A different source of ROS initiates distinct signaling responses in endothelial cells. (A) HAECs were treated with 100 μM H₂O₂ for the indicated times, and total cell lysates were subjected to antibodies specific for pERK, ERK, pcJun, cJun, pp38, or p38. (B) HAECs were treated with 50 ng/ml EGF for the indicated times followed by immunoblotting with antibodies as in A. (C and D) Cells were treated with control adenovirus (Ad-LacZ) or catalase-overexpressing adenovirus (Ad-Cat) for 24 h before treatment with EGF followed by immunoblotting as in A. Error bars represent SD. *, P < 0.05. (E and F) HAECs were transfected with siRNA against nontargeting control (NT), Nox4, or Nox2 for 48 h before extraction of total RNA for RT-PCR and Western blotting (E) or treatment with EGF for 15 min followed by immunoblotting with pERK, ERK, Nox4, and Nox2 antibodies (F). Results are representative of four independent experiments.

PTP1B oxidation by Nox4 is spatially dependent. (A) COS7 cells were transfected with pcDNA3.1/PTP1B wild type or pcDNA3.1/PTP1B-Δ35 in addition to pcDNA3.1/Nox4-V5. After 24 h, cells were fixed and immunostained with anti-V5 (green) for Nox4-V5 and anti-PTP1B for PTP1B wild type and C-terminal deleted PTP1B-Δ35 (red). (B) Cells transfected as in A were also subjected to detergent fractionation followed by immunoblotting with anti-PTP1B and anti-GRP78. Fractions include cytosolic (Cyto), membrane/organelle (MO), nuclear (Nuc), and cytoskeletal/matrix (CSK/MAT). (C) COS-7 cells treated as in A were lysed and labeled with biotin polyethylene oxide maleimide. Subsequent immunoprecipitation with avidin followed by immunoblotting with anti-PTP1B yields a band that represents the reduced, active form of PTP1B. (D) COS-7 cells were transfected with pcDNA3.1/Nox4-V5 or its Nox4i-resistant version pcDNA3.1/Nox4-R in combination with/without pQ/Nox4i. PTP1B oxidation was assessed as in C, and the efficiency of Nox4 knockdown was probed by V5. Total cellular PTP1B is shown in the bottom panel. (E and F) HAECs were transfected with Ad-control siRNA (Ad-Ctli) or Ad-Nox4i for 48 h before lysis and labeling with biotin polyethylene oxide maleimide. Treatment of cells with H₂O₂ at 100 μM for 5 min served as a positive control. Blots are representative of three independent experiments. Error bars represent SD. *, P < 0.05. Bar, 10 μm.

Oxidation of PTP1B trapping mutant by Nox4 attenuates its substrate-binding capacity. (A) COS-7 cells were transfected with expression vectors as indicated for 24 h followed by treatment with or without 50 ng/ml EGF for 30 min. Cell lysates were immunoprecipitated with EGFR antibody and immunoblotted with antiphosphotyrosine. Total cell lysates were also immunoblotted with anti-EGFR as shown in the bottom panel. (B) COS-7 cells were cultured on the glass coverslips and transfected with expression vectors as indicated for 24 h. Imaging was visualized with a two-photon confocal microscope. (C) COS-7 cells were treated as in B, and cell lysates were immunoprecipitated with PTP1B antibody followed by immunoblotting with anti-EGFR and anti-PTP1B. Bar, 10 μm.

EGF-induced endothelial proliferation is mediated by Nox4-dependent PTP1B oxidation. HAECs were transfected with Ad-control siRNA (Ad-Ctli) or Ad-Nox4i for 48 h or adenoviral vectors expressing LacZ (AdLacZ)/Nox4 (AdNox4) for 24 h. (A and B) Total RNA was extracted followed by RT-PCR for measurement of Nox4 mRNA (A) and dichlorofluorescein assay for ROS production in response to 50 ng/ml EGF for 15 min (B). For proliferation assay, HAECs were transfected with adenoviral vectors as in A for 24 h and were seeded into 24-well plates at 2.5 × 10⁴ cells/well. (C and D) Proliferation was measured by BrdU incorporation (C), and cell number was counted at 24 and 48 h after plating (D). Data are shown as representative (gel picture) or statistics of three independent experiments. Error bars represent SD. *, P < 0.05.

Nox4 is localized to the ER. (A) HAECs cultured in EBM2 medium were fixed and subjected to immunostaining with anti-Nox4 (AlexaFluor488; green) and anti-GRP78 (AlexaFluor594; red) followed by two-photon confocal microscopy. (B) HAECs were transfected with adenoviral vector expressing Nox4-V5 (AdNox4-V5). After 24 h, cells were probed with anti-V5 and VE-cadherin antibodies and visualized with AlexaFluor594 (red) and AlexaFluor488 (green), respectively. (C) Cells were transfected with AdNox4-V5 as in B followed by immunogold staining and were visualized by electron microscopy. (D and E) Similarly, HAECs with overexpression of Nox4-V5 were fractionated by biochemical detergent method (D) or Nycodenz gradient centrifugation (E). The obtained fractions cytosolic (Cyto), membrane/organelle (MO), nuclear (Nuc), and cytoskeletal/matrix (CSK/MAT) in D or fractions from top to bottom in E were subjected to immunoblotting with antibodies. Bars: (A) 5 μm; (B) 10 μm.

Nox4-dependent PTP1B oxidation regulates EGF signaling. (A) COS-7 cells were transfected with pcDNA3.1/Nox4-V5 for 24 h followed by 50 ng/ml EGF treatment for the indicated times. The lysates were probed for phosphorylated and total EGFR and ERK. *, P < 0.05 compared with the respective control. (B) COS-7 cells were transfected with pcDNA3.1/PTP1B or pcDNA3.1/PTP1B-Δ35 in addition to pcDNA3.1/Nox4-V5 and treated as in A. Phosphorylation of EGFR was determined by immunoprecipitation with anti-EGFR followed by immunoblotting with antiphosphotyrosine. Total PTP1B and Nox4 levels were examined by using anti-PTP1B and anti-V5, respectively. (C) ERK phosphorylation was measured at 30 min after EGF treatment by immunoblotting with antiphospho-ERK. (D) PTP1B^−/− and PTP1B^−/−(+WT) MEFs were transfected with or without pcDNA3.1/Nox4 followed by treatment and detection as in B. (B–D) *, P < 0.05. (E) PTP1B^−/− MEFs were transfected with pcDNA3.1/PTP1B or pcDNA3.1/PTP1B-Δ35 in addition to pcDNA3.1/Nox4. Cells were collected for assay as in B. *, P < 0.05 compared with the respective control. (F) HAECs were transfected with Ad-control siRNA (Ad-Ctli) or Ad-Nox4i for 48 h and were treated with EGF for the indicated times. Phospho-ERK, total ERK, and Nox4 were determined. *, P < 0.05 compared with the respective Ad-Ctli. The blots are representative of three independent experiments. Error bars represent SD.

ER-targeting antioxidant attenuates Nox4-dependent ROS signaling. (A) COS-7 cells on coverslips were transfected with pcDNA3.1/Nox4-V5 or pcDNA3.1/N35-V5. After 24 h, cells were fixed and immunostained with anti-V5 (green) for fusion proteins and anti-GRP78 (red). (B) COS-7 cells were transfected with overexpression vectors as indicated and were immunostained with anticatalase antibodies. (C) COS-7 cells were transfected with mitochondrial-targeting catalase (MSP-Cat) or ER-targeting catalase (N35-Cat) as indicated for 24 h. Catalase activity was measured in cell lysates using the Amplex red catalase kit. Error bars represent SD. *, P < 0.05 versus control. (D) The Nox4-overexpressing COS-7 cell line was transfected with pcDNA3.1/PTP1B in combination with either MSP-catalase or N35-catalase. After 24 h, cells were treated with 50 ng/ml EGF, and phosphorylation of EGFR and ERK and PTP1B oxidation status were determined as described in Materials and methods. The blots are representative of three independent experiments. Bars, 10 μm.