Patterned human microvascular grafts enable rapid vascularization and increase perfusion in infarcted rat hearts

Abstract

Vascularization and efficient perfusion are long-standing challenges in cardiac tissue engineering. Here we report engineered perfusable microvascular constructs, wherein human embryonic stem cell-derived endothelial cells (hESC-ECs) are seeded both into patterned microchannels and the surrounding collagen matrix. In vitro, the hESC-ECs lining the luminal walls readily sprout and anastomose with de novo-formed endothelial tubes in the matrix under flow. When implanted on infarcted rat hearts, the perfusable microvessel grafts integrate with coronary vasculature to a greater degree than non-perfusable self-assembled constructs at 5 days post-implantation. Optical microangiography imaging reveal that perfusable grafts have 6-fold greater vascular density, 2.5-fold higher vascular velocities and >20-fold higher volumetric perfusion rates. Implantation of perfusable grafts containing additional hESC-derived cardiomyocytes show higher cardiomyocyte and vascular density. Thus, pre-patterned vascular networks enhance vascular remodeling and accelerate coronary perfusion, potentially supporting cardiac tissues after implantation. These findings should facilitate the next generation of cardiac tissue engineering design.

Fig. 1

In vitro anastomosis of human embryonic stem cell-derived endothelial cells (hESC-ECs) in engineered microvessels (µVs). a Schematic of in vitro culture device for μV+SA constructs: mTm-hESC-EC µVs formed via perfusion and attachment with bulk-seeded GFP-hESC-ECs in the surrounding collagen gel. b Maximum intensity projection of stitched large image confocal z-stack of μV+SA construct cultured for 4 days and stained for DsRed (red) and GFP (green) to detect mTm- and GFP-expressing hESC-ECs, respectively. Scale bar, 500 μm. c Outlined region (white box) in b stained for DsRed (red, top left), GFP (green, top right), and VE-cadherin (white, bottom left). Merged image, bottom right. Scale bar, 200 μm. d High magnification images of GFP-hESC-ECs (green) integrated with mTm-hESC-EC (red) patterned vessel in μV+SA constructs. Scale bar, 50 μm. e Quantitation of sprouts from patterned µVs by sprout density (no. of sprouts per vessel surface area), sprout length, and sprout diameter in μV only (blue circles) and μV+SA (green circles) constructs after 4 days and 7 days of culture. N = 6, 7, 4, and 3 biologically independent samples for D4 μV only, D4 μV+SA, D7 μV only, and D7 μV+SA, respectively. p = 0.011 for length and p = 0.007 for diameter for D4 μV only and D7 μV only, p > 0.05 for all others (two-tailed t test). f 3D view of GFP+ de novo lumen integrated with mTm+ microvascular sprout (white arrowheads) stained for CD31 (red) and GFP (green). Scale bar, 100 μm. Representative images for b–d, f from seven biologically independent samples of D4 μV+SA, with similar results. Hoechst-stained nuclei, blue. Error bars, mean ± SEM. *p < 0.05 determined using two-tailed t test. D4 after 4 days of culture, D7 after 7 days of culture

Fig. 2

Fluorescent bead perfusion of engineered microvessels (µVs). a µV+SA construct with perfused anastomotic connections. High magnification views of outlined regions i and ii with corresponding in situ staining for mTm-hESC-ECs (DsRed+, red) and GFP-hESC-ECs (GFP+, green). Scale bars, 100 μm, 40 μm. b High magnification view of fluorescent beads at two time points, t₀ and t₁, 0.27 s apart. Red arrows track the movement of fluorescent beads to calculate velocities. c, d Average bead velocity in the patterned vessel (c) and in sprouts with diameter <50 µm (d) in µV only (blue circles) and µV+SA (green circles) constructs after 4 days and 7 days of culture N = 3, 3, 4, and 3 biological independent samples for D4 µV only, D4 µV+SA, D7 µV only, and D7 µV+SA, respectively. p = 0.004 for D4 µV+SA and D7 µV+SA in sprouts, p = 0.012 for D7 µV and D7 µV+SA in sprouts, p > 0.05 for all others (two-tailed t test). e Scatter plot of bead velocity in relation to sprout diameter N = 15, 20, 37, 53 (number of sprouts analyzed) for D4 µV only, D4 µV+SA, D7 µV only, and D7 µV+SA in 2, 1, 5, and 3 biologically independent samples. Representative images for a, b from three biologically independent samples of D4 µV+SA, with similar results. Hoechst-stained nuclei, blue. Error bars, mean ± SEM. *p < 0.05 determined using two-tailed t test. ns non-significant (p > 0.05). D4 after 4 days of culture, D7 after 7 days of culture

Fig. 3

Citrated whole-blood perfusion in engineered microvessels (µVs). a Brightfield stitched large image of red blood cell-filled pattern and sprouts with magnified view (inset, white dotted boundary) for human embryonic stem cell-derived endothelial cell (hESC-EC)-seeded µV only constructs after 4 days of culture. Scale bar, 200 μm. b Maximum intensity projection of confocal z-stack of constructs with adhered platelets after 30-min perfusion and subsequent phosphate-buffered saline (PBS) washes for untreated hESC-EC-seeded constructs (top) and phorbol myristate acetate (PMA)-treated hESC-EC-seeded constructs (bottom) stained for CD31 (red) and CD41a (green). Scale bar, 200 μm. c Quantification of platelet adhesion on the vessel wall for constructs seeded with human umbilical vein endothelial cells (HUVECs) in control conditions (C–HUVEC), hESC-ECs in control conditions (C–hESC-EC), and hESC-ECs treated with PMA (PMA–hESC-EC) or interleukin (IL)-1β (IL-1β–hESC-EC). Data are expressed as a percentage of the vessel wall surface area. N = 2, 2, 2, and 3 biologically independent samples for C–HUVEC, C–hESC-EC, PMA, and IL-1β, respectively. Representative images for a, b from two biologically independent samples of C–hESC-ECs and two biologically independent samples of PMA–hESC-EC, with similar results. Error bars, mean ± SEM

Fig. 4

Global RNA sequencing reveals differential gene expression profiles among self-assembled (SA), µV only, and µV+SA constructs after 3 days of culture in vitro. a 2D principal component analysis (PCA) of RNA sequencing data for cultured constructs showing clustered groups for each condition. b, c Top genes contributing to PC1: b Size of the circle is proportional to the contribution. c Heat map of the top 75 genes (log₂ CPM) contributing to PC1 in categorized functions. Colormap normalized to minimum and maximum expression. Red, high expression. Blue, low expression. d Gene ontology terminology analysis (GO) showing different gene clusters for μV+SA vs. SA (left) and μV vs. SA (right). Each sample is two constructs pooled. CPM counts per million

Fig. 5

Implanted vascular constructs in infarcted rat heart model with optical microangiography (OMAG) assessment of real-time perfusion 5 days post-implantation. a–c Doppler-based images of vascular flow in Langendorff perfused engrafted hearts for a healthy region, b control graft (self-assembled (SA)), and c patterned, perfusable graft (μV+SA). Top panels, high-resolution, non-quantitative OMAG images. Bottom panels, OMAG-V images with linear correlation between intensity and velocity. Scale bar, 500 μm. d Three-dimensional views of OMAG flow data and overlay of OMAG structure and flow data for SA graft and μV+SA graft. Scale bar, 500 μm. e Quantitation of vessel density of perfused vessels in healthy (red circles) and graft (purple circles) regions for SA and μV+SA. N = 9 and 8 biologically independent animals for SA and μV+SA, respectively. p = 0.0009 for SA healthy and graft, p = 0.0016 for μV+SA healthy and graft, p = 0.0012 for SA and μV+SA graft, p > 0.05 for all others (two-tailed t test). f Velocimetry measurements of blood flow velocity in small arterioles (diameter between 20 and 40 μm) of healthy and graft regions. N = 5 and 6 biologically independent animals for SA and μV+SA, respectively. p = 0.016 for SA healthy and graft, p = 0.023 for SA and μV+SA graft, p > 0.05 for all others (two-tailed t test). g Volumetric perfusion rate in SA and μV+SA grafts from velocimetry data. N = 5 and 6 biologically independent animals for SA and μV+SA, respectively. p = 0.038 for SA and μV+SA graft (two-tailed t test). Representative images for a–d from 5 biologically independent animals containing SA grafts, 6 biologically independent animals containing μV+SA grafts, and 11 corresponding healthy regions from all animals containing grafts, with similar results. Error bars, mean ± SEM. *p < 0.05, **p < 0.01 determined using two-tailed t test

Fig. 6

Detection of human endothelial cells and perfused vessels in grafts 5 days post-implantation. a, b Immunofluorescent-stained paraffin sections of self-assembled (SA) and µV+SA grafts: a human embryonic stem cell-derived endothelial cells (hESC-ECs) (DsRed+, red) and perfused rat and human endothelium (Lectin+, green). White arrows in merged images denote double positive cells. Scale bar, 100 μm. b hESC-ECs (DsRed+, red) and perfused human endothelium (UEA I+, white). Yellow arrows in merged images denote double positive cells. Scale bar, 100 μm. c Quantification of vessel density (number of vessels per mm²) and average vessel size for hESC-EC vessels (DsRed+, red circles) and all perfused vessels (Lectin+, blue circles) in the grafts. N = 6 biologically independent animals for both SA and µV+SA grafts. p = 0.045 for lectin+ vessels for SA and µV+SA, p = 0.018 for DsRed+ and lectin+ for µV+SA, p > 0.05 for all others (two-tailed t test). d Quantification of vessel density and average vessel size of perfused human vessels (UEA I+, grey circles) in the grafts. N = 6 biologically independent animals for both SA and µV+SA grafts, with one animal with µV+SA graft excluded from vessel size calculation due to lack of UEA I+ vessels. p = 0.023 for vessel size for SA and µV+SA, p > 0.05 for vessel density (two-tailed t test). Representative images for a, b from six biologically independent animals containing SA grafts and six biologically independent animals containing μV+SA grafts, with similar results. Hoechst-stained nuclei, blue. Data were collected from at least three confocal images of randomly selected regions per sample and analyzed by a custom lumen identifying code for vessel density and size. Error bars, mean ± SEM. *p < 0.05 determined using two-tailed t test. UEA Ulex europaeus Agglutinin I

Fig. 7

Perfusable cardiac constructs in infarcted rat heart model 5 days post-implantation a, b Immunofluorescent-stained paraffin sections of self-assembled (SA; left) and µV+SA (right) cardiac grafts: a Whole-graft section containing human embryonic stem cell-derived cardiomyocytes (hESC-CMs) (β-MHC, red). Gray dotted line outlines graft tissue. Scale bar, 1 mm. b High magnification images of boxed region in a (yellow box) with hESC-CMs (β-MHC, red, bottom right) and perfused vessels (Lectin+, green, bottom left). Merged, top. Scale bar, 100 μm. c Quantification of cardiomyocyte density (number of β-MHC+ cells per mm²) and β-MHC+ graft size (% LV area). N = 7 and 6 biologically independent animals for SA and µV+SA, with one animal with µV+SA graft excluded from graft size calculation due to partial graft removal during tissue processing. p = 0.015 for density, p > 0.05 for graft size (two-tailed t test). d Quantification of vessel density (number of vessels per mm²) and average vessel size for all perfused vessels (Lectin+) in the cardiac grafts. N = 7 and 6 biologically independent animals for SA and µV+SA, respectively. p = 0.046 for vessel density, p > 0.05 for vessel size (two-tailed t test). e Terminal deoxinucleotidyl transferase-mediated dUTP-fluorescein nick end labeling (TUNEL) assay for apoptotic cells (TUNEL, green) on paraffin sections of SA (left) and µV+SA (right) cardiac grafts. Scale bar, 100 μm. f Quantification of apoptotic cells (TUNEL+) as a percentage of all cells (human and host) in the graft. N = 7 and 6 biologically independent animals for SA and µV+SA, respectively. p > 0.05 (two-tailed t test). Representative images for a, b, e from six biologically independent animals containing SA grafts and five biologically independent animals containing μV+SA grafts, with similar results. Hoechst-stained nuclei, blue. Data were collected from at least two confocal images of randomly selected regions per sample. Images analyzed by a custom lumen identifying code for vessel density and size. Error bars, mean ± SEM. *p < 0.05 determined using two-tailed t test. β-MHC beta-myosin heavy chain