OP 9. Tailoring transcranial current stimulation to modulate cortical oscillations in computer simulations, ferrets, and humans

Abstract

Non-invasive transcranial current stimulation (TCS) has emerged as a highly promising therapeutic neuromodulation approach for a broad range of neurological and psychiatric illnesses. The majority of TCS studies are performed with direct current (tDCs). TCS with sine-wave stimulation waveforms (tACs) represents an attractive alternative to enhance cortical oscillations ( Marshall et al., 2006 ). Yet, little is known about how weak sine-wave electric fields interact with endogenous network dynamics ( Frohlich and McCormick, 2010 ). We hypothesize that more closely matching stimulation waveforms to ongoing cortical activity may increase and focus the effect of electrical stimulation. We utilized a multi-technique approach to determine the interaction dynamics between endogenous cortical activity and applied sine-wave stimulation (large-scale computer simulations, in vivo ferret animal model, awake healthy human subjects). Large-scale computer simulations of cortical networks (Izhikevich neuron models, pyramidal cells and fast-spiking inhibitory interneurons) were performed with CUDA-enhanced, custom-written C. TCS was modeled by injection of electric current equal to that caused by application of the electric field. tACs of varying frequencies and tDCs were used to elucidate the underlying mechanisms of how global weak pertubations of the membrane voltage interact with macroscopic network dynamics. An animal model with a gyrencephalic cortex was used to test tACs in vivo. Isoflurane/xylazine anesthesia induced slow endogenous cortical oscillations. tACs matching intrinsic activity from 0.5 to 2 Hz (in steps of 0.5 Hz) was applied transcranially. Extracellular electrophysiology was conducted to measure modulation of oscillation structure. A pre-clinical study in awake, healthy human subjects assessed differential effects of 0.75 and 40 Hz tACs stimulation. Subjects ( N = 16) were fitted with EEG electrodes and two 5 × 7 cm TCS electrodes. Each subject underwent three sessions of tACs (active sham, 0.75 Hz: frequency did not match intrinsic activity of awake subjects, 40 Hz: frequency matched ongoing activity). Five blocks of stimulation (five minutes each) were interleaved with 1-min blocks of EEG recordings. The computational simulations revealed tACs frequencies matched to intrinsic oscillations require the lowest amplitude to achieve entrainment. tACs was more effective than tDCs at entrainment. Entrainment of endogenous oscillations by weak global perturbations is mediated by a non-linear threshold effect where stimulation increases the number of “hotspots” that give rise to network-wide activation patterns. tACs in ferrets increased oscillation power at the frequency matched to stimulation. In our human study, we found that 40 Hz tACs targets ongoing oscillations in the awake cortex more specifically than 0.75 Hz tACs. 40 Hz tACs amplifies the balance between resting and activated states mediated by alpha and gamma oscillations (ratio of normalized stimulation effect to sham: alpha = .944, p = .027; gamma = 1.08, p = .022). Our computer simulations, in vivo animal experiments, and pre-clinical human trial demonstrate that tACs has the potential to become an effective modulator of cortical oscillations, particularly when stimulation frequency is matched to ongoing endogenous cortical oscillations.