KCC2 knockdown impairs glycinergic synapse maturation in cultured spinal cord neurons.

Synaptic inhibition in the spinal cord is mediated mainly by strychnine-sensitive glycine (GlyRs) and by γ-aminobutyric acid type A receptors (GABAAR). During neuronal maturation, neonatal GlyRs containing α2 subunits are replaced by adult-type GlyRs harboring α1 and α3 subunits. At the same time period of postnatal development, the transmembrane chloride gradient is changed due to increased expression of the potassium-chloride cotransporter (KCC2), thereby shifting the GABA- and glycine-mediated synaptic currents from mostly excitatory depolarization to inhibitory hyperpolarization. Here, we used RNA interference to suppress KCC2 expression during in vitro maturation of spinal cord neurons. Morphological analysis revealed reduced numbers and size of dendritic GlyR clusters containing α1 subunits but not of clusters harboring neonatal α2 subunits. The morphological changes were accompanied by decreased frequencies and amplitudes of glycinergic miniature inhibitory currents, whereas GABAergic synapses appeared functionally unaltered. Our data indicate that KCC2 exerts specific functions for the maturation of glycinergic synapses in cultured spinal cord neurons.

Neuronal chloride transport tuning.

BACKGROUND: Epilepsy is characterised by disturbed neuronal activity in the brain rendering it more susceptible to seizures. An understanding of the molecular mechanisms by which the balance between excitability and inhibition in neuronal networks is controlled will help to devise better treatment options. Hyperpolarising synaptic inhibition through GABAA (γ aminobutyric acid type A) and glycine receptors depends on the presence of the neuronal cation-chloride-cotransporter protein, KCC2. Several transcriptional and post-transcriptional mechanisms have been shown to regulate KCC2 and thereby affect the polarity and efficacy of inhibitory synaptic transmission. However, it is unknown whether regulation of KCC2 enables the transporter to attain different levels of activity, thus allowing a neuron to modulate the strength of inhibitory synaptic transmission to its changing requirements. We therefore investigated whether phosphorylation can allow KCC2 to achieve distinct levels of intracellular chloride ion concentrations in neurons. METHODS: A variety of KCC2 alanine dephosphorylation mimics were created and NH4(+)-induced pHi shifts were used in cultured hippocampal neurons to quantify the rate of KCC2 transport activity exhibited by these mutants. The association between KCC2 transport strength and GABAA receptor-mediated current amplitudes was investigated by performing gramicidine perforated-patch recordings. The correlation between reversal potential of GABAergic currents (EGABA) and NH4(+)-induced pHi shifts enabled an estimate of the range of chloride extrusion possible by kinase-phosphatase regulation of KCC2. Finally, we used the Goldman-Hodgkin-Katz equation to examine how EGABA would vary with increasing concentrations of extracellular K(+) in neurons expressing KCC2 mutants with different rates of transport. FINDINGS: KCC2 transport strength varied considerably in magnitude (from -0·02 to -1·00 pHi shifts) depending on the combination of alanine mutations present on the protein. KCC2 transport strength determined the direction and magnitude of GABAA receptor-mediated current amplitudes and was observed to have a linear correlation with the reversal potential of GABAergic currents. INTERPRETATION: Our findings highlight the considerable potential for regulating the inhibitory tone by KCC2-mediated chloride extrusion. Transport can be enhanced to sufficiently high levels that hyperpolarising GABAA responses can be obtained even at high extracellular K(+) concentrations and in neurons with an extremely negative resting membrane potential. We conclude that cellular signalling pathways might act together to alter the state of KCC2 phosphorylation and dephosphorylation and thereby tune the strength of synaptic inhibition. FUNDING: Royal Society.

Could tuning of the inhibitory tone involve graded changes in neuronal chloride transport?

Hyperpolarizing synaptic inhibition through GABAA and glycine receptors depends on the presence of the neuronal cation-chloride-cotransporter protein, KCC2. Several transcriptional and post-transcriptional mechanisms have been shown to regulate KCC2 and thereby influence the polarity and efficacy of inhibitory synaptic transmission. It is unclear however whether regulation of KCC2 enables the transporter to attain different levels of activity thus allowing a neuron to modulate the strength of inhibitory synaptic transmission to its changing requirements. We therefore investigated whether phosphorylation can allow KCC2 to achieve distinct levels of [Cl(-)]i in neurons. We generated a variety of KCC2 alanine dephosphorylation mimics and used NH4(+)-induced pHi shifts in cultured hippocampal neurons to quantify the rate of KCC2 transport activity exhibited by these mutants. To explore the relationship between KCC2 transport and GABAA receptor-mediated current amplitudes we performed gramicidine perforated-patch recordings. The correlation between EGABA and NH4(+)-induced pHi shifts enabled an estimate of the range of chloride extrusion possible by kinase/phosphatase regulation of KCC2. Our results demonstrate that KCC2 transport can vary considerably in magnitude depending on the combination of alanine mutations present on the protein. Transport can be enhanced to sufficiently high levels that hyperpolarizing GABAA responses may be obtained even in neurons with an extremely negative resting membrane potential and at high extracellular K(+) concentrations. Our findings highlight the significant potential for regulating the inhibitory tone by KCC2-mediated chloride extrusion and suggest that cellular signaling pathways may act combinatorially to alter KCC2 phosphorylation/dephosphorylation and thereby tune the strength of synaptic inhibition.

Intracellular acidification in neurons induced by ammonium depends on KCC2 function.

The Cl(-)-extruding neuron-specific K(+)-Cl(-) cotransporter KCC2, which establishes hyperpolarizing inhibition, can transport NH(4) (+) instead of K(+). It is, however, not clear whether KCC2 provides the only pathway for neuronal NH(4) (+) uptake. We therefore investigated NH(4) (+) uptake in cultured rat brain neurons. In neurons cultured for > 4 weeks, the response to NH(4)Cl applications (5 mM) consisted of an alkaline shift which reversed to an acid shift within seconds. Rebound acid shifts which followed brief applications of NH(4)Cl were blocked by furosemide (100 microM). They were rather insensitive to bumetanide (1 and 100 microM), in contrast to those induced in cultured glial cells. Rebound acid shifts persisted in the presence of 1 mM Ba(2+) and in Na(+)-free solution but were inhibited by extracellular K(+). In neurons with depolarizing GABA responses, indicating the absence of functional KCC2, applications of NH(4)Cl barely induced an acidosis. However, large rebound acid shifts occurred in neurons that had changed their GABA response from Ca(2+) increases to Ca(2+) decreases. Rebound acid shifts continued to increase even after the change in the GABA response had occurred and could be induced earlier in neurons transfected with KCC2 cDNA. We conclude that KCC2 provides the main pathway for fast neuronal NH(4) (+) uptake. Therefore, NH(4)Cl-induced rebound acid shifts can be used to indicate the development of KCC2 function. Further, the well known up-regulation of KCC2 function during development has the inevitable consequence of opening a major pathway for NH(4) (+) influx, which can be relevant under pathophysiological conditions.

Hyperpolarizing inhibition develops without trophic support by GABA in cultured rat midbrain neurons.

During a limited period of early neuronal development, GABA is depolarizing and elevates [Ca2+]i, which mediates the trophic action of GABA in neuronal maturation. We tested the attractive hypothesis that GABA itself promotes the developmental change of its response from depolarizing to hyperpolarizing (Ganguly et al. 2001). In cultured midbrain neurons we found that the GABA response changed from depolarizing to hyperpolarizing, although GABAA receptors had been blocked throughout development. In immature neurons prolonged exposure of the cells to nanomolar concentrations of GABA or brief repetitive applications of GABA strongly diminished the elevation of [Ca+]i by GABA. As revealed by gramicidin perforated-patch recording, reduced [Ca2+]i responses were due to a diminished driving force for Cl-. This suggests that immature neurons do not have an efficient inward transport that can compensate the loss of cytosolic Cl-resulting from sustained GABAA receptor activation by ambient GABA. Transient increases in external K+, which can induce voltage-dependent Cl- entry, restored GABA-induced [Ca2+]i elevations. In mature neurons, GABA reduced [Ca2+]i provided that background [Ca2+]i was elevated by the application of an L-type Ca2+ channel agonist. This was probably due to a hyperpolarization of the membrane by Cl- currents. K(+)-Cl- cotransport maintained the gradient for hyperpolarizing Cl-currents. We conclude that in immature midbrain neurons an inward Cl- transport is not effective although the GABA response is depolarizing. Further, GABA itself is not required for the developmental switch of GABAergic responses from depolarizing to hyperpolarizing in cultured midbrain neurons.

KCC2 mediates NH4+ uptake in cultured rat brain neurons.

Elevated levels of NH4+ in the brain impair neuronal function. We studied the effects of NH4+ on postsynaptic inhibition of cultured rat brain neurons using whole cell recording under nominally HCO3- -free conditions. Application of NH4+ shifted the reversal potentials for spontaneous inhibitory postsynaptic currents and currents elicited by dendritic GABA applications in a positive direction because [Cl-]i increased. The positive shift of the reversal potentials of GABA-induced Cl- currents was equal on equimolar elevation of [NH4+]o or [K+]o, respectively. The NH4+-induced increase in [Cl-]i was reversed by an inhibitor of cation-anion cotransport, furosemide (0.1 mM), but not by bumetanide (0.01 mM) or by replacement of [Na+]o by Li+. We conclude that neuron-specific K-Cl cotransporter (KCC2) transports NH4+ similar to K+. Despite this fact, the small increase of [NH4+]o during metabolic encephalopathies will barely elevate [Cl-]i. However, an impairment of neuronal function may result because KCC2 provides a pathway to accumulate NH4+, and thereby, a continuous acid load to neurons.