| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Department of Biosciences, Kyushu Dental College, 2-6-1 Manazuru, Kokurakitaku, Kitakyushu, 803-8580, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
(Received 9 May 2005;
accepted after revision 24 August 2005;
first published online 25 August 2005)
Corresponding author K. Inenaga: Department of Biosciences, Kyushu Dental College, 2-6-1 Manazuru, Kokurakitaku, Kitakyushu, 803-8580, Japan. Email: ine{at}kyu-dent.ac.jp
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
-subunit families, Shaker (Kv1.4 (Rettig et al. 1994)), Shaw (Kv3.3 and Kv3.4 (Rudy & McBain, 2001)) and Shal (Kv4.1, Kv4.2 and Kv4.3 (Pak et al. 1991)). These homomeric channels are distinguished pharmacologically by the following features: Kv3 channels are highly sensitive (micromolar level) to TEA and 4-aminopyridine (4-AP) (Tseng et al. 1996; Rudy & McBain, 2001) while both Kv1.4 and Kv4 are TEA-resistant less sensitive (millimolar level) to 4-AP (Pak et al. 1991; Rettig et al. 1994). Although there are many papers on SFO neurones, there is no consensus on the pharmacological properties of the IA. It has been reported that the IA in dissociated SFO neurones is partially suppressed by either TEA or 4-AP at 0.1 mM (Schmid, 1998), and the IA in the slice preparation, like the delayed rectifier K+ current (IK), is completely blocked by 20 mM TEA (Ferguson & Li, 1996). On the other hand, in studies using dissociated cultured SFO neurones, IAs are clearly observed even in the presence of high-dose TEA (5–20 mM; Washburn et al. 1999; Johnson et al. 1999; Washburn & Ferguson, 2001). Further features of the IAs that distinguish between Kv1 and Kv4 channels include the well-known and important difference between their time courses of recovery from inactivation (Frank An et al. 2000; Amberg et al. 2003). The recovery time constant from inactivation of the Kv1-derived IA is slow (of the order of 1 s), and that of the Kv4-derived IA is fast (of the order of 100 ms). In a previous study (Washburn et al. 1999), the recovery time constant of IA in SFO neurones has been reported to be approximately 50 ms. This suggests that the IA in SFO neurones is generated by a member of the Kv4 family although this has not been studied in detail.
In the present study, we investigated the pharmacological and electrophysiological properties of the IA in SFO neurones, together with the mRNA expression of IA-generating Kv -subunits using molecular biological techniques. We then studied the effect of ANG on the IA.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Dissociation of SFO neurones
The dissociation procedure used was modified from that of Renganathan et al. (Renganathan et al. 2000; Hiyama et al. 2002). Male adult Wistar rats weighing 150–250 g were deeply anaesthetized with ketamine (250 mg kg–1, S.C.) and decapitated. Slices 300 µm thick were prepared in a cold PBS, which contained (mM): NaCl 137, KCl 2.68, Na2HPO4 8.1, and KH2PO4 1.47. After making the slices, the SFO tissue was dissected away from the hippocampal commissure and corpus callosum under a microscope. The SFO tissue was then digested enzymatically for 15 min at 37°C with collagenase A (1 mg ml–1; Roche Diagnosistics Co., USA) in PBS, and for 15 min at 37°C with collagenase D (1 mg ml–1; Roche Diagnosistics Co., USA) and papain (30 U ml–1, Worthington Biochemical Co., NJ, USA) in PBS containing 0.5 mM EDTA and 2 mg ml–1 cysteine. The preparations were washed in culture medium (DMEM and F12 in a ratio of 1: 1, 10% fetal calf serum), and then carefully placed in a new culture medium using a Pasteur pipette. They were dispersed mechanically with fire-polished glass pipettes (tip i.d. ranging from 200 to 500 µm) and plated onto dishes (Falcon, 3801). The dissociated neurones were maintained in a humidity-controlled incubator gassed with 5% CO2 at 37°C for 3–6 h prior to electrophysiological experimentation.
Whole-cell patch clamp recording
For all dissociated cell experiments, cells were submerged in a chamber with a volume of approximately 0.5 ml and perfused (flow rate 1 ml min–1) at room temperature (22–24°C). The cells for recording were observed using an upright microscope (TMD, Nikon, Japan) and monitored with a CCD camera (MTV-7480ND, Mintron, Taiwan or DS-5M-L1, Nikon, Japan). Immediately after the end of recording, digital images of all cells were made and the cell diameters were measured as mean values along their major and minor axes. Drugs were purchased from the following pharmaceutical companies; TTX from Wako (Japan), CdCl2 from Sigma (USA), blood depressing substance-I (BDS-I) from Anemone Laboratory (Israel), 4-AP and TEA chloride from Kanto Chemical (Japan) and ANG from the Peptide Institute (Japan). All drugs were applied to the cells by the Y-tube method (Nakagawa et al. 1990). In solutions containing TEA chloride (10 mM and 30 mM), equimolar NaCl was used to adjust the osmolarity.
Following our dissociation protocol, the preparation contained other cells, for example, glial and epithelial cells in addition to neurones. Recording cells were judged to be neurones by the following criteria: they were visually round and bright, and had TTX (0.3 µM)-sensitive Na+ inward current in voltage-clamp recordings or generated action potentials in current-clamp recordings. After the identity of the cells had been established, TTX was used throughout all voltage-clamp recordings. The normal perfusion solution contained (mM): NaCl 124, KCl 3, NaHCO3 26, glucose 10, MgSO4 1.3, KH2PO4 1.24, CaCl2 2.1. In most experiments, CdCl2 at 300 µM was added as a Ca2+ channel blocker to eliminate Ca2+ currents and intracellular Ca2+-sensitive K+ currents. Since Cd2+ has a direct effect on Kv1.4- and Kv4.2-derived IA (Yellen et al. 1994; Fiset et al. 1997; Song et al. 1998), a Ca2+-free solution with calcium replaced on an equimolar basis by Mg2+ was used in some experiments. The perfusion solution was saturated with a mixture of 95% O2 and 5% CO2. Patch pipettes were double-pulled (Sutter Instrument, P-97, USA) from thick glass capillaries (GD-1.5, Narishige, Japan), and were adjusted to 2–5 M
when filled with a solution containing (mM): K gluconate 145, MgCl2 3, EGTA 0.2, Hepes 10 (pH 7.2 adjusted with KOH). The series resistance was less than 10 M and compensated up to 70%. The cell capacitance was monitored throughout the recordings. The values of the membrane potentials were corrected for the liquid junction potentials (11 mV) present in the pipette. Currents and voltage were recorded using a whole-cell clamp amplifier (HEKA Elektronik, EPC-8, Germany). In voltage-clamp mode, three holding potentials (–40, –70 and –100 mV) were used. Voltage-steps of 250 or 500 ms were applied to all SFO neurones at 5 s intervals. In current-clamp mode, more than 70% of neurones displayed spontaneous firing and sometimes bursts at the resting membrane potentials of between –45 and –70 mV. Accordingly, neurones were manually clamped below –80 mV to avoid the occurrence of spontaneous firing or bursts, and 20 or 40 pA currents were injected into cells from the recording pipette to study the effect of the drugs on action potential generation.
Conventional RT-PCR analysis
The total RNA from the SFO was analysed with a protocol similar to that reported previously (Honda et al. 2003). SFO tissues were prepared by the same methods used for cell dissociation (see above). The cerebellum was used as a positive control for Kv3.3. Total RNA was extracted using RNeasy Mini Kits (QIAGEN, Japan). Reverse transcription of the total RNA (40 ng) was performed in a final volume of 20 µl using oligo-dT12–18 primer (0.5 µg µl–1) and RNasin (10 U; TaKaRa, Japan) with sensiscript RT kit (QIAGEN, Japan). PCR was performed with a thermal cycler (PCR Thermal Cycler Dice; TaKaRa, Japan). Specific primers for Kv1.4, Kv3.4, Kv4.1, Kv4.2, Kv4.3 (Song et al. 1998), Kv3.1, Kv3.2 (Lien et al. 2002), Kv3.3 (Ohya et al. 1997), KChIP1 and KChIP2 (Boland et al. 2003) were used in published sequences. PCR was performed with a PCR buffer containing 10 pmol primers, 2.5 U Taq DNA polymerase (Hot Start Version; TaKaRa, Japan) and each transcribed cDNA, in a final volume of 50 µl. Single-stranded cDNA products were denatured and subjected to PCR amplification (40 cycles). Each PCR cycle consisted of denaturation at 94°C for 20 s, annealing at 58–62°C for 30 s and finally extension at 70°C for 35 s. Total RNA samples were used as negative controls, and there was no signal for all primer sets. The PCR products were separated by electrophoresis on 2% agarose gels and visualized by ethidium bromide staining. All positive amplicons were sequenced with a dye termination procedure and these sequences were found to closely match published sequences, except for a middle band on the line of Kv3.4.
Single-cell multiplex RT-PCR analysis
After electrophysiological recordings, the cell content was aspirated into the electrode pipette by applying negative pressure. Electrodes contained approximately 6 µl of sterile pipette solution (see above) to which was added RNasin (2 U) to suppress mRNA degradation during the time of recording. After aspiration, the tip of the electrode was broken, and its contents were injected into tube containing oligo-dT12–18 primer (0.5 µg µl–1) and RNasin (5 U) with sensiscript RT kit to give a final volume of 10 µl. Single-stranded cDNA was synthesized in an incubator at 37°C overnight. To identify mRNAs of five IA-generating Kv -subunits coexpressed in single cells, a two-step protocol was used. In the first round, each single-cell cDNA sample was used as a template in a multiplex (five pairs of primers, each 10 pmol) PCR reaction in a final volume of 100 µl. Twenty cycles were performed at 58°C. In the second round, 2 µl of the first reaction mixture sample was used as a template with each pair of specific primers, as well as in the conventional PCR. To ensure that genomic DNA did not contribute to the PCR products, four cells were aspirated after patch-clamp recording, and processed in the normal manner without the reverse transcriptase. These procedures prevented any PCR products from being detected.
Data analysis
Data were analysed off-line by using Pulse Fit (HEKA Elektronik, Germany), Microsoft Excel (Microsoft, Japan), Prism (GraphPad Software, USA) software and an analog–digital converter (MacLab/8, ADI, Australia). After converting the current and voltage traces into ASCII files, calculation and measurement were carried out using a custom macro-program in Microsoft Excel. The equilibrium potential of K+ (EK) is –88 mV using the Nernst equation. The conductance of the outward current (G) was calculated as G = Ipeak(Vc – EK) where Ipeak is the peak amplitude of the current evoked by command voltage-steps (Vc). The relationships between normalized conductance and membrane potential were fitted by the Boltzmann equation: G/Gmax = {1 + exp[(V* – Vc)/k]}–1 where Gmax is the maximal membrane conductance, V* is the midpoint potential and k is the slope coefficient. Inactivation curves were fitted to the equation: I/Imax = {1 + exp[(V* – Vpre)/k]}–1 where I is the current evoked by voltage-step of +50 mV following a 500 ms prepulse (Vpre) to a different voltage and Imax is the maximal peak amplitude of the current. Recovery time courses of electrically isolated IA from inactivation were fitted to single- and double-exponential functions and best-fit recovery time constants (R
s) were calculated. The numerical data are given as mean ± S.E.M., and n represents the number of neurones tested. For statistical analysis, we used Student's t test and the Bonferroni post hoc test following two-way ANOVA. The F-test was used to evaluate curve fitting of two exponential functions.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
We used conventional RT-PCR for the -subunits of the Kv3 family channels to investigate the molecular basis of the highly TEA-sensitive currents (Fig. 1A). Kv3.1 and Kv3.2 channels generated highly TEA-sensitive IK and Kv3.3 and Kv3.4 channels generated highly TEA-sensitive IA (Rudy & McBain, 2001). The mRNAs for Kv3.1, Kv3.2 and Kv3.4 were expressed in the SFO. However, the mRNA of Kv3.3 was not detected in the SFO, although it was found in the cerebellum. PCR primers for Kv3.4 were targeted to identify the splice variants, Kv3.4a and Kv3.4c, and both mRNAs were expressed in the SFO. A middle band on the line of Kv3.4 was identified as a non-specific sequence.
|
Kv3.4 specific blocker BDS-I sensitivity of SFO neurones
TEA at 1 mM blocks not only Kv3.4-derived IA, but also Kv3.1- and/or Kv3.2-derived IK (Rudy & McBain, 2001). Blood depressing substance-I (BDS-I) from sea anemones has been reported to be a specific blocker for the Kv3.4-derived IA (Diochot et al. 1998). Hence, we investigated the effects of BDS-I at 2 µM on the whole-cell currents in seven SFO neurones. Voltage steps were applied to neurones from –70 mV to +50 mV in 20 mV increments for 500 ms duration, followed by a constant voltage step to +50 mV (Fig. 2Aa inset). In four of seven neurones, BDS-I at 2 µM inhibited IA-like currents (BDS-I-sensitive IA-like currents, Fig. 2A). The currents recovered to over 80% of maximal inhibition after washout. However, peak amplitudes of the BDS-I-sensitive IA-like currents were very small (0.11 ± 0.03, 0.24 ± 0.06 and 0.33 ± 0.07 nA in response to voltage steps of +10, +30 and +50 mV, respectively). The voltage dependence of the BDS-I-sensitive IA-like currents is shown in Fig. 2B. These data could be fitted well to a Boltzmann function, having a V* of 12.6 ± 13.4 mV and a k of 13.9 ± 1.3 mV in the activation curve, and V* of –8.8 ± 1.6 mV and a k of –11.7 ± 1.5 mV in the inactivation curve. In three BDS-I-insensitive neurones, TEA at 1 mM was further applied after washout of BDS-I. TEA-sensitive currents in all BDS-I-insensitive neurones failed to show exponential decay. In current-clamp mode in the normal perfusion solution, BDS–I at 2 µM broadened action potentials in three of seven SFO neurones (Fig. 2C).
|
Next, to investigate TEA-resistant IA, expressions of the mRNAs of the -subunits of Kv1.4, Kv4.1, Kv4.2 and Kv4.3, channels were examined by conventional RT-PCR (Fig. 3A). All mRNAs were expressed in the SFO. PCR primers for Kv4.3 were targeted to identify the splice variants, Kv4.3L and Kv4.3M (Ohya et al. 1997). Kv4.3L was abundantly expressed in the SFO, but Kv4.3M was not detected.
|
Voltage dependence of TEA-resistant IA
Voltage-steps were applied to neurones from –100 mV to +50 mV in 10 mV increments for 500 ms duration and followed by a constant voltage-step to +50 mV in the Ca2+-free solution including TEA at 1 mM (Fig. 4Aa inset). The net 4-AP-sensitive and TEA-resistant IAs (Fig. 4Ac) were obtained by subtracting the currents before (Fig. 4Aa) and after (Fig. 4Ab) the application of 4-AP at 5 mM. The voltage dependence of activation and inactivation curves for the 4-AP-sensitive and TEA-resistant IA are shown in Fig. 4C (n = 6). These data fitted well to Boltzmann functions, having V* of –23.4 ± 2.3 mV and k of 22.6 ± 2.6 mV in the activation curve and V* of –71.5 ± 0.7 mV and k of –10.3 ± 0.5 mV in the inactivation curve, respectively. Even in the presence of 30 mM TEA, the activation curve of 4-AP-sensitive and TEA-resistant IA was almost the same as that in the presence of 1 mM TEA (n = 5, V* of –20.0 ± 1.6 mV and k of 20.9 ± 1.8 mV, data not shown). However, maximal values of Gmax in the presence of TEA 30 mM (5.8 ± 0.6 pS) were significantly lower (P < 0.05, unpaired t test) than those in the presence of TEA 1 mM (9.1 ± 1.1 pS). This suggests that 4-AP-sensitive IA is partially suppressed by high concentration of TEA. In Cd2+-containing solution, the voltage dependence of the 4-AP-sensitive and TEA-resistant IA was significantly shifted to the right (n = 5, Fig. 4C; V* of –6.7 ± 2.3 mV and k of 19.4 ± 2.3 mV in the activation curve and V* of –56.0 ± 1.5 mV and k of –10.7 ± 1.3 mV in the inactivation curve), compared with the results in Ca2+-free solution (P < 0.05 and P < 0.001, respectively, by two-way ANOVA). There was no significant difference of maximal values of Gmax between Ca2+-free and Cd2+-containing solutions (11.9 ± 2.0 pS and 10.7 ± 1.2 pS) in the presence of TEA at 1 mM.
|
To study roles of highly TEA-sensitive currents or 4-AP-sensitive and TEA-resistant IAs in action potential generation, in the normal perfusion solution, we investigated the effect of TEA and 4-AP on the first spike latency and first interspike interval in six SFO neurones. By the application of TEA at 1 mM, the first interspike interval was significantly expanded, compared with the control (Fig. 5Bb), although the first spike latency was not changed (Fig. 5Ba). Like the application of BDS-I, TEA at 1 mM broadened action potentials (Fig. 5A), but the effects were observed in all neurones. Both the first spike latency and the first interspike interval were shortened by the application of 4-AP at 5 mM in the presence of TEA, and recovered after washout (Fig. 5Ba and Bb). In all neurones, the application of 4-AP substantially broadened action potentials (Fig. 5A).
|
The time courses of recovery from inactivation of the Kv1- and the Kv4-derived IAs are different from each other (Frank An et al. 2000; Amberg et al. 2003). We used a double-pulse protocol with 5 s intervals for 19 SFO neurones to measure the recovery from inactivation. This experiment was done in the presence of TEA at 1 mM in a Cd2+-containing solution, because removal of extracellular Ca2+ typically shortened the duration of recordings, compared with the normal solution. To isolate TEA-resistant IA electrically, we subtracted whole-cell currents between holding potentials of –70 mV and –40 mV (Fig. 6Aa inset). Neurones were divided into two types, based on whether their recovery ratio was more than 0.95 (n = 13) or less than 0.8 (n = 6) with a 700 ms interval and described as fast recovery (F)- or slow recovery (S)- neurones, respectively. Figure 6Aa and Ab shows typical traces for the two cell types. Figure 6B shows single- or double-exponential curve fittings to the mean recovery time course of both F- and S-type cells. When the recovery time courses of both F- and S-type neurones were fitted to single-exponential functions, the best fit R of F-type was 71 ± 2 ms and that of S-type was 470 ± 15 ms. When the values were fitted to double-exponential functions, the best fit Rs of F-type cells were 52 ± 9 ms (78 ± 18%) and 184 ± 100 ms and those of S-type were 177 ± 70 ms (43 ± 22%) and 1.1 ± 0.7 s. Using the F-test, double-exponential fitting curves matched significantly better with the real data (F-type, P < 0.0014 and S-type, P < 0.0001) than single-exponential curves. These data suggest that F-type neurones display only fast recovery components (both R values were under 200 ms) and S-type neurones display both fast (under 200 ms) and slow (more than 1 s) recovery components. Both F- (n = 3) and S- (n = 2) type neurones were seen even in a Ca2+-free solution (data not shown).
|
s of Kv4-derived IA are shortened to 50–130 ms from 200–500 ms by binding to Kv channel-interacting protein (KChIP; Frank An et al. 2000; Nadal et al. 2001; Hatano et al. 2002). We therefore investigated whether the mRNA for KChIP was present, using conventional RT-PCR. The SFO contained the mRNAs for both KChIP1 and KChIP2 (Fig. 6C). PCR primers for KChIP1 were targeted to identify the splice variants, KChIP1a and KChIP1b (Boland et al. 2003). However in our hands, only KChIP1a was expressed in the SFO, and KChIP1b was not. Characteristics in TEA-resistant IA between F- and S-types
As shown in Fig. 6B, the maximum difference of relative currents between two cell types occurred when the interval between the pulses of double-pulse stimulation was around 200 ms. Thus to differentiate cell types more easily, the values for the relative currents with a pulse interval of 200 ms were investigated. For this purpose, we used a triple-pulse stimulation protocol and we differentiated the cell types by measuring cell diameter and capacitance. The triple-pulse stimulation protocol consisted of three 250 ms duration pulses of +20 mV, from a holding potential of –70 mV. The first interval had a duration of 200 ms and after 250 ms at –70 mV, there was a prepulse of –40 mV for 500 ms before the third pulse (Fig. 7Aa inset). We electrically isolated the IAs by subtracting the third current from the first and second currents (Fig. 7A and B). Figure 7C shows the ratios of the second currents against the first currents. The values were clearly divided into two groups, by whether or not the ratio was over 0.7. It thus appears that whether or not the ratio of the second current against the first was above 0.7 with a 200 ms interval, provides useful a criterion for separating the cells into F- and S-types (Fig. 7A: F-type and Fig. 7B: S-type). The membrane capacitance and cell diameter of the F-type (4.4 ± 0.3 pF and 14.2 ± 0.3 µm, n = 37) cells were significantly larger (P < 0.0001) than those of the S-type (2.6 ± 0.3 pF and 11.6 ± 0.2 µm, n = 17) cells. Although peak amplitude of the electrically isolated IA in F-type cells (0.95 ± 0.12 nA) was significantly greater (P < 0.05) than that of the S-type cells (0.53 ± 0.06 nA), there was no significant difference in peak amplitude per membrane capacitance for the two cell types (0.21 ± 0.02 nA pF–1 and 0.27 ± 0.04 nA pF–1, respectively). Figure 7D shows the relationships between peak amplitude of electrically isolated IA and membrane capacitance of F- and S-type neurones. In 14 neurones (F-type, n = 9 and S-type, n = 5), we examined the expression of the mRNAs for Kv1.4, Kv3.4 and Kv4-family -subunits by single-cell multiplex RT-PCR after the electrophysiological recordings. Figure 7E shows a representative expression pattern of mRNAs for an F-type neurone. Figure 7F shows detection frequencies of Kv -subunits for all 14 neurones tested. The PCR product of Kv4.2 was detected in all cells and Kv4.3L in half the cells tested and there seemed no difference in the expression of Kv1.4, Kv3.4, Kv4.2 and Kv4.3L between the two types of cell. We detected the PCR products of Kv3.4c in one or more neurones of both F- and S-types, but never found that of Kv3.4a in any cells.
|
In 15 neurones, we investigated the effect of ANG on the TEA-resistant IAs of F- and S-type neurones.
After confirming that the current traces were stable for 1–2 min, we repeatedly delivered the triple-pulse stimulation protocol every 10 s and then applied ANG at 30 nM for 10 s. We judged neurones responsive when mean peak amplitude of the first electrically isolated current after application of ANG decreased by more than 10% when compared with that for 30 s before ANG application (control). Application of ANG inhibited TEA-resistant IAs in half of the F-type neurones tested (n = 5 of 10) and in all of the S-type neurones (n = 5) (Fig. 8A). A small degree of recovery was observed in three S-type neurones only. We have reported strong tachyphylaxis after application of ANG to SFO neurones (Ono et al. 2001). Figure 8B shows the time course of the first and second electrically isolated currents, normalized to the currents just before the application of ANG. The time course of the group of responsive neurones was significantly different from that of the non-responsive neurones (P < 0.005 in both types, by two-way ANOVA). There was also a slight difference in time course of the inhibition of ANG shown by the two cell types, but the difference was not significant (two-way ANOVA). The inhibition ratio following ANG administration for the second TEA-resistant IA in S-type cells (but not F-type cells) was significantly smaller (repeated measures two-way ANOVA, P < 0.05, Fig. 8Bb) than that of the first TEA-resistant IA. ANG at 30 nM also induced a small inward current (–10 ± 3 pA) in four neurones out of the ten that responded (two neurones of each cell type).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The Kv3.1- and Kv3.2-derived currents in transfected cells are highly TEA-sensitive IKs and are activated at potentials more positive than –20 mV (Rettig et al. 1992; Rudy & McBain, 2001; Baranauskas et al. 2003). In the present study, the threshold of highly TEA-sensitive IKs was near –20 mV, and mRNAs for Kv3.1 and Kv3.2 were both found in the SFO using conventional RT-PCR. The highly TEA-sensitive IKs in SFO neurones are therefore probably a consequence of the activation of Kv3.1 and/or Kv3.2 channels. We also detected the mRNA of Kv3.4 in the SFO by conventional RT-PCR and in SFO neurones by single-cell RT-PCR. TEA at 1 mM blocks Kv3.4-derived IA, as well as Kv3.1- and Kv3.2-derived IKs (Rettig et al. 1992; Rudy & McBain, 2001). Thus the highly TEA-sensitive IA-like currents in some SFO neurones probably included Kv3.4-derived IA, as well as Kv3.1 and/or Kv3.2-derived IKs. The BDS-I-sensitive IA-like current in Fig. 2Ac appeared to have a relatively slow inactivation with an IK-like sustained component. Some studies (Rudy & McBain, 2001; Baranauskas et al. 2003) have reported that heteromeric Kv3.4/Kv3.1-derived IA shows slow inactivation with an IK-like component, although homomeric Kv3.4-derived IA in transfected cells shows only fast inactivation without an IK-like component. We were thus able to confirm the nature of the kinetics of the heteromeric Kv3.4/Kv3.1 channels that were reported earlier. This probably means that the BDS-I-sensitive IA-like currents found in some SFO neurones are generated through a heteromeric Kv3.4/Kv3.1 channel, rather than a homomeric Kv3.4 channel. The present study cannot however, exclude the possibility that there are hybrid currents consisting of homomeric Kv3.1 and Kv3.4 channels.
In voltage-clamp mode, TEA- and BDS-I-sensitive IA-like currents were observed in 16 of 60 (27%) and 4 of 7 (57%) neurones, respectively. In current-clamp mode, 3 of 7 (42%) neurones displayed action potentials that were broadened by the presence of BDS-I. The proportion of neurones that were sensitive to BDS-I was greater than the proportion that showed a TEA-sensitive IA-like current. These results suggest that, in some neurones, a large TEA-sensitive IK masks a small Kv3.4-related IA with an exponential decay. Further with single-cell RT-PCR, the PCR product of Kv3.4c was detected in only 2 of 14 (14%) of neurones. The PCR primer set for Kv3.4 in the present study detected three PCR products (Kv3.4a, Kv3.4c and non-specific bands) seen in conventional RT-PCR. This might be because the detection efficiency of the PCR amplification for Kv3.4c in the single-cell RT-PCR was too low, so that it was undetectable because of the production of other PCR products. Further single-cell RT-PCR experiments using more specific PCR primer sets for Kv3.4c and Kv3.4a together with electrophysiological and pharmacological experiments for Kv3.4 may be needed to settle this issue. We thus cannot be certain from the results of the present study what proportion of SFO neurones have detectable Kv3.4-related IA.
Two previous studies using dissociated SFO neurones (Schmid, 1998) and slice preparations of the SFO (Ferguson & Li, 1996) have reported that IA-like currents are partially inhibited by TEA at 0.1 mM and completely blocked by TEA at 20 mM, respectively. By contrast, recent studies (Johnson et al. 1999; Washburn et al. 1999; Washburn & Ferguson, 2001) have reported that IAs in dissociated SFO neurones were resistant to TEA at 5–20 mM. Although this discrepancy in the effect of TEA on IA in SFO neurones was not resolved earlier, our results may explain it. The IC50 of TEA for Kv3.4-derived IA has been reported to be 0.3 mM (Rettig et al. 1992). It is thus not surprising that TEA at 0.1 mM partially inhibited IA-like currents in SFO neurones (Schmid, 1998). In the previous study showing complete blockage of IA-like currents by TEA at 20 mM (Ferguson & Li, 1996), the holding potential was –52 mV. The voltage may explain the discrepancy. Our voltage dependence of inactivation curves implies that the holding potential strongly inhibits TEA-resistant IA (Fig. 4C), while it does not affect the generation of a BDS-I-sensitive IA-like current (Fig. 2B). It thus seems that the IA-like current seen at a holding potential of –52 mV (Ferguson & Li, 1996) was probably a Kv3.4-related IA rather than a TEA-resistant IA, and it might be completely blocked by TEA even at 1 mM. On the other hand, at a holding potential lower than –70 mV, Kv3.4-related IA became a minor current in SFO neurones. Furthermore, as described above, not all SFO neurones displayed a Kv3.4-related IA. This may be why the recent studies (Johnson et al. 1999; Washburn et al. 1999; Washburn & Ferguson, 2001) failed to observe a highly TEA-sensitive IA in SFO neurones.
TEA-resistant IA
In conventional RT-PCR using SFO tissue, we detected both Kv1 and Kv4 channels that were thought to be the channels that generated the TEA-resistant IA. Since it was known that the recovery time constant from inactivation of the Kv1-derived IA is slow (of the order of 1 s), and that of the Kv4-derived IA is fast (of the order of 100 ms) (Frank An et al. 2000; Amberg et al. 2003), we examined the time course of the recovery from inactivation of the TEA-resistant IA and were able to divide SFO neurones into two types. The best-fit R values of recovery time course for F-type cells were 52 ms and 184 ms, and those for S-type were 177 ms and 1.1 s, although it has been reported that R value of IA in SFO neurones was near 50 ms (Washburn et al. 1999). Since the both values in F-type and that of the fast component in S-type were below 200 ms, the TEA-resistant IA in F-type and the fast component of TEA-resistant IA in S-type neurones were expected to be Kv4-derived. On the other hand, the slow component in S-type was expected to be Kv1-derived, because the R value was 1.1 s. Although there was a significant difference of the peak amplitudes of TEA-resistant IA between F- and S-type neurones, there was no difference in the peak amplitudes per cell membrane capacitance. This implies that there is a different population of the IA-generating Kv channels in single SFO neurones, rather than a variation in the density of Kv channels. It also indicates that the different populations of SFO neurones cannot be distinguished by cell size alone, as shown in Fig. 7D.
To examine the expression of Kv channel subunits in single SFO neurones, we performed single-cell RT-PCR after the electrophysiological recordings. In our hands the mRNA for Kv4.2 was detected in all SFO neurones but that for Kv4.3L was detected in only half of them. It is well known that the Kv4 family of channels generates low threshold currents (Nadal et al. 2001; Hatano et al. 2002). In the present study, 4-AP-sensitive and TEA-resistant IA was activated from near –70 mV at a holding potential of –100 mV in a Ca2+-free solution. Moreover, it has been reported that the voltage dependence of the Kv4.2-derived current is shifted by micromolar concentrations of extracellular Cd2+ in rat ventricular cells (Fiset et al. 1997) and neostriatal cholinergic interneurones (Song et al. 1998). Similarly, in the present study, the voltage dependence of 4-AP-sensitive and TEA-resistant IA showed a shift in Cd2+-containing solution. Both lines of electrophysiological evidence suggest that part of the TEA-resistant IA in SFO neurones is generated by the Kv4.2 channel, consistent with the result of single-cell RT-PCR. R values of Kv4.2- and Kv4.3 L-derived IA in transfected cells were reported to be approximately 200–500 ms and could be shortened to 50–130 ms by binding KChIP to the -subunits (Frank An et al. 2000; Nadal et al. 2001; Hatano et al. 2002). The range was similar to the R values of the fast components of both groups of cells in the present study. Because the mRNAs for KChIP1a and KChIP2 can be found in the SFO, channels of the Kv4 family may bind to KChIPs in SFO neurones. It is known that Kv4.2 binds to KChIP2 and Kv4.3 to KChIP1 (Frank An et al. 2000). From these results, it seems likely that the fast components of TEA-resistant IAs in F- and S-type cells are mainly generated by Kv4.2 and Kv4.3L channels that bind KChIPs.
We tested the possibility that Kv1.4 might be responsible for the slow component in S-type neurones. It has been reported that the R value of Kv1.4 is 1–2 s in normal solution (Rettig et al. 1994; Frank An et al. 2000) and is greatly prolonged by Cd2+ (up to 10 s) (Yellen et al. 1994). However, in the present study, the R value of the slow component in the S-type was 1.1 s in the Cd2+-containing solution. It is likely that this value is too small, when the prolongation of R values by Cd2+ is considered. Moreover, we detected the mRNA for Kv1.4 both in F-type neurones and in S-type neurones using single-cell RT-PCR. These results suggest that Kv1.4 is not an essential channel for generating the slow components in S-type cells. In anatomical studies, the Kv1.4 channel is specifically localized on axons and near synaptic terminals in the hippocampus (Cooper et al. 1998), although Kv4 channels are found primarily in postsynaptic membranes (Sheng et al. 1992). It would thus probably be difficult to record a Kv1.4-derived current from acutely dissociated neurones that have lost their axons. Further, other Kv1 family channels may be thought of as candidates that might mediate the slow components. Kv1.1, Kv1.2, Kv1.3 and Kv1.5 can generate TEA-resistant IA when they coexist with Kvß subunits and their Rs may be over 1 s (Rettig et al. 1994; Heinemann et al. 1996; Leicher et al. 1998). It is thus possible that these channels may be involved in the generation of the slow components of the IA in SFO neurones.
Roles of TEA-sensitive and TEA-resistant IA on action potential generation
Outward K+ channels play a major role in controlling resting membrane potentials and shaping action potentials. It is thought that Kv3 channels enable the fast repolarization of action potentials without compromising spike initiation due to high-threshold channels, and consequently determine repetitive firing at high frequencies (Rudy & McBain, 2001). In fact, it has been reported that Kv3 channels are expressed in native fast-spiking neurones (Baranauskas et al. 2003). Conversely, subthreshold-activating IA (presumed Kv4 channels) causes delayed excitation and slowed interspike interval as a result of increasing afterhyperpolarizations (Nadal et al. 2001; Amberg et al. 2003). In the present study, application of TEA at 1 mM completely blocked Kv3 channels, induced broadening of the action potentials and expanded the first interspike interval without changing the latency to the first spike in all neurones tested. On the other hand, application of 4-AP at 5 mM, in the presence of TEA at 1 mM, a protocol that was designed to suppress 4-AP-sensitive and TEA-resistant IA (presumed Kv4 and/or Kv1), shortened the latency to the first spike and the first interspike interval. These results are consistent with a general knowledge of the contribution of Kv3 and Kv4 channels to action potential generation, and also support the existence of these channels in SFO neurones. However, since we manually clamped the membrane potentials below –80 mV to avoid spontaneous firing and bursting in the present experiments, we cannot determine clearly the extent to which these channels modulate the firing frequency at resting membrane potentials.
Angiotensin II inhibits presumed Kv4-derived IA
Using the triple-pulse protocol, we found that ANG inhibited the second current more strongly than the first in all S-type neurones, although there was no such difference in a half of F-type neurones that were responsive to ANG. We think that the second current in S-type cells consists of a relatively pure fast recovery component when compared with the first current. This may imply that the fast component in S-type cells, as well as F-type cells, is inhibited by application of ANG. In other words, ANG inhibited presumed Kv4-derived IA in SFO neurones, regardless of whether they were F-type or S-type. Since it has been reported that protein kinase C inhibits Kv4-derived IA in rat ventricular myocytes (Nakamura et al. 1997), activation of ANG receptor in SFO neurones might inhibit IA by activating protein kinase C. However, we cannot exclude the involvement of other IA-generating channels in the ANG-induced response because we used TEA at 1 mM, which eliminated the Kv3.4-derived IA, in the experiments with ANG, and we did not fully investigate the slow component of S-type neurones.
In previous extracellular and patch-clamp studies, 66% and 63% of SFO neurones, respectively, were excited by ANG (Okuya et al. 1987; Ferguson et al. 1997), which is consistent with the proportion of ANG responsive cells that we found (67%, n = 10 of 15). Half of the F-type neurones were not sensitive to ANG, although most had a fast recovery IA. Either these neurones do not have ANG receptors in nature or they had lost the activation. Another possibility is that the pathways activated by ANG were already inactivated for some reason, and that the effect was occluded. In our previous study using the slice preparation (Ono et al. 2001), ANG dose-dependently increased the amplitude of the inward current and the number of neurones that responded (42% at 10 nM and 50% at 100 nM). On the other hand, in four neurones (27%), ANG at 30 nM induced a small inward current in the present study. The proportion of neurones showing detectable inward current following ANG application in dissociated SFO neurones was lower than that seen in the slice preparation. This may indicate that the procedure for dissociating the cells damages the mechanism that induces ANG-related inward current. Similarly, in other studies using dissociated SFO neurones, no ANG-induced inward current was observed, although there was inhibition of the IA-like current (Schmid, 1998). It is also worth noting that the threshold concentration of ANG for inhibition of the IA is probably lower than that for induction of the inward current.
In summary, we have demonstrated that some SFO neurones have a relatively small highly TEA-sensitive Kv3.4-derived IA and all neurones have TEA-resistant IAs. We have seen two different types (F- and S-types) of recovery from inactivation in TEA-resistant IAs, and at least some of the fast components in both types were probably derived from Kv4 family channels. Furthermore, we have demonstrated that ANG inhibited fast recovery components (presumed Kv4) of TEA-resistant IA in half of F-type and all S-type neurones.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Baranauskas G, Tkatch T, Nagata K, Yeh JZ & Surmeier DJ (2003). Kv3.4 subunits enhance the repolarizing efficiency of Kv3.1 channels in fast-spiking neurons. Nat Neurosci 6, 258–266.[CrossRef][Medline]
Boland LM, Jiang M, Lee SY, Fahrenkrug SC, Harnett MT & O'Grady SM (2003). Functional properties of a brain-specific NH2-terminally spliced modulator of Kv4 channels. Am J Physiol Cell Physiol 285, C161–C170.
Cooper EC, Milroy A, Jan YN, Jan LY & Lowenstein DH (1998). Presynaptic localization of Kv1.4-containing A-type potassium channels near excitatory synapses in the hippocampus. J Neurosci 18, 965–974.
Diochot S, Schweitz H, Béress L & Lazdunski M (1998). Sea anemone peptides with a specific blocking activity against the fast inactivating potassium channel Kv3.4. J Biol Chem 273, 6744–6749.
Ferguson AV, Bicknell RJ, Carew MA & Mason WT (1997). Dissociated adult rat subfornical organ neurons maintain membrane properties and angiotensin responsiveness for up to 6 days. Neuroendocrinology 66, 409–415.[Medline]
Ferguson AV & Li Z (1996). Whole cell patch recordings from forebrain slices demonstrate angiotensin II inhibits potassium currents in subfornical organ neurons. Regul Pept 66, 55–58.[CrossRef][Medline]
Fiset C, Clark RB, Shimoni Y & Giles WR (1997). Shal-type channels contribute to the Ca2+-independent transient outward K+ current in rat ventricle. J Physiol 500, 51–64.[Medline]
Frank An W, Bowlby MR, Betty M, Cao J, Ling HP, Mendoza G et al. (2000). Modulation of A-type potassium channels by a family of calcium sensors. Nature 403, 553–556.[CrossRef][Medline]
Hatano N, Ohya S & Imaizumi Y (2002). Functional interaction between KChIP1 and GFP-fused Kv4.3L co-expressed in HEK293 cells. Pflügers Arch 444, 80–88.[CrossRef][Medline]
Heinemann SH, Rettig J, Graack HR & Pongs O (1996). Functional characterization of Kv channel ß-subunits from rat brain. J Physiol 493, 625–633.[Medline]
Hiyama TY, Watanabe E, Ono K, Inenaga K, Tamkun MM, Yoshida S et al. (2002). Nax channel involved in CNS sodium-level sensing. Nat Neurosci 5, 511–512.[CrossRef][Medline]
Honda E, Ono K, Toyono T, Kawano H, Masuko S & Inenaga K (2003). Activation of muscarinic receptors in rat subfornical organ neurones. J Neuroendocrinol 15, 770–777.[Medline]
Johnson RF, Beltz TG, Jurzak M, Wachtel RE & Johnson AK (1999). Characterization of ionic currents of cells of the subfornical organ that project to the supraoptic nuclei. Brain Res 817, 226–231.[CrossRef][Medline]
Leicher T, Bahring R, Isbrandt D & Pongs O (1998). Coexpression of the KCNA3B gene product with Kv1.5 leads to a novel A-type potassium channel. J Biol Chem 273, 35095–35101.
Lien CC, Martina M, Schultz JH, Ehmke H & Jonas P (2002). Gating, modulation and subunit composition of voltage-gated K+ channels in dendritic inhibitory interneurones of rat hippocampus. J Physiol 538, 405–419.
Nadal MS, Amarillo Y, Vega-Saenz de Miera E & Rudy B (2001). Evidence for the presence of a novel Kv4-mediated A-type K+ channel-modifying factor. J Physiol 537, 801–809.
Nakagawa T, Shirasaki T, Wakamori M, Fukuda A & Akaike N (1990). Excitatory amino acid response in isolated nucleus tractus solitarii neurons of the rat. Neurosci Res 8, 114–123.[CrossRef][Medline]
Nakamura TY, Coetzee WA, Vega-Saenz De Miera E, Artman M & Rudy B (1997). Modulation of Kv4 channels, key components of rat ventricular transient outward K+ current, by PKC. Am J Physiol 273, H1775–H1786.[Medline]
Ohya S, Tanaka M, Oku T, Asai Y, Watanabe M, Giles WR et al. (1997). Molecular cloning and tissue distribution of an alternatively spliced variant of an A-type K+ channel -subunit, Kv4.3 in the rat. FEBS Lett 420, 47–53.[CrossRef][Medline]
Okuya S, Inenaga K, Kaneko T & Yamashita H (1987). Angiotensin II sensitive neurons in the supraoptic nucleus, subfornical organ and anteroventral third ventricle of rats in vitro. Brain Res 402, 58–67.[CrossRef][Medline]
Ono K, Honda E & Inenaga K (2001). Angiotensin II induces inward currents in subfornical organ neurones of rats. J Neuroendocrinol 13, 517–523.[CrossRef][Medline]
Pak MD, Baker K, Covarrubias M, Butler A, Ratcliffe A & Salkoff L (1991). mShal, a subfamily of A-type K+ channel cloned from mammalian brain. Proc Natl Acad Sci U S A 88, 4386–4390.
Renganathan M, Cummins TR, Hormuzdiar WN & Waxman SG (2000). -SNS produces the slow TTX-resistant sodium current in large cutaneous afferent DRG neurons. J Neurophysiol 84, 710–718.
Rettig J, Heinemann SH, Wunder F, Lorra C, Parcej DN, Dolly JO et al. (1994). Inactivation properties of voltage-gated K+ channels altered by presence of ß-subunit. Nature 369, 289–294.[CrossRef][Medline]
Rettig J, Wunder F, Stocker M, Lichtinghagen R, Mastiaux F, Beckh S et al. (1992). Characterization of Shaw-related potassium channel family in rat brain. EMBO J 11, 2473–2486.[Medline]
Rudy B & McBain CJ (2001). Kv3 channels: voltage-gated K+ channels designed for high-frequency repetitive firing. Trends Neurosci 24, 517–526.[CrossRef][Medline]
Schmid HA (1998). Effect of glutamate and angiotensin II on whole cell currents and release of nitric oxide in the rat subfornical organ. Amino Acids 14, 113–119.[Medline]
Sheng M, Tsaur ML, Jan YN & Jan LY (1992). Subcellular segregation of two A-type K+ channel proteins in rat central neurons. Neuron 9, 271–284.[CrossRef][Medline]
Song WJ, Tkatch T, Baranauskas G, Ichinohe N, Kitai ST & Surmeier DJ (1998). Somatodendritic depolarization-activated potassium currents in rat neostriatal cholinergic interneurons are predominantly of the A type and attributable to coexpression of Kv4.2 and Kv4.1 subunits. J Neurosci 18, 3124–3137.
Tseng GN, Jiang M & Yao JA (1996). Reverse use dependence of Kv4.2 blockade by 4-aminopyridine. J Pharmacol Exp Ther 279, 865–876.
Washburn DL, Beedle AM & Ferguson AV (1999). Inhibition of subfornical organ neuronal potassium channels by vasopressin. Neuroscience 93, 349–359.[CrossRef][Medline]
Washburn DL & Ferguson AV (2001). Membrane properties of subfornical organ neutrons. Clin Exp Pharmacol Physiol 28, 575–580.[CrossRef][Medline]
Yellen G, Sodickson D, Chen TY & Jurman ME (1994). An engineered cysteine in the external mouth of a K+ channel allows inactivation to be modulated by metal binding. Biophys J 66, 1068–1075.
|
Acknowledgements |
|---|
This article has been cited by other articles:
|
|
 |
|
  I. H. Zucker Novel Mechanisms of Sympathetic Regulation in Chronic Heart Failure Hypertension, December 1, 2006; 48(6): 1005 - 1011. [Full Text] [PDF]
|
|
|
|
 |
|
 E. Honda, K. Ono, S. Kataoka, and K. Inenaga Activation of subfornical organ neurons in rats through pre- and postsynaptic {alpha}-adrenoceptors Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2006; 290(6): R1646 - R1653. [Abstract] [Full Text] [PDF]
|
|
| ||||
) and IA- like (•) currents. Error bars show standard error.
