Lidocaine Inhibits HCN Currents in Rat Spinal Substantia Gelatinosa Neurons.

BACKGROUND: Lidocaine, which blocks voltage-gated sodium channels, is widely used in surgical anesthesia and pain management. Recently, it has been proposed that the hyperpolarization-activated cyclic nucleotide (HCN) channel is one of the other novel targets of lidocaine. Substantia gelatinosa in the spinal dorsal horn, which plays key roles in modulating nociceptive information from primary afferents, comprises heterogeneous interneurons that can be electrophysiologically categorized by firing pattern. Our previous study demonstrated that a substantial proportion of substantia gelatinosa neurons reveal the presence of HCN current (Ih); however, the roles of lidocaine and HCN channel expression in different types of substantia gelatinosa neurons remain unclear.
METHODS: By using the whole-cell patch-clamp technique, we investigated the effect of lidocaine on Ih in rat substantia gelatinosa neurons of acute dissociated spinal cord slices.
RESULTS: We found that lidocaine rapidly decreased the peak Ih amplitude with an IC50 of 80 μM. The inhibition rate on Ih was not significantly different with a second application of lidocaine in the same neuron. Tetrodotoxin, a sodium channel blocker, did not affect lidocaine's effect on Ih. In addition, lidocaine shifted the half-activation potential of Ih from -109.7 to -114.9 mV and slowed activation. Moreover, the reversal potential of Ih was shifted by -7.5 mV by lidocaine. In the current clamp, lidocaine decreased the resting membrane potential, increased membrane resistance, delayed rebound depolarization latency, and reduced the rebound spike frequency. We further found that approximately 58% of substantia gelatinosa neurons examined expressed Ih, in which most of them were tonically firing.
CONCLUSIONS: Our studies demonstrate that lidocaine strongly inhibits Ih in a reversible and concentration-dependent manner in substantia gelatinosa neurons, independent of tetrodotoxin-sensitive sodium channels. Thus, our study provides new insight into the mechanism underlying the central analgesic effect of the systemic administration of lidocaine.

Lidocaine is one of the most widely used local anesthetics in surgical anesthesia and in the management of acute postoperative and chronic pain syndromes.[1-3] Its anesthetic action is typically produced by blocking voltage-gated sodium channels that are responsible for neuronal signal propagation.[4] Moreover, increasing evidence has indicated that lidocaine affects other channels such as calcium channels,[5-7] potassium channels,[8-11] and transient receptor potential channels.[12-15] Recently, the hyperpolarization-activated cyclic nucleotide (HCN) channel has been identified as a novel target of lidocaine.[16-19]

HCN channel family (HCN1–4)–mediated current (Ih) is a slowly activated, mixed inward current carried by Na+ and K+.[20,21] Diverse functions of Ih have been described, including modulation of the resting membrane potential (RMP), action potential firing frequency, neuronal network oscillation, and dendritic integration.[22-25] These functions contribute to nociception, thalamocortical oscillations, and hippocampal plasticity. Recently, attention has been paid to the effect of the HCN channel on inflammatory, neuropathic, and postoperative pain. For instance, abundant axonal accumulation of HCN channels has been observed after nerve injury.[26-29] Concomitantly, systemic administration of ZD7288, a selective inhibitor of HCN channels, has been shown to significantly alleviate mechanical allodynia, likely because of the suppression of ectopic discharges in dorsal root ganglion (DRG) neurons.[30,31] In addition, HCN1 gene knockout was observed to partially prevent the development of mouse cold allodynia.[32,33] Similarly, in HCN2 genetically deleted mice, neuropathic pain was inhibited because of reduced Ih currents and action potential firing rate.[34] These observations indicate that HCN channels are involved in pain generation and maintenance.

It has been reported that lidocaine can inhibit Ih in Xenopus oocytes,[18] DRG,[16,19] and thalamocortical neurons.[17] In both sciatic nerve block and intrathecal anesthesia experiments, the anesthetic durations of lidocaine in HCN1−/− mice has been observed to be shorter than that in wild-type mice.[16] Moreover, studies have shown that the anesthetic duration of lidocaine is prolonged by the coadministration of ZD7288.[16,35,36] In contrast, forskolin, a potent nonspecific adenylyl cyclase activator that can enhance Ih, has been shown to reduce the pinprick blockade duration of lidocaine.[16] Therefore, it is possible that suppression of Ih could be another analgesic mechanism of lidocaine.

Rexed laminae II in the spinal dorsal horn, also referred to as the substantia gelatinosa (SG), integrates nociceptive information from the periphery to the brain and plays a crucial role in nociceptive transmission. Neuronal circuitry in the SG is complicated by a predominance of excitatory interneurons and can be classified according to discharge pattern.[37-40] Our previous study showed that a substantial proportion of SG neurons expresses the HCN channel.[41] Furthermore, it has been demonstrated that systemic administration of lidocaine can suppress spontaneous action potential firings of SG neurons, indicating that SG neurons may be affected by lidocaine through this route of administration.[42]

Given that HCN channels and spinal SG neurons play important roles in inflammatory and neuropathic pain, we hypothesize that lidocaine can block HCN channels in SG neurons. Thus, the aim of this study was to investigate the effect and molecular mechanisms of lidocaine on the HCN channel and to determine the HCN channel distribution in SG subtypes classified by discharge pattern.

METHODS

Preparation of Spinal Cord Slices

All experimental procedures were in accordance with protocols approved by the Institutional Animal Care and Use Committee of Nanchang University Guidelines. Spinal cord slices were prepared from male Sprague Dawley rats (3–5 weeks old) as described previously.[41,43] Briefly, rats deeply anesthetized with urethane (1.5 g/kg, intraperitoneally) were perfused transcardially with ice-cold carbogenated (95% O2 and 5% CO2) sucrose-substituted artificial cerebrospinal fluid (s-ACSF) containing the following (in millimolar): 240 sucrose, 2.5 KCl, 0.5 CaCl2, 3.5 MgCl2, 1.25 NaH2PO4, 25 NaHCO3, 0.4 ascorbate acid, and 2 pyruvate. The lumbosacral spinal cord was immediately dissected and immersed in the same s-ACSF. Animals were then killed by decapitation after extraction and while still under anesthesia. Parasagittal slices measuring 300 μm in thickness were cut with a microslicer (VT1000S; Leica, Nussloch, Germany). The slices were incubated at 32°C for at least 30 minutes in normal carbogenated ACSF containing the following (in millimolar): 117 NaCl, 3.6 KCl, 2.5 CaCl2, 1.2 MgCl2, 1.2 NaH2PO4, 25 NaHCO3, 11 d-glucose, 0.4 ascorbic acid, and 2 pyruvate (pH = 7.4).

Electrophysiologic Recordings

Whole-cell patch-clamp recordings were conducted as in our previous study.[41] After incubation, 1 spinal cord slice was transferred to a recording chamber and continuously perfused with ACSF (2–4 mL/min) at room temperature. The SG neurons in lumbar segments L4 to L5 were visualized with an IR-DIC camera (IR-1000; Dage, Michigan City, IN). Patch pipettes were pulled from the borosilicate glass (World Precision Instruments, Sarasota, FL) on a micropipette puller (P-97; Sutter Instrument, Novato, CA). Typical resistances ranged from 3 to 5 MΩ when filled with a solution containing the following (in millimolar): 130 K-gluconate, 5 KCl, 4 Mg-ATP, 10 phosphocreatinine, 0.5 EGTA, 0.3 Li-GTP, 10 HEPES (pH = 7.3, adjusted with KOH, 300 mOsm). Signals were amplified with an EPC-10 amplifier and Patchmaster software (HEKA Electronik, Lambrecht, Germany). Series resistances typically measured between 20 and 30 MΩ and were monitored throughout the recording period. Data were excluded if the series resistance changed by >20%. RMP was measured in the current clamp with no holding current.

Chemicals

All drugs were obtained from Sigma-Aldrich (St. Louis, MO), except for ZD7288 and tetrodotoxin (TTX), which were obtained from Tocris Bioscience (Bristol, UK). Lidocaine, ZD7288, and TTX were dissolved in distilled water at 1000 times the concentration to be used and stored at −20°C. Before application, these drugs were immediately diluted to the respective concentrations in ACSF solution.

Statistical Analysis

Ih currents and membrane voltage responses were analyzed using Patchmaster and GraphPad Prism 5.0 software (GraphPad Prism Software, Inc., La Jolla, CA). Ih is composed of an instantaneous (Iinst) and a steady-state component (Iss).[44,45] The amplitude of Ih was calculated as the difference between Iss and Iinst at the evoked voltage of −130 mV unless indicated otherwise. The current density was calculated by dividing the amplitude of Ih by the cell capacitance at every test potential.

To determine IC50, a dose-dependent curve was fitted with the Hill equation as follows: y = Imax/(1 + IC50/x), where Imax represents the maximal current amplitude, IC50 is the half-maximal inhibitory concentration, and x is the concentration of lidocaine in micromolar.

To estimate V0.5 (the voltage at which the current is one-half of its maximal level), the activation curves were fitted using GraphPad Prism with the Boltzmann sigmoidal equation: I/Imax = 1/(1 + exp[(V0.5 − V)/k], where Imax represents the maximal current amplitude, V0.5 is the midpoint potential, V is the membrane potential, and k is the slope factor.

The time constant (τ) of Ih activation was obtained by fitting the current traces (from instantaneous to 500 milliseconds) with a single exponential equation as follows: I = Ih × exp(−t/τ) + Iss, where I is the amplitude of the current at time t, Ih represents the current amplitude, and τ is the time constant.[46,47]

To determine the reversal potential (Vrev), the recorded tail current amplitudes during deactivation were plotted against each test potential to construct I-V curves. Vrev is the intersection of the I-V curve with the x-axis. The input resistance (Rin) was calculated based on the current change during a 10 mV hyperpolarizing pulse.

SPSS version 17.0 (SPSS Inc., Chicago, IL) was used for all the statistical analysis, except where noted. Data are expressed as mean ± SEM unless indicated otherwise, and n represents the number of neurons recorded. Two-sample paired Student t tests were used for comparison between 2 dependent groups, and 2-sample unpaired Student t tests were used for comparison between 2 independent groups. Wilcoxon signed rank tests were used for 2 dependent groups when the normality test failed using Shapiro-Wilk tests; for all pairwise differences tested using Student t test, P ≥ 0.110. One-way analysis of variance with post hoc of Bonferroni correction was used to compare >2 groups. For all the cases, P < 0.05 was considered as statistically significant.

RESULTS

Lidocaine Inhibits Ih in SG Neurons

As shown at the bottom of Figure 1A, Ih currents were evoked by hyperpolarizing voltage steps from −60 to −130 mV in 10-mV increments (1-second duration) from a holding potential of −50 mV. Lidocaine (100 μM) dramatically decreased Ih (Fig. 1A). To test whether the currents recorded in our study were HCN channel mediated, we further applied ZD7288 (10 μM; 7 minutes) to the same neuron. ZD7288 reduced the current amplitude from 319 ± 64 to 19 ± 4 pA (n = 10 neurons from 8 rats; P = 0.001, paired t test), showing an inhibition of approximately 93% relative to the control. This result demonstrates that the current recorded in our study was produced by the HCN channel. To investigate the time course of Ih inhibition by lidocaine, we perfused lidocaine for 4 minutes. As illustrated in Figure 1D, the amplitude of the Ih currents decreased after lidocaine treatment, reaching a maximal level within 3 to 4 minutes and gradually returned to the control level in 5 to 6 minutes after washout without apparent rundown (Fig. 1, A and D). Thus, in this study, we perfused lidocaine for 4 minutes, and the amplitude of Ih under the action of lidocaine was measured after 4 ± 1 minutes.

Figure 1.

Lidocaine inhibits Ih in substantia gelatinosa neurons. A, Representative current responses to hyperpolarization voltage steps in the absence (control) and presence of lidocaine (100 μM), washout (recovery), and the following application of ZD7288 (10 μM) in a same neuron (upper). Lower panel shows the Ih evoking voltage protocol. Open circles in the lowest Ih trace indicate the instantaneous (Iinst) and steady state (Iss) of Ih at −130 mV. B, Sample traces under control condition, the first perfusion of lidocaine (100 μM), washout, and the second perfusion of lidocaine. C, Superimposed traces of Ih (at −130 mV) in (B). D, Time course of the inhibition of lidocaine on Ih amplitude recorded from the same neuron in (B). Gray bars represent the periods of bath application of lidocaine. E, Summary data showing the percentage change in the peak Ih amplitude after application of lidocaine. In this and the following figures, *P < 0.05, **P < 0.01, ***P < 0.001. n.s. = no significant difference.

To investigate whether desensitization is involved in lidocaine-induced inhibition of Ih, we applied lidocaine twice to the same neuron (Fig. 1B). The amplitude of Ih was significantly reduced to 44% ± 5% that of the control (n = 6 neurons from 6 rats; 146 ± 22 pA; P = 0.007, 1-way analysis of variance with post hoc of Bonferroni) and recovered to 97% ± 1% (143 ± 23 pA; P = 0.911) after washout (Fig. 1, B and E). When applying lidocaine to the same neuron once again, Ih was still reduced to 46% ± 6% that of the control (P = 0.009). No significant difference in the Ih currents was observed between the 2 perfusions of lidocaine (P = 0.976; Fig. 1, C and E). This finding suggests that lidocaine markedly reduces Ih in the SG neurons, and the effect is rapid, reversible, and nondesensitized.

Lidocaine-Induced Ih Inhibition Is Not Mediated by Sodium Channels

To examine whether sodium channels could affect lidocaine’s inhibition of Ih, we compared the Ih alterations under the coapplication of 0.5 μM TTX and 100 μM lidocaine (Fig. 2, A and B). Lidocaine still inhibited the amplitude of Ih to 52% ± 3% (n = 9 neurons from 4 rats) that of the control (221 ± 62 pA; P = 0.005, paired t test, Fig. 2, A–D) in the presence of TTX, which was not significantly (P = 0.193, unpaired t test, Fig. 2F) different from the inhibition of Ih caused by lidocaine alone (n = 22 neurons from 11 rats; 47% ± 2% that of the control) (296 ± 37 pA; P < 0.0001; paired t test, Fig. 2E). These data confirm that lidocaine directly blocks HCN channels without the involvement of TTX-sensitive voltage-gated sodium channels.

Figure 2.

Lidocaine-induced Ih inhibition is not affected by tetrodotoxin (TTX). A and B, In the presence of TTX (0.5 μM) in artificial cerebrospinal fluid solution, lidocaine (100 μM) still decreased the peak amplitude of Ih. C, Superimposed traces of Ih (at −130 mV) recorded in (A and B). D and E, Averaged percentage of Ih inhibition by lidocaine in the presence and absence of TTX, respectively. F, Differences of Ih amplitude under the treatment of lidocaine in the presence and absence of TTX. In this and the other figures, *P < 0.05, **P < 0.01, ***P < 0.001.

Lidocaine Inhibits Ih in SG Neurons in Concentration-Dependent Manner

To determine IC50 for the inhibitory effect of lidocaine on Ih, we perfused lidocaine in increasing concentrations (1–1000 μM). As illustrated in Figure 3, lidocaine inhibited Ih in a dose-dependent manner (IC50 = 80 μM and Hill coefficient = 0.63).

Figure 3.

Lidocaine blocks Ih in substantia gelatinosa neurons in a concentration-dependent manner. A, Example traces before and after bath application of increasing concentrations of lidocaine (in micromolar): 10, 100, and 600, which are from the same neuron. B, Superimposed traces of Ih (at −130 mV) recorded in (A). C, Dose-response curve for Ih amplitude under the effect of lidocaine. The values in parentheses indicate the number of cells examined.

Lidocaine Shifts Ih Activation Curve to More Hyperpolarized Potential

To investigate whether lidocaine modifies the kinetics of HCN channels, Ih activation currents were evoked by the protocol shown at the bottom of Figure 4A. Both Ih and tail currents were significantly decreased by lidocaine (Fig. 4, A–C). The activation curves were plotted using the Boltzmann equation under the control conditions and during the perfusion of 600 μM lidocaine (Fig. 4D). Lidocaine significantly shifted V0.5 toward more negative values from −109.7 ± 0.9 to −114.9 ± 1.1 mV, a shift of −5.2 mV (n = 23 neurons from 6 rats; P < 0.0001, paired t test; Fig. 4D and Table 1). Moreover, lidocaine significantly decreased the current density by 55% to 73% relative to that measured for the control neurons over the voltage range of −70 to −130 mV (n = 18 neurons from 6 rats; Fig. 4E and Table 1). As shown in Figure 4F, lidocaine increased the time constant to 127% to 148% that of the control (n = 19 neurons from 6 rats). For example, at −130 mV, the time constant was significantly lower in the presence of lidocaine (91.4 ± 9.2 milliseconds) than that of the control (69.8 ± 3.2 milliseconds; P = 0.003, paired t test; Fig. 4F and Table 1).

Figure 4.

Lidocaine shifts Ih activation to more hyperpolarized potentials in substantia gelatinosa neurons. A, Control activation of Ih current (top) responses to the evoking protocol by hyperpolarizing voltage steps over the range from −60 to −130 mV in 10-mV increments for 1 s from the holding potential of −50 mV, which were then subjected to a voltage jump to −130 mV to obtain full activation (bottom). B, Enlargement of the tail currents in the rectangle shown in (A). Tail currents (marked as the arrow) were normalized to the maximum values [(Itail, max − Itail)/Itail, max] to plot the activation curve shown in (D). C, Lidocaine markedly decreased the amplitude of tail currents (same neuron in A). D, Lidocaine clearly shifted V0.5 to more negative values. E, Plot of Ih current density against the membrane potentials. Lidocaine reduced the current density from −70 to −130 mV. F, Time constant (τ) of Ih activation against the membrane potentials. Lidocaine significantly increased τ from −60 to −130 mV. In this and the other figures, *P < 0.05, **P < 0.01, ***P < 0.001.

Table 1.

Effects of Lidocaine on HCN Channels’ Kinetics of Activation in SG Neurons

Lidocaine Shifts the Reversal Potential of Ih Current to More Negative Values

To further examine the effects of lidocaine on Vrev of Ih, we ran the protocol shown at the bottom of Figure 5A. After the bath application of lidocaine (600 μM), the magnitudes of the tail currents were clearly reduced (Fig. 5, A-C). The I-V curves were plotted by recording the tail currents in 24 SG neurons (Fig. 5D). The values of Vrev were −31.1 ± 1.3 and −38.6 ± 1.1 mV in the absence and presence of lidocaine, respectively (P < 0.0001, paired t test).

Figure 5.

Lidocaine shifts the reversal potential of Ih current to more negative values in substantia gelatinosa neurons. A, Sample control traces of the deactivation of Ih current (top) and the evoking protocol (bottom: the membrane potential was stepped to −130 mV for 1 s from the holding potential of −50 mV to fully activate Ih, followed by a series of depolarization test potentials from −120 to −50 mV in 10-mV increments). B, The expanded traces of the tail currents in the rectangle shown in (A). Tail currents (marked as arrow) were measured at the onset of the test potentials. C, Lidocaine greatly reduced the amplitude of the tail currents (the same neuron in A); D, I-V curves constructed from the tail currents in the absence (black) and presence of lidocaine (gray) to each test potential. Lidocaine shifts the reversal potential to more negative values.

Lidocaine Suppresses Burst Firing in SG Neurons

To investigate the effects of lidocaine on firing properties, voltage responses to the current injections were recorded in a current-clamp mode by using the protocol shown at the bottom of Figure 6C, which generates both action potential firing and a rebound spike.[17,48] Lidocaine at each concentration tested (100, 600, and 1000 μM) largely decreased the frequency of action potential firing (Fig. 6, A–C, left). In addition, lidocaine prolonged the latency of the rebound depolarization (Fig. 6D and Table 2) and decreased the rebound spike frequencies (Fig. 6E and Table 2). As shown in Figure 6F and Table 2, the variation of RMP toward hyperpolarization increased with the concentration of lidocaine. In addition, lidocaine significantly increased Rin (Fig. 6G and Table 2).

Figure 6.

The effect of lidocaine on firing properties in substantia gelatinosa neurons. A–C (left), Voltage responses to the current commands shown at the bottom during control (black), and the administration of different concentrations of lidocaine (red): 100, 600, and 1000 μM, respectively. Right, Enlargement of rectangular areas shown in (A–C), with a trace of recovery after washout (blue). Lidocaine reduced the frequency of sodium-dependent action potentials and the rebound firings and increased the latency of the rebound firings. Bottom, Voltage responses were recorded under a 1-s depolarization current pulse from 0 to 150 pA, followed by a 1-s hyperpolarization current pulse from 0 to −150 pA. D–G, Grouped data show the percentage change in rebound depolarization latency, frequency, resting membrane potential (RMP) changes, and Rin after application of different concentrations of lidocaine. In this and the other figures, *P < 0.05, **P < 0.01, ***P < 0.001.

Table 2.

Effects of Lidocaine on Latency and Frequency of RD, RMP, and Rin in SG Neurons

Ih Expression Varies with SG Neuron Firing Pattern

To determine Ih expression in various types of SG neurons, action potentials were elicited with a depolarizing current (120 pA) injection with a duration of 1 second (Fig. 7, Aa–h). On the basis of previous studies,[39,40] we categorized SG neurons (n = 102) into the following 7 groups: tonic firing (63%), delayed firing (14%), single spike (10%), initial burst (8%), phasic firing (5%), gap firing (2%), and reluctant firing (2%) neurons (Fig. 7A). Among these groups of neurons, 64% of tonic-firing neurons, 21% of delayed-firing neurons, 70% of single-spike neurons, 63% of initial-burst neurons, 50% of phasic-firing neurons, 100% of gap-firing neurons and no reluctant-firing neurons were recorded with Ih (Fig. 7B). Conversely, HCN channels consisted of 4 subtypes with different time constants.[49] To roughly examine the HCN channel subtypes in SG neurons, we measured the time constant of Ih at −130 mV, which ranged from 49 to 1289 milliseconds (n = 51; 231 ± 40 milliseconds), with most values being <400 milliseconds (Fig. 7C). These results suggest that most of the HCN channels in SG are probably HCN1 and HCN2 like.

Figure 7.

Ih expression in different firing patterns of substantia gelatinosa (SG) neurons. A, Representative firing patterns in SG neurons: tonic-firing (a), delayed-firing (b), single-spike (c), initial-burst (d), phasic-bursting (e), gap-firing (f), and reluctant-firing (g) neurons evoked by the protocol in (h). B, Summary bar graph showing numbers of neurons expressing Ih with respect to cell electrophysiologic classification. C, Histogram figure showing τ values (at −130 mV) in the subtypes of SG neurons.

DISCUSSION

In this study, we demonstrated that lidocaine strongly and rapidly blocks Ih in SG neurons of the spinal dorsal horn in a reversible and concentration-dependent manner. Our results show that lidocaine could downregulate the excitability of SG neurons by inhibiting HCN currents, providing new insight into the mechanism underlying the analgesic effect of lidocaine.

It is generally believed that sodium channels are the main target of local anesthetics, including lidocaine. By blocking sodium channels, lidocaine inhibits action potential propagation and neuronal excitability. Recently, lidocaine was found to block HCN channels in oocytes and HEK 293 cells,[18] DRG neurons,[16,19] and thalamocortical neurons.[17] In this study, we demonstrated for the first time that lidocaine can decrease the amplitude of Ih in SG neurons. The concentrations of lidocaine used in our study are clinically relevant to spinal and epidural anesthesia, without being toxic to the cardiovascular or central nervous systems.[50] The maximal blocking effect of lidocaine on the amplitude of Ih in SG neurons was approximately 92%, indicating a high efficacy of lidocaine toward HCN channels. In addition, lidocaine-induced Ih inhibition was not affected by TTX. It has been reported that voltage-gated sodium channels in laminae I/II cells of the spinal cord are primarily TTX-sensitive isoforms.[51] Thus, it is clear that sodium channels are not involved in lidocaine-induced Ih inhibition.

Ion channel activation is an important aspect of channel kinetics. The negative shift of Ih activation would decrease the probability that HCN channels are open in the resting state and thus decrease neuronal excitability. In this study, we found that the Ih activation curve is shifted toward negative values by lidocaine. Meng et al.[18] also reported that, in HEK 293 cells, lidocaine could negatively shift Ih activation for homomeric HCN1 channels or heteromeric HCN1 to HCN2 channels. The time constant is another key issue associated with channel kinetics. The speed of HCN channel activation decreases with an increasing time constant. Consistent with its action on the Ih activation curve, lidocaine increased the time constant of Ih at all test potentials in SG neurons.

Vrev is determined by ion channel selectivity. HCN channel opening allows for a greater influx of Na+ and a lower efflux of K+, with a net inward current.[20] Theoretically, the Vrev value of HCN channels can be calculated by using the Goldman-Hodgkin-Katz equation. However, unlike that of cloned cells, the exact ratio of Na+/K+ permeability through HCN channels could not be determined in slice preparations because the distribution of HCN channel subtypes in different cell types is not the same. Therefore, in this and our previous studies,[41] the x-intercept of the Ih I-V curve is represented as Vrev. Lidocaine shifted the Vrev value of HCN channels from −31.1 to −38.6 mV in SG neurons, which resembles the effect of bupivacaine on HCN channels in DRG.[19] Such an effect corresponds to altered Na+ and K+ selectively of HCN channels. Alternately, effects on background currents could be responsible for the apparent alteration of Vrev.

In the nervous system, rebound depolarization plays a pivotal role in neuronal excitability.[52,53] The latency of rebound depolarization is determined by Ih.[54] In our study, lidocaine not only delayed the latency of rebound depolarization but also reduced the number of rebound spikes, in line with previous studies on thalamocortical neurons.[5,17,55] These data further indicate that lidocaine can suppress the excitability of SG neurons by blocking HCN channels.

Previous studies have shown that lidocaine can regulate the resting properties of neurons.[16,17] Similarly, in this study, lidocaine markedly blocked HCN channels in SG neurons, causing the RMP to more hyperpolarized voltages. In addition, Rin increased significantly with the concentration of lidocaine from 100 to 1000 μM in our study.

The composition of SG neurons is complex and has not been clearly elucidated to date. Our data show that Ih can be recorded for approximately 58% of SG neurons. Among these neurons, tonic-firing cells have the highest expression level (64%). It has been reported that most tonic-firing SG neurons are excitatory interneurons.[38] Therefore, the blocking of Ih by lidocaine may lead to the inhibition of SG neuron excitability. In addition, most of the time constants of the Ih currents we recorded here were <400 milliseconds, suggesting higher contents of the HCN1 and HCN2 subtypes in SG neurons. Further experiments, such as immunohistochemistry and single-cell reverse transcription polymerase chain reaction, are required to obtain the exact distribution of HCN channel subtypes in SG neurons.

Taken together, our results demonstrate that lidocaine is an effective blocker of HCN channels in SG neurons. This inhibition may downregulate the excitability of SG neurons by decreasing the rebound depolarization frequency and prolonging rebound depolarization latency. Given that SG neurons play crucial roles in pain modulation, our observations may suggest a novel cellular mechanism underlying the analgesic effects of lidocaine in spinal fluid.

DISCLOSURES

Name: Tao Hu, MD.

Contribution: This author helped conduct the study, collect and analyze the data, and write the manuscript.

Attestation: Tao Hu has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Nana Liu, MD.

Contribution: This author helped conduct the study and analyze the data.

Attestation: Nana Liu has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Minhua Lv, MD.

Contribution: This author helped conduct the study and analyze the data.

Attestation: Minhua Lv has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Longxian Ma, MD.

Contribution: This author helped write the manuscript.

Attestation: Longxian Ma approved the final manuscript.

Name: Huizhen Peng, MD.

Contribution: This author helped write the manuscript.

Attestation: Huizhen Peng approved the final manuscript.

Name: Sicong Peng

Contribution: This author helped review the analysis of the data.

Attestation: Sicong Peng approved the final manuscript.

Name: Tao Liu, PhD, MD.

Contribution: This author helped design the study, review the analysis of the data, and write the manuscript.

Attestation: Tao Liu has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

This manuscript was handled by: Markus W. Hollmann, MD, PhD, DEAA.

ACKNOWLEDGMENTS

We thank Hua Lu for critical discussion, comments, and/or editing of the manuscript.