| Commentary Cellscience Reviews Vol 1 No.1 ISSN 1742-8130 |
|
|
Introduction
Until recently the acronym GABA had become synonymous with inhibitory neurotransmission. A fresh spate of reports however, has made this generalization feel increasingly uncomfortable. Even a decade ago, inhibition in the CNS was a seemingly simple affair, with synaptic inputs from GABAergic and Glycinergic neurons providing a calming counterbalance to excitation through glutamatergic, cholinergic and serotoninergic influences. GABA receptor classification was a matter of A, B and C, all of which were believed inhibitory in their mechanism (Bormann, 2000). Excepting further consideration of the metabotropic GABAB receptor, which mediates inhibition through the activation of K+ channels and the suppression of Ca2+ channel activity (Bowery & Enna, 2000; Couve et al., 2000), GABAA and GABAC receptors are ligand-operated ion channels which regulate the flow of Cl- (and HCO3-) ions across the cell membrane. As under most experimental recording conditions the intracellular Cl- concentration [Cl-]i, or activity, is below the predicted electrochemical equilibrium potential†, GABAA and GABAC receptors are assumed to exert an inhibitory influence by mediating the GABA-gated influx of negatively charged Cl- ions.
Intracellular Cl- activity is however the great variable of cellular electrochemistry. Whilst cells ubiquitously maintain large electrochemical gradients for Na+ and Ca2+, and the equilibrium (reversal) potential for K+ is usually near or beyond (i.e. more negative than) the observed resting membrane potential of the cell*, commonly encountered Cl- reversal potentials fall well within the operational range of membrane potentials (-80 to +80mV) observed in most excitable cells. Furthermore it is easier and more energy efficient for a cell to establish and maintain a [Cl-]i which is above or below the predicted Cl- electrochemical equilibrium, than it is to maintain the high Na+, K+ and Ca2+ gradients using energy hungry Na+/K+ and Ca2+ ATPases. Thus Cl- provides cells with a flexible alternative in regulating their excitability and in the generation of functional diversity within the nervous system. Whilst it is well known that intracellular chloride is accumulated above electrochemical equilibrium within fluid and electrolyte secreting epithelia (Walters et al., 1992) by means of a Na+K+2Cl- cotransporter (O’Brien et al., 1993), many reports are now appearing to suggest that [Cl-]i is also accumulated above equilibrium within a variety of cell types which functionally express GABAA receptors.
GABA excitation and cortical development
During early development [Cl-]i is elevated above equilibrium within immature neurons, causing GABAA receptor activation to be depolarizing (Ben-Ari, 2002). In fact GABA assumes a depolarizing role in the early development of many types of neurons (Leinekugel et al., 1999), including those of the lateral superior olive (Balakrishnan et al., 2003). Within immature neocortical neurons this is achieved by increasing early expression of the Cl- accumulating Na+K+2Cl- cotransporter (NKCC1), and by delaying the expression of the Cl- extruding K+-Cl- cotransporter (KCC2, Ben-Ari, 2002; Yamada et al., 2004). Indeed GABAergic synapses are often formed before glutamatergic contacts, allowing developing neurons to be become excitable and thereby permitting growth and synapse formation (Ben-Ari, 2002). By so delaying the formation of inhibitory influences, cortical networks may later be fashioned in a competitive and activity-dependent manner.
GABA - a bipolar neurotransmitter?
Whilst it has been known for some time that GABA may be depolarizing at early stages of development (Cherubini et al., 1991), many different types of fully differentiated adult cells have now been shown to be excited by GABA. Indeed a cluster of recent studies has firmly demonstrated that this phenomenon is not restricted to developing neurons, and may be plastic in nature. Many findings have demonstrated that GABA has a depolarizing action upon mature cortical neurons (Stein & Nicoll, 2003). For instance GABAA may transiently act as an excitatory (depolarizing) transmitter after intense bursts of GABAA receptor activation within the adult brain. This appears to be due to an activity-dependent shift in the Cl- reversal potential, for example in mature hippocampal CA1 pyramidal cells receiving intense synaptic inputs (Isomura et al., 2003). Such GABAergic excitation participates in the expression of seizure-like rhythmic synchronization (“afterdischarge”) in the mature hippocampal CA1 region (Fujiwara-Tsukamoto et al., 2003). It appears that the depolarizing potentials observed during epileptiform activity induced within hippocampal slices reflect not only glutamatergic activity but also GABAAergic inputs from both stratum oriens-alveus interneurons (Perez Velazquez, 2003) and the stratum pyramidale (Sun & Alkon, 2001). This of course has profound implications for the treatment of epilepsy with GABAergic agonists (Ashton & Young, 2003).
GABA-mediated excitability appears not to be confined to epileptiform activity, and occurs in adult rat hippocampal CA1 pyramidal cells at theta frequencies (5-10 Hz) associated with GABAergic postsynaptic depolarization and a shift of the reversal potential from the equilibrium potential of Cl- towards that of HCO3-, whose electrochemical gradient is regulated by carbonic anhydrase activity (Sun et al., 2001). This theta activity was abolished by GABAA receptor antagonists and also by carbonic anhydrase inhibitors, but was largely unaffected by blocking glutamate receptors (Sun et al., 2001). In fact spatial learning in a water maze appears to be regulated through such a reversal in the polarity of GABAergic postsynaptic responses by increases in carbonic anhydrase activity (Sun et al., 2001). Carbonic anhydrase also functions as a molecular switch in the development of synchronous gamma-frequency firing (40 Hz) of hippocampal CA1 pyramidal cells (Ruusuvuori et al., 2004). As a molecular synaptic switch carbonic anhydrase appears to change the function of the GABAergic synapses from one of an excitation filter to that of an amplifier (Sun & Alkon, 2001), a switch which appears critical for gating the synaptic plasticity that underlies spatial memory formation, and which may also enhance perception, processing, and the storing of temporally associated signals (Sun & Alkon, 2001).
It is hypothetically reasonable to argue, due to the fine and complex architecture of neurons, that asymmetric intracellular chloride activities may be created within the dendrites, axons and synaptic terminals of many neurons. This may occur due to highly localized variations in Cl- fluxes through ion channels, or by the asymmetric activities and expression of Cl- transport mechanisms and carbonic anhydrase activity, even whilst GABA maintains a hyperpolarizing action at the soma of the same neuron. Indeed the activation of presynaptic GABAA receptors depolarizes presynaptic glutamatergic nerve terminals projecting to ventromedial hypothalamic (VMH) neurons, thereby facilitating spontaneous glutamate release by activating excitable Na+ and Ca2+ channels (Jang et al., 2001). These regional elevations in [Cl-]i are generated by the activity of NKCCs and were responsible for the GABA-induced presynaptic depolarization (Jang et al., 2001). It is perhaps inevitable that many more such examples will be found. Some unpublished reports have suggested that longer axons may be depolarized by GABA (Robert Halliwell, University of Durham, UK).
Excitation of glia by GABA
Glial cells are known to interact extensively with neurons within the brain, and strongly influence their activity. Astrocytes, which are associated with synapses, serve to integrate neuronal inputs, and actively release transmitters which modulate synaptic sensitivity (Hansson & Ronnback, 2003). Glial cells not only participate in the formation and rebuilding of synapses, but also play a key role in neuroprotection and the restructuring of nervous tissue following injury (Hansson & Ronnback, 2003). Astrocytes and glial cells express many of the same ionotropic receptors as neurons, and NG2 cells engage in rapid signaling with GABAergic neurons through direct neuron-glia synapses (Fraser et al., 1994). Reports are beginning to emerge that the activation of GABAA receptors in such glial cells may also be excitatory. Indeed GABAA receptor activation depolarizes both proteoglycan NG2 expressing cells (Lin & Bergles, 2004) and acutely isolated hippocampal astrocytes (Fraser et al., 1995).
GABA evokes hormone release from neuroendocrine cells
GABAA receptors are expressed upon a wide range of neuroendocrine cells including insulin-secreting pancreatic B-cells (Glassmeier et al., 1998) and catecholamine releasing adrenal chromaffin cells (Peters et al., 1989; Walters et al., 2002a). In both these neuroendocrine cell types GABA is known to evoke a depolarization as [Cl-]i is maintained above equilibrium. This of particular interest in pancreatic β-cells which possess a pool of GABA-containing synaptic-like microvesicles which are distinct from the population containing insulin granules (Sorenson et al., 1991). In human insulinoma cells, which also express functional GABAA receptors, GABA application evokes a membrane depolarization which activates excitable calcium channels and evokes insulin secretion (Glassmeier et al., 1998). This GABA evoked depolarization is also observed in pancreatic β-cells which have been engineered to express GABAA-receptors at supranormal densities (Braun et al., 2004). It however remains unclear whether the quantities of GABA normally released from this vesicular pool within the pancreatic β-cells would be sufficient to depolarize the β-cell via its autoreceptors to the threshold of voltage-gated Ca2+ channel activation, which would trigger insulin secretion, or whether GABAergic afferents instead might induce insulin secretion under physiological conditions. It has been shown that GAD-like immunoreactivity, indicative of the presence of GABAergic afferent innervation, is found to be associated only with the insulin-secreting cells of rat and mouse pancreas (Gilon et al., 1991). Irrespective of whether the mechanism is via autoreceptors or by direct synaptic input, or both, GABAergic stimulation would appear likely to evoke the release of insulin in vivo. Thus pharmacological modulators of GABAA receptor activity, such as ethanol, neuroactive steroids or benzodiazepines, may significantly alter patterns of insulin release in vivo.
Perhaps the most dramatic illustration that GABA serves as an excitatory transmitter in adult tissues is the presence of functional GABAA receptors upon catecholaminergic adrenal chromaffin cells of the adrenal medulla (Peters et al., 1989). In the adrenal chromaffin cell the application of GABA elevates free [Ca2+]i (Doroshenko, 1989) by evoking membrane depolarization (Busik et al., 1996) to the known activation threshold for voltage-gated Ca2+ channels (Doroshenko, 1989). This indicates not only that [Cl-]i is accumulated above equilibrium in chromaffin cells, but also that an increase in GABAergic activity might be expected to augment or evoke catecholamine release. Immunohistochemical analysis has indicated the presence of both GABAergic afferent fibres and GABA-containing chromaffin cells in canine adrenal glands (Kataoka et al., 1986). A dense network of GABAergic fibers is present both at the boundary between medullary and cortical cells, and within the medullary tissue itself. GABAergic fibres are observed to surround the chromaffin cells, some of these fibers entering the adrenal medulla together with splanchnic cholinergic nerves (Kataoka et al., 1986).
Adrenal medullary chromaffin cells, together with adrenaline (epinephrine)-producing cells of the medulla oblongata, are one of only two cell types in the body known to functionally express the full complement of enzymes of the catecholaminergic pathway. This pathway converts the amino acid L-Tyrosine sequentially via L-Dopa (Tyrosine Hydroxyase), Dopamine (Dopa DeCarboxylase), Noradrenaline (norepinephrine, Dopamine β-Hydroxylase) into Adrenaline (Epinephrine, phenylethanolamine N-methyl-transferase). Adrenaline is the primary catecholamine released from the adrenal medulla in response to stimulation, and is responsible for increases in neural activity (alertness), cardiac output and blood pressure (e.g. Graham, 1990). Moreover, the expression of the enzymes of the catecholaminergic pathway are variably regulated and tightly controlled by a number of hormones and transmitters, including insulin (Walters et al., 2002b).
The natural question would be to ask whether the stimulation of GABAergic afferents actually increases the release of adrenaline into the bloodstream. This however is a fait accompli, as the functional role of the GABAergic system in the regulation of catecholamine release from adrenal chromaffin cells has already been demonstrated. Kataoka and co-workers studied the catecholamine output of canine adrenal glands in situ using an autoperfusion system (Kataoka et al., 1986). They demonstrated that GABA modulates the spontaneous release of catecholamines, and that adrenaline release is elicited in response to electrical stimulation of the splanchnic nerve. Administration of GABAA receptor agonists such as THIP or muscimol increased the catecholamine content in adrenal effluent blood, whilst bicuculline, a GABAA receptor antagonist, reduced it. As a control baclofen, a GABAB receptor agonist, failed to alter the catecholamine content of adrenal effluent blood, and denervation of the adrenal glands did not prevent the THIP-elicited release of catecholamines (Kataoka et al., 1986). Whilst these findings may have been received cautiously at the time, the plausibility of the contention that GABAA receptor activation is directly implicated in the stimulation of adrenaline release has gained weight. As GABA, together with acetylcholine, is effectively established as a major stimulator of catecholaminergic release, all classical GABA receptor modulators such as the BDZs (Walters et al., 2000), ethanol and neuroactive steroids may also be expected to modulate adrenaline output, especially as a BDZ-sensitive alpha subunit has been shown to be expressed both in the adrenal and PC12 cells (Walters et al., 2002a). This would at first appear to give rise to a potential therapeutic paradox, given that BDZs are employed clinically as anxiolytics. As adrenaline is the ‘fast’ stress hormone, then those therapeutic agents which target the GABA molecule might therefore also be predicted to influence stress physiology.
Summary
Clearly GABA can no longer be simply viewed as an inhibitory neurotransmitter, for GABA does merely as the intracellular chloride activity directs. As far as evidence for a widespread excitatory role for GABA in the nervous and neuroendocrine systems is concerned, we have only recently started to look and thus we have only just begun to find. Perhaps the most intriguing paradigm is that the regulation of the intracellular chloride activity provides cells with an efficient, plastic and very dynamic means of regulating changes in their excitability.
* Some investigators have claimed that localized outward fluxes of K+ ions in intercellular spaces may cause instances when the [K+]i is below equilibrium in a localized region of the cell membrane resulting in a K+ influx through inwardly rectifying K+ channels, although this is contentious.
† The intracellular Cl- concentration (activity) at equilibrium will be determined by the transmembrane potential and the concentrations of Cl- on either side of a Cl- permeable membrane. This is predicted by the Nernst equation wherein the Cl- equilibrium potential (ECl) is given by: ECl = RT/zF.ln.([Cl-]o/[Cl-]i). If [Cl-]i is higher than the predicted ECl, then [Cl-]i is said to be above equilibrium and Cl- will leave the cell upon the opening of Cl- selective ion channels, exerting a depolarizing influence upon the cell membrane potential, thereby exciting the cell, and vice-versa.
References
Ashton H, Young AH. GABA-ergic drugs: exit stage left, enter stage right. J Psychopharmacol. 2003 17(2):174-8. Review.
Balakrishnan V, Becker M, Lohrke S, Nothwang HG, Guresir E, Friauf E. Expression and function of chloride transporters during development of inhibitory neurotransmission in the auditory brainstem. J Neurosci. 2003 15;23(10):4134-45.
Ben-Ari Y. Excitatory actions of GABA during development: the nature of the nurture. Nat Rev Neurosci. 2002 3(9):728-39. Review.
Bormann J. The 'ABC' of GABA receptors. Trends Pharmacol Sci. 2000 21(1):16-9.
Bowery NG, Enna SJ. Gamma-aminobutyric acid(B) receptors: first of the functional metabotropic heterodimers. J Pharmacol Exp Ther. 2000 292(1):2-7. Review
Braun M, Wendt A, Birnir B, Broman J, Eliasson L, Galvanovskis J, Gromada J, Mulder H, Rorsman P. Regulated exocytosis of GABA-containing synaptic-like microvesicles in pancreatic β-cells. J Gen Physiol. 2004 123(3):191-204.
Busik, J., Nakamura, M., Abe, Y., Shibuya, I., Kanno, T. Effects of GABA on spontaneous [Ca2+]c dynamics and electrical properties of rat adrenal chromaffin cells. 1996 Brain Res. 11;739(1-2):97-103.
Cherubini E, Gaiarsa JL, Ben-Ari Y. GABA: an excitatory transmitter in early postnatal life. Trends Neurosci. 1991 14(12):515-9. Review.
Couve A, Moss SJ, Pangalos MN. GABAB receptors: a new paradigm in G-protein signaling. Mol Cell Neurosci. 2000 16(4):296-312. Review.
Doroshenko, P.A.g-aminobutyric acid elevates cytosolic Ca2+ in bovine chromaffin cells. 1989 Neurosci Lett. 25;104(1-2):83-7.
Fraser DD, Duffy S, Angelides KJ, Perez-Velazquez JL, Kettenmann H, MacVicar BA. GABAA/benzodiazepine receptors in acutely isolated hippocampal astrocytes. J Neurosci. 1995 15(4):2720-32.
Fraser DD, Mudrick-Donnon LA, MacVicar BA. Astrocytic GABA receptors. Glia. 1994 11(2):83-93. Review.
Fujiwara-Tsukamoto Y, Isomura Y, Nambu A, Takada M. Excitatory GABA input directly drives seizure-like rhythmic synchronization in mature hippocampal CA1 pyramidal cells. Neuroscience. 2003;119(1):265-75.
Gilon P, Tappaz M, Remacle C. Localization of GAD-like immunoreactivity in the pancreas and stomach of the rat and mouse. Histochemistry. 1991;96(4):355-65.
Glassmeier G, Hopfner M, Buhr H, Lemmer K, Riecken EO, Stein H, Quabbe HJ, Rancso C, Wiedenmann B, Scherubl H. Expression of functional GABAA receptors in isolated human insulinoma cells. Ann N Y Acad Sci. 1998 17;859:241-8.
Graham, R.M. Adrenergic receptors: structure and function. Cleve Clin J Med. 1990 57(5):481-91. Review.
Hansson E, Ronnback L. Glial neuronal signaling in the central nervous system. FASEB J. 2003 17(3):341-8. Review.
Isomura Y, Sugimoto M, Fujiwara-Tsukamoto Y, Yamamoto-Muraki S, Yamada J, Fukuda A. Synaptically activated Cl- accumulation responsible for depolarizing GABAergic responses in mature hippocampal neurons. J Neurophysiol. 2003 90(4):2752-6.
Kataoka Y, Fujimoto M, Alho H, Guidotti A, Geffard M, Kelly GD, Hanbauer I. Intrinsic gamma aminobutyric acid receptors modulate the release of catecholamine from canine adrenal gland in situ. J Pharmacol Exp Ther. 1986 239(2):584-90.
Leinekugel X, Khalilov I, McLean H, Caillard O, Gaiarsa JL, Ben-Ari Y, Khazipov R. GABA is the principal fast-acting excitatory transmitter in the neonatal brain. Adv Neurol. 1999;79:189-201. Review.
Lin SC, Bergles DE. Synaptic signaling between neurons and glia. Glia. 2004 15;47(3):290-8.
O'Brien JA, Walters RJ, Valverde MA, Sepulveda FV. Regulatory volume increase after hypertonicity- or vasoactive-intestinal-peptide-induced cell-volume decrease in small-intestinal crypts is dependent on Na(+)-K(+)-2Cl- cotransport. Pflugers Arch. 1993 423(1-2):67-73.
Perez Velazquez JL. Bicarbonate-dependent depolarizing potentials in pyramidal cells and interneurons during epileptiform activity. Eur J Neurosci. 2003 18(5):1337-42.
Peters, J.A., Lambert, J.J., Cottrell, G.A. An electrophysiological investigation of the characteristics and function of GABAA receptors on bovine adrenomedullary chromaffin cells. 1989 Pflugers Arch. 415(1):95-103.
Ruusuvuori E, Li H, Huttu K, Palva JM, Smirnov S, Rivera C, Kaila K, Voipio J. Carbonic anhydrase isoform VII acts as a molecular switch in the development of synchronous gamma-frequency firing of hippocampal CA1 pyramidal cells. J Neurosci. 2004 17;24(11):2699-707.
Sorenson RL, Garry DG, Brelje TC. Structural and functional considerations of GABA in islets of Langerhans. Beta-cells and nerves. Diabetes. 1991 40(11):1365-74.
Stein V, Nicoll RA. GABA generates excitement. Neuron. 2003 6;37(3):375-8. Review.
Sun MK, Zhao WQ, Nelson TJ, Alkon DL. Theta rhythm of hippocampal CA1 neuron activity: gating by GABAergic synaptic depolarization. J Neurophysiol. 2001 85(1):269-79.
Sun MK, Alkon DL. Pharmacological enhancement of synaptic efficacy, spatial learning, and memory through carbonic anhydrase activation in rats. J Pharmacol Exp Ther. 2001 297(3):961-7.
Walters R.J., Rosenbaum C., Willenberg H.S., Scherbaum W.A., Bornstein, S.R. Insulin, Nerve Growth Factor and PACAP Modulate the Expression of GABAA Subunits in PC12 Cells. 2002(a). Vol 1. www.neuroscience.com
Walters, R.J., Brandt M.V., Nastos, A., Hiroi, N., Willenberg, H.S., Oeff, M.K., Scherbaum, W.A., Bornstein, S.R. Insulin modulates the expression levels of dopamine b hydroxylase and phenylethanolamine N-methyl-transferase within PC12 cells. 2002(b). Cellscience
Walters RJ, Hadley SH, Morris KD, Amin J. Benzodiazepines act on GABAA receptors via two distinct and separable mechanisms. Nat Neurosci. 2000 3(12):1274-81.
Walters RJ, O'Brien JA, Valverde MA, Sepulveda FV. Membrane conductance and cell volume changes evoked by vasoactive intestinal polypeptide and carbachol in small intestinal crypts. Pflugers Arch. 1992 421(6):598-605.
Yamada J, Okabe A, Toyoda H, Kilb W, Luhmann HJ, Fukuda A. Cl- uptake promoting depolarizing GABA actions in immature rat neocortical neurones is mediated by NKCC1. J Physiol. 2004 15;557(Pt 3):829-41. Epub 2004 16.