Mice
LepR-flox (LepRf/f) and LepR-Cre lines were previously generated and validated25,26 (JAX stock# 008327 and 008320). Cre-dependently re-activatable LepR-TB mouse line was previously generated27 and kindly provided by Dr. Joel Elmquist at UT Southwestern Medical Center (JAX stock# 018989). All mice were housed in University of Iowa’s vivarium in temperature-controlled environment (12h light and 12h dark cycle) with ad libitum access to food and water. All animal protocols were approved by the University of Iowa’s Institutional Animal Care and Use Committee and are in accordance with NIH guidelines for the use and care of Laboratory Animals. LepRf/f and LepR-Cre mice used in the present study were maintained on a mixed background while LepR-TB mice were maintained on a pure C57BL/6 background. Male mice were used for chemogenetics/optogenetics, fiber photometry, and electrophysiology experiments, while both sexes of mice were used for body weight and sleep architecture analysis in LepR-flox and LepR-TB mice. Overall health of experimental mice was monitored on a daily basis and any mice displaying obvious signs of distress or weakness were removed from the study. Additionally, sentinel mice were periodically screened for pathogenesis.
Stereotaxic surgery
Stereotaxic surgery was performed as previously reported28,29. Mice were deeply anesthetized by intraperitoneal (IP) injection of ketamine/xylazine (100:10 mg/kg) and placed on a double arm Kopf stereotaxic apparatus (David Kopf Model #902). Following standard disinfection procedure, a ~ 1.0 cm incision was made to expose the skull and a small hole was drilled into the skull at defined positions (LHA, AP -1.5 mm, ML + 1.9 mm, DV -5.3 mm with 10-degree injection arm; VTA, AP -3.9 mm, LM + 1.2 mm, DV -3.9 mm with 10-degree injection arm; VLPO, AP 0.1 mm, ML + 1.6 mm, DV -5.6 mm, with 10-degree injection arm). A Hamilton Neuros syringe (Hamilton; Part #65458-01) or a pulled glass pipette filled with viral vector was slowly inserted to reach the targeted brain region and a small volume injection (200–250 nl) was made either by pressing the plunger (for Hamilton syringe) at the rate of ~ 50 nl/min or by applying pulse pressure (for glass pipette) with microINJECTOR™ pressure system (Tritech Research). After 10 minutes of waiting to ensure full penetration of AAV into the targeted area, the needle was slowly removed, and the incision was closed by wound clips. Mice were then kept on a warming pad until awake and then returned to their home cages.
For deletion or restoration of LepRs in the LHA, both sexes of 8-week-old LepR-flox or LepR-TB mice were received bilateral microinjection of either AAV2-CMV-eGFP (Addgene #105530) or AAV2-CMV-eGFP-Cre (Addgene #105545) into the LHA using a Hamilton Neuros syringe. After viral injection, mice were group housed and underwent weekly body weight measurements for up to 11 weeks, followed by sleep recording as described below.
For chemogenetic activation experiments, 10-12-week-old LepR-Cre+ male mice were received bilateral microinjection of either AAV2-hM3Dq-mCherry (Addgene #44361)) or AAV2-DIO-mCherry (Addgene #50459) into the LHA using a Hamilton Neuros syringe. After 2–3 weeks of recovery from the surgery, mice were subjected to chemogenetic activation experiments by receiving an intraperitoneal (IP) injection of clozapine-N-oxide (CNO, Tocris #4936).
For optogenetic circuit mapping experiments, 10-12-week-old LepR-Cre+ male mice were injected with AAV2-DIO-hChR2(E123T/T159C)-eYFP (UNC Vector Core) into the LHA. Optical cannulas (Ø200 um, 0.22 NA; Thor lab #CFML22U) were then implanted to target either VLPO (AP 0.1 mm, ML + 1.6 mm, DV = -5.6 mm) or VTA (AP -3.9 mm, LM + 1.2 mm, DV -3.9 mm) and permanently fixed in place by applying dental cement for subsequent photoactivation.
For in vivo Ca2+ imaging by fiber photometry, 10-12-week-old LepR-Cre+ male mice were unilaterally injected with AAV1-hSyn-GCaMP6m-GFP (Penn Vector Core) into the LHA using a Hamilton Neuros syringe. Four weeks after viral infection, mice were implanted with a fiber optic cannula (Ø200 µm core and 0.39 NA; Thor labs, #CFML14U-20) to target the LHA for subsequent Ca2+ imaging.
For anterograde tract-tracing of LHALepR+ neurons, 10–12-week-old LepR-Cre+ male mice were unilaterally injected with AAV2-DIO-hChR2(E123T/T159C)-eYFP (UNC Vector Core) into the LHA using a pulled glass pipette connected to a microINJECTOR™ pressure system (Tritech Research). Four weeks after viral transduction, mice were transcardially perfused, and the brains were extracted and processed for FIHC to detect eYFP-positive fibers throughout the brain as described below.
EEG/EMG headmount implantation for polysomnographic recording and sleep architecture analysis
Mice were deeply anesthetized with IP injection of ketamine/xylazine (100:10 mg/kg) and placed on a Kopf stereotaxic apparatus (David Kopf Model #902). Following standard disinfection procedure, ~ 1.5 cm incision was made to expose the skull and prefabricated EEG/EMG headmounts (Pinnacle Technology; #8201) were fixed on the skull by four anchoring and conducting screws (2 in frontal cortices and 2 in parietal cortices; Pinnacle Technology; #8209 and #8212). EMG electrodes were inserted into the nuchal muscle and sutured in place. A layer of dental cement was added to protect the exposed skull and secure the headmount. Mice were then kept on a warming pad until fully conscious before returning to their home cages.
EEG/EMG signals were recorded with a wireless EEG/EMG amplifier (Pinnacle, #8200-K9-SL). Signals were amplified and digitized at 512 Hz using Sirenia Acquisition software. Polysomnographic recordings were then scored semi-automatically: states were initially classified by cluster scoring through Sirenia Sleep Pro (Pinnacle Technology), then manually reviewed and scored based on the visual signatures of the EEG and EMG waveforms, as well as the power spectra of 10s epoch. Wakefulness was defined by high-frequency, low-amplitude EEG, and high tonic activity EMG. NREM was defined by low-frequency (0.5–4 Hz), high-amplitude EEG, and low activity EMG when compared to wakefulness. REM was defined as having heightened theta power (5–9 Hz) and low activity EMG. A bout of sleep or wake was automatically defined by Sirenia Sleep Pro as a period during which three or more consecutive 10s epochs were observed. All mice were habituated to the wireless EEG/EMG headset and housing chambers at least overnight before sleep recording. Mice were then recorded for at least for one full light (6 AM – 6 PM) and dark (6 PM – 6 AM) cycle as follows. For the evaluation of the effects of LHALepR−KO on sleep-wake regulation, mice were subjected to a 24h baseline polysomnographic recording and then received IP injection of leptin (3 mg/kg) first at light cycle (9 AM) and then again at dark cycle (6 PM). For the evaluation of the effects of LHALepR−Re on sleep-wake regulation, mice were subjected to a 24h baseline polysomnographic recording. For the chemogenetic activation of LHALepR+ neurons, mice were received IP injection of CNO (2 mg/kg) first at light cycle (9 AM) and then again at dark cycle (6 PM).
Food intake measurement
The effect of chemogenetic activation of LHALepR+ neurons on feeding was evaluated in customized home cages outfitted with in-cage Feeding Experimentation Devices (FED3.1, Kravitz lab) as described previously28,30. Mice were given 2 days habituation period to consume 20 mg food pellet (F0163, Bio-Serv) from FED3 pellet tray, under fixed ratio 1 (FR1) schedule (one correct nose poke to deliver one 20 mg food pellet). On the experiment day, mice were given an IP injection of CNO (2 mg/Kg) either at the onset of the dark cycle (6 PM) or at light cycle (9 AM) and the consumption of pellets was continuously monitored during the subsequent 5-hour period.
Locomotor activity counts
Locomotor activity was evaluated using multi-channel activity monitor (Columbus Instruments #Opto M4). Mice were habituated overnight inside of customized chambers placed within a beam break system with food and water provided ad lib. On the testing day, mice were received a single IP injection of CNO (2 mg/Kg) at 9 AM and the activity was monitored during the subsequent 4-hour period. Data was acquired and analyzed using Columbus instruments Oxymax software.
Fiber photometry recording and analysis
Three weeks after AAV1-hSyn-GCaMP6m-GFP infection into the LHA as described above, mice were implanted with a fiber optic (Ø200 µm core and 0.5 NA; Thor labs, #CFML14U-20) and an EEG/EMG mouse headmount (Pinnacle, #8201-SS-270) for simultaneous calcium and polysomnographic recordings. A Sirenia sleep preamplifier was connected to the EEG/EMG headset along with a low autofluorescence mono fiber-optic patch cord (400 mm, 0.5 NA, Doric Lenses Inc. FCM-MF 1.25_LAF) which was mated on to the ferrule fixed on the skull for GCaMP recording. A TDT RZ10X system was used for time-locked Ca2 + and EEG/EMG recordings. Calcium activity was simultaneously acquired at a sampling frequency of 1017 Hz using TDT Synapse software. Excitation wavelengths used were 470 nm for the calcium-dependent GCaMP6m signal, temporally interleaved with 405 nm light at 1017 Hz for the calcium-independent (isosbestic) signal. Custom MATLAB script used to analyze the photometry data was adopted from TDT’s analysis pipeline (https://www.tdt.com/docs/fiber-photometry-user-guide/data-analysis/). Briefly, data was first downsampled to 1 Hz and corrected for nonspecific autofluorescence and movement artifacts31. For each recording, ΔF/F (%) was calculated using a least-squares linear fit, after which the 405 nm signal was scaled and fitted to the 470 nm one before subtraction to correct for any motion artifacts and normalize the 470 nm signal, as follows: ΔF/F = 100 * (GCaMP signal - fitted Isosbestic signal) / fitted Isosbestic signal. Each mouse underwent multiple times of a 2h recording between 10 AM – 2 PM, of which only sessions during which the animal entered all sleep-wake states (wake, NREM, REM) were included in the final analysis. In subsequent analysis, we calculated the onset mean ΔF/F (%) per bout for wake, NREM, and REM for all the animals. For state transition analysis, we identified all points of state transition and aligned the ΔF/F trace around these times. All transitions across state were pooled to calculate the mean response to the respective change in state in each animal.
Simultaneous miniscope Ca2+ imaging of LHALepR+ neurons and polysomnographic recording
LepR-Cre+ male mice were received a microinjection of AAV1-hSyn-GCaMP6m into the LHA (AP -1.5 mm, ML + 1.9 mm, DV -5.3 mm with 10-degree injection arm) using a glass pipette injection system (Tritech Research, Inc). Immediately following viral injection, mice underwent the implantation of three anchoring stainless screws into the skull (A/P: 1.5 mm, M/L: -1.5 mm; A/P: 1.5 mm, M/L: +1.5 mm; A/P: -3.5 mm, M/L: -3 mm) for subsequent EEG/EMG recording. They then underwent tissue suction (approximately 1.5-2 mm) using controlled vacuum connected to sterilized 27G needle to release intracranial pressure before the implantation of Gradient index (GRIN) lens. The GRIN lens (600 µm diameter, 7.3 mm length; Inscopix, CA, USA) was slowly inserted (over the course of 15 min) to target the LHA and then secured in place first with superglue followed by dental cement. GRIN lens was topped first with thick layer of Kwik-cast (Kwik-Cast Silicone Elastomer, World Precision Instrument) and then with dental cement to prevent the lens damage. After 3–4 weeks of recovery from the surgery, mice underwent baseplate implantation, and the GCaMP signal was visualized by a miniscope V4 (Open Ephys) coupled to the GRIN lens. The implantation of customized EEG/EMG headcounts was only performed for those mice (n = 4) that showed a reasonable number of neurons with Ca2 + signal in the field of view (FOV). A total of 8 mice, which showed either no or few neurons with GCaMP signal in the FOV, were excluded from the study. For simultaneous miniscope and EEG/EMG recordings, mice that received EEG/EMG headmount implantation surgery were first habituated to wearing the miniscope and EEG/EMG wireless devices for at least 3–5 days in open-top home cages before the actual experimental recording. About 1–2 hours of simultaneous miniscope and EEG/EMG recordings were then performed between 10AM – 2PM. Each mouse underwent a total of five trials of 1–2 hours recording. A 5V TTL pulse was used to synchronize Ca2+ recording to the polysomnography traces. Sleep stages were scored offline based on EEG/EMG traces with 10s epoch, and only trials showing all three sleep-wake states (NREM, REM and Wake) were processed for the GCaMP data analysis as follows. The acquired Ca2 + signals were motion corrected to stabilize the image sequences using rigid version of NoRMCorre algorithm32. The neuron identification and demixed GCaMP activity time series were extracted with CNMF-E software package33. The CNMF-E algorithm uses constrained non-negative matrix factorization and optimized to single-photon endoscopic data. The contours of neurons were generated by the CNMF from spatial component matrix. The timestamps of GCaMP traces and EEG signals for sleep stages were synchronized, and the subsequent analysis of calcium transients and z-score integrals of each neuron followed the procedures as previously reported34. Wake- and/or REM-active neurons are defined by an increase in average GCaMP activity during wakefulness or REM sleep that exceeds 60% compared to NREM sleep.
Sleep Deprivation
Sleep deprivation was carried out using automated sleep deprivation chambers (Pinnacle; #9000-K5-S). The chambers were 10” in diameter and equipped with a motorized rotating bar programmed to reverse rotation direction automatically every 10–40 seconds. To identify LepR projections to sleep active neurons (determined by c-Fos staining during sleep recovery), we first injected AAV2-DIO-hChR2(E123T/T159C)-eYFP (UNC Vector Core) into the LHA of LepR-Cre+ mice and then divided them into two groups: a 24h sleep deprivation (SD) group and an SD plus 4h sleep recovery group. All the mice were first acclimatized to the SD chambers within a controlled environment of light and temperature, while having ad libitum access to food and water and then underwent 24h SD from 6 AM. After 24h SD period, one group of mice were immediately sacrificed, while another group of mice were allowed to sleep for 4h before being sacrificed. The brains were then processed for histological examination of eYFP and c-Fos to assess the projections of LHALepR+ neurons to the POA, where sleep-active neurons are located.
Functional optogenetic circuit mapping
Three weeks after AAV2-DIO-hChR2(E123T/T159C)-eYFP injection into the LHA as described above, mice were implanted with a fiber optic cannula (Ø200 um, 0.22 NA; Thor lab #CFML22U) and an EEG/EMG mouse headmount (Pinnacle, #8201-SS-270) for simultaneous optogenetic stimulation and polysomnographic recordings. All optogenetic circuit mapping experiments were conducted during the light cycle. Each testing mouse was first equipped with a wireless EEG/EMG recording device and connected to an optic patch cable (200 um, 0.22 NA; Doric Lenses) for the delivery of laser light (Shanghai Laser & Optics Century Co., Ltd), and then acclimatized overnight in a customized round acrylic chamber prior to optogenetic stimulation. On the testing day, NREM sleep state was first identified by ongoing EEG/EMG traces and the laser light (blue stimulation light at 473 nm or green control light at 520 nm) with 15 mW power and 5 ms light pulses at either 5 or 20 Hz was turned on manually for optogenetic stimulation after confirming at least 30s of NREM sleep. Light stimulation was stopped immediately if an arousal response was detected or continued for up to 100s in the absence of an obvious arousal response. Arousals were defined as periods of at least 5s of wakefulness, determined by EEG/EMG traces and power spectral analysis. At least 5 minutes of rest were provided between each light stimulation trial. Trials in which the mouse woke up immediately after stimulation were discarded due to the stimulation overlapping with the mouse’s natural wake behavior.
Electrophysiology
LepRCre+ male mice were injected with 200 nl of rgAAV-FLEX-tdTomato (Addgene #28306) into the POA (AP 0.1 mm; ML + 1.6 mm; DV -5.5 mm) or the VTA (AP -3.9 mm; LM + 1.2 mm; DV -3.9 mm) with 10-degree needle angle using pulled glass pipette connected to microINJECTOR™ pressure system (Tritech Research). After four weeks of viral injection, mice were sacrificed, and acute brain slices were prepared for slice electrophysiological recordings. Brains were immersed in NMDG-HEPES aCSF cutting solution (in mM): 92 NMDG, 2.5 KCl, 1.25 NaH2PO4, 30 NaHCO3, 20 HEPES, 25 glucose, 2 thiourea, 5 Na-ascorbate, 3 Na-pyruvate, 0.5 CaCl2∙2H2O, and 10 MgSO4∙7H2O. The brain was kept in 95% O2 / 5% CO2 aerated ice-cold cutting solution during the slice preparation with vibratome (Campden Instruments; #7000smz-2). Fresh brain slices (approximately 250 µm in thickness) containing LHA were obtained and transferred to a 95% O2/5% CO2 aerated HEPES artificial cerebrospinal fluid (aCSF) incubation solution consisting of (in mM): 92 NaCl, 2.5 KCl, 1.25 NaH2PO4, 30 NaHCO3, 20 HEPES, 25 glucose, 2 thiourea, 5 Na-ascorbate, 3 Na-pyruvate, 2 CaCl2∙2H2O, and 2 MgSO4∙7H2O. Brain sections were incubated in this solution for at least 30 min and then placed in the recording chamber which has the recording aCSF consisting of (in mM): 124 NaCl, 2.5 KCl, 1.25 NaH2PO4, 24 NaHCO3, 12.5 glucose, 5 HEPES, 2 CaCl2∙2H2O, and 2 MgSO4∙7H2O. Loose-seal (cell-attached) recordings were performed on retrogradely labeled tdTomato-expressing LHALepR+ neurons using electrodes with 4–5 MΩ tip resistance. aCSF was used as the pipette solution. Synaptic blockers CNQX (10 µM) + AP5 (50 µM) added to block excitatory transmission and PTX (50 µM) was added to block GABAA-receptors. For each cell, at least 5 min of baseline spontaneous activity was recorded to achieve a stable baseline followed by perfusion of 100 nM leptin in the bath. MultiClamp 700B Amplifier (Molecular Devices, San Jose, CA) and Axon pCLAMP 11 software (Molecular Devices, San Jose, CA) were used to obtain and analyze data.
Histology
All mice received the stereotaxic microinjections of AAVs or the implantations of fiber optic cannula were subjected to histological verification to confirm viral expression and precise stereotaxic targeting. Mice were transcardially perfused with ice-cold PBS and then with 10% neutralized formalin for fixation. In the case of leptin-induced pSTAT3 analysis, mice received an IP injection of leptin (3 mg/kg) one hour prior to perfusion. The extracted brains were fixed in 10% neutralized formalin for an additional 3–4 hours before being immersed overnight in a 25% sucrose solution in PBS buffer (pH 7.4). Brains were then sectioned into five series of 30 µm sections using a freezing microtome (Leica SM2010 R) and stored in cryoprotectant solution at -20˚C until further processed for histological examination. For the verification of correct targeting of fiber optic cannula, brain sections were directly mounted on a coated glass slide to evaluate the insertion track of fiber optic cannula. For the visualization of other marker proteins of interest, brain sections were processed for fluorescence immunohistochemistry (FIHC) as previously reported29,35. Briefly, one series of brain sections stored in cryoprotectant solution were rinsed with PSB 10 min for 6 times and then incubated in blocking solution (3% normal donkey serum and 0.3% Triton X-100, 0.002% NaN3) for 1 hour at room temperature (RT). Sections were then incubated with primary antibodies against c-Fos (goat anti-c-Fos antibody, 1:1000, Santa Cruz #sc-52, or rabbit anti-c-Fos antibody, 1:1000, CalBioChem #PC38), mCherry (rabbit anti-DsRED antibody, 1:1000, Clontech #632496), GFP (chicken anti-GFP Aves Lab Inc, 1:1000, #GFP-1020,), or orexin (goat anti-Orexin-A antibody, 1:1000, Santa Cruz #SC-8070) overnight at 4˚C. FIHC for pSTAT3 was followed as previously reported29,35. Briefly, brain sections were rinsed in PBS and treated with 1% NaOH, 1% H2O2 in distilled water for 20 minutes at RT, and then incubated with 0.3% glycine in PBS for 10 minutes, followed by 10 minutes incubation in 0.03% sodium dodecyl sulfate (SDS) in PBS and then blocking solution (3% normal donkey serum and 0.3% Triton X-100, 0.002% NaN3) for 30 minutes at RT. Sections were then incubated with primary antibody against pSTAT3 (rabbit anti-pSTAT3, 1/1000, Cell Signaling #9131) for 24 hours at RT plus an additional 48 hours at 4 ◦C. After the primary antibody reaction, all sections were washed and incubated with secondary antibodies conjugated to Alexa Fluor 488, Alexa Fluor 594, or Dylight 649 (1:500, Jackson ImmunoResearch) for fluorescent visualization as per the manufacturer’s instructions. All sections were then mounted on the gelatin-coated slides and visualized using Olympus slide scanner (VS120, Olympus). Slides were carefully examined to ensure precise stereotaxic targeting of viral vector. All miss-targeted cases were excluded for subsequent data analysis.
Cell counting
Images were acquired using Olyvia software associated with an Olympus slide scanner. One series of sections covering the rostral to caudal extent of the LHA were exported as TIFF files and analyzed using Adobe Photoshop. All images were adjusted for the same levels of intensity and contrast. Manual cell counting was performed using the cell counter plugin using Adobe Photoshop.
Statistical analysis
All data sets are presented as the mean ± SEM, and individual data points are shown for behavioral experiments wherever necessary. Multiple unpaired t-tests, one-way or two-way ANOVAs with Sidak’s or Tukey’s multiple comparisons tests, were used to compare sleep stages, body weight, food intake, and GCaMP activity changes between two or three groups, as necessary. Experimenters were blinded to the group analysis. Data from the electrophysiological experiment were analyzed using Clampfit v10.6, and leptin-induced changes in firing rate from the baseline were compared using a paired t-test. All statistical analyses were performed using Prism 9 software by GraphPad.