comment in

Similar articles

See all similar articles

Cited by 6 articles

See all "Cited by" articles

References

    1. Knutson KL, Van Cauter E. Associations between sleep loss and increased risk of obesity and diabetes. Ann N Y Acad Sci. 2008;1129: 287–304. 10.1196/annals.1417.033 - DOI - PMC - PubMed
    1. Chaput J-P, Després J-P, Bouchard C, Tremblay A. Association of sleep duration with type 2 diabetes and impaired glucose tolerance. Diabetologia. 2007;50: 2298–2304. 10.1007/s00125-007-0786-x - DOI - PubMed
    1. Taheri S, Lin L, Austin D, Young T, Mignot E. Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Med. 2004;1: 210–217. 10.1371/journal.pmed.0010062 - DOI - PMC - PubMed
    1. Arble DM, Bass J, Behn CD, Butler MP, Challet E, Czeisler C, et al. Impact of Sleep and Circadian Disruption on Energy Balance and Diabetes: A Summary of Workshop Discussions. Sleep. 2015;38:1849–1860. 10.5665/sleep.5226 - DOI - PMC - PubMed
    1. Schmidt MH. The energy allocation function of sleep: a unifying theory of sleep, torpor, and continuous wakefulness. Neurosci Biobehav Rev. 2014;47: 122–53. 10.1016/j.neubiorev.2014.08.001 - DOI - PubMed

Publication types

LinkOut - more resources

Feedback