Full range physiological mass transport control in 3D tissue cultures

Abstract

We report the first demonstration of a microfluidic platform that captures the full physiological range of mass transport in 3-D tissue culture. The basis of our method used long microfluidic channels connected to both sides of a central microtissue chamber at different downstream positions to control the mass transport distribution within the chamber. Precise control of the Péclet number (Pe), defined as the ratio of convective to diffusive transport, over nearly five orders of magnitude (0.0056 to 160) was achieved. The platform was used to systematically investigate the role of physiological mass transport on vasculogenesis. We demonstrate, for the first time, that vasculogenesis can be independently stimulated by interstitial flow (Pe > 10) or hypoxic conditions (Pe < 0.1), and not by the intermediate state (normal living tissue). This simple platform can be applied to physiological and biological studies of 3D living tissue followed by pathological disease studies, such as cancer research and drug screening.

Fig. 1

(A) Schematic illustrating the use of a long microchannel to control the physiological environment in a microtissue chamber. The long microfluidic channel creates a large range of pressures. The microtissue compartment can be positioned at different locations along the microfluidic channel resulting in either a large (left chamber) or small (right chamber) pressure drop across the tissue and thus a large or small convective flow. Capillary morphogenesis in the necking channel (B₁) adjacent to microchamber and 3–4 mm away (B₂) is suppressed. (C) The capillary burst valve design for the communication pore, (D) one of the microfabricated microplatform and media reservoirs.

Fig. 2

Conceptual illustrations of the 3D microfluidic tissue model with 8 communication pores, where cell construct is formed by fibrin gel (A1) seeded with NHFLs (A2) and ECs (A3) and loaded through necking channel (A4) and wedge-shaped channel (A5). The physiological conditions are controlled by the magnitude of pressure facing into different communication pores (A6) along the microtissue through the fluidic side channels (A7–A8).

Fig. 3

Microfluidic configurations and the pressure fields of the convection-dominated microplatform for generating high interstitial flow (Pe > 10). The interstitial flow can be controlled to be aligned in the (A_1–3) transverse direction or in the (B_1–3) longitudinal direction (black lines are streamlines, and black arrows indicate the direction of interstitial flow). The microfluidic channel and microtissue compartment can also be configured to create a pressure field in which (C_1–3) convection and diffusion are of similar magnitude (Pe ~ 1), or (D_1–3) the transverse pressure gradient is zero, and the longitudinal pressure gradient is near zero and thus diffusion dominates (Pe < 0.1). Figures numbered with subscripts 4 and 6 are experimental patterns of dextran flowing into each microtissue chamber, subscripts 5 and 7 are the corresponding numerical simulation results. CD31-labeled endothelial cells after 10 days of culture demonstrate variable capillary morphogenesis at 40X (A₈, B₈, C₈, D₈), where scale bars are 500 µm. Fluorescent images of the microtissue stained with hyproxyprobeTM-1 under 5% oxygen (A₉, B₉, C₉, D₉) are compared to the positive hypoxic. control 1% oxygen (e1). (e) The histogram of hyproxyprobeTM-1 for the 5% oxygen and the 1% oxygen condition (gray line) in log scale.

Fig. 4

Comparison of (A) total capillary length, (B) total capillary segments, and (C) total number of nuclei as a function of the Pe. Blue diamonds and red circles are data from platforms with 8 and 16 pores, respectively. At Pe >10, filled and open symbols correspond to transverse and longitudinal interstitial flow, respectively. The x-axes are in log scale.

Fig. 5

(A) and (B) are micrographs of microvasculature developed for 21 days by using interstitial flow dominant microplatform shown in Fig. 2A [transversely] and Fig 2B [longitudinally]. (C) and (D) are cross-sectional confocal images of developed microvessels of (A) and (B) at red lines. The scale bars are 500 µm (A) and (B), and 40 µm for confocal images (C) and (D).