Lifting the veil on the dynamics of neuronal activities evoked by transcranial magnetic stimulation

Bingshuo Li, Juha Virtanen, A Oeltermann, Cornelius Schwarz, Martin A. Giese, Ulf Ziemann, Alia BenalieLife 2017 · DOI

TL;DR

This paper introduces a novel method for simultaneously recording electrophysiological activity from individual neurons and applying transcranial magnetic stimulation (TMS) in rodents. This breakthrough allows researchers to observe neuronal responses to TMS with millisecond precision, revealing previously inaccessible dynamics of brain activation and modulation.

What Problem Does This Paper Address?

Transcranial magnetic stimulation (TMS) is a powerful, non-invasive tool for studying and treating brain disorders. However, understanding precisely how TMS affects neuronal activity at the individual neuron level has been a major challenge. The strong electromagnetic fields generated by TMS interfere with conventional electrophysiological recording techniques, masking the very signals researchers want to observe. Previous attempts to overcome this have been technically demanding, limited in scope, or unable to achieve the necessary temporal resolution. This leaves a critical gap in our understanding of TMS-brain interactions, hindering the development of more targeted and effective TMS therapies.

Methods

The core contribution of this paper is a novel TMS-electrophysiology recording system designed to overcome the significant artifact issues associated with TMS. Key methodological innovations include:

  • Gated Multi-Stage TMS-EEP Amplifier: This custom amplifier uses a solid-state analog switch (SSSW) to temporarily ground the input to the high-gain filter stages during the TMS pulse. This effectively attenuates the massive induction and electric field coupling artifacts, allowing for recording as early as 0.8-1 ms after TMS onset.
  • Electrical Shielding: A custom-designed, weakly conductive shield for the TMS coil was developed to attenuate electric field coupling without significantly impacting the magnetic output.
  • Low-Noise Miniature Coaxial Cables: These cables, featuring a semiconducting graphite layer, minimize triboelectric noise generated by vibrations from the TMS coil's click sound.
  • Inadvertent Charge Injection Measurement: A low-gain monitoring channel was incorporated into the amplifier to quantify any unwanted current injection into the brain, a critical factor for ensuring the safety and validity of the recordings.
  • Rodent Model Translation: The method was successfully translated to Sprague-Dawley rats, providing a widely accessible platform for studying basic neurophysiology and TMS mechanisms.

Datasets

The study utilized data from six male Sprague-Dawley rats for method evaluation and optimization, seven for ML experiments, and four for PA experiments. These animals were used to record electrophysiological data (EEP) and electromyography (EMG) signals in response to various TMS stimuli. No specific pre-existing datasets were used for training or evaluation; rather, the study generated its own experimental data.

Key Insights and Ideas

  • Direct Neuronal Access: The most significant insight is the successful development of a method that provides direct, in vivo electrophysiological access to neuronal activity within milliseconds of a TMS pulse. This opens a new window into understanding TMS mechanisms.
  • Time-Grained Neuronal Dynamics: The study revealed that TMS evokes complex, time-grained neuronal activity patterns, not just a simple excitation. This includes high-frequency spiking within the first few milliseconds and a multiphasic rhythm of excitation and inhibition lasting for hundreds of milliseconds.
  • Stimulus Orientation Matters: The orientation of the induced current in the brain significantly influences the short-latency neuronal responses. PA-oriented stimulation elicited robust, high-frequency spiking, potentially reflecting direct activation of corticospinal tract neurons, while ML-oriented stimulation evoked minimal early responses.
  • Multiphase Response: Beyond the initial milliseconds, TMS triggers a longer-lasting, multiphasic response characterized by excitation, inhibition, and rebound excitation. This suggests TMS engages complex network dynamics, potentially involving cortico-basal ganglia-thalamo-cortical loops.
  • Bridging the Gap: The findings in rodents show remarkable parallels with human TMS responses, suggesting that this animal model can effectively elucidate the neurophysiological underpinnings of human TMS applications.

Key Results

  • Early Neuronal Activity: The method successfully recorded neuronal activity as early as 0.8-1 ms after TMS onset, free from significant artifacts.
  • Short-Latency Spiking: PA-oriented TMS stimulation evoked robust, high-frequency neuronal firing in layer V of the motor cortex, with peaks at 1-1.5 ms and 3.2-4.2 ms after TMS onset.
  • Long-Latency Multiphasic Response: A consistent multiphasic response (excitation-inhibition-excitation) was observed in layer V neurons between 6 and 300 ms post-TMS, regardless of stimulus orientation.
  • Muscle Activation Correlation: TMS-evoked muscle activation in the contralateral forelimb was observed with latencies consistent with cortically evoked responses, correlating with neuronal activity.
  • Artifact Reduction: The developed method significantly reduced induction, electric field coupling, and vibration artifacts, enabling reliable EEP recordings.

Limitations and Future Work

  • Rodent Skull Size: The authors acknowledge that the TMS coil size relative to the rodent brain is a potential limitation for achieving fine spatial selectivity. However, they argue that careful coil positioning and the use of smaller, custom coils in future work can mitigate this.
  • Antidromic Spikes: Neuronal activity within the first 1 ms after TMS onset is not visible, meaning any antidromic spikes evoked by direct axonal activation could not be recorded.
  • Future Directions: The authors propose several avenues for future research:
  • Investigating the circuit selectivity of different stimulus orientations to pinpoint direct activation loci.
  • Elucidating the anatomical and functional organization principles of the M1 microcircuitry that drive high-frequency neuronal responses.
  • Extending the investigation to other cortical areas using TMS as a probe.
  • Further characterizing the neuronal mechanisms underlying the long-latency multiphasic response, potentially involving cortico-basal ganglia-thalamo-cortical loops.
  • Combining this method with optogenetics, transgenic tools, and computational modeling to gain deeper insights.
  • Applying this methodology to study the modulatory contributions of TMS in the treatment of neurological and psychiatric conditions.

Summary updated

Abstract

Transcranial magnetic stimulation (TMS) is a widely used non-invasive tool to study and modulate human brain functions. However, TMS-evoked activity of individual neurons has remained largely inaccessible due to the large TMS-induced electromagnetic fields. Here, we present a general method providing direct in vivo electrophysiological access to TMS-evoked neuronal activity 0.8-1 ms after TMS onset. We translated human single-pulse TMS to rodents and unveiled time-grained evoked activities of motor cortex layer V neurons that show high-frequency spiking within the first 6 ms depending on TMS-induced current orientation and a multiphasic spike-rhythm alternating between excitation and inhibition in the 6-300 ms epoch, all of which can be linked to various human TMS responses recorded at the level of spinal cord and muscles. The advance here facilitates a new level of insight into the TMS-brain interaction that is vital for developing this non-invasive tool to purposefully explore and effectively treat the human brain.
BibTeX
@inproceedings{li2017lifting,
  author = {Bingshuo Li and Juha Virtanen and A Oeltermann and Cornelius Schwarz and Martin A. Giese and Ulf Ziemann and Alia Benali},
  title = {{Lifting the veil on the dynamics of neuronal activities evoked by transcranial magnetic stimulation}},
  booktitle = {eLife},
  year = {2017},
  doi = {10.7554/elife.30552},
}
Figure