QS2: DIY Bioreactors for Tissue Engineering - An Ultra-low-cost Platform for Perfusion Cell Culture and Live Imaging

Abstract

Purpose: While tissue engineering offers the promise of revolutionary innovation, scalable three-dimensional tissue culture is limited by the diffusion of nutrients and oxygen making media perfusion obligatory. Unfortunately, the cost of bioreactors for large construct tissue culture can be prohibitive, with a typical perfusion chamber costing several thousand dollars, and even small petri-dish-sized devices costing hundreds of dollars each. We have developed a low-cost perfusion setup that seals collagen-based perfusable cellular constructs within a sterile PDMS well between coverslips, allowing for repeated live-imaging of perfused 3D engineered tissues. Herein we describe fabrication of this novel system and validate its utility. Methods: Molds and frames were designed on 3D-modeling software (Fusion 360) and printed on a Prusa i3 MK3S 3D printer in poly(lactic acid) (PLA). Molds were filled with poly(dimethyl siloxane) (PDMS), which was cured to form chambers, bubble traps, mason jar lid chambers, and media reservoir lid adapters. In total, the tissue culture chamber device, mason jar lid inset, media reservoir lid, and bubble trap require 4, 1, 2, and 4 unique printed components, respectively. Results: Each perfusion chamber can be assembled for under 8 USD per device and reused repeatedly. The current model has a tissue chamber custom-built with 18x10x4 (LxWxH)mm3 dimensions, but this chamber can be readily customized to experiment-specific dimensions. These devices allow cellular hydrogel constructs to be maintained in a sterile environment after assembly, perfused at varying rates to expose cells to different levels of shear stress, and the cells can be intermittently imaged with light, fluorescent or confocal microscopy - an unparalleled benefit for monitoring of experiments and collection of timepoint imaging data. The perfusion circuit consists of autoclavable glass and PDMS components, including a bubble trap, a crucial component of the circuit for preventing air bubbles that can damage cells and block microchannels, and a lid adapter, which allows 50 mL conical tubes to serve as self-oxygenating media reservoirs. Media changes can be performed via peristaltic pump perfusion or with syringe-based cell culture techniques for static culture. Constructs have been perfused within standard incubators for up to 14 days demonstrating normal cell viability without contamination or evidence of infection, with longer perfusion culture intervals (>1 month) currently being tested. Conclusion: The increasing accessibility of 3D-modeling and 3D-printing has enabled rapid prototyping of devices to address the problems that we face as surgeon-scientists. We have developed a low-cost tissue-engineering perfusion circuit that facilitates 3D-tissue-culture while allowing for repeated live-imaging as the cultured tissue develops. The instructions for our setup can be utilized to replicate our devices in other labs, and these designs can be readily customized to meet the needs of specific experimental aims.