Direct conversion of adult human fibroblasts into functional endothelial cells using defined factors.

We aimed to directly convert adult human dermal fibroblasts (aHDFs) into functional endothelial cells (ECs). Lentiviral vectors encoding endothelial transcription factors (TFs) were constructed. We examined whether five TFs (FOXO1, ER71, KLF2, TAL1, and LMO2) used for the generation of mouse induced ECs (iECs) could convert the aHDFs into human iECs. Twenty-eight days after transduction with lentiviral constructs, 32.1 ± 5.1% cells expressed vascular endothelial (VE)-cadherin. Factor screening revealed that only three factors (3F: ER71, KLF2, and TAL1) were necessary to induce VE-cadherin (+) cells (49.4 ± 3.5%). However, whole transcriptome sequencing showed that VE-cadherin (+) cells were not completely reprogrammed. Mature iECs double-positive for VE-cadherin/Pecam1 (DP cells) with a cobblestone appearance were obtained at a frequency of only 5.1 ± 0.6%. Using whole transcriptome analysis, the potential factors that could block the conversion were screened. Among candidates TWIST1-knockdown enhanced efficiency of conversion. Rosiglitazone, an inhibitor of epithelial-mesenchymal transition (EMT), also improved the conversion efficiency. Moreover, a 2nd second-stage conversion process, in which VE-cadherin (+) cells were incubated for additional two weeks, further enhanced the efficiency. The final protocol for 6 weeks yielded a conversion rate of 19.6 ± 3.0% iECs, defined by DP cells depicting the nature of mature ECs in various analyses. Further analyses revealed that the genetic and epigenetic profiles of iECs resembled those of functional ECs. Collectively, aHDFs can be converted into functional ECs through the transduction of ER71, KLF2, and TAL1, combined with two EMT inhibitors (siTWIST1 and rosiglitazone), followed by 2nd stage conversion.

Interaction between platelets and endothelial progenitor cells via LPA-Edg-2 axis is augmented by PPAR-δ activation.

BACKGROUND: Peroxisome proliferator-activated receptor (PPAR)-δ is a nuclear receptor regulating cell metabolism. The role of PPAR-δ in late endothelial progenitor cells (EPCs) has not been fully elucidated. We aim to understand the effects of PPAR-δ activation on late EPC and to reveal the underlying mechanism. METHODS AND RESULTS: Treatment with a highly selective PPAR-δ agonist (GW501516) induced proliferation of late EPCs and enhanced their vasculogenic potential. Search for the target molecule of PPAR-δ activation revealed endothelial differentiation gene (Edg)-2. Chromatin immunoprecipitation and promoter assays demonstrated that Edg-2 gene was specifically induced by PPAR-δ through direct transcriptional activation. Lysophosphatidic acid (LPA), an Edg ligand, mimicked the pro-vasculogenic effects of GW501516 in late EPCs whereas Edg antagonist (Ki16425) blocked these effects. Edg-2 is a membrane receptor for LPA which is a major growth factor from activated platelets. Thus, the interaction between platelets and late EPCs via the LPA-Edg-2 axis was assessed. Platelet supernatant boosted the pro-vasculogenic effects of GW501516, which was reversed by antagonist to PPAR-δ (GSK0660) or Edg (Ki16425). Both of in vivo Matrigel plug model and mouse skin punch-wound model demonstrated that the combination of platelets and PPAR-δ-activated late EPCs synergistically enhanced vascular regeneration. CONCLUSIONS: There exists a synergistic interaction between human platelets and late EPCs leading to vascular regeneration. This interaction consists of LPA from platelets and its receptor Edg-2 on the surface of EPCs and can be potentiated by PPAR-δ activation in EPCs. A PPAR-δ agonist is a good candidate to achieve vasculogenesis for ischemic vascular disease.

Direct conversion of adult skin fibroblasts to endothelial cells by defined factors.

BACKGROUND: Cell-based therapies to augment endothelial cells (ECs) hold great therapeutic promise. Here, we report a novel approach to generate functional ECs directly from adult fibroblasts. METHODS AND RESULTS: Eleven candidate genes that are key regulators of endothelial development were selected. Green fluorescent protein (GFP)-negative skin fibroblasts were prepared from Tie2-GFP mice and infected with lentiviruses allowing simultaneous overexpression of all 11 factors. Tie2-GFP(+) cells (0.9%), representing Tie2 gene activation, were detected by flow cytometry. Serial stepwise screening revealed 5 key factors (Foxo1, Er71, Klf2, Tal1, and Lmo2) that were required for efficient reprogramming of skin fibroblasts into Tie2-GFP(+) cells (4%). This reprogramming strategy did not involve pluripotency induction because neither Oct4 nor Nanog was expressed after 5 key factor transduction. Tie2-GFP(+) cells were isolated using fluorescence-activated cell sorting and designated as induced ECs (iECs). iECs exhibited endothelium-like cobblestone morphology and expressed EC molecular markers. iECs possessed endothelial functions such as Bandeiraea simplicifolia-1 lectin binding, acetylated low-density lipoprotein uptake, capillary formation on Matrigel, and nitric oxide production. The epigenetic profile of iECs was similar to that of authentic ECs because the promoters of VE-cadherin and Tie2 genes were demethylated. mRNA profiling showed clustering of iECs with authentic ECs and highly enriched endothelial genes in iECs. In a murine model of hind-limb ischemia, iEC implantation increased capillary density and enhanced limb perfusion, demonstrating the in vivo viability and functionality of iECs. CONCLUSIONS: We demonstrated the first direct conversion of adult fibroblasts to functional ECs. These results suggest a novel therapeutic modality for cell therapy in ischemic vascular disease.