Canagliflozin Inhibits Human Endothelial Cell Inflammation through the Induction of Heme Oxygenase-1

Abstract

Sodium-glucose co-transporter 2 (SGLT2) inhibitors improve cardiovascular outcomes in patients with type 2 diabetes mellitus (T2DM). Studies have also shown that canagliflozin directly acts on endothelial cells (ECs). Since heme oxygenase-1 (HO-1) is an established modulator of EC function, we investigated if canagliflozin regulates the endothelial expression of HO-1, and if this enzyme influences the biological actions of canagliflozin in these cells. Treatment of human ECs with canagliflozin stimulated a concentration- and time-dependent increase in HO-1 that was associated with a significant increase in HO activity. Canagliflozin also evoked a concentration-dependent blockade of EC proliferation, DNA synthesis, and migration that was unaffected by inhibition of HO-1 activity and/or expression. Exposure of ECs to a diabetic environment increased the adhesion of monocytes to ECs, and this was attenuated by canagliflozin. Knockdown of HO-1 reduced the anti-inflammatory effect of canagliflozin which was restored by bilirubin but not carbon monoxide. In conclusion, this study identified canagliflozin as a novel inducer of HO-1 in human ECs. It also found that HO-1-derived bilirubin contributed to the anti-inflammatory action of canagliflozin, but not the anti-proliferative and antimigratory effects of the drug. The ability of canagliflozin to regulate HO-1 expression and EC function may contribute to the clinical profile of the drug.

Keywords: bilirubin; canagliflozin; endothelial cells; heme oxygenase-1; inflammation; migration; proliferation.

Figure 1

Canagliflozin (Cana) stimulates HO-1 expression in human ECs. (A) Cana (50 µM) stimulates a time-dependent increase in HO-1 mRNA expression. (B) Effect of Cana (0–50 µM) exposure for 8 h on HO-1 mRNA expression. (C) Effect of Cana (0–50 µM) exposure for 24 h on HO-1 protein expression. (D) Effect of Cana (50 µM) exposure for 24 h on HO activity. Results are mean ± SEM (n = 4–6). * Statistically significant effect of Cana.

Figure 2

Canagliflozin (Cana) inhibits human EC proliferation, DNA synthesis and migration in a HO-1-independent manner. (A) Effect of Cana (0–50 µM) treatment for 96 h on cell proliferation in the presence and absence of the HO inhibitor SnPP (10 µM). (B) Effect of Cana (0–50 µM) treatment for 24 h on DNA synthesis in the presence and absence of SnPP (10 µM). (C) HO-1 expression in ECs transfected with NT or HO-1 siRNA (100 nM) and then exposed to CANA (50 µM) for 24 h. (D) Effect of HO-1 silencing on Cana-mediated inhibition of EC DNA synthesis. Cells were transfected with NT and HO-1 siRNA (100 nM) and then exposed to Cana (50 µM) for 24 h. (E) Effect of Cana (50 µM) treatment for 24 h on the migration of ECs in the presence and absence of SnPP (10 µM). (F) Effect of HO-1 silencing on Cana-mediated inhibition of EC migration. Cells were transfected with NT and HO-1 siRNA (100 nM) and then exposed to Cana (50 µM) for 24 h. Results are mean ± SEM (n = 4–6). * Statistically significant effect of Cana. ⁺ Statistically significant effect of HO-1 siRNA.

Figure 3

HO-1 contributes to the anti-inflammatory action of canagliflozin (Cana). (A) Effect of Cana (0–50 µM) on monocyte adhesion following treatment of ECs with TNFα (10 ng/mL) and a high concentration of glucose (HG; 25 mM) for 24 h. (B) Effect of HO-1 inhibition on Cana-mediated blockade of monocyte adhesion. Cells were exposed to TNFα (10 µM)/HG (25 mM) in the absence and presence of Cana (50 µM) and/or SnPP (10 µM) for 24 h. (C) Effect of HO-1 silencing on Cana-mediated inhibition of monocyte adhesion. Cells were transfected with NT or HO-1 siRNA (100 nM) and then exposed to TNFα (10 µM)/HG (25 mM) in the absence and presence of Cana (50 µM), CORM2 (CM2; 20 µM), and/or bilirubin (20 µM) for 24 h. Results are mean ± SEM (n = 6). * Statistically significant effect of TNFα/HG. ⁺ Statistically significant effect of Cana. ^# Statistically significant effect of SnPP or HO-1 siRNA.