Keigo Kohara

A Local Reduction in Cortical GABAergic Synapses after a Loss of Endogenous Brain-Derived Neurotrophic Factor, as Revealed by Single-Cell Gene Knock-Out Method

To address questions of whether brain-derived neurotrophic factor (BDNF) released from active excitatory neurons acts locally only on GABAergic presynaptic terminals contacting these neurons or generally also on GABAergic terminals contacting other inactive neurons, we developed a single-cell gene knock-out method in organotypic slice culture of visual cortex of floxed BDNF transgenic mice. A biolistic transfection of Cre recombinase with green fluorescence protein (GFP) plasmids to layer II/III of the cortex resulted in loss of BDNF in a single neuron or a small number of neurons, which expressed GFP at 13–14 d in vitro. Analysis with in situ hybridization and immunohistochemistry confirmed that neurons expressing GFP lacked BDNF mRNA and protein, respectively. Analysis with immunohistochemistry using antibody against GABA synthesizing enzyme showed that the number of GABAergic terminals on the soma of BDNF knock-out neurons was smaller than that of neighboring control neurons. Morphological analysis indicated that there was no significant difference in the soma size and branch points and length of dendrites between the BDNF knock-out and control neurons. Recordings of miniature IPSCs (mIPSCs) showed that the frequency of mIPSCs of BDNF knock-out neurons was lower than that of control neurons, although the amplitude was not significantly different, suggesting the smaller number of functional GABAergic synapses on whole the BDNF knock-out neuron. The present results suggest that BDNF released from postsynaptic target neurons promotes the formation or proliferation of GABAergic synapses through its local actions in layer II/III of visual cortex.

Difference in trafficking of brain-derived neurotrophic factor between axons and dendrites of cortical neurons, revealed by live-cell imaging

BackgroundBrain-derived neurotrophic factor (BDNF), which is sorted into a regulated secretory pathway of neurons, is supposed to act retrogradely through dendrites on presynaptic neurons or anterogradely through axons on postsynaptic neurons. Depending on which is the case, the pattern and direction of trafficking of BDNF in dendrites and axons are expected to be different. To address this issue, we analyzed movements of green fluorescent protein (GFP)-tagged BDNF in axons and dendrites of living cortical neurons by time-lapse imaging. In part of the experiments, the expression of BDNF tagged with cyan fluorescent protein (CFP) was compared with that of nerve growth factor (NGF) tagged with yellow fluorescent protein (YFP), to see whether fluorescent protein-tagged BDNF is expressed in a manner specific to this neurotrophin.ResultsWe found that BDNF tagged with GFP or CFP was expressed in a punctated manner in dendrites and axons in about two-thirds of neurons into which plasmid cDNAs had been injected, while NGF tagged with GFP or YFP was diffusely expressed even in dendrites in about 70% of the plasmid-injected neurons. In neurons in which BDNF-GFP was expressed as vesicular puncta in axons, 59 and 23% of the puncta were moving rapidly in the anterograde and retrograde directions, respectively. On the other hand, 64% of BDNF-GFP puncta in dendrites did not move at all or fluttered back and forth within a short distance. The rest of the puncta in dendrites were moving relatively smoothly in either direction, but their mean velocity of transport, 0.47 ± 0.23 (SD) μm/s, was slower than that of the moving puncta in axons (0.73 ± 0.26 μm/s).ConclusionThe present results show that the pattern and velocity of the trafficking of fluorescence protein-tagged BDNF are different between axons and dendrites, and suggest that the anterograde transport in axons may be the dominant stream of BDNF to release sites.

Brain-derived neurotrophic factor increases inhibitory synapses, revealed in solitary neurons cultured from rat visual cortex

To elucidate chronic actions of brain-derived neurotrophic factor (BDNF) on GABAergic synapses, we examined effects of a long-term application of BDNF for 10-15 days on autapses (synapses) of solitary GABAergic neurons cultured from rat visual cortex. Solitary neuron preparations were used to exclude a possible contamination of BDNF actions on excitatory neurons in dissociated neuron culture or slice preparations. Neurons were confirmed to be GABAergic pharmacologically with bicuculline, a selective antagonist for GABAA receptors and immunocytochemically with antibody against glutamic acid decarboxylase 65, a GABA synthesizing enzyme. To evaluate GABAergic synaptic function, evoked and/or miniature inhibitory postsynaptic currents (IPSCs) were recorded in the whole-cell voltage-clamp mode. The treatment with BDNF at a concentration of 100 ng/ml enhanced the amplitude of evoked IPSCs and the frequency of miniature IPSCs. In contrast, BDNF did not have a detectable effect on the amplitude of miniature IPSCs and the paired pulse ratio of IPSCs evoked by two, successive activations. To evaluate morphological changes, neurons were immunocytochemically stained with antibodies against microtubule-associated protein 2, to visualize somatodendritic region and synapsin I, to visualize presynaptic sites. The quantitative analysis indicated that BDNF increased the area of soma, the numbers of primary dendrites and dendritic branching points, the total length of dendrites and the number of synaptic sites. Such an action of BDNF was seen in both subgroups of GABAergic neurons, parvalbumin-positive and -negative neurons. To visualize functionally active presynaptic sites, neurons were stained with a styryl dye, FM1-43. BDNF increased the number of stained sites that was correlated with the frequency of miniature IPSCs. These results suggest that the chronic treatment with BDNF promotes dendritic and synaptic development of GABAergic neurons in visual cortex.

Increase in number of functional release sites by cyclic AMP-dependent protein kinase in cultured neurons isolated from hippocampal dentate gyrus

The enhancement of synaptic exocytosis is one form of long-term potentiation (LTP) of synaptic transmission. As possible mechanisms underlying this enhancement, increases in the release probability and/or the number of release sites are suggested. To obtain direct evidence for the increase in the number of functional release sites induced by protein kinase A (PKA) cascade, we attempted to visualize functional release sites using styryl dyes, FM4-64 and FM1-43, and investigated the effects of PKA on the release sites. A PKA activator FSK increased the number of active release sites by approximately 20-30%. A direct PKA activator, Sp-cAMPS, showed the same effect, which was blocked by a PKA inhibitor, KT5720, suggesting that this effect was mediated by PKA. This PKA-dependent increase in the number of release sites requires Ca(2+) in the bath solution, and Sr(2+) can not be a substitute for Ca(2+). Since the number of functional release sites is approximately half the total number of synaptophysin-immunoreactive sites, the PKA dependent activation of silent release sites of DG neuron terminals is suggested.

Activity-dependent survival of rat cerebellar granule neurons is not associated with sustained elevation of intracellular Ca2+

Ca2+ plays a pivotal role for the activity-dependent survival of neurons. In primary culture of cerebellar granule neurons, we found that there is no significant difference in intracellular Ca2+ level in the survival-promoting condition (cultures in the presence of 25 mM KCl) and that in the apoptosis-inducing condition (cultures in the presence of 5 mM KCl). This was not due to the inactivation of voltage-dependent L-type Ca2+ channels in the survival-promoting condition, but due to the enhanced rate of the influx and the efflux of Ca2+ in the survival-promoting condition compared to that in the apoptosis-inducing condition. These results suggest that the activity-dependent survival of the granule neurons is not associated with sustained rise of intracellular Ca2+ but associated with the enhanced turnover rate of Ca2+.

Activity-dependent survival and enhanced turnover of calcium in cultured rat cerebellar granule neurons

Neurons survive when their activity is maintained. An influential hypothesis on the cellular mechanism underlying this phenomenon is that there is an appropriate range of intracellular Ca2+ concentration ([Ca2+]i) for survival. The rat cerebellar granule neuron in culture serves as the most often used model system for the analysis of activity-dependent survival, since it does not survive unless an excitant (KCl or glutamate) is added to the culture medium. Against the above-mentioned hypothesis, we found in our previous examination no difference between steady-state [Ca2+]i in granule neurons cultured under high KCl (i.e., survival) and low KCl (i.e., death) conditions. In this report, we present the quantitative background of unchanged [Ca2+]i between the two culture conditions. Influx of Ca2+ due predominantly to L-type voltage-dependent calcium channels was higher in high KCl cultures than in low KCl cultures. At the same time, efflux of Ca2+ due to the activity of Ca2+/Na+ antiport was also higher in high KCl cultures. Additionally, we found that the endocytotic activity was greater in high KCl cultures than in low KCl cultures, as monitored by the rate of uptake of horseradish peroxidase added to medium. Since the uptake was blocked by an internal Ca2+ chelator, the increased endocytotic activity in high KCl cultures might be a consequence of the enhanced Ca2+ turnover.

Increased exocytotic capability of rat cerebellar granule neurons cultured under depolarizing conditions

To obtain insights into the mechanisms underlying activity-dependent survival of neurons, we surveyed various indices of cellular activity in rat cerebellar granule neurons cultured under conditions advantageous and disadvantageous for survival. Previously, we reported that the turnover of Ca2+ (both influx and efflux) is activated in raised K+-cultures (survival condition), although the cytoplasmic Ca2+ concentration is not affected. We also reported that endocytotic activity was high in the high K+-cultures. In the present study, we used the release of FM1-43 dye [N-(3-triethylammoniumpropyl)-4-(4-dibutylamino)styryl)py ridium bromide] to determine the exocytotic capabilities of neurons cultured in normal K+ (death condition), high K+ (survival condition) and brain-derived neurotrophic factor-supplemented (survival condition) media. The FM1-43 releases triggered by K+-induced depolarization and glutamate exposure were significantly higher in the high K+-cultures than in normal K+-cultures. Interestingly, the neurons whose survival was supported by brain-derived neurotrophic factor did not show high exocytotic capability, indicating that the high exocytotic capability is not a mere result of viability. However, the number of synaptic sites per cell (as monitored by synaptophysin immunopositivity) was unaffected by culture conditions. The present results suggest that an enhanced exocytotic activity supported by a strengthened exocytotic capability may underlie the high viability of rat cerebellar granule neurons cultured under depolarizing conditions.

Response: BDNF as an anterophin; a novel neurotrophic relationship between brain neurons

New evidence supports the notion that BDNF can act in an anterograde direction in addition to the more well-known retrograde action where BDNF is released by the target cell and acts presynaptically.