Abstracts

Rationale:
The Responsive Neurostimulation System (RNS; NeuroPace) & the Deep Brain Stimulation (DBS; Medtronic) System are the only direct brain modulation therapies approved by the U.S. FDA for the treatment of intractable partial-onset epileptogenic networks. The depth electrodes’ neural volume of cortical activation (VOCA) in grey matter determined by the activation function (AF) only extends ~4mm from the surface of the stimulating electrode contacts. We propose that extending the radius of the AF by 1-2mm around a depth lead implanted in white matter can markedly increase the VOCA while modulating distant neural targets. One method to increase the VOCA is to enhance tissue conductivity (σ), therefore reducing tissue impedance adjacent to the electrode. Our previous work demonstrates injecting fluorescein-thiosemicarbazide (FTSC)-functionalized metallic carbon nanotubes (fCNTs) adjacent to the depth lead alters tissue impedance with stable long-term neural localization & biocompatibility in vitro. This in situ replication study uses: 1) electrical impedance spectroscopy to evaluate fCNT-induced changes in σ in a 0.6% agarose gel; 2) in situ immunohistochemistry after fCNT injection to evaluate long-term tissue localization and stability; 3) computational Hodgkin and Huxley (HH) cable modeling.    
Method:
To assess changes in σ, electrical impedance spectroscopy (50Hz-1MHz) was carried out in a 0.6% agarose gel in which an 8-contact platinum cylindrical depth lead (10mm center-to-center) was embedded. Nine 2µl aliquots of metallic fCNT (250 µg/ml) were injected into the gel, spaced perpendicularly at 1mm intervals (FIG 1A). Gel impedances were measured again after injection. Impedance values between controls & fCNT samples were compared. Immunofluorescent fCNT cellular uptake staining was performed in this study to observe localization of 99%-pure fCNTs injected stereotactically into the hippocampus of F344 rats, 4wks before sacrifice. Astrocytes & neurons were immunolabeled with fluorescent GFAP/Cy5(λ=665nm) & anti-synaptophysin(λ=405nm) in fresh frozen 20µm thick brain sections, respectively. fCNT clusters were localized using fluorescence microscopy(100x). Finally, a HH impulse-propagation cable model was used to model the effect of fCNT-induced σ changes in the axonal compartment. Changes in action potential production & propagation velocity were analyzed. 
Results:
fCNT cytotoxicity studies using human astrocytes (HA) were completed in our previous work. HA were cultured, then treated with raw and fCNTs of 90, 95 & 99% purity without evidence of cytotoxicity. Impedance measurements showed a statistically significant difference between the fCNT sample and controls (p< 0.05). The HH cable model showed tissue σ changes, enhanced the induced E-field surrounding the electrode, increased axon excitability & propagation velocity, & E-field inhomogeneities (FIG 1B&C). Immunostaining revealed FTSC-fCNT clusters in the hippocampus with fCNT uptake by both neurons (FIG 2A; arrows) & astrocytes (FIG 2B) adjacent to the electrode.
Conclusion:
1) fCNTs are biocompatible following targeted injection into the brain & remain anchored long-term in neural tissue, while chronically modifying the biophysical properties of neural cells. 2) This study replicates & strengthens the hypothesis that neural tissue impedances can be chronically altered at the cellular level. Such a strategy can form novel bridges with distant neural targets & maximizes modulation of extensive epileptogenic networks. 3) Additionally, the biophysical properties of CNT can enable the delivery of greater charge densities while maintaining safety parameters of direct brain stimulation therapy.
Funding:
:Foglia Family Foundation, Mary Keane Fund.