Modeling angiogenesis in the human brain in a tissue-engineered post-capillary venule

Abstract

Angiogenesis plays an essential role in embryonic development, organ remodeling, wound healing, and is also associated with many human diseases. The process of angiogenesis in the brain during development is well characterized in animal models, but little is known about the process in the mature brain. Here, we use a tissue-engineered post-capillary venule (PCV) model incorporating stem cell derived induced brain microvascular endothelial-like cells (iBMECs) and pericyte-like cells (iPCs) to visualize the dynamics of angiogenesis. We compare angiogenesis under two conditions: in response to perfusion of growth factors and in the presence of an external concentration gradient. We show that both iBMECs and iPCs can serve as tip cells leading angiogenic sprouts. More importantly, the growth rate for iPC-led sprouts is about twofold higher than for iBMEC-led sprouts. Under a concentration gradient, angiogenic sprouts show a small directional bias toward the high growth factor concentration. Overall, pericytes exhibited a broad range of behavior, including maintaining quiescence, co-migrating with endothelial cells in sprouts, or leading sprout growth as tip cells.

Keywords: Angiogenesis; Angiogenic sprout; Blood-brain barrier; Pericytes; Tip cell; Tissue engineering.

Fig. 1

Growth factor perfusion promotes angiogenesis in single-channel iBMEC microvessels. a Schematic illustration of single-channel iBMEC microvessel fabrication. ECM, extracellular matrix; iBMEC, induced brain microvascular endothelial-like cells. b Schematic illustration of growth factor perfusion in the microvessel and the initiation of angiogenesis. iBMECs, induced brain microvascular endothelial-like cells. c Phase/fluorescence overlay images following perfusion of an iBMEC microvessel with 20 kDa dextran-FITC at different time points. (Right) Fluorescence intensity profile at the locations indicated by the dotted vertical arrows. d Representative images of RFP-iBMEC microvessels after perfusion with 10 ng mL⁻¹ bFGF at different time points. e Number of sprouts per field of view in a 10 × image. n = 3 images across 3 microvessels. f Length of angiogenic sprouts. n = 27 sprouts for day 1 and n = 32 sprouts for day 2. g Confocal images of angiogenic sprout growth and anastomosis with neighboring sprouts. Confocal images at day 1 and day 4 were taken from different regions. h Confocal image showing the lumen structure of angiogenic sprouts in iBMEC microvessels on day 4, 30 min following perfusion with 500 kDa dextran-FITC (green). i Immunostaining of sprouts for BMEC markers on day 4 following perfusion with 10 ng mL⁻¹ bFGF

Fig. 2

Influence of growth factor gradient on angiogenesis in a iBMEC microvessel in a triple-channel device. a Schematic illustration of the process for fabrication of triple-channel iBMEC microvessel model with source and sink channels to establish a growth factor gradient. b Schematic illustration of the triple-channel device for establishing a growth factor concentration gradient. The source channel (top) is perfused by growth factors and the sink channel (bottom) maintains a gradient between them. The microvessel is located between the source and sink channels. c Phase/fluorescence overlay images showing the distribution of fluorescently-labeled 20 kDa dextran-FITC after 1 h, 24 h, and 48 h. The source channel was perfused with medium supplemented with 10 ng mL⁻¹ bFGF and 1 μM 20 kDa dextran-FITC. The central channel was seeded with RFP-iBMECs (Red). d Fluorescence intensity profiles between the source and sink channels obtained from fluorescence images following perfusion of the source channel with 20 kDa dextran-FITC. e Representative phase/fluorescence overlay images of angiogenic sprouts in an iBMEC microvessel at different time points following perfusion of the source channel with 10 ng mL⁻¹ bFGF and 20 kDa dextran-FITC (f) Number of sprouts growing toward the source or sink channel per field of view in a 10 × image. n = 9 images across 3 microvessels. g Length of angiogenic sprouts. The length is defined as the vertical distance from the front of tip cell to the wall of microvessel. h Acute angle of the longest tip cell to the microvessel. n = 43 (day 1 source), 46 (day 2 source), 26 (day 1 sink), and 33 (day 2 sink) across 3 microvessels for g and h

Fig. 3

Angiogenesis in a post-capillary venule (PCV) model under a growth factor concentration gradient in a three-channel device. a Post-capillary venule in a triple-channel microfluidic chip. iPC, induced pericyte-like cells. b Schematic illustration of the classic model of brain angiogenesis involving BMEC tip cells and a model involving pericyte tip cells. PC, pericytes; EC, endothelial cells (c) Confocal images of a PCV perfused with pericyte medium 1 day after seeding showing sparse abluminal pericytes surrounding the endothelium. iBMEC, induced brain microvascular endothelial-like cell; iPC, induced pericyte-like cells. d Representative phase /fluorescence overlay images of angiogenic sprouts in a PCV 4 days after perfusion of the source channel with 10 ng mL⁻¹ bFGF. e–g Characterization of angiogenic sprouts in a PCV on days 1 and 2 after perfusion of the source channel with 10 ng mL⁻¹ bFGF. (e) Number of angiogenic sprouts led by BMECs or PCs. n = 9 images across 3 microvessels. (f) Length of sprouts. g Angle of angiogenic sprouts at the tip cell. n = 43 (PC tip cells on day 1), 46 (PC tip cells on day 2), 24 (EC tip cells on day 1), and 22 (EC tip cells on day 2) for f and g

Fig. 4

Morphology of angiogenic sprouts in a iBMEC/iPC PCV model following perfusion of the source channel with 10 ng mL⁻¹ bFGF. a Confocal images of a PCV model on day 1. Angiogenic sprouts were led by endothelial (red) or pericyte (green) tip cells. b Confocal images of a PCV microvessel on day 2 showing the growth of angiogenic sprouts. c Lumen structure of an angiogenic sprout on day 2. (Right panel) Cross-section image showing the lumen structure of the sprout

Fig. 5

Barrier function of PCV models during angiogenesis. a Phase/fluorescence overlay images of a control PCV during perfusion of the source channel with basal medium on day 2. 500 kDa dextran-FITC and 70 kDa dextran-Texas Red were perfused into the microvessels for 10 min. b, c Fluorescence intensity as a function of distance from the microvessel wall (i.e., perpendicular to the microvessel axis) in control PCVs for b 500 kDa dextran-FITC and c 70 kDa dextran-Tex red. d Phase/fluorescence overlay images of a PCV during perfusion of the source channel with 10 ng mL⁻¹ bFGF on day 2. e, f Fluorescence intensity as a function of distance from the microvessel wall in PCVs for e 500 kDa dextran-FITC and f 70 kDa dextran-Tex red. Two dashed lines indicate the region of the PCV