Investigating the role of hypoxia-inducible factor-1 activation in the vascularization of modular tissue engineered constructs

摘要

Vascularization is crucial to the fabrication of engineered tissues of therapeutically relevant size and cell loads. While prolonged hypoxia is detrimental to cell survival in these constructs, it is also a necessary driver of angiogenesis, exerting many of its effects via the hypoxia inducible factor (HIF) -1 pathway. Modular tissue engineered constructs are formed by random assembly of micrometer-scale cylindrical collagen pieces ("modules') embedded with adipose-derived mesenchymal stromal cells (adMSC) and enveloped with endothelial cells (EC); injected modules remodel in vivo to form a vascularized organoid. HIF-1 is hypothesized to govern: 1) the survival of graft-derived cells immediately following implantation, and 2) subsequent recruitment of angiogenic myeloid cells to the hypoxic site. When cell densities, module diameter and implant volume were varied, immunohistochemical analyses showed that graft-derived vascularization was most robust (p＜0.01) in implants with pronounced nuclear expression of HIF-1α at day 7, as with "large" volume implants (0.10 mL, 1.5×10~7 adMSC/mL, 3.9×10~4 EC/mL) and "small" volume implants with the highest cell density (0.01 mL, 1.5×10~8 adMSC/mL, 9.8×10~4 EC/mL). Moreover, many HIF-1α~+ cells were EC lining new vessels, suggesting an important role of HIF-1 activation in vessel assembly. A pharmacological inhibition study was then undertaken to elucidate a causal role of HIF-1 on module-induced vascularization: SCID-bg mice were provided daiiy injections of 2 mg/kg digoxin or saline starting 2 days prior to implantation of modules (in vitro preconditioned in 100 nM digoxin or saline for 6 hours), and continuing on to 7 days post-implantation. Digoxin-treated implants showed reduced graft-derived and total vessel formation compared with saline-treated controls at day 7 (19.4±1.9 vs. 44.0±9.9 UEA1~+ graft-derived vessels/mm~2; 19.4±3.0 vs. 46.0±10.0 CD31~+ total vessels/mm~2; n=3). Preliminary flow cytometry results also showed that effects of HIF-1 inhibition were correlated with reduced mobilization of CXCR4* myeloid cells (CD45~+CXCR4~+) into peripheral blood of animals collected 2 days after the start of the drug regimen (1.2±0.3 vs. 2.0±0.3 ×10~4 cells/mL blood; n=6). Together, these results point to HIF-1 as a driver of module-induced vascularization, potentially exerting its effects by modulating mobilization and recruitment of angiogenic CXCR4~+ myeloid cells to the implant. Currently, HIF-1α is being stably knocked down in graft-derived adMSC and/or EC using lentiviral delivery of short hairpin RNA. We expect that the rate and density of in vivo graft- and host-derived vascularization will be negatively affected. Knowledge gained from these studies may be harnessed to tune vessel formation in engineered tissues, for better application of such platforms in tissue therapy.