HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 7, 4945-4950, February 15, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/7/4945    most recent M109439200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ueno, T. Articles by Nabekura, J. Search for Related Content PubMed PubMed Citation Articles by Ueno, T. Articles by Nabekura, J. Social Bookmarking                      What's this?

Diversity of Neuron-specific K⁺-Cl Cotransporter Expression and Inhibitory Postsynaptic Potential Depression in Rat Motoneurons^*

Tsuyoshi Ueno, Akihito Okabe^§, Norio Akaike, Atsuo Fukuda^§, and Junichi Nabekura^¶

From the  Department of Cellular and System Physiology, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan and ^§ Department of Physiology, Hamamatsu University School of Medicine, 20-1 Handayama 1-chome, Hamamatsu, Shizuoka 431-3192, Japan

Received for publication, October 1, 2001, and in revised form, November 28, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Motoneurons receive a robust recurrent synaptic inhibition by -aminobutyric acid and glycine, which activate Cl channels. Thus, Cl homeostasis determines the efficacy of synaptic inhibition in the motoneurons. In situ hybridization reveals that the neuronal K⁺-Cl cotransporter isoform 2 (KCC2), a major mechanism in maintaining a low Cl concentration in neurons, is abundantly expressed in the facial, hypoglossal (XII), and spinal motoneurons innervating striated muscle, whereas the dorsal vagal motoneurons (DMVs) controlling smooth muscle exhibited little expression of KCC2. This raises a general interest in the correlation between KCC2 expression and inhibitory postsynaptic potential (IPSP) performance in the native circuits. Intracellular and whole-cell patch recordings revealed that an activity-dependent depression of IPSPs and positive shift of IPSP reversal potentials were more prominent in the DMV than in the XII. Cl influx through Cl channels was extruded more potently in the XII than in the DMV, suggesting that differences in Cl extrusion account for these dynamic differences of IPSP. Cl extrusion was inhibited by either furosemide or an increase in extracellular potassium concentrations. Thus, the rigid maintenance of IPSP and rapid Cl extrusion in the XII reflects an intense expression of KCC2. KCC2 expression may strongly influence the IPSP depression and functional properties of the motoneurons innervating striated muscles.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Motor neurons innervating striated muscles are generally known to receive robust recurrent inhibitory inputs mediated by -aminobutyric acid (GABA)¹-ergic and glycinergic neurons, such as Renshaw cells. These inhibitory transmitters activate Cl channels, resulting in hyperpolarization of the motoneurons. Thus, the efficacy of these inhibitions is largely influenced by intracellular Cl concentration ([Cl]_i) homeostasis. Among the many molecules involved in [Cl]_i homeostasis, K⁺-Cl cotransporter isoform 2 (KCC2) is a principal molecule for maintenance of low [Cl]_i in central neurons (1).

The distribution of Cl across neuronal membranes is not constant and is reflected in the variety of reversal potentials of Cl-mediated inhibitory postsynaptic potentials (E_IPSPs) reported in different neurons (2-7). Such differences importantly affect the characteristics of IPSPs, including their capacity to both hyperpolarize and shunt inhibition of postsynaptic neurons (8). The developmental shift of GABA responses from depolarization to hyperpolarization was first demonstrated in hippocampal neurons (9). [Cl]_i decreases developmentally as a result of changes in such Cl regulators as Na⁺-K⁺-Cl cotransporter (NKCC) isoform 1 and KCC2 (1, 10, 11). The developmental induction of KCC2 function evokes a marked negative shift in the reversal potential of GABA responses (E_GABA) in mature neurons and promotes fast hyperpolarizing GABAergic and glycinergic postsynaptic inhibition (1, 9, 12).

In addition to the developmental changes of GABA and glycine responses associated with changes in KCC and NKCC function, GABA type A (GABA_A) receptor-mediated responses differ regionally. In nucleus reticularis thalami neurons, GABAergic IPSPs induce inhibition mainly by shunting rather than by overt hyperpolarization, whereas in thalamocortical neurons, IPSPs are markedly hyperpolarizing (6). Such differences in GABA action reflect very different E_GABAs between thalamocortical and reticularis thalamus neurons. In dorsal root ganglion neurons, the high [Cl]_i maintained by NKCC1 causes depolarizing GABA responses (13). Thus, these regional differences in GABAergic action may indicate variations of [Cl]_i regulation across these brain nuclei.

A recent study demonstrated the essential role of KCC in synaptic inhibition. KCC2 knockout mice have severe motor deficits and abnormal neuronal activity of spinal motoneurons and die due to a failure of respiration (14). Thus, KCC2 is critically involved in the neuronal function of motor neurons innervating striated muscles. In the present study, we evaluated the differences in GABAergic IPSP performance in two groups of motoneurons, one innervating striated muscle and the other controlling smooth muscle, and compared their [Cl]_i regulation. Across these two pools of motoneurons, we found a close correspondence between differences in their Cl transport properties and activity-dependent IPSP depression along with their KCC2 expression. The motoneurons innervating striated muscles exhibit a potent [Cl]_i controlling system, resulting in the maintenance of constant IPSP efficacy.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Tissue Preparation-- Our experimental protocol was approved by the Ethics Review Committee for Animal Experimentation of the Japanese Physiological Society. Wistar rats, 16-18 days old, were decapitated under anesthesia with pentobarbitone sodium (55 mg/kg, intraperitoneal), and 400-µm transverse slices of the brain including the hypoglossal (XII) and dorsal vagal motor (DMV) nuclei were prepared with a microslicer (VT-1000S; Leica, Nussloch, Germany). For whole-cell patch clamp recording, XII and DMV neurons were dissociated as described in our previous report (15). To assure that the tissue was taken from the XII or DMV nucleus, the micropunches were taken from within the boundaries of the respective nuclei. The resulting isolated neurons retained several original morphological features resembling those described previously for XII and DMV neurons (16-19).

Solutions-- The incubation solution for slices contained 124 mM NaCl, 5 mM KCl, 1.2 mM KH₂PO₄, 1.3 mM MgSO₄, 2.4 mM CaCl₂, 10 mM glucose, and 24 mM NaHCO₃ (pH 7.45; 95% O₂:5% CO₂). The standard external solution used for gramicidin-perforated whole-cell patch clamp recordings contained 150 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 2 mM CaCl₂, 10 mM glucose, and 10 mM HEPES. The pH was adjusted to 7.4 with Tris-hydroxymethyl aminomethane (Tris-base). Because GABA_A receptor-gated chloride channels have a permeability to Cl and HCO, we measured E_GABA in the absence of HCO to estimate [Cl]_i. For our high K⁺ extracellular solution, KCl was substituted for equimolar NaCl. The micropipettes for intracellular recording were filled with 0.5 M potassium acetate. The patch pipette solution for gramicidin-perforated patch recording contained 150 mM KCl and 10 mM HEPES (pH 7.2 with Tris-base). Gramicidin was first dissolved in methanol to prepare a stock solution of 10 µg/ml and then diluted to a final concentration of 100 µg/ml for the pipette solution. The gramicidin-containing solution was prepared just before the experiment.

Electrophysiology-- For intracellular recording, the electrolyte filling the pipette (50-80 megaohms) was connected via Ag-AgCl electrode to the input stage of an intracellular recording amplifier (IR283; Neuro Data). IPSPs were evoked by a bipolar electrode placed near the nucleus of interest and activated by electrical pulses (100-µs duration) at 32 °C to 34 °C.

For gramicidin-perforated whole-cell recordings, ionic currents and voltage were measured with a patch clamp amplifier (EPC-7; List-Electronic) as described in the previous report (11). All experiments on isolated neurons were carried out at room temperature.

Drugs-- Drugs used in the present experiments were gramicidin D, 6-cyano-7-nitroquinoxaline-2,3-dione, and D,L-2-amino-5-phosphovaleric acid from Sigma, Pronase from Calbiochem, furosemide from Tokyo Kasei (Tokyo, Japan), and tetrodotoxin from Sankyo (Tokyo, Japan). Rapid change of the external solution was performed with "Y-tube" method described previously (20).

Statistical Analysis-- The data were presented as the means ± S.E., with n equal to the number of cells studied. Differences between the groups were analyzed for statistical significance using Student's t test. Values of p < 0.05 were taken as the standard for a significant difference.

In Situ Hybridization-- Frozen sections of brainstem from 16-day-old Wistar rats were prepared for in situ hybridization. To detect the KCC2 transcript, a 189-bp cDNA fragment of rat KCC2 (GenBank^TM accession number U55816; position 197-385) was amplified by reverse transcription-PCR from adult rat cerebral RNA (5' primer, TTCATCAACAGCACGGACAC; 3' primer, CTTCTTCTTTCCGCCCTCAT). Digoxigenin-labeled cRNA was used, and in situ hybridizations were performed as described previously (21).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

KCC2 mRNA Expression among Motor Neurons-- KCC2 is a neuron-specific K⁺-Cl cotransporter involved in active Cl extrusion (22). We assessed the expression levels of KCC2 among XII, facial nuclei, and spinal ventral horn as well as the DMV by in situ hybridization in rats (Fig. 1). KCC2 mRNA was expressed abundantly in very large neurons of spinal ventral horn, XII, and facial nuclei, whereas KCC2 expression is low in the DMV. This result supports the general hypothesis that motoneurons innervating striated muscle express KCC2 intensely.

View larger version (175K): [in this window] [in a new window]   Fig. 1.   KCC2 mRNA expression in XII, DMV, spinal, and facial motoneurons demonstrated by in situ hybridization using digoxigenin-labeled KCC2 antisense RNA probe. A, remarkable differences in KCC2 expression were demonstrated between XII and DMV neurons. The areas surrounded by dashed lines indicate the regions of XII and DMV nuclei. CC, central canal; scale bar, 100 µm. B, the large neurons in the facial nucleus and ventral horn of the spinal cord demonstrated high KCC2 expression compared with surrounding neurons. Scale bar: 100 µm, spinal cord; 300 µm, facial nucleus.

To examine whether the diversity of KCC2 expression correlates with the difference in Cl homeostasis and IPSP performance between the two sets of motor neurons, electrophysiological approaches were employed on XII neurons and the DMV.

Differential Patterns of IPSP Depression during Repetitive Stimulation in XII and DMV Neurons-- Local field stimulation evoked GABAergic IPSPs in both the XII and DMV neurons. These IPSPs were isolated pharmacologically by the presence of glutamatergic blockade with 10⁵ M 6-cyano-7-nitroquinoxaline-2,3-dione and 10⁵ M D,L-2-amino-5-phosphovaleric acid. Varying membrane potentials induced by passing current through the recording electrode altered the amplitude and direction of the IPSPs. E_IPSPs in the XII and DMV neurons were 73 ± 4 mV (n = 7) and 64 ± 3 mV (n = 6), respectively (Fig. 2). These very different E_IPSPs (p < 0.05) occurred despite no differences in the resting membrane potential between XII and DMV neurons (63 ± 4 mV (n = 7) and 62 ± 3 mV (n = 6), respectively).

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Depression of IPSP amplitude and change of EIPSP by repetitive stimulation (3.3 Hz) in XII and DMV neurons. A and B, hyperpolarization of IPSPs persisted during stimulation in the XII, although their amplitudes also gradually became smaller (A). The amplitudes of IPSP gradually and greatly decreased during stimulation in the DMV (B). Relative amplitude of each IPSP was calculated as a ratio to that of the first IPSP during repetitive stimulation. C and D, the amplitude of IPSPs before () and after repetitive stimulation () was plotted as a function of membrane potential in XII (C) and DMV neurons (D). EIPSP shifted more positive in the DMV neurons but not in XII neurons after repetitive stimulation. Stimulation was applied at the resting membrane potential in XII neurons (open arrowhead in C). On the other hand, membrane potential was depolarized by about 10 mV (closed arrowhead in D) just before repetitive stimulation by passing current in DMV neurons because EIPSPs before repetitive stimulation were too close to the resting membrane potential in DMV neurons (open arrowhead in D) to observe apparent hyperpolarization of IPSP. Membrane depolarization by about 10 mV during stimulation (Depo. in B) revealed obvious hyperpolarizing IPSPs. Vertical and horizontal bars for the inserted trace of IPSP (A) are 4 mV and 20 ms, respectively, which are applied to all IPSP traces in A and B.

Repetitive activation of the GABAergic afferents at 3.3 Hz for 30-60 s evoked characteristically different responses in the two types of neuron. During repetitive stimulation, IPSP amplitude gradually declined to reach a steady value during stimulation (Fig. 2, A and B) in both XII and DMV neurons. However, the degree of decline in IPSP amplitude was significantly greater in DMV neurons (23 ± 7% of the first IPSP; n = 6; Fig. 2B) than in XII neurons (66 ± 11% of the first IPSP; n = 7; Fig. 2A).

In the recovery phase, at 20 s after the end of repetitive stimulation (<1 min), E_IPSP averaged 72 ± 3 mV (n = 7) in XII neurons and 56 ± 4 mV (n = 6) in DMV neurons (Fig. 2, C and D). E_IPSPs before and after repetitive stimulation were different in DMV (p < 0.05), but not in XII (p > 0.1). Multiple possible mechanisms, such as transmitter release probability and GABA_A receptor desensitization, might contribute to this difference in IPSP depression during repetitive stimulation. However, a depolarizing shift of E_IPSP could account for the rapid loss of efficacy of the IPSP that was greater in the DMV than in XII neurons. In the presence of 1 mM furosemide, E_IPSP shifted significantly to a more depolarized level even in the XII (8.1 ± 3.1 mV; n = 3).

Changes in [Cl]_i during Repetitive GABA Applications-- To examine the potential mechanisms underlying the differences in E_IPSP performance between XII and DMV neurons, we assessed [Cl]_i using gramicidin-perforated patch recordings of acutely dissociated XII and DMV neurons. E_GABAs were measured using voltage ramps (from 100 to 0 mV, 2-s duration) applied before and during 10 µM GABA application starting from a holding potential (V_H) of 50 mV (Fig. 3A, right panel). GABA induced outward currents at a V_H of 50 mV in both XII and DMV neurons (Fig. 3A, left panel). GABA-induced currents (I_GABAs) were eliminated by bicuculline (1 µM), a GABA_A receptor antagonist, in all neurons. As a result, I_GABA was found to be a Cl current conducted through GABA_A receptors. Cl moves through GABA-activated channels according to its electrochemical potential across the membrane. The recovery of the neuron from such fluxes to its resting state occurs due to various means of Cl transport. We repetitively applied 10 µM GABA to XII and DMV neurons at an interval of 5 min and measured the amplitude of each I_GABA. In DMV neurons, the amplitude of I_GABA gradually declined, and I_GABA was nearly completely depressed at the fifth GABA application (Fig. 3A, left panel). In contrast, I_GABA was maintained at an almost constant amplitude in XII neurons.

View larger version (37K): [in this window] [in a new window]   Fig. 3.   Changes in the [Cl]i by repetitive GABA application in XII and DMV neurons. A, left panel, GABA-induced outward currents in XII and DMV neurons under gramicidin-perforated patch recording at the first (left trace) and fifth (right trace) applications. GABA was applied at an interval of 5 min. In all of the following experiments, transient vertical lines before (i) and during (ii) GABA-induced currents (IGABA) are the current responses to steps in ramp voltage. A, right panel, EGABA was determined as a membrane potential at which current-voltage relationships obtained in the resting conductance (i) and resting plus GABA-activated conductance (ii) intersected with each other. Voltage clamp recording was performed at a holding potential (VH) of 50 mV unless noted otherwise. B, increase of [Cl]i ([Cl]i) over that measured at the first GABA application was plotted as a function of time (n = 3). C, changes of [Cl]i during repetitive GABA application at various intervals were plotted as a function of GABA trial in XII and DMV neurons ( and , respectively). The interval of GABA application is indicated to the right of each plot.

E_GABA was assumed to equal the Cl equilibrium potential (E_Cl), and this allowed us to calculate the [Cl]_i by using the Nernst equation using our measured E_GABA and known extracellular Cl concentration ([Cl]_o = 161 mM) in each neuron (11, 23). The calculated [Cl]_i shifted after each GABA application, but the magnitude of the change was very different between XII and DMV neurons (Fig. 3B). The changes in [Cl]_i between the first and fifth GABA applications were 0.35 ± 0.39 mM (n = 4; p > 0.1) and 4.4 ± 0.12 mM (n = 4; p < 0.01) at XII and DMV neurons, respectively. This indicates that although the GABA-associated Cl influx leads to a net accumulation of intracellular Cl (minimizing the steady-state Cl gradient across the membrane), this loaded Cl was extruded more effectively across from XII neurons to allow recovery within the 5-min test cycle. However, increasing the rate of repetitive GABA application to 1-min intervals effectively raised [Cl]_i even in XII neurons (Fig. 3C, ). Conversely, decreasing the interval to 20 min allowed DMV neurons to maintain an almost constant [Cl]_i (Fig. 3C, ).

To more directly examine the efficacy of Cl extrusion, we measured the recovery of E_GABA using a defined level of loading Cl. Increasing the duration of application of 3 × 10⁴ M GABA (approximately 1 min) brought E_GABA close to V_H (50 mV) and resulted in diminished GABA responses in both XII and DMV neurons (Fig. 4, A and B, i). At 5 min after GABA application, E_GABA achieved more negative values in the XII than in the DMV (Fig. 4B, ii). The calculated [Cl]_i reduction 5 min after Cl loading was significantly greater in the XII (6.36 ± 0.82 mM; n = 7) than in the DMV (2.23 ± 0.91 mM; n = 7; p < 0.05, unpaired t test). Such results are consistent with the substantial difference in the capacity for Cl extrusion in these two groups of neurons.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Differences in Cl extrusion efficiency in XII and DMV. A and B, long (approximately 1-min) application of 3 × 104 M GABA gradually decreased IGABA and finally brought EGABA to VH (50 mV, B) in both neurons (measured at (i)). Five min later after washing out GABA (ii), an apparent GABA response reappeared, and EGABA was more hyperpolarized than VH in XII (B).

Furosemide-sensitive and K+-dependent Cl Transport-- In XII neurons, the amplitude of I_GABA was maintained at a constant level during repetitive GABA applications at an interval of 5 min (Fig. 3A, left panel). However, in the presence of 1 mM furosemide, the amplitude of I_GABA decreased rapidly (Fig. 5A, top panel). This decrease in I_GABA resulted from a marked shift of E_GABA toward V_H (50 mV; Fig. 5B, top panel). At DMV neurons, furosemide also shifted E_GABA toward V_H (n = 3). These results indicate that furosemide-sensitive Cl transport mechanisms play a major role in regulating [Cl]_i in both XII and DMV neurons. The differences in Cl regulation between both types of neurons could be accounted for by differences in the expression of furosemide-sensitive mechanisms, i.e. cation-Cl cotransporters.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Effects of furosemide and increased [K+]o. A, changes of IGABAs in the presence of 1 mM furosemide (open bar) and 30 mM K+ (solid bar). GABA was applied at an interval of 5 min. Note that IGABA became inward in 30 mM [K+]o. B, the changes in EGABA are plotted as a function of time before, during, and after application of furosemide (top panel) and high K+ (bottom panel; , DMV;  and , XII). One representative example is shown in this figure.

To verify that a high expression of KCC2 reflects low [Cl]_i in XII neurons, we examined the effect of extracellular K⁺ on Cl regulation. The efficacy and direction of KCC activity are dependent on the driving force for K⁺ across the membrane (5, 11, 24). In XII neurons, increasing the extracellular K⁺ concentration ([K⁺]_o) from 5 to 30 mM promptly reduced the amplitude of I_GABA and finally reversed GABA responses inward at a V_H of 50 mV (Fig. 5A, bottom panel). E_GABA shifted to more depolarized values from 70 mV with 5 mM K⁺ to 43 mV at the third GABA application with 30 mM K⁺ in response to the associated reversal of the driving force for Cl (Fig. 5B, bottom panel, ). However, 20 mM [K⁺]_o simply reduced the amplitude of I_GABA but failed to reverse the GABA response and driving force for Cl (Fig. 5B, bottom panel, ). Returning to [K⁺]_o of 5 mM promptly returned E_GABA to a value similar to that seen before the increased [K⁺]_o. Such results are consistent with the participation of the K⁺-dependent and furosemide-sensitive [Cl]_i extrusion of KCC. Although a similar alteration of E_GABA by [K⁺]_o manipulation was observed in DMV neurons, the effect of [K⁺]_o change on E_GABA appeared to be greater in the XII (Fig. 5B, bottom panel).

Resting [Cl]_i of XII and DMV Neurons-- Resting values for E_IPSP (Fig. 2, C and D) and E_GABA were more negative in the XII than in the DMV. E_GABAs measured with the first application of GABA after dissociation were 81.4 ± 2.8 mV (n = 8) in XII and 62.2 ± 3.6 mV (n = 8) in DMV. Calculated resting [Cl]_i for XII neurons was significantly lower than that of DMV neurons (5.6 ± 0.6 mM (n = 8) and 12.5 ± 1.9 mM (n = 8), respectively; p < 0.01). However, the different steady-state [Cl]_i could not be accounted for simply by differing KCC function. The E_Cl driven only by KCC was calculated to be 85 mV in both types of neurons under our experimental conditions, in which intracellular K⁺ concentration ([K⁺]_i), [K⁺]_o, and [Cl]_o were 150, 5, and 161 mM, respectively (25).

Increases in [Cl]_i by GABA_A receptor activation shifted E_GABA to approach V_H (60 mV) in the presence of furosemide and resulted in diminished GABA responses (Fig. 6A, i). Thereafter, V_H was stepped to 40 mV. Significant increases of E_GABA at 5 min after this voltage jump and in the presence of furosemide were observed in both groups of neurons (p < 0.01, paired t test; Fig. 6B). Such results indicate the existence of furosemide-insensitive [Cl]_i regulation, including passive Cl conductance, in both neurons that brings E_Cl toward V_H.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Furosemide-insensitive Cl regulation in XII and DMV. A, IGABA was nearly abolished during repetitive application of GABA in the presence of 1 mM furosemide at a VH of 60 mV (i), and then VH was immediately altered to a more depolarized level (40 mV). Five min after VH change (40 mV), we observed GABA-induced current and measured EGABA (ii). B, furosemide-insensitive Cl regulator shifted EGABA toward VH in both types of neurons (n = 4).

Together, the results suggest a diversity of KCC2 function and different resting states for [Cl]_i between the XII and DMV. This seems to be the result of different balances between Cl extrusion by KCC2 and Cl accumulation by furosemide-insensitive mechanisms. More active KCC2 function could contribute to a more negative set point of resting E_Cl in the XII.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

[Cl]_i and Cl Homeostasis in Two Sets of Motor Neurons-- KCC, NKCC, Cl-HCO exchange, Na⁺-dependent Cl-HCO exchange, Cl-ATPase, and passive Cl conductance are involved in [Cl]_i regulation of neurons (26, 27). In our experiments, we used HEPES-buffered solutions without HCO. Thus, HCO-dependent Cl regulators are likely to be negligible contributors under the conditions of the present studies. In the present study, E_GABA was brought close to 53.3 ± 0.1 mV (n = 3) with 20 mM [K⁺]_o in the XII neurons (Fig. 5B, bottom panel). This value (E_GABA = 53.3 mV) was similar to V_H (50 mV) and the calculated value of E_Cl regulated only by KCC with 20 mM [K⁺]_o and 150 mM [K⁺]_i (51 mV). In addition, increasing [K⁺]_o to 30 mM reversed GABA responses inward (Fig. 5A), a finding that could be explained by the reversed direction of Cl transport by KCC (24). Thus, our results indicate that the furosemide-sensitive and K⁺-dependent Cl extrusion mechanism observed in both neurons is mainly KCC. In evaluating the activity of extrusion of accumulated Cl, repetitive GABA application at an interval of 5 min gradually increased [Cl]_i in the DMV neurons, but not in the XII neurons (Fig. 3B). With longer intervals of GABA application, an increase of [Cl]_i was less apparent at DMV neurons (Fig. 3C). Thus, we suggest that a furosemide- and [K⁺]_o-sensitive extrusion Cl mechanism also exists in the DMV neuron but is less active than that in the XII. In addition to regional diversity of [Cl]_i regulation (Figs. 2 and 3), frequency-dependent [Cl]_i shift by repetitive Cl channel activation is also demonstrated in developing neurons, in which a young neuron with a less efficient Cl regulation mechanism needs a longer interval to maintain original [Cl]_i than mature neurons (28).

Relation between Cl Extrusion and KCC2 mRNA Expression-- KCC2 is specifically localized to neurons and displays unique functional characteristics, including a lack of swelling activation and a high affinity for external K⁺ (22, 29, 30). The NKCC isoform expressed in the brain is NKCC1 (31-33). In the present study, KCC2 mRNA was more prominently expressed in XII neurons than in DMV neurons (Fig. 1). On the other hand, the expression level of NKCC1 mRNA was lower in both neuron types (data not shown). The possible mechanisms for the apparent difference in activity of Cl extrusion between XII and DMV neurons are: 1) differences in the density of the molecule (KCC2) expressed in the membrane, and 2) differences in the activity of each molecule to extrude Cl. The activity of KCC1 is regulated by protein phosphorylation (34, 35). Whether neuronal KCC, KCC2, is affected by phosphorylation in function is controversial. In Xenopus oocytes, tyrosine phosphorylation does not affect KCC2 function (36), whereas tyrosine kinase may regulate Cl extrusion in cultured hippocampal neurons (37). Our electrophysiological and in situ hybridization results suggest that the amount of KCC2 expression appears to correlate with the functional activity of Cl extrusion. XII neurons possess more active Cl extrusion driven by more abundant KCC2 molecules than DMV neurons.

A Variety of IPSP Properties among Neurons-- GABAergic and glycinergic IPSP inhibition varies in such properties as amplitude and duration. One of the postsynaptic mechanisms that affect the properties of IPSPs is the difference between membrane potential and E_Cl. We observed differing decreases in GABAergic IPSP magnitude during repetitive stimulation in XII and DMV neurons (Fig. 2). The depression of GABAergic IPSPs might be accounted for by several mechanisms, i.e. an accumulation of [Cl]_i, an elevation of [K⁺]_o that may decrease the driving force for KCC (5), a decrease in the presynaptic GABA release induced by GABA_B receptor (38), or a desensitization of postsynaptic GABA_A receptor (39, 40). However, the present study suggests that greater subsynaptic [Cl]_i (decreased driving force) brought about by less Cl extrusion (less KCC2) might contribute to rapid IPSP depression in the DMV because KCC2 is well colocalized with GABA_A receptor in the neurons (29).

Differences in excitable properties between DMV and XII motoneurons have been reported. DMV neurons control smooth muscle activity with a transient K⁺ current, resulting in a long after hyperpolarization, thus leading to a slow reduction of the firing rate. On the other hand, XII neurons innervating striated muscle lack this transient K⁺ current, and this favors a higher frequency firing without any delay upon depolarization (41). In the study using KCC2 knockout mice, KCC2 is vital for motoneurons to maintain normal striated muscle activity (14). In the present studies, XII neurons with more KCC activity displayed more hyperpolarization of GABAergic responses and less depression of IPSP during repetitive GABAergic afferent stimulation than DMV neurons. Together with high expression of KCC mRNA in facial and spinal motoneurons, a larger hyperpolarization and less depression of GABAergic IPSPs might contribute to a rapid and constant on-off of firing in motoneurons driving striated muscles.

	ACKNOWLEDGEMENTS

We thank Profs. M. C. Andressen and K. Kaila for critical discussion. In addition, the excellent technical support of Dr. A. Furuta is gratefully acknowledged.

	FOOTNOTES

^* This work was supported by Grants-in-aid for Scientific Research 13210108 on Advanced Brain Research and 13035036 on Integrated Brain Research (to J. N.) from the Ministry of Education, Science and Culture, Japan.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EBI Data Bank with accession number(s) U55816.

^¶ To whom correspondence should be addressed. Tel.: 81-92- 642-6090; Fax: 81-92-642-6094; E-mail: Nabekura@mailserver.med.kyushu-u.ac.jp.

Published, JBC Papers in Press, December 3, 2001, DOI 10.1074/jbc.M109439200

	ABBREVIATIONS

The abbreviations used are: GABA, -aminobutyric acid; IPSP, inhibitory postsynaptic potential; E_IPSP, reversal potential of IPSP; GABA_A, GABA type A; I_GABA, GABA-induced current; E_GABA, reversal potential of I_GABA; KCC, K⁺-Cl cotransporter; NKCC, Na⁺-K⁺-Cl cotransporter; XII, hypoglossal; DMV, dorsal vagal motor; V_H, holding potential; E_Cl, Cl equilibrium potential; [Cl]_i, intracellular Cl concentration; [Cl]_o, extracellular Cl concentration; [K⁺]_i, intracellular K⁺ concentration; [K⁺]_o, extracellular K⁺ concentration; KCC2, K⁺-Cl cotransporter isoform 2.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				G. Huberfeld, L. Wittner, S. Clemenceau, M. Baulac, K. Kaila, R. Miles, and C. Rivera Perturbed Chloride Homeostasis and GABAergic Signaling in Human Temporal Lobe Epilepsy J. Neurosci., September 12, 2007; 27(37): 9866 - 9873. [Abstract] [Full Text] [PDF]

				H. Wake, M. Watanabe, A. J. Moorhouse, T. Kanematsu, S. Horibe, N. Matsukawa, K. Asai, K. Ojika, M. Hirata, and J. Nabekura Early Changes in KCC2 Phosphorylation in Response to Neuronal Stress Result in Functional Downregulation J. Neurosci., February 14, 2007; 27(7): 1642 - 1650. [Abstract] [Full Text] [PDF]

				Y. Isomura, M. Sugimoto, Y. Fujiwara-Tsukamoto, S. Yamamoto-Muraki, J. Yamada, and A. Fukuda Synaptically Activated Cl- Accumulation Responsible for Depolarizing GABAergic Responses in Mature Hippocampal Neurons J Neurophysiol, October 1, 2003; 90(4): 2752 - 2756. [Abstract] [Full Text] [PDF]

				R. A. Wardle and M.-m. Poo Brain-Derived Neurotrophic Factor Modulation of GABAergic Synapses by Postsynaptic Regulation of Chloride Transport J. Neurosci., September 24, 2003; 23(25): 8722 - 8732. [Abstract] [Full Text] [PDF]

				C. Vale, J. Schoorlemmer, and D. H. Sanes Deafness Disrupts Chloride Transporter Function and Inhibitory Synaptic Transmission J. Neurosci., August 20, 2003; 23(20): 7516 - 7524. [Abstract] [Full Text] [PDF]

				J. Nabekura, T. Ueno, A. Okabe, A. Furuta, T. Iwaki, C. Shimizu-Okabe, A. Fukuda, and N. Akaike Reduction of KCC2 Expression and GABAA Receptor-Mediated Excitation after In Vivo Axonal Injury J. Neurosci., June 1, 2002; 22(11): 4412 - 4417. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/7/4945    most recent M109439200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ueno, T. Articles by Nabekura, J. Search for Related Content PubMed PubMed Citation Articles by Ueno, T. Articles by Nabekura, J. Social Bookmarking                      What's this?