Glucose Promotes EMMPRIN/CD147 and the Secretion of Pro-Angiogenic Factors in a Co-Culture System of Endothelial Cells and Monocytes

Abstract

Vascular complications in Type 2 diabetes mellitus (T2DM) patients increase morbidity and mortality. In T2DM, angiogenesis is impaired and can be enhanced or reduced in different tissues ("angiogenic paradox"). The present study aimed to delineate differences between macrovascular and microvascular endothelial cells that might explain this paradox. In a monoculture system of human macrovascular (EaHy926) or microvascular (HMEC-1) endothelial cell lines and a monocytic cell line (U937), high glucose concentrations (25 mmole/L) increased the secretion of the pro-angiogenic factors CD147/EMMPRIN, VEGF, and MMP-9 from both endothelial cells, but not from monocytes. Co-cultures of EaHy926/HMEC-1 with U937 enhanced EMMPRIN and MMP-9 secretion, even in low glucose concentrations (5.5 mmole/L), while in high glucose HMEC-1 co-cultures enhanced all three factors. EMMPRIN mediated these effects, as the addition of anti-EMMPRIN antibody decreased VEGF and MMP-9 secretion, and inhibited the angiogenic potential assessed through the wound assay. Thus, the minor differences between the macrovascular and microvascular endothelial cells cannot explain the angiogenic paradox. Metformin, a widely used drug for the treatment of T2DM, inhibited EMMPRIN, VEGF, and MMP-9 secretion in high glucose concentration, and the AMPK inhibitor dorsomorphin enhanced it. Thus, AMPK regulates EMMPRIN, a key factor in diabetic angiogenesis, suggesting that targeting EMMPRIN may help in the treatment of diabetic vascular complications.

Keywords: AMPK; CD147/EMMPRIN; diabetes mellitus; dorsomorphin/compound C; endothelial cells; glucose; metformin.

Figure 1

Glucose affects only microvascular endothelial cell viability at high concentrations. Cells of the three cell lines (20,000 cells/well) were seeded separately in low-glucose DMEM, with increasing concentrations of glucose, and after 24 h of incubation, cell viability was determined using the XTT assay. Results are presented as fold from control of each cell line. The one-way ANOVA followed by Dunnett’s multiple comparisons test was used (n = 7 for each cell line at each time point). *, p < 0.05, **, p < 0.01 relative to no treatment.

Figure 2

High glucose concentrations enhance the expression of EMMPRIN, MMP-9, and VEGF in endothelial cells. The human macrovascular endothelial cells EaHy926, microvascular endothelial cells HMEC-1, and the monocytic-like cell line U937 were incubated as single cultures (100,000 cells each/well/500 μL) in serum-starvation medium with low-glucose DMEM, and glucose was added to achieve the indicated concentrations. Cells were incubated for 24 h, and then superannuates were collected and the concentrations of (A–C) EMMPRIN, (D–F) MMP-9, and (G–I) VEGF, were determined using ELISA. The means ± SEM are presented (n = 7). Data were analyzed using the one-way ANOVA followed by Bonferroni’s post hoc test.

Figure 3

High glucose concentrations and co-culturing of endothelial cells with monocytes enhance the secretion of EMMPRIN, MMP-9, and VEGF. The human endothelial cells EaHy926 and HMEC-1, derived from macrovasculature or microvasculature, respectively, were incubated for 24 h alone (100,000 cells each/500 μL in a 24-well plate) or in a co-culture with the human monocytic-like cell line U937 (100,000 cells) in low-glucose DMEM serum-starvation medium, with increasing glucose added to achieve the indicated concentrations. At the end of the incubation, superannuates were collected and the concentrations of (A,D) EMMPRIN, (B,E) MMP-9, and (C,F) VEGF, were determined using ELISA. The means ± SEM are presented (n = 6). Data were analyzed using the one-way ANOVA followed by Bonferroni’s post hoc test.

Figure 4

The anti-EMMPRIN antibody (h161-pAb) decreases VEGF and MMP-9 and cell migration. (A–D) The human endothelial cells EaHy926 or HMEC-1 (100,000 cells/well) were incubated in co-culture with the U937 cells at a 1:1 ratio, with glucose added to achieve the indicated concentrations, and with or without the addition of the anti-EMMPRIN antibody (h161-pAb, 2 ng/mL). Cells were incubated for 24 h, and the concentrations of (A,B) MMP-9 and (C,D) VEGF were determined using ELISA. The means ± SEM are presented (n = 4–7). Data were analyzed using the one-way ANOVA followed by Bonferroni’s post hoc test. (E–G) The human endothelial cell line EaHy926 (20,000 cells/well) was seeded and allowed to grow to confluency overnight. A scratch was made with a toothpick, and non-adherent cells were washed away. The cells were then incubated for 18 h with full medium and supernatants (diluted at a ratio of 2.5:1) derived from EaHy926 cells that were previously incubated alone or in co-culture with glucose and with or without the h161-pAb (2 ng/mL). Representative images of (E) assay at 5 mmole/L glucose, (F) assay at 25 mmole/L glucose, and (G) quantitative analysis of the assays (n = 8–12). The distance to which cells migrated was calculated from images taken at 0 h and 18 h. The means ± SEM are presented, and data were analyzed using the one-way ANOVA followed by Bonferroni’s post hoc test.

Figure 5

Metformin inhibits the secretion of the pro-angiogenic factors. (A–F) The human endothelial cells EaHy926 or HMEC-1 (100,000 cells/well/500 μL) were co-cultured with the U937 cells at a 1:1 ratio, with increasing glucose added to achieve the indicated concentrations, and with or without the addition of metformin (2 mmole/L). Cells were incubated for 24 h, and the concentrations of (A,D) EMMPRIN, (B,E) MMP-9, and (C,F) VEGF were determined using ELISA. The means ± SEM are presented (for EaHy926 cells, n = 10–15 in each group; for HMEC-1 cells, n = 5–6 in each group). Data were analyzed using the one-way ANOVA followed by Bonferroni’s post hoc test. (G–J) The human endothelial cells EaHy926 (100,000 cells/well/1000 μL) were co-cultured with the U937 cells at a 1:1 ratio in inserts (0.4 μm pore size) with 25 mmole/L glucose, and with or without the addition of metformin (2 mmole/L), the AMPK inhibitor (0.5 μM), or their combination. Cells were incubated for 24 h, and then supernatants were collected and cells were lysed. (G) The phosphorylation of AMPK was determined using ELISA (n = 8–12). The concentrations of (H) EMMPRIN, (I) VEGF, and (J) MMP-9 in the supernatants were determined using ELISA (n = 7–9 in each group). The means ± SEM are presented. Data were analyzed using the one-way ANOVA followed by Bonferroni’s post hoc test.

Figure 5

Metformin inhibits the secretion of the pro-angiogenic factors. (A–F) The human endothelial cells EaHy926 or HMEC-1 (100,000 cells/well/500 μL) were co-cultured with the U937 cells at a 1:1 ratio, with increasing glucose added to achieve the indicated concentrations, and with or without the addition of metformin (2 mmole/L). Cells were incubated for 24 h, and the concentrations of (A,D) EMMPRIN, (B,E) MMP-9, and (C,F) VEGF were determined using ELISA. The means ± SEM are presented (for EaHy926 cells, n = 10–15 in each group; for HMEC-1 cells, n = 5–6 in each group). Data were analyzed using the one-way ANOVA followed by Bonferroni’s post hoc test. (G–J) The human endothelial cells EaHy926 (100,000 cells/well/1000 μL) were co-cultured with the U937 cells at a 1:1 ratio in inserts (0.4 μm pore size) with 25 mmole/L glucose, and with or without the addition of metformin (2 mmole/L), the AMPK inhibitor (0.5 μM), or their combination. Cells were incubated for 24 h, and then supernatants were collected and cells were lysed. (G) The phosphorylation of AMPK was determined using ELISA (n = 8–12). The concentrations of (H) EMMPRIN, (I) VEGF, and (J) MMP-9 in the supernatants were determined using ELISA (n = 7–9 in each group). The means ± SEM are presented. Data were analyzed using the one-way ANOVA followed by Bonferroni’s post hoc test.