Guidance of Regenerating Peripheral Axon Subtypes Using Molecular Cues
Abstract
There are approximately 1.9 million amputees in the USA, and approximately 185,000 new amputations occur every year. Recent progress in robotic limbs provides the possibility of interfacing with the peripheral nerve to replicate natural sensory and motor functions of the human hand. While robotic prosthesis has advanced significantly, providing up to 22 degrees of freedom equipped with multiple sensors for position, temperature, and pressure, eliciting naturalistic sensation has yet to be achieved. Electrodes interfaced at the somatic peripheral nerves are intended to decode motor intent and elicit sensation through selective stimulation of sensory axons. However, current peripheral nerve interfacing technology elicits unnatural sensations through fascicular stimulation of mixed modality axon bundles. In addition, recording motor intent is more difficult due to the sparse arrangement of motor neurons. Therefore, peripheral nerve interfacing technology that selectively records motor activity and elicits naturalistic sensory percepts remains a challenge. This dissertation takes a unique approach to this problem by evaluating the use of molecular cues to selectively guide regenerating motor and sensory axons into Y-shaped conduits to modulate the sensory to motor ratio of regenerative neural interfaces. In Chapter 2 we tested the efficacy of single neurotrophic factors (NTFs) in modulating the sensory/motor axon ratio in using a Y-shaped nerve conduit. The functionality of the regenerated segments was confirmed by evoked compound nerve action potentials in 97% of the nerve fascicles. Retrograde labeling of ventral motor neurons showed that glial derived neurotrophic (GDNF) induced a 3-fold increase in motor neurons compared to bovine serum albumin (BSA) compartment, while sensory neuron regeneration showed no difference. The sensory/motor ratio however was not significantly altered by single molecular attractants. In Chapter 3 we evaluated the response of regenerating axons using motor axon guidance molecules brain–derived neurotrophic factor (BDNF)/GDNF in one compartment and sensory axon attractants nerve growth factor (NGF) or pleiotrophin (PTN) in the adjacent arm. The combination of BDNF/GDNF versus PTN showed a significant (p< 0.05) change in the sensory/motor axon ratio between the nerve fascicles. In Chapter 4 we evaluated the effectiveness of Semaphorin3A, a nociceptive sensory axons molecular repellent, in combination with BDNF, an attractant for mechanoreceptors, in one regenerative chamber compared to an adjacent one containing NGF, an attractant for nociceptive axons. Sema3A significantly reduced the number of small myelinated axons compared to the BDNF only compartment, but also affected the NGF chamber, indicating cross-diffusion of the repellent to the other arm. In summary, this body of work supports the notion that molecular attractant cues are able to influence the regeneration of functional sensory/motor axons in nerve fascicles using a Y-conduit in vivo, in a nerve amputee model. Furthermore, chemorepellent molecules were shown to induce an inhibitory effect on small myelinated axons for the first time in a peripheral nerve regeneration paradigm. Together, this work contributes to establish the feasibility of molecularly guided regenerative peripheral neural interfaces, and to define the potential of this technology towards achieving intuitive motor control and conveying natural feel to amputee users of advanced bionic limbs.