Modulation of motor excitability by cortical optogenetic theta burst stimulation

Abstract

Intermittent theta burst stimulation (iTBS) and continuous theta burst stimulation (cTBS) are protocols used in repetitive transcranial magnetic stimulation (rTMS) or cortical electrical stimulation (CES) to facilitate or suppress corticospinal excitability. However, rTMS and CES excite all types of neuron in the target cortex probed by the coil or electrode, making it difficult to differentiate the effect of TBS on specific neural circuits involved in motor plasticity. In this study, TBS protocols were converted into an optogenetic model to achieve focalized and cell-type-specific cortical modulation. Light-sensitive channelrhodopsin-2 (ChR2) was expressed in the glutamatergic neuron in the primary motor cortex (M1) driven by the CaMKIIα promoter. A custom-made optrode comprising an optical fiber and a metal cannula electrode was fabricated to achieve optogenetic stimulation and simultaneous local field potential (LFP) recording. Single-pulse CES was delivered into M1 to elicit motor-evoked potential (MEP), which served as an indicator of motor excitability, before and after TBS intervention. Results show that both CES-iTBS and optogenetic iTBS (Opto-iTBS) can potentiate MEP activity. However, CES-cTBS suppressed MEP activity whereas Opto-cTBS enhanced it. This discrepancy may have resulted from the different neural networks targeted by the two TBS modalities, with CES-cTBS exciting all types of neuron and Opto-cTBS targeting excitatory neuron specifically. The results support the idea that intra-cortical networks determine the variation of TBS-induced neuroplasticity. This study shows that focalized and cell-type-specific brain stimulation using the optogenetic approach is viable and can be extended for further exploration of neuroplasticity.

Fig 1. Optrode design for simultaneous optogenetic stimulation and LFP recording.

A Schematic diagram of fiber-based opto-electrical neural interface showing (a) optical fiber, (b) fiber housing, (c) fiber cap, (d) pedestal of optrode, (e) connector for CES or LFP recording, and (f) stainless steel cannula (serving as cortical electrode). Close-up schematic (bottom right) shows polished surface of optrode tip. B Connection of optrode to patch cord fiber. C Optical intensity curve. Power density of blue light emitted from optrode is plotted (mean ± SD) versus laser-driven current.

Fig 2. Implantation of optical neural interface into rat’s brain.

A Schematic of optrode and reference electrode mounted on rat skull corresponding to M1 forelimb area. Lentivirus carrying ChR2 gene fused to eYFP under control of CaMKIIα promoter was injected into M1 right before optrode implantation. B Close-up images of intrinsic eYFP fluorescence and C bright-field image of acute brain slice showing rat M1. ChR2-eYFP was expressed at layer 2 through layer 6. Insertion of optrode caused lesion on cortex surface, which indicates position of implantation.

Fig 3. Recordings of optogenetic-evoked potentials under various laser intensities and pulse widths.

A Representative LFP traces (grey line) and averaging waveform (black line) recorded during 1-ms single-pulse optical stimulation in ChR2-negative (ChR2-) and ChR2-positive (ChR2+) rats. Light-induced artifacts in LFP traces (marked with ↓) can be observed. Peak-to-peak amplitudes were measured as potential difference between N1 and P1. B Dose-response curve of peak-to-peak amplitudes evoked by various power densities on logarithmic scale (log10). Linear least squares fitting curve is plotted in solid line (R² = 0.91). Each point corresponds to mean of amplitudes (μV) ± standard error of the mean. C Amplitudes evoked by 100-mW/mm² optical pulses with various durations. Each bar corresponds to mean of normalized LFP amplitude ± standard error of the mean. *p < 0.05.

Fig 4. Experimental design of CES- and Opto-iTBS/cTBS protocols applied to ChR2-expressed rats.

Averaged LFP trace is shown as black line in top-right schematic, demonstrating response of ChR2-positive M1 during Opto-TBS intervention. Thirty minutes after full anesthetization, MEPs were recorded for 10 min as baseline. After TBS intervention, 30-min MEPs were recorded for comparison with baseline. Total of five sets of experiments were performed, including CES-iTBS, CES-cTBS, Opto-iTBS, Opto-cTBS, and sham (no stimulation) sessions.

Fig 5. Representative MEP traces before and 30 min after sham, CES-iTBS, CES-cTBS, Opto-iTBS, and Opto-cTBS treatments.

MEP waveforms appear within 10 to 20 ms of EMG signal after single-pulse CES. No obvious change occurred after sham stimulation. MEP traces show increased amplitude after CES-iTBS, Opto-iTBS, and Opto-cTBS treatments, and reduced amplitude after CES-cTBS.

Fig 6. Effects of TBS on MEPs.

Fold changes were observed in MEP amplitude after A CES-iTBS/-cTBS and B Opto-iTBS/-cTBS interventions. Each data point corresponds to average ± standard error of the mean in normalized MEP amplitude. The significant differences were marked as #p < 0.05 versus sham; *p < 0.05 versus baseline (-10 min) by post hoc Dunn’s test.