The Journal of Neurophysiology Vol. 77 No. 6 June 1997, pp. 2910-2924
Copyright ©1997 by the American Physiological Society

Multiple Effects of Serotonin on Membrane Properties of Trigeminal Motoneurons In Vitro

C. F. Hsiao, P. R. Trueblood, M. S. Levine, and S. H. Chandler

Department of Physiological Science and the Mental Retardation Research Center, University of California at Los Angeles, California 90095

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Hsiao, C. F., P. R. Trueblood, M. S. Levine, and S. H. Chandler. Multiple effects of serotonin on membrane properties of trigeminal motoneurons in vitro. J. Neurophysiol. 77: 2910-2924, 1997. Intracellular recordings from guinea pig trigeminal motoneurons (TMNs) in brain stem slices were used to determine the underlying ionic mechanisms responsible for our previously demonstrated enhancement of TMN excitability during jaw movements by serotonin (5-HT). 5-HT (0.5-100 µM) depolarized motoneurons and increased input resistance in the majority of neurons tested. Additionally, 5-HT reduced the amplitude of the postspike medium-duration afterhyperpolarization, decreased the current threshold for maintained spike discharge, and increased the maximum slope of the steady-state spike frequency-current relationship. Under voltage clamp, from holding potentials close to resting potential, 5-HT produced an inward current and a decrease in instantaneous slope conductance, suggesting a reduction in a resting K⁺ leak conductance (I_leak). The instantaneous current-voltage (I-V) relationship for the inward 5-HT current (I_5-HT) was linear throughout most of the voltage range tested. However, the steady-state I-V relationship showed some degree of inward rectification at potentials starting around 70 mV. The mean reversal potential for the instantaneous I_5-HT was 86.2 ± 4.5 (SE) mV (n = 9), a value slightly negative to the predicted potassium equilibrium potential of 82 mV in these neurons. In the presence of 2 mM Ba²⁺, 5-HT application did not produce a further reduction in input conductance, but did expose a Ba²⁺-insensitive residual inward current that was resistant to Cs⁺ application. The instantaneousI-V relationship during 5-HT application in the presence of Ba²⁺ was shifted downward and parallel to control, suggesting that Ba²⁺ and 5-HT block the same resting I_leak. The residual Ba²⁺- and Cs⁺-insensitive component of the total inward I_5-HT was voltage independent and was blocked when the extracellular Na⁺ was replaced by choline, suggesting that the predominant charge carrier for this residual current is Na⁺. 5-HT enhanced a hyperpolarization-activated cationic current, I_h. In the presence of Ba²⁺, the time course of I_5-HT resembled that of I_h and showed a similar voltage dependence that was blocked by extracellular Cs⁺ (1-3 mM). The effects of 5-HT on membrane potential, input resistance, and I_h were partially mimicked by 5-HT₂ agonists and suppressed by 5-HT₂ antagonists. It is concluded that 5-HT enhances TMN membrane excitability through modulation of multiple intrinsic membrane conductances. This provides for a mechanism(s) to fine tune the input-output discharge properties of these neurons, thus providing them with greater flexibility in output in response to time-varying synaptic inputs during various movements of the jaw.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Trigeminal motoneurons (TMNs) are involved in production of oral-motor behaviors such as mastication, swallowing, and suckling, among others. Compared with spinal motoneurons, much less is known about the factors that control the discharge patterns of TMNs during movement. A complete understanding of the factors that control TMN discharge will require, at a minimum, a knowledge of the properties and discharge characteristics of the inputs to these motoneurons, as well as a detailed characterization of the properties of the intrinsic membrane conductances TMNs possess and their potential for modulation. The electrical properties and geometry of TMNs have been determined in detail and those studies have provided us with significant information (Curtis and Appenteng 1993; Moore and Appenteng 1991, among others). With respect to the location and discharge characteristics of the inputs to the trigeminal motor nucleus, a number of studies has also provided valuable information (reviewed in Goldberg and Chandler 1990; Lund 1991; Nakamura and Katakura 1995). However, fewer studies have actually identified specific intrinsic membrane conductances TMNs possess, and their potential for neurochemical modulation (Chandler and Trueblood 1995; Chandler et al. 1994; Hsaio and Chandler 1995; Kim and Chandler 1995). Information of this type is necessary to generate any comprehensive models of TMN discharge during various types of oral-motor behaviors.

There is evidence that serotonergic systems are important in control of TMN output. The trigeminal nucleus receives a dense serotonergic input (Kolta et al. 1993; Saha et al. 1991), most likely originating from raphe nuclei (Li et al. 1993), and contains serotonergic receptors (Kolta et al. 1993). Some serotonergic raphe cells have been shown to increase their discharge during oral-motor activity (Fornal et al. 1996; Veasey et al. 1995). Furthermore, in the guinea pig, iontophoretic application of serotonin (5-HT) or its agonists onto individual motoneurons exhibiting rhythmic burst discharge during cortically induced rhythmic jaw movements (RJMs) potently facilitates their discharge over many minutes (Katakura and Chandler 1990). Clearly, a facilitory role for serotonergic systems in control of TMNs has been established. However, the underlying ionic mechanism(s) of the facilitation has only recently been addressed (Chandler and Trueblood 1995; Trueblood et al. 1996).

Indirect evidence obtained in TMNs from brain stem slices suggests that a reduction in a resting leakage K⁺ conductance is the basis for 5-HT-induced increase in excitability (Chandler and Trueblood 1995), similar to that shown for facial motoneurons and spinal motoneurons (Aghajanian and Rasmussen 1989; Elliott and Wallis 1992; Larkman and Kelly 1992; Wang and Dun 1990; White and Fung 1989, among others). However, we have recently demonstrated that a prominent inward rectification mediated by a hyperpolarization-activated inward rectifier I_h is present in TMNs (Chandler et al. 1994). Therefore the possibility exists that under the appropriate conditions 5-HT could enhance this current and contribute to membrane depolarization, similar to results demonstrated recently in facial motoneurons (Larkman and Kelly 1992) and neonatal spinal motoneurons (Takahashi and Berger 1990).

In the present study we investigated, in vitro, the ionic basis for the previously demonstrated 5-HT-induced increase in excitability of TMNs at rest (Kurasawa et al. 1990) and during both reflex-induced (Ribeiro-do-Valle et al. 1991) and RJMs (Katakura and Chandler 1990). The results suggest that 5-HT increases membrane excitability through effects on a variety of intrinsic membrane conductances, such as 1) reduction of a resting barium-sensitive leak K⁺ current, 2) enhancement of I_h, 3) activation of a Ba²⁺- and Cs⁺-insensitive Na⁺ current, and 4) reduction of a calcium-dependent K⁺ current underlying the postspike afterhyperpolarization (AHP).

	METHODS

Abstract Introduction Methods Results Discussion References

Slice preparation

Transverse slices of brain stem containing the trigeminal motor nucleus were obtained from male albino guinea pigs (150-250 g) ~20 days old as described previously (Chandler et al. 1994). Briefly, the guinea pigs were anesthetized with ketamine HCl (130 mg/kg im) after an initial dose of chlorpromazine HCL (12.5 mg/kg im). The brain was removed and placed in cold modified artificial cerebrospinal fluid (M-ACSF) containing sucrose iso-osmotically substituted for Na⁺, and was then sectioned coronally through the trigeminal motor nucleus into 450-µm slices. Slices were then transferred into a holding chamber at 32-34°C in M-ACSF for 20 min, followed by 50% M-ACSF and 50% normal artificial cerebrospinal fluid (N-ACSF) for 20 min, and then placed in 100% N-ACSF for an additional 1 h of recovery before recording.

N-ACSF was composed of the following (in mM): 130 NaCl, 3.0 KCl, 2.4 CaCl₂, 1.3 MgSO₄, 20 NaHCO₃, 1.25 KH₂PO₄, and 10 D-glucose, bubbled with 95% O₂-5% CO₂ to maintain pH at 7.35-7.4. M-ACSF consisted of the following (in mM): 5.0 KCl, 0.2 CaCl₂, 4 MgSO₄, 20 NaHCO₃, 1.25 KH₂PO₄, 10 D-glucose, and 130 sucrose. The sucrose substitution method used was modified from Aghajanian and Rasmussen (1989) and was found essential for long-term survival of TMNs in slices (Chandler et al. 1994). Slices were then transferred individually to the stage of a modified Haas-type gas interface brain slice chamber (Haas et al. 1979). The recording chamber was superfused at a rate of 2 ml/min and kept at 32-34°C, with humidified 95% O₂-5% CO₂ flowing over it.

Electrophysiological recordings

Intracellular recordings were obtained with the use of an Axoclamp 2A (Axon Instruments, Burlingame, CA) with micropipettes fabricated from conventional thin-wall glass (1.5 mm OD, 15-20 M) pulled on a Brown-Flaming horizontal puller (Sutter Instruments, San Francisco, CA). Microelectrodes were filled with 2 or 3 M KCl, or 2% biocytin in 3 M KCl (for morphological identification). The tips of recording electrodes were coated with RTV silicone sealant (Dow Corning 734) to reduce capacitance and improve sample rate.

Neurons were impaled by applying small-capacitance buzz and advancing the electrode in 5-µm steps through the trigeminal motor nucleus. The recordings were performed in "bridge" mode (output bandwidth: 10 kHz), discontinuous current-clamp mode (3-kHz filter), or single-electrode voltage-clamp mode (1-kHz filter). The sampling rate ranged between 5 and 10 kHz. During the discontinuous recordings, the headstage voltage was continuously monitored to ensure complete settling of the voltage transients before the sampled voltage measurements (Finkel and Redman 1985). The voltage-clamp data presented should be interpreted with caution and treated as qualitative because the point clamp at the soma will not allow adequate space clamp in TMNs having extensive processes (Chandler et al. 1994).

Drugs

Tetrodotoxin (TTX, Sigma), 4-aminopyridine (4-AP, Sigma), serotonin creatine sulphate (5-HT, Sigma), 5-carboxamidtryptamine (5-CT, RBI), (±)1-(2,5-dimethyoxy-4-iodophenyl)-2-aminopropane HCl (DOI, RBI), N-[(8)-1,6-dimethylergolin-8-yl]N , N - d i m e t h y l s u l f a m i d e  h y d r o c h l o r i d e  (m e s u l e r g i n e ,  R B I),2-methyl-5-HT maleate (RBI), and ketanserin tartrate (RBI) were dissolved in distilled water from a 10 mM stock solution, whereas spiperone (RBI) was dissolved in ethyl alcohol. Tetraethylammonium chloride (TEA, Sigma) was substituted in equimolar amounts for NaCl. When Mn²⁺ was substituted for Ca²⁺, the H₂PO₄ was omitted and substituted with Cl to avoid precipitation. All solutions were added directly to the perfusate. The new classification scheme for 5-HT receptors approved by the International Union of Pharmacology (Hoyer et al. 1994) is used throughout the text.

To determine the effects of drug application on membrane properties during current-clamp experiments, the membrane potential was adjusted back to the original control value by extrinsic current application through the recording pipette. The estimated equilibrium potential for K⁺ (E_K⁺) was calculated from the Nernst equation to be 82 mV assuming internal K⁺ concentration of 94 mM (Jiang and Haddad 1991) and external K⁺ concentration of 4.25 mM recorded at 34°C. In those experiments in which the effects of 5-HT were examined on spike discharge, the region of maximum slope of the steady-state frequency-current (f-I) relationship was determined by eye and then, on the basis of at least five consecutive points, the slope was calculated by linear regression. In all instances, the maximal slope occurred within the first segment of the f-I relationship.

Data were stored on a multichannel video cassette recorder (digitized at 44 kHz, A. R. Vetter, Rebersburg, PA) and PC microcomputer with the use of pClamp (Axon Instruments), analyzed, and displayed with the use of Sigmaplot (Jandel Scientific) and Corel Draw (Corel, Ontario, Canada) software. Statistical comparisons were made with the use of Student's paired and unpaired t-test. Data are reported as means ± SE.

	RESULTS

Abstract Introduction Methods Results Discussion References

The effects of 5-HT were tested in a total of 136 TMNs from 126 animals. In those cells bathed in N-ACSF and recorded with electrodes filled with 3 M KCl (n = 46),resting membrane potentials were at least 55 mV(67.6 ± 0.90 mV, mean ± SE), input resistance (R_inp) [determined by measuring the voltage deflection to a <1-nA current pulse delivered at resting potential or from the slope of the steady-state voltage-current (V-I) plot generated by 150-ms incrementing current pulses around resting potential] ranged from 4.0 to 20.0 M (10.3 ± 0.56 M), and action potential amplitudes (measured from resting potential to peak) were 70 mV (94.9 ± 0.18 mV). These values were similar to those previously reported (Chandler et al. 1994). The lag time for 5-HT to get to the slice was ~3 min. The effects of 5-HT generally occurred within 5 min of the start of application and peaked at ~10 min.

Current-clamp experiments

SUBTHRESHOLD MEMBRANE PROPERTIES. Bath application of 0.5-100 µM 5-HT produced a significant membrane depolarization (6.4 ± 0.14 mV for all concentrations; P  0.01, mean range 1.4-16.7 mV, n = 43) and increase in R_inp (mean increase for all concentrations: 2.9 ± 0.34 M; P  0.01; mean range 0.59-6.32 M) in 43 of 46 cells examined. The half-maximal effective dose for these effects was estimated from the dose-response curve to be ~14 µM (not shown), with a peak response occurring at ~50 µM. To limit the changes in holding current and input conductance occurring after 5-HT during voltage clamp experiments, all subsequent experiments were performed with the use of 10 µM 5-HT. In a separate set of neurons tested (n = 43), bath application of 10 µM 5-HT produced in all neurons a significant membrane depolarization (Fig. 1A; 7.1 ± 0.05 mV from 67.6 ± 0.9 mV, P  0.001; range 3-15 mV) and a 37% increase in R_inp (11.3 ± 0.7 to 15.3 ± 1.0 M, P  0.01; range 0.8-11 M). A representative example of the effects of 5-HT on membrane potential and R_inp is shown in Fig. 1. The steady-state V-I relationship was determined from a series of constant current pulse injections before and during 5-HT application (Fig. 1C). These effects occurred in the presence and absence of TTX (n = 6), as well as in low-Ca²⁺/Mn²⁺ solutions containing a mixture of known K⁺ channel blockers (TEA and 4-AP, n = 7). In 2 of the original 46 neurons tested, 5-HT produced a hyperpolarization accompanied by an increase in R_inp before membrane depolarization, and in 1 neuron 5-HT produced a depolarization accompanied by a decrease in R_inp. Because these effects occurred in a small population of neurons, they were not investigated further.

View larger version (31K): [in this window] [in a new window]   FIG. 1. Effects of serotonin (5-HT) on membrane potential and membrane current. A and B: bath application of 10 µM 5-HT in current-clamp (A) and voltage-clamp (B) modes. Holding potential:72 mV (A), 68 mV (B). C: voltage-current(V-I) relationship obtained from family of constant current pulses. Intersection of curves before and during 5-HT indicates apparent reversal potential for 5-HT effects (E5-HT). Bars: duration of 5-HT application. Downward deflections are responses to either constant current (A) or voltage (B) pulses. A-C taken from different cells.

The observation that 5-HT depolarized the majority of TMNs and increased their R_inp suggests that this could be the result of a reduction in a leak K⁺ conductance. This is supported by the observation that the estimated mean reversal potential for the 5-HT effects (E_5-HT) determined from the intersection of the extrapolated lines of regression from the slope of the linear region of the V-I plots was 93.8 ± 0.41 mV (n = 17; Fig. 1C), which is close to the estimated E_K+ of 82 mV for the present experimental conditions (see METHODS). However, the fact that the estimated E_5-HT was more negative than E_K+ suggests the possibility that an additional cationic conductance with a reversal potential positive to resting potential was activated simultaneously in some motoneurons.

ACTION POTENTIAL AND REPETITIVE FIRING CHARACTERISTICS. In 13 neurons the effects of 5-HT were examined on single-spike and repetitive discharge characteristics. Previously, we showed that in addition to a fast TEA-sensitive AHP (fAHP), a slow-onset, apamin-sensitive AHP of medium duration (mAHP) is present in TMNs (Chandler et al. 1994). In many TMNs the fAHP and mAHP were separated by a depolarizing afterpotential (Fig. 2A). Bath application of 10 µM 5-HT had effects on all of these action potential characteristics. In general, the effects of 5-HT on action potential characteristics evoked by a short 2-ms pulse were a mean 51% reduction in the mAHP peak amplitude in 10 of 13 neurons (3.4 ± 0.4 to 1.7 ± 0.4 mV, n = 10, P < 0.003; Fig. 2B) and a 45% reduction in rheobasic current in 13 of 13 neurons tested (2.1 ± 0.3 to 1.2 ± 0.3 nA, n = 13, P  0.003). In contrast, there was no significant reduction in either spike height or spike half-amplitude duration, suggesting that the reduction was a result of a reduction in the calcium-dependent K⁺ current underlying the mAHP as opposed to a change in Ca²⁺ influx during the action potential.

View larger version (34K): [in this window] [in a new window]   FIG. 2. Effects of 5-HT on action potential properties and repetitive discharge characteristics. A: superimposed records of subthreshold passive membrane response and action potential response elicited by a 2-ms current pulse. B: action potentials elicited before and during application of 10 µM 5-HT. C and D: repetitive discharge evoked by 1-s constantcurrent pulse before (C) and during (D)5-HT application. E and F: frequency-current (f-I) relationship for 1st interspike interval (ISI) and steady-state discharge before and during 5-HT application. Resting potential before 5-HT was 70 mV and membrane potential was adjusted to 70 mV by current injection after 5-HT application. Dotted lines: linear region of maximum slope estimated by eye. All data taken from same cell. mAHP, medium-duration afterhyperpolarization.

The effects of 5-HT on repetitive discharge in response to a constant current pulse were examined in 13 neurons and consisted of a 62 ± 6.7% decrease in rheobasic current for maintained minimum discharge (Fig. 2, C and D; 1.5 ± 0.2 to 0.6 ± 0.1 nA, P  0.003). This was exhibited as a shift in the instantaneous and steady-state f-I relationships to the left (Fig. 2, E and F). Additionally, a 54% increase in the maximum slope (16.5 ± 1.7 to 25.6 ± 3.5 Hz/nA, P  0.002) of the steady-state relationship was observed after 5-HT application. This is shown in Fig. 2F.

Voltage-clamp experiments

In 9 of 9 cells examined, 10 µM 5-HT produced an inward current (mean 1.0 ± 0.07 nA) from holding potentials between 59 and 65 mV (62.0 mV; Fig. 1B). Figure 3, A and B, shows an example of a family of current responses evoked by voltage steps from a holding potential of 61 mV. As shown previously (Chandler et al. 1994), these cells exhibited a time-dependent inwardly rectifying I_h. In response to 5-HT a steady inward current was evident as a maintained inward shift in the holding current (Fig. 3, A and B). Examination of the linear instantaneous current-voltage (I-V) relationship shows that 5-HT reduced the instantaneous conductance by ~42.0% (149.9 ± 18 to 85.3 ± 7.9 nS, P  0.0004, n = 9). The reversal potential for this 5-HT current (I_5-HT) was obtained by determining the intersection of the linear regression lines for the instantaneous I-V relationship before and during the peak of the 5-HT effects (Fig. 3C). In nine cells examined, the reversal potential for I_5-HT was 86.2 ± 1.5 mV, which is close to the estimated E_K+ of 82 mV in TMNs.

View larger version (26K): [in this window] [in a new window]   FIG. 3. Effects of 5-HT on instantaneous and steady-state current-voltage (I-V) relationships. A and B: current traces in response to a family of voltage steps from a holding potential of 61 mV in control (A) and during 5-HT (B). Note shift in holding current (Ihold) to more negative values and the slow onset of a hyperpolarization-activated inward rectifier Ih in response to hyperpolarizing voltage steps. C: instantaneous I-V relationship before and during 5-HT. Note decrease in slope after 5-HT and the intersection of the regression lines, indicating E5-HT. D: steady-state I-V relationships. Note presence of inward rectification, and its enhancement after 5-HT. E: instantaneous and steady-state I-V relationship for 5-HT current (I5-HT) obtained by subtraction. For this and subsequent figures, membrane potential and current were measured within the 1st 5 ms when the capacity transient settled for instantaneous I-V plots, and at the end of the voltage step for steady-state I-V plots. Lines in C-E are best fits of linear regions around resting potential and were determined by eye. Erev, reversal potential.

In contrast to the instantaneous I-V relationship, the steady-state I-V relationship exhibited an inward rectification at potentials more negative than 70 mV (Fig. 3D). In all cells tested (n = 9), 5-HT application shifted the steady-state I-V relationship downward throughout the voltage range examined and enhanced the region of inward rectification. Figure 3E shows both the instantaneous and steady-state I_5-HT obtained by subtraction of the curves (5-HT minus control) before and during the peak of the 5-HT effects. It is evident that the instantaneous I_5-HT was linear throughout most of the voltage range examined, whereas the steady-state I_5-HT showed inward rectification at voltages more negative than 70 mV. These results suggest that the membrane depolarization produced by 5-HT occurs through a reduction of a leakage K⁺ conductance in conjunction with a simultaneous enhancement of an I_h conductance at membrane potentials negative to rest. To differentiate these effects, additional experiments were performed.

5-HT REDUCES A LEAKAGE K⁺ CONDUCTANCE. To determine whether the inward current and reduction in input conductance in response to 5-HT occurred, in part, from a reduction in a leakage K⁺ conductance, the following experiments were performed. Barium is known to block many types of K⁺ conductances including a resting leak K⁺ conductance (Jones 1989). We determined whether barium blocks a resting leak K⁺ conductance in TMN by examining the instantaneous I-V relationship before and during application of barium (100 µM-2 mM). Figure 4, A and B, shows a representative example of the effects of Ba²⁺ on the current responses evoked by a family of voltage steps from a holding potential of 61 mV. Generally, Ba²⁺ produced a steady inward shift in the holding current, and a reduction in the input conductance, as evidenced by the decrease in size of the instantaneous current jumps (Fig. 4B), and a reduction in the slope of the instantaneous I-V relationship (Fig. 4C). The mean reduction in instantaneous input conductance for all cells tested was 53% (144.6 ± 8.1 to 67.9 ± 8.4 nS, n = 4, P  0.005). As shown in Fig. 4C, the instantaneous I-V relationship for this neuron was well fit by a linear regression line. The barium equilibrium potential as determined by the intersection of the regression lines before and during Ba²⁺ application was 81 mV. For all neurons tested, the mean intersection of the curves occurred at 82.0 ± 2.0 mV, (n = 4), which again was equal to the estimated E_K+ of 82 mV in TMNs. Figure 4D shows the voltage dependence of the instantaneous inward current produced by Ba²⁺ (I_Ba2+) (obtained by subtraction), which is remarkably similar to the instantaneous I-V relationship for I_5-HT shown in Fig. 3E. These data indicate that TMNs possess a barium-sensitive leak K⁺ conductance.

View larger version (22K): [in this window] [in a new window]   FIG. 4. Barium reduces a resting leakage K+ current. A and B: currents in response to a family of voltage step commands before (A) and during (B) Ba2+ application. Note the increase in inward Ihold after Ba2+. C: instantaneous I-V relationship in control and during Ba2+. D: instantaneous I-V relationship for the inward current produced by Ba2+ (IBa2+) obtained by subtraction of curves in C. Dotted lines: barium equilibrium potential (EBa2+).

To determine whether 5-HT and Ba²⁺ block the same resting K⁺ conductance in TMNs, occlusion experiments were performed. Figure 5 shows an example of such an experiment. In normal Ringer solution, the effects of 5-HT on the instantaneous I-V relationship demonstrated the usual decrease in input conductance (Fig. 5A). However, when the same cell was then bathed in a solution containing 2 mM Ba²⁺, subsequent 5-HT application did not produce a further reduction in conductance (Fig. 5B). The mean decrease in conductance in response to 10 µM 5-HT in the presence of Ba²⁺ for all neurons tested was ~6% (38.8 ± 2.5 to 36.3 ± 2.6 nS, n = 3), compared with 34% (110.8 ± 9.7 to 72.9 ± 6.7 nS) in normal Ringer solution in response to 5-HT. Therefore these data show that Ba²⁺ and 5-HT block the same resting leakage K⁺ conductance.

View larger version (22K): [in this window] [in a new window]   FIG. 5. Barium and 5-HT reduce the same leakage K+ conductance. A and B: instantaneous I-V relationships in normal Ringer solution (A) and in the presence of Ba2+ (B) before and during 5-HT. C: instantaneous I-V relationship for totalI5-HT and Ba2+-insensitive component of I5-HT. D: instantaneous I-V relationship for Ba2+-sensitive component of total I5-HT obtained by subtraction of curves in C.

Interestingly, although Ba²⁺ blocked the change in conductance produced by 5-HT, it did not completely block the steady-state inward current produced by 5-HT. This residual inward current was manifest as a parallel shift downward of the I-V relationship (Fig. 5B, ), suggesting that the total I_5-HT obtained in normal Ringer solution is composed of barium-sensitive and -insensitive components. Figure 5C shows the Ba²⁺-insensitive I_5-HT component () and the total I_5-HT (). The Ba²⁺-insensitive I_5-HT was derived in the presence of Ba²⁺ by subtraction of the current before 5-HT application (Fig. 5B, ) from that obtained during the peak of the 5-HT effect (Fig. 5B, ). The barium-sensitive I_5-HT component was derived by subtraction of the Ba²⁺-insensitive I_5-HT (Fig. 5C, ) from the total I_5-HT (Fig. 5C, ) and is plotted in Fig. 5D. This current had similar properties to that shown for I_Ba2+; a linear voltage dependence and a reversal potential close to E_K+. For all three cells examined in this way, the mean barium-sensitive E_5-HT was found to be 81.7 ± 0.9 mV (n = 3), which is remarkably close to the estimated E_K+ of 82 mV.

BA²⁺- AND CS⁺-INSENSITIVE I_5-HT IS CARRIED BY SODIUM IONS. In the presence of Ba²⁺, 5-HT produced a residual inward current that was manifest as a parallel shift downward in the instantaneous I-V relationship, as mentioned above (Fig. 5B). This current was not associated with any appreciable change in input conductance, suggesting that the current might be carried by a cation with an equilibrium potential far from resting potential. To determine whether Na⁺ is responsible for this current, in the presence of Ba²⁺ and Cs⁺ (to block the barium-sensitive I_5-HT and I_h, respectively), the effects of 5-HT on the instantaneous I-V relationship were examined after substitution of choline chloride for NaCl in the extracellular medium. In five of six cells tested the Ba²⁺- and Cs⁺-insensitive I_5-HT was substantially reduced or blocked after substitution of choline for Na⁺ (Fig. 6). In the presence of choline the mean change in current after 5-HT application from a holding potential of 70 mV was 0.058 ± 0.06 nA (n = 6), compared with that observed in normal Ringer solution (0.226 ± 0.06 nA, n = 8). This difference was significant (P  0.01), indicating that Na⁺ is the main charge carrier for the Ba²⁺- and Cs⁺-insensitive I_5-HT.

View larger version (12K): [in this window] [in a new window]   FIG. 6. Na+ is the predominate charge carrier for the Ba2+-insensitive I5-HT. A: instantaneous I-V relationship in presence of 2 mM Ba2+ and 3 mM Cs+ before and during 5-HT. B: same as A, except choline was substituted for Na+. Holding potential: 60 mV.

5-HT ENHANCES I_h. To determine whether I_h potentially contributes to the 5-HT depolarization, the effects of 5-HT on I_h and inward rectifier conductance (G_h) activation were examined. Figure 7, A and B, shows the current responses to a family of hyperpolarizing voltage steps from a holding potential of 60 mV in the absence and presence of 5-HT. I_h was defined as the difference between the steady-state current at the end of the voltage step and instantaneous current, and a composite for all cells examined is plotted as a function of voltage in Fig. 7C. It is clear from the I-V relationship for this current that 5-HT enhanced this current. For all neurons tested, I_h was enhanced by ~75% after 5-HT (0.6 ± 0.1 to 1.0 ± 0.1 nA, n = 6, P  0.0004) when measured at 85 mV from holding potentials around rest. Because 5-HT enhances R_inp, it is possible that the enhanced I_h resulted from an improved space clamp after 5-HT application. Although this cannot be entirely excluded, 5-HT also increased I_h by 41% (n = 6, P < 0.01) when measured at 85 mV under Ba²⁺ conditions in which the 5-HT-induced increase in R_inp was eliminated, suggesting that 5-HT does enhance I_h to some extent.

View larger version (23K): [in this window] [in a new window]   FIG. 7. 5-HT enhances Ih. A and B: current traces in response to a family of voltage steps before (A) and during (B)5-HT application. Ab: tail current relaxations taken from data shown in Aa (dashed box). C: summary relationship between Ih and membrane potential before and during 5-HT for 6 cells. D: composite plot for 6 neurons showing relationship between normalized conductance for Ih, obtained from tail currents, and prepulse potential before and during 5-HT. Boltzmann fits superimposed over data points. E: composite plot of time constant vs. command potential before and during 5-HT application for 6 cells. Bars on points in C-E: SE. Vh, holding potential; , time constant of activation; Iin, instantaneous current; V1/2, potential for half maximal activation.

The effect of 5-HT on the voltage dependence of G_h activation was determined from the amplitude of the tail current relaxations to determine whether the enhancement by 5-HT resulted from a change in the gating properties of the channels or an increase in the maximal level of I_h. A representative example is shown in Fig. 7Ab. These experiments were performed in artificial cerebrospinal fluid that contained 10 mM TEA and 3 mM 4-AP to eliminate contamination of the tail currents by activation of additional outward currents (Hsaio and Chandler 1995). The activation curve was constructed by applying a family of voltage steps from a holding potential around 60 mV to various levels to activate G_h. The activation voltage steps were then followed by a test pulse to 75 mV to elicit tail current relaxations. The amplitude of these relaxations is indicative of the level of G_h activation for a given prepulse voltage step. The conductance was normalized according to the equation where R is the tail current amplitude for a given prepulse step voltage command, R_min is the minimum amplitude inward tail current in response to stepping from depolarizing potentials to the test pulse potential, and R_max is the maximum amplitude in response to stepping from hyperpolarizing potentials to the test pulse potential. The normalized values were plotted as a function of initial step potential and the subsequent curves were fitted with a Boltzmann function of the form where V is the prepulse command potential, V_1/2max the potential for half-maximal activation of G_h, and k is a slope constant that determines the steepness of activation. Examination of the composite activation curve for six cells (Fig. 7D, ) shows that G_h activates around 70 mV with a V_1/2max of 87.6 ± 0.94 mV. Application of 5-HT did not significantly shift the activation curve (Fig. 7D, ,V_1/2max = 86.4 ± 0.90 mV), suggesting that the enhancement of I_h by 5-HT was not a result of a shift in the activation properties of the conductance, but rather of an increase in the maximal level of I_h activation.

Although 5-HT did not significantly shift the activation curve for I_h, there was a reduction in the time constant of I_h activation at any given potential (Fig. 7E). In seven cells, the time constant of I_h activation was determined by fitting the time course of the current activation by a single-exponential function of the form where I_t is the current amplitude at time t, I_ss is the steady-state current at the end of the voltage step, I_h is the difference between the membrane current at the end of the voltage step and the instantaneous current after the capacitive transient has recovered, and  is the time constant for activation. A composite plot of  versus membrane potential for all cells examined shows that the time constant is voltage dependent over the range examined, decreasing at more negative clamp potentials (Fig. 7E, ). In the presence of 5-HT,  became more rapid at any given voltage (Fig. 7E, ). In general, 5-HT reduced  by 36% (n = 7, P < 0.01) in normal Ringer solution and by 18% (n = 6, P < 0.01) in the presence of extracellular Ba²⁺ when recorded at 85 mV, suggesting that part of the reduction in  resulted from 5-HT effects on the kinetics of I_h activation as opposed to improvement in the space clamp after 5-HT.

The component of I_5-HT that resembles I_h is best seen in the presence of Ba²⁺ to eliminate the simultaneous reduction in K⁺ leak conductance (I_leak) and changes in the space clamp conditions. Figure 8Aa shows the steady-state I-V relationship for the total membrane current produced by a family of hyperpolarizing voltage steps from a holding potential of 55 mV before and in the presence of 5-HT, whereas Fig. 8Ab shows the I_5-HT traces obtained by subtraction. The time course of I_5-HT resembles the time course of activation of I_h in response to hyperpolarizing voltage steps (compare Fig. 8Ab with Fig. 7A). Furthermore, the activation threshold of I_5-HT for this cell was approximately 75 mV (Fig. 8Ac), which is similar to the mean activation of 70 mV obtained for I_h (Fig. 7D). This current was completely blocked by cesium, as evidenced by the absence of any inward rectification in the steady-state I-V relationship for total membrane current (Fig. 8Ba) and I_5-HT (Fig. 8Bc), or of development of time-dependent inward current (Fig. 8Bb) after 5-HT application.

View larger version (23K): [in this window] [in a new window]   FIG. 8. In the presence of Ba2+, I5-HT resembles Ih. Aa: steady-state I-V plot of membrane current obtained before and during 5-HT in the presence of Ba2+. Note presence of inward rectification, and its enhancement after 5-HT. Ab: barium-insensitive I5-HT traces, obtained by subtraction, in response to a series of hyperpolarizing voltage steps. Note time-dependent development of inward current resembling Ih. Ac: steady-state I-V plot of I5-HT obtained in Ab. Note inward rectification starting around 75 mV is similar to Ih (Fig. 7C). Ba: same as Aa, except 3 mM Cs+ was added to bath. Note block of inward rectification. Bb: same as Ab, except in presence of Cs+. Bc: same as Ac, except in presence of Cs+. Note block of inward rectification. Holding potential: 65 mV. Residual inward Na+ current is shown.

Receptor pharmacology

The effects of various 5-HT agonists and antagonists were examined to determine the nature of the 5-HT receptor subtype(s) responsible for the 5-HT-induced membrane depolarization and change in R_inp in TMNs.

5-HT₂ RECEPTORS MEDIATE MEMBRANE DEPOLARIZATION, INCREASE IN R_inp, AND I_h. The 5-HT-induced depolarization and increase in R_inp were partially mimicked by the 5-HT₂ agonist DOI. DOI at 10 µM produced a mean depolarization of 4.8 mV from a mean resting potential of 60.6 ± 1.9 mV (P < 0.0001) and an 18% increase in R_inp (8.9 ± 0.8 to 10.5 ± 1.0 M, n = 11, P < 0.002), respectively. However, these values were less than that produced by 10 µM 5-HT, suggesting participation of other receptor subtypes. These effects were antagonized by the 5-HT₂ antagonist ketanserin (n = 1) and the more selective 5-HT_2C antagonist mesulergine (n = 4, data not shown). The 5-HT_1/2C agonist 5-CT (10 µM) also increased R_inp by ~17% from an initial 8.5 M (n = 10, P < 0.004), but, in contrast to DOI, produced a much smaller change in membrane depolarization (1.6 ± 0.51 mV, n = 10, P < 0.01). Mesulergine also significantly antagonized the 5-HT-induced changes in membrane depolarization (7.1 ± 0.43 mV, n = 43 vs.1.2 ± 0.50 mV, n = 5 in the presence of mesulergine, P  0.01) and R_inp (37.4 ± 3.14%, n = 43 vs. 3.6 ± 1.52% increase in the presence of mesulergine, n = 5, P  0.01), suggesting participation of 5-HT_2C receptors.

In contrast to DOI and 5-CT, other agonists including 5HT_1A [8-hydroxy-2-(n-dipropylamino)tetralin (8-OH-DPAT),n = 3], 5-HT_1B/1D (n-trifluoromethylphenyl piperazine hydrochloride; n = 5), and 5-HT₃ (2-methyl-5HT; n = 3) were without affect on membrane potential and R_inp. Furthermore, spiperone, a 5-HT_2A/1A antagonist, was without significant effect on the 5-HT-induced change in R_inp and membrane depolarization.

The 5-HT₂ agonists DOI (10 µM) and 5-CT (10 µM) enhanced I_h by 47% (0.5 ± 0.1 to 0.8 ± 0.1 nA, P  0.02, n = 5) and 64% (0.6 ± 0.1 to 0.9 ± 0.1 nA, n = 9, P < 0.002), respectively, when measured at 85 mV from holding potentials around rest. In the presence of mesulergine the enhancement of I_h by 5-HT was antagonized. For all neurons tested (n = 5), 5-HT enhanced I_h by only 6.5 ± 2.4% from a mean value of 0.9 nA in the presence of mesulergine, compared with 74.8 ± 5.2% from a mean value of 0.6 nA in its absence (P  0.002), suggesting an involvement of 5-HT_2C receptors. In contrast, in the presence of spiperone (n = 4), I_h was enhanced by 38.3 ± 7.3% by 5-HT, which was not statistically different from the values found for5-HT alone.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Our previous study and those of others showed that direct application of 5-HT to TMNs increased their burst discharge and overall excitability for prolonged periods at rest (Kurasawa et al. 1990), during reflex activation (Ribeiro-do-Valle et al. 1991), or during cortically induced RJMs in cats and guinea pigs (Katakura and Chandler 1990). The main findings of the present study now provide an ionic basis for the findings of those studies. In particular, the present study shows that 5-HT produces an increase in membrane excitability through a membrane depolarization of slow onset and prolonged time course accompanied by an increase in membrane resistance. The membrane depolarization is complex in nature and is mediated by some combination of at least three mechanisms: 1) a reduction in a resting Ba²⁺-sensitive, voltage-independent K⁺ conductance, 2) an increase in a Ba²⁺- and Cs⁺-insensitive, steady-state, TTX-resistant Na⁺ current, and 3) an enhancement of I_h. The demonstration that three different mechanisms potentially contribute to the 5-HT-induced membrane depolarization is a unique finding for mammalian motoneurons and is similar to what is observed in hypoglossal motoneurons in response to the peptide thyrotropin-releasing hormone (TRH) (Bayliss et al. 1992).

5-HT effects on resting K⁺ conductance

5-HT application depolarized TMNs and was accompanied by an increase in R_inp, suggesting that a reduction in a resting K⁺ conductance in part underlies the 5-HT depolarization. From holding potentials out of the range of I_h activation, the instantaneous I_5-HT obtained by subtraction demonstrated a linear I-V relationship throughout most of the voltage range examined and reversed at 86 mV, which was only slightly negative to the predicted E_K+ of 82 mV for TMNs (see METHODS). The slightly more negative reversal potential could suggest an action of 5-HT on a part of the neuron electrically remote from the recording site or the presence of additional currents. It is more likely that the total I_5-HT is actually composed of two components: a Ba²⁺-sensitive inward current that results from closure of leakage K⁺ channels and a Ba²⁺/Cs⁺-insensitive inward current with a more positive reversal potential. This is supported by the observation that the Ba²⁺-sensitive component of the instantaneous I_5-HT was linear throughout the voltage range tested, and reversed at 83 mV, close to the predicted E_K+. Furthermore, Ba²⁺ application is known to block leakage K⁺ channels (Jones 1989), and the Ba²⁺-sensitive component of I_5-HT had properties similar to those of I_Ba2+, which reversed at 82 mV, consistent with the predicted E_K+.

A 5-HT-induced membrane depolarization has been demonstrated in spinal (Elliott and Wallis 1992; Takahashi and Berger 1990; Wang and Dun 1990; White and Fung 1989; Ziskind-Conhaim et al. 1993), facial (Aghajanian and Rasmussen 1989; Larkman and Kelly 1992; Vandermaelen and Aghajanian 1982), and neonatal hypoglossal (Berger et al. 1992) motoneurons, among others, and is typically but not always (Berger et al. 1992; Takahashi and Berger 1990) associated with a decrease in membrane conductance. In some of those studies the precise K⁺ conductance affected was not determined. However, in other central neurons, such as nucleus accumbens, the 5-HT depolarization was previously shown to result from block of a fast barium-sensitive inward rectifier current active at rest (North and Uchimara 1989). This is unlikely for TMNs, because the instantaneous I-V relationship from resting potential to around 90 mV showed no evidence of inward rectification. Likewise, we have no evidence that a reduction in an M current, as shown in rat hippocampal neurons (Colino and Halliwell 1987), contributes to the 5-HT depolarization, because muscarine had no effect on membrane potential in TMNs (unpublished observations). Although 5-HT did reduce the mAHP following an action potential, it is also unlikely that a Ca²⁺-dependent K⁺ conductance active at rest contributes to I_5-HT, because the depolarization was not affected by altering the external Ca²⁺ concentration. On the basis of the properties of the barium-sensitive I_5-HT, it is likely that part of the 5-HT membrane depolarization results from a reduction in a resting, barium-sensitive leakage K⁺ conductance. It is interesting to note that this conductance appears to be a universal convergence point for modulation by various neuromessengers, because in spinal and hypoglossal motoneurons as well as other CNS neurons a leak K⁺ conductance with similar properties was suppressed by TRH (Bayliss et al. 1992; Fisher and Nistri 1993), substance P (Fisher and Nistri 1993), norepinephrine (Larkman and Kelly 1992; McCormick 1992), and 5-HT (Elliott and Wallis 1992; Larkman and Kelly 1992), among others. Determination of the intracellular signal transduction pathway(s) that mediates suppression of this conductance will provide insight into the precise mechanism for this convergence.

A Ba²⁺/Cs⁺-insensitive component of the total inward I_5-HT was demonstrated and would be expected to contribute to the membrane depolarization. This component of the total I_5-HT was readily observed when the effects of 5-HT on the resting K⁺ conductance were occluded by solutions containing Ba²⁺ or Ba²⁺ and Cs⁺ (to block any contribution from I_h; see below). Under these conditions the instantaneous I-V relationship after 5-HT was shifted downward and parallel to the I-V curve observed before 5-HT. The charge carrier for this barium-insensitive component is most likely Na⁺, because replacement of Na⁺ with choline completely blocked this component. Presently, we cannot rule out a Ca²⁺ dependence to this Na⁺ current, as demonstrated in hippocampal neurons in response to muscarinic agonists (Colino and Halliwell 1993), because we did not specifically determine the presence of this current in low-Ca²⁺ solutions or solutions that contained inorganic Ca²⁺ channel blockers. The inability to observe a change in the slope conductance and obtain a reversal potential for this component of I_5-HT could have resulted from activation of a current with a very small change in conductance and with a reversal potential far more depolarized from rest, or from activation of a conductance on a dendrite that was inadequately space clamped, a possibility that we cannot rule out presently. Unfortunately, our voltage-clamp conditions precluded command depolarizations to levels sufficient to determine the reversal potential of this current.

To our knowledge, a Ba²⁺- and TTX-resistant Na⁺ component to the total I_5-HT has not been demonstrated in adult mammalian motoneurons, although a current with similar properties was shown to occur in neonatal rat hypoglossal motoneurons in response to 5-HT (Berger et al. 1992). A similar current was recently demonstrated in rat hypoglossal motoneurons in response to norepinephrine (Parkis et al. 1995), and was suggested as a contributor to the TRH-induced depolarization observed in hypoglossal motoneurons (Bayliss et al. 1992). Moreover, currents with similar properties have also been found in other types of CNS neurons in response to different neuromessengers (Alreja and Aghajanian 1994; Colino and Halliwell 1993; Jiang et al. 1994; Shen and North 1992, among others). It remains to be determined whether this current contributes to the 5-HT depolarization observed in spinal (Elliott and Wallis 1992; Wang and Dun 1990; Ziskind-Conhaim et al. 1993) and facial (Aghajanian and Rasmussen 1989; Larkman and Kelly 1992) motoneurons.

5-HT effects on I_h

On the basis of the present data, I_h, previously characterized in TMNs (Chandler et al. 1994), would only contribute to the 5-HT depolarization in a small percentage of TMNs at rest. This most likely explains our difficulty, and that of others (Larkman et al. 1989), in obtaining a reversal potential for I_5-HT with the use of current-clamp methods. Furthermore, it is unlikely that the enhancement of I_h by 5-HT resulted entirely from an improvement in the space clampbecause, in the presence of Ba²⁺, which increased R_inp,5-HT still enhanced I_h, although to a lesser extent, in the absence of any further change in R_inp.

The evidence to suggest that I_h is enhanced by 5-HT and could contribute to the 5-HT depolarization comes from the following observations: 1) the steady-state I-V plot obtained in voltage clamp shows a time-dependent inward rectification that is enhanced after 5-HT application, 2) the voltage and time dependence of steady-state I_5-HT obtained by subtraction are similar to I_h, and 3) I_h and the inwardly rectifying component of the steady-state I_5-HT obtained in Ba²⁺ conditions are blocked completely by Cs⁺, a conventional blocker of I_h (Halliwell and Adams 1982). The voltage dependence of I_h suggests that in some motoneurons it is partially active at rest and could contribute modestly to the 5-HT depolarization. Although V_1/2max was around 86 mV, I_h activated around 70 mV, which is close to the mean resting potential of 67 mV in these motoneurons. In other neuron types (Bobker and Williams 1989; McCormick and Pape 1990; Nedergaard et al. 1991), including facial motoneurons (Garratt et al. 1993; Larkman and Kelly 1992), 5-HT was shown to enhance I_h. Interestingly, in neonatal spinal motoneurons (Takahashi and Berger 1990) and nucleus prepositus hypoglossi neurons (Bobker and Williams 1989), 5-HT depolarization was accompanied by an increase in membrane conductance and was associated with activation of an inward rectifier current with properties similar to I_h, suggesting developmental and neuron-specific effects of 5-HT on membrane conductances.

In a comparable study on facial motoneurons (Larkman and Kelly 1992), enhancement of I_h by 5-HT was associated with a mean shift to the right of 3.4 mV for V_1/2max (n = 3) in the absence of changes in the maximal amplitude of the current, and a decrease in time constant for activation at a given membrane potential, suggesting that monoamine effects occurred on the voltage dependence of channel gating. These effects were suggested as the basis for the monoamine-induced depolarization observed in that study. In the present study, although the kinetics of I_h were increased after 5-HT, the depolarization and enhancement of I_h were not associated with a statistically significant shift in the activation curve (V_1/2max shift of 1.2 mV, n = 6), suggesting that 5-HT increased the functional number of active channels at rest, as opposed to changes in channel gating mechanisms. This is supported by the observation that at potentials between 95 and 100 mV, where the I_h conductance would be expected to be nearly maximally activated, 5-HT still produced a substantial increase in I_h. Similar observations were obtained in hippocampal CA1 neurons in response to muscarinic receptor activation (Colino and Halliwell 1993) and in nucleus prepositus hypoglossi neurons in response to 5-HT (Bobker and Williams 1989). Although changes in voltage gating were not excluded, and likely did contribute to the transmitter effects, it was suggested in those studies that changes in the functional number of active channels by the neuromessengers could have also contributed to the transmitter-induced membrane depolarization. The fact that we found no shift in activation curve for I_h, which contrasts with the observation in facial motoneurons (Larkman and Kelly 1992), further emphasizes that caution must be exerted in generalizing effects of transmitters on different motoneuronal types and, therefore, the need to examine in seemingly similar motoneuronal types the effects of transmitters on particular intrinsic currents. To resolve some of these differences, single-channel analyses will be necessary.

Receptor subtypes mediating the effects of 5-HT

Although not extensively examined in the present study, the effects of 5-HT on I_h and resting membrane potential were most likely mediated by some combination of 5-HT₂receptor subtypes. This is consistent with the action of5-HT on resting potential and I_h recorded in spinal (Elliott and Wallis 1992; Wang and Dun 1990) and facial (Garratt et al. 1993; Larkman and Kelly 1992; Rasmussen and Aghajanian 1990) motoneurons. This is not to suggest that all effects of 5-HT on membrane conductances are mediated through 5-HT₂ receptor subtypes. Although not determined in the present study, the mAHP following an action potential in hypoglossal motoneurons was shown to be selectively reduced by 5-HT_1A receptor subtypes (Bayliss et al. 1995). In the present study, DOI (a 5-HT₂ agonist) (see Hoyer et al. 1994 for receptor classification) and 5-CT (a 5-HT_1/2C agonist) mimicked 5-HT effects on membrane potential and R_inp, although to a lesser extent, suggesting possible involvement of other receptor subtypes. In contrast, 8-OH-DPAT, a 5-HT_1A agonist, was without effect. Similarly, the antagonists ketansarin (5-HT₂) and mesulergine (5-HT_2C) blocked the 5-HT depolarization and increase in R_inp, whereas spiperone, a 5-HT_1A/2A antagonist, was ineffective. This is in contrast to what has been observed in neonatal rat spinal motoneurons recorded in vitro (Takahashi and Berger 1990), where the 5-HT depolarization was associated with a conductance increase and was blocked by spiperone and mimicked by 8-OH-DPAT. In that study the depolarization was solely attributed to activation of an inward rectifier current. Developmental regulation or species differences could account for these differences, because our data are from mature guinea pigs as opposed to 3- to 4-day-old rats.

Similarly, the 5-HT enhancement of I_h was mimicked byDOI and 5-CT and blocked by mesulergine, suggesting5-HT₂ receptor involvement consistent with that recently demonstrated in facial motoneurons (Garratt et al. 1993) but in contrast to that observed in spinal neonatal motoneurons (Takahashi and Berger 1990) or nucleus prepositus hypoglossi neurons (Bobker and Williams 1989) recorded in vitro. In those studies, I_h was blocked by the 5-HT_1A antagonist spiperone and mimicked by 8-OH-DPAT. Interestingly, we found some differences between 5-CT and DOI with respect to efficacy of production of membrane depolarization and enhancement of I_h: 5-CT was generally more effective in enhancing I_h and less effective in producing membrane depolarization compared with DOI, suggesting that different 5-HT receptor subtypes are associated with K⁺ channel closure and enhancement of I_h. A similar proposal has been made for the differential effects of 5-HT agonists on K⁺ channel closure and enhancement of I_h in facial motoneurons (Garratt et al. 1993; Larkman and Kelly 1992). More selective 5-HT agonists and antagonists will be needed to resolve this issue for TMNs.

Functional consequences

The present study demonstrates that 5-HT increases TMN excitability through modulation of multiple intrinsic membrane conductances. Therefore activation of serotonergic systems that project to the trigeminal motor nucleus (Li et al. 1993) would be expected to play a role in shaping the final discharge patterns of these motoneurons during various oral-motor behaviors (Fornal et al. 1996; Veasey et al. 1995). However, the precise role will vary according to which conductance channels 5-HT effects.

Although 5-HT depolarized the motoneurons, which will enhance neuronal excitability by bringing the motoneuron closer to spike threshold, the increase in R_inp, through closure of leakage K⁺ channels, and the modulation of I_h could be more significant with respect to controlling membrane excitability and shaping the final discharge pattern over time. For instance, increases in R_inp produced by 5-HT would shift the input-output relationship of the motoneuron to the left, thus necessitating less synaptic current to bring the membrane potential to spike threshold and, subsequently, produce higher-frequency spike trains.

An increase in R_inp by 5-HT will also increase the efficacy of ongoing synaptic activity by increasing the amplitude of both depolarizing and hyperpolarizing synaptic potentials. Recently it was shown that synaptic transmission between mesencephalic nucleus of V afferents and TMNs is facilitated by 5-HT application (Trueblood et al. 1996), and that was partly attributed to an increase in R_inp produced. Similarly, in the presence of serotonergic activation, facilitation of ongoing synaptic activity in TMNs would be expected to occur during RJMs, and might function as a form of contrast enhancement between antagonist motoneuronal groups. Because 5-HT receptors are uniformly distributed within the motor nucleus and 5-HT has similar effects on guinea pig opener and closer motoneurons (Kurasawa et al. 1990), it is likely that the increase in R_inp would serve to simultaneously enhance the excitatory and inhibitory synaptic activity occurring in antagonist motoneuronal groups during RJMs. Contrast enhancement was recently proposed for a function of TRH on hypoglossal motoneurons controlling antagonistic muscles of the tongue (Bayliss et al. 1992).

A reduction in the mAHP following an action potential could partly underlie the increased spike discharge observed in TMNs during iontophoretic application of 5-HT during mastication (Katakura and Chandler 1990). Because the mAHP is an important factor controlling TMN discharge (Chandler et al. 1994), 5-HT-induced alterations of the mAHP following an action potential would be expected to contribute to shaping motoneuron discharge patterns. In most neurons tested in the present study, 5-HT reduced the amplitude of the mAHP and increased the slope of the instantaneous and steady-state f-I curves. A similar increase in slope of the f-I relationship in response to 5-HT was observed in neonatal hypoglossal (Berger et al. 1992) and spinal (Hounsgaard and Kiehn 1989; Wallen et al. 1989; White and Fung 1989) motoneurons after reduction of the mAHP.

On the basis of its voltage dependence for activation, a role for I_h and its modulation by 5-HT would be expected to contribute under certain conditions to control of motoneuronal excitability during RJMs, but will vary according to the level of membrane potential during the jaw movement cycle. For I_h to modulate spike discharge during RJMs, the membrane potential would have to traverse the activation range of I_h. This does not occur during prolonged depolarizing bursts lasting hundreds of milliseconds within an RJM cycle. However, during the interburst periods in jaw closer motoneurons, when the membrane potential is hyperpolarized below resting potential (Chandler and Goldberg 1982; Goldberg et al. 1982; Gurahian et al. 1989), and presumablywithin the range of I_h activation, modulation of I_h by 5-HTwould be expected to contribute to shaping membrane potential trajectory. This would occur by limiting the degree of membrane hyperpolarization (thus counterbalancing the indirect effects of an increase in R_inp on inhibitory synaptic potentials) and increasing the rate of membrane repolarization to spike threshold, thus facilitating the onset of the subsequent burst discharge during RJMs. Therefore modulation of this conductance would serve as an additional mechanism for control of interburst period duration during the RJM cycle.

In conclusion, we have shown that 5-HT modulates multiple intrinsic membrane conductances in TMNs, resulting in an increase membrane excitability. The conditions under which each of these modulatory mechanisms are operative during oral-motor behaviors, and how they are used to fine tune motoneuronal discharge, await further development of reduced preparations (Katakura et al. 1995; Kogo et al. 1996) in which membrane conductances can be altered selectively and membrane potential trajectories and discharge characteristic during particular oral-motor behaviors, such as RJMs, can be examined.

	ACKNOWLEDGEMENTS

  We thank M. Castillo for technical assistance.

  This work was supported by National Institute of Dental Research Grant RO1 DE-06193 and a grant from the Physical Therapy Foundation.

	FOOTNOTES

  Address for reprint requests: S. H. Chandler, Dept. of Physiological Science/UCLA, 2851 Slichter Hall, Los Angeles, CA 90095.

  Received 26 July 1996; accepted in final form 4 February 1997.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• AGHAJANIAN, G. K., RASMUSSEN, K. Intracellular studies in the facial nucleus illustrating a simple new method for obtaining viable motoneurons in adult rat brain slices. Synapse. 3: 331-338, 1989.[Medline]
• ALREJA, M., AGHAJANIAN, G. K. QX-314 blocks the potassium but not the sodium-dependent component of the opiate response in locus coeruleus neurons. Brain Res. 639: 320-324, 1994.[Medline]
• BAYLISS, D. A., UMEMIYA, M., BERGER, A. J. Inhibition of N- and P-type calcium currents and the afterhyperpolarization in rat motoneurones by serotonin. J. Physiol. Lond. 485: 635-647, 1995.[Abstract/Free Full Text]
• BAYLISS, D. A., VIANA, F., AND BERGER, A. J. Mechanisms underlying excitatory effects of thyrotropin-releasing hormone on rat hypoglossal motoneurons in vitro. J. Neurophysiol. 1733-1745, 1992.
• BERGER, A. J., BAYLISS, D. A., VIANA, F. Modulation of neonatal rat hypoglossal motoneuron excitability by serotonin. Neurosci. Lett. 143: 164-168, 1992.[Medline]
• BOBKER, D. H., WILLIAMS, J. T. Serotonin augments the cationic current I_h in central neurons. Neuron 2: 1535-1540, 1989.[Medline]
• CHANDLER, S. H., GOLDBERG, L. J. Intracellular analysis of synaptic mechanisms controlling spontaneous and cortically induced rhythmical jaw movements in the guinea pig. J. Neurophysiol. 48: 126-138, 1982.[Free Full Text]
• CHANDLER, S. H., HSAIO, C. F., INOUE, T., GOLDBERG, L. J. Electrophysiological properties of guinea pig trigeminal motoneurons recorded in vitro. J. Neurophysiol. 71: 129-145, 1994.[Abstract/Free Full Text]
• CHANDLER, S. H., TRUEBLOOD, P. R. Pharmacological mechanisms controlling trigeminal motoneuronal excitability. In: Brain and Oral Function: Oral Motor Function and Dysfunction, edited by T. Morimoto, T. Matsuya, and K. Takada . Amsterdam: Elsevier, 1995, p. 99-105
• COLINO, A., HALLIWELL, J. V. Differential modulation of three separate K-conductances in hippocampal CA1 neurons by serotonin. Nature Lond. 328: 73-77, 1987.[Medline]
• COLINO, A., HALLIWELL, J. V. Carbachol potentiates Q current and activates a calcium-dependent non-specific conductance in rat hippocampus in vitro. Eur. J. Neurosci. 5: 1198-1209, 1993.[Medline]
• CURTIS, J. C., APPENTENG, K. The electrical geometry, electrical properties and synaptic connections onto rat V motoneurones in vitro. J. Physiol. Lond. 465: 85-119, 1993.[Abstract/Free Full Text]
• ELLIOTT, P., WALLIS, D. I. Serotonin and L-norepinephrine as mediators of altered excitability in neonatal rat motoneurons studied in vitro. Neuroscience 47: 533-544, 1992.[Medline]
• FINKEL, A. S., REDMAN, S. J. Optimal voltage clamping with a single microelectrode. In: Voltage and Patch Clamping With Microelectrodes, edited by T. G. Smith Jr., H. Lecar, S. J. Redman, and P. W. Gage . Bethesda, MD: Am. Physiol. Soc., 1985, p. 95-101
• FISHER, N. D., NISTRI, A. Substance P and TRH share a common effector pathway in rat spinal motoneurones: an in vitro electrophysiological investigation. Neurosci. Lett. 153: 115-119, 1993.[Medline]
• FORNAL, C. A., METZLER, C. W., MARROSU, F., RIBIERO-DO-VALLE, L. E., JACOBS, B. L. A subgroup of dorsal raphe serotonergic neurons in the cat is strongly activated during oral-buccal movements. Brain Res. 716: 123-133, 1996.[Medline]
• GARRATT, J. C., ALREJA, M., AGHAJANIAN, G. K. LSD has high efficacy relative to serotonin in enhancing the cationic current I_h: intracellular studies in rat facial motoneurons. Synapse 13: 123-134, 1993.[Medline]
• GOLDBERG, L. J., CHANDLER, S. H. Central mechanisms of rhythmic trigeminal activity. In: Neurophysiology of Jaws and Teeth, edited by and A. Taylor . London: Macmillan, 1990, p. 268-321
• GOLDBERG, L. J., CHANDLER, S. H., TAL, M. Relationship between jaw movements and trigeminal motoneuron membrane-potential fluctuations during cortically induced rhythmical jaw movements in the guinea pig. J. Neurophysiol. 48: 110-125, 1982.[Free Full Text]
• GURAHIAN, S. M., CHANDLER, S. H., GOLDBERG, L. J. Intracellular analysis of trigeminal motoneuron rhythmical activity during stimulation of the ponto-medullary reticular formation in the anesthetized guinea pig. J. Neurophysiol. 62: 1225-1236, 1989.[Abstract/Free Full Text]
• HAAS, H. L., SCHAERER, B., VOSMANSKY, M. A simple perfusion chamber for the study of nervous tissue slices in vitro. J. Neurosci. Methods 1: 323-325, 1979.[Medline]
• HALLIWELL, J. V., ADAMS, P. R. Voltage-clamp analysis of muscarinic excitation in hippocampal neurons. Brain Res. 250: 71-92, 1982.[Medline]
• HOUNSGAARD, J., KIEHN, O. Serotonin-induced bistability of turtle motoneurones caused by a nifedipine-sensitive calcium plateau potential. J. Physiol. Lond. 414: 265-282, 1989.[Abstract/Free Full Text]
• HOYER, D., CLARKE, D. E., FOZARD, J. R., HARTIG, P. R., MARTIN, G. R., MYLECHARANE, E. J., SAXENA, P. R., HUMPHREY, P. P. International Union of Pharmacology classification of receptors for 5-hydroxytryptamine (serotonin). Pharmacol. Rev. 46: 157-203, 1994.[Abstract]
• HSIAO, C. F., CHANDLER, S. H. Characteristics of a fast transient outward current in guinea pig trigeminal motoneurons. Brain Res. 695: 217-226, 1995.[Medline]
• JIANG, C., HADDAD, G. G. Effect of anoxia on intracellular and extracellular potassium activity in hypoglossal neurons in vitro. J. Neurophysiol. 66: 103-111, 1991.[Abstract/Free Full Text]
• JIANG, Z., PESSIA, M., NORTH, R. A. Neurotensin excitation of rat ventral tegmental neurones. J. Physiol. Lond. 474: 119-129, 1994.[Abstract/Free Full Text]
• JONES, S. W. On the resting potenial of isolated frog sympathetic neurons. Neuron 3: 153-161, 1989.[Medline]
• KATAKURA, N., CHANDLER, S. H. An iontophoretic analysis of the pharmacologic mechanisms responsible for trigeminal motoneuronal discharge during masticatory-like activity in the guinea pig. J. Neurophysiol. 63: 356-369, 1990.[Abstract/Free Full Text]
• KATAKURA, N., JIA, L., NAKAMURA, Y. NMDA-induced rhythmical activity in XII nerve of the isolated CNS from newborn rats. Neuroreport 6: 601-604, 1995.[Medline]
• KIM, Y. I., CHANDLER, S. H. NMDA-induced burst discharge in guinea pig trigeminal motoneurons In Vitro. J. Neurophysiol. 74: 334-346, 1995.[Abstract/Free Full Text]
• KOGO, M., FUNK, G., CHANDLER, S. H. Rhythmical oral-motor activity recorded in an in vitro brainstem preparation. J. Somatosens. Mot. Res. 13: 39-48, 1996.
• KOLTA, A., DUBUC, R., LUND, J. P. An immunocytochemical and autoradiographic investigation of the serotoninergic innervation of trigeminal mesencephalic and motor nuclei in the rabbit. Neuroscience 53: 1113-1126, 1993.[Medline]
• KURASAWA, I., TODA, K., NAKAMURA, Y. Non-reciprocal facilitation of trigeminal motoneurons innervating jaw-closing and jaw-opening muscles induced by iontophoretic application of serotonin in the guinea pig. Brain Res. 515: 126-134, 1990.[Medline]
• LARKMAN, P. M., KELLY, J. S. Ionic mechanisms mediating 5-hydroxytryptamine- and noradrenaline-evoked depolarization of adult rat facial motoneurones. J. Physiol. Lond. 456: 473-490, 1992.[Abstract/Free Full Text]
• LARKMAN, P. M., PENINGTON, N. J., KELLY, J. S. Electrophysiology of adult rat facial motoneurones: the effects of serotonin (5-HT) in a novel in vitro brainstem slice. J. Neurosci. Methods 28: 133-146, 1989.[Medline]
• LI, Y.-Q., TAKADA, M., MIZUNO, N. The sites of origin of serotoninergic afferent fibers in the trigeminal motor, facial, and hypoglossal nuclei in the rat. Neurosci. Res. 17: 307-313, 1993.[Medline]
• LUND, J. P. Mastication and its control by the brain stem. Crit. Rev. Oral Biol. Med. 2: 33-64, 1991.[Abstract/Free Full Text]
• MCCORMICK, D. A. Cellular mechanisms underlying cholinergic and noradrenergic modulation of neuronal firing mode in the cat and guinea pig dorsal lateral geniculate nucleus. J. Neurosci. 12: 278-289, 1992.[Abstract]
• MCCORMICK, D. A., PAPE, H. C. Noradrenergic and serotonergic modulation of a hyperpolarization-activated cation current in thalamic relay neurones. J. Physiol. Lond. 431: 319-342, 1990.[Abstract/Free Full Text]
• MOORE, J. A., APPENTENG, K. The morphology and electrical geometry of rat jaw-elevator motoneurones. J. Physiol. Lond. 440: 325-343, 1991.[Abstract/Free Full Text]
• NAKAMURA, Y., KATAKURA, N. Generation of masticatory rhythm in the brainstem. Neurosci. Res. 23: 1-19, 1995.[Medline]
• NEDERGAARD, S., FLATMAN, J. A., ENGBERG, I. Excitation of substantia nigra pars compacta neurones by 5-hydroxytryptamine in vitro. Neuroreport 2: 329-332, 1991.[Medline]
• NORTH, R. A., UCHIMURA, N. 5-hydroxytryptamine acts at 5-HT₂ receptors to decrease potassium conductance in rat nucleus accumbens neurones. J. Physiol. Lond. 417: 1-12, 1989.[Abstract/Free Full Text]
• PARKIS, M. A., BAYLISS, D. A., BERGER, A. J. Actions of norepinephrine on rat hypoglossal motoneurons. J. Neurophysiol. 74: 1911-1919, 1995.[Abstract/Free Full Text]
• RASMUSSEN, K., AGHAJANIAN, G. K. Serotonin excitation of facial motoneurons: receptor subtype characterization. Synapse 5: 324-332, 1990.[Medline]
• RIBEIRO-DO-VALLE, L. E., METZLER, C. W., JACOBS, B. L. Facilitation of masseter EMG and masseteric (jaw-closure) reflex by serotonin in behaving cats. Brain Res. 550: 197-204, 1991.[Medline]
• SAHA, S., APPENTENG, K., BATTEN, T. F. Light and electron microscopical localisation of 5-HT-immunoreactive boutons in the rat trigeminal motor nucleus. Brain Res. 559: 145-148, 1991.[Medline]
• SHEN, K. Z., NORTH, R. A. Muscarine increases cation conductance and decreases potassium conductance in rat locus coeruleus neurones. J. Physiol. Lond. 455: 471-485, 1992.[Abstract/Free Full Text]
• TAKAHASHI, T., BERGER, A. J. Direct excitation of rat spinal motoneurones by serotonin. J. Physiol. Lond. 423: 63-76, 1990.[Abstract/Free Full Text]
• TRUEBLOOD, P. R., LEVINE, M. S., CHANDLER, S. H. Dual-component excitatory amino acid mediated responses in trigeminal motoneurons and their modulation by serotonin in vitro. J. Neurophysiol. 76: 2461-2473, 1996.[Abstract/Free Full Text]
• VANDERMAELEN, C. P., AGHAJANIAN, G. K. Serotonin-induced depolarization of rat facial motoneurons in vivo: comparison with amino acid transmitters. Brain Res. 239: 139-152, 1982.[Medline]
• VEASEY, S. C., FORNAL, C. A., METZLER, C. W., JACOBS, B. L. Response of serotonergic caudal raphe neurons in relation to specific motor activities in freely moving cats. J. Neurosci. 15: 5346-5359, 1995.[Abstract]
• WALLEN, P., BUCHANAN, J. T., GRILLNER, S., HILL, R. H., CHRISTENSON, J., HOKFELT, T. Effects of 5-hydroxytryptamine on the afterhyperpolarization, spike frequency regulation, and oscillatory membrane properties in lamprey spinal cord neurons. J. Neurophysiol. 61: 759-768, 1989.[Abstract/Free Full Text]
• WANG, M. Y., DUN, N. J. 5-hydroxytryptamine responses in neonate rat motoneurones in vitro. J. Physiol. Lond. 430: 87-103, 1990.[Abstract/Free Full Text]
• WHITE, S. R., FUNG, S. J. Serotonin depolarizes cat spinal motoneurons in situ and decreases motoneuron afterhyperpolarizing potentials. Brain Res. 502: 205-213, 1989.[Medline]
• ZISKIND-CONHAIM, L., SEEBACH, B. S., GAO, B. X. Changes in serotonin-induced potentials during spinal cord development. J Neurophysiol. 69: 1338-1349, 1993.[Abstract/Free Full Text]

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