In this proposal we leverage our recent developments in MRI acquisition methods and new animal models to attempt direct detection of neuronal currents with Magnetic Resonance Imaging (MRI). If successful, this project will have a high impact for scientific investigations of the nervous system, forming a powerful new tool for human neuroscience by supplementing the commonly applied hemodynamic based MR blood oxygenation level dependent (BOLD) functional MRI (fMRI) technique. Although fMRI is a primary tool for human neuroscience, it has severe shortcomings primarily arising from the complex coupling between neuronal activity (electrophysiology) and the hemodynamic response. Although the two are correlated, it is clear that information is lost when viewed through the hemodynamic filter. In short, we seek a true MR based non-invasive electrophysiology. Other non-invasive imaging modalities such as Magnetoencephalography (MEG) and Electroencephalography (EEG) do directly measure the neuronal activity in the form of the electric and magnetic fields produced by post-synaptic currents. While proven clinical and research tools, they also have limitations. Because they detect the field outside the head, they are sensitive only to large groups of neurons firing coherently with certain total post-synaptic current distributions. For example MEG is not sensitive to currents oriented radially with respect to the spherical head. The currents derive from pyramidal cells oriented normal to the cortical surface and the non-radial constraint causes multiple blind spots where the cortex folds in such a way that the pyramidal cells are oriented radially. EEG shares a similar problem. Finally, both modalities share limited spatial resolution complicated by an ill posed inverse problem which makes it impossible to determine the source locations without assumptions about their nature and distribution. The net effect is a modality with direct access to the temporal dynamics, but limited imaging ability compared to MRI. In this proposal, we outline a research project that pursues a novel MRI technique for the direct detection of neuronal currents. Its successful development would bypass many of the problems of MEG and EEG by probing the neuronal magnetic fields locally within each MRI voxel thus providing the high spatial resolution without the complications of the hemodynamic filter. Many attempts using MRI have been pursued over the last 15 years without clear success. We propose a new MR method based on the ability of a spin- lock MRI sequence (stimulus-induced rotary saturation: SIRS) to interact resonantly with magnetic fields oscillating at 10-100Hz. The sensitivity to oscillating fields is a significant improvement over other MR detection methods which require un-realistic DC fields. While we have shown that the sequence's minimum detection threshold is as low as the previous methods, it is nevertheless an exploratory proposal that will benefit from a concerted effort and a strong, readily controllable stimulus. Thus, we plan to test our method using an optogenetic rat model we have previously used for fMRI. In this preparation, neurons in somatosensory cortex are transfected to express ChannelRhodopsin-2 (ChR2) channels that excite neurons when illuminated. We show that optical stimulation produces extremely strong electrophysiological neural signals in synchrony with controlled optical stimuli. The optogenetic model is MR compatible allows millisecond control of the activation and elicits strong coherence of the neuronal firing compared to other stimuli. We propose a focused investigation using carefully controlled animal experiments to avoid contamination with the standard hemodynamic signal. For example, using both electrophysiology and fMRI, we demonstrated that the complete elimination of concomitant BOLD activation can be achieved via carbogen inhalation (CO2/O2) in rat models. As a central focus of this proposal, SIRS will be implemented and used to detect neuroelectric signals in the rat models transfected with ChR2 genes. Particular attention will be paid to remove confounders such as hemodynamic MRI signals in conjunction with independent electrophysiology to verify the characteristics of neural activities induced by optogenetics. The output of this 2 year project will be a validated MRI method that can then be used to quantify and spatially map the electrophysiological activity.
Aim 1 : Optimize carbogen (5% CO2/ 95% O2) dose in optogenetically modified rats to eliminate BOLD activation while maintaining the neuroelectric activity in response to optogenetic stimuli. We hypothesize that: Inhalation of carbogen removes the BOLD activation induced by optogenetic stimuli; Inhalation of carbogen does not affect the electrophysiological activity induced by optogenetic stimuli.
Aim 2 : Application of the Stimulus Induced Rotary Saturation (SIRS) acquisition to optogentically modified rats using long block duration experiments capturing neuronal current effects. We hypothesize that: During systemic hyperoxia/hypercapnia, optogenetic stimuli produce SIRS-based MRI signal; The optical stimulation amplitude is proportional to the neural activity; Optical stimulation with unmatched frequency does not result in SIRS-based neural signal.
Although fMRI is a primary tool for human neuroscience, it has severe shortcomings primarily arising from the complex coupling between neuronal activity (electrophysiology) and the hemodynamic response. Although the two are correlated, it is clear that information is lost when viewed through the 'hemodynamic filter.' Using recent advancements in MRI and neuroscience technology, we seek a true MR based 'non-invasive electrophysiology' for the direct detection of neuronal currents.
| Shim, Woo H; Suh, Ji-Yeon; Kim, Jeong K et al. (2016) Enhanced Thalamic Functional Connectivity with No fMRI Responses to Affected Forelimb Stimulation in Stroke-Recovered Rats. Front Neural Circuits 10:113 |
