Investigation Of Axon Growth In Asymmetric Microfluidic Channels

Abstract

Event Abstract Back to Event Investigation Of Axon Growth In Asymmetric Microfluidic Channels Alexey Pimashkin1*, Eugene Malishev2, Anton Bukatin2, Arseniy Gladkov1, Yana Pigareva1, Victor Kazantsev1 and Irina Mukhina1 1 N.I. Lobachevsky State University of Nizhny Novgorod, Neurocenter, Russia 2 Saint-Petersburg National Research Academic University of the Russian Academy of Sciences, Russia Motivation Microfluidic chips combined with microelectrode arrays are widely used for neural culture morphology manipulation and investigation of functional connectivity [1,2]. Specific designs of microfluidic channels which couple neural sub-populations provide an unique approach to model brain areas, study interaction and coupling of various cell types and formation of functional signalling pathways in the brain. Materials and methods Microfluidic devices consist of two chambers and microchannel arrays, providing unidirectional axon growth between Source and Target neural networks (figure 1A). Microchannels’ design is based on a sequence of various triangular segments: for backward growing axons each microchannel broadening with subsequent bottleneck significantly reduces the outgrowth probability, while the number of axons, reached the "Target" chamber, is enough for signal propagation due to "guiding" effect of convergent walls (figure 1B, C) [3]. PDMS microfluidic chips were fabricated by two layer lithography and PDMS molding techniques (see figure. 2.a). Mold design contained: first 5 µm-thick layer, which formed microchannel structure and second 50 ?m-thick layer, which formed chambers. For mold fabrication, Silicon wafers and negative photoresists SU8 (MicroChem, USA) were used. Microchannels’ structure was based on several types of segments (see figure 1D) and their length varied from 600 µm to 700 µm (3 - 11 segments) which is enough to provide a growth only for axons. Each PDMS chip was positioned and mounted onto the surface of a planar microelectrode array (MEA, Multichannel systems, Germany) with 60 electrodes of 30 µm in diameter, so as to locate several electrodes in the microchannels. Before cell plating the device was covered with polyethyleneimine. Dissociated hippocampal neurons were plated into separate subcompartments (chambers). Electrical activity was recorded after 7 days in vitro by USB-MEA system (Multichannel systems, Germany). Results First we studied axon growth dynamics in various microchannel shapes in order to find optimal design for unidirectional connectivity between neuronal sub-populations (figure 1 D). We proposed three types of segment shapes and each type was implemented with three segment lengths of 67 µm, 100 µm and 200 µm. and two diameters of “bottleneck” - 5 ?m and 7 µm. Each day after plating we analyzed individual neural branches in microchannels. Axons passed the “bottleneck” and in most cases grew alongside the sidewalls of the segments. Smaller and narrow segments showed better results in growth from source to target chamber, axons preserved the direction in most of the segments. Next we examined axons growth in “backward” direction during first three days. At least one segment at the end of microchannel of any shape was filled with axons. After 5-7 days in vitro whole microchannels were filled with axons and we suggested that the neuronal subpopulations formed morphological connections. To estimate an efficacy of microchannel to guide axons in preferred direction we measured the average distance of “backward” growth of axons from “target chamber” while its distinguishable from the axons originated from the “source” chamber. We found that “zig-zag” shaped segments with 5 and 7 µm “bottleneck” and 100 µm length segments were the most effective. In general, 67 and 100 µm segments of all types showed similar results, while large and wide segments were least effective (figure 2 B). Next, we tested synaptic connectivity between chambers using microelectrode arrays. After 20 DIV spontaneously emerged bursts of spikes in the "Source" chamber consequently evoked the bursting activity in the "Target" chamber through axons in the microchannels of narrow “straight” segment shape (figure 2 C). Discussion and Conclusion In this study we developed microfluidic chips with dual chambers coupled by microchannels for axon outgrowth between sub-populations of dissociated hippocampal neurons. We found optimal design of the shapes of microchannels to achieve unidirectional axon growth and spiking activity propagation between isolated neuronal clusters (figure 2). Such microfluidic device can be used in the research of neural network dynamics in a realistic morphology, network-wide synaptic plasticity and interaction between various cell types to study higher cognitive functions in the brain.