An Experimental Platform to Study Axonal Guidance and Function

Studying neurons and axons in vivo poses a significant challenge due to the complex nature of the nervous system. The living environment makes it difficult to observe and manipulate individual axons effectively. In response to these limitations, researchers turn to in vitro models to gain insights into axon function and development from the bottom up. Our approach involves culturing neurons within polydimethylsiloxane (PDMS) microstructures on multielectrode arrays (MEAs) [1], [2], providing a controlled environment for studying axonal growth and activity. By utilizing these easily controllable and tunable systems, we aim to uncover the spatial limitations of axonal growth and determine the optimal spatial constraints for promoting healthy and robust axonal growth.<br/>In this work, we study axonal growth dynamics using rat primary neurons cultured in custom-designed PDMS microfluidic devices containing nanochannels [3] of minimal size 100 nm. To generate the nanoscale features, a thin 350 nm layer of silicon dioxide (SiO 2 ) was deposited on a silicon wafer followed by a thin layer of photoresist that was then patterned using E-beam before the structures were etched in the SiO 2 using reactive ion etching. After stripping the remaining photoresist, the two subsequent layers were patterned in SU8 using conventional photolithography. The resulting microstructure design allows for separation of cell soma from axons and thus studying axons in detail. By varying the nano-/microchannel size, we show the spatial limitations for axonal growth and observe changes in axonal guidance depending on the channel width. We note that there is almost no penetration in the channels below 400 nm in width. Furthermore, we analyze the formation and structuring of axonal bundles, their dependence on cell density, and on the type of spatial constraint. We observe more frequent formation of bundles in cases where we topologically force the narrowing. The opposite is seen in the channels that are orders of magnitude larger than the average axon size, where they tend to spread along the available space. We also see that bundling is present in the nodes with lower cell density. We also see that axons grow the longest in the channels of 10-12 µm width range. Finally, we assess neuronal activity by culturing neurons in PDMS microstructures on MEAs. The temporal precision of action potentials recorded by MEAs in combination with spatial precision of axonal growth within the PDMS, enabled to analyze action potential fidelity, e.g. it’s velocity, shape and amplitude depending on the the number of cells, the width of the formed axonal bundle and the age of an in vitro culture.<br/>Looking forward, we aim to improve the system further by modifying substrate stiffness to make it more physiologically relevant, by adding hydrogels into the microstructures. The potential to expand our findings on human induced pluripotent stem cell (hiPSC) derived neurons and the diseased cells, offers yet another application. Additionally, a promising direction is the investigation of axon fasciculation [4] by alternating the angle at which axons approach one another with PDMS. With its wide range of capabilities, this highly tunable system holds promise for unlocking new insights and advancing our understanding of axon function and pathology.<br/><br/>References<br/>[1] J. Duru, J. et al, Frontiers in Neuroscience, 2022<br/>[2] S. J. Ihle et al, Biosensors and Bioelectronics, 2022<br/>[3] J. Mateus and S. Weaver et al, ACS Nano, 2022<br/>[4] M. A. Breau and A. Trembleau, Semin Cell Dev Biol, 2022