We investigated subcellular localization of total brain-derived neurotrophic factor (BDNF) mRNA (panBDNF) and its different 5' exon-specific transcripts in cultured hypothalamic neurons. Non-isotopic in situ hybridization (DIG-labeled exon-specific riboprobes) associated with immunocytochemical MAP2 or GFAP labeling was used for detection. We found that under basal conditions panBDNF mRNA was localized in neuronal soma and in primary dendritic processes. Transcripts I and II were weakly expressed in neuronal soma while transcripts IV and VI mRNA were strongly expressed. panBDNF mRNA and transcript VI mRNA were detected in proximal dendritic processes and in astrocytes. N-methyl-D: -aspartate (NMDA) treatment decreased the dendritic label of panBDNF and transcript VI mRNA. In contrast, MK-801 (NMDA antagonist) treatment extended the labeling of all the transcripts in dendrites while K(+) depolarization only extended the dendritic labeling of panBDNF and transcript VI mRNAs. These results suggest a NMDA-receptor dependent inhibitory mechanism for dendritic destination of BDNF transcripts in hypothalamic neurons.

The contribution of distinct Ca(2+)-sensitive protein kinases to the regulation of the expression of the synaptosomal-associated protein SNAP-25 was examined in bovine chromaffin cells. Prolonged incubation with high K(+) (38 mM) or 1,1-dimethyl-4-phenyl-piperazinium (DMPP), a nicotinic receptor agonist, significantly increased SNAP-25 protein and mRNA expression, as assessed by immunoblotting and semi-quantitative RT-PCR analysis. Both stimuli preferentially enhanced mRNA coding for the SNAP-25a isoform. Increase of SNAP-25 expression induced by K(+) or DMPP was inhibited over 70% by KN-62 and KN-93, two Ca(2+)/calmodulin-dependent protein kinase (CaMK) inhibitors, whereas the inactive analogue KN-92 only reduced the expression by 34%. The three compounds also inhibited the high K(+)-elicited [Ca(2+)](i) signal by 40%, suggesting that the effect of KN-62 and KN-93 was a combination of CaMK/ Ca(2+) influx inhibitory actions. Incubation of the cells with mitogen-activated protein kinase (MAPK) inhibitors PD98059 and U0126 reduced protein expression elicited by high K(+) by 50%, but did not modify the response to DMPP. Interestingly, although protein kinase A (PKA) inhibition by H-89 did not affect the high K(+) or DMPP-induced SNAP-25 expression, basal protein levels were significantly modified upon activation or inhibition of this pathway. Basal expression of SNAP-25 was also modified by the protein kinase C (PKC) activator, phorbol 12-myristate 13-acetate, but not by Gö6976, a PKC-alpha inhibitor, suggesting that the Ca(2+)-insensitive PKC-epsilon isoform control basal expression of SNAP-25 in these cells. Taken together, these results provide the first evidence that diverse protein kinases might converge in the induction of SNAP-25 expression in chromaffin cells. The preferential contribution of one or another kinase would depend on the physiological or experimental conditions.

The contribution of Ca 2+ entry through different voltage-activated Ca 2+ channel (VACC) subtypes to the phosphorylation of extracellular signal regulated kinase (ERK) was examined in bovine adrenal-medullary chromaffin cells. High K + depolarization (40 mM, 3 min) induced ERK phosphorylation, an effect that was inhibited by specific mitogen-activated protein kinase kinase inhibitors. By using selective inhibitors, we observed that depolarization-induced ERK phosphorylation completely depended on protein kinase C-a (PKC-a), but not on Ca 2+ /calmodulin-dependent protein kinase nor cyclic AMPdependent protein kinase. Blockade of L-type Ca 2+ channels by 3 lM furnidipine, or blockade of N channels by 1 lM x-conotoxin GVIA reduced ERK phosphorylation by 70%, while the inhibition of P/Q channels by 1 lM x-agatoxin IVA only caused a 40% reduction. The simultaneous blockade of L and N, or P/Q and N channels completely abolished this response, yet 23% ERK phosphorylation remained when L and P/Q channels were simultaneously blocked. Confocal imaging of cytosolic Ca 2+ elevations elicited by 40 mM K + , showed that Ca 2+ levels increased throughout the entire cytosol, both in the presence and the absence of Ca 2+ channel blockers. Fifty-eight percent of the fluorescence rise depended on Ca 2+ entering through N channels. Thus, ERK phosphorylation seems to depend on a critical level of Ca 2+ in the cytosol rather than on activation of a given Ca 2+ channel subtype. The expression of L, N, P/Q, R and T subtypes of voltageactivated Ca 2+ channels (VACC) in neurons (Olivera et al. 1994;García et al. 2000) poses the interesting question of their specialization to control different cell functions. For instance, N and P/Q-type channels, which are predominantly found along the length of apical dendrites and in axon terminals that synapse on dendrites (Westenbroek et al. 1992), control the release of various neurotransmitters (Olivera et al. 1994;Wheeler et al. 1994;García et al. 2000). In contrast, L-type channels located on proximal dendrites and neuronal cell bodies Ahlijanian et al. 1990;Westenbroek et al. 1992;Waterman 1997;Westenbroek et al. 1998; Timmerman et al. 2002) have been associated with the regulation of gene expression and enzyme activity in cortical and hippocampal neurons (Murphy et al. 1991;Bading et al. 1993; Elliot et al. 1995;Westenbroek et al. 1995;Deisseroth et al. 1998).The particular segregation of Ca 2+ channel subtypes to dendrites, axon terminals, or somata facilitates their specialization to accomplish such specific functions in various neuronal cell types. However, specialization of the different

We report the establishment of continuously growing cell lines from spinal cords of normal and trisomy 16 fetal mice. We show that both cell lines, named M4b (derived from a normal animal) and MTh (trisomic) possess neurological markers by immunohistochemistry (neuron specific enolase, synaptophysin, microtubule associated protein-2 [MAP-2], and choline acetyltransferase) and lack glial traits (glial fibrillary acidic protein and S100). MTh cells were shown to overexpress mRNA of Cu/Zn superoxide dismutase, whose gene is present in autosome 16. We also studied intracellular Ca2+ signals ([Ca2+]i) induced by different agonists in Indo-1 loaded cells. Basal [Ca2+]i was significantly higher in MTh cells compared to M4b cells. Glutamate (200 microM) and (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid (ACDP) (100 microM) induced rapid, transient increases in [Ca2+]i in M4b and MTh cells, indicating the presence of glutamatergic metabotropic receptors. N-methyl-D-aspartate (NMDA) and kainate, but not alpha-amino-hydroxy-5-methylisoxazole-4-propionic acid (AMPA), produced [Ca2+)]i rises in both cell types. MTh cells exhibited faster time-dependent decay phase kinetics in glutamate-induced responses compared to M4b cells. Nicotine induced a transient increase in [Ca2+]i in M4b and MTh cells, with significantly greater amplitudes in the latter compared to the former. Further, both cell types responded to noradrenaline. Finally, we examined cholinergic function in both cell lines and found no significant differences in the [3H]-choline uptake, but fractional acetylcholine release induced by either K+, glutamate or nicotine was significantly higher in MTh cells. These results show that M4b and MTh cells have neuronal characteristics and the MTh line shows differences which could be related to neuronal pathophysiology in Down's syndrome.

No abstract