Mapping convective and diffusive transport in 3D printed vascularized tissues

摘要

Introduction: 30 printing provides a promising method to establish complex, customizable vascular networks within engineered tissues. Though computational modeling offers unique understanding of essential parameters within complex tissues, few research groups first verify their computational models in vitro or in vivo, a necessary step for using computational models to aid in the design and production of 3D printed tissues.In this work, we begin to verify computational models of 3D printed vascularized tissues to understand the role of vascular architecture in altering the cellular environment. Methods: Using an open-source stereolithography printer developed in our lab, we printed PEG-DA gels containing branched micro-channel networks. To assess convective transport through the channels, we tracked the flow of fluorescent beads using an epiftuorescent microscope and PIVlab software. To map diffusion rates of different sized molecules into the gels, we assessed the transport of fluorescently labeled dextrans (10 kDa and 3 kDa) along with methacrylated rhodamine into the gels. Finally, to assess cell viability following the printing process, we printed gels containing 293T cells transduced to express Luc2P. We then measured the luminescent output of the gels in the presence of luciferin after 5 days in culture. To develop parallel computational models for these gels, we used a 3D model of our printed channel networks to develop computational models for flow and diffusion through the gels (Fig 1). Results and Discussion: Results from bead tracking data for 5 printed gels indicate that average flow rates varied noticeably between channels, with the first and last channels having the highest flow rates (Fig 1). After normalizing the inflow rate of our computational model to that of our printed gels, the computational models corresponded strongly in vitro data (Fig 1). These results show the capability of computational modeling to demonstrate non-obvious flow patterns in 3D printed tissues along with the necessity for in vitro verification. For initial diffusion data, trends show that small molecules diffused rapidly (within 10 min) to distances of 200 μm, while larger dextrans took upwards of 60 min (Fig 2). These results demonstrate the need for improved transport mechanisms to deliver proteins and larger molecules to interstitial cells, possibly through the use of superphysiological flow rates or the addition of capillary networks. Finally, after 5 days in culture, a strong luminescent output was still detected from printed cells, suggesting that cells can survive the initial printing process and may be viable in long term culture. For future work we will compare computational and experimental models for diffusion in printed gels, and will develop computational models for cell viability. With this groundwork, computational models are an invaluable resource in making high throughput optimizations for the design of 3D printed tissues.