Pamela Graney1 Shahar Ben-Shaul2 Shulamit Levenberg2 Kara Spiller1

1, Drexel University, Philadelphia, Pennsylvania, United States
2, Technion-Israel Institute of Technology, Haifa, , Israel

A major challenge in engineering biomaterials for regenerative medicine is achieving sufficient vascularization to support tissue survival and integration in vivo. Functional blood vessel networks must not only form within the tissue (angiogenesis), but also connect with the existing host vasculature (anastomosis) to achieve graft perfusion upon implantation. Macrophages, the primary cell of the innate immune system, strongly affect the outcome of implanted biomaterials and have been implicated in both angiogenesis and anastomosis. However, the role of macrophage phenotype in these processes is unclear. Previously, we have shown that macrophages activated with pro-inflammatory stimuli interferon-gamma (IFNg) and lipopolysaccharide (LPS) (M1) secrete factors involved in the early stages of angiogenesis, whereas macrophages activated with interleukin-4 (IL4) and IL13 (M2a) secrete factors involved in the stabilization of newly sprouted vessels1. The purpose of this study was to delineate the contribution of macrophage phenotype to biomaterial vascularization, using in vitro-polarized macrophages in a previously developed 3D model of in vitro blood vessel network formation2.

Human adipose microvascular endothelial cells expressing tdTomato together with human adipose-derived mesenchymal stem cells were pre-seeded on porous Gelfoam® scaffolds (Pfizer, New York, NY) to generate self-assembled vascular networks that can be used to examine vascularization dynamics in vitro. PMA-activated, GFP-expressing THP1-derived M0, M1, or M2a macrophages were added to the pre-vascularized constructs on days 3 or 6 of vessel growth, and changes in network development were monitored over 14 days using confocal microscopy. Images were analyzed in 2D and 3D using Angiotool3 and Matlab software, respectively, in terms of vessel density, length, extent of branching and number of endpoints. Statistical analysis was completed in GraphPad Prism 7.0 using analysis of variance with Tukey’s post-hoc analysis (n ≥ 3). Image analysis revealed a significant (p < 0.05) increase in vessel sprouting by M1 macrophages, while both M1 and M2a phenotypes increased vessel connections 1 day post-seeding relative to control constructs without macrophages. However, phenotype-specific changes in network morphology were indistinguishable after 72 hours, despite qualitative differences in M0, M1, and M2a morphology and localization. Interestingly, all macrophage-seeded constructs exhibited significant vessel regression independent of phenotype after 4 days in vitro; this regression was not observed in control constructs. Ongoing work aims to characterize the temporal changes in gene expression of macrophages and endothelial cells during angiogenesis. Ultimately, this work will aid in designing immunomodulatory biomaterials that promote vascularization and integration.

1. Spiller KL et al. Biomaterials 2014; 35.
2. Freiman A et al. Stem Cell Res Ther 2016; 7.
3. Zudaire E et al. PLoS One 2011; 6.