Hemi-parkinsonian Rat Experiments

Our research using animal models to study the motor circuit and deep brain stimulation (DBS) is divided into two parts: (1) behavioral studies of DBS targeting the globus pallidus interus (GPi) and (2) recording from GPi and the motor cortex (M1-Layer V) in rodents during a trained behavior.

The GPi (i.e. entopeduncular nucleus in rats) is the main output nucleus of the basal ganglia (BG), which is associated with a variety of functions including motor performance and cognition. The GPi is one of the primary targets of DBS in patients with Parkinson's disease (PD), along with the STN. However, the therapeutic mechanism of DBS is poorly understood and rodent models of GPi-DBS have not been previously characterized. Cognitive side effects, such as impulsivity and depression, of DBS treatment for PD are known, but their relationship to the efficacy of the treatment is not well explained.

We have assayed rodents (n = 21) in a variety of different trained/untrained behaviors, including a reaction-time lever pressing task, rotation test, open field test, cylinder test, and sucrose preference test. These behaviors were performed on naive, hemi-parkinsonian and/or hemi-parkinsonian with GPi-DBS rodents.

The goal of this work is to illuminate the effects of GPi-DBS on the relationship between motor and cognitive function. In this work, we have studies the motor performance of the rodents in multiple behaviors and correlated it with several measures of impulsivity and depression. These results directly establish a connection between stimulating the GPi, motor performance and cognition. Additionally, for the first time it is verified that GPi-DBS is effective in the rodent model.

The second part of our research involves recording unilaterally from the GPi and bilaterally from M1-Layer V (output layer) in rodents that are performing a trained task. We have designed a behavior which isolates movement of one forelimb and are recording while the rat behaves. We have implanted microdrives with 14 recording tetrodes on naive rodents and recorded while they are healthy. Current work is to lesion the rats through injection of 6-OHDA via a cannula positioned in the medial forebrain bundle and continue recording while the rat performs the same task two weeks following the 6-OHDA injection.

Carbon Nanotube Fiber Electrodes

Implantable electrodes are the basic tool for the development of brain machine interfaces and DBS systems. Small surface area ( < 4000 um^2), flexible microelectrodes are desireable for high-resolution stimulation with minimal damage at the electrode-tissue interface. Carbon nanotube fibers (CNTf), manufactured in the Pasquali lab from carbon nanotubes, combine the specific electrical conductivity of metals with the typical specific strength of carbon fibers.

In collaboration with the Pasquali lab, we have developed a fabrication method for stereotrodes made from 40 um diameter CNTfs which are coated in an insulating elastomer (2 um layer), leaving only the tips exposed. In vitro and in vivo have been conducted to characterize the electrochemical stability of the electrodes. Over a week of chronic current pulsing in phosphate buffered saline (PBS), we have shown with impedance spectroscopy and cyclic voltammetry that the CNTfs are stable. They are also shown to have a larger charge storage capactiy and water window relative to platinum iridium (PtIr) stereotrodes of similar size. These stereotrodes have been implanted in the GPi of a cohort of rats (n = 4) for 6 weeks and voltage transients were recorded regularly over this time period to confirm in vivo stability.

The rats implanted with CTNf stereotrodes were unilaterally lesioned via a 6-OHDA injection in the hemisphere ipsalteral to the electrode. A PtIr electrode was implanted contralerally and voltage transient measurements were collected for this electrode for comparison. The rats were subjected to a rotation test with methamphetamine and we confirmed that behavioral results with this new technology were identical to previous results of performing DBS with PtIr stereotrodes over a wide range of frequencies (85 - 175 Hz).

Post-mortem histological analysis was performed to investigate signs of inflammation, blood brain barrier disruption, and gliosis. We hypothesize that, due to its mechanical flexibility, the CNTf flex with micromovements of the brain, thus causing minimal damage to the surrounding neural tissue.

Basal Ganglia Computational Modeling

Deep brain stimulation uses electrical stimulation to modulate neural activity in order to reduce motor symptoms associated with Parkinson's disease. The design of the electrical stimulation signal used is strongly linked to the efficacy of such a treatment. We construct computational models of core brain structures impacted by Parkinson's disease (STN, GPi, GPe, Thalamus, and M1) which are modulated, directly and indirectly, by electrical current injections from chronically implanted electrodes as a part of deep brain stimulation.

These computational models serve as a testbed for novel stimulation signal designs. Our particular interest is in investigating the use of current pulses that do not occur at a fixed rate. Although regularization of neural activity has been linked to therapeutic benefits of DBS and occurs as a result of regular inter-current pulse intervals, this technique highjacks brain structures so that useful information cannot be passed through the system. We have shown that using a high frequency signal with small random deviations in pulse timing can simultaneously partially induce regularized activity associated with improved motor symptoms and permit greater variation in neural responses, which are necessary to encode neural information.