Q/R editing of the rat GluR5 and GluR6 kainate receptors in vivo and in vitro: evidence for independent developmental, pathological and cellular regulation.

Kainate (KA) is a potent neuroexcitatory agent in several areas of the adult brain, with convulsant and excitotoxic properties that increase as ontogeny proceeds. Besides its depolarizing actions, KA may enhance intracellular accumulation of Ca2+ to promote selective neuronal damage. The effects of KA are mediated by specific receptors recently considered to be involved in fast neurotransmission and that can be activated synaptically. KA receptors, e.g. GluR5 and GluR6 have been characterized by molecular cloning. Structure-function relationships indicate that in the MII domain of these KA receptors, a glutamine (Q) or arginine (R) residue determines ion selectivity. The arginine stems from post-transcriptional editing of the GluR5 and GluR6 pre-RNAs, and the unedited and edited versions of GluR6 elicit distinct Ca2+ permeability. Using a PCR-based approach, we show that in vivo, Q/R editing in the GluR5 and GluR6 mRNAs is modulated during ontogeny and differs substantially in a variety of nervous tissues. GluR5 editing is highest in peripheral nervous tissue, e.g. the dorsal root ganglia, where GluR6 expression is barely detectable. In contrast, GluR6 editing is maximal in forebrain and cerebellar structures where GluR5 editing is lower. Intra-amygdaloid injections of KA provide a model of temporal lobe epilepsy, and we show that following seizures, the extent of GluR5 and GluR6 editing is altered in the hippocampus. However, in vitro, high levels of glutamate and potassium-induced depolarizations have no effect on GluR5 and GluR6 Q/R editing. GluR6 editing is rapidly enhanced to maximal levels in primary cultures of cerebellar granule neurons but not in cultured hippocampal pyramidal neurons. Finally, we show that cultured glial cells express partially edited GluR6 mRNAs. Our results indicate that Q/R editing of GluR5 and GluR6 mRNAs is structure-, cell type- and time-dependent, and suggest that editing of these mRNAs is not co-regulated.

Developmentally regulated alternative splicing of mRNAs encoding N-terminal tau variants in the rat hippocampus: structural and functional implications.

Tau protein variants are axonal microtubule-associated phosphoproteins whose expression correlates with developmentally regulated neurite outgrowth. A single gene encodes multiple tau transcripts via complex alternative splicing. We studied the expression of the mRNAs encoding N-terminal variants of tau, and we showed distinct alternative splicing of exons 2 and 3 in nervous tissues of the adult rat, including the inner ear, hippocampus, cortex, striatum, brainstem, cerebellum, olfactory bulb and retina. Using the reverse transcriptase-coupled polymerase chain reaction and in situ hybridization, we then focused our developmental study on hippocampal neurons, both in vivo and in vitro, to address the developmental and spatial expression of the alternatively spliced mRNAs encoding N-terminal variants of tau. Tau mRNAs devoid of exons 2 and 3 were present throughout development, although their levels decreased in adults. Those containing exon 2 but not exon 3 were already present in the hippocampus of newborn rats and their levels increased during the first postnatal week, mainly in the pyramidal cell layer. Tau RNAs containing exons 2 and 3 appeared at the end of this period in the pyramidal cell layer and in the dentate granule cells. Exon 2-containing mRNAs seemed to be associated with cells undergoing axonal sprouting, while exon 3-containing RNAs were expressed in mature neurons that had established their connections. The timing and pattern of tau alternative splicing were maintained in cultured hippocampal neurons, suggesting that splicing processes are independent of the organized connectivity and of the environmental cues provided in vivo. Secondary structure predictions of tau variants revealed that the insertion of the exon 3-encoded domain substantially modifies the secondary structure of the N-terminal region of tau. This N-terminal heterogeneity may confer distinct regulatory roles on the tau variants during ontogeny and may contribute to plasticity in the adult rat brain.

Acidic calponin cloned from neural cells is differentially expressed during rat brain development.

Calponin is an actin-, tropomyosin- and Ca2+ calmodulin-binding protein that inhibits in vitro the actomyosin MgATPase. Basic and acidic variants of calponin have been described to date. Although the cerebral expression of calponin remained controversial for some time, transcripts encoding acidic calponin in the adult rat brain and in cultured cerebellar cells have been reported. In the present work, we report the expression of acidic calponin mRNAs and the isolation of cDNAs encoding the full-length acidic calponin in cultured neuronal and glial cells and in adult rat brain. Sequence analysis reveals that acidic calponin in the brain is identical to that previously described in rat aortic vascular smooth muscle. In situ hybridization shows that calponin is highly expressed during ontogenesis in granule cells of the dentate gyrus of the hippocampus, in all layers of the olfactory bulb and in cerebellar granule neurons of the external and internal layers. In the adult rat brain, calponin expression decreased in these fields, but increased in choroid plexus cells. Bergmann glial cells were also labelled by a calponin probe. The reverse transcription-coupled polymerase chain reaction confirms that calponin mRNA levels are highest in the early stages of hippocampal development and that expression levels are low in adult hippocampi. The developmental expression pattern of brain acidic calponin suggests that calponin could be involved in contractile activity associated with neural cell proliferation or neuronal migration.

Molecular correlates between reactive and developmental plasticity in the rat hippocampus.

Area CA3 of the hippocampus is the most epileptogenic structure of the brain. Various studies have shown that kainate-induced experimental epilepsy in rats and human cases of epilepsy are associated with sprouting of the mossy fibers of the dentate granule neurons and selective loss of pyramidal neurons, notably in the CA3-CA4 areas of Ammon's horn. In experimental models of epilepsy, brief seizure activity initiates a cascade of molecular alterations that will contribute to changes in the expression of numerous genes, which can last several weeks. The products of some of these genes will contribute to the permanent state of enhanced synaptic efficiency, to the sprouting and formation of novel excitatory synapses, and possibly to neuronal cell loss. The expression of genes encoding transcription factors and numerous growth factors is rapidly altered following seizure episodes. Based on observations in vivo and in vitro in cultured hippocampal neurons, it is hypothesized that an interplay between transcription and growth factors, because of their pleiotropic effects on the regulation of effector genes, may be instrumental in coupling transient extracellular stimuli to irreversible cellular alterations.

Calcium-dependent inactivation of heteromeric NMDA receptor-channels expressed in human embryonic kidney cells.

1. Whole-cell current through heteromeric NR1-NR2A and NR1-NR2B subunit combinations of NMDA channels transiently expressed in human embryonic kidney cells (HEK 293) were studied using the patch-clamp technique. 2. With 4 mM Mg-ATP in the internal pipette solution, the responses of cells expressing NR1-NR2A channels to glutamate application gradually decreased, reaching 50% of control during the first 20 min of recording. This process was accompanied by acceleration of desensitization. 3. Conditioning (5-15 s) applications of glutamate (100 microM) induced a transient inactivation of NR1-NR2A and NR1-NR2B channels (20-40%) with a slow time course of recovery (tau r = 10-60 s). Both the degree of inactivation and the time constant of recovery increased with the duration of conditioning applications of glutamate, and with an elevation of Ca2+ in the external solution. 4. These results show that both NR1-NR2A and NR1-NR2B recombinant NMDA receptor-channels expressed in HEK 293 cells can be transiently inhibited by Ca2+ ions in a similar way to that described for hippocampal neurones.

aFGF, bFGF and flg mRNAs show distinct patterns of induction in the hippocampus following kainate-induced seizures.

We report that kainic acid-induced seizures lead to marked increases in mRNAs encoding basic and acidic fibroblast growth factors (bFGF and aFGF, respectively) and flg, one of their receptors, in the rat hippocampus. Anticonvulsant pretreatment inhibits the up-regulation of these mRNAs. The observed increase in flg mRNA levels involves the pyramidal cells of all hippocampal subfields and the granular cells of the dentate gyrus. The increased expression of aFGF and bFGF mRNAs is limited to neuron populations that are resistant to seizure-induced injury, the granular cells of dentate gyrus and pyramidal cells of CA1 region, respectively. The results suggest that the increase in the FGFs and flg may play pivotal roles in neuron survival and in long-term changes occurring in the hippocampus following seizure activity.

A new method for the measurement of endogenous transmitter release in localized regions of hippocampal slices.

We describe here a method that allows measurement of the release of endogenous amino acids from localized regions of brain slices combined with conventional electrophysiological experiments. Hippocampal slices were placed in fully submerged chambers and a cannula was positioned just above the dendritic layers of CA1. The cannula was connected to a peristaltic pump and the content of amino acids in the perfusate was measured by HPLC. Extracellular field potentials were concomitantly recorded. Stable levels of aspartate and glutamate were found above the stratum radiatum of CA1. No detectable release was found when the cannula was located above the alveus, the fimbria or in the effluent of the slice. A pulse of K+ (50 mM) produced a brief 3-fold increase in glutamate, aspartate and a detectable release of GABA in CA1. Brief high frequency trains (10 Hz) also increased significantly the release. This method will be useful in determining alterations in transmitter release in the slice in relation to anoxia, epilepsy and long term potentiation.

Selective release of endogenous zinc from the hippocampal mossy fibers in situ.

The release of endogenous zinc was studied in the hippocampus of the anesthetized rat. Push-pull cannulae were bilaterally introduced in the hippocampus and zinc concentrations determined by atomic absorption spectrophotometry. We first studied the regional distribution of K+-evoked release of zinc. A 2 min (30 mM) K+ pulse produced a release of endogenous zinc, when the push-pull cannulae were located in the vicinity of the mossy fibers (CA3 or hilus) but not in other regions (including CA1, fimbria, molecular layer of the fascia dentata, thalamus etc...). In CA3 the maximal release was of 2000 ng/ml (200 times the levels of zinc present in the cerebrospinal fluid (CSF)). Destruction of the mossy fibers by a local unilateral injection of 1.5 micrograms of colchicine dissolved in 0.6 microliter of a saline solution, eliminated this release without affecting the release in the control (contralateral) side. Electrical stimulation of the perforant path at 1 Hz did not evoke a release of zinc. In contrast at 10 Hz this stimulation produced a burst of population spike and a significant release of zinc in the mossy fibers (and not in other regions of the hippocampus). These experiments provide direct evidence that zinc is selectively released from the mossy fibers.

Effect of seizures induced by intra-amygdaloid kainic acid on kainic acid binding sites in rat hippocampus and amygdala.

[3H]Kainic acid binding sites with a slow dissociation rate in the rat limbic system were investigated in detail. Extensively washed membranes prepared from the hippocampal formation and from the region comprising the amygdala and the piriform cortex yielded non-linear Scatchard plots. Microdissection showed that the high-affinity component (affinity constant around 1 nM) was present in the hippocampal CA3 region (4.2 fmol/mg wet tissue) and the amygdaloid complex (4.6 fmol/mg wet tissue), whereas the remaining part of the hippocampal formation and the piriform lobe contained the low-affinity component (affinity constant 5-20 nM; 11.6 and 11.3 fmol/mg wet tissue, respectively). In the lateral + medial septum we detected only the low-affinity component. Severe limbic seizures, induced by unilateral injection of 0.7 or 0.8 microgram kainic acid in 0.3 microliter of phosphate-buffered saline into the amygdala, reduced kainic acid binding sites in the ipsilateral amygdala and CA3 region. The decline of kainic acid binding sites in the injected amygdala was followed by a similar effect in the contralateral amygdala ("mirror focus") and later by a moderate loss also in the contralateral CA3 region. Kainic acid receptor autoradiography demonstrated that binding sites were lost from the stratum lucidum in hippocampus. Septal lesion had no effect on kainic acid binding sites in the hippocampus. Comparison with previous results on the histopathological changes after this lesion shows that high-affinity kainic acid binding sites are preferentially located on neurons that undergo selective degenerations after severe kainic acid-induced seizures.

Spontaneous and evoked release of endogenous Zn2+ in the hippocampal mossy fiber zone of the rat in situ.

In rats anaesthetized with urethane, push pull cannulae were stereotaxically introduced in the hippocampus (bilaterally) and Zn2+ assayed in the perfusate by atomic absorption spectrophotometry. When the cannula was located in the immediate vicinity of the mossy fiber zone, both spontaneous and K+ evoked release of Zn2+ were observed, this release was associated with a reduction in the histologically demonstrable Zn2+ as assessed by means of the Timm's stain. Neither spontaneous nor evoked release of Zn2+ was observed when the cannula was located in the medial part of CA1, the fimbria or the thalamus. These observations suggest that Zn2+ is released in the mossy fiber zone.

Maturation of kainic acid seizure-brain damage syndrome in the rat. II. Histopathological sequelae.

The histopathological sequelae of parenteral administration of kainic acid were investigated in immature rats (3-35 days of age). The brains were fixed 1-14 days after the administration of kainate and the damage evaluated by means of argyrophylic (Fink-Heimer, Gallyas or Nauta-Gygax) and Nissl stains. In animals of less than 18 days of age there was no sign of damage even after 1-2 h of severe tonico-clonic convulsions. Between 18 and 35 days after birth, there was a progressive increase in the severity of the damage, the adult pattern being reached at the latter age. As in adult animals, brain damage was most severe in structures which are part of the limbic system, i.e. the hippocampal formation, lateral septum, amygdaloid complex, claustrum, piriform cortex, etc. In addition to neuronal abnormalities, the following reactions were observed: hypertrophy and swelling of satellite oligodendroglia, proliferation of hypertrophic microglia, proliferation of astroglia and hypertrophy of endothelial cells in the capillary wall. The latter type of change, together with local coagulative necrosis, was almost exclusively restricted to the granular and molecular layers of the fascia dentata. In the hippocampal formation we found a temporal gradient of vulnerability. The earliest and most consistent neuronal alterations were largely restricted to interneurons of the hilar region and to a lesser extent to non-pyramidal neurons of strata oriens and radiatum. The severe necrotic destruction of the pyramidal layer of CA3 is conspicuous at a later age (postnatal day 30-35) and with longer survival times. Our results suggest that: (1) the neurotoxin only induces brain damage once it also causes limbic motor seizures and its associated metabolic activations, notably in the amygdala; (2) the earliest pathological sequelae occur in interneurons of the hilar region and (3) sclerosis of the vulnerable region of the Ammon's horn--the CA3 region--is only obtained once the dentate granules and their mossy fibres are fully operational, thereby reflecting the crucial role of this axonal connection in eliciting hippocampal damage.

Anatomical correlates of morphine-withdrawal syndrome: differential participation of structures located within the limbic system and striatum.

Morphine-withdrawal signs have been induced, in morphine-dependent rats, by microinjection of naloxone in various diencephalic and telencephalic structures. A differential participation of the central amygdala, lateral septum, dorsal hippocampus, medial thalamic nuclei, globus pallidus and caudateputamen has been observed for the following signs: jumping, wet-dog shakes, paw tremor, chewing and diarrhea. Amygdala, medial thalamus and globus pallidus were the most sensitive to local injection of naloxone.

Alterations in local glucose consumption following systemic administration of kainic acid, bicuculline or metrazol.

Using the autoradiographic deoxyglucose method, local consumption of glucose has been visualized during convulsive seizures induced by systemic administration of kainic acid, bicuculline or metrazol. A striking correlation was found between the time course of the electrographic and metabolic alterations in limbic structures following kainic acid administration. In addition, pathological alterations were almost exclusively observed in structures in which there had been a rise in metabolism. In contrast, following bicuculline or metrazol a rise in metabolism was conspicuous primarily in the cerebellum, vestibular nuclei and neocortex.