During the development of the nervous system, a complex system of chemical and mechanical cues guide nerve cell projections toward appropriate targets by eliciting attractive or repulsive responses from navigating structures called growth cones. Current in vitro models of neural guidance lack the capacity to provide multiple cues within the same substrate. This dissertation presents in vitro models with tunable presentation of multiple chemical and mechanical guidance cues in a single substrate for elucidating a more complete understanding of growth cone responses to the extracellular environment.

In the first study, our model utilized a light-sensitive agarose gel as the growth substrate and a polyethylene glycol gel that contained three dimensional neural growth in specific geometries. This model incorporated molecular cues as immobilized proteins and gradients of soluble factors in a quantifiable manner. The agarose gel presented issues with gelation and supporting neural growth, so we sought to improve upon our initial design by changing the growth substrate.

The second study employed a novel light-sensitive dextran gel in place of the agarose of the first study. The chemoattractant neurotrophin-3 and chemorepellant semaphorin3A were immobilized within this dextran gel in a spatially defined manner, and the response from growth cones quantified. The growth cones did not respond to the control protein NeutrAvidin, exhibited a moderate response to neurotrophin-3, and showed a strong repulsive response to semaphorin 3A.

The third study investigated neural responses to changes in substrate stiffness. Dextran gels with light-degradable crosslinks exhibited different elastic moduli based upon irradiation time. Specified regions of dextran gels were irradiated with light to present growth cones with a choice between two different elastic moduli. Growth cones exhibited preference for a narrow range of elastic moduli spanning less than 200 Pa, an observation unattainable with other in vitro models.

The results establish our models as possible platforms for observing and manipulating specific growth cone responses to both chemical and mechanical cues. Understanding how the growth cone interacts with its environment may lead to improved wound healing therapies, and expanding our model into a high-throughput assay could contribute to the development of these new treatments