J Sleep Res. 2020;00:e12976. wileyonlinelibrary.com/journal/jsr  | 1 of 11https://doi.org/10.1111/jsr.12976

© 2020 European Sleep Research Society

Received:15September2019  |  Revised:9December2019  |  Accepted:10December2019DOI:10.1111/jsr.12976

R E G U L A R R E S E A R C H P A P E R

Targeted recombination in active populations as a new mouse genetic model to study sleep-active neuronal populations: Demonstration that Lhx6+ neurons in the ventral zona incerta are activated during paradoxical sleep hypersomnia

Hyun-Sook Lee1,2,3 | Risa Yamazaki1 | Dianru Wang1 | Sébastien Arthaud1 | Patrice Fort1  | Laura A. DeNardo4 | Pierre-Hervé Luppi1

H-S Lee and R. Yamazaki contributed equally.

1Centre de Recherche en Neurosciences de Lyon (CRNL), Université Claude Bernard Lyon1,CNRSUMR5292,INSERMU1028,Bron, France2DepartmentofAnatomy,SchoolofMedicine, Konkuk University, Seoul, Korea3ResearchInstituteofMedicalScience,SchoolofMedicine,KonkukUniversity,Seoul, Korea4DepartmentofPhysiology,UniversityofCaliforniaLA,LosAngeles,CA,USA

CorrespondencePierre-HervéLuppi,CentredeRechercheen Neurosciences de Lyon (CRNL), Bâtiment 462-Neurocampus Michel Jouvet, 69500 Bron, France.Email:luppi@sommeil.univ-lyon1.fr

Funding informationUniversite Claude Bernard Lyon 1; The JapanSocietyforthePromotionofScience;PostdoctralFellowshipfromtheUeharaMemorialFoundation;FrenchEmbassyinJapan; Konkuk University-*14

AbstractThecFosimmunostainingallowedtheidentificationofmultiplepopulationsofneu-ronsinvolvedinthegenerationofparadoxicalsleep.Weadoptedthetransgenic(tar-getedrecombinationinactivepopulations)mousemodel,whichfollowinginjectionoftamoxifen,allowsexpressionofCre-dependentreporterconstructs(i.e.,mCherry)inneuronsexpressingcFosduringwakingorparadoxicalsleephypersomniafollow-ingautomaticparadoxicalsleepdeprivation.Threegroupsofmiceweresubjectedtotwoperiodsofwaking,oneperiodofwakingandoneofparadoxicalsleephypersom-nia,ortwoperiodsofparadoxicalsleephypersomnia.Ahighpercentageofdouble-labelledneuronswasobservedinthelateralhypothalamicareaandzonaincertaoftwoperiodsofwakingandtwoperiodsofparadoxicalsleephypersomnia inmice,butnot in thoseofoneperiodofwakingandoneofparadoxicalsleephypersom-nia in animals. Melanin-concentrating hormone neurons in the lateral hypothalamic areaandLhx6+cellsinthezonaincertaconstituted5.7±1.5%and8.8±2.3%ofallmCherry+cellsand20.6±4.8%and24.6±5.9%ofallcFos+neuronsintwoperiodsofparadoxicalsleephypersomniainanimals.Inaddition,melanin-concentratinghor-monecellsaswellasLhx6+neuronsrarelyexpressedmCherry(orcFos)inthewak-ingcondition, incontrasttoorexinneurons,whichconstitutedapproximately30%ofmCherry+andcFos+neurons.Our resultsvalidate theTRAPmethodologyandopenthewaytouseitforidentifyingtheneuronsactivatedduringwakingandpara-doxicalsleephypersomnia.Furthermore,theyindicateforthefirsttimethatLhx6+neurons in the zona incerta, like melanin-concentrating hormone cells in the lateral hypothalamicarea,areactivatedduringparadoxicalsleephypersomniabutnotdur-ingwaking.TheseresultsindicatethatLhx6+neuronsmightplayaroleinthecontrolofparadoxicalsleep,likethemelanin-concentratinghormonecells.

K E Y W O R D S

hypothalamic neuropeptides, sleep–wake cycle

2 of 11  |     LEE Et aL.

1  | INTRODUC TION

ThecFosimmunostaininghaslongbeenutilizedtoidentifytheneu-ronsactivatedduringwaking(W),slow-wavesleep(SWS)andpar-adoxical(i.e.,rapideyemovement[REM])sleep(PS).Bythismeans,we and others (Blanco-Centurion et al., 2019; Varin, Luppi, & Fort, 2018;Verretetal.,2003)identifiedmultiplepopulationsofneuronsplayingaroleinthegenerationofPSlikethemelanin-concentratinghormone(MCH)cellslocatedinthelateralhypothalamicarea(LHA)andzonaincerta(ZI). Itwas laterconfirmed,usingunitrecordings,that theseneuronsare specifically activeduringPS (Hassani, Lee,& Jones, 2009) and that their opto- and chemogenetic activation inducesanincreaseinPSquantities(Jegoetal.,2013;Varinetal.,2018).TheseresultsvalidatedthestrategyofusingthecFosstain-ingtoidentifypopulationsofPS-activeneurons.However,thevastmajorityofcFos+neuronslocalizedintheLHAandZIafterPShy-persomnia have not been functionally characterized, although ithasbeenshownthata largenumberexpressGAD67andarethusGABAergic (Brown et al., 2017; Sapin, Berod, Léger, & Herman,2010; Sapin et al., 2009). Optogenetic stimulation of GABAergicneuronsintheLHAandZIinducedWratherthanPS,indicatingthatpart of theseGABA neurons isW-inducing (Herrera et al., 2016).Interestingly,ithasbeenrecentlyreportedthatGABAergicneuronsco-containingLIMhomeobox6(Lhx6)intheventralZIexpresscFosattheendofthedarkperiodorduringtotalsleepdeprivationandthattheiractivation(orinhibition)increases(ordecreases)SWSandPSquantities(Liuetal.,2017).

Although cFos immunostaining allowed the identification ofneuronsactivatedduringPS in theLHA,ZIandmanyotherstruc-tures, the identification of their role was strongly limited due tothe lackofspecificmarkers.Anewgeneticstrategy,so-calledtar-getedrecombinationinactivepopulations(TRAP),mightbeanidealchoice tostudy theseneurons (DeNardo&Luo,2017;Guenthner,Miyamichi, Yang,Heller,& Luo, 2013). TheTRAPmice have beensuccessfullyusedtocharacterizeneuronsinvolvedinsensorycod-ing,motorbehaviour,memory,valence-encodingandSWSgenesis(Allenetal.,2017;DeNardoetal.,2019;Girasoleetal.,2018;Tasakaetal.,2018;Zhangetal.,2019).Thestrikingadvantageofsuchtrans-genic mice is that they allow Cre-dependent reporter constructs in theneuronsexpressingcFosatagiventimebymeansof injecting4-OH-tamoxifen(4-OHT).Inparticular,permanentGolgi-likelabel-ling(usingmCherryasareportergene)ofactivatedneuronscanbeobtained,whichcanbecombinedwithcFoslabellingiftheanimalswere perfused right after the second stimulus occurring days (orweeks)afterthefirststimulus.

TheaimofthepresentstudywastovalidatetheTRAPmeth-odology in studying the role of LHA and ZI neurons activatedduringPShypersomniaandW.Tothisaim,weallowedtheexpres-sionofmCherryandcFosinthreegroupsofmice.Thefirstgroupwasputtwice(1weekapart)inanopenfield.Itreceived4-OHTafterthefirstWperiodandwasperfusedattheendofthesecondWperiod.Bythismeans,wewereabletodeterminewhetherthesamecellswerelabeledwithmCherryandcFos.Wealsocombined

stainingofbothmarkerswiththatoforexin(ORX)toconfirmthatthelabelledneuronswereW-activeneurons.Inthesecondgroup,weperformed two (1weekapart)PSdeprivation-rebound (PSR)procedures,whichallowedustoshowwhethermCherry+neuronswereexpressingcFosduringthesecondPShypersomnia.Wever-ified that these neuronsmight correspond to neurons activatedduringPShypersomniabycombiningmCherryandcFoslabellingwithMCHimmunostaining.Thethirdgroupofmicewasfirstex-posed to theopen field and then1week later beforeperfusionexposedtoPSR.Double-labellinginthesemiceallowedustocon-firmthatneuronsactivatedduringWwerenotactivatedduringPShypersomnia.WefinallydeterminedwhetherLhx6+neuronsareactiveduringWorPShypersomnia.

2  | METHODS

2.1 | Animals

WefollowedEUaswellasFrenchguidelinesforanimalcareinsci-entific research. The experimental protocol was approved by theresearch ethical committee C2EA-55 of University-Lyon 1 (No.DR2014-45).Micewerehousedingroupsoftwobeforesurgeryandinindividualbarrelsafterwards.Animalsweremaintainedat22±1°Cunder12hrlight/darkcycles(lightsonfrom08:00to20.00hours)withfreeaccesstofoodandwater.ThetransgenicFos2A-iCreER/+ mice (TRAP2,foractivity-dependentgeneticlabelling,kindlydonatedbyDr Liqun Luo from StanfordUniversity)were crossed toR26AI14/+ (AI14) mice to obtain double heterozygous (TRAP2;Ai14, TRAP-RED) mice (Allenetal.,2017;DeNardoetal.,2019).Activity-dependentexpressionofdoubletransgenesisdescribedinFigure1a.Animalsat8–12weeksofage(n=21;25–30g)wereutilizedandeveryef-fort was made to minimize the number of mice sacrificed (W-W,n=3;W-PSR,n=8;PSR-PSR,n=10).Amongthem, fourW-PSR(M1270,1283,M1301and1361)andthreePSR-PSR(M1268,1269and1312)micewereselected.Criteriaforinclusionforanalysisweremicethatexhibitat least24%oftheamountofPSduringthefirst30-minperiodduring2-hrPShypersomnia; afterperfusion, theseanimalsdidshowoptimalcFosandmCherrystaining (Figures1‒5;Tables 1 and 2).

2.2 | Surgical procedures

Amixtureofketamine (100mg/kg)andxylazine (10mg/kg)wasinjected intraperitoneally to induce anaesthesia. Using a stere-otaxicdevice,threeelectroencephalogram(EEG)electrodeswereimplanted: two above the frontal (1 mm lateral to midline and2 mm anterior to bregma) and parietal (1.5 mm lateral to midline and −2mm from bregma) cortices and the third above the cer-ebellum(2mmlateraltomidlineand−6mmfrombregma)asref-erence.Electrodesforelectromyogram(EMG)wereinsertedintoneck muscles.

     |  3 of 11LEE Et aL.

2.3 | Polysomnography

Animalswereallowedto recover fromsurgery for1week in theirhomecagebeforebeinghabituatedtotherecordingconditionsfor4 days. They were connected to a cable attached to a slip-ring com-mutator to allow free movement within the barrel. Unipolar EEGandbipolarEMGsignalswereamplified(MCP-PLUS;Alpha-OmegaEngineering, Israel), digitized at 512 Hz and transferred to SlipAnalysisv2.9.8software.Vigilancestateswerescoredusinga5-swindow frame according to standard criteria (Libourel, Corneyllie,Luppi,Chouvet,&Gervasoni,2015;Renouardetal.,2015).

2.4 | Tamoxifen injection

Astocksolution(20mg/ml)of4-OHT(Sigma;H6278)waspreparedbyadding200-proofethanol,vortexed,andplacedonashakerat37°Cfor15minuntilitdissolved.Itwascoveredwithaluminiumfoiltominimize

lightexposure.Aworkingsolution(10mg/ml)waspreparedinoil;the4-OHT/ethanolmixture combinedwith cornoil (Sigma;C8267)wasputintoaspeed-vacfor2–3hruntilethanolevaporated.Thedrugwasdelivered intraperitoneally based on the time schedule (Figure 1b).

2.5 | Protocols for PS deprivation and rebound and W induction

Deprivationofparadoxicalsleepwasachievedbyanautomatedmethod(Libourel et al., 2015) employing an online signal analysis with Sleepscore (Viewpoint)software.Briefly,severaldiscriminantparametersofEEG/EMGrecordingswerecalculatedbasedonmanual scoringof the re-cordingsinordertoautomaticallyassignavigilancestate(W,SWSorPS)forevery1-sepoch.WhenaPSepisodewasdetected,atransistor–transistorlogic(TTL)signalwassenttoastimulationboxthroughasoft-ware developed under Matlab, waking up the animal by vertical shaking ofthebarrelfloorviaanelectromagnet.AutomaticPSdeprivationfor

F I G U R E 1  Abinary(i.e.,mCherryandcFos)expressionsystemoftargetedrecombinationinactivepopulations(TRAP)ofmice.(a)TRAPrequirestwotransgenes:onethatexpressesCreERT2(i.e.,tamoxifen-inducibleCrerecombinase)fromanactivity-dependentFospromoterandtheotherthatallowsexpressionofaneffectorgene(i.e.,mCherry)bymeansofCreERT2recombination.Without4-hydroxytamoxifen(4-OHT),CreERT2(blue,pacman-shapedstructure)isretainedinthecytoplasmofactivecells,whereitisexpressed,thusnorecombinationcanoccur(a,thesecondrow).Inthepresenceof4-OHT,CreERT2recombination(i.e.,bindingofCreERT2intofloxedalleles)canoccur,inactivatingthestopcodonandresultinginpermanentexpressionofmCherry(redstaining,a,thefirstrow).Non-activecellsdonotundergorecombinationduetotheabsenceofCreERT2expression(picturesnotshown).(b)TimesequenceofexperimentalprotocolsforTRAPing.(c)BasedonthedatafromTable1,temporalevolutionfortheamount(%)ofparadoxicalsleep(PS)wasrepresentedforcontrol(Ctrl)versusexperimentalstates.Becausethesecond(i.e.,cFos-labelled)eventwasadoptedfromwake-PSRanimals(i.e.,M1270,1283,1301and1361),thecorrespondingsecond(i.e.,cFos-labelled)conditionwasselectedfromPSR-PSRanimals(M1268,1269and1312).Forstatistics,thepaired t test was employed (***p < .001; **p < .01; *p<.05).PSR,paradoxicalsleepdeprivation-rebound

4 of 11  |     LEE Et aL.

48hrwashighlyefficient inmice, leavingaresidualamountofPSofonly2.2%(Arthaudetal.,2015).PSdeprivationstartedat10:00hours,whenamountsofPSarehigh.After48hrofPSdeprivation,micewere

returned to their home barrel, where they could sleep ad libitum during a2-hrrecoveryperiod(Figure1b).InPSR-PSRanimals,4-OHTwasin-jectedafter2hrofPSR.Oneweeklater,theanimalsweresubjectedto

F I G U R E 2  Orexin(ORX)neuronsactivated during wake-wake versus wake-PSRperiods.Inawake-wakeanimal(M1259;A–B4),manymCherry+(red)orcFos+(green)ORX(blue)neuronswereobserved in the lateral hypothalamic area (LHA).Inaddition,triple-labelledneuronscharacterized by a white nucleus with violetcytoplasmwereprominent(AandB4).Inawake-PSRanimal(M1270;C–D4),mCherry+(i.e.,wake-active)ORXneuronswereprominent,whereascFos+(i.e.,PSR-active)ORXcellswereminimal(CandD4).AandCrepresenttiledconfocalimages.3V,thirdventricle;f,fornix;ic,internalcapsule;PSR,PSR,paradoxicalsleepdeprivation-rebound. Scale bars = 200 μm (A),20μm (B2)

(a)

(b1) (b2)

(b3) (b4)

(d1) (d2)

(d3) (d4)(c)

F I G U R E 3  Theaveragenumbersofsingle-, double- or triple-labelled cells inwake-wake,wake-PSRandPSR-PSRanimals are presented as Venn diagrams, whereas standard error means (SEM) are described in the Results section. Based onthesenumbers,theproportions(%)oflabelled neuronal groups were calculated andstatisticallyanalysedinTable2.PSR,paradoxicalsleepdeprivation-rebound.Red:mCherry,blue,Lhx6,CFos,green

     |  5 of 11LEE Et aL.

PSdeprivationandperfusedafter2hrofPSR(i.e.,atnoon).InW-PSRorW-Wanimals,micewereplacedfor2hrinawhiteopen-fieldbox(base,45×45cm)containingavarietyoftoys.Theanimalsweremonitoredviavideocameraandpolysomnography(Liboureletal.,2015).Wheneverasleep,theywereawakenedbyslowrepositioningoftoysorbygentletouchusingasofttissue.Oneweeklater,W-Wanimalsweresubmittedtothesameprocedure,whereastheW-PSoneswerePSdeprivedfor48hrandallowedtorecoverfor2hrbeforeperfusion.

2.6 | Perfusion and cryostat sectioning

Miceweretranscardiallyperfusedwithheparin-addedRinger'ssolu-tionand4%ofparaformaldehyde (0.01Mphosphate-bufferedsa-line,PBS;pH7.4).Brainswereremovedandstoredin30%sucroseuntil they settled to the bottom. Using a cryostat, eight consecutive seriesofhypothalamicsections(extendingfromthecaudalpoleoftheparaventricularnucleustotherostralpoleoftuberomammillary

F I G U R E 4   Melanin-concentrating hormone (MCH) neurons activated duringPSR-PSRversuswake-PSRperiods.InaPSR-PSRanimal(M1312;A–B4),mCherry+(red)aswellascFos+(green) MCH (blue) neurons were observedintheLHA(A),whereastriple-labelled neurons characterized by a white nucleus with violet cytoplasm were rare (B4).Inawake-PSRanimal(M1361;C–D4),mCherry+(i.e.,wake-active)MCHneuronswererare,butcFos+(i.e.,PSR-active) MCH cells were substantial (C andD4).AandCrepresenttiledconfocalimages.3V,thirdventricle;f,fornix;ic,internal capsule. Scale bars = 200 μm(A),20 μm(B1).PSR,PSR,paradoxicalsleepdeprivation-rebound

(a) (b3) (b4)

(b1) (b2)

(d1) (d2)

(d3) (d4)(c)

F I G U R E 5  Lhx6+neuronsactivatedduringPSR-PSRversuswake-PSRperiods.InaPSR-PSRanimal(M1269;A–C4),asubstantialnumberofmCherry+(B4,red)orcFos+(C4,blue)Lhx6(green)neuronswasobservedintheZI.Triple-labelledneuronscharacterizedbyawhitenucleuswithredcytoplasmwerealsosubstantial(A,B4andC4).Inawake-PSRanimal(M1361;D–F4),alargenumberofcFos+(i.e.,PSR-active)Lhx6+cellswasobserved(E4andF4).AandCrepresenttiledconfocalimages.3V,thirdventricle;f,fornix;ic,internalcapsule.Scalebars = 100 μm(A),20μm(B1).PSR,PSR,paradoxicalsleepdeprivation-rebound

(a)

(b1) (b2) (b3) (b4)

(e1) (e2) (e3) (e4)

(f1) (f2) (f3) (f4)

(c1) (c2) (c3) (c4)

(d)

6 of 11  |     LEE Et aL.

TAB

LE 1

 Theproportion(%)ofparadoxicalandslow-wavesleepinwake-PSR(paradoxicalsleepdeprivation-rebound)versusPSR-PSRanimalsduring2hrofcontrolorPSreboundperiods

Cas

esW

ake-

PSR

(n =

4)

PSR-

PSR

(n =

3)

Ani

mal

sM

1270

M12

83M

1301

M13

61

M12

68M

1269

M13

12

1st

PSR

2n

d PS

R

1st

PSR

2n

d PS

R

1st

PSR

2n

d PS

R

Slee

p st

ates

PS

(%)a

SWS

(%)a

PSSW

SPS

SWS

PSSW

SPS

SWS

PSSW

SPS

SWS

PSSW

SPS

SWS

PSSW

S

Con

trol

sta

tesb

0–30

min

0.00

65.5

79.

4467

.77

9.73

65.83

4.72

34.7

310.83

61.6

7—

—17.78

71.6

7—

—12

.22

75.2

7—

—

30–6

016

.94

75.0

07.

2263

.03

3.07

55.5

77.

5084.70

11.3

963

.33

——

10.83

81.93

——

0.00

16.1

0—

—

60–9

00.28

15.0

00.

000.83

11.1

078.07

8.33

78.87

0.00

31.9

3—

—0.

5621

.13

——

10.0

080.57

——

90–1

201.

1147

.50

6.39

71.9

38.07

70.5

70.

008.9

13.89

71.9

7—

—6.

3981.67

——

14.7

266

.10

——

Exp.states(i.e.,PSR)

0–30

min

53.0

641

.11

29.1

761

.39

24.1

771

.67

25.5

659

.17

44.1

750.68

29.7

266

.39

27.2

242

.22

39.1

755.28

30.83

47.5

049

.72

38.61

30–6

036

.94

58.88

32.2

258.89

13.89

61.9

421

.39

69.1

61.

3968.05

17.2

260.28

26.1

153.89

24.4

472

.50

29.4

458.05

37.5

056

.95

60–9

019

.44

73.6

122.78

65.28

23.0

669

.44

21.3

968.34

10.83

24.7

29.

1758.05

30;0

059

.17

28.89

66.3

91.

9456

.39

36.1

156

.11

90–1

2022.78

68.05

11.1

161

.94

1.11

61.6

712

.50

45.83

20.5

675

.00

15.0

081.67

0.00

6.94

25.83

71.1

117.78

72.5

031

.67

63.0

6

a Theproportion(%)ofPSandSW

Sduring2hrofcontrolorPSreboundperiodswasobtainedfrompolysomnographicdata.Theproportion(%)ofthewakeperiodwasomittedforsimplicity,whichcould

beeasilyobtainedbysubtractingthesumofPSandSW

Sfrom100(i.e.,thetotalpercentage).

b Controldatawereobtainedduringthesame2-hrperiodonthepreviousdaybeforetheexposuretothewakeorPSRconditions);thus,thevalueswereabsent(—)inthe2ndPSRofM1268,1269and

1312

.

     |  7 of 11LEE Et aL.

nucleus)wereobtained at the thicknessof30μm; thus each well containedaseriesof240µm-apartsections.

2.7 | Immunofluorescence

Sections were washed with PBS containing 0.3% Triton X-100(PBST) and then incubated in 10% normal serum for 1 hr. Theywere then reacted with primary antibodies, including rabbit anti-cFos (1:2,500; Millipore, ABE457), goat anti-MCH (1:250;

Santa Cruz, 14509), goat anti-orexin (1:250; Santa Cruz, 8070)ormouseanti-Lhx6 (1:250;SantaCruz,271433) for48hr (4°C).Following washes, sections were incubated in 1:500 dilution ofAlexaFluor488-or647-conjugatedsecondaryantibodies(JacksonImmunoresearch)for2hr.Sectionsweremounted,coverslippedwith ProlongGold (Molecular Probes, P36930), and cured over-night in the dark (4°C). A positive control experiment for theimmunostainingwasperformedbyidentifyingpositiveimmunore-action in other brain regions that have been known to contain the antigen.Anegativecontrolwasperformedbyincubatingsections

TA B L E 2  Activatedorexin(ORX),melanin-concentratinghormone(MCH)orLIMhomeobox6(Lhx6)neuronsduringwake(W)-W,W-paradoxicalsleeprebound(PSR)orPSR-PSRconditionsintargetedrecombinationinactivepopulations(TRAP)mice

PeptidesExp. conditions Animals

Proportions (%)

Total (mCherry+ peptide+) over total mCherry

Total (mCherry+ peptide+) over total peptide

Total (cFos+ peptide+) over total cFos

Total (cFos+ peptide+) over total peptide

Triple over total peptide

ORX W-W M1258 39.3 25.3 31.8 34.8 13.2

M1259 28.3 29.5 30.0 49.6 19.8

M1363 25.6 42.0 30.4 44.3 25.7

X±SEM (n = 3)

31.1±4.2 32.3±5.0 30.7±0.5a 42.9±4.3b 19.6±3.6c

W-PSR M1270 26.9 26.9 4.3 1.9 0

M1301 22.6 34.0 6.8 1.3 0

M1361 30.5 28.4 7.4 1.6 0

X±SEM (n = 3)

26.7±2.3 29.8±2.2 6.2±0.9a 1.6±0.2b 0c

MCH PSR-PSR M1268 6.4 13.0 11.0 9.5 0

M1269 2.8 4.0 26.1 8.8 0

M1312 7.9 10.2 24.7 31.6 1.2

X±SEM (n = 3)

5.7±1.5d 9.1±2.7e 20.6±4.8 16.6±7.5 0.4±0.4

W-PSR M1270 1.2 2.0 25.5 24.8 0

M1283 1.5 1.9 24.2 19.3 0

M1361 2.2 2.2 21.3 21.6 0

X±SEM (n = 3)

1.6±0.3d 2.0±0.1e 23.7±1.2 21.9±1.6 0

Lhx6 PSR-PSR M1268 6.0 4.2 35.4 5.0 2.7

M1269 13.4 6.5 23.5 7.2 2.6

M1312 6.9 3.0 14.9 5.4 2.4

X±SEM (n = 3)

8.8±2.3f 4.6±1.0g 24.6±5.9 5.9±0.7h 2.6±0.1i

W-PSR M1283 2.8 2.4 31.5 11.7 0.3

M1301 2.0 1.6 24.7 8.4 0.4

M1361 1.5 1.0 12.9 10.4 0.6

X±SEM (n = 3)

2.1±0.4f 1.7±0.4g 23.0±5.4 10.2±1.0h 0.4±0.1i

Note: Proportions(%)werecalculatedfromthenumberofsingle-,double-ortriple-labelledcellsindicatedinFigure3.Hypothalamicsectionsbetweenbregma−1.34and−1.82mmwereobtained,wherecellcountsweremadeunilaterallyusingeveryeighthsection(atotalofthreesectionsperanimal;sectionthickness,30µm).BasedontheMann–WhitneyU-test,meandifferencebetweenthetwosamesuperscripts(a–i)isstatisticallysignificant(p < .05).

8 of 11  |     LEE Et aL.

intheincreasingdilutionsofprimaryantibodiesuntilimmunoreac-tiondisappearedintheabsenceoftheantibody.

2.8 | Confocal microscopy and cell counting

ZEN 2010 software was utilized with a confocal laser scanningmicroscope (LSM 800; Zeiss), which was equipped with blueargon (488 nm), green helium neon (543 nm) and red heliumneon(633nm)lasers.IntheSMARTsetup,“bestsignal”wascho-sen. Toobtain high-resolution images, a specific frame size (i.e.,x*y = 1,024 × 1,024) at speed 4 was selected with an averaging process (n=4).Double-ortriple-labelledneuronswererecheckedfor genuine labelling using a single channel illumination.Opticalsectionthicknesswasbetween0.6and1.8µm.Tilescanwasse-lected to obtain awider view; three in the LHAwere taken forORX orMCH neurons, whereas three in the ZI were taken forLhx6+cells.Thecounting tool in theAdobePhotoshopCS6 (64Bit)wasused;cellswereconsideredlabelledwhentheyexhibitedclear cytoplasmic (i.e.,mCherry+,MCHorORX) or nuclear (i.e.,cFos+orLhx6+)morphology.

2.9 | Data analysis

The paired t test was employed to analyse the values in Table 1, be-causethedatawereobtainedfromthesameindividualsatthesametimeof theday andondifferentdays.TheMann–WhitneyU-test was employed to analyse the data shown in Table 2.

3  | RESULTS

TheTRAPmicewereused in thisstudyto identify theneurons intheLHAandZIactivatedduringWandPShypersomnia.Themeth-odology allowed us to label neurons expressing cFos at two dif-ferent time-points.ThemCherry labelling (so-calledTRAPing)wasinducedinactivatedneuronswhen4-OHTwasinjected(Figure1a).Neurons were labelled with cFos when they were activated just be-fore perfusion. Three protocols were used: PSR-PSR,W-PSR andW-W(Figure1b).Basedonpolysomnographicdata(Table1),theef-ficiencyof48-hrPSdeprivationwasassessedasameansofinduc-ingPShypersomnia (Figure1c).Seven representativePSRanimalswere selected for analysis; the proportions of PS during the first30-minperiodvaried from0%to17.78% in thecontrol condition,whereas they were between 24.17% and 53.06% during the PSRcondition(Table1).TheamountofPSwassignificantlyhigherinPSR(Figure 1c, blue line) than in the control condition (Figure 1c, black line)duringallfour30-minperiods.Thus,48-hrPSdeprivationwasefficientininducingasignificantPShypersomniaduringthesubse-quent 2-hr rebound period.

We compared the distribution of mCherry+ (or cFos+) ORXneuronsinW-WversusW-PSRanimals(Figure2).InW-Wanimals

(Figure2B4),manyORXneuronswereeithercFos+ormCherry+.Alargenumberofcellsweretriple-labelled,exhibitingawhitenucleuswith violet cytoplasm. InW-PSR animals (Figure 2D4), mCherry+ORXneuronsrepresentingcellsactivatedduringWwerenumerous,whereas cFos+ORX cellswere rare. Triple-labelled neuronswerenot observed in this case.

Quantitativeanalysiswasperformedtocharacterizetheneu-rons activated inW-WandW-PSRconditions (Figure3, the toprow;Table2, rows3–10). InW-Wmice, theaveragenumbersofmCherry+andcFos+neuronsintheLHAwere325±93(n = 3) and 405±41(n=3),respectively.Theaveragenumberofdouble-la-belled (mCherry+ and cFos+) cells was 123 ± 22, which corre-spondstoapproximately37.8%ofthemCherry+cellsand30.4%ofthecFos+cells.InW-PSRmice,theaveragenumbersofmCherry+andcFos+neuronswere281±40and71±22,respectively.Onaverage, only three neurons were double-labelled, indicating that differentgroupsofcellsareactivatedduringWandPSR.Basedon thenumberof labelled cells (Figure3, the top row), thepro-portionsoflabelledneuronalgroupswereanalysed(Table2,rows3–10). In W-W animals, the average percentages of double-la-belled (mCherry+ andORX+) neurons over totalmCherry+ cellsandtotalORXneuronswere31.1±4.2%and32.3±5.0%,respec-tively.ThevalueswereinthesamerangeinW-PSRanimals.ThepercentagesofcFos+ORXneuronsovertotalcFos+cellsandtotalORXneuronsinW-Wanimalswere30.7±0.5%and42.9±4.3%,respectively. In contrast, inW-PSR mice, the percentages were6.2±0.9%and1.6±0.2%,respectively,indicatingthatORXneu-rons are not activated during PSR. In linewith this assumption,theaveragenumberoftriple-labelled(mCherry+,cFos+andORX+)neuronswas57±12inW-Wanimals(Figure3,thetoprow),butit was zero in theW-PSRmice. Impressively, the percentage oftriple-labelledneuronsinW-Wmiceconstituted19.6±3.6%ofallORXneurons(Table2).

WethendeterminedwhethertheneuronsactivatedduringPSRcorrespondtoMCHneurons.InaPSR-PSRanimal,alargenumberofcFos+(ormCherry+)MCHneuronswereobserved,buttherewereonlyafewtriple-labelledMCHcells(Figure4A–B4).InaW-PSRani-mal,however,alargeproportionofMCHcellswerecFos+,buttherewerefewmCherry+MCHcells(Figure4C–D4),indicatingthatMCHneuronsareactivatedduringPSRbutnotduringW.

QuantitativeanalysiswasperformedtocomparethenumberofMCHneuronsactivated inPSR-PSRversusW-PSRmice (Figure3,thesecondrow;Table2,rows11–18). InPSR-PSRanimals,theav-erage numbers ofmCherry+ and cFos+ neurons in the LHAwere369±27 (n =3) and192±64 (n = 3), respectively (Figure 3, the secondrow).Thenumberofdouble-labelled(mCherry+andcFos+)cells was 53 ± 11, which corresponds to 14.4% of themCherry+neuronsand27.6%ofthecFos+cells.InW-PSRanimals,thenum-bersofmCherry+andcFos+neuronswere273±29and197±27,respectively.Onaverage,only10neuronsweredouble-labelled inW-PSR animals, indicating that different populations of neuronsareactivatedduringWandPSR.Basedonthenumberof labelledcells(Figure3,thesecondrow),theproportionsoflabelledneuronal

     |  9 of 11LEE Et aL.

groupswereanalysed (Table2, rows11–18).Theaveragepercent-ages of mCherry+ MCH cells over total mCherry+ neurons andtotalMCHcellsinPSR-PSRanimalswere5.7±1.5%and9.1±2.7%,respectively. In contrast, these values were only 1.6 ± 0.3% and2.0±0.1%inW-PSRanimals.Finally,thepercentagesofcFos+MCHneuronsovertotalcFos+cellsandtotalMCHneurons inPSR-PSRanimalswere20.6±4.8%and16.6±7.5%,respectively.Theseval-uesweresimilarinW-PSRanimals(i.e.,23.7±1.2%and21.9±1.6%).Anextremely limitednumberof triple-labelledcellswasobservedonlyinonePSR-PSRanimal(Table2,M1312).

WefinallyexaminedwhetherLhx6+neuronsmainlylocalizedintheZIareactivatedduringWorPShypersomnia.Themicrograph(Figure5A–C4)showsasubstantialnumberoftriple-labelled,Lhx6+neuronsinaPSR-PSRanimal.BecauseLhx6isanuclearprotein(likecFos),triple-labelledcellsintheventralZIexhibitedred(i.e.,mCher-ry-labelled) cytoplasm with a white nucleus; many mCherry+ (orcFos+)Lhx6-immunoreactiveneuronsarealsovisibleintheregion.InaW-PSRanimal,however,manycFos+Lhx6neuronsarevisiblebutonlyafewmCherry+Lhx6cells(Figure5D–F4).

QuantitativeanalysiswasperformedtoexaminetheactivationofLhx6+neuronsinPSR-PSRversusW-PSRanimals(Figure3,thebot-tomrow;Table2,rows19–26).InPSR-PSRanimals,theaveragenum-bersofmCherry+andcFos+neuronsintheZIwere193±21(n = 3) and100±27 (n = 3), respectively (Figure 3, the bottom row). The numberofdouble-labelled (mCherry+andcFos+) cellswas17±2,whichcorrespondsto8.8%ofthemCherry+neuronsand17.0%ofthecFos+cells.InW-PSRanimals,theaveragenumbersofmCherry+and cFos+neuronswere218±15 and147±51, respectively.Onaverage, only nine neurons were double-labelled (i.e., mCherry+andcFos+),indicatingthatdifferentgroupsofneuronsareactivatedduringWandPSR.Basedonthenumberoflabelledcells(Figure3,thebottomrow),theproportionsoflabelledneuronalgroupswereana-lysed (Table2, rows19–26).TheaveragepercentagesofmCherry+Lhx6neuronsovertotalmCherry+cellsandtotalLhx6+neuronsinPSR-PSRanimalswere8.8±2.3%and4.6±1.0%,respectively.Thesevaluesweresignificantly(p<.05)higherthanthoseseeninW-PSRanimals(i.e.,2.1±0.4%and1.7±0.4%),indicatingthatLhx6+neu-ronsareactivatedduringPSRratherthanduringW.Inaddition,thepercentagesofdouble-labelled(cFos+andLhx6+)neuronsovertotalcFos+cellsandtotalLhx6+neuronswere24.6±5.9%and5.9±0.7%,respectively.ThesevalueswereinthesamerangeasW-PSRanimals(i.e.,23.0±5.4%and10.2±1.0%).Altogether,theseresultsindicatethatasubstantialproportionoftheZIneuronsactivatedduringPSR(butnotduringW)correspondtoLhx6+neurons.Inlinewiththeseresults,theaveragepercentageoftriple-labelledneuronsovertotalLhx6+cellswassignificantly(p<.05)largerinPSR-PSRthaninW-PSRanimals (Table 2, the last column).

4  | DISCUSSION

Inthepresentstudy,wefirstreportthatalargeproportionofneu-ronslabelledwithmCherryduringW(37.8%)orPSR(14.4%)express

cFoswhentheanimalsarere-exposedtothesameconditions(i.e.,W-WorPSR-PSR) justbeforeperfusion,whereas theproportionsaremuchsmaller(i.e.,1.1%and3.7%)whentheanimalsareexposedtotwodifferentconditions(i.e.,W-PSR).Inaddition,wefoundthatORXneuronsknowntobespecificallyactiveduringW(Lee,Hassani,&Jones,2005)expresscFosduringWbutnotduringPSR,whereasthoseexpressingMCHknowntobespecificallyactiveduringPSR(Gao,2009;Peever&Fuller,2016;Verretetal.,2003)aremCherry+during PSR and not duringW. These results first indicate that alargeproportionof theneuronsexpressingmCherryexpresscFoswhentheanimalsarere-exposedtothesamecondition,validatingtheTRAPmethodologyforstudyingneuronsactivatedduringWorPSR.Inaddition,weobservedthat8.8%ofZIneuronslabelledwithmCherryduringPSRexpressedLhx6,incontrasttotheWcondition,inwhichonly2.1%ofthemCherry+cellsexpressedLhx6.Thesere-sults indicate that Lhx6+neuronsare specifically activatedduringPSR.ThesevaluesareevenhigherthanthoseseenwithMCH,sug-gestingthatLhx6+neuronsmightbespecificallyactiveduringPS,like the MCH cells.

The TRAPmethod has never been utilized to study neuronsactivatedduringPShypersomniaandW.WeshowedherethatalargepercentageoftheneuronslabelledwithmCherryarecFos+inW-W and PSR-PSR but not inW-PSR conditions. Further,weshowedthata largepercentageofthemCherry+andcFos+cellslabelled after theW condition expressORX and notMCH, indi-cating thatmCherry is expressed inW-active neurons, and thusvalidatingthespecificityofthestainingobtained.Further,wecon-firmedthespecificityforactivatedneuronsofthemCherrylabel-lingobtainedafterPSRbecausealargenumberofthemCherry+neuronsexpressedMCH.Inaddition,thefactthatnearlynoORXneuronswerelabelledinthePSRconditionindicatesthatalthoughtherearesomeWepisodesduringPSR,itisnotsufficienttoinducemCherry or cFos staining.

Insummary,ourresultsclearly indicatethattheTRAPmethodcanbeusedtostudyneuronsactivatedduringPSRandW,introduc-ingapowerfulnewmethodtostudysuchneurons.Thisisinlinewitha recent studyusingTRAPmice to identifyapreviouslyunknownpopulationofglutamatergicneurons involved in thegenerationofSWS (Zhang et al., 2019).Our studyopens the possibility to fullycharacterize PS-activated neurons across the brain, by combiningtheTRAPmethodologywithmRNAssequencing,“invitro”record-ings, and optogenetic and chemogenetic approaches. Furthermore, we demonstrated for the first time that Lhx6+ neurons of the ZIwere activated duringPSR like theMCHneurons, and not duringW. We previously showed that most of the neurons expressingcFosintheZIduringPSRexpressGAD67(Sapinetal.,2010).Ithasbeen recently shown that Lhx6+ neurons of the ZI express vGATandGAD67 and are distinct fromMCH andORX neurons (Liu etal.,2017;Mickelsenetal.,2019).ItisthereforelikelythattheLhx6+neurons in theZI constitute a sub-population of thePS-activatedGABAergicneurons.

Inapreviousreport,Liuetal.(2017)demonstratedthatcFosexpressionisincreasedinLhx6+neuronsatlightonsetandafter

10 of 11  |     LEE Et aL.

6hroftotalsleepdeprivation,withorwithout1hrofsleeprecov-ery.From these results, theyconcluded thatLhx6+neuronsareactivatedbysleeppressure.Althoughthisisnotdirectlycompa-rabletoourresults,wedidfindthatcFosexpressionwasstronglyincreased inLhx6+neuronsafter2hrofPSRandnotafter2hrofW.OurresultsthereforeindicatethattheseneuronsaremoreactiveduringPSR thanduringW.Unit recordingsof theseneu-ronsareeventuallyneededtoconfirmthishypothesis.Liuetal.(2017)furthershowedthatthelossofLhx6+neurons(whichwasdirectlyinducedbygenedeletionfromthedevelopingdienceph-alon) caused an increase inW and a reduction of SWS and PS,witharelativedecreaseofPSgreaterthanthatofSWS.Likewise,clozapineNoxide (CNO) injectionaftertransfectingLhx6+neu-rons with excitatory designer receptors exclusively activatedbydesignerdrugs (DREADDS) inducedan increase inPSand toaminor extent in SWS,whereas inhibitionof Lhx6+ cells trans-fectedwithinhibitoryDREADDSreducedsleep(PSinparticular).TheyfurthershowedthatLhx6+cellsprojecttotheORXcellsaswellasmonoaminergicW-activeneuronsandarelikelytoinducesleep by inhibiting them. These observations are similar to what hasbeenreportedfortheactivationorinhibitionofMCHneurons(Jego et al., 2013;Varin et al., 2018).Altogether, these andourresultssuggestthatlikeMCHneurons,Lhx6+cellsintheZImightcontributetoPS.

In addition to these two populations of neurons, other neu-rons in the LHA and ZI express cFos following PSR; thus, furtherexperimentsareneededtoidentifytheseneuronsandcharacterizetheirspecificfunctions.Inconclusion,thepresentfindingsdemon-strate that TRAPmethodology combinedwith our automatic PS-deprivationsystemconstitutesapromisingnewstrategytoidentifyvarioustypesofneuronsinvolvedinPS.

ACKNOWLEDG EMENTSThis paper was written as part of Konkuk University's researchsupport programme for its faculty (Hyun-Sook Lee) on sabbaticalleavein2018.ThisworkwassupportedbytheCentreNationaldelaRechercheScientifique,UniversitédeLyonandUniversitéLyon1. RY was funded by the JSPS Overseas Challenge Program forYoung Researchers from The Japan Society for the Promotion ofScienceandaPostdoctoralFellowship from theUeharaMemorialFoundation,FrenchEmbassyinJapan.

CONFLIC T OF INTERE S TNoconflictsofinterestdeclared.

AUTHOR CONTRIBUTIONSPHL, PF and LAD designed the experiment. RY,DW and SA per-formedphysiologicalrecordings.HSLperformedanatomicalworkswithdataanalysis.HSLandPHLwrotethemanuscript.

ORCIDPatrice Fort https://orcid.org/0000-0003-1211-8631 Pierre-Hervé Luppi https://orcid.org/0000-0002-0503-423X

R E FE R E N C E SAllen,W.E.,DeNardo,L.A.,Chen,M.Z.,Liu,C.D.,Loh,K.M.,Fenno,

L. E., … Luo, L. (2017). Thirst-associated preoptic neurons encodean aversive motivational drive. Science, 357, 1149–1155. https ://doi.org/10.1126/scien ce.aan6747

Arthaud, S., Varin, C.,Gay,N., Libourel, P.-A., Chauveau, F., Fort, P.,…Peyron,C. (2015).Paradoxical (REM)sleepdeprivation inmiceusing the small-platforms-over-water method: Polysomnographicanalyses and melanin-concentrating hormone and hypocretin/orexinneuronalactivationbefore,duringandafterdeprivation.The Journal of Sleep Research, 24, 309–319. https ://doi.org/10.1111/jsr.12269

Blanco-Centurion,C.,Luo,S.,Spergel,D.J.,Vidal-Ortiz,A.,Oprisen,S.,VandenPol,A.N.,…Shiromani,P.J.(2019).Dynamicnetworkacti-vationofhypothalamicMCHneuronsinREMsleepandexploratorybehavior. The Journal of Neuroscience, 39, 4986–4998. https://doi.org/10.1523/JNEUROSCI.0305-19.2019

Brown,R.E.,Yang,C.,Winston,S.,Anderson-Chernisof,M.,McNally,J.M.,&McKenna,J.T.(2017).Investigationofthedevelopmentalor-iginofforebrainGABAergicneuronsinvolvedinsleep-wakecontrolusingafate-mappingapproach.Sleep, 40(suppl.1),A17.https://doi.org/10.1093/sleepj/zsx050.042

DeNardo,L.A.,Liu,C.D.,Allen,W.E.,Adams,E.L.,Friedmann,D.,Fu,L.,…Luo,L.(2019).Temporalevolutionofcorticalensemblespromotingremote memory retrieval. Nature Neuroscience, 22, 460–469. https ://doi.org/10.1038/s41593-018-0318-7

DeNardo,L.A.,&Luo,L.(2017).Geneticstrategiestoaccessactivatedneurons. Current Opinion in Neurobiology, 45, 121–129. https ://doi.org/10.1016/j.conb.2017.05.014g/10.1038/s41593-018-0318-7

Gao,X.-B.(2009).ElectrophysiologicaleffectsofMCHonneuronsinthehypothalamus. Peptides, 30, 2025–2030. https ://doi.org/10.1016/j.pepti des.2009.05.006

Girasole,A.E.,Lum,M.Y.,Nathaniel,D.,Bair-Marshall,C.J.,Guenthner,C.J.,Luo,L.,…Nelson,A.B.(2018).Asubpopulationofstriatalneu-rons mediates levodopa-induced dyskinesia. Neuron, 97, 787–795.https://doi.org/10.1016/j.neuron.2018.01.017

Guenthner,C.J.,Miyamichi,K.,Yang,H.H.,Heller,H.C.,&Luo,L.(2013).Permanent genetic access to transiently active neurons via TRAP:Targeted recombination in active populations. Neuron, 78,773–784.https ://doi.org/10.1016/j.neuron.2013.03.025

Hassani,O.K.,Lee,M.G.,&Jones,B.E.(2009).Melanin-concentratinghormoneneuronsdischargeinareciprocalmannertoorexinneuronsacross the sleep-wake cycle. Proceedings of the National Academy of Sciences of the United States of America, 106,2418–2422.https://doi.org/10.1073/pnas.0811400106

Herrera,C.G.,Cadavieco,M.C.,Jego,S.,Ponomarenko,A.,Korotkova,T.,&Adamantidis,A. (2016).Hypothalamicfeed-forward inhibitionof thalamocortical network controls arousal and consciousness.Nature Neuroscience, 19,290–298.https://doi.org/10.1038/nn.4209

Jego,S.,Glasgow,S.D.,Herrera,C.G.,Ekstrand,M.,Reed,S.J.,Boyce,R.,…Adamantis,A.R.(2013).Optogeneticidentificationofarapideye movement sleep modulatory circuit in the hypothalamus. Nature Neuroscience, 16,1637–1643.https://doi.org/10.1038/nn.3522

Lee,M.G.,Hassani,O. K., & Jones, B. E. (2005).Discharge of identi-fiedorexin/hypocretinneuronsacross the sleep-wakingcycle.The Journal of Neuroscience, 25, 6716–6720. https ://doi.org/10.1523/JNEUROSCI.1887-05.2005

Libourel, P. A., Corneyllie, A., Luppi, P. H., Chouvet, G., & Gervasoni,D. (2015).Unsupervisedonline classifier in sleep scoring for sleepdeprivation studies. Sleep, 38, 815–828. https://doi.org/10.5665/sleep.4682

Liu,K.,Kim,J.,Kim,D.W.,Zhang,Y.S.,Bao,H.,Denaxa,M.,…Blackshaw,S.(2017).Lhx6-positiveGABA-releasingneuronsofthezonaincertapromote sleep. Nature, 548,582–587.https://doi.org/10.1038/nature23663

     |  11 of 11LEE Et aL.

Mickelsen, L. E., Bolisetty,M., Chimileski, B. R., Fujita, A., Beltrami, E.J.,Costanzo,J.T.,…Jackson,A.C.(2019).Single-celltranscriptomicanalysisofthelateralhypothalamicarearevealsmolecularlydistinctpopulationsofinhibitoryandexcitatoryneurons.Nature Neuroscience, 22,642–656.https://doi.org/10.1038/s41593-019-0349-8

Peever, J., & Fuller, P. M. (2016). The biology of REM sleep. Current Biology, 26, R34–R35. https ://doi.org/10.1016/j.cub.2015.11.011

Renouard, L., Billwiller, F., Ogawa, K., Clément, O., Camargo, N.,Abdelkarim,M.,…Luppi,P.H.(2015).Thesupramammillarynucleusand the claustrum activate the cortex during REM sleep. Science Advances, 1, e1400177. https ://doi.org/10.1126/sciadv.1400177

Sapin,E.,Berod,A.,Léger,L.,&Herman,P.A.(2010).Averylargenum-berofGABAergicneuronsareactivatedinthetuberalhypothalamusduringparadoxical (REM)sleephypersomnia.PLoS One, 5, e11766. https://doi.org/10.1073/pnas.0811400106

Sapin,E.,Lapray,D.,Bérod,A.,Goutagny,R.,Léger,L.,Ravassard,P.,…Luppi, P.-H. (2009). Localization of the brainstemGABAergic neu-ronscontrollingparadoxical(REM)sleep.PLoS One, 4, e4272. https ://doi.org/10.1371/journ al.pone.0004272

Tasaka, G., Guenthner, C. J., Shalev, A., Gilday,O., Luo, L., &Mizrahi,A. (2018). Genetic tagging of active neurons in auditory cor-tex reveals maternal plasticity of coding ultrasonic vocalizations.Nature Communications, 9, 871–884. https://doi.org/10.1038/s41467-018-03183-2

Varin, C., Luppi, P.-H., & Fort, P. (2018). Melanin-concentrating hor-mone-expressing neurons adjust slow-wave sleep dynamics tocatalyze paradoxical (REM) sleep. Sleep, 41(6), 1–12. https ://doi.org/10.1093/sleep/zsy068

Verret,L.,Gautagny,R.,Fort,P.,Cagnon,L.,Salvert,D.,Léger,L.,…Luppi,P.-H.(2003).Aroleofmelanin-concentratinghormonepro-ducingneuronsinthecentralregulationofparadoxicalsleep.BMC Neuroscience, 4,19.https://doi.org/10.1186/1471-2202-4-19

Zhang,Z.,Zhong,P.,Hu,F.,Barger,Z.,Ren,Y.,Ding,X.,…Dan,Y.(2019).An excitatory circuit in the perioculomotormidbrain for non-REMsleep control. Cell, 177, 1293–1307. https ://doi.org/10.1016/j.cell.2019.03.041

How to cite this article:LeeH-S,YamazakiR,WangD,etal.Targeted recombination in active populations as a new mouse genetic model to study sleep-active neuronal populations: DemonstrationthatLhx6+neuronsintheventralzonaincertaareactivatedduringparadoxicalsleephypersomnia.J Sleep Res. 2020;00:e12976. https ://doi.org/10.1111/jsr.12976