5-HT modulates multiple conductances in immature rat rostral ventrolateral medulla neurones in vitro

doi:10.1111/j.1469-7793.1999.0217z.x

Journal List > J Physiol > v.517(Pt 1); May 15, 1999

J Physiol. 1999 May 15; 517(Pt 1): 217–228.

doi: 10.1111/j.1469-7793.1999.0217z.x.

PMCID: PMC2269332

5-HT modulates multiple conductances in immature rat rostral ventrolateral medulla neurones in vitro

L L Hwang and N J Dun

Department of Anatomy & Neurobiology, Medical College of Ohio, 3000 Arlington Avenue, Toledo, OH 43699, USA

Corresponding author N. J. Dun: Department of Pharmacology, James H. Quillen College of Medicine, East Tennessee State University, PO Box 70577, Johnson City, TN 37614, USA. Email: dunnae/at/etsu.edu

Received November 26, 1998; Accepted January 29, 1999.

This article has been cited by other articles in PMC.

Abstract

Whole-cell patch-clamp recordings were made from rostral ventrolateral medulla (RVLM) neurones of brainstem slices from 8- to 12-day-old rats. In the presence of tetrodotoxin (0·5 μM), 5-HT (50 μM) elicited an outward current (I_5-HT,outward) (10/44 neurones) associated with an increase in membrane conductance, and an inward current (I_5-HT,inward) (29/44 neurones) accompanied by a decrease or no significant change in membrane conductance.
The steady-state I-V relationship of I_5-HT,outward showed an inward rectification; the 5-HT-induced current, which reversed at -87·9 ± 3·0 mV, was suppressed by 0·1 mM Ba²⁺.
Two types of steady-state I-V relationship for I_5-HT,inward were noted: type I I_5-HT,inward was characterized by a significant decrease in membrane conductance and reversed at a potential close to or negative to the theoretical K⁺ equilibrium potential (E_K), -94 mV, in 8/17 neurones; type II I_5-HT,inward was not associated with a significant change in membrane conductance and was relatively independent of membrane potential.
Both type I and type II I_5-HT,inward were significantly reduced in a low [Na⁺]_o solution. In this solution, I_5-HT,inward decreased with hyperpolarization and had a linear steady-state I-V relationship with a reversal potential of approximately -110 mV. The reversal potential of type I I_5-HT,inward shifted to about -80 mV as the [K⁺]_o was increased from 3·1 to 7·0 mM in low [Na⁺]_o solution. The type II I_5-HT,inward did not reverse at the estimated E_K in the same solution.
While not affected by externally applied Cs⁺ (1 mM), I_5-HT,inward was significantly smaller in RVLM neurones patched with Cs⁺-containing electrodes; the current reversed at -11·9 ± 6·4 mV in 8/15 responsive neurones.
It may be concluded that in rat RVLM neurones 5-HT increases an inwardly rectifying K⁺ conductance which may underlie the I_5-HT,outward and that a combination of varying degrees of K⁺ conductance decrease and a Cs⁺-insensitive, non-selective cation conductance increase may account for the two types of conductance change associated with I_5-HT,inward.

Rostral ventrolateral medulla (RVLM) neurones are a group of neurones situated in the ventral quadrant of the rostral medulla that are thought to play an important role in cardiovascular, respiratory and nociceptive functions (Dampney, 1994). RVLM neurones receive a moderately dense network of 5-HT-immunoreactive fibres arising from dorsal raphe nuclei (Steinbusch, 1981; Vertes & Kocsis, 1994). The importance of the central serotonergic system, especially the B3, B7 and B8 raphe nuclei, in cardiovascular function has been recognized (Chalmers et al. 1988). More importantly, microiontophoretic application of 5-HT or 5-HT receptor agonists to RVLM neurones increased or decreased the blood pressure, heart rate and sympathetic outflow that are mediated by 5-HT₂- and 5-HT_1A-like receptors (Gillis et al. 1989; Lovick, 1989a,b; Mandal et al. 1990; Nosjean & Guyenet, 1991; Helke et al. 1992). In addition, the selective 5-HT₂ antagonist ketanserin has been proposed as an innovative drug therapy for systemic hypertension and other cardiovascular disorders (Frishman et al. 1995).

5-HT reportedly interacts with seven subtypes of cell surface receptor (Hoyer et al. 1994). Operationally, except for 5-HT₃ receptors which are ligand-gated cation channels, all are G-protein coupled receptors. Activation of different types of 5-HT receptor results in a decrease and/or increase of membrane conductance to several ions (Bobker & Williams, 1990). Ion channels associated with 5-HT₁ and 5-HT₂ receptors appear to be diverse (Bobker & Williams, 1990). For example, activation of 5-HT_1A receptors opens an inwardly rectifying K⁺ conductance in several types of central neurone (Colino & Halliwell, 1987; Williams et al. 1988; Pan et al. 1993; Penington et al. 1993; Okuhara & Beck, 1994). On the other hand, stimulation of 5-HT_1A receptors may inhibit Ca²⁺ currents in a number of neurones including rat dorsal raphe (Penington & Kelly, 1990), hypoglossal (Bayliss et al. 1995) and dorsal root ganglion (Mar et al. 1994).

While activation of 5-HT₂ receptors is generally associated with a decrease of K⁺ conductance, the type(s) of K⁺ conductance appears to vary with different neurones. Thus, activation of 5-HT₂ receptors decreases a leak K⁺ current in facial motoneurones (Larkman & Kelly, 1992), an inwardly rectifying K⁺ conductance in rat nucleus accumbens neurones (North & Uchimura, 1989) and an outwardly rectifying K⁺ conductance in rat sympathetic preganglionic neurones (Pickering et al. 1994). On the other hand, 5-HT has been shown to depolarize several types of central neurone by augmenting a hyperpolarization-activated, Cs⁺-sensitive, non-selective cationic current, I_h (Bobker & Williams, 1989; Pape & McCormick, 1989; Takahashi & Berger, 1990; Larkman & Kelly, 1992).

Recently we have reported that 5-HT interacting with 5-HT₂- and 5-HT_1A-like receptors depolarizes and hyperpolarizes RVLM neurones of brainstem slices of immature rats (Hwang & Dun, 1998). The present study was undertaken to characterize further the ionic mechanisms underlying these two responses.

Some of the results have been published as an abstract (Hwang & Dun, 1997).

METHODS

Coronal brainstem slices were prepared from immature (8- to 12-day-old) Sprague-Dawley rats as previously described (Hwang & Dun, 1998; Lin et al. 1998). Rats were anaesthetized with ether and decapitated immediately. The brainstem was removed and 500 μm coronal sections were prepared with the use of a vibratome. The first two slices rostral to the area postrema were obtained and incubated in oxygenated Krebs solution at room temperature (21 ± 1°C) for at least 1 h prior to the start of experiments. The Krebs solution had the following composition (mM): 127 NaCl, 1.9 KCl, 1.2 KH₂PO₄, 2.4 CaCl₂, 1.3 MgCl₂, 26 NaHCO₃ and 10 glucose, and was saturated with 95 % O₂-5 % CO₂. One slice was transferred to the recording chamber, held in place between two grids of fine nylon mesh and superfused with oxygenated Krebs solution at a rate of 3-6 ml min⁻¹. All experiments were carried out at room temperature. Low Na⁺ (26 mM)-Krebs solution was prepared by replacing NaCl with Tris buffer titrated to pH 7.4 with HCl. In preparing Ca²⁺-free-high Mg²⁺ (10.9 mM) solution, CaCl₂ was removed and MgCl₂ was increased accordingly.

The RVLM was recognized in the slice as an area ventral to the nucleus ambiguus, which appears as a slightly dark area in a freshly prepared medullary slice, and lateral to the paragigantocellular nucleus at the level rostral to the area postrema (Hwang & Dun, 1998; Lin et al. 1998). Patch electrodes filled with a solution containing (mM): 130 potassium gluconate, 1 MgCl₂, 2 CaCl₂, 4 ATP, 0.3 GTP, 10 EGTA and 10 Hepes had a resistance of 2-5 MΩ; the pH of the solution was adjusted to 7.2 with NaOH. In the series of experiments where potassium ions were replaced by caesium ions in the patch-electrode solution, caesium methanesulfonate of equal molarity was substituted for potassium gluconate.

Whole-cell recordings from neurones in the brainstem slices were similar to that described by Blanton et al. (1989). The resistance of the pipette was continuously monitored by observing the size of the current in response to a -3 mV test pulse across the pipette. With a positive pressure applied to the patch pipette, the pipette tip was gently lowered into the solution and then into the area of interest in the slice until the size of the current response decreased to about one-third of the original size. Gentle suction was then applied to the pipette and a seal began to form. Upon forming a gigaohm seal, the current response almost disappeared, leaving only small pipette capacitance transients in response to each test pulse, which were then zeroed out using the pipette capacitance compensation control on the amplifier. After setting the holding potential at -60 mV, short pulses of suction were applied to break the patch of membrane inside the pipette tip and the whole-cell configuration was then established. Signals were recorded using an Axopatch-1C (Axon Instruments) in voltage-clamp mode, low-pass filtered at 2 kHz and acquired using a personal computer and pCLAMP software (version 5.7.1 or 7.0; Axon Instruments) for later analysis. The output was monitored on a Gould digital storage oscilloscope 1620 and recorded with a two-channel Gould chart recorder RS3200. Membrane potentials reported in the text have been corrected for the liquid junction potential. The access resistance was less than 25 MΩ and was routinely compensated by 50-70 %. Only those cells with action potentials in excess of 50 mV were included in the study. The steady-state current-voltage (I-V) relationship was obtained by a series of 2 s clamp steps every 10 s from the resting or holding potential to different potentials before and during the application of 5-HT in tetrodotoxin (TTX, 0.5 μM)-containing Krebs solution. Currents elicited by such voltage commands in control media were subtracted from their counterparts in the presence of 5-HT to yield steady-state I-V curves of 5-HT-sensitive currents. Experimental protocols were controlled by a personal computer with pCLAMP software.

5-HT and other compounds were dissolved in Krebs solution and applied by superfusion. The following compounds were used: TTX, pindobind-5-HT_1A (PBD) and ketanserin from Research Biochemicals Inc.; 5-HT creatinine sulfate and other chemicals from Sigma. Data are expressed as means ±s.e.m. and were analysed statistically using Student's t test, Student's paired t test or one-way ANOVA with a significance level of P < 0.05.

RESULTS

RVLM neurones had a mean resting potential, spike height and input resistance of -61.7 ± 1.6 mV, 73.0 ± 3.3 mV and 577.8 ± 84.5 MΩ (n = 20), which are comparable to the values reported earlier (Hwang & Dun, 1998).

General features of 5-HT-induced inward and outward currents

In the presence of TTX (0.5 μM), the direct effects of 5-HT on 44 RVLM neurones, held at the original resting membrane potential, were examined under whole-cell voltage-clamp conditions. Similar to the results obtained under current-clamp conditions (Hwang & Dun, 1998), 5-HT (50 μM) caused an outward (10/44 neurones, 23 %) or inward (29/44 neurones, 66 %) shift of holding currents in RVLM neurones (Fig. 1), and no detectable change in holding currents in the remaining five neurones (11 %). The 5-HT-induced outward current (I_5-HT,outward) had a mean peak amplitude of 28.0 ± 7.3 pA in 10 of the 44 neurones. The I_5-HT,outward was associated with an increase in membrane conductance (51.0 ± 15.7 %) in 7 of the 10 neurones (Fig. 1A). Of the 29 neurones in which 5-HT caused an inward current (I_5-HT,inward), the mean peak amplitude of I_5-HT,inward was 52.9 ± 9.4 pA. The I_5-HT,inward was accompanied by a membrane conductance decrease (29.5 ± 4.3 %) in 14 of the 29 neurones (Fig. 1B), and was not associated with a detectable change in membrane conductance in the remaining 15 neurones (Fig. 1C).

Figure 1

Outward and inward currents caused by 5-HT in RVLM neurones

In addition to a monophasic depolarization or hyperpolarization, 5-HT caused a biphasic response in a small population of RVLM neurones (< 10 %), which was due to the activation of both 5-HT_1A- and 5-HT₂-like receptors in a single neurone (Hwang & Dun, 1998). Here, 5-HT by superfusion caused a biphasic current in three neurones. To eliminate possible contamination by the activation of undesired 5-HT receptors, the 5-HT₂ receptor antagonist ketanserin or the 5-HT_1A receptor antagonist PBD was included in the perfusing solution so that the conductances underlying I_5-HT,outward or I_5-HT,inward could be pharmacologically isolated in the following experiments.

Steady-state I-V relationship of I_{5-{kern 0 0}HT,outward}

In the presence of ketanserin (2 μM) and TTX, the steady-state I-V relationship of I_5-HT,outward showed an inward rectification. I_5-HT,outward reversed at a potential of -87.9 ± 3.0 mV (n = 8) (Fig. 2A). The estimated K⁺ equilibrium potential (E_K) is -94 mV at 20°C. The 5-HT-induced current was markedly suppressed by extracellular Ba²⁺ (0.1 mM, n = 3; Fig. 2B). The amplitude, steady-state I-V relationship and reversal potential (control, -86.7 ± 0.3 mV vs. low Na⁺, -88.7 ± 1.3 mV; n = 3) of I_5-HT,outward was not significantly affected by lowering [Na⁺]_o from 153 to 26 mM.

Figure 2

Steady-state I-V relationship of I_5-HT,outward and sensitivity to extracellular Ba²⁺

Steady-state I-V relationship of I_5-HT,inward

In the presence of TTX and PBD (2 μM), 5-HT induced an inward current (I_5-HT,inward) with a peak amplitude of 50.7 ± 7.1 pA in 35 of 54 neurones, which were clamped at the original resting membrane potential; a detectable response was not noted in the remaining 19 neurones. Two types of steady-state I-V relationship for I_5-HT,inward were observed, type I and type II, which differed with respect to their slope conductance. The type I or type II response pattern was based on the shape of the I-V curve. The slope conductance was measured in the potential range between the holding potential (i.e. original resting membrane potential) and 10 mV negative to the holding potential. The slope conductance, which was obtained before and during the application of 5-HT in the same cell, was compared using Student's paired t test.

Type I I_5-HT,inward

5-HT (50 μM) caused an inward shift of the holding current (peak amplitude of 43.5 ± 11.9 pA at the resting potential of -59.5 ± 1.8 mV) accompanied by a significant decrease in membrane conductance (slope conductance, 4.0 ± 0.7 nS in control vs. 2.8 ± 0.5 nS in 5-HT) in 17 of the 35 neurones in which 5-HT evoked an inward current. The decrease in membrane conductance is shown as a reduction in the slope of the steady-state I-V curve obtained in the presence of 5-HT as compared with that obtained in the control (Fig. 3A and B). Subtracting the steady-state I-V curve obtained in the control media from that obtained in the presence of 5-HT yields the steady-state I-V relationship of type I I_5-HT,inward (shown as ‘difference’I-V curve in Fig. 3), which declined with membrane hyperpolarization. The type I I_5-HT,inward reversed polarity in 8/17 neurones and showed a nearly linear steady-state I-V relationship over the entire range of potentials tested (Fig. 3A). The reversal potential of -92.6 ± 1.5 mV was close to the estimated E_K (-94 mV), and the peak amplitude of 14.0 ± 4.2 pA measured at the resting potential was relatively small in five of these eight neurones. In the other three neurones, the reversal potential was more negative, i.e. -111.0 ± 3.5 mV, and the peak amplitude of 70.7 ± 17.3 pA was larger than that obtained in the previous group of RVLM neurones. In the remaining nine neurones, the I_5-HT,inward was not reversed at a potential as negative as -130 mV (Fig. 3B).

Figure 3

Steady-state I-V relationship of type I (A and B) and II (C and D) I_5-HT,inward in RVLM neurones

The mean resting membrane potential, amplitude of the I_5-HT,inward and membrane conductance for each type of response are shown in Table 1.

Table 1

Membrane properties and peak amplitudes of I_5-HT,inward in RVLM neurons

Type II I_5-HT,inward

In 18 of 35 neurones responding to 5-HT, the I_5-HT,inward (peak amplitude of 57.5 ± 8.2 pA at the resting potential of -59.0 ± 1.3 mV) was not associated with a significant change in membrane conductance (slope conductance, 4.4 ± 0.6 vs. 4.3 ± 0.5 nS). In this group of neurones, the steady-state I-V curves obtained in the presence of 5-HT were parallel to the control I-V curves over nearly the entire potential range tested, and the I_5-HT,inward was relatively independent of membrane potential. Two representative experiments are shown in Fig. 3C and D.

The amplitude of type I I_5-HT,inward, which had a reversal potential (-92.6 ± 1.5 mV) close to the estimated E_K, was usually small (14.0 ± 4.2 pA) as mentioned above. For this reason, neurones with this type of response were not examined further in the following experiments. In addition, due to the infrequent occurrence (3 in 54 neurones), we were not able to evaluate further the larger amplitude (70.7 ± 17.3 pA) type I I_5-HT,inward, which reversed at the more negative potential (-111.0 ± 3.5 mV). Therefore, only non-reversed type I and type II I_5-HT,inward were investigated in the following experiments.

Steady-state I-V relationship of I_5-HT,inward in low Na⁺ solutions

To assess the contribution of Na⁺ conductance, the amplitude and steady-state I-V relationships of I_5-HT,inward were examined in neurones bathed in a low Na⁺ (26 mM) solution. Neurones were clamped at the original resting potentials. The amplitudes of both non-reversed type I and type II I_5-HT,inward were significantly reduced by the low Na⁺ solution, as shown in Fig. 4A and B. The mean amplitudes of type I and type II I_5-HT,inward were reduced to 43.4 ± 10.4 % (n = 4) and 23.0 ± 4.9 % (n = 4) of normalized control values (Fig. 4B). In the low Na⁺ solution, the I_5-HT,inward of both type I and type II decreased with hyperpolarization and had a nearly linear steady-state I-V relationship, with a reversal potential of -110.0 ± 2.0 and -111.0 ± 5.6 mV, respectively, as shown in Fig. 4C and D.

Figure 4

Sodium and potassium dependence of I_5-HT,inward in RVLM neurones

To determine whether the residual I_5-HT,inward obtained in low Na⁺ solution was mainly contributed by a decrease of K⁺ conductance, the [K⁺]_o was increased from 3.1 to 7 mM in the low Na⁺ solution. Neurones were first exposed to 5-HT in normal Krebs solution containing 153 mM Na⁺ and 3.1 mM K⁺. After washout of 5-HT, the neurones were superfused with modified Krebs solution containing 26 mM Na⁺ and 7 mM K⁺ and then exposed to 5-HT again. Under these conditions, the amplitudes of both non-reversed type I and type II I_5-HT,inward were significantly reduced (Fig. 4E and F). The steady-state I-V curve of non-reversed type I I_5-HT,inward reversed polarity at -82.0 ± 4.6 mV (n = 4) in the modified Krebs solution (Fig. 4E). The estimated E_K under these experimental conditions is -74 mV. In contrast, the steady-state I-V relationship of type II I_5-HT,inward showed small conductance changes over the entire voltage range in modified Krebs solution in four of the five neurones examined (Fig. 4F); 5-HT caused an inward current with conductance increase in modified Krebs solution in one neurone (not shown). For all these five neurones, the steady-state I-V curve did not reverse polarity close to the estimated E_K.

Effects of Ba²⁺ on I_5-HT,inward

In order to determine whether the K⁺ conductance affected by 5-HT was sensitive to Ba²⁺, the effect of Ba²⁺ (1 mM) on non-reversed type I I_5-HT,inward was examined. Ba²⁺ slightly decreased the slope of the steady-state I-V curve (n = 6, Fig. 5). In this series of experiments, neurones were held at -60 mV.

Figure 5

Effect of Ba²⁺ on non-reversed type I I_5-HT,inward

Effects of Cs⁺ on I_5-HT,inward

In this series of experiments, the slices were superfused with a modified Krebs solution that contained 0 mM Ca²⁺-10.9 mM Mg²⁺ and 0.5 μM TTX to block Ca²⁺ and voltage-dependent Na⁺ channels. In addition, potassium gluconate in the patch electrode solution was replaced by caesium methanesulfonate to block K⁺ channels. Under these conditions, an inward shift in holding currents occurred immediately after the establishment of whole-cell recordings and reached a steady state in 20 min (not shown). The 5-HT responses and steady-state I-V relationships were determined 30 min after the establishment of whole-cell recording conditions at the holding potential of -60 mV. 5-HT caused an inward current in 15/23 neurones (65 %) with a mean peak amplitude of 26.5 ± 5.9 pA at the holding potential. While the percentage of responsive neurones was the same as that obtained in control solution (35/54; 65 %) without caesium, the peak amplitude was significantly smaller than that (50.7 ± 7.1 pA) obtained in control solution, which was recorded at the resting membrane potential of -59.3 ± 1.1 mV (Fig. 6A). In 8 of the 15 5-HT-responsive neurones recorded with the Cs⁺-filled patch electrode, 5-HT produced an inward current that had a reversal potential of -11.9 ± 6.4 mV, while four neurones had a I_5-HT,inward that was not reversed. No information was obtained from the remaining three neurones. Two types of steady-state I-V relationship were observed in the eight neurones in which the I_5-HT,inward was reversed (Fig. 6B and C). The I_5-HT,inward in the RVLM neurone shown in Fig. 6B declined with depolarizing membrane potentials over almost the entire potential range examined, whereas the I_5-HT,inward evoked in the neurone shown in Fig. 6C decreased in the range between -50 and 0 mV.

Figure 6

I-V relationship of I_5-HT,inward recorded with a Cs⁺-containing pipette in RVLM neurones

A representative steady-state I-V curve of one of the four neurones without a reversal is shown in Fig. 6D, which is similar to that of type II I_5-HT,inward.

Involvement of a cation current, I_h

A hyperpolarization-induced inward current relaxation with characteristics similar to the cation current I_h, which has been reported to be enhanced by 5-HT (Bobker & Williams, 1989; Pape & McCormick, 1989; Takahashi & Berger, 1990; Larkman & Kelly, 1992), was noted in about half of RVLM neurones sampled here. The amplitude and rate of development of inward current relaxation were increased at more negative voltages (Fig. 7A and B). The inward current relaxation was well fitted over the entire voltage range tested by a single-exponential function:

A mathematical equation, expression, or formula that is to be displayed as a block (callout) within the narrative flow. The name of referred object is tjp0517-0217-mu1.jpg

where I_t is the amplitude of the current at time t, A and B are constants and τ is the time constant. The amplitude of inward current relaxation, which represents the amplitude of I_h, was calculated by subtracting the instantaneous current (I_i) from the steady-state current (I_ss) measured at the end of the command (I_ss - I_i). Due to the presence of transient currents at the onset of the voltage steps, the I_i at time zero was calculated by the fitted equation as mentioned above. To determine whether I_h was involved, we examined the effect of 5-HT on the amplitude of I_h and the effect of Cs⁺ on I_5-HT,inward in six neurones in which type II I_5-HT,inward was produced in response to 5-HT. In this series of experiments, neurones were held at -60 mV and a series of hyperpolarization steps (from -70 to -120 mV) was applied to obtain the I-V relationships. Neurones were first perfused with 5-HT-containing solution in the absence of Cs⁺ (1 mM) and then 5-HT was re-applied to the same neurones in the presence of Cs⁺. The amplitude of I_h was not increased during the course of type II I_5-HT,inward in any of the six neurones tested (Fig. 7A (left) and B). In addition, I_h has been shown to be sensitive to extracellular Cs⁺ in a number of excitable membranes (Hagiwara et al. 1976; DiFrancesco, 1981; Mayer & Westbrook, 1983; Xi-Moy & Dun, 1995). Here, in all six neurones exhibiting I_h, Cs⁺ (1 mM) suppressed the inward current relaxation (Fig. 7A (right) and C), without significantly affecting the amplitude of I_5-HT,inward in the potential range (-60 to -120 mV) over which I_h could be activated (Fig. 7D).

Figure 7

Effect of 5-HT on I_h in RVLM neurones

In the present study, 81 neurones were found to exhibit I_h. When comparing the amplitude of I_h in control and in the presence of 5-HT in all 81 neurones, which included those neurones not treated with PBD, 5-HT did not increase the amplitude of I_h in 77 of 81 neurones. 5-HT increased I_h amplitude in only four neurones, of which three displayed I_5-HT,inward (Fig. 7E, inset traces). The effect of Cs⁺ was tested on one neurone in which 5-HT enhanced I_h during the course of type II I_5-HT,inward, and it was found that Cs⁺ reduced the I_5-HT,inward at hyperpolarizing potentials (-70 to -120 mV) as shown in Fig. 7E.

DISCUSSION

In our earlier study, activation of 5-HT₂ and 5HT_1A receptors was found to cause a depolarization and hyperpolarization in RVLM neurones (Hwang & Dun, 1998). The purpose of this study was to characterize further the ionic mechanism underlying these responses. To minimize the potential interference by subtypes of 5-HT receptors coexisting in a single RVLM neurone, experiments were conducted in the presence of the 5-HT_1A receptor antagonist PBD or the 5-HT₂ receptor antagonist ketanserin to isolate the specific response. Under these conditions, 5-HT evoked an inward or outward shift of the holding current in RVLM neurones.

Ionic basis of I_5-HT,outward

Hyperpolarizations in response to 5-HT were reduced by a low concentration (0.1 mM) of Ba²⁺, suggesting the possible involvement of an inwardly rectifying K⁺ conductance in mediating the responses (Hwang & Dun, 1998). In the present study, the steady-state I-V relationship of I_5-HT,outward displayed an inward rectification, which was suppressed by a low concentration of Ba²⁺. Collectively, these results support the idea that activation of 5-HT_1A receptors may increase an inward rectifier, as has been reported in other central neurones (Colino & Halliwell, 1987; Williams et al. 1988; Pan et al. 1993; Penington et al. 1993; Okuhara & Beck, 1994). Furthermore, our results show that both the steady-state I-V curve and the reversal potential of I_5-HT,outward were not changed in low Na⁺ solution.

Ionic basis of I_5-HT,inward

Results from ion substitution studies and the observation of two different membrane potential-depolarizing response relationships in our previous study suggest that 5-HT depolarizations may be caused by an opening or closing of several different channels, for example Na⁺ and K⁺ (Hwang & Dun, 1998). Here, the types of conductance affected by 5-HT were investigated by analysing the steady-state I-V relationships of I_5-HT,inward.

Characteristics and possible mechanisms of type I and type II I_5-HT,inward

Two types of steady-state I-V relationship for I_5-HT,inward were observed in RVLM neurones. Type I I_5-HT,inward was characterized by a decline with hyperpolarizing membrane potential and by a significant decrease of membrane conductance. A decrease of K⁺ is likely to be the major ionic mechanism. However, the response became smaller in a low Na⁺ solution, indicating the involvement of a Na⁺ or cation conductance, in addition to a decrease of K⁺ conductance. Type II I_5-HT,inward exhibited no significant change of membrane conductance and was relatively independent of membrane potential. These types of response may be explained by (1) activation of an active transport mechanism, or (2) simultaneous activation of two (or more) conductances, for example, a combination of cation conductance increase and K⁺ conductance decrease. Results obtained in this study are consistent with the latter hypothesis.

Involvement of K⁺ in I_5-HT,inward

A decrease of K⁺ conductance is thought to be the primary mechanism underlying depolarizations elicited by slow excitatory neurotransmitters including 5-HT (Nicoll et al. 1990). Results obtained in the present study are consistent with the hypothesis that reduction of a voltage-independent K⁺ conductance is one of the major mechanisms underlying I_5-HT,inward in RVLM neurones. First, type I I_5-HT,inward had a linear I-V relationship, declined with hyperpolarization and reversed polarity at a potential close to or negative to the theoretical E_K in 8/17 neurones. Second, both the type I and type II I_5-HT,inward declined with hyperpolarization and had a linear I-V relationship with a reversal potential of -110.0 ± 2.0 and -111.0 ± 5.6 mV, respectively, in a low Na⁺ solution. This linear steady-state I-V curve may represent the residual K⁺ component in a low Na⁺ solution. The more negative reversal potential may be explained by contamination by the cation component. Third, a dependence of the reversal potential on [K⁺]_o was demonstrated in a low Na⁺ and high K⁺ solution for type I I_5-HT,inward where a decrease in K⁺ conductance is likely to be the predominant ionic mechanism. Even though the amplitude was significantly reduced, the steady-state I-V curve of type II I_5-HT,inward did not reverse at a potential close to the estimated E_K (-74 mV) in low Na⁺ and high K⁺ solution. Compared with that of the type I I_5-HT,inward, the ionic basis of type II I_5-HT,inward is probably a larger cation component and a smaller K⁺ component. The K⁺ component was revealed in the low Na⁺ solution. In the low Na⁺ solution, an increase in [K⁺]_o would cause a decrease in the amplitude of the K⁺ component, in addition to a shift of the reversal potential in the positive direction. An increase of [K⁺]_o could also increase the cation component on the assumption that both K⁺ and Na⁺ may be involved in the cation conductance affected by 5-HT. Therefore, the observation that type II I_5-HT,inward was not reversed at a potential close to the estimated E_K in the low Na⁺ and high K⁺ solution may be due to a relatively small K⁺ component and a simultaneous decrease of K⁺ and increase of the cation component in high K⁺ solution. Lastly, 5-HT produced a significantly smaller inward current in RVLM neurones recorded with electrodes in which K⁺ in the patch solution was replaced with Cs⁺, which blocks K⁺ channels. The present results together with our previous findings that depolarizations in response to 5-HT were slightly reduced by an elevated K⁺ solution but not by low Cl⁻ or Ca²⁺-free solution (Hwang & Dun, 1998), suggest that a decrease of K⁺ conductance contributes to type I and type II I_5-HT,inward in RVLM neurones. In addition, our results show that Ba²⁺ at 1 mM only partially suppressed the K⁺ conductance(s) involved in I_5-HT,inward.

Involvement of cation conductances in I_5-HT,inward

In addition to a reduction of K⁺ conductance, an increase of cation conductance has also been suggested to underlie the excitatory effect of several putative transmitters including acetylcholine, noradrenaline and 5-HT (Nicoll et al. 1990). In the present study, I_5-HT,inward was significantly reduced by low Na⁺ solution, indicating a substantial contribution of a Na⁺ or cation conductance. To study the Na⁺ or cation conductance in isolation, the voltage-dependent Na⁺ and Ca²⁺ currents were eliminated by TTX and Ca²⁺-free-high Mg²⁺ solution and K⁺ current was abolished by intracellular infusion of Cs⁺. Under these conditions, the I_5-HT,inward decreased with depolarization and reversed at -11.9 ± 6.4 mV in 67 % of neurones. On the basis of steady-state I-V relationships and the mean reversal potential, the I_5-HT,inward obtained under these conditions appeared to be the result of activation of a cation conductance. The I_5-HT,inward was unaffected by membrane potential in the entire testing range in the remaining 33 % of neurones: this characteristic resembles that of type II I_5-HT,inward. A possible explanation is that the K⁺ component may not have been completely removed by intracellular perfusion of Cs⁺. An incomplete replacement of K⁺ might be due to a long diffusion distance between the patched site and the responsive site and/or limited permeability of Cs⁺ through K⁺ channels (Latorre & Miller, 1983). Therefore, the I_5-HT,inward shown in Fig. 6B and C may represent a predominant cation component. Furthermore, the similarity between the I-V relationships of the I_5-HT,inward obtained with Cs⁺-containing pipettes and of the I_5-HT,inward sensitive to low Na⁺ when K⁺-containing pipettes were used (i.e. the current differences shown as dashed lines in Fig. 4C and D) provides additional evidence of the involvement of a cation conductance. It may be concluded that a concomitant decrease of K⁺ conductance and increase of cation conductance is the primary mechanism underlying type I and type II I_5-HT,inward. The varying degree of contribution of these two components may explain the difference in the steady-state I-V relationship.

Bobker & Williams (1989) have shown that 5-HT augmented a time- and voltage-dependent, non-selective cation current, I_h, which was sensitive to blockade by Cs⁺ in rat nucleus propositus hypoglossi. A similar response has been reported in lateral geniculate neurones of the guinea-pig and cat (Pape & McCormick, 1989), rat spinal motoneurones (Takahashi & Berger, 1990) and rat facial motoneurones (Larkman & Kelly, 1992). The I_h found to be present in about half of the RVLM neurones studied here was sensitive to Cs⁺. However, the cation conductance affected by 5-HT in RVLM neurones is not likely to be I_h because, in most of the RVLM neurones, 5-HT did not increase the hyperpolarization-induced inward current relaxation and I_5-HT,inward was not affected by Cs⁺ in the potential range that I_h could be activated. Takahashi & Berger (1990) reported that the 5-HT effect on I_h is mediated by 5-HT_1A receptors. In the present study, the 5-HT_1A antagonist PBD was present in all the experiments where the ionic mechanism of I_5-HT,inward was studied. However, PBD was not present in experiments where I_5-HT,outward was examined and 5-HT did not increase I_h in any of these neurones.

A voltage-independent and a voltage-dependent non-selective cation conductance activated by muscarine have been reported in rat locus coeruleus neurones (Shen & North, 1992) and association cortex (Haj-Dahmane & Andrade, 1996). 5-HT and muscarine may modulate the same cation conductance. As shown in Fig. 6B and C, the two steady-state I-V curves recorded with the Cs⁺-containing electrode graphically resembled the voltage-independent and voltage-dependent cation currents, respectively. However, a decrease in K⁺ conductance occurring simultaneously with a cation conductance increase would result in a progressive decrease of the inward current at hyperpolarized potentials, leading to the conclusion that the underlying cation conductance is voltage dependent. As a potassium component cannot be completely excluded even with the use of a Cs⁺-containing patch solution, a detailed analysis of the characteristics of the cation conductance affected by 5-HT in RVLM neurones was not performed.

A combined K⁺ conductance decrease and cation conductance increase has been suggested to underlie the slow excitatory effect of muscarine in bullfrog ganglion cells (Kuba & Koketsu, 1978; Tsuji & Kuba, 1988) and rat locus coeruleus neurones (Shen & North, 1992), and of substance P on guinea-pig inferior mesenteric ganglion cells (Dun & Minota, 1981). A similar mechanism has also been reported for the actions of substance P, 5-HT, muscarine, vasoactive intestinal polypeptide, forskolin and slow excitatory synaptic transmission in guinea-pig submucosal neurones (Shen & Surprenant, 1993). The present study provides evidence that this type of mechanism may contribute to the depolarizing action of 5-HT in RVLM neurones. It remains to be determined whether a single class of 5-HT₂ receptors may couple to different channels or whether the channels are linked to subclasses of 5-HT₂ receptors.

Acknowledgments

This study was supported by NIH grant HL51314 from the Department of Health and Human Services.

References

Bayliss, DA; Umemiya, M; Berger, AJ. Inhibition of N- and P-type calcium currents and the after-hyperpolarization in rat motoneurones by serotonin. The Journal of Physiology. 1995;485:635–647. [PubMed]
Blanton, MG; LoTurco, JJ; Kriegstein, AR. Whole-cell recording from neurons in slices of reptilian and mammalian cerebral cortex. Journal of Neuroscience Methods. 1989;30:203–210. [PubMed]
Bobker, DH; Williams, JT. Serotonin augments the cationic current I_h in central neurons. Neuron. 1989;2:1535–1540. [PubMed]
Bobker, DH; Williams, JT. Ion conductances affected by 5-HT receptor subtypes in mammalian neurons. Trends in Neurosciences. 1990;13:169–173. [PubMed]
Chalmers, JP; Pilowsky, PM; Minson, JB; Kapoor, V; Mills, E; West, MJ. Central serotonergic mechanisms in hypertension. American Journal of Hypertension. 1988;1:79–83. [PubMed]
Colino, A; Halliwell, JV. Differential modulation of three separate K conductances in hippocampal CA1 neurones by serotonin. Nature. 1987;328:73–77. [PubMed]
Dampney, RAL. The subretrofacial vasomotor nucleus: Anatomical, chemical and pharmacological properties and role in cardiovascular regulation. Progress in Neurobiology. 1994;42:197–227. 10.1016/0301-0082(94)90064-7. [PubMed]
DiFrancesco, D. A study of the ionic nature of the pace-maker current in calf Purkinje fibres. The Journal of Physiology. 1981;314:377–393. [PubMed]
Dun, NJ; Minota, S. Effects of substance P on neurones of the inferior mesenteric ganglia of the guinea-pig. The Journal of Physiology. 1981;321:259–271. [PubMed]
Frishman, WH; Huberfeld, S; Okin, S; Wang, Y-H; Kumar, A; Shareef, B. Serotonin and serotonin antagonism in cardiovascular and non-cardiovascular disease. Journal of Clinical Pharmacology. 1995;35:541–572. [PubMed]
Gillis, RA; Hill, KJ; Kirby, JS; Quest, JA; Hamosh, P; Normal, WP; Kellar, KJ. Effect of activation of central nervous system serotonin 1A receptors on cardiorespiratory function. Journal of Pharmacology and Experimental Therapeutics. 1989;248:851–857. [PubMed]
Hagiwara, S; Miyazaki, S; Rosenthal, NP. Potassium current and the effect of caesium on this current during anomalous rectification of the egg cell membrane of a starfish. Journal of General Physiology. 1976;67:621–638. 10.1085/jgp.67.6.621. [PubMed]
Haj-Dahmane, S; Andrade, R. Muscarinic activation of a voltage-dependent cation nonselective current in rat association cortex. Journal of Neuroscience. 1996;16:3848–3861. [PubMed]
Helke, CJ; McDonald, CH; Phillips, ET. Hypotensive effects of 5-HT_1A receptor activation: ventral medullary sites and mechanisms of action in the rat. Journal of the Autonomic Nervous System. 1992;42:177–188. 10.1016/0165-1838(93)90048-Y. [PubMed]
Hoyer, D; Clarke, DE; Fozard, JR; Hartig, PR; Martin, GR; Mylecharane, EJ; Saxena, PR; Humphrey, PPA. VII. International union of pharmacology classification of receptors for 5-hydroxytryptamine (serotonin). Pharmacological Reviews. 1994;46:157–203. [PubMed]
Hwang, LL; Dun, NJ. Multiple effects of 5-HT on conductances of rat rostral ventrolateral medulla neurons. Society for Neuroscience Abstracts. 1997;23:1229.
Hwang, LL; Dun, NJ. 5-Hydroxytryptamine responses in immature rat rostral ventrolateral medulla neurons in vitro. Journal of Neurophysiology. 1998;80:1033–1041. [PubMed]
Kuba, K; Koketsu, K. Synaptic events in sympathetic ganglia. Progress in Neurobiology. 1978;11:77–169. 10.1016/0301-0082(78)90010-2. [PubMed]
Larkman, PM; Kelly, JS. Ionic mechanisms mediating 5-hydroxytryptamine- and noradrenaline-evoked depolarization of adult rat facial motoneurones. The Journal of Physiology. 1992;456:473–490. [PubMed]
Latorre, R; Miller, C. Conduction and selectivity in potassium channels. Journal of Membrane Biology. 1983;71:11–30. [PubMed]
Lin, HH; Wu, S; Lai, CC; Dun, NJ. GABA- and glycine-mediated IPSPs in neonatal rat rostral ventrolateral medulla neurons in vitro. Neuroscience. 1998;82:429–442. 10.1016/S0306-4522(97)00294-7. [PubMed]
Lovick, TA. Cardiovascular responses to 5-HT in the ventrolateral medulla of the rat. Journal of the Autonomic Nervous System. 1989a;28:35–41. 10.1016/0165-1838(89)90005-2. [PubMed]
Lovick, TA. Systemic and regional haemodynamic responses to microinjection of 5-HT agonists in the rostral ventrolateral medulla in the rat. Neuroscience Letters. 1989b;107:157–161. 10.1016/0304-3940(89)90809-4. [PubMed]
Mandal, AK; Kellar, KJ; Norman, WP; Gillis, RA. Stimulation of serotonin₂ receptors in the ventrolateral medulla of the cat results in nonuniform increases in sympathetic outflow. Circulation Research. 1990;67:1267–1280. [PubMed]
Mar, LPD; Cardenas, CG; Scroggs, RS. Serotonin inhibits high-threshold Ca²⁺ channel currents in capsaicin-sensitive acutely isolated adult rat DRG neurons. Journal of Neurophysiology. 1994;71:2551–2554. [PubMed]
Mayer, ML; Westbrook, GL. A voltage-clamp analysis of inward (anomalous) rectification in mouse spinal sensory ganglion neurones. The Journal of Physiology. 1983;340:19–45. [PubMed]
Nicoll, RA; Malenka, RC; Kauer, JA. Functional comparison of neurotransmitter receptor subtypes in mammalian central nervous system. Physiological Reviews. 1990;70:513–565. [PubMed]
North, RA; Uchimura, N. 5-Hydroxytryptamine acts at 5-HT₂ receptor to decrease potassium conductance in rat nucleus accumbens neurones. The Journal of Physiology. 1989;417:1–12. [PubMed]
Nosjean, A; Guyenet, PG. Role of ventrolateral medulla in sympatholytic effect of 8-OH-DPAT in rats. American Journal of Physiology. 1991;260:R600–609. [PubMed]
Okuhara, DY; Beck, SG. 5-HT_1A receptor linked to inwardly-rectifying potassium current in hippocampal CA3 pyramidal cells. Journal of Neurophysiology. 1994;71:2161–2167. [PubMed]
Pan, ZZ; Wessendorf, MW; Williams, JT. Modulation by serotonin of the neurons in rat nucleus raphe magnus in vitro. Neuroscience. 1993;54:421–429. 10.1016/0306-4522(93)90263-F. [PubMed]
Pape, HC; McCormick, DA. Noradrenaline and serotonin selectivity modulate thalamic burst firing by enhancing a hyperpolarization-activated cation current. Nature. 1989;340:715–718. 10.1038/340715a0. [PubMed]
Penington, NJ; Kelly, JS. Serotonin receptor activation reduces calcium currents in an acutely dissociated adult central neuron. Neuron. 1990;4:751–758. [PubMed]
Penington, NJ; Kelly, JS; Fox, AP. Whole-cell recordings of inwardly rectifying K⁺ currents activated by 5-HT_1A receptors on dorsal raphe neurones of the adult rat. The Journal of Physiology. 1993;469:387–405. [PubMed]
Pickering, AE; Spanswick, D; Logan, SD. 5-Hydroxytryptamine evokes depolarizations and membrane potential oscillations in rat sympathetic preganglionic neurones. The Journal of Physiology. 1994;480:109–121. [PubMed]
Shen, K-Z; North, RA. Muscarine increases cation conductance and decreases potassium conductance in rat locus coeruleus neurones. The Journal of Physiology. 1992;455:471–485. [PubMed]
Shen, K-Z; Surprenant, A. Common ionic mechanisms of excitation by substance P and other transmitters in guinea-pig submucosal neurones. The Journal of Physiology. 1993;462:483–501. [PubMed]
Steinbusch, HWM. Distribution of serotonin-immunoreactivity in the central nervous system of the rat cell bodies and terminals. Neuroscience. 1981;6:557–618. 10.1016/0306-4522(81)90146-9. [PubMed]
Takahashi, T; Berger, AJ. Direct excitation of rat spinal motoneurones by serotonin. The Journal of Physiology. 1990;423:63–76. [PubMed]
Tsuji, S; Kuba, K. Muscarinic regulation of two ionic currents in the bullfrog sympathetic neurone. Pflügers Archiv. 1988;411:361–370.
Vertes, RP; Kocsis, B. Projections of the dorsal raphe nucleus to the brainstem: PHA-L analysis in the rat. Journal of Comparative Neurology. 1994;340:11–26. [PubMed]
Williams, JT; Colmer, WF; Pan, ZZ. Voltage- and ligand-activated inwardly rectifying currents in dorsal raphe neurons in vitro. Journal of Neuroscience. 1988;8:3499–3506. [PubMed]
Xi-Moy, SX; Dun, NJ. Potassium currents in adult rat intracardiac neurones. The Journal of Physiology. 1995;486:15–31. [PubMed]