Postnatal changes in somatic c-aminobutyric acid signalling in the rat hippocampus Roman Tyzio, 1 Marat Minlebaev, 1 Sylvain Rheims, 1 Anton Ivanov, 1 Isabelle Jorquera, 1 Gregory L. Holmes, 2 Yuri Zilberter, 1 Yehezkiel Ben-Ari 1 and Rustem Khazipov 1 1 Inmed ⁄ Inserm U901, Universite ´ de la Me ´ diterrane ´ e, 163, Avenue de Luminy, 13273 Marseille, France 2 Department of Neurology, Neuroscience Center at Dartmouth, Dartmouth Medical School, Lebanon, NH, USA Keywords: development, membrane potential, patch-clamp, single channels Abstract During postnatal development of the rat hippocampus, c-aminobutyric acid (GABA) switches its action on CA3 pyramidal cells from excitatory to inhibitory. To characterize the underlying changes in the GABA reversal potential, we used somatic cell-attached recordings of GABA(A) and N-methyl-d-aspartate channels to monitor the GABA driving force and resting membrane potential, respectively. We found that the GABA driving force is strongly depolarizing during the first postnatal week. The strength of this depolarization rapidly declines with age, although GABA remains slightly depolarizing, by a few millivolts, even in adult neurons. Reduction in the depolarizing GABA driving force was due to a progressive negative shift of the reversal potential of GABA currents. Similar postnatal changes in GABA signalling were also observed using the superfused hippocampus preparation in vivo, and in the hippocampal interneurons in vitro. We also found that in adult pyramidal cells, somatic GABA reversal potential is maintained at a slightly depolarizing level by bicarbonate conductance, chloride-extrusion and chloride-loading systems. Thus, the postnatal excitatory-to-inhibitory switch in somatic GABA signalling is associated with a negative shift of the GABA reversal potential but without a hyperpolarizing switch in the polarity of GABA responses. These results also suggest that in adult CA3 pyramidal cells, somatic GABAergic inhibition takes place essentially through shunting rather than hyperpolarization. Apparent hyperpolarizing GABA responses previously reported in the soma of CA3 pyramidal cells are probably due to cell depolarization during intracellular or whole- cell recordings. Introduction c-Aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the adult brain (Farrant & Kaila, 2007). However, early in development, GABA, acting via chloride-permeable GABA(A) chan- nels, exerts an excitatory action due to elevated [Cl ) ] i in the immature neurons (Cherubini et al., 1991; Ben-Ari et al., 1997; Ben Ari, 2002; Owens & Kriegstein, 2002; Ben Ari et al., 2007). Maturation of GABAergic signalling in CA3 pyramidal cells of the rat hippocampus is characterized by two excitatory-to-inhibitory (E–I) switches in GABA’s action. The first near-term E–I switch is transient and is mediated by the action of maternal oxytocin (Tyzio et al., 2006). The second developmental E–I switch is permanent, occurring during the second postnatal week (Khazipov et al., 2004; Tyzio et al., 2007); and is due to progressive changes in the expression of the chloride membrane transporters (Rivera et al., 1999; Ganguly et al., 2001; Payne et al., 2003; Yamada et al., 2004; Dzhala et al., 2005). Both the near-term and postnatal E–I switches in GABA action on CA3 pyramidal cells are thought to be due to the switch in the polarity of GABAergic responses from depolarizing to hyperpolarizing (Obata et al., 1978; Mueller et al., 1984; Ben-Ari et al., 1989; Swann et al., 1989; Luhmann & Prince, 1991; Psarropoulou & Descombes, 1999; Rivera et al., 1999; Banke & McBain, 2006; Tyzio et al., 2006). The action of GABA depends on the direction of the transmembrane current elicited by GABA, its driving force (DF GABA ) being the difference between the GABA(A) reversal potential (E GABA ) and resting membrane potential (E m ). Using noninvasive cell-attached recordings of single GABA(A) and N-methyl-d-aspartate (NMDA) channels from CA3 pyramidal cells to monitor DF GABA and E m , it has been shown that the near-term E–I switch in the GABA action is associated with a change in the polarity of DF GABA from depolarizing to hyperpolarizing due to a negative shift of E GABA (Tyzio et al., 2006). Studies using sharp electrodes and gramicidin-perforated patch recordings have also suggested that the postnatal E–I switch is associated with a change in the polarity of DF GABA from depolarizing to hyperpolarizing (Mueller et al., 1984; Ben-Ari et al., 1989; Swann et al., 1989; Psarropoulou & Descombes, 1999; Rivera et al., 1999; Banke & McBain, 2006). However, intracellular and gramicidin- perforated patch recordings may introduce several sources of error in the estimation of GABA actions, including neuronal depolarization and modification of [Cl ) ] i , both of which may affect measurements in small immature cells (Ben-Ari et al., 1989; Barry & Lynch, 1991; Staley et al., 1992; Tyzio et al., 2003). Therefore, it remains unclear whether the postnatal E–I switch in the GABA action is associated with a depolarizing-to-hyperpolarizing switch in GABA responses. Maturation of GABA signalling in specific cell compartments is yet another poorly understood phenomenon. A considerable amount of Correspondence: R. Khazipov, as above. E-mail: khazipov@inmed.univ-mrs.fr Received 19 January 2007, revised 26 March 2008, accepted 31 March 2008 European Journal of Neuroscience, Vol. 27, pp. 2515–2528, 2008 doi:10.1111/j.1460-9568.2008.06234.x ª The Authors (2008). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience