Differential effect of anesthesia on visual cortex neurons with diverse population coupling

Abstract

Introduction: Firing rate (FR) and population coupling (PC) are intrinsic properties of cortical neurons.Neurons with different FR and PC have diverse excitability to stimulation, tuning curve, and synaptic plasticity. Therefore, investigation of the effect of anesthesia on neurons with different FR and PC would be important to understand state-dependent information processing in neuronal circuits.Methods: To test how anesthesia affects neurons with diverse PC and FR, we measured single-unit activities in deep layers of primary visual cortex at three levels of anesthesia with desflurane and in wakefulness. Based on PC and FR in wakefulness, neurons were classified into three distinct groups: high PC-high FR (HPHF), low PC-high FR (LPHF), and low PC-low FR (LPLF) neurons. Results: Applying repeated light flashes as visual stimuli, HPHF neurons showed the strongest early response (FR at 20-150ms post-stimulus) among the three groups, whereas the response of LPHF neurons persisted longest (up to 440ms). Anesthesia profoundly altered PC and FR, and differently affected the three neuron groups: (i) PC and FR became strongly correlated suppressing population-independent spike activity; (ii) Pairwise correlation of spikes between neurons could be predicted by a PC-based raster model suggesting uniform neuron-to-neuron coupling; (iii) Contrary to evoked-potential studies under anesthesia, the flash-induced early response of HPHF neurons was attenuated, and their spike timing was split and delayed; (iv) Late response (FR at 200-400ms post-stimulus) was suppressed both in HPHF and LPHF neurons.Conclusions: Anesthetic-induced association between PC and FR suggests reduced information content in the neural circuit. Altered response of HPHF neurons to visual stimuli suggests that anesthesia interferes with conscious sensory processing in primary sensory cortex.

Anesthesia alters cortical spike sequences in rodent visual cortex

“…In Sweeney and Clopath's mathematical model, low-firing, low-population coupling neurons function as a backbone stimulus representation, driven by slow synaptic change, while high-firing, high-population coupling neurons function as a flexible substrate driven by fast synaptic changes (Bachatene et al, 2015;Sweeney and Clopath, 2020). Furthermore in our previous study, the role of low/high firing and low/high population coupling was sustained in graded anesthetic states (Lee et al, 2020b), suggesting the function of neurons remained intact in anesthesia. Our results and previous studies suggest low-firing excitatory neurons have unique role to learning by synaptic modification in neuron networks (Bachatene et al, 2015;Sweeney and Clopath, 2020).…”

“…Excluding low firing rate neurons also tend to reduce dimensionality and enhance the statistical power of the analysis (Altman and Krzywinski, 2018). The low firing neurons also reveal unique functions with precise and selective response to sensory stimuli (Bachatene et al, 2015; Clawson et al, 2018; Lee et al, 2020b). Both the exclusion and inclusion of low firing rate neurons, therefore, reveal unique aspects of the cortical spiking activity.…”

“…The administration of anesthetics provides a well-controlled pharmacological approach to alter the state of cortex, neuronal excitability, population activity, and sensory function (Vizuete et al, 2012;Vizuete et al, 2014;Lee et al, 2020b;Lee et al, 2020a).…”

“…The administration of anesthetics provides a well-controlled pharmacological approach to alter the state of cortex, neuronal excitability, population activity, and sensory function (Vizuete et al, 2012; Vizuete et al, 2014; Lee et al, 2020b; Lee et al, 2020a). Anesthetic agents inevitably modify the neuronal circuitry to slow electrophysiological oscillations (Purdon et al, 2015) and decrease the excitatory and inhibitory spike rates (Taub et al, 2013).…”