Synaptic pruning through glial synapse engulfment upon motor learning.

Synaptic pruning is a fundamental process of neuronal circuit refinement in learning and memory. Accumulating evidence suggests that glia participates in sculpting the neuronal circuits through synapse engulfment. However, whether glial involvement in synaptic pruning has a role in memory formation remains elusive. Using newly developed phagocytosis reporter mice and three-dimensional ultrastructural characterization, we found that synaptic engulfment by cerebellar Bergmann glia (BG) frequently occurred upon cerebellum-dependent motor learning in mice. We observed increases in pre- and postsynaptic nibbling by BG along with a reduction in spine volume after learning. Pharmacological blockade of engulfment with Annexin V inhibited both the spine volume reduction and overnight improvement of motor adaptation. These results indicate that BG contribute to the refinement of the mature cerebellar cortical circuit through synaptic engulfment during motor learning.

Optogenetic stimulus-triggered acquisition of seizure resistance.

Unlike an electrical circuit, the hardware of the brain is susceptible to change. Repeated electrical brain stimulation mimics epileptogenesis. After such "kindling" process, a moderate stimulus would become sufficient in triggering a severe seizure. Here, we report that optogenetic neuronal stimulation can also convert the rat brain to a hyperexcitable state. However, continued stimulation once again converted the brain to a state that was strongly resistant to seizure induction. Histochemical examinations showed that moderate astrocyte activation was coincident with resilience acquisition. Administration of an adenosine A1 receptor antagonist instantly reverted the brain back to a hyperexcitable state, suggesting that hyperexcitability was suppressed by adenosine. Furthermore, an increase in basal adenosine was confirmed using in vivo microdialysis. Daily neuron-to-astrocyte signaling likely prompted a homeostatic increase in the endogenous actions of adenosine. Our data suggest that a certain stimulation paradigm could convert the brain circuit resilient to epilepsy without exogenous drug administration.

Glial amplification of synaptic signals.

KEY POINTS: Recent studies have repeatedly demonstrated the cross-talk of heterogeneous signals between neuronal and glial circuits. Here, we investigated the mechanism and the influence of physiological interactions between neurons and glia in the cerebellum. We found that the cerebellar astrocytes, Bergmann glial cells, react to exogenously applied glutamate, glutamate transporter substrate (d-aspartate) and synaptically released glutamate. In response, the Bergmann glial cells release glutamate through volume-regulated anion channels. It is generally assumed that all of the postsynaptic current is mediated by presynaptically released glutamate. However, we showed that a part of the postsynaptic current is mediated by glutamate released from Bergmann glial cells. Optogenetic manipulation of Bergmann glial state with archaerhodpsin-T or channelrhodopsin-2 reduced or augmented the amount of glial glutamate release, respectively. Our data indicate that glutamate-induced glutamate release in Bergmann glia serves as an effective amplifier of excitatory information processing in the brain. ABSTRACT: Transmitter released from presynaptic neurons has been considered to be the sole generator of postsynaptic excitatory signals. However, astrocytes of the glial cell population have also been shown to release transmitter that can react on postsynaptic receptors. Therefore, we investigated whether astrocytes take part in generation of at least a part of the synaptic current. In this study, mice cerebellar acute slices were prepared and whole cell patch clamp recordings were performed. We found that Bergmann glial cells (BGs), a type of astrocyte in the cerebellum, reacts to a glutamate transporter substrate, d-aspartate (d-Asp) and an anion conductance is generated and glutamate is released from the BGs. Glutamate release was attenuated or augmented by modulating the state of BGs with activation of light-sensitive proteins, archaerhodopsin-T (ArchT) or channelrhodopsin-2 (ChR2) expressed on BGs, respectively. Glutamate release appears to be mediated by anion channels that can be blocked by a volume-regulated anion channel-specific blocker. Synaptic response to a train of parallel fibre stimulation was recorded from Purkinje cells. The latter part of the response was also attenuated or augmented by glial modulation with ArchT or ChR2, respectively. Thus, BGs effectively function as an excitatory signal amplifier, and a part of the 'synaptic' current is actually mediated by glutamate released from BGs. These data show that the state of BGs have potential for having direct and fundamental consequences on the functioning of information processing in the brain.

Targeted expression of step-function opsins in transgenic rats for optogenetic studies.

Rats are excellent animal models for experimental neuroscience. However, the application of optogenetics in rats has been hindered because of the limited number of established transgenic rat strains. To accomplish cell-type specific targeting of an optimized optogenetic molecular tool, we generated ROSA26/CAG-floxed STOP-ChRFR(C167A)-Venus BAC rats that conditionally express the step-function mutant channelrhodopsin ChRFR(C167A) under the control of extrinsic Cre recombinase. In primary cultured cortical neurons derived from this reporter rat, only Cre-positive cells expressing ChRFR(C167A) became bi-stable, that is, their excitability was enhanced by blue light and returned to the baseline by yellow~red light. In bigenic pups carrying the Phox2B-Cre driver, ChRFR(C167A) was specifically expressed in the rostral parafacial respiratory group (pFRG) in the medulla, where endogenous Phox2b immunoreactivity was detected. These neurons were sensitive to blue light with an increase in the firing frequency. Thus, this transgenic rat actuator/reporter system should facilitate optogenetic studies involving the effective in vivo manipulation of the activities of specific cell fractions using light of minimal intensity.

Effects of 3,3',5-triiodothyronine on microglial functions.

L-tri-iodothyronine (3, 3', 5-triiodothyronine; T3) is an active form of the thyroid hormone (TH) essential for the development and function of the CNS. Though nongenomic effect of TH, its plasma membrane-bound receptor, and its signaling has been identified, precise function in each cell type of the CNS remained to be investigated. Clearance of cell debris and apoptotic cells by microglia phagocytosis is a critical step for the restoration of damaged neuron-glia networks. Here we report nongenomic effects of T3 on microglial functions. Exposure to T3 increased migration, membrane ruffling and phagocytosis of primary cultured mouse microglia. Injection of T3 together with stab wound attracted more microglia to the lesion site in vivo. Blocking TH transporters and receptors (TRs) or TRα-knock-out (KO) suppressed T3-induced microglial migration and morphological change. The T3-induced microglial migration or membrane ruffling was attenuated by inhibiting Gi /o -protein as well as NO synthase, and subsequent signaling such as phosphoinositide 3-kinase (PI3K), mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinase (ERK). Inhibitors for Na(+) /K(+) -ATPase, reverse mode of Na(+) /Ca(2+) exchanger (NCX), and small-conductance Ca(2+) -dependent K(+) (SK) channel also attenuated microglial migration or phagocytosis. Interestingly, T3-induced microglial migration, but not phagocytosis, was dependent on GABAA and GABAB receptors, though GABA itself did not affect migratory aptitude. Our results demonstrate that T3 modulates multiple functional responses of microglia via multiple complex mechanisms, which may contribute to physiological and/or pathophysiological functions of the CNS.

[Novel function of astrocytes revealed by optogenetics].

Astrocytes respond to neuronal activity. However, whether astrocytic activity has any significance in brain function is unknown. Signaling pathway leading from astrocytes to neurons would be required for astrocytes to participate in neuronal functions and, here, we investigated the presence of such pathway. Optogenetics was used to manipulate astrocytic activity. A light-sensitive protein, channelrhodopsin-2 (ChR2), was selectively expressed in astrocytes. Photostimulation of these astrocytes induced glutamate release which modulated neuronal activity and animal behavior. Such glutamate release was triggered by intracellular acidification produced by ChR2 photoactivation. Astrocytic acidification occurs upon brain ischemia, and we found that another optogenetic tool, archaerhodopsin (ArchT), could counter the acidification and suppress astrocytic glutamate release. Controlling of astrocytic pH may become a therapeutic strategy upon ischemia.

Optogenetic countering of glial acidosis suppresses glial glutamate release and ischemic brain damage.

The brain demands high-energy supply and obstruction of blood flow causes rapid deterioration of the healthiness of brain cells. Two major events occur upon ischemia: acidosis and liberation of excess glutamate, which leads to excitotoxicity. However, cellular source of glutamate and its release mechanism upon ischemia remained unknown. Here we show a causal relationship between glial acidosis and neuronal excitotoxicity. As the major cation that flows through channelrhodopsin-2 (ChR2) is proton, this could be regarded as an optogenetic tool for instant intracellular acidification. Optical activation of ChR2 expressed in glial cells led to glial acidification and to release of glutamate. On the other hand, glial alkalization via optogenetic activation of a proton pump, archaerhodopsin (ArchT), led to cessation of glutamate release and to the relief of ischemic brain damage in vivo. Our results suggest that controlling glial pH may be an effective therapeutic strategy for intervention of ischemic brain damage.

Expression, subunit composition, and function of AMPA-type glutamate receptors are changed in activated microglia; possible contribution of GluA2 (GluR-B)-deficiency under pathological conditions.

Microglia express AMPA (α-amino-hydroxy-5-methyl-isoxazole-4-propionate)-type of glutamate (Glu) receptors (AMPAR), which are highly Ca(2+) impermeable due to the expression of GluA2. However, the functional importance of AMPAR in microglia remains to be investigated, especially under pathological conditions. As low expression of GluA2 was reported in some neurodegenerative diseases, GluA2(-/-) mice were used to show the functional change of microglial AMPARs in response to Glu or kainate (KA). Here we found that Glu-induced currents in the presence of 100 µM cyclothiazide, an inhibitor of AMPAR desensitization, showed time-dependent decrease after activation of microglia with lipopolysaccharide (LPS) in GluA2(+/+) microglia, but not in GluA2(-/-) microglia. Upon activation of microglia, expression level of GluA2 subunits significantly increased, while expression of GluA1, A3 and A4 subunits on membrane surface significantly decreased. These results suggest that nearly homomeric GluA2 subunits were the main reason for low conductance of AMPAR in activated microglia. Increased expression of GluA2 in microglia was also detected partially in brain slices from LPS-injected mice. Cultured microglia from GluA2(-/-) mice showed higher Ca(2+) -permeability, consequently inducing significant increase in the release of proinflammatory cytokine, such as TNF-α. The conditioning medium from KA-treated GluA2(-/-) microglia had more neurotoxic effect on wild type cultured neurons than that from KA-treated GluA2(+/+) microglia. These results suggest that membrane translocation of GluA2-containing AMPARs in activated microglia has functional importance and thus, dysfunction or decreased expression of GluA2 may accelerate Glu neurotoxicity via excess release of proinflammatory cytokines from microglia.

Application of an optogenetic byway for perturbing neuronal activity via glial photostimulation.

Dynamic activity of glia has repeatedly been demonstrated, but if such activity is independent from neuronal activity, glia would not have any role in the information processing in the brain or in the generation of animal behavior. Evidence for neurons communicating with glia is solid, but the signaling pathway leading back from glial-to-neuronal activity was often difficult to study. Here, we introduced a transgenic mouse line in which channelrhodopsin-2, a light-gated cation channel, was expressed in astrocytes. Selective photostimulation of these astrocytes in vivo triggered neuronal activation. Using slice preparations, we show that glial photostimulation leads to release of glutamate, which was sufficient to activate AMPA receptors on Purkinje cells and to induce long-term depression of parallel fiber-to-Purkinje cell synapses through activation of metabotropic glutamate receptors. In contrast to neuronal synaptic vesicular release, glial activation likely causes preferential activation of extrasynaptic receptors that appose glial membrane. Finally, we show that neuronal activation by glial stimulation can lead to perturbation of cerebellar modulated motor behavior. These findings demonstrate that glia can modulate the tone of neuronal activity and behavior. This animal model is expected to be a potentially powerful approach to study the role of glia in brain function.