Serum oxidative stress-induced repression of Nrf2 and GSH depletion: a mechanism potentially involved in endothelial dysfunction of young smokers

Abstract

Background: Although oxidative stress plays a major role in endothelial dysfunction (ED), the role of glutathione (GSH), of nuclear erythroid-related factor 2 (Nrf2) and of related antioxidant genes (ARE) are yet unknown. In this study we combined an in vivo with an in vitro model to assess whether cigarette smoking affects flow-mediated vasodilation (FMD), GSH concentrations and the Nrf2/ARE pathway in human umbilical vein endothelial cells (HUVECs).

Methods and results: 52 healthy subjects (26 non-smokers and 26 heavy smokers) were enrolled in this study. In smokers we demonstrated increased oxidative stress, i.e., reduced concentrations of GSH and increased concentrations of oxidation products of the phospholipid 1-palmitoyl-2-arachidonyl-sn-glycero-3-phosphorylcholine (oxPAPC) in serum and in peripheral blood mononuclear cells (PBMC), used as in vivo surrogates of endothelial cells. Moreover we showed impairment of FMD in smokers and a positive correlation with the concentration of GSH in PBMC of all subjects. In HUVECs exposed to smokers' serum but not to non-smokers' serum we found that oxidative stress increased, whereas nitric oxide and GSH concentrations decreased; interestingly the expression of Nrf2, of heme oxygenase-1 (HO-1) and of glutamate-cysteine ligase catalytic (GCLC) subunit, the rate-limiting step of synthesis of GSH, was decreased. To test the hypothesis that the increased oxidative stress in smokers may have a causal role in the repression of Nrf2/ARE pathway, we exposed HUVECs to increasing concentrations of oxPAPC and found that at the highest concentration (similar to that found in smokers' serum) the expression of Nrf2/ARE pathway was reduced. The knockdown of Nrf2 was associated to a significant reduction of HO-1 and GCLC expression induced by oxPAPC in ECs.

Conclusions: In young smokers with ED a novel further consequence of increased oxidative stress is a repression of Nrf2/ARE pathway leading to GSH depletion.

Figure 1. Endothelial function and correlation between flow-mediated vasodilation (FMD) and concentrations of GSH in peripheral blood mononuclear cells (PBMC) of non-smokers and smokers.

FMD and glyceryl trinitrate (GTN)-induced vasodilation in non-smokers and smokers (A); correlation between FMD and intracellular concentrations of GSH in PBMC of non-smokers and smokers (B). Data are presented as mean±SD; FMD and GTN are expressed as maximal percentage change in brachial artery dilation. *P<0.001 versus non-smokers.

Figure 2. Effect of serum derived from non-smokers and smokers and of lipoprotein-depleted serum (LPDS) derived from smokers on intracellular GSH concentration, on intracellular reactive oxygen species (ROS) and on nitric oxide (NO) formation in human umbilical vein endothelial cells (HUVECs).

Confluent HUVECs were incubated without and with 10, 30 and 50% serum derived from non-smokers and smokers and with the corresponding LPDS derived from smokers for 12 hours. Figure shows intracellular GSH concentration (A), intracellular ROS (B) and cumulative basal and bradykinin-stimulated NO production, evaluated by measuring levels of nitrite in the supernatants (C). Data are presented as mean±SD of measurements performed in triplicate in four different occasions; *P<0.01 versus control (no addition of serum derived from the subjects) non-smokers' serum and smokers' LPDS.

Figure 3. Effect of serum derived from non-smokers and smokers and of lipoprotein-depleted serum (LPDS) derived from smokers on Nrf2, GCLC and HO-1 expression in human umbilical vein endothelial cells (HUVECs).

Confluent HUVECs were incubated without and with increasing amounts (10, 30 and 50%) of serum derived from smokers (serum S 10, serum S 30, serum S 50), with 50% serum derived from non-smokers (serum N–S 50) and with 50% LPDS derived from smokers (LPDS S 50) for 12 hours. mRNA for Nrf2, GCLC and HO-1 (A–C) was analysed by quantitative Real-Time PCR. Normalised gene expression levels were given as the ratio between the mean value for the target gene and that for beta-actin in each sample. Results are reported as the mean±SD of measurements performed in triplicate. *P<0.01 versus non-smokers and LPDS. (D) shows a representative Western blot analysis of three independent experiments for nuclear Nrf2, for GCLC and HO-1.

Figure 4. Effect of increasing concentrations of oxPAPC on intracellular GSH concentration, on intracellular reactive oxygen species (ROS), and on nitric oxide (NO) production in human umbilical vein endothelial cells (HUVECs).

Confluent HUVECs were incubated without and with increasing concentrations (25–150 µg/mL) of oxPAPC for 6 hours. Figure shows intracellular GSH concentration (A), intracellular ROS (B) and cumulative basal and bradykinin-stimulated NO production, evaluated by measuring levels of nitrite in the supernatants (C). Data are presented as mean±SD of measurements performed in triplicate in four different occasions. *P<0.01 versus oxPAPC 0 e 75 µg/mL. # P<0.01 versus basal NO.

Figure 5. Effect of increasing concentrations of oxPAPC on Nrf2, GCLC and HO-1 expression in human umbilical vein endothelial cells.

mRNA (A–C) was analysed by quantitative Real-Time PCR. Normalised gene expression levels were given as the ratio between the mean value for the target gene and that for beta-actin in each sample. Results are presented as the mean±SD of measurements performed in triplicate.*P<0.01 versus oxPAPC 0 and 150 µg/mL. Figure (D) shows a representative Western blot analysis of three independent experiments for nuclear Nrf2, GCLC and HO-1.

Figure 6. Effect of small interfering (si) and scrambler (sc) RNA against Nrf2 on oxPAPC-dependent expression of GCLC and HO-1 in human umbilical vein endothelial cells.

mRNA (a) was analyzed by quantitative Real-Time PCR. Normalized gene expression levels were given as the ratio between the mean value for the target gene and that for the beta-actin in each sample. Figure shows a representative Western blot analysis for GCLC and HO-1 and the average quantification obtained by densitometric analysis of all the samples (b). Data on Western blot analysis are expressed as the density ratio of target to control (beta-actin) in arbitrary units ×10. Results are the mean±SD of measurements performed in triplicate in four different occasions. *P<0.01 versus control; †P<0.01 versus scRNA; #P<0.01 versus oxPAPC+scRNA.