Matrigel tube formation assay. (A) Schematic diagram of the tube formation assay chamber. HUVECs were seeded in 24-well plates containing GeltrexTM matrix in the lower chamber. The upper chamber was prepared by plating a monolayer of BMC, WBM or MSC onto a very thick layer followed by incubation at 37°C for eight hours. Following incubation, the Boyden chamber containing the cells was removed and the plate with HUVEC cells was stained for 30 minutes with calcein-AM dissolved in DMSO. The number of tube-like structures that formed in the gel was measured by total length per field (X 200). (B) Phase contrast micrographs show HUVECs grown in matrix gel and tubular structure induced by different kinds of stem cells: (I) negative control (Medium 200); (II) positive control (Medium 200 containing 2%(v/v) FBS and bFGF (3 ng/μl); (III) BMC: whole bone marrow cultured cells; (IV) whole bone marrow and (V) mesenchymal stem cells. (C) Quantitative analysis of angiogenesis in vitro by measuring the occupied area due to the tubular structures formed by HUVECs. Photographs were taken eight hours after the HUVECs were seeded and inserts containing the stem cells were placed in the 24 well plates. In this experiment we used Medium 200 containing 2% (v/v) FBS and bFGF (3 ng/μl) as the positive control and Medium 200 from Invitrogen as the negative control. The induction of matrigel tube formation is higher in the presence of WBM compared with BMC and MSC. WBM versus BMC P <0.05, WBM versus MSC P <0.05. Data are expressed as mean +/− SEM of five measurements. bFGF, basic fibroblast growth factor; BMC, whole bone marrow cultured cells; DMSO, dimethyl sulfoxide; FBS, fetal bovine serum; HUVECs, human umbilical vein endothelial cells; MSC, mesenchymal stem cells; SEM, standard error of the mean; WBM, whole bone marrow.