Report

Thalamic Control of Visceral Nociception Mediated by T-Type Ca2+ Channels

See allHide authors and affiliations

Science  03 Oct 2003:
Vol. 302, Issue 5642, pp. 117-119
DOI: 10.1126/science.1088886

This article has a correction. Please see:

Abstract

Sensations from viscera, like fullness, easily become painful if the stimulus persists. Mice lacking α1G T-type Ca2+ channels show hyperalgesia to visceral pain. Thalamic infusion of a T-type blocker induced similar hyperalgesia in wild-type mice. In response to visceral pain, the ventroposterolateral thalamic neurons evokeda surge of single spikes, which then slowly decayed as T type–dependent burst spikes gradually increased. In α1G-deficient neurons, the single-spike response persisted without burst spikes. These results indicate that T-type Ca2+ channels underlie an antinociceptive mechanism operating in the thalamus andsupport the idea that burst firing plays a critical role in sensory gating in the thalamus.

Low voltage–activated (LVA) T-type Ca2+ channels play crucial roles in the control of cellular excitability under diverse physiological and pathological processes (1, 2). Recently, studies revealed a novel role of T-type Ca2+ channel in the pain sensory pathway by showing that this channel facilitates pain signals in peripheral nociceptors (3, 4) and in the spinal cord (5). T-type channels are also highly expressed in the thalamus (6), through which noxious signals from the spinal cords must pass before reaching the cortex (7). When the thalamocortical relay neurons receive sensory inputs, they respond in dual firing modes: either in singular action potentials or in a burst of action potentials clustered together as a high-frequency discharge (810). T-type Ca2+ channels are known to excite hyperpolarized thalamic neurons to generate bursts of action potentials. There has been much debate about the role of the thalamic burst firing in the sensory processing (11, 12). Therefore, whether thalamic T-type channels would contribute to the nociceptive signal processing as a signal enhancer or a suppressor is an open question.

Mice homozygous for a null mutation of the α1G (CaV3.1) gene showed a functional deletion of T-type currents and lacked low threshold burst firing in the thalamocortical relay neurons (13). We measured the sensitivity of the α1G-deficient mice (α1G–/–) by delivering thermal or mechanical stimuli delivered either on the palm or tail (supporting online material). No significant difference was observed between the mutants and their wild-type littermates in these assays (Fig. 1, A to C). Hyperalgesia to cutaneous pain, as measured by the relative enhancement of the pain response by a subcutaneous injection of complete Freund's adjuvant (CFA) before pain tests (14), also did not significantly differ between the wild type and the mutant (Fig. 1D). Next, we examined the sensitivity of the mice to visceral pain induced by intraperitoneal administration of either acetic acid (Fig. 1E) or MgSO4 solution (Fig. 1F) as previously described (15). The wild-type mice showed typical pain behaviors characterized by writhing, such as abdominal stretching and constriction in response to these two chemicals, with MgSO4-induced pain responses terminated earlier than those by acetic acids (15). However, compared with the wild type, the mutant mice showed stronger writhing responses to both acetic acid [Fig. 1E, two-way analysis of variance (ANOVA) for genotype, F(1,22) = 10.23, P < 0.01] and MgSO4 treatments [Fig. 1F, two-way ANOVA for genotype, F(1,6) = 4.32, P < 0.01], which suggests that T-type channels inhibit visceral nociception. This was unexpected because T-type channels have been thought to be involved in the strong excitatory pathway boosting pain signals (3).

Fig. 1.

Pain responses of α1G–/– mice to noxious stimuli. (A) Responses to mechanical stimuli with von Frey filaments. (B) Tail flick responses to thermal stimuli. (C) Paw withdrawal responses to infrared thermal stimuli at two different intensities. (D) One day after injection of CFA (1×) in the left paw, infrared thermal stimuli were delivered either to the injected paw (ipsilateral) or the opposite uninjected paw (contralateral). Visceral pain induced by intraperitoneal injection of either acetic acid (E) or MgSO4 solution (F). Writhing responses were examined for 20 min after acetic acid injection or for 10 min after MgSO4 injection. Error bars indicate SEM. Two-tailed t test, *P < 0.01; **P > 0.05.

To define the site affected by the α1G-null mutation, we did phenocopy experiments in wild-type mice using a T-type channel inhibitor, mibefradil, which does not cross the blood-brain barrier (16). An intraperitoneal administration of mibefradil caused dose-dependent analgesia to visceral pain, an effect opposite that in the α1G–/– mice (Fig. 2, A and C). These results along with similar findings by others (4) indicate a pain-enhancing role of T-type channels in the periphery, and they suggest that the hyperalgesia of the α1G–/– mice to visceral pain might not be associated with a peripheral mechanism. The hyperalgesia cannot be explained by the role of these channels in spinal processing of pain signals, because previous studies have shown that injection of T-type inhibitor in the spinal cord reduced neuronal response to cutaneous pain (5) or showed no significant effects on visceral nociception (17).

Fig. 2.

Opposite effects of mibefradil on visceral pain depending on the route of delivery. (A and C) Dose-dependent analgesic effect of mibefradil injected at the periphery. Visceral pain responses measured for 1 hour after intraperitoneal injections of mibefradil at different concentrations. (B and D) Pain-enhancing effect of mibefradil in the thalamus. About 0.5 μl mibefradil (1 mM) or saline was infused into the left and right ventroposterior thalamic regions. +/+, wild type; –/–, mutant. Two-tailed t test for saline control, *P < 0.01; **P > 0.05.

Recently, a novel central pathway for visceral pain has been revealed: noxious signals from the viscera activated ventroposterolateral (VPL) thalamocortical neurons via excitation of the dorsal column of the spinal cord (18). The α1G subunits are predominantly expressed in the thalamocortical neurons (6) and play a critical role in the modulation of intrinsic firing of these neurons (13). A focal transient infusion of mibefradil into the VPL thalamus (supporting online material) enhanced pain responses in the wild-type mice in a dose-dependent manner (Fig. 2, B and D), which suggests that the thalamus was, at least in part, involved in the enhanced response to visceral pain in α1G–/– mice.

To explore the role of T-type channels in the response of VPL neurons to visceral pain, we examined the response mode of the neurons with single-unit recordings in vivo after acetic acid injection under urethane anesthesia (supporting online material). First, we analyzed the baseline firing patterns of the neurons before acetic acid injection. Low-threshold burst spikes were defined by the presence of a preceding silent period longer than 100 ms, high-frequency spikes of 200 to 400 Hz, a shortening of the first interspike interval, and a progressive prolongation of successive interspike intervals (Fig. 3A), as previously described in the cat lateral geniculate nucleus neurons (19). The total baseline firing frequency (Hz) before the treatment was significantly lower in the mutant than in the wild type (Fig. 3C). There was no significant difference in the firing rate of single spikes between the two groups, whereas burst spikes were basically depleted in the mutant (Fig. 3C). These findings indicate that the decrease in the total firing frequency in the mutant could be accounted for by the depletion of burst spikes in the α1G–/– VPL neurons. The lack of burst firing in the mutants was consistent with the previous findings in brain slices that thalamic burst firings depended on the activation of T-type calcium channels (8, 13, 20).

Fig. 3.

Firing pattern changes in VPL neurons triggered by visceral pain in wild-type and α1G–/– VPL neurons. (A) Sample tracings obtained from the three experimental groups (+/+, saline; +/+, acetic acids; –/–, acetic acids) are shown for three periods: before the injection, –0.5 min to ∼0; immediately after the injection, 0 to ∼0.5 min; at a later time, 5.0 to ∼5.5 min. Each spike is marked as either single (blue) or burst (red) with vertical strokes below the trace: (top) wild type injected with saline; (middle), wild type injected with acetic acids; (bottom), mutants injected with acetic acids. A vertical bar crossing the trace marks the time of injection. Sample traces for either single or burst spike marked by dots on the left of the top panel are shown with an expanded time scale on the right at the bottom. (B) The temporal change of the firing patterns (supporting online material). The baseline total firing rate (C) is taken as the reference point and marked by a horizontal dotted line. Point C on the x axis indicates before treatment (Top) Wild type injected with saline; (middle) wild type injected with acetic acids; (bottom) mutants injected with acetic acids. Vertical blue lines mark the time of intraperitoneal injection. (C) Baseline firing rates obtained from results collected for 5 min preceding the acetic acid injection. Total, burst, and single firing rates in Hz are compared for the two genotypes. Spike numbers of each burst event that occurred are summarized at 2-min intervals. Two-tailed t test, **P > 0.05; *P <0.001. (D) Prolonged single-spike response in α1G–/– VPL neurons to visceral pain. Relative single-spike firing rates in reference to the baseline single-spike firing were calculated for consecutive 2-min intervals for each recorded neuron. Point C on the x axis is the average of baseline single-spike firing frequency taken as 1.0 (horizontal line) for each neuron. Open circle refers to mutant and closed circle, to wild type. Repeated ANOVA for genotype (post hoc test), **P > 0.1; *P < 0.05.

Next, we examined the firing patterns of the neurons after induction of visceral pain. As previously reported in other species (18, 21), a subset of single units in the VPL region of the thalamus showed characteristic changes in the firing rate after induction of visceral pain. These “visceral pain–responsive” neurons were analyzed in wild-type and mutant mice (Table 1 and Fig. 3A). In these neurons, an intraperitoneal injection of acetic acid evoked an early-onset surge of action potentials, which consisted mainly of unitary single spikes at a significantly higher level when compared with saline control [Fig. 3, A and B, top and middle, two-way ANOVA for treatment between acetic acid and saline, F(1,13) = 11.84, P < 0.005]. This single-spike surge induced after the treatment was followed by a gradual increase of burst spikes, which reflected the activation of T-type channels [Fig. 3, A and B, ANOVA for time, F(4,56) = 3.07, P < 0.05]. However, as the burst activity increased, the firing rate of single spikes significantly decayed [ANOVA for time, F(4,20) = 5.03, P < 0.01]. The increased burst activity was not due to a generalized effect of anesthesia, because this was not observed in mice injected with saline (Fig. 3, A and B, top).

Table 1.

Response of VPL neurons to visceral pains. One single-unit recording was obtained from each mouse before treatments (number of neurons shown in first row). The second row shows the number of neurons showing spike frequency above 0.1 Hz for 5 min that were used in visceral pain induction. The third row shows the number of neurons showing a change in firing rates in response to visceral pain induction; visceral pain–responsive neurons characterized by a change of firing rate more than 2 times the value of the SD for baseline firing rates recorded during the 5 min before treatment.

Recorded neurons Saline MgSO4 Acetic acids
+/+ +/+ +/+ -/-
Total single-unit obtained (n) 12 9 26 23
Used in visceral pain induction (n) 10 9 20 18
Firing rate changed (n) 3 6 16 10

The firing pattern of the α1G–/– VPL neurons is shown in Fig. 3, A and B, bottom. Even after the acetic acid injection, α1G–/– VPL neurons showed no significant change in the burst activity [Fig. 3B, ANOVA for time, F(4,48) = 65.32, P > 0.1], but showed the early-onset increase of single spikes as examined in the wild type (Fig. 3, A and B, bottom). An important difference was noted in the late period: The induced single-spike response persisted without showing a significant decay within the recording time in α1G–/– VPL neurons [Fig. 3B, bottom, ANOVA for time, F(4,28) = 5.98, P > 0.1]. A direct comparison of single-spike activities alone revealed this point more clearly: the induction level of the early-onset peak was comparable between wild type and mutants [Fig. 3D, ANOVA for genotype, F(1,12) = 5.87, P > 0.1], but the induced single-spike activities were sustained at a higher level in the mutant than in the wild type after 5 min (Fig. 3D).

These results suggest that T-type Ca2+ channels in VPL thalamocortical neurons are first activated after a surge of pain signal influx from the viscera and then play an inhibitory role in the processing of those signals, thereby suppressing pain responses. It has been well demonstrated that the low-threshold burst spikes evoked by activation of T-type Ca2+ channel in the thalamocortical relay neurons can efficiently activate nRT neurons, which are the major inhibitory neurons innervated to VPL regions. The excitation of these neurons causes hyperpolarization and rebound burst spikes again in the thalamocortical neurons via reciprocal connections between the two neuronal groups (22). Those reciprocal interactions have been implicated as a pacemaker activity for sleep spindles or spike-and-wave discharges during sleep or absence seizures, respectively (2, 23). Interestingly, however, a recent study using artificial and biological hybrid network in vitro showed that hyperpolarization and/or burst sequences induced by reciprocal interactions could also contribute to the inhibitory processing of sensory signals by reducing the responsiveness of thalamocortical neurons to the artificial sensory inputs (24). Because the hyperpolarization is associated with an increased membrane conductance and a burst has a long refractory period (170 to 200 ms) (25), it is conceivable that the hyperpolarization and/or burst sequences prevent rapidly recurring sensory signal influxes to thalamocortical relay neurons (26, 27). Thus, the suppression of T-type channels in VPL neurons by genetic or pharmacological means could have reduced the hyperpolarization and/or burst sequences and thereby interfere with the reciprocal interactions necessary for the “sensory gating,” which leads to an enhancement of noxious signals.

The gain of inhibition in the thalamus has been implicated in focal attention by decoupling other sensory inputs except for the sensory modality of interest (28). The present study suggests that the thalamic decoupling per se could be a robust pain-relieving mechanism. The antinociceptive mechanism in the thalamus, however, may not be effective for controlling acute pain responses that are mediated by fast reflexes at the periphery, because it is turned on late after a persistent influx of noxious signals into the thalamus. The reports of similar T-type channel activities in the thalamus of patients with other chronic pain syndromes (29) raise the possibility that the thalamic mechanism of antinociception plays a role in pain responses beyond visceral pain.

Supporting Online Material

www.sciencemag.org/cgi/content/full/302/5642/117/DC1

Materials and Methods

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article