An oxidized extracellular oxidation-reduction state increases Nox1 expression and proliferation in vascular smooth muscle cells via epidermal growth factor receptor activation

Abstract

Objective: To examine the effect of an oxidized extracellular oxidation-reduction (redox) state (E(h)) on the expression of NADPH oxidases in vascular cells.

Methods and results: The generation of reactive oxygen species by NADPH oxidase (Nox)-based NADPH oxidases activates redox-dependent signaling pathways and contributes to the development of "oxidative stress" in vascular disease. An oxidized plasma redox state is associated with cardiovascular disease in humans; however, the cellular mechanisms by which the extracellular redox state may cause disease are not known. Aortic segments and cultured aortic smooth muscle cells were exposed to E(h) between -150 mV (reduced) and 0 mV (oxidized) by altering the concentration of cysteine and its disulfide, cystine, the predominant redox couple in plasma. A more oxidized E(h) increased the expression of Nox1 and resulted in Nox1-dependent proliferation of smooth muscle cells. Oxidized E(h) rapidly induced epidermal growth factor receptor phosphorylation via shedding of epidermal growth factor-like ligands from the plasma membrane and caused extracellular signal-regulated kinase 1/2-dependent phosphorylation of the transcription factors activating transcription factor-1 and cAMP-response element-binding protein. Inhibition of epidermal growth factor receptor or extracellular signal-regulated kinase 1/2 activation, or addition of small interference RNA to activating transcription factor-1, prevented the increase in Nox1 expression.

Conclusions: Our results identify a novel mechanism by which extracellular oxidative stress increases expression and activity of Nox1 NADPH oxidase and contributes to vascular disease.

Figure 1

Oxidized extracellular E_h increases Nox1 expression and growth of vascular cells. Isolated aortae and SMCs were exposed to varying concentrations of Cys and CySS to obtain a range of E_h from −150 mV to 0 mV. (A) Representative images show aortic outgrowth on Matrigel. Mean outgrowth area was calculated from 8 aortic rings in each group obtained from two animals. Nox1 mRNA expression was measured by qRT-PCR (n=4). (B) Nox1 and Nox4 mRNA expression in SMCs was measured by qRT-PCR (n=4). (C) Nox1 protein expression in SMCs was measured by Western blotting. Results are normalized to GAPDH expression (n=6). (D) Superoxide levels in SMCs were assessed by DHE fluorescence. (E) Mean DHE fluorescence was obtained from 14–16 fields per group in n=2 independent experiments. (F) SMC proliferation was measured by [³H]-thymidine incorporation (n=4). Data are presented as mean ± SE. * p < 0.05 vs. −150 mV, Δ p < 0.05 vs. 0 mV Ad-empty.

Figure 2

EGFR is phosphorylated by oxidized extracellular E_h via activation of metalloproteinases. (A) SMCs were treated with E_h=0 mV and p-EGFR expression analyzed by Western blotting. Results are normalized to total EGFR expression (n=4). (B) SMCs were incubated with 20 μM GM6001, then treated with E_h=0 mV for 1 min. Phospho-EGFR was analyzed by Western blotting. Results are normalized to total EGFR expression (n=4). Blots shown under A and B were run on the same gel but cut for better presentation. (C) Fluorogenic peptide substrates were added to E_h=0 mV media, then collected at specific time points for measurement of metalloproteinase activity. An appropriate blank (without fluorogenic substrate) was subtracted from each measurement (n=5). (D) Gelatin zymography of matrix metalloproteinase activity after control (c) and E_h=0 mV (t). Representative gel of n=3 independent experiments is shown. Summary data are presented as mean ± SE. * p < 0.05 vs. untreated (0 min); Δ p < 0.05 vs. 0 mV vehicle.

Figure 3

Oxidized extracellular E_h increases Nox1 expression via the metalloproteinase-EGFR pathway. SMCs were pretreated with 10 μM AG1478 (A) or 20 μM GM6001 (B) and then exposed to E_h=0 mV for 24 hours. Nox1 mRNA expression was measured by qRT-PCR (n=4). SMCs were transfected with Nox1 promoter luciferase or control plasmid, then pretreated with 10 μM AG1478 (C) or 20 μM GM6001 (D) and exposed to E_h=0 mV for 8 hours. Firefly luciferase activity was normalized to Renilla luciferase (n=4). Data are presented as mean ± SE. * p < 0.05 vs. control; Δ p < 0.05 vs. 0 mV vehicle.

Figure 4

Oxidized extracellular E_h activates ERK1/2 and Akt via the metalloproteinase-EGFR pathway, but only ERK1/2 mediates the associated increase in Nox1 expression. SMCs were exposed to E_h=0 mV and p-ERK1/2 (A, B) or p-Akt (D, E) were analyzed by Western blotting after pretreatment with 20 μM GM6001 (A, D) or 10 μM AG1478 (B, E). Results are normalized to total ERK (A, B) or GAPDH (D, E) expression (n=3 to 5). Nox1 promoter luciferase activity was measured after pretreatment with 10 μM U0126 (C) or 1 μM Wortmannin (F) and exposed to E_h=0 mV for 8 hours. Firefly luciferase activity was normalized to Renilla luciferase (n=4 to 6). Data are presented as mean ± SE. * p < 0.05 vs. control; Δ p < 0.05 vs. 0 mV vehicle.

Figure 5

Oxidized extracellular E_h activates transcription factors CREB and ATF-1; however, only ATF-1 contributes to the increase in Nox1 expression. (A) SMCs were treated with E_h=0 mV and the nuclear fraction was subjected to Western blotting for p-CREB and p-ATF-1. Representative blot of n=3 independent experiments is shown. SMCs were transfected with 50 nM siRNA to ATF-1 (B) or CREB (C) and exposed to E_h=0 mV for 24 hours. Nox1 mRNA expression was measured by qRT-PCR. Scrambled siRNA served as a control. The data are presented as mean ± SE (n=5). * p < 0.05 vs. control; Δ p < 0.05 vs. 0 mV control siRNA.

Figure 6

ATF-1, but not CREB, is activated via the EGFR-ERK1/2 pathway in response to oxidized extracellular E_h. SMCs were pretreated with 10 μM AG1478 (A, D), 10 μM U0126 (B, F) or 1 μM wortmannin (C, E) and stimulated with E_h=0 mV. Summary data of p-ATF-1 and p- CREB expression analyzed by Western blotting are shown. Results are normalized to GAPDH expression and presented as mean ± SE (n=3 to 5). * p < 0.05 vs. control; Δ p < 0.05 vs. 0 mV vehicle.