Speaker
Description
Introduction: Tissue function depends on the intricate 3D organization of cells, matrix components, bioactive cues, and dynamic factors like nutrient and oxygen gradients, and mechanical forces. Advances in bioprinting have enhanced cellular organization, yet full tissue maturation, critical for biological function, often requires post fabrication measures. Here, we introduce a novel method to converge volumetrically bioprinted chip constructs with increasing channel complexity and a custom bioreactor to enable microfluidic perfusion. Establishing imaging-based analyses, we provide benchmarks to assess the developed platform and apply this workflow to the creation of a human mammary gland model. Here, the effects of perfusion underline the influence of mechanical stimulation on cell organization. Finally, to increase complexity of the model, we demonstrate the co-culture of two ductal structures—a mammary and an endothelial duct— as a step towards vascularized mammary models.
Methods: To fabricate the chips, volumetric bioprinting (VBP) was performed with gelatin methacryloyl supplemented with photoinitiator lithium phenyl-2,4,6-trimethylbenzoyl-phosphinate. Printing accuracy was assessed through computational nominal actual comparison of the printed object (from lightsheet microscopy images) to the original STL file. Cell seeding optimization used human umbilical vein endothelial cells (HUVECs) and MCF10A cells at varying densities (7.5–60 million cells/mL) with channel coatings of poly-L-lysine, collagen type I, or both. A custom platform enabled automated chip rotation, and cell coverage was analyzed from segmented confocal images. To enable dynamic culture, an autoclavable and transparent CNC-milled perfusion bioreactor was developed. Chips were seeded with MCF10A to assess monolayer formation and cell polarization. As a proof of concept, a dual-channel chip incorporating an MCF10A and an endothelial channel was fabricated.
Results and discussion: Chip constructs with 500-1000µm-diameter channels were printed with minimal deviations (1.94±2.96%) compared to the STL file (Figure 1A). Seeded chips were rotated for 6 hours (90°/10min), and specific coating and density conditions were optimized for both MCF10A and HUVECs (>80% channel coverage). Inclusion of the chips into the bioreactor enabled leak-free perfusion for 21 days, as well as efficient imaging without compromising sterility (Figure 1B). Channels seeded with MCF10A formed a tight epithelial monolayer when cultured dynamically compared to static controls. Furthermore, cell polarization was enhanced in perfused constructs, as observed by CK14 (basal) and CK8/18 (luminal) localization, and cell alignment could be controlled by changing the positioning of the print in the printing reservoir. In a two-channel chip, both endothelial and mammary epithelial cells could be cultured in independent channels, as close as 200 µm from each other, with the mammary channel possessing lobular structures to better mimic mammary duct architecture (Figure 1C,D).
Conclusions: We present a biofabrication pipeline combining VBP and dynamic perfusion in a custom platform. Using a mammary gland model, we highlight the impact of perfusion on cell organization and polarization, offering a valuable tool for studying healthy and diseased tissues. Open source, imaging-based workflows were developed for chip characterization, and the platform's modifiable digital models enable adaptation for diverse tissue engineering applications, advancing disease modeling and drug screening through native like dynamic culture.
64057822267
