HIF-1alpha induced-VEGF overexpression in bone marrow stem cells protects cardiomyocytes against ischemia

Abstract

Hypoxia inducible factor-1alpha (HIF-1alpha) is a proangiogenic transcription factor stabilized and activated under hypoxia. It regulates the expression of numerous target genes, including vascular endothelial growth factor (VEGF) and other cytoprotective proteins. In this study, we hypothesized that bone marrow stem cells (BMSCs) secrete growth factors which protect cardiomyocytes via HIF-1alpha pathway. BMSCs were obtained from transgenic mice overexpressing green fluorescent protein (GFP). The study was carried out in vitro using co-culture of BMSCs with cardiomyocytes. LDH release, MTT uptake, DNA fragmentation and annexin-V positive cells were used as cell injury markers. The level of HIF-1alpha protein as well as its activated form and VEGF were measured by ELISA. The expression of HIF-1alpha and VEGF in BMSCs was analyzed by quantitative PCR and cellular localization was determined by immunohistochemistry. LDH release was increased and MTT uptake was decreased after exposure of cardiomyocytes to hypoxia for 30 h, which were prevented by co-culturing cardiomyocytes with BMSCs. Cardiomyocyte apoptosis induced by hypoxia and H(2)O(2) was also reduced by co-culture with BMSCs. VEGF release from BMSCs was significantly increased in parallel with high level of HIF-1alpha in BMSCs following anoxia or hypoxia in a time-dependent manner. Although no significant up-regulation could be seen in HIF-1alpha mRNA, HIF-1alpha protein and its activated form were markedly increased and translocated to the nucleus or peri-nuclear area. The increase and translocation of HIF-1alpha in BMSCs were completely blocked by 2-methoxyestradiol (2-ME2; 5 mumol), a HIF-1alpha inhibitor. Moreover, the protection of cardiomyocytes by BMSC and VEGF secretion was abolished by neutralizing HIF-1alpha antibody in a concentration dependent manner (200-3200 ng/ml). Bone marrow stem cells protect cardiomyocytes by up-regulation of VEGF via activating HIF-1alpha.

Figure 1

BMSCs were obtained from transgenic mice expressing GFP and cardiomyocytes from neonatal rat ventricles. A, Primary cultured GFP-positive BMSCs. B, Same as “A”, but the nuclei of BMSCs were stained with DAPI. C. Cultured myocytes were positive for α-actinin (green). Connexin 43 (red) was observed between myocytes. The nuclei were stained with DAPI.

Figure 2

LDH release and MTT uptake by cultured cardiomyocytes under different treatments. ^# p < 0.05 vs normal control; * p < 0.01 vs myocytes alone; † p < 0.05 vs myocytes co-cultured with BMSC.

Figure 3

Effect of BMSCs on DNA fragmentation under hypoxic conditions. Panel A: myocytes alone, Panel B: Myocytes co-cultured with BMSCs (Lane 1: marker; Lane 2: normal; Lane 3: Hypoxia 24h; Lane 4: Hypoxia 48h; Lane 5: Hypoxia 72h; Lane 6: positive control). Panel C. Myocytes co-cultured with BMSC before and after anti-HIF-1α antibody treatment.

Figure 4

Apoptosis assay after cells were exposed to hypoxia for 30 hours and H₂O₂ (200μmol) for 2 hours. A. Hypoxic treatment. B~G: H₂O₂ treatment. Panel B–D: Annexin V positive cells are shown in red color; Panel E–G: flowcytometry assay. Panels B and E: normal cultured myocytes; Panels C and F: H₂O₂ treated myocytes; Panels D and G: H₂O₂ treated cardiomyocytes co-cultured with BMSCs; Panel H. Percentage of annexin V positive cells after H₂O₂ treatments. ^# p < 0.05 vs normal cultured myocytes; * P < 0.05 vs hypoxia- or H₂O₂-treated myocytes alone. ^† p < 0.05 vs myocytes co-cultured with BMSCs.

Figure 4

Apoptosis assay after cells were exposed to hypoxia for 30 hours and H₂O₂ (200μmol) for 2 hours. A. Hypoxic treatment. B~G: H₂O₂ treatment. Panel B–D: Annexin V positive cells are shown in red color; Panel E–G: flowcytometry assay. Panels B and E: normal cultured myocytes; Panels C and F: H₂O₂ treated myocytes; Panels D and G: H₂O₂ treated cardiomyocytes co-cultured with BMSCs; Panel H. Percentage of annexin V positive cells after H₂O₂ treatments. ^# p < 0.05 vs normal cultured myocytes; * P < 0.05 vs hypoxia- or H₂O₂-treated myocytes alone. ^† p < 0.05 vs myocytes co-cultured with BMSCs.

Figure 5

Activity and distribution of HIF-1α in BMSCs. Panel A: Total HIF-1α (n = 10) and Panel B: Activity of HIF-1α (n = 8) in the BMSCs under hypoxia for 30 hours. It was inhibited by 2-ME2. ^# p < 0.05 vs normal culture and * p < 0.05 vs hypoxia alone. Panel C: Quantitative PCR for HIF-1α mRNA (n = 4). There was no significantly difference among various treatments. Panel D. HIF-1α was scattered in cytosol of normal BMSCs; Panel E: HIF-1 was highly concentrated in peri-nucleus and in nuclei (arrows) after BMSCs were exposed to hypoxia for 30 hours. Panel F. Pretreatment of cells with 2-ME2 (5 μmol) for 16 hours significantly abolished the translocation of HIF-1α. Red: HIF-1α; Blue: DAPI (nuclei).

Figure 6

Secretion of VEGF by BMSCs. Panel A. VEGF release from BMSCs. *, p < 0.05 vs normal culture, respectively. Panel B: Release of VEGF from BMSCs by exposure to 30 hours hypoxia was partially inhibited by 2-ME2 (5μmol). ^# p < 0.05 vs normal culture 10 hours. Panel C. The expression of VEGF in BMSCs. Bars represent the fold increase in VEGF concentrations measured in hypoxic cultured BMSCs relative to normoxic conditions. The expression of VEGF was significantly reduced by specific HIF-1α neutralizing antibodies (3200 ng/ml) and HIF-1α inhibitor 2-ME2 (5 μmol). ^#, p < 0.05 vs normal culture; *, p < 0.05 vs cells exposed to hypoxia.

Figure 7

Effect of VEGF on cardiomyocyte apoptosis. Panel A: VEGF reduced the DNA fragmentation induced by hypoxia. Panel B: VEGF also prevented apoptosis of myocytes exposed to H₂O₂ (200 μmol/L) for 2 hours. ^#, p < 0.05 vs normal culture; *, p < 0.05 vs myocytes exposed to H₂O₂.