Suppression of apoptosis by nitric oxide via inhibition of interleukin-1beta-converting enzyme (ICE)-like and cysteine protease protein (CPP)-32-like proteases

Abstract

Physiological levels of shear stress alter the genetic program of cultured endothelial cells and are associated with reduced cellular turnover rates and formation of atherosclerotic lesions in vivo. To test the hypothesis that shear stress (15 dynes/cm2) interferes with programmed cell death, apoptosis was induced in human umbilical venous cells (HUVEC) by tumor necrosis factor-alpha (TNF-alpha). Apoptosis was quantified by ELISA specific for histone-associated DNA-fragments and confirmed by demonstrating the specific pattern of internucleosomal DNA-fragmentation. TNF-alpha (300 U/ml) mediated increase of DNA-fragmentation was completely abrogated by shear stress (446 +/- 121% versus 57 +/- 11%, P <0.05). This anti-apoptotic activity of shear stress decreased after pharmacological inhibition of endogenous nitric oxide (NO)-synthase by NG-monomethyl-L-arginine and was completely reproduced by exogenous NO-donors. The activation of interleukin-1beta-converting enzyme (ICE)-like and cysteine protease protein (CPP)-32-like cysteine proteases was required to mediate TNF-alpha-induced apoptosis of HUVEC. Endothelial-derived nitric oxide (NO) as well as exogenous NO donors inhibited TNF-alpha-induced cysteine protease activation. Inhibition of CPP-32 enzyme activity was due to specific S-nitrosylation of Cys 163, a functionally essential amino acid conserved among ICE/CPP-32-like proteases. Thus, we propose that shear stress-mediated NO formation interferes with cell death signal transduction and may contribute to endothelial cell integrity by inhibition of apoptosis.

Figure 1

Inhibition of TNF-α–induced apoptosis by shear stress, nitric oxide and ICE-like and CPP-32–like protease inhibitors in human endothelial cells. (A) HUVEC were incubated with LNMA (1 mM) and/or TNF-α (300 U/ml) for 18 h with or without additional shear stress (ss) and DNA fragmentation was determined. Results are means ± SE, with * P <0.05 versus TNF-α; ** P <0.05 versus TNF-α + shear stress. (B) Dosedependent inhibition of TNF-α–induced apoptosis by the NO donors SNP and SNAP. HUVEC were incubated with SNP and SNAP with or without TNF-α (300 U/ml) for 18 h and DNA fragmentation was measured. (C) Effect of protease inhibitors or NO donors on TNF-α induced apoptosis after 18 h of incubation. Substances were added just before TNF-α addition in the following concentrations: SNP and SNAP (10 μM); ICE-like protease inhibitor Ac-YVAD-CHO and CPP-32–like inhibitor Ac-DEVD-CHO (100 μM); 8-bromo-cGMP (1 mM). Results are means ± SE, with * P <0.05 versus TNF-α.

Figure 1

Inhibition of TNF-α–induced apoptosis by shear stress, nitric oxide and ICE-like and CPP-32–like protease inhibitors in human endothelial cells. (A) HUVEC were incubated with LNMA (1 mM) and/or TNF-α (300 U/ml) for 18 h with or without additional shear stress (ss) and DNA fragmentation was determined. Results are means ± SE, with * P <0.05 versus TNF-α; ** P <0.05 versus TNF-α + shear stress. (B) Dosedependent inhibition of TNF-α–induced apoptosis by the NO donors SNP and SNAP. HUVEC were incubated with SNP and SNAP with or without TNF-α (300 U/ml) for 18 h and DNA fragmentation was measured. (C) Effect of protease inhibitors or NO donors on TNF-α induced apoptosis after 18 h of incubation. Substances were added just before TNF-α addition in the following concentrations: SNP and SNAP (10 μM); ICE-like protease inhibitor Ac-YVAD-CHO and CPP-32–like inhibitor Ac-DEVD-CHO (100 μM); 8-bromo-cGMP (1 mM). Results are means ± SE, with * P <0.05 versus TNF-α.

Figure 1

Inhibition of TNF-α–induced apoptosis by shear stress, nitric oxide and ICE-like and CPP-32–like protease inhibitors in human endothelial cells. (A) HUVEC were incubated with LNMA (1 mM) and/or TNF-α (300 U/ml) for 18 h with or without additional shear stress (ss) and DNA fragmentation was determined. Results are means ± SE, with * P <0.05 versus TNF-α; ** P <0.05 versus TNF-α + shear stress. (B) Dosedependent inhibition of TNF-α–induced apoptosis by the NO donors SNP and SNAP. HUVEC were incubated with SNP and SNAP with or without TNF-α (300 U/ml) for 18 h and DNA fragmentation was measured. (C) Effect of protease inhibitors or NO donors on TNF-α induced apoptosis after 18 h of incubation. Substances were added just before TNF-α addition in the following concentrations: SNP and SNAP (10 μM); ICE-like protease inhibitor Ac-YVAD-CHO and CPP-32–like inhibitor Ac-DEVD-CHO (100 μM); 8-bromo-cGMP (1 mM). Results are means ± SE, with * P <0.05 versus TNF-α.

Figure 2

(A) CPP-32–like activity in HUVEC treated with NO donors (10 μM) or Ac-DEVD (100 μM) and TNF-α (300 U/ml) for 18 h. (B) Northern blot analysis of CPP32. RNA from HUVEC was prepared after 6 h of incubation with TNF-α (300 U/ml), SNP and SNAP (10 μM) as indicated. 10 μg of total RNA was resolved, blotted and sequentially hybridized to full-length human cDNA of CPP-32 and GAPDH.

Figure 2

(A) CPP-32–like activity in HUVEC treated with NO donors (10 μM) or Ac-DEVD (100 μM) and TNF-α (300 U/ml) for 18 h. (B) Northern blot analysis of CPP32. RNA from HUVEC was prepared after 6 h of incubation with TNF-α (300 U/ml), SNP and SNAP (10 μM) as indicated. 10 μg of total RNA was resolved, blotted and sequentially hybridized to full-length human cDNA of CPP-32 and GAPDH.

Figure 3

Direct inhibition of ICE-like and CPP-32–like enzyme activity by NO. Inhibition of ICE-like and CPP-32–like activity in cell homogenates by preincubation with various concentrations of SNP for 5 min. Cell homogenates were obtained from TNF-α–stimulated HUVEC (300 U/ml for 18 h).

Figure 4

Purification of p17 subunits and direct inhibition of reconstituted CPP-32- and ICEsubunits by NO. (A) Silver stain of the crude homogenates before purification (60 μg; lane 3), the unbound fraction of the NiNTA columns (60 μg; lane 4) and the purified CPP-32 p17 subunit (10 μg; lane 2) separated by a 12% SDS–polyacrylamide gel. (B) Inhibition of cloned, purified and reconstituted CPP-32 and ICE by incubation with SNP or SNAP for 1 h.

Figure 4

Purification of p17 subunits and direct inhibition of reconstituted CPP-32- and ICEsubunits by NO. (A) Silver stain of the crude homogenates before purification (60 μg; lane 3), the unbound fraction of the NiNTA columns (60 μg; lane 4) and the purified CPP-32 p17 subunit (10 μg; lane 2) separated by a 12% SDS–polyacrylamide gel. (B) Inhibition of cloned, purified and reconstituted CPP-32 and ICE by incubation with SNP or SNAP for 1 h.