GABA, but Not Bestrophin-1, Is Localized in Astroglial Processes in the Mouse Hippocampus and the Cerebellum.

GABA is proposed to act as a gliotransmitter in the brain. Differences in GABA release from astroglia are thought to underlie differences in tonic inhibition between the cerebellum and the CA1 hippocampus. Here we used quantitative immunogold cytochemistry to localize and compare the levels of GABA in astroglia in these brain regions. We found that the density of GABA immunogold particles was similar in delicate processes of Bergman glia in the cerebellum and astrocytes in the CA1 hippocampus. The astrocytic GABA release is proposed to be mediated by, among others, the Ca2+ activated Cl- channel bestrophin-1. The bestrophin-1 antibodies did not show any significant bestrophin-1 signal in the brain of wt mice, nor in bestrophin-1 knockout mice. The bestrophin-1 signal was low both on Western blots and immunofluorescence laser scanning microscopic images. These results suggest that GABA is localized in astroglia, but in similar concentrations in the cerebellum and CA1 hippocampus, and thus cannot account for differences in tonic inhibition between these brain regions. Furthermore, our data seem to suggest that the GABA release from astroglia previously observed in the hippocampus and cerebellum occurs via mechanisms other than bestrophin-1.

A distinct set of synaptic-like microvesicles in atroglial cells contain VGLUT3.

There is increasing evidence for vesicular release of glutamate from astrocytes. We have previously demonstrated existence of VGLUT1 on astrocytic synaptic-like microvesicles (SMLVs) in several brain regions indicating a role in astroglial glutamate release. As VGLUT3 is prominently expressed in non-neuronal cells, this prompted us to investigate whether VGLUT3 is also involved in astroglial release of glutamate. Confocal microscopic investigations revealed that astrocytes in the hippocampus and the frontal cortex, as well as Bergmann glia in the cerebellum were labeled for VGLUT3. Immunogold cytochemistry showed that VGLUT3 gold particles were located over SMLVs in perisynaptic astrocytic and Bergmann glial processes. The specificity of the VGLUT3 immunoreactivity was demonstrated by abolished VGLUT3 labeling in astroglia in VGLUT3 knock-out mice. Double immunogold labeling showed that astrocytic processes contained labeling for VGLUT3 and VGLUT1, but the antibodies labeled separate subpopulations of vesicles in the processes. The ratio of gold particle densities between glial processes and nerve terminals were higher for VGLUT3 than for VGLUT1, suggesting that VGLUT3 is particularly abundant in astrocytic processes. Thus, our data show that VGLUT3 localizes to a distinct set of SMLVs in perisynaptic astroglial processes and suggest that VGLUT3 is important for glutamate release from astrocytes.

VGLUT1 is localized in astrocytic processes in several brain regions.

During the last years, the concept of gliotransmission has been established. Glutamate has been shown to be released from astrocytes by different mechanisms, e.g., in an exocytotic manner. The authors have previously shown that astrocytes in the dentate-molecular layers express vesicular glutamate transporters on synaptic-like microvesicles (SLMVs). By confocal microscopy, the authors, in this study, show that vesicles by a family of glutamate transporters 1 (VGLUT1) labeling was clearly present within astrocytic processes (diameter > 1 µm) in several brain regions; the dentate-molecular layers, the stratum radiatum of CA1 hippocampus, the frontal cortex, and the striatum. At the electron microscopic level, immunogold cytochemistry showed the presence of VGLUT1 gold particles over SLMVs in delicate astrocytic processes (cross-sectional diameter < 500 nm) in all the above-mentioned brain regions. When measuring the distance from the membrane of SLMVs in astrocytes to the closest VGLUT1 gold particle, it turned out that most gold particles (above 95 %) were located within 25 nm from the membrane, strongly suggesting that VGLUT1 is present in SLMVs in the astrocytes. Finally, electron microscopic immunocytochemistry shows that VGLUT1 labeling was concentrated in astrocytic processes from wild type, and not in VGLUT1 knock out hippocampus. The authors have concluded that astrocytes not only in the dentate-molecular layers but also in stratum radiatum of CA1 hippocampus, frontal cortex, and the striatum possess SLMVs carrying VGLUT1, suggesting that astrocytes in all these brain regions are capable of vesicular release of glutamate.

Functional and anatomical identification of a vesicular transporter mediating neuronal ATP release.

ATP is known to be coreleased with glutamate at certain central synapses. However, the nature of its release is controversial. Here, we demonstrate that ATP release from cultured rat hippocampal neurons is sensitive to RNAi-mediated knockdown of the recently identified vesicular nucleotide transporter (VNUT or SLC17A9). In the intact brain, light microscopy showed particularly strong VNUT immunoreactivity in the cerebellar cortex, the olfactory bulb, and the hippocampus. Using immunoelectron microscopy, we found VNUT immunoreactivity colocalized with synaptic vesicles in excitatory and inhibitory terminals in the hippocampal formation. Moreover, VNUT immunolabeling, unlike that of the vesicular glutamate transporter VGLUT1, was enriched in preterminal axons and present in postsynaptic dendritic spines. Immunoisolation of synaptic vesicles indicated presence of VNUT in a subset of VGLUT1-containing vesicles. Thus, we conclude that VNUT mediates transport of ATP into synaptic vesicles of hippocampal neurons, thereby conferring a purinergic phenotype to these cells.

Long-term levetiracetam treatment affects reproductive endocrine function in female Wistar rats.

PURPOSE: Several antiepileptic drugs (AEDs) induce changes in endocrine function in women with epilepsy. Levetiracetam (LEV) is one of the newer AEDs, and to date no endocrine side-effects have been reported in humans. However, a recent study on ovarian follicular cells from prepubertal pigs showed that LEV affected basal steroid hormone secretion. The aim of the present study was to investigate possible effects of the drug on endocrine function and ovarian morphology in non-epileptic rats. METHODS: Thirty female Wistar rats were fed per-orally with either 50mg/kg LEV (n=15) or 150 mg/kg LEV (n=15) twice daily for 90-95 days. Twenty rats received a control solution. The rats were killed in the dioestrus phase of the oestrous cycle. Serum concentrations of testosterone, 17beta-oestradiol, progesterone, follicle stimulating hormone (FSH), luteinizing hormone (LH) and LEV were measured, and the ovaries examined histologically. RESULTS: Mean ovarian weight showed a significant, dose-dependent increase after LEV treatment. Mean numbers of ovarian follicular cysts were not changed, but the numbers of corpora lutea and secondary follicles were significantly higher in the treated animals. Serum testosterone was significantly increased in treated animals (0.50 nmol/l versus 0.16 nmol/l in controls, p<0.05), while oestradiol was reduced (67.4 compared to 257.5 pmol/l in controls, p<0.05). The low-dose group had significantly lower serum progesterone concentrations than the control group (56.8 nmol/l versus 34.7 nmol/l, respectively, p<0.05). FSH was reduced in the treated animals (3.3 ng/ml versus 5.5 ng/ml, p<0.05) while LH was unaffected. CONCLUSION: Our findings indicate a possible effect of LEV on the hypothalamic-pituitary-gonadal (HPG) axis and ovarian morphology in non-epileptic rats. The effects differ from those previously described for other AEDs. Caution must be taken before these results can be applied to humans.