The Journal of Neurophysiology Vol. 77 No. 6 June 1997, pp. 3284-3296
Copyright ©1997 by the American Physiological Society

Two Parallel Signaling Pathways Couple 5HT_1A Receptors to N- and L-Type Calcium Channels in C-Like Rat Dorsal Root Ganglion Cells

Carla G. Cardenas, Lucinda P. Del Mar, and Reese S. Scroggs

Department of Anatomy and Neurobiology, College of Medicine, University of Tennessee, Memphis, Tennessee 38163

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Cardenas, Carla G., Lucinda P. Del Mar, and Reese S. Scroggs. Two parallel signaling pathways couple 5HT_1A receptors to N- and L-type calcium channels in C-like rat dorsal root ganglion cells. J. Neurophysiol. 77: 3284-3296, 1997. The coupling of serotonin receptors to Ca²⁺ channels was studied in a subpopulation of acutely isolated rat dorsal root ganglion (DRG) cell bodies (type 1 DRG cells), which have membrane properties similar to C-type nociceptive sensory neurons. In these cells, serotonin (5HT) inhibited high-threshold Ca²⁺ channel current and decreased action potential duration. The inhibitory effects of 5HT and the 5HT_1A agonist 8-OH-DPAT were shown to be antagonized by the 5HT_1A antagonists spiperone and pindolol, respectively, indicating involvement of a 5HT_1A receptor. Several observations suggest that 5HT_1A receptors couple to N- and L-type Ca²⁺ channels by two different signaling pathways in type 1 DRG cells. The inhibition of Ca²⁺ channel currents produced by 10 µM 5HT occurred in two phases, an initial slowing of current activation rate (kinetic slowing), which was complete within 10 s, and a simultaneous reduction in steady state current amplitude (steady state inhibition), which peaked in ~1 min. The kinetic slowing, but not steady state inhibition, was reversed by a positive prepulse to +70 mV (prepulse). Blockade of N-type Ca²⁺ channels selectively reduced the kinetic slowing and its reversal by prepulses. Chelation of intracellular Ca²⁺ or blockade of L-type Ca²⁺ channels selectively reduced the steady state inhibition. Recordings using the cell-attached patch configuration suggest that steady state inhibition required a component that was diffusible in the cytosol, while kinetic slowing occurred via a membrane delimited pathway. The application of 5HT to the cell body outside the patch pipette reduced macroscopic Ca²⁺ channel currents in 33% of small-diameter DRG cells tested, indicating the participation of a cytosolic diffusible component. Application of 5HT (a membrane impermeant compound) outside the patch pipette produced steady state inhibition only, whereas similar application of membrane permeant 5HT_1A agonists, 8-OH-DPAT or 5-methoxy-N,N-dimethyl-tryptamine, produced kinetic slowing and steady state inhibition. Together these data suggest that 5HT_1A receptors couple negatively to Ca²⁺ channels via two pathways: a membrane-delimited pathway that couples to N-channels and actuates voltage-sensitive kinetic slowing and a pathway dependent on a cytosolic diffusible component and free intracellular Ca²⁺, which couples to L channels and actuates steady state inhibition.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Several different biochemical pathways have been implicated in the coupling of neurotransmitter receptors to Ca²⁺ channels in neurons. Initially "membrane-delimited" pathways were observed in several neuronal types, where the signal traverses a very short distance through the membrane from the receptor to a closely associated ion channel (Bean 1989; Brown 1993). Membrane-delimited pathways have been implicated directly in the actions of several neurotransmitters including, serotonin, neuropeptide Y, -aminobutyric acid, and norepinephrine (Bernheim et al. 1991; Foehring 1996; Forscher et al. 1986; Green and Cottrell 1988; Hirning et al. 1990; Penington et al. 1991).

Subsequently, evidence was obtained in several neuronal types suggesting the coupling of various neurotransmitter receptors to Ca²⁺ channels by cytosolic diffusible pathways. Cytosolic diffusible pathways have been implicated directly in the actions of several neurotransmitters including acetylcholine, angiotensin, dopamine, norepinephrine, and bradykinin (Bernheim et al. 1991; Diverse-Pierliussi et al. 1995; Hille 1994; Shapiro et al. 1994; Surmeier et al. 1995). There appears to be several cytosolic diffusible pathways, including a pathway involving protein kinase A and phosphatase 2A (Surmeier et al. 1995), another pathway involving protein kinase C (Boland et al. 1991; Diverse-Pierliussi and Dunlap 1993), and another pathway that appears not to correspond to any known signaling pathway, but has a dependence on the presence of intracellular Ca²⁺ (Beech et al. 1991; Bernheim et al. 1991; Hille 1992; Howe and Surmeier 1995; Shapiro et al. 1994).

In previous reports, we described the inhibition of high-threshold Ca²⁺ channel currents by serotonin (5HT) in a group of acutely isolated small-diameter rat dorsal root ganglion (DRG) neurons, which had membrane properties consistent with those of C-type nociceptors (referred to as type 1 DRG cells) (Cardenas et al. 1995; Del Mar et al. 1994). Type 1 DRG cells exhibit large high-threshold Ca²⁺ currents of which, on average, 28% is conducted through N-type Ca²⁺ channels, 46% through L-type Ca²⁺ channels, and 26% through other high-threshold Ca²⁺ channel(s) that are resistant to selective L- and N-channel blockers (Cardenas et al. 1995). Below evidence is presented that suggests that in these type 1 DRG cells, 5HT_1A receptors couple to high-threshold Ca²⁺ channels via two parallel pathways, a membrane-delimited pathway as previously reported in dorsal raphe neurons and cortical neurons (Foehring 1996; Penington et al. 1991), and a cytosolic diffusible pathway, which has not been reported elsewhere. Inhibition via the membrane delimited pathway is characterized by a voltage-sensitive slowing of current activation rate and coupling to N-type Ca²⁺ channels. Inhibition via the cytosolic diffusible pathway is characterized by a voltage-insensitive decrease in steady state current amplitude, dependence on the presence of intracellular Ca²⁺, and coupling to L-type Ca²⁺ channels.

Because the two pathways differ in several respects (voltage sensitivity, Ca²⁺ channel coupling, intracellular signaling molecules), modulation via either pathway may be inactivated selectively or enhanced by various physiological input. Such physiological input could include rapid firing rates, depolarization, and hyperpolarization by impinging afferents, cross-talk from other activated second messenger pathways, or prolonged exposure to agonist. Also due to the differences in the two pathways, selective reduction or enhancement of either pathway could modify changes in Ca²⁺ entry patterns produced by activation of 5HT_1A receptors.

	METHODS

Abstract Introduction Methods Results Discussion References

Male rats (75-150 g, Sprague Dawley purchased from Harlan) were rendered unconscious with methoxyflurane, decapitated, and DRG from thoracic and lumbar regions were dissected out. The ganglia were incubated at 36°C for 1 h in Tyrode's solution (composition below) containing 2 mg/ml collagenase (Sigma, Type 1) and 5 mg/ml Dipase II (Boehringer Mannheim). Individual DRG cell bodies were isolated by trituration and adhered to a poly-L-lysine coated coverslip stuck to the bottom of a 35-mm Petri dish and superfused with Tyrode's solution containing (in mM) 140 NaCl, 4 KCl, 2 MgCl₂, 2 CaCl₂, 10 glucose, 10 N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES), adjusted to pH 7.4 with NaOH. Currents were recorded either in the whole cell or cell-attached patch configuration using an Axopatch 200A (Axon Instruments). Voltage and current steps, holding potential, and data acquisition and analysis were controlled by an on-line IBM PC/AT clone computer programmed with Axobasic 1.0 (Axon Instruments).

Electrodes were fabricated from soda lime capillary glass (Scientific Products, B4416-1) using a Narishige 2-stage vertical puller, coated with silicone elastomer (Sylgard) to ~200 µm from the tip, and fire polished to a final resistance of 0.8-2.0 M, using a Narishige microforge. For voltage-clamp experiments, series resistance was estimated from capacity transients, as described previously (Scroggs and Fox 1992a). No data were included where series resistance resulted in greater than a 10-mV error in voltage commands.

Solutions were changed using the "sewer pipe" system, which consisted of a glass capillary tube mounted on a micromanipulator. The end of the glass capillary tube was placed near the cell under study, and the flow from it completely isolated the cell from the background flow of Tyrode's solution, which continuously flowed over the cells via another port. Different solutions were directed out of the capillary tube by means of a small manifold to which various 10-ml aliquots of drug or control solutions were connected. All experiments were done at room temperature (23°C).

Capsaicin (Sigma) was made daily as a 10 mM stock solution in 100% ethanol and diluted to 1 µM in Tyrodes' for experiments. Previous control experiments determined that the ethanol vehicle had no effect by itself (Del Mar et al. 1994). Omega-conotoxin GVIA (BacChem) was dissolved into 0.01 N acetic acid, aliquoted, lyophilized, and stored at 4°C for future use. Immediately before use the -conotoxin GVIA was dissolved in the external solution used for Ca²⁺ channel isolation. Stock solutions (10 mM) of nimodipine and ±Bay K 8644 (Research Biochemicals) were made in 100% ethanol and stored in the freezer. The dihydropyridines were diluted in external solution for superfusion of cells. Restricted light conditions were present during experiments involving dihydropyridines. (+)8-OH-DPAT, serotonin creatinine sulfate, 5-methoxy-N,N-dimethyl tryptamine, 5-carboxyamidotryptamine, pindolol, and spiperone (Research Biochemicals) first were dissolved as 1- or 10-mM stock solutions (made fresh daily) and then were diluted to appropriate concentrations in the external solution. Pindolol was first dissolved in slightly acidified water to obtain a 1-mM concentration. Spiperone was dissolved in dimethyl sulfoxide to obtain a 10-mM stock solution.

Experiments using the whole cell patch-clamp configuration were restricted to type 1 DRG cells identified by their small diameter, sensitivity to capsaicin, and lack of I_H and I_A, as described previously (Cardenas et al. 1995). For measuring the above mentioned K⁺-dependent phenomena, cells were superfused with Tyrodes externally, and the patch electrodes were filled with (in mM) 120 KCl, 5 2Na-ATP, 0.4 2Na-GTP, 5 ethylene glycol-bis(aminoethylether)-N,N,N,N-tetraacetic acid (EGTA), 2.25CaCl₂, 20 HEPES, adjusted to pH 7.4 with KOH. Total KCl after pH adjustment was 154 mM. Free [Ca²⁺]_i was calculated at 140 nM. Calcium channel currents (carried by Ba²⁺) were isolated by changing the Tyrodes' solution superfusing the outside of the DRG cell under study to one containing (in mM) 160 tetraethylammonium (TEA)-Cl, 2 BaCl₂, 10 HEPES, pH 7.4 with TEA-OH. Evidence that this combination of internal and external solutions adequately isolates Ca²⁺ currents has been presented previously (Del Mar et al. 1994). Furthermore, it has been demonstrated previously that the reduction of Ca²⁺ current amplitude by 5HT under the above conditions is due to a direct action on Ca²⁺ channels rather than activation of an outward K⁺ current (Del Mar et al. 1994). Unless otherwise stated, Ca²⁺ channel currents were evoked using 40-ms test depolarizations from a holding potential of 60 mV.

For cell-attached patch experiments the recording electrode was filled with a solution containing (in mM) 90 BaCl₂, 20 TEA, 10 CsCl, 20 HEPES, pH 7.4 adjusted with TEA-OH. The membrane potential of the cell under study was depolarized to near 0 mV by superfusing the cell with a solution containing (in mM) 140 K⁺ aspartate, 1 MgCl₂, 10 EGTA, 10 HEPES, adjusted to pH 7.4 with KOH. Data acquisition was achieved using the axobasic 1.0 single channel program "pinhed." A transmembrane potential of 60 mV was achieved by clamping the pipette tip to +60 mV. To trigger peak channel opening, the pipette tip potential was changed to around 20 mV. Macroscopic Ca²⁺ channel currents could be observed routinely by using a patch pipette with a relatively large tip, which included numerous Ca²⁺ channels. Peak currents recorded using this technique averaged 45 ± 6.9 pA (mean ± SE; n = 39). Some patches had no detectable current and were not included in the data analysis. In cell-attached patch recordings, DRG cells were not characterized regarding capsaicin sensitivity, I_H, or I_A, but recordings were restricted to small diameter cells to increase the likelihood of patching type 1 DRG cells.

In experiments regarding Ca²⁺ channel current amplitude, measurements were made from plots of current versus time. It was necessary to take into consideration the rate Ca²⁺ channel current rundown, which varied from cell to cell and decreased over time. To adjust for rundown in experiments where Ca²⁺ channel currents were blocked with 5HT agonists or Ca²⁺ channel antagonists, a line was drawn through the control data points and extrapolated out to a position over the peak effect of the drug. This position was considered to be the control Ca²⁺ channel current amplitude and was used to calculate the percent change in Ca²⁺ channel current amplitude produced by the drug.

In most cells, a significant portion of the decrease in peak Ca²⁺ channel current produced by 5HT agonists was due to a reduction in current activation rate (kinetic slowing). The percentage inhibition of peak Ca²⁺ current occurring via kinetic slowing was estimated using the formula: {[(pA reduction of peak current  pA reduction of steady state current)/pA peak current] * 100}, where "peak current" refers to the time point after the beginning of the test depolarization where peak current amplitude was observed under control conditions, and "steady state current" was the current amplitude observed at the end of the test depolarization. This formula likely results in an underestimation of the inhibition occurring via kinetic slowing because in many cells control current, and the current in the presence of 5HT had not quite reached steady state by the end of the 40-ms sweep. However, due to the slow rate of change in current amplitude around this time point, the error is not large.

Concentration-response curves were generated for 5HT in the absence and presence of 1 µM spiperone and for 8-OH-DPAT. Several concentrations of agonist were tested on individual type 1 DRG cells (1 agonist per cell) starting with the lowest concentration and proceeding to higher concentrations in ascending order. For estimation of EC₅₀, the inhibition of Ca²⁺ channel current produced by each concentration of agonist in a given cell was normalized as percent of maximal inhibition observed for each agonist in the same cell. Maximal inhibition for each agonist in individual cells was determined either from a plateau in the concentration response curve or, in some cases regarding 5HT concentration response curves, that observed with 5 or 10 µM 5HT, which were supramaximal concentrations in seven cells where 5HT was tested at higher concentrations. For estimation of the EC₅₀ for 5HT and 8-OH-DPAT, the data points were fitted with the binding isotherm R = R_max/[(1 + EC₅₀)/(5HT)ⁿ] using the nonlinear curve fitting program in Systat (SPSS). n (the number of molecules that have to bind to the receptor to produce activation) was assumed to be 1. For cells where 5HT or 8-OH-DPAT were tested in the presence of an antagonist, the cell first was superfused with the antagonist for ~1 min before testing of agonist. Spiperone and pindolol were without effect by themselves.

	RESULTS

Abstract Introduction Methods Results Discussion References

Whole cell patch-clamp experiments were restricted to type 1 DRG cells, which are characterized by small size, capsaicin sensitivity, lack of I_H, long duration action potentials, and thus resemble C-type nociceptors (Cardenas et al. 1995; Del Mar et al. 1994). We previously reported that 5HT consistently produces a pronounced inhibition of high-threshold Ca²⁺ channel currents in these type 1 DRG cells (Cardenas et al. 1995). Recent studies on action potential duration (see METHODS) in type 1 DRG cells suggests that the inhibition of Ca²⁺ channel current by 5HT produces a simultaneous reduction in action potential duration. Figure 1, A and B, illustrates the inhibition of Ca²⁺ channel current and the marked reduction in action potential duration produced by superfusion with 10 µM 5HT in a type 1 DRG cell. In six type 1 cells, which had an average control action potential duration of 8.2 ± 0.56 ms, the reduction of action potential duration at 1/2 peak amplitude by 10 µM 5HT averaged 33 ± 4.0%. In these same six cells, 10 µM 5HT inhibited peak high-threshold Ca²⁺ channel current by40 ± 4.0%.

View larger version (15K): [in this window] [in a new window]   FIG. 1. Effects of serotonin (5HT) on Ca2+ channel currents and action potential duration in a type 1 dorsal root ganglion (DRG) cell. A: action potential duration under control conditions and during peak effect of 10 µM 5HT. Action potentials were evoked by a 200-pA depolarizing current pulse through recording electrode in current clamp mode. Resting membrane potential was 60 mV. B: peak high-threshold Ca2+ channel current under control conditions and during peak effect of 10 µM 5HT, in same cell depicted in A. For A and B, pipette solution contained (in mM) 120 KCl, 5 2Na-ATP, 0.4 2Na-GTP, 5 MgCl2, 5 ethylene glycol-bis(-aminoethyl ether)-N,N,N,N-tetraacetic acid (EGTA), 2.25 CaCl2, and 20 N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES), adjusted to 7.4 with KOH (total KCl = 154 mM, free [Ca2+]i was calculated to be 140 nM). For A, external solution (Tyrode's) contained (in mM) 140 NaCl, 4 KCl, 2 MgCl2, 2 CaCl2, 10 glucose, 10 HEPES, adjusted to pH 7.4 with NaOH. For B, external solution contained (in mM) 160 tetraethylammonium (TEA)-Cl, 2 BaCl2, 10 HEPES, pH 7.4 with TEA-OH.

RECEPTOR SUBTYPE. A previous study demonstrated that the inhibition of Ca²⁺ channel currents in type 1 DRG cells was mimicked by the putative 5HT_1A selective agonist 8-OH-DPAT and that the inhibition produced by 8-OH-DPAT was antagonized by the putative 5HT_1A-selective antagonist NAN-190 (Cardenas et al. 1995; Cornfield et al. 1991; Del Mar et al. 1994; Glennon et al. 1988). These data were consistent with the idea that the inhibition of Ca²⁺ channel currents in type 1 DRG is mediated by a 5HT_1A receptor. However, recently several novel 5HT receptors (5HT_1D,5HT_1F, 5HT₅, and 5HT₇) have been demonstrated that also may be targets for 8-OH-DPAT and/or NAN-190 (Boess and Martin 1994). Thus the receptor subtype involved was further explored with pindolol, which is selective for 5HT_1A and 5HT_1B receptors versus the above mentioned 5HT receptor subtypes, and spiperone, which is selective for the 5HT_1A receptor versus the 5HT_1B receptor (Boess and Martin 1994; Sills et al. 1984).

Figure 2, A-C, illustrates concentration-response data for the inhibition of peak high-threshold Ca²⁺ channel current by 5HT (taken from a group of 12 type 1 DRG cells) and for 5HT in the presence of 1 µM spiperone (taken from a different group of 6 type 1 DRG cells). The EC₅₀ for 5HT was determined to be 196 nM in the absence of antagonist and was shifted rightward to 2.4 µM in the presence of 1 µM spiperone (Fig. 2, A-C). The antagonism by spiperone appeared to be surmountable by high concentrations of 5HT, consistent with a model of competitive antagonism. In the same type 1 cells where concentration-response was studied, the maximal inhibition of Ca²⁺ channel currents by 5HT alone (1-10 µM) averaged 37 ± 4.1% (n = 12), compared with 49 ± 7.3% (n = 6) observed for 5HT (5-50 µM) in the presence of 1 µM spiperone.

View larger version (31K): [in this window] [in a new window]   FIG. 2. Antagonism of 5HT agonist-induced inhibition of Ca2+ channel current in type 1 DRG cells by spiperone and pindolol. A: concentration-response curves for 5HT and 5HT + 1 µM spiperone, regarding the inhibition of Ca2+ channel current in type 1 DRG cells. Data points represent average percent inhibition produced by each concentration relative to maximal effect observed in each cell (n = 2-12 for each point). Error bars represent SE of mean. Maximal inhibition was determined as defined in METHODS. For estimation of EC50 for 5HT and 5HT + spiperone, data points were fitted with binding isothermR = Rmax/[(1 + EC50)/(5HT)n] as described in METHODS. B and C: representative current traces for 5HT (B) and 5HT + spiperone (C) taken from cells included in concentration response curves in A. For A-C, peak Ca2+ channel current was evoked from a holding potential of 60 mV. Numbers above each trace represent concentration of 5HT tested (in micromolar). D-G: antagonism by pindolol of 8-OH-DPAT-induced inhibition of Ca2+ channel current. D: plot of peak Ca2+ channel current vs. time illustrating effects of 1 µM 8-OH-DPAT on Ca2+ channel current amplitude under control conditions, during coapplication of 500 nM pindolol, and after washout of pindolol. E-G: current traces from experiment depicted in D showing effects of 8-OH-DPAT before, during, and after exposure of cell to pindolol. Solutions were same as in Fig. 1B.

The EC₅₀ for 8-OH-DPAT regarding inhibition of Ca²⁺ channel current was determined to be 276 nM in a group of seven type 1 DRG cells tested (not illustrated). The inhibitory effect of 1 µM 8-OH-DPAT on Ca²⁺ channel current in type 1 cells was antagonized potently by 500 nM pindolol (Fig. 2, D-G). In six cells tested, 1 µM 8-OH-DPAT produced an inhibition of peak Ca²⁺ current averaging 35 ± 3.8% under control conditions compared with an average inhibition of only 9.4 ± 1.9% when added in the presence of 500 nM pindolol. After washout of antagonist for several minutes, the inhibition produced by 8-OH-DPAT recovered to 31 ± 4.6%. The inhibition produced by 8-OH-DPAT after the washout of pindolol was significantly greater than that observed in the presence of pindolol (P < 0.05, paired-difference t-test).

5-carboxyamidotryptamine and 5-methoxy-N,N-dimethyl tryptamine, two additional 5HT agonists known to activate 5HT_1A receptors, also were tested for their ability to inhibit Ca²⁺ channel current in type 1 DRG cells. In group of four cells tested, 5-carboxyamidotryptamine (10 µM) inhibited Ca²⁺ channel current by 77.5 ± 3.1%, whereas 5-methoxy-N,N-dimethyl tryptamine (10 µM) inhibited Ca²⁺ channel current by 41 ± 5.8%, in another group of three cells tested.

EVIDENCE FOR TWO PARALLEL PATHWAYS. Several observations suggest that two parallel signaling pathways couple 5HT receptors to Ca²⁺ channels in type 1 DRG cells. One such observation was that the inhibition of Ca²⁺ channel currents by 5HT appeared to involve two components with different time courses: a slowing of current activation rate with a rapid time course and a decrease in overall current amplitude with a slower time course (Fig. 3, A and B). Similar components of neurotransmitter induced inhibition of Ca²⁺ currents have been observed previously and frequently are referred to as "kinetic slowing" and "steady state inhibition," respectively (Luebke and Dunlap 1994).

View larger version (21K): [in this window] [in a new window]   FIG. 3. Time course and voltage dependence of 5HT-induced inhibition of Ca2+ channel currents in a type 1 DRG cell. A: graph of peak high-threshold Ca2+ current vs. time. Ca2+ channel current was evoked every 10 s with a test depolarization to 10 mV from a holding potential of 60 mV. 5HT (10 µM) was superfused over cell during period indicated by bar. B: individual current sweeps taken from experiment depicted in A under control conditions and 10, 20, 30, 40, and 50 s after exposure to 5HT. Note that kinetic slowing was completed within first 10 s, whereas steady state inhibition persisted over next 4 sweeps. C: current traces before and during a 40-ms prepulse to +70 mV in a type 1 DRG cell. Prepulse produced only a 4.1% increase in current amplitude. D: effects of a prepulse given during peak inhibition of Ca2+ current with 5HT in same cell as C. Solutions were same as in Fig. 1B.

In 37 type 1 DRG cells challenged once with 10 µM 5HT, peak inhibition of Ca²⁺ channel current (averaging 53 ± 2.2%) occurred in an average time of 61 ± 2.5 s (Fig. 3A). Kinetic slowing was completed within the first 10 s after exposure to 5HT, whereas Ca²⁺ channel current amplitude continued to decrease for up to ~1 min via steady state inhibition (Fig. 3B). These two components of inhibition were observed in >90% of type 1 cells, regardless of whether 5HT, 8-OH-DPAT, 5-carboxyamidotryptamine, or 5-methoxy-N,N-dimethyl tryptamine was used as the agonist. As calculated using the formula described in METHODS, kinetic slowing reduced peak Ca²⁺ channel current by11.9 ± 1.4%, which accounted for an average of 22.5 ± 2.6% of the total inhibition (n = 37). Thus steady state inhibition accounted for ~75% of the total decrease in Ca²⁺ channel current amplitude.

The two components of inhibition were further differentiated by their sensitivity to reversal by strong depolarizing voltage commands (prepulses) delivered 3 ms before peak Ca²⁺ channel current was evoked. 5HT-induced kinetic slowing was reversible by strong depolarizing prepulses, whereas steady state inhibition was little affected. Figure 3C illustrates that under control conditions, a prepulse to +70 mV increased peak current amplitude (facilitation) in type 1 DRG cells only slightly by an average of 4.7 ± 0.4%(n = 34). However, when the prepulse was delivered during peak inhibition produced by 10 µM 5HT in the same cells, facilitation (relative to pre-5HT control current amplitude) was increased to 19 ± 1.1% (n = 34; Fig. 3D). During peak inhibition of Ca²⁺ channel currents by 5HT, the prepulse predominately reversed the kinetic slowing (Fig. 3D). These data suggest the participation of parallel pathways: one that mediates voltage-sensitive kinetic slowing and one that mediates steady state inhibition.

The idea of two pathways is supported by the observation that chelation of intracellular Ca²⁺ antagonized steady state inhibition but not voltage-sensitive kinetic slowing. Significantly less inhibition of Ca²⁺ currents was produced by 5HT when the patch electrodes were filled with a solution containing 10 mM bis-(o-aminophenoxy)-N,N,N,Ntetraacetic acid (BAPTA) and no added Ca²⁺ (BAPTA treatment) instead of the usual solution, which buffered intracellular Ca²⁺ at 140 nM (Fig. 4, A, B, and E). In BAPTA-treated type 1 DRG cells, 5HT reduced Ca²⁺ channel current by 27.2 ± 3.9% (n = 8) compared with a reduction of 54.5 ± 4.0% in controls (n = 15; Students' t-test, P < 0.05). This series of experiments included control and BAPTA-treated cells from the same group of rats to reduce the possibility of a sampling error.

View larger version (37K): [in this window] [in a new window]   FIG. 4. Interaction of chelation of intracellular Ca2+ with time course and voltage dependence of 5HT-induced inhibition of Ca2+ channel currents. A and B: plots of Ca2+ channel current vs. time in 2 different type 1 cells illustrating effects of internal Ca2+ chelation on magnitude and time course of 5HT-induced Ca2+ channel current inhibition. In A, internal Ca2+ was buffered to 140 nM with EGTA as per usual procedure. In B, internal Ca2+ was buffered to near 0 using an internal solution containing 10 mM bis(o - aminophenoxy) - N, N, N, N - tetraaceticacid (BAPTA) plus no added CaCl2. C and D: current sweeps taken from cells in A and B, respectively, under control conditions, during peak inhibition by 5HT, and after a prepulse delivered during peak inhibition by 5HT. E-H: summary of effects of BAPTA treatment on 5HT-induced inhibition of Ca2+ channel currents (E), percent of inhibition via kinetic slowing (F), facilitation of Ca2+ current in presence of 5HT, relative to control current amplitude (G), and percent of total inhibition occurring after 10 s post 5HT (H). Asterisk, responses statistically significant from controls, P < 0.05, Student's t-test. For A and C and for "control" in E-H, solutions were same as that used in Fig. 1B. For B and D and for "BAPTA" in E-H, solutions were same as that used in Fig. 1B, except that for internal solution, Ca2+ was omitted and 5 mM EGTA was replaced with 10 mM BAPTA.

A closer inspection of the above data revealed that the chelation of intracellular Ca²⁺ selectively affected steady state inhibition versus voltage-sensitive kinetic slowing. 5HT-induced kinetic slowing occurred in BAPTA-treated cells as well as controls, reducing Ca²⁺ channel current amplitude by 9 ± 1.1% in control cells (n = 15) compared with 11.2 ± 3.6% in BAPTA-treated cells (n = 8; Fig. 4, C, D, and F). Also, the voltage sensitivity of 5HT inhibition was not affected by Ca²⁺ chelation (Fig. 4, C, D, and G). In control cells, a prepulse to +70 mV during peak inhibition by 5HT facilitated Ca²⁺ channel current amplitude by 18 ± 1.9% (n = 12) compared with a facilitation of 16.2 ± 1.6% (n = 8) in BAPTA-treated cells (relative to pre-5HT control Ca²⁺ channel current amplitude). Conversely, BAPTA treatment selectively reduced the inhibition occurring after 10 s post 5HT, where steady state inhibition normally predominated (Fig. 4, A, B, and H). On average, 45.3 ± 4.6% (n = 15) of the total 5HT-induced inhibition occurred after 10 s post 5HT in controls versus only 24.3 ± 4.5% (n = 8) in BAPTA-treated cells (P < 0.5, Students' t-test).

Experiments conducted using the cell-attached patch configuration suggest that steady state inhibition occurred via a diffusible cytosolic pathway, whereas kinetic slowing occurred via a membrane delimited pathway. In the cell-attached patch configuration, superfusion of the cell body outside the patch pipette with 5HT (a lipophobic molecule) can only affect the Ca²⁺ channels in the patch under the pipette via a cytosolic diffusible pathway (Fig. 5F). Under these conditions, 5HT (10 µM) was observed to inhibit Ca²⁺ channel current by an average of 36 ± 5.1% in 13 of 39 small diameter cells tested. On average, peak inhibition occurred in 58 ± 4.5 s (n = 13), which is similar to the 61 ± 2.5 s to peak inhibition observed in the above mentioned whole cell patch experiments.

View larger version (25K): [in this window] [in a new window]   FIG. 5. Inhibition of Ca2+ channel currents by 5HT via a diffusible cytosolic signaling pathway. Ca2+ channel currents were recorded using cell-attached patch configuration (illustrated in F), using large diameter patch pipettes. A: plot of current vs. time in a small-diameter rat DRG cell. At time indicated by first arrow, 10 µM 5HT was applied to cell membrane outside patch pipette and washed away as indicated by second arrow. B: individual current traces taken from plot of current vs. time in A under control conditions and after peak inhibition by 5HT. C: current traces taken from same cell as in B after Ca2+ channel current amplitude recovered from inhibition by 5HT and after peak effect of application of 10 µM 5-methoxy-N,N-dimethyl tryptamine (5-OMe-DMT). D: in another cell, current traces under control conditions and during peak effect of 10 µM 5HT. E: current sweeps from same cell as D after washout of effects of 5HT and during peak effect of 1 µM 8-OH-DPAT (DPAT). F: illustration of 5HT-induced inhibition of Ca2+ channel current in a cell recorded from in cell-attached patch configuration. Molecules of 5HT (pentagons) cannot access receptors under patch. This leaves only a diffusible cytosolic signaling pathway () by which 5HT-receptor interactions can affect channels under patch. Pipette solution contained 90 mM Ba2+, 10 mM TEA, 10 mM CsCl, and 20 mM HEPES, pH 7.4 with TEA-OH. External solution contained (in mM) 140 K+ aspartate, 1 MgCl2, 10 EGTA, 10 HEPES, pH 7.4 with KOH.

An example of the 5HT-induced inhibition of Ca²⁺ channel currents under cell-attached patch conditions is illustrated in Fig. 5, A and B. Note that only steady state inhibition was induced and that no kinetic slowing took place (Fig. 5B). This lack of kinetic slowing was observed in each of the 13 cells where 5HT produced inhibition under cell-attached patch conditions and is consistent with the idea that steady state inhibition observed in whole cell patch experiments was mediated by a diffusible cytosolic pathway. In contrast, both kinetic slowing and steady state inhibition were produced on application of 5-methoxy-N,N-dimethyl tryptamine (a membrane permeant 5HT_1A agonist) outside the patch pipette to the same cell depicted in Fig. 5, A and B (Fig. 5C). Similarly, kinetic slowing and steady state inhibition were produced by 8-OH-DPAT (also membrane permeant) outside the patch pipette in three DRG cells recorded from in cell-attached patch configuration (Fig. 5E). In two of these cells, 5HT also was tested in the same manner and, as usual, only produced steady state inhibition (Fig. 5D). As reported above, both 5-methoxy-N,N-dimethyl tryptamine and 8-OH-DPAT produced kinetic slowing and steady state inhibition in type 1 DRG cells recorded from using the whole cell patch configuration. Thus the induction of kinetic slowing by 5-methoxy-N,N-dimethyl tryptamine and 8-OH-DPAT in the above experiments could be due to their penetration of the cell membrane, allowing access to 5HT_1A receptors under the patch pipette. Activated 5HT_1A receptors under the patch pipette then could couple to immediately adjacent Ca²⁺ channels via a membrane-delimited pathway.

Although recordings were limited to small diameter cells in these cell-attached patch experiments, the DRG cells could not be characterized as to action potential duration, expression of I_H, or sensitivity to capsaicin, The presence of 90 mM Ba²⁺ in the pipette tip precluded whole cell recordings. However, the percentage of cells where 5HT inhibited Ca²⁺ channel currents using the cell-attached patch configuration (13 out of 39 = 33%) was similar to the proportion of type 1 DRG cells randomly recorded from in previous whole cell patch experiments (~1/3), when recordings were limited to small-diameter cells (Cardenas et al. 1995).

CALCIUM CHANNEL COUPLING. To test for 5HT_1A receptor coupling to N- and L-type Ca²⁺ channels, the occlusion of 5HT-induced Ca²⁺ channel current inhibition by 1 µM -conotoxin GVIA (CTX) and 2 µM nimodipine, respectively, was measured (Bean 1991; Boland et al. 1994; McCarthy and TanPiengco 1992). 5HT (10 µM) reversibly blocked 54 ± 3.2% (n = 12) of the Ca²⁺ channel current under control conditions compared with only 34.2 ± 4.3%, when N channels were blocked by 1 µM -conotoxin GVIA (Fig. 6, A-C and G). These data suggest that N channels were coupled to the 5HT_1A receptors. In a similar series of experiments targeting L-type Ca²⁺ channels, 10 µM 5HT inhibited the Ca²⁺ channel current by 49 ± 3.9% (n = 10) under control conditions, compared with only 27 ± 2.2%, when L channels were blocked by 2 µM nimodipine (Fig. 6, D-G). Thus L channels also appeared coupled to 5HT_1A receptors. (Percent inhibition of Ca²⁺ channel current by 5HT after blockade of N or L channels was calculated relative to control peak current observed immediately before treatment with channel blocker.)

View larger version (37K): [in this window] [in a new window]   FIG. 6. Effects of selective N- and L-channel antagonists on 5HT-induced inhibition, kinetic slowing, and facilitation of Ca2+ channel current by prepulses. A: plot of Ca2+ channel current amplitude vs. time illustrating inhibition of Ca2+ current by 5HT and effects of prepulses (PP) before and during N-channel blockade with 1 µM -conotoxin GVIA (CTX). B and C: representative current sweeps taken from experiment depicted in A under different conditions as marked. D-F: same as A-C except that L-channels were blocked with 2 µM nimodipine. G and H: bar graphs summarizing data obtained by repeating experiments depicted in A-F. G summarizes inhibition produced by 5HT under control conditions (Control) and after blockade of N channels (CTX) or L channels (NIM) or both N and L channels (CTX + NIM). Control responses did not vary significantly between these 3 sets of experiments and are grouped together. H-J summarize effects of CTX or nimodipine treatment on percent inhibition occurring via kinetic slowing (H), facilitation by a prepulse during peak inhibition by 5HT (I), and percent of total inhibition that occurs after 10 s post 5HT (J). * and **, responses statistically significant from pair matched controls, P < 0.05, paired difference test. Solutions were same as those used in Fig. 1B.

The effect of 5HT also was tested on the later portion of Bay K 8644 prolonged Ca²⁺ channel tail currents, which are conducted exclusively through L-type Ca²⁺ channels (Plummer et al. 1989). As illustrated in Fig. 7, A-C, 5HT decreased the amplitude of the Bay K 8644 prolonged tail currents at a time point where the tail current was deactivated completely under control conditions (n = 7). Additional experiments were conducted to test whether this effect might be due to poor voltage control. Both the peak current and tail current turned on gradually in response to graduated increases in the amplitude of the depolarizing test pulse, indicating good voltage control (n = 2; Fig. 7, D and E). In addition, 5HT still reduced the amplitude of Bay K 8644 prolonged tail currents in three type 1 cells held at 40 mV; this greatly reduced the amplitude of peak and tail currents (Fig. 7F). These data support the idea that L-type Ca²⁺ channels are coupled to 5HT_1A receptors in type 1 DRG cells.

View larger version (27K): [in this window] [in a new window]   FIG. 7. Blockade of Bay K 8644-induced L-type Ca2+ channel tail currents by 5HT in a type 1 DRG cell. A: plot of current vs. time illustrating effects of 10 µM 5HT on peak Ca2+ channel currents and tail currents in a type 1 DRG cell under control conditions and after superfusion of cell with 100 nM Bay K 8644. Top: open circle, peak whole cell Ca2+ channel current; bottom: filled circle, represents tail current measured at 4 ms after cessation of depolarizing test command. B and C: current traces from experiment depicted in A before and after 5HT under control conditions and before and after 5HT in presence of 5HT. Letters near each current trace indicate where traces were taken from in plot of current vs. time shown in A. D: current voltage relationship of Ca2+ current in presence of 100 nM Bay K 8644. Series resistance was 0.98 M. E: current traces from I-V curve depicted in D. F: effects of 10 µM 5HT on Bay K 8644 prolonged tail currents in a type 1 cell held at 40 mV to reduce current amplitude. Solutions were same as in Fig. 1B.

A small amount of 5HT-induced inhibition of Ca²⁺ channel currents persisted in cells simultaneously superfused with 2 µM nimodipine and 1 µM CTX (summarized in Fig. 6G). In four type 1 DRG cells tested, 5HT inhibited Ca²⁺ channel currents by 56 ± 4.9% under control conditions and by 7.5 ± 0.6% (relative to control Ca²⁺ current amplitude before addition of antagonists) after superfusion with nimodipine and CTX. Thus it is possible that some non-N/L Ca²⁺ channel type is coupled to 5HT_1A receptors in these cells. An alternate possibility is that 5HT was modulating L channels that were not blocked by 2 µM nimodipine, due to a reduction in efficacy of the dihydropyridine antagonist at negative membrane potentials (Bean 1983). It is unlikely that the nimodipine/CTX-resistant modulation was achieved by inhibition of T-type Ca²⁺ channels, which were little affected in type 1 cells where 5HT produced a large inhibition of high-threshold Ca²⁺ channel current (n = 2; Fig. 8).

View larger version (14K): [in this window] [in a new window]   FIG. 8. Lack of effect of 5HT on T-type Ca2+ currents in a type 1 cell. Cell was held at 90 mV and depolarized to 40 mV to evoke T currents. A subsequent depolarization to 10 mV 3 ms later evoked high-threshold Ca2+ channel current. Superimposed current traces are shown under control conditions and after peak effect of 10 µM 5HT on high-threshold Ca2+ channel current. Solutions were same as in Fig. 1B.

Further inspection of the interaction between 5HT inhibition and CTX and nimodipine indicated that 5HT-induced kinetic slowing predominately involved a decrease in N-channel activity, whereas the steady state inhibition included a decrease in L-channel activity. CTX treatment reduced kinetic slowing by 82 ± 3.4% (n = 12) relative to pre-CTX controls (Fig. 6, B, C, and H). Also, relief of 5HT-induced inhibition by a prepulse was abolished completely by CTX treatment (Fig. 6, A-C and I). After CTX treatment, facilitation produced by a prepulse during peak inhibition by 5HT was reduced to an average of 9.7 ± 6.5% (n = 12) of pre-CTX control values (after subtraction of the pre-5HT control facilitation). Consistent with the above mentioned effects, blockade of N channels reduced the 5HT-induced inhibition occurring within the first 10 s after 5HT challenge when kinetic slowing normally occurred. Accordingly, N-channel blockade increased the proportion of inhibition occurring after 10 s post 5HT, when steady state inhibition typically predominated (Fig. 6, A and J).

An opposite pattern was observed regarding the above phenomena in cells where L channels were blocked with nimodipine. Kinetic slowing was reduced by only 34 ± 5.0% (n = 10; Fig. 6, E, F, and H). Facilitation resulting from a prepulse during peak inhibition by 5HT was little affected, averaging 82 ± 8.2% (n = 10) relative to pre-nimodipine controls (Fig. 6, D-F and I). Finally, blockade of L channels increased the proportion of inhibition that occurred within the first 10 s after 5HT challenge (when kinetic slowing took place) and reduced the inhibition of Ca²⁺ channel current that occurred after 10 s post 5HT (when steady state inhibition normally predominated; Fig. 6, D and J). In this regard, the L-channel blocker nimodipine had a similar effect as chelation of intracellular Ca²⁺ with BAPTA, described above.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Several lines of evidence, presented above, suggest that 5HT_1A receptors are coupled to high-threshold Ca²⁺ channels by two parallel pathways in type 1 DRG cells. Regarding the 5HT receptor subtype, in a previous report we concluded that a 5HT_1A receptor was involved, based on the activity of the putative 5HT_1A selective ligands 8-OH-DPAT and NAN-190 (Del Mar et al. 1994; Glennon et al. 1988). However, it is now apparent that the Del Mar et al. (1994) study was inconclusive, due to the demonstration of novel 5HT receptor subtypes (5HT_1D, 5HT₅, and 5HT₇), which potentially are affected by concentrations of 8-OH-DPAT and/or NAN-190 close to those used in the Del Mar et al. (1994) study (Boess and Martin 1994). However our new data, demonstrating that pindolol and spiperone are potent antagonists of 5HT-receptor-mediated inhibition of Ca²⁺ channel current in type 1 DRG cells, reaffirms the involvement of a 5HT_1A receptor. The nearly complete antagonism by 500 nM pindolol of inhibition of Ca²⁺ channel currents by a nearly maximal concentration of 8-OH-DPAT (1 µM), fits best with its affinity at the 5HT_1A receptor (pKi, 7.5) or 5HT_1B receptor (pKi, 6.3) versus its affinity at 5HT₅, 5HT₇, and 5HT_1D receptors (pKi < 6) (Boess and Martin 1994; Dumis et al. 1988; Harwood et al. 1995; Lovenberg et al. 1993; Wisden et al. 1993). The potent antagonism of 5HT's action by 1 µM spiperone fits best with spiperone's affinity for the 5HT_1A receptor (pKi, 7.2) versus the 5HT_1B receptor (pKi, < 5) (Boess and Martin 1994; Sills et al. 1984).

The EC₅₀ regarding 5HT inhibition of Ca²⁺ channel currents in type 1 DRG cells (196 nM) is somewhat higher than the EC₅₀ observed for 5HT regarding some other reputed 5HT_1A-mediated responses including inhibition of Ca²⁺ currents in acutely isolated rat cortical neurons (44 nM), increase in a K⁺ conductance in acutely isolated rat dorsal raphe neurons (30 nM), or inhibition of adenylate cyclase activity in cultured hippocampal neurons (50 nM) (Dumuis et al. 1987; Foehring 1996; Penington et al. 1993). However, the EC₅₀ for 5HT has been shown to vary from low nanomolar to low micromolar concentrations in different expression systems transfected with 5HT_1A receptors (reviewed in Boess and Martin 1994). This variation appears to be related to 5HT_1A receptor density and 5HT_1A receptor coupling to different signaling pathways (Boess and Martin 1994).

The differences in the ED₅₀ regarding the effects of 5HT in the above physiological systems versus that observed regarding type 1 DRG cells also may reflect variations in 5HT_1A receptor density and/or coupling to second messenger systems. The ion channel coupling to 5HT_1A receptors in cortical and dorsal raphe neurons appears to exclusively involve membrane-delimited pathways (Foehring 1996; Penington et al. 1993), whereas in type 1 DRG cells, a cytosolic diffusible component is included. In addition, membrane-permeant adenosine 3,5-cyclic monophosphate analogs neither mimic nor occlude inhibition of Ca²⁺ channel currents by 5HT in type 1 DRG cells, suggesting that adenylate cyclase is not involved (Cardenas, unpublished observations). It is also possible that 5HT_1A receptor density varies between type 1 DRG cells and the above mentioned central neurons.

Another possible source of variation in the affinity of the 5HT_1A receptor for 5HT and other ligands is alteration of the receptor by the enzyme treatment used in the cell dissociation procedure, which differed between the studies on acutely isolated neurons mentioned above and this study (Foehring 1996; Penington et al. 1993). However the pharmacological profile of the putative 5HT_1A receptor in type 1 cells suggests that it is not severely altered. The ED₅₀s for (+)8-OH-DPAT and 5HT and are not greatly different regarding Ca²⁺ channel inhibition in type 1 DRG cells (276 and 196 nM, respectively), similar to the situation at 5HT_1A receptors in rat cortex neurons (15 and 44 nM, respectively), 5HT_1A receptors transfected into hamster CHO cells (73 and 146 nM, respectively), and 5HT_1A receptors expressed by cultured rat hippocampal neurons (8 and 50 nM, respectively) (Dumuis et al. 1987; Foehring 1996; Raymond et al. 1992). Furthermore, the putative 5HT_1A receptor in type 1 DRG cells is sensitive to antagonism by typical 5HT_1A-selective antagonists such as spiperone and NAN-190 (Del Mar et al. 1994; this study). Also the receptor is strongly activated by 5-carboxyamidotryptamine and 5-methoxy-N,N-dimethyl tryptamine, which have been shown to activate 5HT_1A receptors in other systems (Boess and Martin 1994).

The experiments demonstrating that the inhibition of Ca²⁺ channel currents is occluded partially by blockade of Ca²⁺ current with CTX/and or nimodipine indicate that N and L channels are negatively coupled to 5HT receptors in type 1 DRG cells. In addition, the reduction of Bay K 8644 induced tail currents by 5HT is further evidence that L-channel currents are involved. The pharmacological data indicating involvement of N and L channels is consistent with the observation that frequently 5HT_1A agonists inhibit 60-75% of peak high-threshold Ca²⁺ current in type 1 cells. This degree of inhibition is more than can be explained by the blockade of N or L channels alone, because high-threshold Ca²⁺ current in type 1 cells is on average 28% N type and 46% L type and 26% CTX/nimodipine resistant (Cardenas et al. 1995).

Some modulation of Ca²⁺ currents by 5HT was observed even after both N and L channels were blocked with supposedly saturating concentrations of CTX (1 µM) or nimodipine (2 µM) (Bean 1991; Boland et al. 1994; McCarthy and TanPiengco 1992), indicating the possibility that some other high-threshold Ca²⁺ channel type also is coupled to 5HT receptors in type 1 cells. However, because we have not directly determined that 2 µM nimodipine is a saturating concentration regarding L channels at the membrane potential of 60 mV used in the experiment, it is possible that some or all of the residual modulation of Ca²⁺ channel current in the presence of 2 µM nimodipine and 1 µM CTX may be due to modulation of unblocked L channels.

Several observations made in the present study are consistent with the idea that the 5HT_1A receptors couple negatively to N- and L-type Ca²⁺ channels via two different signaling pathways. The inhibition of Ca²⁺ channel currents by 5HT consisted of two components with different time courses: a rapidly occurring kinetic slowing and a slower steady state inhibition. Also, the steady state inhibition was antagonized by chelation of intracellular Ca²⁺, whereas the kinetic slowing was not, suggesting that the two components involve different signaling molecules.

The data obtained from cell-attached patch recordings provides evidence that the kinetic slowing occurred via a membrane delimited pathway, whereas steady state inhibition required a cytosolic diffusible component. When recording in the cell-attached patch configuration, 5HT applied outside the patch produced only steady state inhibition. 5HT is not membrane permeant and thus would not be expected to gain access to the receptors under the patch pipette to activate receptors tightly coupled to channels in a membrane-delimited pathway. Thus only a diffusible cytosolic pathway was available to actuate the 5HT-induced steady state inhibition of Ca²⁺ channels under the patch pipette. On the other hand, membrane-permeant 5HT_1A agonists (8-OH-DPAT and 5-methoxy-N,N-dimethyl tryptamine) produced both steady state inhibition and kinetic slowing. These agonists could be expected to gain access to the receptors under the patch pipette as well as those outside the patch pipette and thus activate both membrane-delimited pathways and cytosolic diffusible pathways.

The interaction of Ca²⁺ channel blockers with 5HT-induced inhibition suggests that the kinetic slowing involved modulation of N-type Ca²⁺ channels, whereas steady state inhibition involved modulation of L channels. Blockade of N-type Ca²⁺ channels with CTX occluded kinetic slowing induced by 5HT. Accordingly, blockade of N channels selectively reduced the proportion of inhibition that occurred in the first 10 s after addition of 5HT, which is the time period when kinetic slowing took place under control conditions. On the other hand, blockade of L-type channels with nimodipine did not greatly affect 5HT-induced kinetic slowing. However, L-channel blockade significantly reduced the proportion of inhibition that occurred after the first 10 s of exposure to 5HT, a time when steady state inhibition predominated under control conditions.

The above data do not answer the question as to whether N channels participate in steady state inhibition. As mentioned above, a small amount of steady state inhibition by 5HT was observed after blockade of N and L channels with nimodipine and CTX, suggesting the possibility that non-N/L channel(s) participate in steady state inhibition. Thus it seems possible that N channels also play a role in steady state inhibition.

Several observations support the idea that the 5HT-induced kinetic slowing was voltage sensitive, whereas the steady state inhibition was not. 5HT-induced kinetic slowing was reversed by prepulses to +70 mV. In addition, when kinetic slowing was abolished by selective blockade of N-type Ca²⁺ channels, prepulses no longer reversed inhibition of Ca²⁺ channel current by 5HT. Thus the voltage sensitivity of 5HT-induced inhibition depended on the presence of kinetic slowing. Conversely, prepulses had little effect on steady state inhibition, which was especially obvious when kinetic slowing was blocked with CTX. Together these data suggest that 5HT_1A receptors couple negatively to Ca²⁺ channels via two pathways: a membrane-delimited pathway, which couples to N channels and actuates voltage-sensitive kinetic slowing, and a pathway dependent on a cytosolic diffusible component and free intracellular Ca²⁺ that couples to L channels and actuates steady state inhibition.

The normal physiological role of the 5HT_1A receptor coupling to Ca²⁺ channels in DRG cell bodies is not clear. It is possible that neuroactive substances are released in the DRG. Several reports have been made of sparsely distributed synapses in DRG including varicosities containing norepinephrine and acetylcholine (Kayahara et al. 1981; Kummer 1994). In addition, serotonergic fibers have been observed in dorsal roots, suggesting the possibility of serotonin release in the DRG (Di Carlo 1983). Also, it is possible that the DRG cell bodies themselves release neuroactive substances, similar to sympathetic and parasympathetic neuronal cell bodies (Johnson and Pilar 1980; Suetake et al. 1981). Thus it is possible that various neurotransmitter-ion channel systems are expressed by subpopulations of DRG cell bodies to transduce this putative chemical input into changes in metabolism, growth, and excitability.

Under physiological conditions, where Ca²⁺ channels are triggered to open by action potentials, the inhibition of Ca²⁺ entry into type 1 DRG cell bodies by 5HT or other neuroactive agents might be exaggerated due to the simultaneous reduction in action potential duration. The change in Ca²⁺ channel activity produced by 5HT may reduce Ca²⁺ entry by two independent mechanisms, a direct decrease in the permeability of the membrane to Ca²⁺ during a depolarization and a secondary reduction resulting from shortening of the action potential duration. Previous studies have demonstrated that decreasing the duration of action potentials, used as depolarizing stimulus in voltage-clamp experiments, can decrease the amount of Ca²⁺ entry into neurons independently of any direct change in channel open probability (McCobb and Beam 1991; Penington et al. 1992; Scroggs and Fox 1992b). On the other hand, a neurotransmitter-induced reduction of Ca²⁺ channel opening probability will reduce Ca²⁺ entry evoked by a constant depolarizing stimulus (Bean 1991). Thus the reduction in Ca²⁺ entry due to 5HT-induced action potential narrowing could be expected to be additive to the reduction in Ca²⁺ entry produced by 5HT-induced inhibition of Ca²⁺ channel opening. Such a scenario has been demonstrated previously regarding 5HT inhibition of Ca²⁺ entry during action potentials in dorsal raphe neurons (Penington et al. 1992).

It is unknown what purpose is served by having parallel signaling pathways coupling 5HT receptors to Ca²⁺ channels in the type 1 DRG cells. The answer may lie in the fact that the two pathways are distinctly different regarding key characteristics such as coupling to Ca²⁺ channel subtypes, susceptibility to reversal by strong positive voltage, and participation of different molecules in the signaling pathway. These differences may permit various physiological input to selectively inactivate one or the other pathway. For example, one might expect rapid trains of action potentials to selectively reduce modulation via the membrane-delimited pathway by relieving the voltage-sensitive kinetic slowing produced by 5HT. Also, the voltage-sensitive kinetic slowing predominately involved N channels, whereas L-channel inhibition participated mainly in steady state inhibition. Thus due to the fact that N channels become inactivated at more negative membrane potentials than L channels (Fox et al. 1987a,b), depolarization or hyperpolarization of the membrane potential could be expected to alter the participation of the membrane-delimited pathway versus the diffusible cytosolic pathway. Finally, it is likely that different molecules are involved in each pathway, as evidenced by the dependence of the diffusible cytosolic pathway on free intracellular Ca²⁺ and the other various different characteristics of the two pathways. Participation of different molecules could lead to variations in the modulation of the two pathways by activation of other receptors. For example, activation of protein kinase C has been observed to reduce neurotransmitter-induced inhibition of Ca²⁺ currents via voltage-sensitive pathways in several neuronal types (Swartz 1993). Possibly activation of other second messenger pathways through synaptic activity or prolonged exposure to 5HT also might inactivate selectively one or the other pathways coupling 5HT_1A receptors to Ca²⁺ channels.

Selective inactivation or enhancement of the membrane-delimited pathway versus the cytosolic diffusible pathway might have important consequences because the two pathways are coupled to different Ca²⁺ channel subtypes, have different overall time courses, and most likely differ regarding the timing of inhibition of Ca²⁺ entry relative to the rise and fall of the action potential. The coupling of the membrane-delimited and the cytosolic diffusible pathways to different Ca²⁺ channel subtypes may be particularly important because different Ca²⁺ channel types have been shown to be variably coupled to the release of different neurotransmitters (Waterman 1996).

If 5HT inhibits Ca²⁺ currents in the afferent terminals of type 1 DRG cells, which resemble nociceptors, the release of 5HT into the spinal cord could be expected to reduce greatly neurotransmitter release from them. This possibly could explain the antinociceptive effect of 5HT release into the spinal cord in some cases (Yaksh and Wilson 1979). Regarding such cases, the antinociceptive effects of 5HT also could be susceptible to impinging signals, which alter firing rates, change membrane potential, or activate alternate signaling pathways. A point of interest here is the possibility that L channels, which are not typically found at synapses, might be involved in neurotransmitter release from some types of nociceptors. This might render certain types of pain susceptible to blockade by dihydropyridines.

	ACKNOWLEDGEMENTS

  This work was supported by National Science Foundation Grant IBN 93-10065 to R. S. Scroggs and National Institute of Neurological Disorders and Stroke Grant NS-30600-01A2 to E. G. Anderson.

	FOOTNOTES

  Address for reprint requests: C. G. Cardenas, Dept. of Anatomy and Neurobiology, College of Medicine, University of Tennessee, Memphis, TN 38163.

  Received 8 November 1996; accepted in final form 19 February 1997.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• BEAN, B. P. Neurotransmitter inhibition of neuronal calcium currents by changes in channel voltage dependence. Nature Lond. 340: 153-156, 1989.[Medline]
• BEAN, B. P. Pharmacology of calcium channels in cardiac muscle, vascular muscle, and neurons. Am. J. Hypertens. 4: 406S-411S, 1991.[Medline]
• BEECH, D. J., BERNHEIM, L., MATHIE, A., HILLE, B. Intracellular Ca²⁺ buffers disrupt muscarinic suppression of Ca²⁺ current and M current in rat sympathetic neurons. Proc. Natl. Acad. Sci. USA 88: 652-656, 1991.[Abstract/Free Full Text]
• BERNHEIM, L., BEECH, D. B., HILLE, B. A diffusible second messenger mediates one of the pathways coupling receptors to calcium channels in rat sympathetic neurons. Neuron 6: 859-867, 1991.[Medline]
• BOESS, F. G., MARTIN, I. L. Molecular biology of 5-HT receptors. Neuropharmacology 33: 275-317, 1994.[Medline]
• BOLAND, L. M., ALLEN, A. C., DINGLEDINE, R. Inhibition by bradykinin of voltage-activated barium current in a rat dorsal root ganglion cell line: role of protein kinase C. J. Neurosci. 11: 1140-1149, 1991.[Abstract]
• BOLAND, L. M., MORRILL, J. A., BEAN, B. P. -Conotoxin block of N-type calcium channels in frog and rat sympathetic neurons. J. Neurosci. 14: 5011-5027, 1994.[Abstract]
• BROWN, A. M. Membrane-delimited cell signaling complexes: direct ion channel regulation by G proteins. Membr. Biol. 131: 93-102, 1993.[Medline]
• CARDENAS, C. G., DEL MAR, L. P., SCROGGS, R. S. Variation of serotonergic inhibition of calcium channel currents in four types of rat sensory neurons differentiated by membrane properties. J. Neurophysiol. 74: 1870-1879, 1995.[Abstract/Free Full Text]
• CORNFIELD, L. J., LAMBERT, G., ARVIDSSON, L., MELLIN, C., VALLGARDA, J., HACKSELL, U., NELSON, D. L. Intrinsic activity of enatiomers of 8-hydroxy-2-(di-n-propylamino)tetralin and its analogs at 5-hydroxytryptamine1A receptors that are negatively coupled to adenylate cyclase. Mol. Pharmacol. 39: 780-787, 1991.[Abstract]
• DEL MAR, L. P., CARDENAS, C. G., SCROGGS, R. S. Serotonin inhibits high-threshold Ca²⁺ channel currents in capsaicin sensitive acutely isolated adult rat sensory neurons. J. Neurophysiol. 72: 2551-2554, 1994.[Abstract/Free Full Text]
• DI CARLO, V. Serotonergic fibers in dorsal root of the spinal cord. Neurosci. Lett. 43: 119-125, 1983.[Medline]
• DIVERSE-PIERLUISSI, M., DUNLAP, K. Distinct, convergent second messenger pathways modulate neuronal calcium currents. Neuron 10: 753-760, 1993.[Medline]
• DIVERSE-PIERLIUSSI, M., GOLDSMITH, P. K., DUNLAP, K. Transmitter-mediated inhibition of N-type calcium channels in sensory neurons involves multiple GTP-binding proteins and subunits. Neuron 14: 191-200, 1995.[Medline]
• DUMUIS, A., SEBBEN, M., BOCKAERT, J. Pharmacology of 5-hydroxytryptamine-1A receptors which inhibit cAMP production in hippocampal and cortical neurons in primary culture. Mol. Pharmacol. 33: 178-186, 1988.[Abstract]
• FOEHRING, R. C. Serotonin modulates N- and P-type calcium currents in neocortical pyramidal neurons via a membrane delimited pathway. J. Neurophysiol. 75: 648-659, 1996.[Abstract/Free Full Text]
• FORSCHER, P., OXFORD, G. S., SCHULTZ, D. Noradrenaline modulates calcium channels in avian dorsal root ganglion cells through tight receptor-channel coupling. J. Physiol. Lond. 379: 131-144, 1986.[Abstract/Free Full Text]
• FOX, A., NOWYCKY, M., TSIEN, R. Kinetic and pharmacological properties distinguish three types of calcium channels in chick sensory neurons. J. Physiol. Lond. 394: 149-172, 1987a.[Abstract/Free Full Text]
• FOX, A., NOWYCKY, M., TSIEN, R. Single channel recordings on three types of calcium channels in chick sensory neurons. J. Physiol. Lond. 394: 173-200, 1987b.[Abstract/Free Full Text]
• GLENNON, R. A., NAIMAN, N. A., PIERSON, M. E., TITELER, M., LYON, R. A., WEISBERG, E. NAN-190: an arylpiperazine analog that antagonizes the stimulus effects of the 5-HT_1A agonist 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT). Eur. J. Pharmacol. 154: 339-341, 1988.[Medline]
• GREEN, K. A., COTTRELL, G. A. Actions of baclofen on components of the Ca-current in rat and mouse DRG neurones in culture. Br. J. Pharmacol. 94: 235-245, 1988.[Medline]
• HARWOOD, G., LOCKYER, M., GILES, H., FAIRWEATHER, N. Cloning and characterization of the rabbit 5-HT_1D and 5-HT_1D receptors. FEBS Lett. 377: 73-76, 1995.[Medline]
• HIRNING, L. D., FOX, A. P., MILLER, R. J. Inhibition of calcium currents in cultured myenteric neurons by neuropeptide Y: evidence for direct receptor/channel coupling. Brain Res. 532: 120-130, 1990.[Medline]
• HILLE, B. G protein-coupled mechanisms and nervous signaling. Neuron 9: 187-195, 1992.[Medline]
• HILLE, B. Modulation of ion-channel function by G-protein-coupled receptors. Trends Neurosci. 17: 531-535, 1994.[Medline]
• HOWE, A. R., SURMEIER, D. J. Muscarinic receptors modulate N-, P-, and L-type Ca²⁺ currents in rat striatal neurons through parallel pathways. J. Neurosci. 15: 458-469, 1995.[Abstract]
• JOHNSON, D. A., PILAR, G. The release of acetylcholine from post-ganglionic cell bodies in response to depolarization. J. Physiol. Lond. 299: 605-619, 1980.[Abstract/Free Full Text]
• KAYAHARA, T., TAKIMOTO, T., SAKASHITA, S. Synaptic junctions in the cat spinal cord. Brain Res. 216: 277-290, 1981.[Medline]
• KUMMER, W. Sensory ganglia as a target of autonomic and sensory nerve fibers in the guinea-pig. Neuroscience 59: 739-754, 1994.[Medline]
• LOVENBERG, T. W., BARON, B. M., DE LECEA, L., MILLER, J. D., PROSSER, R. A., REA, M. A., FOYE, P. E., RACKE, M., SLONE, A. L., SIEGEL, B. W., DANIELSON, P. W., SUTCLIFF, J. G., ERLANDER, M. G. A novel adenylyl cyclase-activating serotonin receptor (5-HT₇) implicated in the regulation of mammalian circadian rhythms. Neuron 11: 449-458, 1993.[Medline]
• LUEBKE, J. I., DUNLAP, K. Sensory neuron N-type calcium channels are inhibited by both voltage-dependent and-independent mechanisms. Pfluegers Arch. 428: 499-507, 1994.[Medline]
• MCCARTHY, R., TANPIENGCO, P. Multiple types of high-threshold calcium currents in rabbit sensory neurons: high affinity block of neuronal L-type by nimodipine. J. Neurosci. 12: 2225-2234, 1992.[Abstract]
• MCCOBB, D. P., BEAM, K. G. Action potential waveform voltage-clamp commands reveal striking differences in calcium entry via low and high-threshold voltage-activated calcium channels. Neuron 7: 119-127, 1991.[Medline]
• PENINGTON, N. J., KELLY, J. S., FOX, A. P. A study of the mechanism of Ca²⁺ current inhibition produced by serotonin in rat dorsal root raphe neurons. J. Neurosci. 11: 3594-3609, 1991.[Abstract]
• PENINGTON, N. J., KELLY, J. S., FOX, A. P. Action potential waveforms reveal simultaneous changes in I_Ca and I_K produced by 5-HT in rat dorsal raphe neurons. Proc. R. Soc. Lond. B Biol. Sci. 248: 171-179, 1992.[Medline]
• PENINGTON, N. J., KELLY, J. S., FOX, A. P. Whole-cell recordings of inwardly rectifying K⁺ currents activated by 5HT_1A receptors on dorsal raphe neurones of the adult rat. J. Physiol. Lond. 469: 387-405, 1993.[Abstract/Free Full Text]
• PLUMMER, M. R., LOGOTHETIS, D. E., HESS, P. Elementary properties and pharmacological sensitivities of calcium channels in mammalian peripheral neurons. Neuron 2: 1453-1463, 1989.[Medline]
• RAYMOND, J. R., ALBERS, F. J., MIDDLETON, J. P. Functional expression of human 5-HT_1A receptors and differential coupling to second messengers in CHO cells. Naunyn-Schmiedebergs Arch. Pharmacol. 346: 127-137, 1992.[Medline]
• SCROGGS, R. S., FOX, A. P. Calcium current variation between acutely isolated adult rat dorsal root ganglion neurons of different size. J. Physiol. Lond. 445: 639-658, 1992a.[Abstract/Free Full Text]
• SCROGGS, R. S., FOX, A. P. Multiple Ca²⁺ currents elicited by action potential waveforms in acutely isolated adult rat dorsal root ganglion neurons. J. Neurosci. 12: 1789-1801, 1992b.[Abstract]
• SHAPIRO, M. S., WOLLMUTH, L. P., HILLE, B. Angiotensin II inhibits calcium and M current channels in rat sympathetic neurons via G proteins. Neuron 12: 1319-1329, 1994.[Medline]
• SILLS, M. A., WOLFE, B. B., FRAZER, A. Determination of selective and nonselective compounds for the 5-HT_1A and 5-HT_1B receptor subtypes in rat frontal cortex. J. Pharmacol. Exp. Ther. 231: 480-487, 1984.[Abstract/Free Full Text]
• SURMEIER, D. J., BARGAS, J., HEMMINGS, H. C., NARIN, A. C., GREENGARD, P. Modulation of calcium currents by a D1 dopaminergic protein kinase/phosphatase cascade in rat neostriatal neurons. Neuron 14: 1-20, 1995.[Medline]
• SUETAKE, K., KOJIMA, H., INANAGA, K., KOKETSU, K. Catecholamine is released from non-synaptic cell-some membrane: histochemical evidence in bullfrog sympathetic ganglion cells. Brain Res. 205: 436-440, 1981.[Medline]
• SWARTZ, K. J. Modulation of Ca²⁺ channels by protein kinase C in rat central and peripheral neurons: disruption of G protein-mediated inhibition. Neuron 11: 305-320, 1993.[Medline]
• WATERMAN, S. A. Multiple subtypes of voltage-gated calcium channel mediate transmitter release from parasympathetic neurons in the mouse bladder. J. Neurosci. 16: 4155-4161, 1996.[Abstract/Free Full Text]
• WISDEN, W., PARKER, E. M., MAHLE, C. D., GRISEL, D. A., NOWAK, H. P., YOCCA, F. D., FELDER, C. C., SEEBURG, P. H., VOIGT, M. M. Cloning and characterization of the rat 5HT_5B receptor: evidence that the 5-HT receptor couples to a G protein in mammalian cell membranes. FEBS Lett. 333: 25-31, 1993.[Medline]
• YAKSH, T. L., WILSON, P. R. Spinal serotonin terminal system mediates antinociception. J. Pharmacol. Exp. Ther. 208: 446-453, 1979.[Free Full Text]

0022-3077/97 $5.00 Copyright ©1997 The American Physiological Society