The Increased Activity of CaV3.2 Channels Contributes to Nociceptors’ Hyperexcitability and Chronic Neuropathic Pain Following Spinal Cord Injury in Mice

Abstract

Following spinal cord injury (SCI), up to 60–70% of patients develop chronic neuropathic pain. SCI-induced neuropathic pain (SCI-NP) is often lifelong and therapeutically intractable, leading to a severe decline in quality of life and increased risk for depression, anxiety, and addiction. Studies in preclinical rat models of SCI support the contribution of nociceptors’ hyperexcitability and their spontaneous activity (SA) in driving the development/maintenance of SCI-NP and suggest that reducing nociceptors’ hyperexcitability could provide a therapeutic strategy to treat SCI-NP. Previous data from our laboratory collected from SCI rats showed that the increased activity of T-type calcium channels induced by SCI contributes to drive nociceptors’ hyperexcitability and their SA in vitro and pain hypersensitivity in vivo, supporting a contribution of T-type calcium channels in driving nociceptors’ hyperexcitability and the development/maintenance of SCI-NP. The data presented here in a mouse model of SCI confirm a similar increase in nociceptors’ excitability, supporting a conserved mechanism(s) between species. Data collected in SCI CaV3.2−/− mice showed reduced nociceptors’ excitability in vitro and reduced mechanical hypersensitivity and spontaneous pain in vivo as compared with SCI wild-type mice, consistent with a contribution of CaV3.2 channels in driving nociceptors’ hyperexcitability and SCI-NP. Data with behavioral pharmacology in vivo showed that TTA-P2 (a blocker of T-type calcium channels) and gabapentin (inhibitor of trafficking of high voltage-activated calcium channels) reduced mechanical hypersensitivity in male and female SCI mice, while only a TTA-P2–sensitive component of spontaneous pain was observed.

Significance Statement

Development of chronic neuropathic pain following spinal cord injury (SCI) is an unmet medical condition affecting up to 60–70% of patients. SCI-induced neuropathic pain (SCI-NP) is often lifelong and therapeutically intractable, leading to a severe decline in quality of life. The lack of adequate treatment for SCI-NP highlights our limited understanding of the underlying mechanism(s) responsible for the development/maintenance of SCI-NP. Nociceptors’ hyperexcitability has been proposed to drive the development/maintenance of SCI-NP. Here we provide evidence that CaV3.2 channels play a contribution in driving nociceptors’ hyperexcitability in vitro and mechanical hypersensitivity and spontaneous pain in vivo in a mouse model of SCI, suggesting a possible pharmacological target to reduce SCI-NP.

Introduction

Chronic neuropathic pain is a severe secondary complication following spinal cord injury (SCI) affecting up to 60–70% of patients (Finnerup et al., 2001, 2015; Siddall and Loeser, 2001; Barrett et al., 2003; Cragg et al., 2015; van Gorp et al., 2015; Burke et al., 2017). Anticonvulsant and antidepressant drugs are used as first analgesic choice to treat SCI-induced neuropathic pain (SCI-NP; Finnerup et al., 2001, 2015; Attal et al., 2002; Putzke et al., 2002; Cardenas et al., 2004; Levendoglu et al., 2004; Siddall et al., 2006; Rintala et al., 2007; Baastrup and Finnerup, 2008; Vranken et al., 2008; Agarwal and Joshi, 2017; Colloca et al., 2017), but patients report unsatisfactory pain relief (usually <30%; Ravenscroft et al., 2000; Murphy and Reid, 2001; Cardenas et al., 2004; Rintala et al., 2007). SCI-NP may become in many cases lifelong and therapeutically intractable (Stormer et al., 1997; Finnerup et al., 2001, 2015; Siddall and Loeser, 2001; Barrett et al., 2003; Werhagen et al., 2004; Jensen et al., 2007; Cohen and Mao, 2014; Andresen et al., 2016; Bavencoffe et al., 2016; Colloca et al., 2017), leading to a severe decline in quality of life and increased risk for depression, anxiety, and addiction (Cragg et al., 2015; Andresen et al., 2016; Velly and Mohit, 2018).

In vivo and in vitro data in a rat model of SCI have provided support that hyperexcitability and spontaneous activity (SA) of nociceptors play a major role in the development/maintenance of SCI-NP (Bedi et al., 2010; Walters, 2012; Yang et al., 2014; Bavencoffe et al., 2016; Wu et al., 2017; Lauzadis et al., 2020; Liu et al., 2023), and there is a strong relationship between the severity of pain and the incidence of SA of nociceptors (Bedi et al., 2010). Several mechanisms have been suggested to trigger nociceptors’ hyperexcitability following SCI, including changes in the activity of NaV1.8, Kv3.4, and TRPV1 channels (Wu et al., 2013; Yang et al., 2014; Ritter et al., 2015), change in the activity of EPAC1 and EPAC2 (Berkey et al., 2020), and continuing cAMP-PKA signaling (Bavencoffe et al., 2016). In addition to the above mechanisms, data from our laboratory collected in a rat model of SCI showed that the increased activity of T-type calcium channels induced by SCI contributes to drive nociceptors’ hyperexcitability and their SA in vitro and pain hypersensitivity in vivo (Lauzadis et al., 2020; Liu et al., 2023), supporting a contribution of T-type calcium channels in driving nociceptors’ hyperexcitability and the development/maintenance of SCI-NP.

All three isoforms of T-type calcium channels (CaV3.1, CaV3.2, and CaV3.3) are expressed in the peripheral nervous system (Talley et al., 1999; Shin et al., 2003; Todorovic and Jevtovic-Todorovic, 2006; Watanabe et al., 2015; Rzhepetskyy et al., 2016; Candelas et al., 2019). CaV3.2 channels play a major role in the development of neuropathic pain in several preclinical animal models of peripheral neuropathies such as the spared nerve injury model (Chen et al., 2018; Kang et al., 2018), the chronic constriction injury model of the sciatic nerve (Bourinet et al., 2005; Jagodic et al., 2008), chemotherapy-induced peripheral neuropathy (Tomita et al., 2019), the spinal nerve injury model (Yue et al., 2013), and diabetic peripheral neuropathy (Messinger et al., 2009; Obradovic et al., 2014; Todorovic and Jevtovic-Todorovic, 2014, suggesting that the increased activity of CaV3.2 channels may also be responsible for driving nociceptors’ hyperexcitability and the development/maintenance of SCI-NP.

Here, we used in vitro patch-clamp electrophysiology and in vivo behavioral pharmacology combined with measurements of mechanical hypersensitivity and spontaneous pain in a mouse model of SCI to determine the contribution of CaV3.2 channels to the development/maintenance of SCI-NP. We found that the increased activity of CaV3.2 channels induced by the injury contributes to drive nociceptors’ hyperexcitability and their SA in vitro. Behavioral experiments in vivo showed reduced neuropathic pain in the SCI CaV3.2−/− mice.

Materials and Methods

Animals

C56BL/6J male and female mice (12–14 weeks old at the time of injury) were used in this study. All procedures were performed in strict accordance with the guidelines of the International Association for the Study of Pain and were approved by the Stony Brook University Institutional Animal Care and Use Committee. Mice were deeply anesthetized with isoflurane before decapitation.

SCI

Animals were anesthetized with isoflurane in 5% O₂. Once there was no longer a paw pinch response, the hairs of the thoracic and lumbar spinal region were sheared. A midline incision was made on the back, and the paravertebral muscles at the level of the T9–T11 vertebrae were gently separated from the spinous processes of the vertebrae and retracted. The T9 and T11 dorsal processes were braced to stabilize the spine, and a dorsal laminectomy was carried out at T10 with a bone rongeur to expose the spinal cord. SCI was then performed by spinal cord contusion by using an Infinite Horizon Impactor (Precision Systems). A steel rod (0.5 mm diameter) attached to the impactor was used to apply a force of 50 kilodynes to the midline dorsal aspect of the spinal cord to generate the injury. Following the injury, the muscles were sutured, and the skin wound was closed with metal wound clips and treated with betadine solution. Animals received baytril (2.5 mg/kg, i.m.) twice daily for 5 d and meloxicam (5 mg/kg, i.m.) once daily for 3 d. Manual expression of the bladder was performed twice daily until voiding was regained (usually 2–3 weeks following the spinal impact).

Sham laminectomy

The surgical procedure was similar to that described above for SCI, but animals received only the laminectomy at T10 without the contusion.

Short-term culture of dorsal root ganglia (DRG) neurons

Mice were deeply anesthetized with isoflurane and decapitated. A dorsal laminectomy was performed to expose the lumbar aspect of the spinal cord and DRG. DRG (L3, L4, and L5) were pulled out and placed in a cold Ca²⁺, Mg²⁺-free Hank's solution containing (in mM): 137 NaCl, 5.3 KCl, 0.33 Na₂HPO₄, 0.44 KH₂PO₄, 5 HEPES, 5.5 glucose, pH 7.4, with NaOH. Ganglia were cut in half and incubated for 20 min at 34°C in Ca²⁺, Mg²⁺-free Hank's solution containing 20 U/ml papain (Worthington Biochemical) and 5 mM dl-cysteine. Ganglia were then treated for 20 min at 34°C with 3 mg/ml collagenase (Type 1, Sigma-Aldrich) and 4 mg/ml Dispase II (Boehringer Mannheim) in Ca²⁺, Mg²⁺-free Hank's solution. Ganglia were then washed with Leibovitz's L-15 medium (Invitrogen). Individual cells were dispersed by mechanical trituration using fire-polished Pasteur pipettes with decreasing bore size and plated on glass coverslip treated with 30–50 μg/ml poly-d-lysine. Cells were incubated in the L-15 solution supplemented with 10 ng/ml of NGF at 34°C (in 5% CO₂) and used within 24 h. Small DRG neurons (diameters < 27 µm) were chosen for recording. Small DRG neurons were selected by measuring the diameter from images captured to a computer by a CCD camera (Oly-150, Olympus Imaging America) using a video acquisition card (dP dPict Imaging). Cell capacitance was measured by integrating the average of 5–10 capacitive currents in response to a −5 mV step from −80 mV filtered at 10 KHz and acquired at 50 KHz.

Cell classification

At the end of each recording, DRG neurons were tested for capsaicin sensitivity and classified as TRPV1(+) (those responding to capsaicin), corresponding to a subset of nociceptors (Cardenas et al., 1995; Caterina et al., 1997; Petruska et al., 2000) or TRPV1(−) (those insensitive to capsaicin). Further classification was made by testing the ability of TRPV1(+) or TRPV1(−) DRG neurons to bind the isolectin B4 (IB4) FITC conjugate. Following the capsaicin test, the cell was challenged for 3 min with a modified Tyrode's solution [containing (in mM) 151 NaCl, 2 CaCl₂, 10 HEPES, 13 glucose, with NaOH], pH 7.4, plus 5 μg/ml IB4 FITC conjugate applied directly on the cell through a quartz fiber flow pipe (320 µm internal diameter). After 3 min, IB4 binding was assessed by florescence (Fig. 1F), and the cell was classified as nonpeptidergic [IB4(+)] (clear presence of a fluorescent ring) or peptidergic [IB4(−)] (Silverman and Kruger, 1990; Stucky and Lewin, 1999; Dirajlal et al., 2003).

Figure 1.

Increased incidence of SA in nociceptors isolated from SCI mice. A, SA recorded in vitro in current clamp from a small-diameter (23 μm) DRG neuron isolated at 3 months post-SCI. B, SA was observed in 22/40 DRG neurons isolated from 12 SCI wild-type mice (with an average frequency of 0.9 ± 0.1 Hz during the first minute in whole), 6/23 DRG neurons isolated from 8 sham animals (with an average frequency of 0.8 ± 0.4 Hz), and 6/26 DRG neurons isolated from 8 SCI CaV3.2−/− mice (with an average frequency of 0.8 ± 0.5 Hz). Fisher's exact test [SCI (W/t) vs sham p = 0.0359; SCI (W/t) vs SCI (CaV3.2−/−) p = 0.0122; sham vs SCI (CaV3.2−/−) p > 0.9999]. C, Representative action potential from the cell in A showing a characteristic shoulder during the repolarization phase and a long duration (5.6 ms, measured at half-maximal amplitude). D, Histogram of the action potentials duration (measured at half-maximal amplitude) for the cell in A during 1 min of recording. E, Sensitivity to 1 µM capsaicin was tested at the end of each recording and cells responding to capsaicin with repetitive firing (suggestive of TRPV1 expression) were classified as TRPV1(+). F, Following the capsaicin test, the live cell was challenged with IB4 FITC-conjugated (5 μg/ml) and classified as IB4(+) or IB4(−). Bright-field (left) and fluorescent (right) images of a live IB4(+) DRG neuron. Scale bar, 20 μm.

Electrophysiology

Whole-cell recordings were made with a Multiclamp 700B amplifier (Molecular Devices). Patch pipettes were pulled from borosilicate glass (A-M Systems) using a Sutter P97 puller (Sutter Instrument). The resistance of the patch pipette was 1.4–1.8 MΩ when filled with the standard internal Cs- or K-based solution. The shank of the patch pipette was wrapped with parafilm to reduce pipette capacitance. In a whole-cell mode, the capacity current was reduced by using the amplifier circuitry. To reduce voltage errors, 70–80% of series resistance compensation was applied. In voltage-clamp experiments, sodium, calcium, and Ih currents active during the interspike interval were isolated by using the action potential clamp technique (Raman and Bean, 1999; Puopolo et al., 2005, 2007; Lauzadis et al., 2020). SA was first recorded from a spontaneously active nociceptor and then used as voltage command in voltage clamp. A segment of SA with regular firing was chosen to facilitate the analysis of the interspike currents, and the same segment was used for all the experiments in voltage clamp. To isolate sodium and calcium currents, we used a Cs-based internal solution containing (in mM): 110 Cs-methanesulfonate, 15 TEA-Cl, 10 NaCl, 2 MgCl₂, 10 EGTA, 10 HEPES, 14 Tris-creatine PO₄, 4 Mg-ATP, and 0.3 Na-GTP, pH 7.2, with CsOH. The standard external solution was a modified Tyrode's solution. To isolate the sodium current, the Tyrode's solution contained the following (in mM): 151 NaCl, 2 CaCl₂, 10 HEPES, 13 glucose, pH 7.4, with NaOH. The tetrodotoxin (TTX)-sensitive sodium current was isolated as the current sensitive to 1 μM TTX added to the external solution. The TTX-resistant sodium current was isolated as the current sensitive to 1 μM A803467 applied on top of 1 μM TTX. To isolate the calcium current, the Tyrode's solution contained the following (in mM): 151 TEA-Cl, 2 CaCl₂, 10 HEPES, 13 glucose, pH 7.4, with TEA-OH. The total calcium current was isolated as the current sensitive to replacement of 2 mM Ca²⁺ with 2 mM Co²⁺ in the external solution. The T-type calcium current was isolated as the current sensitive to 1 μM TTA-P2. To isolate the Ih current, the internal solution contained (in mM) 125 K-methanesulfonate, 10 NaCl, 2 MgCl₂, 10 EGTA, 10 HEPES, 14 Tris-creatine PO₄, 4 Mg-ATP, and 0.3 Na-GTP, pH 7.2, with KOH. The Tyrode's solution contained (in mM) 151 NaCl, 2 CaCl₂, 10 HEPES, and 13 glucose, pH 7.4, with NaOH. The Ih current was isolated as the current sensitive to 3 mM Cs⁺ in the external solution. For current-clamp recordings, the internal solution contained (in mM) 125 K-methanesulfonate, 10 NaCl, 2 MgCl₂, 0.1 EGTA, 10 HEPES, 14 Tris-creatine PO₄, 4 Mg-ATP, and 0.3 Na-GTP, pH 7.2, with KOH. The Tyrode's solution contained (in mM) 151 NaCl, 2 CaCl₂, 10 HEPES, and 13 glucose, pH 7.4, with NaOH. After the whole-cell configuration was established, solutions were applied directly to the cell by using an array of quartz fiber flow pipes (320 µm internal diameter) positioned in the vicinity of the cell (350–500 μm away from the cell) containing the test solution and glued on an aluminum rod controlled with a micromanipulator. Solutions were flowing at a rate of 0.1 ml/min and changed by moving the pipes laterally with the micromanipulator. For the experiments in which the effects of TTA-P2 were tested either in voltage or current clamp, the same amount of vehicle (DMSO 0.01%) used to dissolve TTA-P2 was also included in the control Tyrode's solution without TTA-P2.

Data acquisition and analysis

Currents and voltages were controlled and sampled using a Digidata 1550B interface and pCLAMP 10.6 software (Molecular Devices). Current and voltage signals were filtered at 10 kHz (3 dB, four-pole Bessel) and digitized at 50 kHz. Analysis was performed with Clampfit 10.6 and IGOR Pro (version 6.2; WaveMetrics) by using DataAccess (Bruxton) to import pCLAMP files into IGOR. For analysis of action potential frequencies during spontaneous and evoked firing, action potentials were defined as voltage deflections overshooting 0 mV. Reported voltages were corrected for junction potential between the internal solution and the extracellular solution measured using a flowing 3 M KCl electrode (Neher, 1992). Data are presented as mean ± SEM or mean ± SD.

Behavioral tests

All behavioral experiments were carried out in SCI and sham mice at 3–5 months postinjury, a time when SCI rodents show clear signs of thermal and mechanical hypersensitivity and spontaneous pain (Bedi et al., 2010; Bavencoffe et al., 2016; Lauzadis et al., 2020; Liu et al., 2023) or postlaminectomy, respectively.

Recovery of hindlimb motor function (BMS score)

The Basso Mouse Scale (BMS; Basso et al., 2006) was used for measuring motor deficits following SCI. Animals in the SCI and sham groups were assessed for hindlimb motor function while moving freely in a circular enclosure (1 m diameter). For the SCI group, only mice with a BMS score of zero at Day 0–1 postinjury were included in the study. For the sham group, only mice with a BMS score of 9 at Day 0–1 postlaminectomy and with no signs of motor impairment were included in the study.

Mechanical hypersensitivity

Mechanical hypersensitivity was measured on the hindpaw with von Frey filaments (Stoelting) using the up–down method to determine the 50% mechanical threshold (Christensen et al., 2020). Mice were injected with 300 μl of vehicle or TTA-P2 (10 mg/kg) or gabapentin (50 mg/kg, i.p.) and allowed to rest in their cages for 10–15 min before measurements. Calibrated von Frey filaments, starting at 0.16 g, were applied vertically to the center of the plantar surface of the hindpaw with sufficient force to bend the filament for 5 s. A withdrawal of the hindpaw was considered a positive response, while the bending of the filament without a withdrawal of the hindpaw was considered a negative response. A 3 min interval was observed before the same hindpaw was tested again. Following a positive response, the hindpaw was tested using a filament with decreased force until a negative response was observed. This up–down process was repeated four times after the first positive response, and the 50% mechanical threshold for paw withdrawal was determined by the response pattern and the force of the last von Frey filament tested (Christensen et al., 2020).

Conditioned place preference (CPP) paradigm

The CPP paradigm, using analgesics as the conditioning stimuli, was used to measure spontaneous ongoing pain (King et al., 2009; Yang et al., 2014; Lauzadis et al., 2020; Liu et al., 2023). The CPP box (Harvard Apparatus) with automated data collection has three chambers with equal levels of dim illumination: black and white end chambers and a connecting gray chamber. Mice were accustomed to the CPP box during the preconditioning time (30 min/d for 3 d) and were allowed to explore the CPP box without restriction. After the preconditioning time, on Day 4, each mouse was placed in the gray chamber with unrestricted access to all chambers for 15 min, and the time spent in each chamber was measured as the baseline. Mice that showed a preference of >80% or <20% in one chamber were excluded from the study. The pairing session took place during the next 3 d. Each morning, the mouse was injected with 300 µl of vehicle (intraperitoneal) and allowed to rest in its cage for 10–15 min. After 10–15 min, the mouse was confined in the more-preferred chamber (vehicle-paired chamber) for 60 min before returning to its cage. Four hours later, the same mouse was injected with 300 µl of TTA-P2 (10 mg/kg, i.p.) or gabapentin (50 mg/kg, i.p.) and allowed to rest in its cage for 10–15 min. After 10–15 min, the mouse was confined in the less-preferred chamber (drug-paired chamber) for 60 min. Following the pairing session, on Day 8, no drug or vehicle was administered, and each mouse was placed in the gray chamber with unrestricted access to all chambers for 15 min. The time spent in each chamber was measured and compared with the time spent before pairing. Analgesia was associated with the difference in time spent in the drug-paired chamber. Results are reported as the difference score, which measures the difference in the amount of time spent in each chamber before and after pairing. Sham animals were included in the study to control for addictive, aversive, or other nonpain-related contributions to drug-induced behavioral changes.

Drugs

TTA-P2 and gabapentin were prepared fresh before intraperitoneal injection. Drugs were prepared by a second person in the laboratory and given blindly to the experimenter. TTA-P2 was dissolved in DMSO/Cremophor/saline at a ratio of 10/10/80% to a final concentration of 10 mg/ml and injected intraperitoneally. Based on previous studies (Choe et al., 2011; Joksimovic et al., 2018; Liu et al., 2023), TTA-P2 was injected at a dose of 10 mg/kg, a dose at which mice showed no signs of reduced motor activity or sedation. Gabapentin was dissolved in saline. Based on previous studies (Erichsen and Blackburn-Munro, 2002; Hama and Borsook, 2005; Griggs et al., 2015), gabapentin was injected at a dose of 50 mg/kg since at this dose, gabapentin did not produce sedation (Griggs et al., 2015).

Experimental design and statistical analysis

Figure 1 shows the research design and experimental approaches for the data collected with in vitro electrophysiology. SA was recorded in current clamp and then used as voltage command in voltage clamp (Fig. 2A). At the end of each experiment, the cell was tested for capsaicin and IB4 sensitivity. For the data shown in Figure 1B, 40 DRG neurons were recorded from 12 SCI wild-type mice, 23 DRG neurons were recorded from 8 sham mice, and 26 DRG neurons were recorded from 8 SCI CaV3.2−/− mice, and data were analyzed with Fisher's exact test. For the data shown in Figure 2, 12–15 DRG neurons were recorded for each group from a total of 25 mice, and data were analyzed with paired t test. For the data shown in Figure 3, 12–20 DRG neurons were recorded for each group from a total of eight mice, and data were analyzed with paired t test. For the data shown in Figure 4A–C, 14 DRG neurons were recorded for each group from a total of 12 mice, and data were analyzed with paired t test. For the data shown in Figure 4, H and I, 15–20 DRG neurons were recorded for each group from a total of 18 mice, and data were analyzed with two-way ANOVA followed by Sidak's post hoc comparison. For the data shown in Figure 5, A and B, 12–15 DRG neurons were recorded for each group from a total of nine mice, and data were analyzed with paired t test. For the data shown in Figure 5E, 13 DRG neurons were recorded from a total of 11 mice, and data were analyzed with two-way ANOVA followed by Sidak's post hoc comparison. For the behavioral experiments in vivo, mice were randomized and assigned to either a SCI or a sham group. Mechanical hypersensitivity was measured on the hindpaw with von Frey filaments using the up–down method to determine the 50% mechanical threshold (Christensen et al., 2020); spontaneous pain was measured with the CPP paradigm (King et al., 2009). Sufficient animals were used to detect 30% differences between means with at least 80% power. For the data shown in Figure 6, 12 animals were used in each group, and data were analyzed with one-way ANOVA followed by Tukey's post hoc comparison test. For the data shown in Figure 7, 12 animals were used in each group, and data were analyzed with paired t test. For the data shown in Figure 8, A and B, 12 animals were used in each group, and data were analyzed with one-way ANOVA followed by Tukey's post hoc comparison test. For the data shown in Figure 8, C and D, 12 animals were used in each group, and data were analyzed with paired t test. For the data shown in Figure 9, 12 animals were used in each group, and data were analyzed with one-way ANOVA followed by Tukey's post hoc comparison test. For the data shown in Figure 10, 12 animals were used in each group, and data were analyzed with paired t test. Differences were considered significant at *p < 0.05. Data are reported as mean ± SEM or mean ± SD as stated in the figure legend. Statistical analysis was performed by using the Prism 7 software (GraphPad).

Figure 2.

Interspike currents in nociceptors isolated from SCI and sham mice. A, Here and in the following figures, interspike currents were isolated with the action potential clamp method. Top panel, Action potentials recorded in current clamp from a spontaneously active nociceptor were used as voltage command in voltage clamp. Bottom panel, Representative interspike sodium, calcium, and Ih currents. Interspike currents were isolated by subtracting traces (average of three consecutive traces) before and after application of a specific blocker. The total calcium current (Co²⁺-sen, red trace) was isolated as the current sensitive to replacement of 2 mM Ca²⁺ with 2 mM Co²⁺ in the external solution; the TTX-sensitive sodium current (TTX-sen, gray trace) was isolated as the current sensitive to 1 µM TTX; the TTX-resistant sodium current (TTX-res, black trace) was isolated as the current sensitive to 1 µM A803467 applied on top of 1 µM TTX; the Ih current (Cs⁺-sen, purple trace) was isolated as the current sensitive to 3 mM Cs⁺ included in the external solution. B, Total interspike calcium charge measured in nociceptors isolated from SCI and sham mice by integrating the total interspike calcium current from the afterhyperpolarization of the first action potential to −50 mV before the threshold for the second action potential. The total interspike calcium charge was 69 ± 9 fC/pF in nociceptors isolated from SCI mice (SCI, black column) and 26 ± 6 fC/pF in nociceptors isolated from sham mice (Sham, gray column; n = 12; p = 0.0019; paired t test). C, The interspike Ih charge was 3 ± 4 fC/pF in nociceptors isolated from SCI mice (SCI, black column) and 3 ± 5 fC/pF in nociceptors isolated from sham mice (Sham, gray column; n = 11; p = 0.2878; paired t test). D, The interspike TTX-sensitive sodium charge was 9 ± 5 fC/pF in nociceptors isolated from SCI mice (SCI, black column) and 5 ± 4 fC/pF in nociceptors isolated from sham mice (Sham, gray column; n = 10; p = 0.1975; paired t test). E, The interspike TTX-resistant sodium charge was 9 ± 6 fC/pF in nociceptors isolated from SCI mice (SCI, black column) and 7 ± 4 fC/pF in nociceptors isolated from sham mice (Sham, gray column; n = 10; p = 0.2368; paired t test). Data are reported as mean ± SEM.

Figure 3.

Interspike total calcium current and T-type calcium current in nociceptors isolated from SCI and sham mice. A, Representative interspike total calcium current (Co²⁺-sen, black trace, measured as the current sensitive to replacement of 2 mM Ca²⁺ with equimolar Co²⁺) and T-type calcium current (TTA-P2-sen, gray trace, measured as the current sensitive to 1 µM TTA-P2) recorded in a nociceptor with the action potential clamp method. B, Collected results showing the interspike total calcium current and T-type calcium current in nociceptors isolated from SCI mice. The total interspike calcium current (Co²⁺-sen, black dots) was as follows: −0.03 ± 0.04 pA/pF at −80 mV; −0.03 ± 0.02 pA/pF at −75 mV; −0.04 ± 0.02 pA/pF at −70 mV; −0.11 ± 0.02 pA/pF at −65 mV; −0.18 ± 0.03 pA/pF at −60 mV; −0.38 ± 0.06 pA/pF at −55 mV; and −0.76 ± 0.12 pA/pF at −50 mV. The interspike T-type calcium current (TTA-P2-sen, gray dots) was as follows: −0.06 ± 0.03 pA/pF at −80 mV; −0.06 ± 0.01 pA/pF at −75 mV; −0.06 ± 0.01 pA/pF at −70 mV; −0.06 ± 0.01 pA/pF at −65 mV; −0.09 ± 0.02 pA/pF at −60 mV; −0.15 ± 0.03 pA/pF at −55 mV; and −0.5 ± 0.06 pA/pF at −50 mV [n = 12; p = 0.0089 (at −65 mV); p = 0.0003 (at −60 mV); p = 0.0001 (at −55 mV); p < 0.0001 (at −50 mV); paired t test]. C, Collected results showing the interspike total calcium current and T-type calcium current in nociceptors isolated from sham mice. The total interspike calcium current (Co²⁺-sen, black dots) was as follows: −0.02 ± 0.03 pA/pF at −80 mV; −0.02 ± 0.03 pA/pF at −75 mV; −0.05 ± 0.03 pA/pF at −70 mV; −0.16 ± 0.04 pA/pF at −65 mV; −0.18 ± 0.05 pA/pF at −60 mV; −0.26 ± 0.07 pA/pF at −55 mV; and −0.45 ± 0.11 pA/pF at −50 mV. The interspike T-type calcium current (TTA-P2-sen, gray dots) was as follows: −0.05 ± 0.04 pA/pF at −80 mV; −0.04 ± 0.03 pA/pF at −75 mV; −0.03 ± 0.02 pA/pF at −70 mV; −0.05 ± 0.02 pA/pF at −65 mV; −0.09 ± 0.01 pA/pF at −60 mV; −0.08 ± 0.02 pA/pF at −55 mV; and −0.1 ± 0.01 pA/pF at −50 mV [n = 12; p = 0.041 (at −55 mV); p = 0.0119 (at −50 mV); paired t test]. D, Interspike total calcium charge and T-type calcium charge measured in nociceptors isolated from SCI mice by integrating the current from the afterhyperpolarization of the first action potential to −50 mV before the threshold for the second action potential. The total interspike calcium charge (Co²⁺-sen, black column) was 69 ± 9 fC/pF, and the interspike T-type calcium charge (TTA-P2-sen, gray column) was 32 ± 4 fC/pF (n = 12; p = 0.0005; paired t test). E, Interspike total calcium charge and T-type calcium charge measured in nociceptors isolated from sham mice. The total interspike calcium charge (Co²⁺-sen, black column) was 26 ± 6 fC/pF and the interspike T-type calcium charge (TTA-P2-sen, gray column) was 12 ± 7 fC/pF (n = 12; p = 0.087; paired t test). Data are reported as mean ± SEM.

Figure 4.

TTA-P2 reduces nociceptors’ hyperexcitability. SA recorded in a nociceptor isolated from SCI mice in control (A) and in the presence of 1 µM TTA-P2 (B). C, In collected results, the spontaneous firing frequency was reduced from 1.27 ± 0.14 Hz (in control) to 0.21 ± 0.07 Hz in the presence of 1 µM TTA-P2 (n = 14; p < 0.0001; paired t test). D, Representative traces showing evoked firing upon injection of 150 pA (1 s duration) in a nociceptor isolated from a SCI mouse in control and in the presence of 1 µM TTA-P2 (E). F, Representative traces showing evoked firing upon injection of 150 pA (1 s duration) in a nociceptor isolated from a sham mouse in control and in the presence of 1 µM TTA-P2 (G). H, Collected results showing the f–I relationship in response to current injections from 0 to 800 pA (50 pA increment, 1 s duration) in nociceptors isolated from SCI mice in control (black dots) and in the presence of 1 µM TTA-P2 (gray dots; n = 17). Interaction, F_(20,680) = 4.742; p < 0.0001; post hoc, p < 0.0001; p = 0.0002; p < 0.0001; p < 0.0001; p < 0.0001; p = 0.0011; and p = 0.0442 for current injections of 50, 100, 150, 200, 250, 300, and 350 pA, two-way ANOVA followed by Sidak's post hoc comparison. I, Collected results showing the f–I relationship in response to current injections from 0 to 800 pA (50 pA increment, 1 s duration) in nociceptors isolated from sham mice in control (black dots) and in the presence of 1 µM TTA-P2 (gray dots; n = 11). Interaction, F_(20,320) = 0.3422; p = 0.9968, two-way ANOVA followed by Sidak's post hoc comparison. Here and in the following figures, the f–I curves were generated from the same neurons before and after TTA-P2 application. The f–I curve was first generated in control. Then, the same cell was held at the resting potential and continuously perfused with 1 µM TTA-P2 for 3 min before a second f–I curve was generated in the presence of 1 µM TTA-P2. Data are reported as mean ± SEM.

Figure 5.

Reduced excitability of nociceptors isolated from SCI CaV3.2−/− mice. A, Interspike total calcium current and T-type calcium current in nociceptors isolated from SCI CaV3.2−/− mice. The total interspike calcium current (Co²⁺-sen, black dots) was as follows: −0.09 ± 0.03 pA/pF at −80 mV; −0.09 ± 0.02 pA/pF at −75 mV; −0.09 ± 0.01 pA/pF at −70 mV; −0.07 ± 0.01 pA/pF at −65 mV; −0.07 ± 0.01 pA/pF at −60 mV; −0.09 ± 0.02 pA/pF at −55 mV; and −0.13 ± 0.03 pA/pF at −50 mV. The interspike T-type calcium current (TTA-P2-sen, gray dots) was as follows: −0.08 ± 0.04 pA/pF at −80 mV; −0.06 ± 0.02 pA/pF at −75 mV; −0.04 ± 0.01 pA/pF at −70 mV; −0.03 ± 0.01 pA/pF at −65 mV; −0.04 ± 0.01 pA/pF at −60 mV; −0.05 ± 0.01 pA/pF at −55 mV; and −0.09 ± 0.01 pA/pF at −50 mV (n = 11; *p < 0.05; paired t test). B, Interspike total calcium charge and T-type calcium charge in nociceptors isolated from SCI CaV3.2−/− mice. The total interspike calcium charge (Co²⁺-sen, black column) was 13 ± 4 fC/pF, and the interspike T-type charge (TTA-P2-sen, gray column) was 9 ± 4 fC/pF (n = 11; p = 0.1294; paired t test). C, Representative traces showing evoked firing upon injection of 150 pA (1 s duration) in a nociceptor isolated from a SCI mouse in control and in the presence of 1 µM TTA-P2 (D). E, Collected results showing the f–I relationship in response to current injections from 0 to 800 pA (50 pA increment, 1 s duration) in nociceptors isolated from SCI mice in control (black dots) and in the presence of 1 µM TTA-P2 (gray dots; n = 13). Interaction, F_(20,280) = 0.2454; p = 0.9997, Two-way ANOVA followed by Sidak's post hoc comparison.

Figure 6.

Effects of TTA-P2 and gabapentin on mechanical hypersensitivity in male mice. Here and in the following figures, mechanical hypersensitivity was measured in male mice (at 3–5 months postinjury or postlaminectomy) by using the von Frey filaments and the up–down method for the 50% threshold. A, For the TTA-P2 group, in SCI mice, the 50% mechanical threshold dropped from 1.48 ± 0.09 g at the baseline (open circles) to 0.93 ± 0.11 g following SCI (filled circles) and to 0.95 ± 0.08 g following vehicle injection (open triangles). TTA-P2 (10 mg/kg) increased the 50% mechanical threshold to 1.46 ± 0.11 g at 1 h postinjection (filled triangles) and to 1.11 ± 0.13 g at 3 h postinjection (open diamonds; n = 12; baseline vs vehicle p = 0.0018; vehicle vs TTA-P2 1 h p = 0.0046; one-way ANOVA followed by Tukey's post hoc comparison test). B, In sham mice, the 50% mechanical threshold dropped from 1.43 ± 0.07 g at the baseline (open circles) to 1.31 ± 0.09 g following laminectomy (filled circles) and to 1.20 ± 0.14 g following vehicle injection (open triangles). TTA-P2 (10 mg/kg) changed the 50% mechanical threshold to 1.20 ± 0.11 g at 1 h postinjection (filled triangles), and to 1.14 ± 0.13 g at 3 h postinjection (open diamonds; n = 12, one-way ANOVA followed by Tukey's post hoc comparison test; *p < 0.05). C, For the gabapentin group, in SCI mice, the 50% mechanical threshold dropped from 1.55 ± 0.10 g at the baseline (open circles) to 0.93 ± 0.07 g following SCI (filled circles) and to 0.94 ± 0.09 g following vehicle injection (open triangles). Gabapentin (50 mg/kg) increased the 50% mechanical threshold to 1.58 ± 0.09 g at 1 h postinjection (filled triangles; n = 12; baseline vs vehicle p = 0.0015; vehicle vs Gabap 1 h p < 0.0001; one-way ANOVA followed by Tukey's post hoc comparison test). D, In sham mice, the 50% mechanical threshold dropped from 1.51 ± 0.07 g at the baseline (open circles) to 1.42 ± 0.08 g following laminectomy (filled circles) and to 1.47 ± 0.09 g following vehicle injection (open triangles). Gabapentin (50 mg/kg) changed the 50% mechanical threshold to 1.44 ± 0.09 g at 1 h postinjection (filled triangles; n = 12; one-way ANOVA followed by Tukey's post hoc comparison test; *p < 0.05). Data are reported as mean ± SEM.

Figure 7.

Effects of TTA-P2 and gabapentin on spontaneous pain in male mice. Here and in the following figures, male mice (at 3–5 months postinjury or postlaminectomy) were subjected to the CPP paradigm (used to measure spontaneous ongoing pain) by using a CPP box with three chambers: white, gray, and black. The less-preferred chamber was paired with TTA-P2 (10 mg/kg, i.p.) or gabapentin (50 mg/kg, i.p.), while the more-preferred chamber was paired with vehicle. Results are reported as the difference score (test day minus baseline) during 15 min trial. A, For the TTA-P2 group, SCI mice showed an increase in the TTA-P2–paired chamber of 59 ± 35 s (TTA-P2–paired, open circles) and a decrease in the vehicle-paired chamber of −76 ± 31 s (vehicle-paired, open triangles; n = 12; p = 0.0081; paired t test). B, Sham mice showed an increase in the TTA-P2–paired chamber of 2 ± 13 s (TTA-P2–paired, open circles) and a decrease in the vehicle-paired chamber of −2 ± 17 s (vehicle-paired, open triangles; n = 12; paired t test; *p < 0.05). C, For the gabapentin group, SCI mice showed a decrease in the gabapentin-paired chamber of −5 ± 31 s (Gabap-paired, open circles) and a decrease in the vehicle-paired chamber of −4 ± 33 s (vehicle-paired, open triangles; n = 12; paired t test; *p < 0.05). D, Sham mice showed an increase in the gabapentin-paired chamber of 1 ± 24 s (Gabap-paired, open circles) and a decrease in the vehicle-paired chamber of −7 ± 21 s (vehicle-paired, open triangles; n = 12; paired t test; *p < 0.05). Data are reported as mean ± SEM.

Figure 8.

Effects of TTA-P2 and gabapentin on mechanical hypersensitivity and spontaneous pain in SCI CaV3.2−/− mice. A, For the TTA-P2 group, in SCI mice, the 50% mechanical threshold dropped from 1.54 ± 0.10 g at the baseline (open circles) to 1.33 ± 0.10 g following SCI (filled circles) and to 1.31 ± 0.11 g following vehicle injection (open triangles). TTA-P2 (10 mg/kg) increased the 50% mechanical threshold to 1.35 ± 0.07 g at 1 h postinjection (filled triangles) and to 1.28 ± 0.07 g at 3 h postinjection (open diamonds, n = 12). B, For the gabapentin group, in SCI mice, the 50% mechanical threshold dropped from 1.46 ± 0.11 g at the baseline (open circles) to 1.39 ± 0.07 g following SCI (filled circles) and to 1.40 ± 0.07 g following vehicle injection (open triangles). Gabapentin (50 mg/kg) increased the 50% mechanical threshold to 1.55 ± 0.07 g at 1 h postinjection (filled triangles, n = 12). Data are reported as mean ± SEM. One-way ANOVA followed by Tukey's post hoc comparison test, *p < 0.05. C, For the spontaneous pain, for the TTA-P2 group, SCI mice showed an increase in the TTA-P2–paired chamber of 22 ± 30 s (TTA-P2–paired, open circles) and a decrease in the vehicle-paired chamber of −8 ± 37 s (vehicle-paired, open triangles, n = 12). D, For the gabapentin group, SCI mice showed an increase in the gabapentin-paired chamber of 22 ± 38 s (Gabap-paired, open circles) and a decrease in the vehicle-paired chamber of −21 ± 45 s (vehicle-paired, open triangles, n = 12). Data are reported as mean ± SEM. Paired t test, *p < 0.05.

Figure 9.

Effects of TTA-P2 and gabapentin on mechanical hypersensitivity in female mice. A, For the TTA-P2 group, in SCI mice, the 50% mechanical threshold dropped from 1.38 ± 0.08 g at the baseline (open circles) to 0.94 ± 0.06 g following SCI (filled circles) and to 0.93 ± 0.07 g following vehicle injection (open triangles). TTA-P2 (10 mg/kg) increased the 50% mechanical threshold to 1.44 ± 0.08 g at 1 h postinjection (filled triangles) and to 0.86 ± 0.11 g at 3 h postinjection (open diamonds; n = 12; baseline vs vehicle p = 0.0014; vehicle vs TTA-P2 1 h p < 0.0001; one-way ANOVA followed by Tukey's post hoc comparison test). B, In sham mice, the 50% mechanical threshold dropped from 1.35 ± 0.10 g at the baseline (open circles) to 1.34 ± 0.09 g following laminectomy (filled circles) and to 1.30 ± 0.09 g following vehicle injection (open triangles). TTA-P2 (10 mg/kg) increased the 50% mechanical threshold to 1.31 ± 0.10 g at 1 h postinjection (filled triangles) and to 1.29 ± 0.10 g at 3 h postinjection (open diamonds; n = 12; one-way ANOVA followed by Tukey's post hoc comparison test; *p < 0.05). C, For the gabapentin group, in SCI mice, the 50% mechanical threshold dropped from 1.46 ± 0.11 g at the baseline (open circles) to 0.97 ± 0.04 g following SCI (filled circles) and to 0.95 ± 0.04 g following vehicle injection (open triangles). Gabapentin (50 mg/kg) increased the 50% mechanical threshold to 1.52 ± 0.08 g at 1 h postinjection (filled triangles; n = 12; baseline vs vehicle p = 0.0009; vehicle vs Gabap 1 h p < 0.0001; one-way ANOVA followed by Tukey's post hoc comparison test). D, In sham mice, the 50% mechanical threshold dropped from 1.39 ± 0.10 g at the baseline (open circles) to 1.32 ± 0.09 g following laminectomy (filled circles) and to 1.35 ± 0.08 g following vehicle injection (open triangles). Gabapentin (50 mg/kg) increased the 50% mechanical threshold to 1.45 ± 0.08 g at 1 h postinjection (filled triangles; n = 12; one-way ANOVA followed by Tukey's post hoc comparison test; *p < 0.05). Data are reported as mean ± SEM.

Figure 10.

Effects of TTA-P2 and gabapentin on spontaneous pain in female mice. A, For the TTA-P2 group, SCI mice showed an increase in the TTA-P2–paired chamber of 71 ± 15 s (TTA-P2–paired, open circles) and a decrease in the vehicle-paired chamber of −67 ± 17 s (vehicle-paired, open triangles; n = 12; p < 0.0001; paired t test). B, Sham mice showed an increase in the TTA-P2–paired chamber of 3 ± 29 s (TTA-P2–paired, open circles) and a decrease in the vehicle-paired chamber of −16 ± 3 s (vehicle-paired, open triangles; n = 12; paired t test; *p < 0.05). C, For the gabapentin group, SCI mice showed an increase in the gabapentin-paired chamber of 8 ± 23 s (Gabap-paired, open circles) and a decrease in the vehicle-paired chamber of −12 ± 22 s (vehicle-paired, open triangles; n = 12; paired t test; *p < 0.05). D, Sham mice showed an increase in the gabapentin-paired chamber of 7 ± 21 s (Gabap-paired, open circles) and a decrease in the vehicle-paired chamber of −8 ± 19 s (vehicle-paired, open triangles; n = 12; paired t test; *p < 0.05). Data are reported as mean ± SEM.

Results

In vitro and in vivo electrophysiological recordings have shown hyperexcitability and increased incidence of SA in nociceptors from 3 d up to 8 months following the injury in a rat model of SCI (Bedi et al., 2010; Yang et al., 2014; Bavencoffe et al., 2016; Lauzadis et al., 2020). Therefore, we first determined whether there was hyperexcitability and increased incidence of SA also in nociceptors isolated from SCI mice at 3–5 months postinjury. Figure 1A shows a representative trace recorded in current clamp from small-diameter (23 μm) DRG neuron sensitive to capsaicin isolated from a SCI mouse. SA was observed in 22/40 (55%) of DRG neurons from SCI wild-type mice (Fig. 1B; Table 1), with an average frequency of 0.9 Hz (measured during the first minute in whole-cell current clamp). In contrast, only 6/23 (26%) of DRG neurons from sham mice and 6/26 (23%) of DRG neurons from SCI CaV3.2−/− mice showed SA. Action potentials had a characteristic shoulder during the repolarization phase (Fig. 1C) and a long duration (5.6–6.0 ms) measured at half-maximal amplitude (Fig. 1D). Of the total SCI DRG neurons included in the study, 66% of them were sensitive to capsaicin (as shown by repetitive firing in response to 1 µM capsaicin; Fig. 1E) and therefore were classified as TPRV1(+) nociceptors (Cardenas et al., 1995; Caterina et al., 1997; Petruska et al., 2000). Following the capsaicin test, an additional classification was made by staining the live DRG neuron with IB4 FITC-conjugated (Fig. 1F; Nagy and Hunt, 1982; Silverman and Kruger, 1990; Stucky and Lewin, 1999; Dirajlal et al., 2003), and 82% of DRG neurons included in the study were IB4(+). Taken together, our data show that hyperexcitability and increased incidence of SA of nociceptors are retained also in SCI mice and are consistent with previous findings in SCI rats and mice (Bedi et al., 2010; Yang et al., 2014; Bavencoffe et al., 2016; Berkey et al., 2020; Lauzadis et al., 2020; Hu et al., 2025), suggesting conserved mechanism(s) between species.

Table 1.

Properties of spontaneously active DRG neurons

Property	SCI (W/t)	Sham	SCI (CaV3.2−/−)
Soma diameter (μm)	23.9 ± 0.5 (22, 12)	22.4 ± 1.1 (6, 8)	22.6 ± 0.8 (6, 8)
Capacitance (pF)	20.5 ± 1.6 (22, 12)	18.5 ± 1.7 (6, 8)	18.8 ± 1.7 (6, 8)
Resting membrane potential (RMP; mV)	−57.2 ± 2.2 (22, 12)*	−64.3 ± 2.6 (6, 8)	−69.8 ± 3.2 (6, 8)
Frequency (Hz)	0.9 ± 0.1 (22, 12)	0.8 ± 0.4 (6, 8)	0.8 ± 0.5 (6, 8)
Coefficient of variability of interspike interval (%)	86.8 ± 9.4 (22, 12)	99.6 ± 12.8 (6, 8)	95.1 ± 6.6 (6, 8)
fAHP (mV)	−79.4 ± 4.0 (22, 12)	−84.2 ± 3.2 (6, 8)	−83.3 ± 3.9 (6, 8)
mAHP (mV)	−61.6 ± 3.2 (22, 12)	−60.5 ± 3.9 (6, 8)	−60.9 ± 3.7 (6, 8)
AP threshold (mV)	−34.3 ± 2.7 (22, 12)	−35.9 ± 3.6 (6, 8)	−40.0 ± 2.2 (6, 8)
AP half-maximal amplitude (ms)	5.0 ± 0.3 (22, 12)	4.3 ± 0.7 (6, 8)	5.6 ± 0.8 (6, 8)
TRPV1(+) and IB4(+)	(n = 11)	(n = 3)	(n = 3)
TRPV1(+) and IB4(−)	(n = 2)	(n = 0)	(n = 0)
TRPV1(−) and IB4(+)	(n = 9)	(n = 2)	(n = 2)
TRPV1(−) and IB4(−)	(n = 0)	(n = 1)	(n = 1)

Data are from spontaneously active small DRG neurons (diameter < 27 μm) isolated from male and female mice. Numbers in parenthesis indicate the number of cells and the number of animals used, respectively. Statistical significance was determined with one-way ANOVA followed by Tukey's multiple-comparison test. Data are reported as mean ± SEM. Differences were considered significant at *p < 0.05.

^*Indicates statistical significance for comparison between SCI (W/t) and SCI (CaV3.2−/−); p = 0.0166.

Next, we employed the action potential clamp technique in voltage clamp (Raman and Bean, 1999; Puopolo et al., 2005, 2007; Lauzadis et al., 2020) in combination with pharmacological tools to directly identify the underlying ionic mechanism responsible for driving the SA of nociceptors. Action potentials were first recorded in current clamp from a spontaneously active nociceptor (Fig. 1A) and then used as voltage command in voltage clamp (Fig. 2A, top). Interspike currents recorded from a nociceptor isolated from a SCI mouse at 3 months postinjury showed a major contribution of calcium current during the interspike interval, while there was minimal contribution from the hyperpolarization-activated cation current (Ih), TTX-sensitive (TTX-s), and TTX-resistant (TTX-r) sodium currents (Fig. 2A, bottom). Summary data collected by measuring the interspike charge from the afterhyperpolarization of the first action potential to −50 mV before the second action potential confirmed that the total interspike calcium charge (Fig. 2B) was several folds higher than the interspike Ih charge (Fig. 2C), TTX-s charge (Fig. 2D), and TTX-r charge (Fig. 2E). Taken together, the data collected from nociceptors isolated from SCI mice point to a major contribution of calcium channels in driving nociceptors’ hyperexcitability and their SA.

Previous findings from our laboratory showed that T-type calcium channels carry the bulk of the interspike calcium current in nociceptors in a rat model of SCI (Lauzadis et al., 2020). Therefore, we employed the action potential clamp technique in voltage clamp in combination with TTA-P2, a blocker of T-type calcium channels (Shipe et al., 2008; Kraus et al., 2010; Choe et al., 2011), to isolate and measure the fraction of the total interspike calcium current carried by T-type calcium channels. In our experimental conditions, we found that in nociceptors isolated from SCI mice, the T-type calcium channels account for the bulk (∼60%) of the total interspike calcium current (Fig. 3A,B). In contrast, in nociceptors isolated from sham mice (Fig. 3C), there was little contribution from T-type calcium channels during the interspike interval. This suggests that in nociceptors from SCI mice, similar to previous findings in nociceptors from SCI rats (Lauzadis et al., 2020), the increased activity of T-type calcium channels induced by the injury may contribute to drive nociceptors’ hyperexcitability and their SA. To test this directly, we determined the effects of TTA-P2 on the spontaneous firing and the frequency–current (f–I) relationship during current-clamp recordings. Inhibition of T-type calcium channels with 1 μM TTA-P2 reduced the spontaneous firing rate of nociceptors from 1.3 to 0.2 Hz (Fig. 4A–C). Similarly, upon current injection (Fig. 4D–G, sample traces), 1 μM TTA-P2 reduced the evoked firing rate in nociceptors isolated from SCI mice (Fig. 4H), while it had little effects on the f–I relationship in nociceptors isolated from sham mice (Fig. 4I). Taken together, the data suggest that the increased activity of T-type calcium channels induced by the injury contributes to driving nociceptors’ hyperexcitability and their SA at higher frequencies (see Discussion).

All three isoforms of T-type calcium channels (CaV3.1, CaV3.2, and CaV3.3) are expressed in the peripheral nervous system (Talley et al., 1999; Shin et al., 2003), and CaV3.2 is the predominant isoform in the spinal cord and sensory neurons (Todorovic and Jevtovic-Todorovic, 2006; Watanabe et al., 2015; Rzhepetskyy et al., 2016; Candelas et al., 2019), suggesting that the increased activity of CaV3.2 channels induced by SCI could contribute to drive nociceptors’ hyperexcitability. To test this possibility, we measured the size of the TTA-P2–sensitive current during the interspike interval in nociceptors isolated from SCI CaV3.2−/− mice at 3–5 months postinjury. The total interspike calcium current and the interspike TTA-P2–sensitive calcium current (Fig. 5A) and calcium charge (Fig. 5B) were reduced in size as compared with nociceptors isolated from SCI wild-type mice (compare Fig. 3). In agreement with the voltage-clamp data, recordings in current clamp (Fig. 5C,D) confirmed that nociceptors isolated from SCI CaV3.2−/− mice had reduced excitability as determined with the f–I relationship with only minimal effects of TTA-P2 on firing (Fig. 5E). Taken together, our data point to CaV3.2 as the main isoform of T-type calcium channels responsible for driving nociceptors’ hyperexcitability following SCI.

Previous reports in SCI rats showed a strong relationship between the severity of pain and the incidence of SA of nociceptors (Bedi et al., 2010) and a contribution of T-type calcium channels to the development/maintenance of mechanical hypersensitivity and spontaneous pain following SCI (Lauzadis et al., 2020; Liu et al., 2023). Therefore, we employed behavioral pharmacology combined with measurements of mechanical hypersensitivity and spontaneous pain in vivo (Davoody et al., 2011; Yang et al., 2014; Griggs et al., 2015; Wu et al., 2017; Lauzadis et al., 2020; Liu et al., 2023) to determine whether the increased activity of T-type calcium channels observed in vitro in nociceptors will contribute to the development/maintenance of chronic neuropathic pain in vivo and compared the effects of TTA-P2, a blocker of T-type calcium channels (Shipe et al., 2008; Kraus et al., 2010; Choe et al., 2011), with those of gabapentin, acting mainly by inhibiting trafficking of high voltage-activated (HVA) calcium channels (Field et al., 2006; Li et al., 2006; Hendrich et al., 2008) and known to reduce neuropathic pain (Bannister et al., 2017; Shin et al., 2022). In our experimental conditions, in male mice, both TTA-P2 (10 mg/kg, i.p.) and gabapentin (50 mg/kg, i.p.) reduced mechanical hypersensitivity (Fig. 6), while only TTA-P2 induced CPP (Fig. 7), suggestive of a TTA-P2–sensitive component of spontaneous pain similar to previous results in SCI rats (Liu et al., 2023). Driven by the results in vitro showing a contribution of CaV3.2 channels to nociceptors’ hyperexcitability (Fig. 5), we then carried out behavioral experiments in vivo in SCI CaV3.2−/− mice to determine the contribution of CaV3.2 channels to the development/maintenance of chronic neuropathic pain following SCI. First, we found that SCI CaV3.2−/− mice developed less mechanical hypersensitivity as compared with SCI wild-type mice, and treatment with TTA-P2 or gabapentin had minimal additional effects (Fig. 8A,B). Second, TTA-P2 and gabapentin failed to induce CPP (Fig. 8C,D). Taking together, the results are consistent with a significant contribution of CaV3.2 channels to the development/maintenance of chronic neuropathic pain following SCI in mice.

Finally, we tested whether the effects of TTA-P2 and gabapentin on mechanical hypersensitivity and spontaneous pain in SCI mice are sex dependent. To this purpose, we used the same experimental approach in SCI and sham female mice at 3–5 months postinjury or postlaminectomy, respectively. Like male mice, also in SCI female mice TTA-P2 (10 mg/kg, i.p.) and gabapentin (50 mg/kg, i.p.) reduced mechanical hypersensitivity (Fig. 9), while only TTA-P2 induced CPP (Fig. 10), suggestive of a TTA-P2–sensitive component of spontaneous pain in SCI female mice. Overall, our data collected in SCI mice confirmed a hyperexcitable state and SA of nociceptors as previously reported in SCI rats and mice, pointing to conserved mechanism(s) between species, and support a contribution of CaV3.2 channels in driving nociceptors’ hyperexcitability and the development/maintenance of neuropathic pain following SCI.

Discussion

A wealth of studies in rat models of SCI have shown that nociceptors’ hyperexcitability plays a major contribution to driving the development/maintenance of SCI-NP (Bedi et al., 2010; Walters, 2012; Yang et al., 2014; Bavencoffe et al., 2016; Wu et al., 2017; Lauzadis et al., 2020; Liu et al., 2023). Our data show that SCI triggers nociceptors’ hyperexcitability in mice, consistent with previous studies (Berkey et al., 2020; Hu et al., 2025) and pointing to a conserved mechanism(s) between species. DRG neurons with small diameter (≤27 µm) and a cell capacitance of 23 ± 5 pF (n = 135), of which 66% were TRPV1(+) and 82% were IB4(+), were included in the study, suggestive of putative nociceptors (Caterina et al., 1997; Petruska et al., 2000; Ho and O'Leary, 2011).

Similar to findings in SCI rats (Lauzadis et al., 2020), the data in SCI mice show that T-type calcium channels contribute the bulk of inward current during the interspike interval (Fig. 3), further supporting a conserved mechanism(s) between species. The data in current clamp showed reduced firing in the presence of TTA-P2 (Fig. 4) and reduced nociceptors’ excitability in SCI CaV3.2−/− mice (Fig. 5). In our experimental conditions, we observed frequencies of nociceptors’ SA (average 0.9 ± 0.1 Hz; Table 1) that were higher than frequencies reported previously in SCI rodents (Bedi et al., 2010; Bavencoffe et al., 2016; Hu et al., 2025). Consequently, our data in current clamp (Fig. 4) showed that TTA-P2 reduced the SA of nociceptors mainly at higher frequencies. These observations raise the possibility that the increased activity of T-type calcium channels reported here may contribute to driving nociceptors’ SA more at higher frequencies while depolarizing spontaneous fluctuations may contribute more at lower frequencies (Odem et al., 2018; Tian et al., 2024). Consistent with the possibility that multiple ion channels may contribute to nociceptors’ SA at different frequencies (Tian et al., 2024), a recent study showed that, in addition to T-type calcium channels reported here, also L- and N-type calcium channels contribute to nociceptors’ SA (Garza-Carbajal et al., 2024).

CaV3.2 channels play a critical role in the development of neuropathic pain in models of peripheral neuropathies (Bourinet et al., 2005; Jagodic et al., 2008; Messinger et al., 2009; Yue et al., 2013; Obradovic et al., 2014; Todorovic and Jevtovic-Todorovic, 2014; François et al., 2015; Chen et al., 2018; Kang et al., 2018; Tomita et al., 2019; Fayad et al., 2022). In this respect, our data add to these previous studies demonstrating that CaV3.2 channels contribute to nociceptors’ hyperexcitability and the development/maintenance of SCI-NP, further supporting the role of CaV3.2 channels in the development of neuropathic pain (Zamponi et al., 2009; Todorovic and Jevtovic-Todorovic, 2014; Waxman and Zamponi, 2014; Bourinet et al., 2016; Harding and Zamponi, 2022). However, in our study we used global CaV3.2−/− mice in which we could not exclude the possibility of genetic compensations in response to gene knock-out (El-Brolosy and Stainier, 2017) that could trigger changes in the activity of ion channels and contribute to shape nociceptors’ excitability. Future studies with conditional CaV3 knockin/knock-out mice (François et al., 2015) will be instrumental to determine the contribution of CaV3.2 channels in different subpopulations of sensory neurons to the development/maintenance of SCI-NP.

Spontaneous pain accounts for a major component of SCI-NP (Baastrup and Finnerup, 2008; Masri and Keller, 2012). Anticonvulsant and tricyclic antidepressant drugs are first-line choice for the treatment of SCI-NP (Cardenas et al., 2002; Finnerup et al., 2002, 2015; Putzke et al., 2002; Tai et al., 2002; Levendoglu et al., 2004; Siddall et al., 2006; Rintala et al., 2007; Baastrup and Finnerup, 2008; Vranken et al., 2008; Agarwal and Joshi, 2017; Colloca et al., 2017), but patients report unsatisfactory pain relief and many side effects (Ravenscroft et al., 2000; Murphy and Reid, 2001; Warms et al., 2002; Rintala et al., 2007; Cardenas et al., 2013; Kukkar et al., 2013). Therefore, treatment of SCI-NP remains a major challenge, and improved outcomes will require a better understanding of the underlying mechanism(s) responsible for driving SCI-NP. In line with the contribution of T-type calcium channels in driving peripheral and central neurons’ hyperexcitability following nerve injury (Todorovic and Jevtovic-Todorovic, 2013; Bourinet et al., 2016; Cai et al., 2021; Harding and Zamponi, 2022), blockers of T-type calcium channels, including amiloride, ethosuximide, mibefradil, dihydropyridine, and cannabinoid derivatives (You et al., 2011; Gadotti et al., 2013; Berger et al., 2014; Bladen et al., 2015a,b), and new blockers with increased selectivity such as TTA-P2 and Z944 have shown efficacy in rodent models of chronic inflammatory and neuropathic pain (Ferreira et al., 1999; Dogrul et al., 2003; Flatters and Bennett, 2004; Shipe et al., 2008; Choe et al., 2011; Tringham et al., 2012; Harding et al., 2021). Our findings support and expand on these previous studies by showing the presence of a TTA-P2–sensitive component of mechanical hypersensitivity and spontaneous pain following SCI. Interestingly, similar to a previous study in SCI rats (Liu et al., 2023), we found that, in our experimental conditions, TTA-P2 was more effective than gabapentin in inducing CPP (Fig. 7). However, it is important to note that TTA-P2 acts as a pore-blocking drug of T-type calcium channels (Shipe et al., 2008; Choe et al., 2011), while gabapentin binds to the alpha-2-delta-1 subunit of HVA calcium channels to reduce their trafficking and membrane expression (Field et al., 2006; Li et al., 2006; Hendrich et al., 2008). These observations could suggest that, in our experimental conditions, the lack of a gabapentin-sensitive component of spontaneous pain could be due, at least in part, to insufficient pairing time during the CPP paradigm. Although the effects of gabapentin on HVA calcium channels are well established (Fink et al., 2000; Sutton et al., 2002; Hendrich et al., 2008), gabapentin may also affect other targets such as microglia (Wodarski et al., 2009), astrocytes (Baydas et al., 2005; Kim et al., 2016), and synaptic transmission (Patel et al., 2000; van Hooft et al., 2002; Bayer et al., 2004) that could account for some of the effects of gabapentin reported here. TTA-P2 and gabapentin were injected intraperitoneally, and therefore they will block the activity of calcium channels both in peripheral neurons and in the spinal and supraspinal centers, raising the possibility that some of the effects of TTA-P2 and gabapentin could also be mediated by calcium channels expressed in central targets. In addition, sham laminectomy to expose the spinal cord triggers changes in nociceptors’ SA, gene expression, and changes in operant and reflexive behaviors (Odem et al., 2018, 2019; Cuevas-Diaz Duran et al., 2023). Because our study did not include naive control animals, we cannot exclude a priori the possibility that postsurgical pain may have contributed to some of the SCI effects reported here, although this seems unlikely considering that the behavioral studies in vivo were carried out at 3–5 months postinjury or postlaminectomy.

Epidemiological studies addressing gender differences in SCI patients suffering from neuropathic pain have reported conflicting results, with some studies suggesting higher pain and higher use of analgesics in women (Stormer et al., 1997; Norrbrink Budh et al., 2003), others reporting no differences (Cardenas et al., 2004; Werhagen et al., 2004), and others reporting significantly higher prevalence in men (Finnerup et al., 2001). A conflicting trend has emerged also in preclinical studies carried out in SCI rodents, with some studies reporting significant differences between males and females (Dominguez et al., 2012; Gaudet et al., 2021; Lee et al., 2023), while in others only minimal differences were found (Walker et al., 2019; McFarlane et al., 2020). In this respect, our data show that male mice developed slightly higher mechanical hypersensitivity following SCI (37% drop in mechanical threshold, Fig. 6) as compared with SCI female mice (32% drop in mechanical threshold; Fig. 9), and there was a complete reversal by TTA-P2 and gabapentin, suggestive of a TTA-P2 and a gabapentin-sensitive components of mechanical hypersensitivity. In contrast, only TTA-P2, both in male and female SCI mice, induced CPP, suggestive of a TTA-P2–sensitive component of spontaneous pain.

The increased activity of CaV3.2 channels found in the majority of TRPV1(+) and IB4(+) nociceptors in our study is consistent with previous data in nociceptors from SCI rats (Lauzadis et al., 2020) and from naive rats (Nelson et al., 2005). Nonetheless, genetic approaches in mice showed expression of CaV3.2 in the C- and Aδ-low-threshold mechanoreceptors (LTMRs), with low expression in the TRPV1(+) neurons (François et al., 2015), and in situ hybridization in human DRG neurons showed preferential expression of CaV3.2 in the Aδ-LTMRs [TrkB(+) neurons; Chemin et al., 2025]. These differences may suggest intrinsic species differences between rodents and humans or alternatively increased or de novo expression of CaV3.2 channels in specific subpopulations of DRG neurons induced by the injury. Future studies will be necessary to determine whether there is an increased activity of CaV3.2 channels and whether they contribute to driving the SA of human DRG neurons following SCI.