Characteristic development of the GABA-removal system in the mouse spinal cord.

GABA is a predominant inhibitory neurotransmitter in the CNS. Released GABA is removed from the synaptic cleft by two GABA transporters (GATs), GAT-1 and GAT-3, and their dysfunction affects brain functions. The present study aimed to reveal the ontogeny of the GABA-removal system by examining the immunohistochemical localization of GAT-1 and GAT-3 in the embryonic and postnatal mouse cervical spinal cord. In the dorsal horn, GAT-1 was localized within the presynapses of inhibitory axons after embryonic day 15 (E15), a little prior to GABAergic synapse formation. The GAT-1-positive dots increased in density until postnatal day 21 (P21). By contrast, in the ventral horn, GAT-1-positive dots were sparse during development, although many transient GABAergic synapses were formed before birth. GAT-3 was first localized within the radial processes of radial glia in the ventral part on E12 and the dorsal part on E15. The initial localization of the GAT-3 was almost concomitant with the dispersal of GABAergic neurons. GAT-3 continued to be localized within the processes of astrocytes, and increased in expression until P21. These results suggested the following: (1) before synapse formation, GABA may be transported into the processes of radial glia or immature astrocytes by GAT-3. (2) At the transient GABAergic synapses in the ventral horn, GABA may not be reuptaken into the presynapses. (3) In the dorsal horn, GABA may start to be reuptaken by GAT-1 a little prior to synapse formation. (4) After synapse formation, GAT-3 may continue to remove GABA from immature and mature synaptic clefts into the processes of astrocytes. (5) Development of the GABA-removal system may be completed by P21.

Intracranial self-stimulation enhances neurogenesis in hippocampus of adult mice and rats.

Running is known to promote neurogenesis. Besides being exercise, it results in a reward, and both of these factors might contribute to running-induced neurogenesis. However, little attention has been paid to how reward and exercise relate to neurogenesis. The present study is an attempt to determine whether a reward, in the form of intracranial self-stimulation (ICSS), influences neurogenesis in the hippocampus of adult rodents. We used bromodeoxyuridine labeling to quantify newly generated cells in mice and rats that experienced ICSS for 1 h per day for 3 days. ICSS increased the number of 5-bromodeoxyuridine (Brdu)-labeled cells in the hippocampal dentate gyrus (DG) of both species. The effect, when examined at 1 day, 1 week, and 4 weeks post-ICSS, was predominantly present in the side ipsilateral to the stimulation, although it was distributed to the contralateral side. We also found in rats that, 4 weeks after Brdu injection, surviving newborn cells in the hippocampal DG of the ICSS animals co-localized with a mature neuron marker, neuronal nuclei (NeuN), and these surviving cells in rats were double-labeled with Fos, a marker of neuronal activation, after the rats had been trained to perform a spatial task. The results demonstrate that ICSS can increase newborn neurons in the hippocampal DG that endure into maturity.

Glycine receptors mediate excitation of subplate neurons in neonatal rat cerebral cortex.

The development of the cerebral cortex depends on genetic factors and early electrical activity patterns that form immature neuronal networks. Subplate neurons (SPn) are involved in the construction of thalamocortical innervation, generation of oscillatory network activity, and in the proper formation of the cortical columnar architecture. Because glycine receptors play an important role during early corticogenesis, we analyzed the functional consequences of glycine receptor activation in visually identified SPn in neocortical slices from postnatal day 0 (P0) to P4 rats using whole cell and perforated patch-clamp recordings. In all SPn the glycinergic agonists glycine, beta-alanine, and taurine induced dose-dependent inward currents with the affinity for glycine being higher than that for beta-alanine and taurine. Glycine-induced responses were blocked by the glycinergic antagonist strychnine, but were unaffected by either the GABAergic antagonist gabazine, the N-methyl-d-aspartate-receptor antagonist d-2-amino-5-phosphonopentanoic acid, or picrotoxin and cyanotriphenylborate, antagonists of alpha-homomeric and alpha1-subunit-containing glycine receptors, respectively. Under perforated-patch conditions, glycine induced membrane depolarizations that were sufficient to trigger action potentials (APs) in most cells. Furthermore, glycine and taurine decreased the injection currents as well as the synaptic stimulation strength required to elicit APs, indicating that glycine receptors have a consistent excitatory effect on SPn. Inhibition of taurine transport and application of hypoosmolar solutions induced strychnine-sensitive inward currents, suggesting that taurine can act as a possible endogenous agonist on SPn. In summary, these results demonstrate that SPn express glycine receptors that mediate robust excitatory membrane responses during early postnatal development.

Homogenous glycine receptor expression in cortical plate neurons and Cajal-Retzius cells of neonatal rat cerebral cortex.

Glycinergic membrane responses have been described in cortical plate neurons (CPn) and Cajal-Retzius cells (CRc) during early neocortical development. In order to elucidate the functional properties and molecular identity of glycine receptors in these two neuronal cell types, we performed whole-cell patch-clamp recordings and subsequent single-cell multiplex reverse transcriptase-polymerase chain reaction (RT-PCR) analyses on visually identified neurons in tangential and coronal slices as well as in situ hybridizations of coronal slices from neonatal rat cerebral cortex (postnatal days 0-4). In both CPn and CRc the glycinergic agonists glycine, beta-alanine and taurine induced inward currents with larger current densities in CRc. The functional properties of these currents were similar between CPn and CRc. In both cell types the glycine receptor showed a higher affinity for glycine than for the glycinergic agonists beta-alanine and taurine. The glycinergic responses of both cells were blocked by the glycinergic antagonist strychnine and were unaffected by the GABAergic antagonist bicuculline (100 microM), the N-methyl-D-aspartic acid receptor antagonist (+/-)-2-amino-5-phosphonopentatonic acid (60 microM) and by picrotoxin (30 microM), an antagonist of alpha homomeric glycine receptors. Single-cell multiplex RT-PCR revealed the expression of glycine receptor alpha(2) and beta subunits in CPn and CRc, while no alpha(1) and alpha(3) subunits were observed. In situ hybridization histochemistry showed the expression of mRNAs for alpha(2) and beta subunits within the cortical plate and in large neurons of the marginal zone, while there were no signals for alpha(1) and alpha(3) subunits. In summary, these results suggest that CPn and CRc express glycine receptors with similar functional and pharmacological properties. The correlation of pharmacological properties and mRNA expression suggests that the glycine receptors in both cell types may consist of alpha(2)/beta heteromeric receptors.

L-isoform but not S-isoform of myelin associated glycoprotein promotes neurite outgrowth of mouse cerebellar neurons.

Myelin associated glycoprotein (MAG) has growth promoting effect on mouse cerebellar neurons. In the present study, we examined which isoform of MAG has the effect. cDNA for L-MAG and S-MAG was stably transfected into BALB/c 3T3 cells, on which cerebellar neurons were cultured. The neurons were stained with antibody against microtubule-associated protein-2. Neurites of the neurons cultured on cells expressing L-MAG extended significantly further than those cultured on cells expressing S-MAG or on control cells. Therefore, intracellular domain of MAG may have the potential to affect MAG-neurite interaction.

Expression of the kallikrein gene family in normal and Alzheimer's disease brain.

The human kallikrein gene family consists of 15 serine proteases. We examined the expression of the kallikrein genes in human cerebral cortex and hippocampus by RT-PCR and compared their expression between Alzheimer's disease (AD) and control tissue. KLK1, 4, 5, 6, 7, 8, 10, 11, 13 and 14 are expressed in both cerebral cortex and hippocampus. KLK9 is expressed in cortex but not hippocampus, whereas KLK2, 3, 12 and 15 are not expressed in either tissue. We demonstrate an 11.5-fold increase in KLK8 mRNA levels in AD hippocampus compared to controls. The KLK8 gene product, neuropsin, processes extracellular matrix and is important for neuronal plasticity. Therefore, the increase in KLK8 could have detrimental effects on hippocampal function in AD.