With regard to the functional analysis of DSCs

With regard to the functional analysis of DSCs and HGF-1, several in vitro tests were performed with endothelial Nanaomycin A (HMEC-1) to assess their influence on the different steps of the angiogenic process. Despite the expression of multiple mitogens such as VEGF, EDN1, DPPIV and ANG, DSCs were not able to increase the proliferation of HMEC-1. A potential explanation for this discrepancy could be the concentration of the secreted proliferation-stimulating factors, which was presumably too low to cause a sustainable effect. Another explanation may lie in the expression of proliferation-inhibiting factors, such as PTX3 and THBS1; the actions of these proteins could be predominant in comparison to the angiogenesis-promoting factors and tip the angiogenic balance towards an inhibitory state. In contrast, Iohara et al., demonstrated a significant increase of human umbilical vein endothelial cell (HUVEC) proliferation following incubation with conditioned medium of porcine pulp-derived CD31− CD146− stem cells (Iohara et al., 2008). However, besides considering the potential species-related differences between swine and human, it also has to be kept in mind that the aforementioned subset of DPSCs could potentially display more pronounced angiogenic properties than DPSCs in general (Nakashima et al., 2009). In comparison, studies regarding the proliferation-stimulating capacity of BMSCs yielded similar conflicting results. While Potapova et al. and others mentioned a significant increase in HUVEC proliferation caused by BMSCs, Gruber et al., demonstrated no proliferation-promoting effect of BMSC conditioned medium (Chen et al., 2008; Gruber et al., 2005; Kinnaird et al., 2004; Potapova et al., 2007). However, the constitution of the applied conditioned medium appears to play an important role, as the addition of fetal bovine serum can increase the expression of proliferation-stimulating factors such as VEGF and in that way bias the outcome of the experiment (Potapova et al., 2007). For that reason, the conditioned medium in this study only contained 0.1% FBS in order to avoid artificial upregulation of VEGF or other mitogens. Another important aspect within the angiogenic cascade is endothelial migration. Since the expression analysis demonstrated that DSCs and HGF-1 express multiple factors which are known to affect migration, such as ANGPT1, EDN1, IGFBP3, uPA and VEGF, a transwell migration assay was carried out to assess their chemotactic potential. Following 24h of incubation, DPSCs, SCAPs and HGF-1 significantly increased endothelial transmigration, while FSCs had no substantial impact. Given the high secretion of IGFBP3 by FSCs, the lack of a pronounced migration-stimulating effect was rather unexpected. However, the aforementioned dual role of IGFBP3 taken together with the lower secretion of VEGF and ANGPT1 probably established suboptimal conditions for endothelial migration. HGF-1 on the other hand, display a secretion profile similar to FSCs in terms of VEGF, IGFBP3 and ANGPT1 and do significantly enhance endothelial migration. This discrepancy can be explained by the potential contribution of other (yet to be identified) angiogenic factors which influence endothelial migration. With regard to the chemotactic properties of BMSCs, similar observations were made by Potapova et al. and others, who mentioned a significant increase in HUVEC transmigration caused by stromal cell-conditioned medium (Gruber et al., 2005; Potapova et al., 2007). In terms of tubulogenesis, functional assays showed a pronounced effect of DPSCs on endothelial tube formation, an outcome which also differed significantly from SCAPs and HGF-1. Earlier studies of Tran-Hung et al. and others, reported a similar increase and stabilization of endothelial tubular structures following direct co-culture of HUVECs and DPSCs, indicating a more pericyte-like behavior of DPSCs (Dissanayaka et al., 2012; Janebodin et al., 2013; Tran-Hung et al., 2006). Human and murine BMSCs on the other hand, are also capable of promoting endothelial tube formation, as was shown by a number of studies (Estrada et al., 2009; Lin et al., 2012; Sorrell et al., 2009; Wu et al., 2007). Since ANGPT1 and VEGF play an important role in the induction of tubulogenesis, the aforementioned increase can probably be explained by the angiogenic secretion profile of the different cell populations as DPSCs displayed a notably higher VEGF secretion compared to SCAPs, FSCs and HGF-1.