SB202190 inhibits endothelial cell apoptosis via induction of autophagy and heme oxygenase-1

Abstract

Activation of the p38 mitogen-activated protein kinase (MAPK) pathway has been implicated in various detrimental events finally leading to endothelial dysfunction. The present study therefore investigates the impact of the p38 MAPK inhibitor SB202190 on the expression of the cytoprotective enzyme heme oxygenase-1 (HO-1) as well as metabolic activity, apoptosis and autophagy of endothelial cells. Using human umbilical vein endothelial cells (HUVEC) SB202190 was found to cause a time- and concentration-dependent induction of HO-1 protein. Induction of HO-1 protein expression was mimicked by SB203580, another p38 MAPK inhibitor, but not by SB202474, an inactive structural analogue of p38 MAPK inhibitors. HO-1 induction by both SB202190 and SB203580 was also demonstrated by analysis of mRNA expression. On the functional level, SB202190 was shown to increase metabolic activity and autophagy of HUVEC along with diminishing basal apoptosis. Treatment of cells with tin protoporphyrin IX (SnPPIX), a well-characterised HO-1 enzymatic inhibitor, or HO-1 siRNA left SB202190-modulated metabolic activity and autophagy virtually unaltered but caused a significant reversal of the anti-apoptotic action of SB202190. Conversely, however, HO-1 expression by SB202190 became completely suppressed by the autophagy inhibitor bafilomycin A₁. Bafilomycin A₁ likewise fully reversed effects of SB202190 on metabolic activity and apoptosis, albeit significantly inducing apoptosis per se. Collectively, this work demonstrates SB202190 to confer upstream induction of autophagy followed by HO-1 induction resulting in potential protective effects against apoptosis. On the other hand, our data oppose HO-1 to contribute to SB202190-mediated increases in metabolic activity and autophagy, respectively.

Keywords: SB202190; apoptosis; autophagy; heme oxygenase-1; p38 MAPK.

Figure 1. Effect of p38 MAPK inhibitors on HO-1 protein and mRNA expression in HUVEC

Cells were incubated with p38 MAPK inhibitors (SB202190, SB203580) or inactive structural analogue SB202474 at indicated concentrations (A) or at 10 μM (B, C, D) for 24 h (A, C, D) or as indicated (B). After incubation, cells were analysed for protein or mRNA expression of HO-1. Protein and mRNA expression values were normalised to β-actin. Percent control represents comparison with the respective vehicle-treated time-matched group (set as 100%). Values are means ± SEM of n = 3 (A), n = 5 (B, C) or n = 5–6 (D) experiments. ^*P < 0.05 vs. time-matched vehicle control; one-way ANOVA plus post hoc Dunnett test (A, C, D) or Student’s two-tailed t test (B).

Figure 2. Impact of SB202190 on metabolic activity, apoptosis and cell cycle progression of HUVEC

Cells were incubated with increasing concentrations of SB202190 or vehicle control for 24 h (A, C, E) or with 10 μM SB202190 for the indicated times (B, D). Following incubation, cells were analysed for metabolic activity using WST-1 colorimetric assay (A, B), DNA fragmentation (C, D) or cell cycle distribution using flow cytometry (E). Exemplary images of flow cytometry analysis are shown for vehicle control and 10 μM SB202190, respectively (F). Percent control represents comparison with the respective vehicle-treated time-matched group. Values are means ± SEM of n = 15–18 (A), n = 19–22 (B), n = 7–8 (C), n = 11–12 (D) and n = 3 (E) experiments. ^*P < 0.05 vs. time-matched vehicle control; one-way ANOVA plus post hoc Dunnett test (A, C, E) or Student’s two-tailed t test (B, D).

Figure 3. SB202190 triggers activation of autophagy in HUVEC

Cells were incubated with increasing concentrations of SB202190 or vehicle control for 24 h (A) or with 10 μM SB202190 for the indicated times (B). Following incubation, cells were harvested and lysates were analysed for autophagy-related protein LC3A/B-I/II. Protein expression values were normalised to β-actin. Percent control represents comparison with the respective vehicle-treated time-matched group (set as 100%). Values are means ± SEM of n = 4 (A) or n = 3 (B) experiments. Statistical analysis revealed no significant differences between SB202190-treated groups and time-matched vehicle controls.

Figure 4. Impact of the HO-1 inhibitor SnPPIX on effects of SB202190 on metabolic activity, apoptosis and autophagy

HUVEC were pre-incubated with the HO-1 activity inhibitor SnPPIX (25 μM) for 1 h followed by addition of SB202190 (10 μM) and continuation of incubation for another 24 h. Thereafter, cells were analysed for metabolic activity (A) and DNA fragmentation (B). Lysates of cells were analysed for autophagy-related protein LC3A/B-I/II (C) and HO-1 (D). Protein expression values were normalised to β-actin. Percent control represents comparison with vehicle-treated group (set as 100%). Values are means ± SEM of n = 21–24 (A), n = 10–12 (B), n = 12 (C) or n = 16 (D) experiments. ^*P < 0.05 vs. vehicle control; ^#P < 0.05 vs. SB202190; one-way ANOVA plus post hoc Bonferroni test.

Figure 5. Impact of HO-1 siRNA on effects of SB202190 on metabolic activity, apoptosis and autophagy

HUVEC were transfected with HO-1-specific siRNA or non-targeting siRNA (NON) 24 h prior to stimulation with SB202190 (10 μM). Following a 24-h incubation, cells were analysed for metabolic activity (A) and DNA fragmentation (B). Lysates of cells were analysed for autophagy-related protein LC3A/B-I/II (C) and HO-1 (D). Protein expression values were normalised to β-actin. Percent control represents comparison with vehicle-treated group (set as 100%). Values are means ± SEM of n = 13–16 (A), n = 18–20 (B) or n = 20 (C, D) experiments. ^*P < 0.05 vs. vehicle control; ^#P < 0.05 vs. SB202190; one-way ANOVA plus post hoc Bonferroni test.

Figure 6. Impact of the autophagy inhibitor bafilomycin A₁ on effects of SB202190 on metabolic activity, apoptosis, autophagy and HO-1 expression

HUVEC were pre-incubated with the autophagy inhibitor bafilomycin A₁ (2.5 nM) for 1 h followed by addition of SB202190 (10 μM) and continuation of incubation for another 24 h. Thereafter, cells were analysed for metabolic activity (A) and DNA fragmentation (B). Lysates of cells were analysed for protein expression of LC3A/B-I/II (C) and HO-1 (D). Protein expression values were normalised to β-actin. Percent control represents comparison with vehicle-treated group (set as 100%). Values are means ± SEM of n = 19–25 (A), n = 14–15 (B), n = 11 (C) and n = 12 (D) experiments. ^*P < 0.05 vs. vehicle control, ^#P < 0.05 vs. SB202190; one-way ANOVA plus post hoc Bonferroni test.

Figure 7. Proposed mechanism underlying the cytoprotective action of SB202190 in HUVEC

The p38 MAPK inhibitor SB202190 inhibits apoptosis of endothelial cells by activation of autophagy followed by induction of the cytoprotective enzyme HO-1. Activation of autophagy was substantiated by increased metabolic activity and enhanced phosphatidyl ethanolamine (PE)-conjugation of LC3-I protein, resulting in LC3-II protein [43]. Inhibition or knockdown of HO-1 reversed anti-apoptotic effects, but not autophagy activation mediated by SB202190. Inhibition of autophagosome formation by late-phase autophagy inhibitor bafilomycin A₁ [44, 48] reversed pro-metabolic effects and HO-1 induction by SB202190 thereby abolishing anti-apoptotic effects of the p38 MAPK inhibitor in HUVEC.