Early alterations in a mouse model of Rett syndrome: the GABA developmental shift is abolished at birth.

Genetic mutations of the Methyl-CpG-binding protein-2 (MECP2) gene underlie Rett syndrome (RTT). Developmental processes are often considered to be irrelevant in RTT pathogenesis but neuronal activity at birth has not been recorded. We report that the GABA developmental shift at birth is abolished in CA3 pyramidal neurons of Mecp2-/y mice and the glutamatergic/GABAergic postsynaptic currents (PSCs) ratio is increased. Two weeks later, GABA exerts strong excitatory actions, the glutamatergic/GABAergic PSCs ratio is enhanced, hyper-synchronized activity is present and metabotropic long-term depression (LTD) is impacted. One day before delivery, maternal administration of the NKCC1 chloride importer antagonist bumetanide restored these parameters but not respiratory or weight deficits, nor the onset of mortality. Results suggest that birth is a critical period in RTT with important alterations that can be attenuated by bumetanide raising the possibility of early treatment of the disorder.

GABA: an excitatory transmitter in early postnatal life.

In the adult mammalian CNS, GABA is the main inhibitory transmitter. It inhibits neuronal firing by increasing a Cl- conductance. Bicuculline blocks this effect and induces interictal discharges. A different picture is present in neonatal hippocampal neurones, where synaptically released or exogenously applied GABA depolarizes and excites neuronal membranes--an effect that is due to a different Cl- gradient. In fact, during the early neonatal period, GABA acting on GABAA receptors provides most of the excitatory drive, whereas excitatory glutamatergic synapses are quiescent. It is suggested that during development GABA exerts mainly a trophic action through membrane depolarization and a rise in intracellular Ca2+.

GABAA, NMDA and AMPA receptors: a developmentally regulated 'ménage à trois'.

The main ionotropic receptors (GABAA, NMDA and AMPA) display a sequential participation in neuronal excitation in the neonatal hippocampus. GABA, the principal inhibitory transmitter in the adult CNS, acts as an excitatory transmitter in early postnatal stage. Glutamatergic synaptic transmission is first purely NMDA-receptor based and lacks functional AMPA receptors. Therefore, initially glutamatergic synapses are 'silent' at resting membrane potential, NMDA channels being blocked by Mg2+. However, when GABA and glutamatergic synapses are coactivated during the physiological patterns of activity, GABAA receptors can facilitate the activation of NMDA receptors, playing the role conferred to AMPA receptors later on in development. Determining the mechanisms underlying the development of this 'ménage à trois' will shed light not only on the wide range of trophic roles of glutamate and GABA in the developing brain, but also on the significance of the transition from neonatal to adult forms of plasticity.

Mechanisms of induction and expression of long-term depression at GABAergic synapses in the neonatal rat hippocampus.

Synaptic plasticity at excitatory glutamatergic synapses is believed to be instrumental in the maturation of neuronal networks. Using whole-cell patch-clamp recordings, we have studied the mechanisms of induction and expression of long-term depression at excitatory GABAergic synapses in the neonatal rat hippocampus (LTD(GABA-A)). We report that the induction of LTD(GABA-A) requires a GABA(A) receptor-mediated membrane depolarization, which is necessary to remove the Mg(2+) block from postsynaptic NMDA receptors. LTD(GABA-A) is associated with an increase in the coefficient of variation of evoked GABA(A) receptor-mediated synaptic currents and a decrease in the frequency, but not amplitude, of Sr(2+)-induced asynchronous GABA(A) quantal events. We conclude that LTD(GABA-A) induction requires the activation of both GABA(A) and NMDA postsynaptic receptors and that its expression is likely presynaptic.

Effect of neonatal degranulation on the morphological development of rat CA3 pyramidal neurons: inductive role of mossy fibers on the formation of thorny excrescences.

Intracellular peroxidase injections were made on CA3 pyramidal neurons from slices obtained from 3-60 day old control and experimental rats. Experimental rats had been irradiated at birth with gamma-rays to destroy the granule cells of the fascia dentata. Neonatal gamma-ray irradiation induced a clear cut reduction of the total number of granule cells (80% decrease at P60) and mossy fibers as revealed by Nissl and Timm stains. The number of branching points and segments, total dendritic length, and density of dendritic spines in the stratum radiatum and stratum oriens on CA3 pyramidal cells were not significantly different between control and irradiated rats. To quantify the development of thorny excrescences we measured the number and total area of the thorny excrescences per neuron at different stages of development. gamma-Ray irradiation induced a reduction in the number and total area of thorny excrescences. This effect was readily apparent by postnatal day (P) 12, and a reduction of 70% in the total area was observed at P30. In conclusion, gamma-ray irradiation destroys the majority of granule cells and induces a reduction in the development of thorny excrescences. Our data strongly suggest that this effect is directly caused by the lack of mossy fibers rather than by the irradiation itself since other parameters (i.e., the number of segments and branching points, density of non-mossy spines, and total dendritic length) were not affected. Therefore, we suggest that mossy fibers play an inductive role in the formation of thorny excrescences.

GABAergic mechanisms in the CA3 hippocampal region during early postnatal life.

The developmental pattern of GABAergic neurons in the rat hippocampus during the first week of postnatal life shows several particularities both from a morphological and physiological point of view: (1) GABA immunoreactive neurons which are initially localized in a deep and superficial layer, progressively disappear from these two layers. From the end of the first postnatal week, GABAergic neuronal somata appear throughout the whole hippocampus, but GABA immunoreactive terminal structures are not frequent until the second postnatal week. (2) Intracellular observations in slices reveal the presence in CA3 pyramidal neurons between P0 and P6 (postnatal days) of spontaneous giant depolarizing potentials (GDPs); these are mediated by GABA acting on GABAA receptors and modulated presynaptically by NMDA receptors. During this period of development, GABA and GABAA analogues have a depolarizing action at resting membrane potential. Bicuculline at this developmental stage blocks completely spontaneous and evoked synaptic potentials. During the second postnatal week, when GABA responses shift from depolarizing to hyperpolarizing, bicuculline induces spontaneous interictal discharges. It is suggested that the positive feedback of the GABAergic interneuron on the pyramidal neuron during the first week of life may account for the generation of GDPS which may play an important role in synaptogenesis.

gamma-Aminobutyric acid (GABA): a fast excitatory transmitter which may regulate the development of hippocampal neurones in early postnatal life.

The properties of neonatal GABAergic synapses were investigated in neurones of the hippocampal CA3 region. GABA, acting on GABAA receptors, provides most of the excitatory drive on immature CA3 pyramidal neurones at an early stage of development, whereas glutamatergic synapses (in particular, those mediated by AMPA receptors) are mostly quiescent. Thus, during the first postnatal week of life, bicuculline fully blocked spontaneous and evoked depolarising potentials, and GABAA receptor agonists depolarised CA3 pyramidal neurones. GABAA mediated currents also had a reduced sensitivity to benzodiazepines. In the presence of bicuculline, between P0 and P4, increasing the stimulus strength reveals an excitatory postsynaptic potential which is mostly mediated by NMDA receptors. During the same developmental period, pre- (but not post) synaptic GABAB inhibition is present. Intracellular injections of biocytin showed that the axonal network of the GABAergic interneurones is well developed at birth, whereas the pyramidal recurrent collaterals are only beginning to develop. Finally, chronic bicuculline treatment of hippocampal neurones in culture reduced the extent of neuritic arborisation, suggesting that GABA acts as a trophic factor in that period. In conclusion, it is suggested that during the first postnatal week of life, when excitatory inputs are still poorly developed, GABAA receptors provide the excitatory drive necessary for pyramidal cell outgrowth. Starting from the end of the first postnatal week of life, when excitatory inputs are well developed, GABA (acting on both GABAA and GABAB receptors) will hyperpolarise the CA3 pyramidal neurones and, as in the adult, will prevent excessive neuronal discharges. Our electrophysiological and morphological studies have shown that hippocampal GABAergic interneurones are in a unique position to modulate the development of CA3 pyramidal neurones. Developing neurones require a certain degree of membrane depolarisation, and a consequent rise in intracellular calcium, for stimulating neurite outgrowth; the GABAergic network, which develops prior to the glutamatergic one, appears to provide this depolarisation. Starting from the end of the first postnatal week of life, at a time when excitatory pathways are developing, GABA (acting on both GABAA and GABAB receptors) would reverse its action, and start to play its well-known role as an inhibitory neurotransmitter.

GABA mediated excitation in immature rat CA3 hippocampal neurons.

Intracellular recordings from rat hippocampal neurons in vitro during the first postnatal week revealed the presence of spontaneous giant depolarizing potentials (GDPs). These were generated by the synchronous discharge of a population of neurons. GDPs reversed polarity at -27 and -51 mV when recorded with KCl or K-methylsulphate filled electrodes, respectively. GDPs were blocked by the GABAA receptor antagonist bicuculline (10 microM). Iontophoretic or bath applications of GABA (10-300 microM) in the presence of tetrodotoxin (1 microM), induced a membrane depolarization or in voltage clamp experiments an inward current which reversed polarity at the same potential as GDPs. The response to GABA was blocked in a non-competitive manner by bicuculline (10 microM) and did not desensitize. GABA mediated GDPs were presynaptically modulated by N-methyl-D-aspartate (NMDA) and non-NMDA receptors. Their frequency was reduced or blocked by NMDA receptor antagonists and by the rather specific non-NMDA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX). The frequency of GDPs was enhanced by glycine and D-serine (10-30 microM) in a strychnine insensitive manner. This effect was blocked by AP-5, suggesting that it was mediated by the allosteric modulatory site of the NMDA receptor. These observations suggest that most of the 'excitatory' drive in immature neurons is mediated by GABA acting on GABAA receptors; furthermore excitatory amino acids modulate the release of GABA by a presynaptic action on GABAergic interneurons.

Restoration of mossy fiber projection in slice co-cultures of dislocated dentate gyrus and degranulated hippocampus.

Regional specificity of the mossy fiber projection is a well described feature of hippocampal intrinsic connectivity. Possible mechanisms involved in the formation of this specific projection include attraction molecules localized in the target area or repulsive cues preventing from ingrowth in non-target areas. To test this hypothesis, using organotypic co-cultures of dentate gyrus and irradiated degranulated hippocampal slices, we have disrupted the pathway normally taken by mossy fibers. The dentate gyrus explant was ectopically placed facing the alveus/stratum oriens of the irradiated hippocampal slice forcing the mossy fibers to cross the stratum oriens to reach their target area. Extensive plexuses of labeled mossy fibers were observed in the hilus and adjacent pyramidal cell layer of non-irradiated dentate gyrus explants. A few mossy fibers crossed the border between the co-cultures and reached their specific termination area in the irradiated hippocampus where they formed characteristic multiple synaptic contacts on their target cells. In addition to mossy fibers, numerous thin and varicose non-mossy fibers invade all parts of the co-cultured hippocampus establishing symmetric synapses. From these data we assume that mossy fiber axons emerging from dislocated non-irradiated dentate gyrus explants find their normal termination zone in the co-cultured degranulated hippocampal slice even if they are forced to run an unusual pathway. These results support the idea that an attraction signal arising from the target area is involved in the formation of this specific projection.

Morphology of CA3 non-pyramidal cells in the developing rat hippocampus.

Although several investigations have shown that the local GABAergic circuit in the rat hippocampus is functional very early in development, this result has not been yet completed by the investigation of the full dendritic and axonal arborization of the neonatal interneurones. In the present study, intracellular injection of biocytin was used to assess the branching pattern of interneurones in the hippocampal CA3 region of rat between 2 and 6 days of age. Based on their dendritic morphology, the biocytin-filled interneurones were divided into four classes: bipolar, stellate, pyramidal-like and fusiform interneurones. About half of the biocytin-filled neonatal interneurones exhibited dendritic or somatic filopodial processes. The axonal arbors of the filled-interneurones were widely spread into the CA3 region, and in four out of nine cases extended beyond the CA3 region to branch into the CA1 region. These results show that, despite immature features, the filopodial processes, the hippocampal interneurones are well developed early in development at a time when their target cells, the pyramidal neurones, are still developing. These observations are consistent with a trophic role that GABA may play early in development.

Involvement of GABAA receptors in the outgrowth of cultured hippocampal neurons.

Whereas GABA is a major inhibitory neurotransmitter in the adult central nervous system, recent experiments performed in our laboratory have shown that the activation of GABAA receptors in the hippocampus leads to excitatory effects during the early post-natal period. The possible consequence of a depolarizing effect of GABA was assessed on the neuritic outgrowth of embryonic hippocampal neurons in culture. No morphological alterations were observed when hippocampal neurons were cultured for three days in the presence of muscimol, a GABAA receptor agonist. In contrast, the neuritic outgrowth of cultured hippocampal neurons was profoundly affected by the presence of bicuculline in the culture medium. In the presence of this GABAA receptor antagonist neurons displayed a reduction in the number of primary neurites and branching points, resulting in a concomitant decrease of the total neuritic length. Thus, this study suggests that GABA, acting on GABAA subtype of receptors, is able to affect the development of the hippocampus.

Contributions of AMPA and GABA(A) receptors to the induction of NMDAR-dependent LTP in CA1.

The contributions of (+/-)-alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) and gamma-aminobutyric acid (GABA[A]) receptors in the induction of long-term potentiation (LTP) have been studied in the CA1 region of the rat hippocampus. The results suggest that: (1) in physiological conditions, AMPARs are necessary for the induction of N-methyl-D-aspartate receptor (NMDAR)-dependent LTP since LTP cannot be elicited in the presence of the AMPAR antagonist, 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX). Although a NMDAR-dependent LTP occurs in the presence of a GABA(A) antagonist and high concentrations of divalents cations, blockade of AMPARs leads to a voltage-dependent calcium channels (VDCC)-dependent LTP since its induction is blocked by nifedipine and not by APV. (2) The bicarbonate-induced GABA(A) receptor-mediated depolarizing response is not necessary in the induction of NMDAR-dependent or VDCC-dependent LTP since induction of these two types of LTP were not blocked by acetazolamide or in a nominally bicarbonate-free solution.

GABA is the principal fast-acting excitatory transmitter in the neonatal brain.

gamma-aminobutyric acid (GABA) is the principal neurotransmitter of inhibition in the adult mammalian brain. However, at early stages of development, including the embryonic period and first week of postnatal life, GABA plays the role of main neurotransmitter of excitation. The paradoxical excitatory effect of GABA is caused by an inverted chloride gradient and, therefore, a depolarizing direction of GABA type A (GABAA) receptor mediated responses. In addition, another type of GABAergic inhibition mediated by postsynaptic GABA type B (GABAB) receptors is not functional at early stage of life. In the neonatal rat hippocampus, GABA, acting via GABAA receptors, activates voltage-gated sodium and calcium channels and potentiates the activity of N-methyl-D-aspartate (NMDA) receptors by reducing their voltage-dependent Mg2+ block. The temporal window when GABA exerts excitatory actions coincides with a particular pattern of activity of hippocampal neuronal network that is characterized by periodical giant depolarizing potentials (GDPs) reminiscent of interictal-like epileptiform discharges. Recent studies have shown that GDPs result from the synchronous discharge of GABAergic interneurons and principal glutamatergic pyramidal cells, and they are mediated by the synergistic excitatory actions of GABAA and glutamate receptors. GDPs provide synchronous intracellular Ca2+ oscillations and may, therefore, be implicated in hebbian modulation of developing synapses and activity-dependent formation of the hippocampal network.

Epileptiform activity triggers long-term plasticity of GABA(B) receptor signalling in the developing rat hippocampus.

GABA(B) receptor (GABA(B)R)-mediated presynaptic inhibition regulates neurotransmitter release from synaptic terminals. In the neonatal hippocampus, GABA(B)R activation reduces GABA release and terminates spontaneous network discharges called giant depolarizing potentials (GDPs). Blocking GABA(B)Rs transforms GDPs into longer epileptiform discharges. Thus, GABA(B)R-mediated presynaptic inhibition of GABA release (GABA auto-inhibition) controls both spontaneous network activity and excitability in the developing hippocampus. Here we show that extensive release of endogenous GABA during epileptiform activity impairs GABA auto-inhibition, but not GABA(B)R-mediated inhibition of glutamate release, leading to hyperexcitability of the neonatal hippocampal network. Paired-pulse depression of GABA release (PPD) and heterosynaptic depression of glutamate release were used to monitor the efficacy of presynaptic GABA(B)R-mediated inhibition in slices. PPD, but not heterosynaptic depression, was dramatically reduced after potassium (K+)-induced ictal-like discharges (ILDs), suggesting a selective impairment of GABA(B)R-dependent presynaptic inhibition of GABAergic terminals. Impairing GABA auto-inhibition induced a 44% increase in GDP width and the appearance of pathological network discharges. Preventing GABA-induced activation of GABA(B)Rs during ILDs avoided PPD loss and most modifications of the network activity. In contrast, a partial block of GABA(B)Rs induced network discharges strikingly similar to those observed after K+-driven ILDs. Finally, neither loss of GABA auto-inhibition nor network hyperexcitability could be observed following synchronous release of endogenous GABA in physiological conditions (during GDPs at 1 Hz). Thus, epileptiform activity was instrumental to impair GABA(B)R-dependent presynaptic inhibition of GABAergic terminals. In conclusion, our results indicate that endogenous GABA released during epileptiform activity can reduce GABA auto-inhibition and trigger pathological network discharges in the newborn rat hippocampus. Such functional impairment may play a role in acute post-seizure plasticity.

Plasticity of GABAergic synapses in the neonatal rat hippocampus.

While the development and plasticity of excitatory synaptic connections have been studied into detail, little is known about the development of inhibitory synapses. As proposed for excitatory synapses, recent studies have indicated that activity-dependent forms of synaptic plasticity, such as long-term potentiation and long-term depression, may play a role in the establishment of functional inhibitory synaptic connections. Here, I review these different forms of plasticity and focus on their possible role in the developing neuronal network.

Activity- and age-dependent GABAergic synaptic plasticity in the developing rat hippocampus.

Activity-dependent plasticity of GABAergic synaptic transmission was investigated in rat hippocampal slices obtained between postnatal day (P) 0-15 using the whole-cell patch-clamp recording technique. Spontaneous GABA(A) receptor-mediated postsynaptic currents (sGABA(A)-PSCs) were isolated in the presence of ionotropic glutamate receptor antagonists. A conditioning protocol relevant to the physiological condition, consisting of repetitive depolarizing pulses (DPs) at 0.1 Hz, was able to induce long-lasting changes in both frequency and amplitude of sGABA(A)-PSCs between P0 and P8. Starting from P12, DPs were unable to induce any form of synaptic plasticity. The effects of DPs were tightly keyed to the frequency at which they were delivered. When delivered at a lower (0.05 Hz) or higher (1 Hz) frequency, DPs failed to induce any long-lasting change in the frequency or amplitude of sGABA(A)-PSCs. In two cases, DPs were able to activate sGABA(A)-PSCs in previously synaptically silent cells at P0-1. These results show that long-term changes in GABAergic synaptic activity can be induced during a restricted period of development by a conditioning protocol relevant to the physiological condition. It is suggested that such activity-induced modifications may represent a physiological mechanism for the functional maturation of GABAergic synaptic transmission.

Neonatal irradiation prevents the formation of hippocampal mossy fibers and the epileptic action of kainate on rat CA3 pyramidal neurons.

1. The effects of unilateral gamma-ray irradiation at birth on the properties of adult CA3 pyramidal neurons have been studied in hippocampal slices. 2. Neonatal gamma-ray irradiation reduced by 80% the number of granule cells and prevented the formation of mossy fiber synapses without reducing the number of CA3 pyramidal cells. The destruction of the mossy fibers was also confirmed with extracellular recordings. 3. Excitatory and inhibitory postsynaptic potentials (EPSPs and IPSPs) evoked by stimulation of the stratum radiatum had similar properties in nonirradiated and irradiated hippocampi: the EPSP reversed polarity near 0 mV, was reduced in amplitude by 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 microM) and D(-)-2-amino-5-phosphonovalerate (APV, 50 microM); the fast and slow IPSPs reversed at -75 and -100 mV, were blocked by bicuculline (10 microM), and reduced by phaclofen (0.5 mM), respectively. 4. Bath application of kainate (300-500 nM) evoked epileptiform activity in 81.5% of nonirradiated hippocampal CA3 regions and only in 29% of the irradiated CA3 regions. In contrast, bath application of high potassium (7 mM) and bicuculline (10 microM) generated spontaneous and evoked epileptiform activity in both nonirradiated and irradiated CA3 regions. 5. In nonirradiated and irradiated CA3 regions, kainate (200-300 nM) reduced the amplitude of the fast and slow IPSPs, reduced spike accommodation, and increased the duration of the action potential generated by a depolarizing pulse. 6. The postsynaptic responses of CA3 neurons to bath application of glutamatergic agonists were similar in nonirradiated and irradiated hippocampi in terms of amplitude, reversal potential, and pharmacology. 7. It is concluded that the most conspicuous effect of neonatal gamma-ray irradiation is to prevent the epileptic action of kainate. We propose that kainate generates epileptiform activity in the intact CA3 region by activating high-affinity binding sites located on the mossy fiber terminals.

Postnatal development of pre- and postsynaptic GABAB-mediated inhibitions in the CA3 hippocampal region of the rat.

1. Intracellular recordings were made from adult and neonatal rat hippocampal slices to study the postnatal development of GABAB-mediated inhibition in CA3 pyramidal neurons. 2. In the presence of glutamatergic receptor antagonists, direct electrical stimulation of the interneurons induced a biphasic GABAA- and GABAB-mediated inhibitory postsynaptic potential in adult [postnatal day (P) 30-P40] and young (P6-P8) CA3 pyramidal neurons. In contrast, in pups (P0-P3), electrical stimulation only induced a bicuculline-sensitive depolarizing GABAA synaptic potential. 3. The outward postsynaptic currents generated by bath-applications of baclofen (30 microM, 30 s) at P3 (78 +/- 60 pA, mean +/- SE) were 4 to 5 times smaller than those evoked between P6 (329 +/- 32 pA) and P30 (412 +/- 44 pA). At P0, baclofen failed to induce a postsynaptic current. 4. The outward currents generated by serotonin (50 microM, 30 s) and the A1 receptor agonist N-cyclopentyladenosine (40 microM, 30 s) ranged between 0 and 50 pA at P3 and between 200 and 400 pA at P6 and P30 (holding potential = -60 +/- 2 mV). 5. In the presence of potassium channel blockers, the amplitude of calcium current elicited by a depolarizing voltage step command (1 s) from a holding potential of -60 mV to a test potential of 0 mV was 2 +/- 0.15 nA at P6 (n = 9) and 0.73 +/- 0.14 nA at P3 (n = 8). Baclofen reversibly reduced the amplitude of calcium currents in young rats but not in pups. 6. Baclofen reversibly reduced the amplitude of the evoked GABAA-mediated and glutamatergic synaptic events at all developmental stages. These effects were dose dependent and antagonized by P-alpha 3-aminopropyl-P-diethoxymethyl-phosphinic acid (CGP) 35348 (500 microM). 7. We conclude that postsynaptic GABAB-mediated inhibition is absent or minimal during the first postnatal days in the CA3 region. In contrast, presynaptic GABAB inhibition is present at birth. We discuss the mechanisms and physiological consequences of these observations.