Abstract
Ex vivo tissue culture can model tissue physiology under well-controlled conditions and is especially promising for understanding the complex mechanisms of the brain. Three-dimensional (3D) printing has immense potential to accelerate microfluidic technology development, especially for ex vivo tissue culture devices where miniaturization is ultimately limited by the physical dimensions of tissue explants. Here we describe the development of a 3D printed microfluidic perfusion device for ex vivo brain slices that utilizes media droplets segmented by oxygen bubbles, a perfusion technique we call "bubble perfusion". Device design considerations are described, including materials property challenges associated with 3D printed plastic, such as wetting behavior and thermal conductivity challenges. Integrating a heated water circulation chamber and media prewarming chambers yielded media droplets delivered to brain slice explants at a temperature of 36.8 ± 0.13 °C, with tissue experiencing a temperature drift of 0.5 ± 0.09 °C over the course of a 60 s media droplet exposure. Murine brain tissue explants containing the suprachiasmatic nucleus (SCN) or entorhinal cortex (EC) were observed to be viable within the perfusion system by fluorescence imaging of intracellular Ca(2+) flux induced by single-droplet stimulus of 60 mM KCl. Robust Ca(2+) flux was observed for perfusion experiments lasting up to 12 h, with sequential droplet observations indicating the temporal dynamics of Ca(2+) responses. End-point propidium iodide staining was used to characterize the health of EC and SCN tissue, with ca. 60% of cells in both regions showing no sign of membrane damage after 12 h of perfusion. The utility of the perfusion system toward pharmacological studies was demonstrated by comparing the Ca(2+) flux induced by stimulus with 50 μM cannabidiol (CBD) vs 50 μM anandamide (AEA). Interestingly, similar magnitude and temporal dynamics of Ca(2+) flux were observed for both CBD and AEA stimuli despite differential proposed mechanisms of action with respect to the CB1 receptor. These studies demonstrate the utility of the 3D printed bubble perfusion system toward the study of receptor-binding ligands that induce relatively modest magnitudes of Ca(2+) flux.