Connection-type-specific biases make uniform random network models consistent with cortical recordings.

Uniform random sparse network architectures are ubiquitous in computational neuroscience, but the implicit hypothesis that they are a good representation of real neuronal networks has been met with skepticism. Here we used two experimental data sets, a study of triplet connectivity statistics and a data set measuring neuronal responses to channelrhodopsin stimuli, to evaluate the fidelity of thousands of model networks. Network architectures comprised three neuron types (excitatory, fast spiking, and nonfast spiking inhibitory) and were created from a set of rules that govern the statistics of the resulting connection types. In a high-dimensional parameter scan, we varied the degree distributions (i.e., how many cells each neuron connects with) and the synaptic weight correlations of synapses from or onto the same neuron. These variations converted initially uniform random and homogeneously connected networks, in which every neuron sent and received equal numbers of synapses with equal synaptic strength distributions, to highly heterogeneous networks in which the number of synapses per neuron, as well as average synaptic strength of synapses from or to a neuron were variable. By evaluating the impact of each variable on the network structure and dynamics, and their similarity to the experimental data, we could falsify the uniform random sparse connectivity hypothesis for 7 of 36 connectivity parameters, but we also confirmed the hypothesis in 8 cases. Twenty-one parameters had no substantial impact on the results of the test protocols we used.

Microcircuits of excitatory and inhibitory neurons in layer 2/3 of mouse barrel cortex.

Synaptic interactions between nearby excitatory and inhibitory neurons in the neocortex are thought to play fundamental roles in sensory processing. Here, we have combined optogenetic stimulation, whole cell recordings, and computational modeling to define key functional microcircuits within layer 2/3 of mouse primary somatosensory barrel cortex. In vitro optogenetic stimulation of excitatory layer 2/3 neurons expressing channelrhodopsin-2 evoked a rapid sequence of excitation followed by inhibition. Fast-spiking (FS) GABAergic neurons received large-amplitude, fast-rising depolarizing postsynaptic potentials, often driving action potentials. In contrast, the same optogenetic stimulus evoked small-amplitude, subthreshold postsynaptic potentials in excitatory and non-fast-spiking (NFS) GABAergic neurons. To understand the synaptic mechanisms underlying this network activity, we investigated unitary synaptic connectivity through multiple simultaneous whole cell recordings. FS GABAergic neurons received unitary excitatory postsynaptic potentials with higher probability, larger amplitudes, and faster kinetics compared with NFS GABAergic neurons and other excitatory neurons. Both FS and NFS GABAergic neurons evoked robust inhibition on postsynaptic layer 2/3 neurons. A simple computational model based on the experimentally determined electrophysiological properties of the different classes of layer 2/3 neurons and their unitary synaptic connectivity accounted for key aspects of the network activity evoked by optogenetic stimulation, including the strong recruitment of FS GABAergic neurons acting to suppress firing of excitatory neurons. We conclude that FS GABAergic neurons play an important role in neocortical microcircuit function through their strong local synaptic connectivity, which might contribute to driving sparse coding in excitatory layer 2/3 neurons of mouse barrel cortex in vivo.

Parameter extraction and classification of three cortical neuron types reveals two distinct adaptation mechanisms.

Cortical information processing originates from the exchange of action potentials between many cell types. To capture the essence of these interactions, it is of critical importance to build mathematical models that reflect the characteristic features of spike generation in individual neurons. We propose a framework to automatically extract such features from current-clamp experiments, in particular the passive properties of a neuron (i.e., membrane time constant, reversal potential, and capacitance), the spike-triggered adaptation currents, as well as the dynamics of the action potential threshold. The stochastic model that results from our maximum likelihood approach accurately predicts the spike times, the subthreshold voltage, the firing patterns, and the type of frequency-current curve. Extracting the model parameters for three cortical cell types revealed that cell types show highly significant differences in the time course of the spike-triggered currents and moving threshold, that is, in their adaptation and refractory properties but not in their passive properties. In particular, GABAergic fast-spiking neurons mediate weak adaptation through spike-triggered currents only, whereas regular spiking excitatory neurons mediate adaptation with both moving threshold and spike-triggered currents. GABAergic nonfast-spiking neurons combine the two distinct adaptation mechanisms with reduced strength. Differences between cell types are large enough to enable automatic classification of neurons into three different classes. Parameter extraction is performed for individual neurons so that we find not only the mean parameter values for each neuron type but also the spread of parameters within a group of neurons, which will be useful for future large-scale computer simulations.

In vivo optogenetic stimulation of neocortical excitatory neurons drives brain-state-dependent inhibition.

BACKGROUND: Synaptic interactions between excitatory and inhibitory neocortical neurons are important for mammalian sensory perception. Synaptic transmission between identified neurons within neocortical microcircuits has mainly been studied in brain slice preparations in vitro. Here, we investigate brain-state-dependent neocortical synaptic interactions in vivo by combining the specificity of optogenetic stimulation with the precision of whole-cell recordings from postsynaptic excitatory glutamatergic neurons and GFP-labeled inhibitory GABAergic neurons targeted through two-photon microscopy. RESULTS: Channelrhodopsin-2 (ChR2) stimulation of excitatory layer 2/3 barrel cortex neurons evoked larger and faster depolarizing postsynaptic potentials and more synaptically driven action potentials in fast-spiking (FS) GABAergic neurons compared to both non-fast-spiking (NFS) GABAergic neurons and postsynaptic excitatory pyramidal neurons located within the same neocortical microcircuit. The number of action potentials evoked in ChR2-expressing neurons showed low trial-to-trial variability, but postsynaptic responses varied strongly with near-linear dependence upon spontaneously driven changes in prestimulus membrane potential. Postsynaptic responses in excitatory neurons had reversal potentials, which were hyperpolarized relative to action potential threshold and were therefore inhibitory. Reversal potentials measured in postsynaptic GABAergic neurons were close to action potential threshold. Postsynaptic inhibitory neurons preferentially fired synaptically driven action potentials from spontaneously depolarized network states, with stronger state-dependent modulation in NFS GABAergic neurons compared to FS GABAergic neurons. CONCLUSIONS: Inhibitory neurons appear to dominate neocortical microcircuit function, receiving stronger local excitatory synaptic input and firing more action potentials compared to excitatory neurons. In mouse layer 2/3 barrel cortex, we propose that strong state-dependent recruitment of inhibitory neurons drives competition among excitatory neurons enforcing sparse coding.

Membrane potential dynamics of GABAergic neurons in the barrel cortex of behaving mice.

Computations in cortical circuits are mediated by synaptic interactions between excitatory and inhibitory neurons, and yet we know little about their activity in awake animals. Here, through single and dual whole-cell recordings combined with two-photon microscopy in the barrel cortex of behaving mice, we directly compare the synaptically driven membrane potential dynamics of inhibitory and excitatory layer 2/3 neurons. We find that inhibitory neurons depolarize synchronously with excitatory neurons, but they are much more active with differential contributions of two classes of inhibitory neurons during different brain states. Fast-spiking GABAergic neurons dominate during quiet wakefulness, but during active wakefulness Non-fast-spiking GABAergic neurons depolarize, firing action potentials at increased rates. Sparse uncorrelated action potential firing in excitatory neurons is driven by fast, large, and cell-specific depolarization. In contrast, inhibitory neurons fire correlated action potentials at much higher frequencies driven by slower, smaller, and broadly synchronized depolarization.