Microelectrode cluster technology for precise interactions with neuronal circuits. Towards highly specific adaptive deep brain stimulation.

Neuro-electronic interfaces, which can be used for stable communication between neurons-computers over long periods of time, would be valuable for understanding and interacting with the nervous system. A major challenge has been to overcome the tissue reactions towards implanted electrodes. Flexible microelectrodes that cause less implantation injury and which can follow the micromotions of the brain have been considered as a solution to achieve stable neuronal recordings and stimulations. The aim of this thesis work was therefore to develop and evaluate biocompatible neuro-electronic interfaces, as well as introduce new implantation methods which together allow stable recordings and spatially precise stimulation of the brain
To this end, we have developed a new generation of ultrathin flexible electrode arrays based on 12.5 µm thin wires embedded in a gelatin vehicle providing structural support during implantation. The gelatin embedded electrodes were implanted in rat brains via a narrow track line and spread out as a cluster in the target area. In the first study, we evaluated the performance of the neural recordings for eight weeks with respect to impedance, signal amplitudes and noise levels. We found impedance, and signal to noise ratio of single units to be quite stable, suggesting high biocompatibility. In the second study, we developed a gelatin embedded microelectrode array consisting of 16 microelectrodes, distally equipped with silicone cushions to reduce vascular damage. This array was implanted medial to the subthalamic nucleus, in 6-hydroxydopamine lesioned rats (a classical animal model for Parkinson’s disease), and the effects of deep brain stimulation were evaluated for 6 weeks. Stimulation with subsets of 4-8 electrodes evoked specific motor behaviors in all the tested rats. Depending on the exact electrode combination, stimulation elicited either improvement of locomotion, or grooming and rearing, increased turning, dyskinesia, or no movement. These results suggest that improved stimulation specificity can be obtained by choosing the right group of electrodes from the cluster. In the third study, we hypothesized that reducing the tissue resistance during the insertion of the electrodes would minimize the implantation injury. To address this problem, we coated gelatin embedded needles with a layer of ice, which on melting, provided a super slippery surface during insertion into the brain. The addition of a layer of melting ice decreased the insertion force by approximately 50%, significantly reduced neuronal loss, as well as the astrocytic response, but did not have any obvious effect on microglial activation.
In conclusion, this thesis presents a novel design for implantable and biocompatible neuro-electronic interfaces comprising highly flexible microelectrodes rendering stable recording properties and improved stimulation specificity. In addition, a novel implantation vehicle was developed to reduce the acute tissue reactions in response to the implantation