Generation of functionally competent and durable engineered blood vessels from human induced pluripotent stem cells

Abstract

Efficient generation of competent vasculogenic cells is a critical challenge of human induced pluripotent stem (hiPS) cell-based regenerative medicine. Biologically relevant systems to assess functionality of the engineered vessels in vivo are equally important for such development. Here, we report a unique approach for the derivation of endothelial precursor cells from hiPS cells using a triple combination of selection markers--CD34, neuropilin 1, and human kinase insert domain-containing receptor--and an efficient 2D culture system for hiPS cell-derived endothelial precursor cell expansion. With these methods, we successfully generated endothelial cells (ECs) from hiPS cells obtained from healthy donors and formed stable functional blood vessels in vivo, lasting for 280 d in mice. In addition, we developed an approach to generate mesenchymal precursor cells (MPCs) from hiPS cells in parallel. Moreover, we successfully generated functional blood vessels in vivo using these ECs and MPCs derived from the same hiPS cell line. These data provide proof of the principle that autologous hiPS cell-derived vascular precursors can be used for in vivo applications, once safety and immunological issues of hiPS-based cellular therapy have been resolved. Additionally, the durability of hiPS-derived blood vessels in vivo demonstrates a potential translation of this approach in long-term vascularization for tissue engineering and treatment of vascular diseases. Of note, we have also successfully generated ECs and MPCs from type 1 diabetic patient-derived hiPS cell lines and use them to generate blood vessels in vivo, which is an important milestone toward clinical translation of this approach.

Keywords: diabetes; reprogramming; vascular endothelial cells.

Fig. 1.

Differentiation and characterization of hiPS cell-derived ECs. (A–C) 2D EC differentiation from nondiseased hiPS cells. (A) HS27-iPS cell colonies grown on feeder cells (mouse embryonic fibroblasts). (B) Day 10 in differentiation medium (50 ng/mL bone morphogenetic protein-4 added on day 1). (C) Putative EC morphology. Immunocytochemical staining of HS27-iPS-ECs for CD31 (D), vascular endothelial cadherin (E), and von Willebrand factor (F). (G) HS27-iPS-ECs show uptake for Ac-LDL. (H) HS27-iPS-ECs show tube formation on Matrigel. (Magnification: A–G, 10×.)

Fig. 2.

In vivo imaging of hiPS cell-derived engineered blood vessels. (A) Multiphoton laser-scanning microscopy image of HS27-iPS-ECs (green) and 10T1/2 cells (blue) coembedded in a fibronectin/collagen I tissue-engineered vessel construct and inoculated in SCID mice in a cranial window. These cells developed functional perfused blood vessels [red, 1,1-dioctadecyl-3,3,3,3-tetramethylindodicarbocyanine perchlorate (DID)-labeled RBCs] in vivo (day 14). (B–D) Functional assessment of iPS cell-derived engineered blood vessels. (B) Perfused vessels engineered from HS27-iPS-ECs (green) imaged after injection of DiD-RBCs (red) and Alexa 647-BSA (blue). (C) Map of RBC velocity quantified by recently established line and full-field RBC velocity-scanning techniques (45). Engineered vessels are well perfused. (D) Histogram of RBC velocity profile in iPS cell-derived vessels. HS27-iPS-ECs with 10T1/2 cells (blue bars), T1D-iPS-ECs with 10T1/2 cells (brown bars), and HS27-iPS-ECs with HS27-iPS-mesenchymal cells (green bars) are shown. (E) Map of vessel permeability to Alexa 647-labeled BSA quantified based on the extravasation of BSA from individual vessels over time. Engineered vessels have a segment with higher permeability compared with endogenous normal vessels. Multiphoton imaging was carried out on a custom-built multiphoton laser-scanning microscope using a confocal laser-scanning microscope body and a broadband femtosecond laser source. Imaging studies in A–C and E were performed at a magnification of 20×, using a 0.95-N.A. water immersion objective. Two-micron-thick optical sections were taken. The imaging field of view was 660 μm × 660 μm × 155 μm with a resolution of 1.3 μm × 1.3 μm × 2 μm.

Fig. 3.

Tissue imaging of engineered vessel construct. (A) Confocal whole-mount image of the tissue-engineered vessel construct (collagen 1 gel) extracted from the cranial window of an SCID mouse after 6 mo. Eleven images (all 20x) were manually aligned to form this mosaic image. EGFP⁺ HS27-iPS-ECs (green) are forming vessels with DsRed⁺ 10T1/2 supporting cells (red). Mouse CD31⁺ host ECs (blue) infiltrated in the gel and anastomosed with engineered vasculature (green). (B) High-power confocal image of the excised engineered vessel construct with EGFP⁺ HS27-iPS-ECs (green) and DsRed⁺ 10T1/2 supporting cells (red). In this image, desmin-positive perivascular cells surrounding engineered vessels (blue) are shown.

Fig. 4.

In vivo lifetime of engineered vessels. (A) In vivo Kaplan–Meier survival curves of ECs derived from HS27-iPS cell 2D differentiation with different sorting protocols. The fraction of mice containing viable ECs in each group at a given time point is presented. Maximum in vivo EC survival was seen with the CD34⁺KDR⁺NRP1⁺ protocol group (up to 280 d). (B) In vivo Kaplan–Meier survival curves of engineered vessels derived from HS27-iPS-ECs coimplanted with mesenchymal progenitor cells for different sources. HS27-iPS cell-derived multipotent mesenchymal progenitor cells showed longer duration of durable blood vessels (∼28 d) compared with bone marrow-derived human MSCs (hMSCs), which could support engineered vessels for only ∼11 d.