Chronic treatment with N-acetyl-cystein delays cellular senescence in endothelial cells isolated from a subgroup of atherosclerotic patients

Abstract

Endothelial senescence may contribute to the pathogenesis of age-related vascular disorders. Furthermore, chronic exposure to risk factors for cardiovascular disease (CVD) accelerates the effects of chronological aging by generating stress-dependent damages, including oxidative stress, therefore promoting stress-induced premature senescence. Our objective was to determine whether a chronic treatment with an antioxidant (N-acetyl-cystein, NAC) could delay senescence of endothelial cells (EC) isolated and cultured from arterial segments of patients with severe coronary artery disease. If EC were considered as one population (n=26), chronic NAC treatment slightly shortened telomere attrition rate associated with senescence but did not significantly delay the onset of endothelial senescence. However, in a subgroup of NAC-treated EC (n=15) cellular senescence was significantly delayed, NAC decreased lipid peroxidation (HNE), activated the catalytic subunit of telomerase (hTERT) and inhibited telomere attrition. In contrast, in another subgroup of EC (n=11) characterized by initial short telomeres, no effect of NAC on HNE and high levels of DNA damages, the antioxidant was not beneficial on senescence, suggesting an irreversible stress-dependent damage. In conclusion, chronic exposure to NAC can delay senescence of diseased EC via hTERT activation and transient telomere stabilization, unless oxidative stress-associated cell damage has become irreversible.

Figures

Fig. 1

Dual effect of chronic treatment with N-Acetyl-Cystein on the time course of senescence. Two populations of EC were isolated according to their response to NAC (10 μM) on senescence: time course of senescence was either unaffected (n = 11) (A and B) or delayed (n = 15) (C and D). Unaffected senescence was evidenced by a slightly shorter time needed to reach 50% of senescence (SA-β-Gal positive cells) ( *p < 0.05 compared to control untreated cells) with no change in the replication potential (A) or cell morphology (B). Delayed senescence was demonstrated by a longer time needed to reach 50% of β-Gal positive cells ( p < 0.05) and an increased replication potential (C) ( *p < 0.05 compared to control untreated cells) as well as an improvement in cell morphology (D).

Fig. 2

Dual effect of chronic treatment with N-Acetyl-Cystein on telomere length. Telomere length (RFL, bp) was measured by Southern blot in control and NAC-treated cells, initially (initial passage) and when cells reached senescence (final passage). (A) In NAC-unaffected EC, RFL were not affected by the antioxidant. (B) In cells where NAC-delayed senescence, telomere shortening was prevented by NAC. Representative Southern blot and the distribution (%) of short (5–10 kbp), intermediate (10–15 kbp) and long (15–20 kb) telomeric fragments length are shown in control non-treated EC (C), NAC-treated unaffected EC (D) and NAC-delayed EC. (E) *p < 0.05 final passage vs. initial passage in control cells.

Fig. 3

Modulation of oxidative stress by NAC. (A) Typical paired measurement (Control/NAC) of lipid peroxidation in fixed EC by HNE immunostaining in control and NAC-treated cells. (B) Average levels of HNE (intensity of fluorescence detected by immunofluorescent staining, a.u.) in control (n = 7–9) and NAC-treated (n = 11) EC. *p < 0.05 compared to untreated cells in the “delayed senescence” group. †p < 0.05 compared to NAC-treated cells in “unaffected senescence” group.

Fig. 4

Modulation of DNA damage makers by NAC. Immunostaining of H2AX and ATM in NAC-treated EC, at low passage (P4) (A) and at senescence (final P) (B), in cells in which senescence was either unaffected or delayed by the chronic antioxidant treatment. Bar graphs of H2AX (C) and ATM (D) staining of EC in the “delayed” group (n = 10) and in the “unaffected” group (n = 11). Data represent the ratio of nuclear/cytosolic fluorescent signal corrected by the number of cells. Bars = 20 μm. *p < 0.05 compared to EC at passage 4, in the “unaffected” or “delayed” category. †p < 0.05 compared to EC in the “unaffected” category.

Fig. 5

Positive correlation between ATM and H2AX. Immunostaining of H2AX (red), ATM (green) and TOPRO-3 (nuclear staining in blue) in NAC-treated EC, at senescence (final P), in cells in which senescence was either unaffected or delayed by the chronic antioxidant treatment. H2AX and ATM positively correlate (n = 11). Data represent the ratio of nuclear/cytosolic fluorescent signal corrected by the number of cells. Bars = 20 μm.

Fig. 6

Modulation cellular damage makers by NAC. Immunostaining of PML bodies (red), p53 (green) and TOPRO-3 (nuclear staining in blue) (A) in control senescent EC and in NAC-treated EC in which senescence was delayed by the chronic antioxidant treatment. (B) Bar graphs of PML and p53 staining of EC (n = 14). (C) p53 and PML positively correlate. Data represent the ratio of nuclear/cytosolic fluorescent signal corrected by the number of cells. Bars = 20 μm. *p < 0.05 compared to control EC.

Fig. 7

Modulation of hTERT by NAC. Confocal images of hTERT-immunostained cells in control and NAC-treated EC. Translocation of the enzyme from the cytosol to the nucleus was observed if the antioxidant delayed the appearance of senescence (B) but not if NAC did not affect the time course of senescence (A). Cytosolic and nuclear hTERT activity (a.u.) detected by real-time-TRAP assay, in control and NAC-treated cells (C) not affected (n = 5) or positively affected (n = 5) by NAC. The bar scale represents 10 μm. *p < 0.05 compared to control cells. †p < 0.05 compared to nuclear hTERT activity in NAC-treated cells from the “unaffected senescence” group.

Fig. 8

Schematic representation of molecular pathways of senescence. In control untreated EC, p53-dependent stress-induced senescence (SIS) can be triggered either by ROS-dependent caveolin-1 activation, or by ROS-dependent accumulation of DNA damage. Replicative senescence is mainly triggered by excessive telomere shortening which is recognized as DNA damage as reflected by markers such as ATM/H2AX. When treated chronically with NAC, caveolin-1 expression was abolished in both the unaffected and the delayed senescence group. We showed however, that when the levels of HNE and p53, both suppressors of hTERT activity, are too high, hTERT is not activated, telomere shortening is not prevented, DNA damage accumulates and senescence is unaffected. In contrast, when HNE and p53 levels are lowered, NAC activated hTERT and this was associated with a delay in the onset of senescence. These EC however, still accumulate DNA damages associated with a loss of endogenous DNA repair mechanism and thus, finally, reach senescence.