Dendritic spines have been proposed to transform synaptic signals through chemical and electrical compartmentalization. However, the quantitative contribution of spine morphology to synapse compartmentalization and its dynamic regulation are still poorly understood.We used time-lapse superresolution STED imaging in combination with FRAP measurements, 2-photon glutamate uncaging, electrophysiology and simulations to investigate the dynamic link between nanoscale anatomy and compartmentalization in live spines of CA1 neurons in mouse brain slices.We report a diversity of spine morphologies that argues against common categorization schemes, and establish a close link between compartmentalization and spine morphology, where spine neck width is the most critical morphological parameter. We demonstrate that spine necks are plastic structures that become wider and shorter after LTP. These morphological changes are predicted to lead to a substantial drop in spine head EPSP, while leaving overall biochemical compartmentalization preserved. 3 Dendritic spines form the postsynaptic component of most excitatory synapses, whose plasticity is essential for brain development and higher brain functions 1,2 . In addition to the molecular composition of the synapse, the morphology of spines is thought to be critical for synaptic function, as spine head size correlates with synaptic strength 3,4 and undergoes changes during synaptic plasticity [5][6][7][8] . Even so, our understanding of how spine structure shapes synapse function remains fragmented.It is well established that spines compartmentalize biochemical signals 9 . By contrast, the quantitative contribution of spine morphology to compartmentalization is still unknown, and only moderate correlations between spine neck length or head volume and chemical diffusion have been reported [9][10][11] . It is an open question to what extent biochemical compartmentalization is determined primarily by spine geometry or intracellular factors such as organelles or protein assemblies.Concerning electrical compartmentalization, it is not clear how electrical signals are transformed by the spine neck 9,[12][13][14] . This is an important question because synaptic strength may be adjusted through structural changes in spine necks, which has been a long-standing hypothesis 15,16 . An early electron microscopy study reported that the average spine head becomes larger and the neck wider and shorter after the induction of long-term plasticity (LTP) 17 , which was corroborated more recently by work based on 2-photon microscopy 6,18,19 . However, it is not known how these structural changes might affect biochemical and electrical compartmentalization, because 2-photon microscopy does not have sufficient spatial resolution to properly resolve spines and electron microscopy cannot be combined with functional assays. 4 Here, we combined stimulated emission depletion (STED) microscopy, fluorescence recovery after photo-bleaching (FRAP) experiments, 2-photon glutamate uncaging, and patch-clam...

Cortical synapses display remarkable structural, molecular and functional heterogeneity. Our knowledge regarding the relationship between the ultrastructural and functional parameters is still fragmented. Here we asked how the release probability and presynaptic [Ca2+] transients relate to the ultrastructure of rat hippocampal glutamatergic axon terminals. Two-photon Ca2+ imaging-derived optical quantal analysis and correlated electron microscopic reconstructions revealed a tight correlation between the release probability and the active zone area. The peak amplitude of [Ca2+] transients in single boutons also positively correlated with the active zone area. Freeze-fracture immunogold labeling revealed that the voltage-gated Ca2+ channel subunit Cav2.1 and the presynaptic protein Rim1/2 are confined to the active zone and their numbers scale linearly with the active zone area. Gold particles for Cav2.1 showed a nonrandom distribution within the active zones. Our results demonstrate that the number of several active zone proteins, including presynaptic Ca2+ channels, docked vesicles and the release probability scales linearly with the active zone area.

Microglia are the main immune cells of the brain and contribute to common brain diseases. However, it is unclear how microglia influence neuronal activity and survival in the injured brain in vivo. Here we develop a precisely controlled model of brain injury induced by cerebral ischaemia combined with fast in vivo two-photon calcium imaging and selective microglial manipulation. We show that selective elimination of microglia leads to a striking, 60% increase in infarct size, which is reversed by microglial repopulation. Microglia-mediated protection includes reduction of excitotoxic injury, since an absence of microglia leads to dysregulated neuronal calcium responses, calcium overload and increased neuronal death. Furthermore, the incidence of spreading depolarization (SD) is markedly reduced in the absence of microglia. Thus, microglia are involved in changes in neuronal network activity and SD after brain injury in vivo that could have important implications for common brain diseases.

Extracting neuronal spiking activity from large-scale two-photon recordings remains challenging, especially in mammals in vivo, where large noises often contaminate the signals. We propose a method, MLspike, which returns the most likely spike train underlying the measured calcium fluorescence. It relies on a physiological model including baseline fluctuations and distinct nonlinearities for synthetic and genetically encoded indicators. Model parameters can be either provided by the user or estimated from the data themselves. MLspike is computationally efficient thanks to its original discretization of probability representations; moreover, it can also return spike probabilities or samples. Benchmarked on extensive simulations and real data from seven different preparations, it outperformed state-of-the-art algorithms. Combined with the finding obtained from systematic data investigation (noise level, spiking rate and so on) that photonic noise is not necessarily the main limiting factor, our method allows spike extraction from large-scale recordings, as demonstrated on acousto-optical three-dimensional recordings of over 1,000 neurons in vivo.

The understanding of brain computations requires methods that read out neural activity on different spatial and temporal scales. Following signal propagation and integration across a neuron and recording the concerted activity of hundreds of neurons pose distinct challenges, and the design of imaging systems has been mostly focused on tackling one of the two operations. We developed a high-resolution, acousto-optic two-photon microscope with continuous three-dimensional (3D) trajectory and random-access scanning modes that reaches near-cubic-millimeter scan range and can be adapted to imaging different spatial scales. We performed 3D calcium imaging of action potential backpropagation and dendritic spike forward propagation at sub-millisecond temporal resolution in mouse brain slices. We also performed volumetric random-access scanning calcium imaging of spontaneous and visual stimulation-evoked activity in hundreds of neurons of the mouse visual cortex in vivo. These experiments demonstrate the subcellular and network-scale imaging capabilities of our system.

Organic electrochemical transistors are integrated on depth probes to achieve localized electrical stimulation of neurons. The probes feature a mechanical delamination process which leaves only a 4 μm thick film with embedded transistors inside the brain. This considerably reduces probe invasiveness and correspondingly improves future brain-machine interfaces.

Inhibitory interneurons are considered to be the controlling units of neural networks, despite their sparse number and unique morphological characteristics compared with excitatory pyramidal cells. Although pyramidal cell dendrites have been shown to display local regenerative events-dendritic spikes (dSpikes)-evoked by artificially patterned stimulation of synaptic inputs, no such studies exist for interneurons or for spontaneous events. In addition, imaging techniques have yet to attain the required spatial and temporal resolution for the detection of spontaneously occurring events that trigger dSpikes. Here we describe a highresolution 3D two-photon laser scanning method (Roller Coaster Scanning) capable of imaging long dendritic segments resolving individual spines and inputs with a temporal resolution of a few milliseconds. By using this technique, we found that local, NMDA receptor-dependent dSpikes can be observed in hippocampal CA1 stratum radiatum interneurons during spontaneous network activities in vitro. These NMDA spikes appear when approximately 10 spatially clustered inputs arrive synchronously and trigger supralinear integration in dynamic interaction zones. In contrast to the one-to-one relationship between computational subunits and dendritic branches described in pyramidal cells, here we show that interneurons have relatively small (∼14 μm) sliding interaction zones. Our data suggest a unique principle as to how interneurons integrate synaptic information by local dSpikes.3D scanning | hippocampus | uncaging | signal integration | modeling I nterneurons are critically important for synaptic plasticity and synchronization of oscillatory activities and have been suggested to provide temporal control for principal cells (1). Although the output of interneurons is more thoroughly studied, there is only sparse evidence on how the inputs on their dendrites act when activated in concert under in vitro conditions. In contrast, there are a number of phenomena explored concerning signal integration on principal cells that have not been described on interneurons. Spatial clustering of synchronized synaptic inputs can lead to nonlinear integration and regenerative events in dendrites of principal neurons, increasing their computational power (2-8). On the contrary, interneurons were previously suggested to have more linear or sublinear integration propertiesfeatures that might imply a passive involvement in neuronal operations (9). Dendritic integration in principal cells is mediated by multiple layers of logical integrators (2, 6, 10), the first layer being the apical Ca 2+ and axonal Na + integration zone, generating relatively more global propagating spikes. At the same time, both tuft and basal thin dendritic branches are able to generate NMDA spikes, providing an integration method for distant inputs to overcome strong dendritic filtering (11) and drive the output of the cell (10). On the contrary, we are aware of no studies to date that have shown that local regenerative spikes (evoked or spontaneou...

How neuronal computations in the sensory periphery contribute to computations in the cortex is not well understood. We examined this question in the context of visual-motion processing in the retina and primary visual cortex (V1) of mice. We disrupted retinal direction selectivity – either exclusively along the horizontal axis using FRMD7 mutants or along all directions by ablating starburst amacrine cells – and monitored neuronal activity in layer 2/3 of V1 during stimulation with visual motion. In control mice, we found an overrepresentation of cortical cells preferring posterior visual motion, the dominant motion direction an animal experiences when it moves forward. In mice with disrupted retinal direction selectivity, the overrepresentation of posterior-motion-preferring cortical cells disappeared, and their response at higher stimulus speeds was reduced. This work reveals the existence of two functionally distinct, sensory-periphery-dependent and -independent computations of visual motion in the cortex.