Pain hypersensitivity is associated with increased activity of peripheral and central neurons along the pain neuroaxis. We show that at the peak of acute inflammatory pain, superficial medullary dorsal horn projection neurons (PNs) that relay nociceptive information to the parabrachial nucleus reduce their intrinsic excitability and, consequently, action potential firing. When pain resolves, the excitability of these neurons returns to baseline. Using electrophysiological and computational approaches, we found that an increase in potassium A-current (I_A) underlies the decrease in the excitability of medullary dorsal horn PNs in acute pain conditions. In chronic pain conditions, no changes of I_A were observed, and medullary dorsal horn PNs exhibit increased intrinsic excitability and firing. Our results reveal an opposite modulation of the excitability of medullary dorsal horn projection neurons in acute and chronic pain conditions, suggesting a regulatory mechanism that, in acute pain conditions, fine tunes the output from the dorsal horn network and, if lacking, could facilitate pain chronification.

Methods

Animals 

Adult (9-13 weeks) C57BL/6J mice were used in all experiments unless stated otherwise. To specifically express Channelrhodopsin 2 (ChR2) in transient receptor potential vanilloid receptor 1 (TRPV1) expressing neurons, we used homogeneous TRPV1^cre mice (48) that were purchased from the Jackson Laboratory (#017769) and bred them in-house. Animals were group-housed under a controlled temperature (21 ± 2°C) and humidity (55 ± 10%) on a 12h-12h light/dark cycle, with ad libitum access to food and water. All experimental procedures were approved by the Institutional Animal Care and Use Committees at the Hebrew University of Jerusalem (study approval number MD-21-16152).

Pain models

Acute pain model: UV-photokeratitis model

Corneal photokeratitis was induced as previously described (28, 29). Briefly, mice were anesthetized with a combination of 70 mg/kg ketamine and 0.8 mg/kg medetomidine, intraperitoneally (i.p.). Mice were placed on a heating pad, and one eye was exposed to UV-C radiation (254 nm, 450 mJ/cm²), delivered through a UV light source (VL-4.C 230V 50/60Hz; Viber Lourmoat, Marne-la-Vallee, France), from a distance of 17 cm for 22 minutes. Following UV radiation, mice were injected with 1.3 mg/kg atipamezole for anesthesia reversal. Sham mice underwent the same procedure but without the UV radiation.

Chronic pain model: Chronic constriction injury of the distal infraorbital nerve (CCI-dION)

The CCI-dION model was performed as previously described (49, 50). In short, mice were anesthetized with a combination of 70 mg/kg ketamine and 0.8 mg/kg medetomidine (i.p.) and subcutaneous injection of 1-2 mg/kg meloxicam for analgesia. Then, mice were placed on a heating pad, and the skin below the infraorbital foramen (3 mm) was shaved and cleansed using 10% betadine. A 0.5 cm incision was made, and the nerve was gently exposed by blunt preparation. A chromic gut ligature (5-0) was loosely tied around the distal part of the ION. The skin incision was sutured, and mice were injected with 1.3 mg/kg atipamezole for anesthesia reversal. Sham mice underwent the same procedure of nerve exposure but without ligation.

Behavioral tests

All mice were handled for 5 consecutive days before the behavioral experiments. For 3 days before the experiments, as well as during experimental days, mice were habituated to head restriction by the experimenter and to the behavioral room (≥30 min) and recording chambers used (≥10 min).

The experimenter was blinded to mice groups in all the experiments and analyses.

 Assessment of ongoing pain: Eye-closing ratio

The eye-closing ratio is a measurement of the orbital tightening used to quantify ongoing pain (51–53). The eye-closing ratio was calculated as the ratio of the height and width of the tested eye. The height was defined as the distance between the edge of the upper and lower eyelids, and the width as the distance between the internal and external canthus. For eye-closing ratio measurements over the course of corneal UV-photokeratitis, mice were head-restricted and photographed. For eye-closing ratio measurements in the CCI-dION model, measurements were done 4 weeks following the CCI-dION / sham procedure. On a separate set of experiments, measurements were taken for baseline, 1, 2, and 7 days following the CCI-dION / sham procedure.

Assessment of nociceptive hyperalgesia

To assess the changes in sensitivity to a noxious stimulus, behavioral responses produced by corneal instillation of capsaicin were recorded under control and acute pain conditions. After habituation, mice were treated with a 10 µl drop of vehicle (0.1% ethanol in standard extracellular solution, SES) onto the tested eye and video recorded for 5 minutes in the recording chamber (a custom-made 10×20×25 cm transparent plexiglass). SES was composed of (in mM): 145 NaCl, 5 KCl, 2 CaCl₂, 1 MgCl₂, 10 glucose, and 10 HEPES. One hour later, 10 µl of 1 µM capsaicin (in SES) was applied to the tested eye, and the mouse was video recorded for 5 minutes in the recording chamber. Eye wipes were manually counted post hoc. Mice (N=2) that showed apathetic behavior, manifested by reduced motor activity, were noted by the blinded experimenter and were excluded from the analysis. Mice (N=2) that showed no eye wipe responses to both vehicle and capsaicin in baseline (day 0) were excluded from the analysis. In the trials in which mice were not in front of the camera, preventing correct detection of eye wipes for at least 30 seconds out of the first minute following capsaicin application, were excluded from analysis.

Assessment of mechanical allodynia 

To assess changes in sensitivity to innocuous mechanical stimulus, withdrawal responses to von Frey filaments applied to the middle area of the whisker pad were measured in mice in sham and CCI-dION conditions. Mechanical sensory thresholds were assessed using von Frey filaments of 0.07g to 6g by the simplified up-down (SUDO) method (54) while immobilizing the mice through neck pinch. A positive response was recorded if the mouse withdrew its head in 3 out of 5 trials for each filament.

Assessment of nocifensive behavior using optogenetics in freely behaving mice

4-6 weeks after TG injections, for 3 consecutive days prior to the experiment, mice were taken to the behavioral room, where we inserted into their home cage a custom-made miniature netted apparatus (size of 11×6×6 cm) for mice acclimatization with the apparatus and the room. After 1 hour, each individual mouse was put inside the apparatus for an additional 1 hour. 2-3 days prior to the experiment, the mice's cranium was shaved for the experiment. First, mice were anesthetized with a combination of 70 mg/kg ketamine and 0.8 mg/kg medetomidine (i.p.). Mice's cranium was carefully shaved, and the depilation was completed with hair removal cream. Finally, mice were injected with 1.3 mg/kg atipamezole for anesthesia reversal. On an experimental day, mice were taken to the behavioral room and acclimatized to the testing apparatus, as explained above. A 470-nm ultra high power collimated LED (Prizmatix, Israel) coupled to a multimode optical fiber (1 mm core diameter, 1-m length, prizmatix, Israel) was used for light stimulation (0.5-5 mW/mm²) onto the cranium of the tested mouse. The optical fiber tip was carefully positioned above the cranium with extra care so as not to touch the skin. We waited at least 30 seconds between light stimulations. Each combination of light intensity and duration was repeated 3-5 times. Experiments were video recorded and analyzed post hoc. For analysis, the percentage of withdrawal response was calculated for each combination of light intensity and duration. 1-3 days after the experiment, all tested mice were validated for successful viral infection by imaging the expression of EYFP in the fibers innervating the cornea using in-vivo corneal imaging (55, 56). We used the expression of the EGFP in the corneal afferents as evidence for the successful expression of the virus in TG neurons and nerve endings. Mice that did not show expression of the fluorophore in the corneal terminals were excluded from the analysis. During behavioral experiments and analysis, the experimenter was blinded to ChR2 / control injections.

Stereotactic injections

5-6 weeks old mice were anesthetized with a combination of 60 mg/kg ketamine, 0.5 mg/kg medetomidine, and 0.4-0.6% isoflurane (i.p.), pretreated with a subcutaneous injection of 1-2 mg/kg meloxicam and placed on a heating pad in a stereotactic frame (KOPF instruments). The cranium was shaved, and the skin was cleansed using 70% ethanol and 10% betadine. A 1-1.5 cm midline incision was made, followed by drilling a small hole in the skull using a fine drill burr. Unilateral injections to the PBN were delivered at the following coordinates (from bregma): ML: 1.3 mm, AP: −5.2 mm, DV: −3.75 mm using a microliter syringe (35g nanofil needle, World Precision Instruments, Sarasota, FL) loaded with retro-AAV-CaMKII-tdTomato (titer 4.5×10¹² genome copies per ml, ELSC vector Core Facility, Israel). 250 nl of the virus was injected at 100nl/min via UltraMicroPump (World Precision Instruments, Sarasota, FL). Injections to the TG were delivered at the following coordinates (from bregma): ML: 1.35 mm, AP: 0.4 mm, DV: −6.38. For TG injections, a pulled glass pipette was loaded with AAV1-EF1a-double floxed-hChR2(H134)-EYFP-WPRE-HGHpA or AAV1-EF1a-DIO-EYFP (titer 6×10¹¹ - 7×10¹¹ genome copies per ml, gc ml^-1) (Addgene, #20298, #27056 and ELSC vector Core Facility, Israel), and ~2µL/ganglion were injected using pressure. For both TG and PBN injections, pipette/needle were left at the injection site for an additional 5-10 minutes before being slowly retracted. Finally, the incision was sutured using chromic gut ligature (5-0) or glued by tissue adhesive (VetBond), and 1 mg/kg atipamezole was injected for anesthesia reversal. For both electrophysiological and immunohistological experiments, we injected the retro-AAV into either the right or left PBN and examined the corresponding ipsilateral medullary dorsal horn (24, 47). Dual injections to the PBN and TG (fig. S1D) were performed unilaterally, such that both injections were performed on the same side (right or left). For optogenetics behavioral experiments, the virus containing ChR2/EYFP was injected bilaterally.

 Electrophysiological recordings

Slice preparation

3-7 weeks after stereotactic injections, mice were deeply anesthetized with isoflurane, decapitated and the brain stem was quickly removed into a warm (34°C) artificial cerebrospinal fluid (aCSF) solution containing (in mM): 126 NaCl, 3 KCl, 1.3 MgSO₄, 1.2 KH₂PO₄, 26 NaHCO₃, 10 glucose, 2.4 CaCl₂, saturated with 95% O₂/5% CO₂. 300 µm coronal slices of caudal brain stem containing the medullary dorsal horn were prepared using a vibratome (VT 1200S, Leica). Slices were incubated in warm (34°C) aCSF solution for 1 hour, after which they were maintained at room temperature (23 ± 2°C) until recording.

Whole-cell recordings

Slices were transferred into a recording chamber and continuously perfused with oxygenated aCSF (the same as cutting solution). PNs of the medullary dorsal horn were visualized using differential interference contrast infrared microscopy (BX61WIF, Olympus, Tokyo, Japan) and identified by location and tdTomato expression using a CoolLed fluorescence excitation system. Borosilicate fire-polished patch pipettes (4-6 MΩ, pulled on a P-1000 puller, Sutter Instrument, USA) were used for all experiments. Signals were amplified using Multiclamp 700B (Molecular Devices) and digitized at 20-50 kHz (and low-pass filtered at 3 kHz) with Digidata 1440A (Molecular Devices) interfaced with pClamp 10.3 (Molecular Devices). Electrode series resistance (R_s) was routinely checked, and recordings were discarded if: (1) R_s was more than 30 MΩ; (2) R_s changed >15% during recordings, and (3) if resting membrane potential (RMP) was higher than −50 mV.

Current clamp recordings

The intrinsic properties of medullary dorsal horn PNs were recorded using internal solution containing either (in mM): 130 K-gluconate, 4 Na₂ATP, 0.5 NaGTP, 20 HEPES, 0.5 EGTA, 5 KCl (internal solution 1, 283 mOsm, pH was adjusted to 7.3 with KOH), or (in mM): 130 K-gluconate, 4 MgATP, 0.3 NaGTP, 10 HEPES, 0.5 EGTA, 5 KCl, 10 Na₂-phosphocreatine (internal solution 2, 291 mOsm, pH was adjusted to 7.3 with KOH). The recordings with either solution 1 or 2 produced similar results, and we combined the data from both experimental sets.

The resting membrane potential (RMP) was measured within 1-2 minutes after seal breaking. Once the membrane potential was stable, 3-10 seconds of the membrane potential was averaged and determined as the RMP.

To calculate input resistance (R_in), the neuron was held approximately at −60 mV, and a hyperpolarizing step of −10 pA for 400 ms was given. R_in was calculated by dividing the voltage response at the steady state by the current injection amplitude. At least 2 repetitions were used and averaged to calculate R_in.

For the f-I (frequency-intensity) curve, the membrane of medullary dorsal horn PNs was held approximately at −70 mV, and a 500 ms current step was given at 20 pA increments every 10 seconds. Firing type was classified by the following criteria (57): (1) tonically firing cells showed AP discharge throughout the current injection; (2) delayed firing cells had a delayed onset to the first AP and a hyperpolarizing notch after the initial depolarization at the start of the current injection; (3) phasically firing cells ceased AP firing in the second half of the current injection or showed only an initial burst of APs; (4) H-current cells showed a sag potential that peaked at ~55 ms following termination of the current injection; (5) Cells that could not be classified by one of the criteria above, were named unidentified. Only the properties of medullary dorsal horn PNs exhibiting delayed firing patterns were analyzed.

Action potential (AP) frequency was calculated by dividing the number of APs by the duration of the current injection (500 ms).

Instantaneous firing frequency was calculated as the reciprocal for the inter-spike intervals (ISIs) and averaged for each current step.

Regularity index was calculated by dividing the average instantaneous firing frequency by the regular firing frequency described above.

Coefficient of variation (CV) of firing was calculated by dividing the standard deviation of the ISIs by the average of the ISIs.

Adaptation ratio was calculated as the ratio of the first and last instantaneous firing frequency.

f-I curve slope was calculated by extracting the slope of the linear fit for the firing frequencies starting in the pulse preceding the first action potential firing.

Latency to the first AP was calculated as the time interval between the beginning of the current injection and the time point of the first AP threshold.

To calculate AP threshold, half-width, and after-hyperpolarization (AHP), the first spike of the first suprathreshold voltage response in the f-I protocol was analyzed. AP threshold was calculated using a phase plot analysis as previously described (14). The point in the phase plot where the increase in dV/dt was larger than 8 mV/ms was determined as the AP threshold.

AP half-width was determined as the width at half amplitude between the AP threshold and peak.

AP afterhyperpolarization (AHP) was determined as the difference between the AP threshold and the minimal membrane potential 5-10 ms after the AP peak.

AP amplitude and dV/dt_max were calculated for APs evoked by a short, 10 ms, current steps increasing by 10 pA increments from a holding potential of approximately −70 mV. The AP of the first suprathreshold voltage response was analyzed. The AP Amplitude was determined as the amplitude between the AP threshold and peak. dV/dt_max was determined as the maximal point of phase plot analysis.

Capacitance was calculated by first extracting the time constant from the protocol used to measure input resistance. To determine the time constant, we fitted a 1-exponential fit on the voltage response. Capacitance was calculated by dividing the time constant by input resistance.

Recordings of AP firing before and after application of 5 mM 4-aminopyridine (4-AP) were performed in the presence of 10 µM NBQX to block excitatory postsynaptic potentials (EPSPs). The recordings were performed at least 5 minutes following the perfusion of 4-AP.

Voltage-clamp recordings

For voltage-clamp recordings, data were sampled at 50 kHz and low pass filtered at 2 kHz. Recordings commenced at least 3 minutes after breaking the seal. Capacitive transient and leak currents were not compensated, and recordings were discarded if R_s was >20 MΩ or changed >15% during recordings.

For I_A recordings, 0.5 µM tetrodotoxin (TTX), 100 µM CdCl₂, 10 µM NBQX, and 100 µM picrotoxin were added to bath solution to block voltage-gated Na⁺ currents, Ca²⁺ currents, excitatory and inhibitory postsynaptic currents (EPSCs and IPSCs), respectively. The internal solution contained (in mM): 130 K-gluconate, 4 MgATP, 0.3 NaGTP, 10 HEPES, 0.5 EGTA, 5 KCl, 10 Na₂-phosphocreatine (291 mOsm, pH was adjusted to 7.3 with KOH).

To isolate I_A, membrane voltage was held at –80 mV in voltage-clamp mode. A two-step voltage protocol was used to determine the voltage-dependence of activation with a series of 500 ms depolarizing voltage steps in +10 mV increments to a maximum of +40 mV at 8-second intervals (fig. S11A, left). Next, a corresponding paired voltage step was given, consisting of the same voltage command but with a 150 ms prepulse to −10 mV (fig. S11A, right). I_A was extracted by the digital subtraction of each pair of pulses. I_A amplitudes were calculated from the isolated I_A traces as the current difference between the minimal current after the transient capacitive current and the peak of the current.

Voltage-dependence of I_A activation: I_A amplitudes were converted to conductances using the equation g = I/(V_m – V_rev) where V_m is the membrane voltage at the voltage step, and V_rev is the calculated K⁺ reversal potential (−88.85 mV). Normalized conductances were fitted with a Boltzmann equation: g/g_max = 1/(1+exp((V_1/2 – V) /k)), where g_max is the maximal conductance, V_1/2 is the half-maximal voltage and k is the slope factor.

Voltage-dependence of I_A inactivation was determined using incremental conditioning command of 150 ms prepulses from −100 mV to +20 mV, followed by a voltage step to +40 mV for 500 ms at 8-second intervals (fig. S11B). All traces were subtracted with the trace obtained at the final voltage step to zero residual currents. Current amplitudes were calculated as the difference between the peak amplitude and the mean steady-state current in the last 20 ms of the voltage command. Normalized currents were fitted similarly with the Boltzmann equation: I/I_max = 1/(1+exp((V_1/2 – V) /k)), where I_max is the maximal current, V_1/2 is the half-maximal voltage and k is the slope factor.

I_A window current was calculated as the area under the activation and inactivation curves for each cell.

 I_A kinetics were quantified for the isolated I_A traces at the peak conductance (+10 mV). Decay current was fitted with 1 or 2 exponents, depending on the fit with the minimal residual sum of squares between the data and the fit. For most cells, 2 exponents resulted in the best fit (Control: 7/9; Inflammation: 6/10; Sham: 8/14; CCI-ION: 5/7).

I_A charge was calculated as the area under the curve (integral) of the smoothed current response to the +10 mV voltage step.

For recording excitatory and inhibitory synaptic inputs following optogenetic stimulation of TRPV1-expressing afferents, we used an internal solution that contained (in mM): 115 Cs⁺-methansulfonate, 4 MgATP, 0.3 NaGTP, 10 HEPES, 1 EGTA, 1.5 MgCl₂, 10 Na₂-phosphocreatine, 2 QX 314-Cl, 10 BAPTA tetracesium (297 mOsm, pH was adjusted to 7.3 with CsOH). To isolate evoked excitatory and inhibitory synaptic responses, the membrane voltage was held at −70 mV and 0 mV, respectively. A 470-nm ultra high power collimated LED (Prizmatix, Israel) was used for light stimulation under the control of Master-9 (A.M.P.I, Israel). For the stimulation, we used the parameters that evoked nocifemsive responces in vivo (10 ms, 5 mW/mm²). Light was delivered through the optical path of the microscope and the ×40 objective. For each holding potential, five optogenetic stimulation were given with >30 seconds intervals between stimulation. We classified the response type to optogenetic stimulation by the following criteria: (1) an excitatory response was determined for neurons that showed 5/5 responses of inward currents (at a holding potential of −70 mV) with peak amplitudes that were >15 pA and <20 ms latencies (2) an inhibitory response was determined for neurons that showed 5/5 responses of outward currents (at a holding potential of 0 mV) with peak amplitudes that were >5 pA and because of the polysynaptic nature of the events latencies of <100 ms (3) excitatory and inhibitory response was determined for cells that passed both criteria 1 and 2.

All electrophysiological data were analyzed using custom scripts in Matlab (MathWorks, Natick, MA, USA).

Validation of functional expression of ChR2 in TRPV1-expressing TG neurons

ChR expression (conjugated with EYFP, see above the “Stereotactic injections” section) and function in TRPV1-expressing TG neurons were evaluated by recording from acutely dissociated EYFP-expressing TG neurons and behavioral assays (detailed above in the “Behavioral tests” section).

Electrophysiological recordings from EYFP-expressing acutely dissociated TG neurons

1.     Trigeminal ganglion cell culture preparation

Primary trigeminal ganglion (TG) neurons were isolated from TRPV1Cre-ChR2 mice 4-6 weeks following stereotactic injections. TG neurons (of both right and left TGs) were removed and placed into DMEM with 1% penicillin-streptomycin, then digested in 5 mg/ml collagenase, 1 mg/ml Dispase II (Roche). Cells were triturated with Pasteur pipettes in the presence of DNase I (250 U) and centrifuged through 15% BSA. The cell pellet was re-suspended in 1 ml Neurobasal media, containing B27 supplement (Invitrogen), 1% penicillin-streptomycin, 10 mM AraC, 2.5S NGF (100 ng/ml, Promega), and GDNF (2 ng/ml). Cells were plated onto poly-D-lysine (100 g/ml) and laminin (1 mg/ml) coated 35 mm tissue culture dishes. Cells were kept in the incubator at 37°C with 5% carbon dioxide.

2.      Whole-cell recordings

Recordings were performed from dissociated EYFP-expressing TG neurons, up to 24 hours after plating. Cells were visualized using an inverted eclipse Ti Nikon. Whole-cell membrane voltages were recorded using a Multiclamp 700B amplifier (Molecular Devices), at room temperature (24 ± 2°C). Data were sampled at 20 kHz and were low-pass filtered at 500 Hz (−3 dB, 8 pole Bessel filter). Patch pipettes (2-5 MΩ) were pulled from borosilicate glass capillaries (1.5 mm/1.1 mm OD/ID, Sutter Instrument, USA) on a P-1000 puller (Sutter Instrument, USA) and fire-polished (LWScientific). Data were digitized with Digidata 1440A A/D interface (Molecular Devices), interfaced with pClamp 10.3 (Molecular Devices).

Internal solution for whole cell patch recordings contained (in mM): 130 K-gluconate, 8 NaCl, 6 KCl, 10 HEPES, 2 Mg-ATP (290 mOsm, pH 7.25 adjusted with KOH).

The extracellular solution contained (in mM): 145 NaCl, 5 KCl, 1 MgCl2, 2 CaCl2 10 HEPES, 10 D-Glucose, pH was adjusted to 7.4 with NaOH.

Electrode series resistance (Rs) was routinely checked, and recordings were discarded if: (1) Rs was more than 20 MΩ; (2) Rs changed >15% during recordings, and (3) if resting membrane potential (RMP) was higher than −50 mV.

A 470-nm ultra-high power collimated LED (Prizmatix, Israel) was used for light stimulation under the control of pClamp 10.3 (Molecular Devices).

Action potential firing was examined in current clamp mode, using short (5 ms) optogenetic stimulation (5 mW/mm²), which was delivered through the optical path of the microscope and the ×40 objective. Next, we switched to voltage clamp mode in −70 mV holding potential and recorded light-evoked inward current in response to long (1 s) optogenetic stimulation. Finally, capsaicin (1 µM) was applied to determine that the neurons were TRPV1-sensitive nociceptors. 4 out of 5 EYFP-expressing neurons that responded to optogenetic stimulation by inward current and action potential firing also showed capsaicin-evoked inward current (fig. S1B).

Immunofluorescence staining

For spontaneous c-Fos expression, mice perfused 2 days after sham or UV radiation. For evoked c-Fos expression, 2 days after sham or UV radiation, 10 µl of 1 µM capsaicin (in SES) was applied to the eye, and mice were put in a separate cage for 5 minutes before putting back in the home cage. Mice were perfused 90 minutes after capsaicin instillation. In both spontaneous and evoked c-Fos expression experiments, mice were not habituated or handled before the experiments.

Mice were deeply anesthetized with isoflurane and perfused transcardially with 20 ml of cold phosphate-buffered saline (PBS) followed by 20 ml of cold 4% paraformaldehyde (PFA). Trigeminal ganglia (TGs) and brains were extracted and post-fixed in 4% PFA 2 hours and overnight, respectively, at 4°C. Next, tissues were rinsed in PBS (3×10 min), followed by cryoprotection in 30% (w/v) sucrose in PBS for at least 2 nights at 4°C. Next, brains and TGs were frozen in OCT for sectioning. For TG staining, 15 µm free-floating TG sections were washed (3x10 min) in PBS. Following the initial washing steps, the sections were blocked with CAS-Block universal blocking agent (#008120, ThermoFisher Scientific) for 10 min at room temperature. The sections were then incubated in goat anti-c-Fos primary antibody (1:500, #sc-52-G, Santa Cruz Biotechnology) diluted in the same blocking buffer overnight at 4°C. On the next day, the slices were washed (3x10 min) in PBS and incubated with Alexa Fluor 488-conjugated donkey anti-goat secondary antibodies (1:1000, #705545147, Jackson ImmunoResearch) diluted in freshly prepared PBS in 0.3% Triton X-100 (#X100, Sigma Aldrich) for 2 hours at room temperature. Final wash steps were performed (3x10 min) in PBS. All the above steps were carried out on a shaker with gentle shaking at 70 rpm. Finally, the sections were mounted onto the microscope slides and air-dried before covering them with VECTASHIELD antifade mounting media with DAPI (#EW9395224, Vector Laboratories). The slices were then coverslipped for imaging. Imaging was done with a confocal microscope (Nikon Ti Eclipse) using a 20X objective at a resolution of 2048 pixels. c-Fos expressing cells were identified using custom scripts in Cellprofiler (58) and validated by eye. For medullary dorsal horn staining, 30 µm free-floating sections (3-5 slices for each mouse) were washed in PBS and blocked in a solution containing 10% normal donkey serum and 0.3% Triton X-100 in PBS for 1 hour at room temperature. Rabbit anti-c-Fos primary antibody (1:500, #226 003, Synaptic Systems) was diluted in a solution containing 5% normal donkey serum and 0.3% Triton X-100 in PBS and incubated with sections overnight at 4°C. The following day, sections were rinsed in PBS (3x10 min) and incubated with donkey anti-rabbit Alexa Fluor 647 (1:1000, ab150063, abcam) secondary antibody diluted in a solution containing 5% normal donkey serum and 0.3% Triton X-100 in PBS for 2 hours at room temperature. Next, slices were incubated in DAPI (2 µg/ml, 10236276001, Roche Merck) in PBS (10 min), followed by two additional washes in PBS (2x10 min). Finally, slices were mounted (Dako fluorescence mounting medium) and coverslipped. Slices were imaged using Olympus IX83P2ZF with a 10x objective lens. For analysis, the medullary dorsal horn region was cropped for each slice using ImageJ. Using Cellprofiler custom scripts, c-Fos and tdTomato expressing cells were identified and validated by eye. For each mouse, the percentage of co-labeled cells was calculated as the number of c-Fos+ and tdTomato expressing cells from all slices out of the total number of tdTomato expressing cells or out of the total number of c-Fos+ cells. TG and medullary dorsal horn c-Fos stainings were done by an experimenter blinded to mice groups.

Computational modeling

Projection neurons were simulated using NEURON simulation environment. The mathematical parameters model are detailed in the Matherials and Methods section of the article and in the code in the database

All model parameters and the complete code used for simulation are available on Git Hub.

Statistics

Experiments were performed with side-by-side control and in a random order. No power analysis was used to determine experimental sample sizes in advance, but sample sizes were similar to those reported previously. The number of mice, slices, and cells are indicated in the Supplementary Tables accompanying each figure. Parametric and non-parametric tests were used where appropriate. All statistical tests, statistics (F or t), and P values are reported in figure legends or in the Supplementary Tables accompanying each figure. For the statistical analysis of eye wipes over time (Fig. 2C, middle) we fitted a linear mixed-effects model (MATLAB, function fitlme) using group, day, and the interaction of them for fixed effects and mouse ID for a random effect. For the statistical analysis of f-I curves, latency to 1^st AP, and intrinsic properties between 3 groups (Fig. 4), we fitted a linear mixed-effects model using group or condition (for fig. S10B) for a fixed effect. In repeated measurements structured experiments (f-I curves and latency to 1^st AP), we usually used mouse ID, neuron ID, and current injection for random effects unless stated otherwise in the statistics table. For intrinsic properties analysis in Fig. 4, B (inset), D (right) and E, we used mouse ID for a random effect. The random effects were varied specifically for each case to improve the fit of the model to the data. This was examined using compare function in MATLAB. All models are depicted in the statistics table. All statistical analyses were conducted with Matlab (MathWorks, Natick, MA, USA).