Aqueous CS extracts (CSE) were prepared by the method described previously, with a modification 12. Immediately before use, mainstream smoke was generated with one cigarette (Peace^®; Japan Tobacco Inc.) by drawing consecutive puffs into a 20-ml plastic syringe with a stopcock connected through one port to a glass vessel containing 3 ml of phosphate-buffered saline (PBS). A 20-ml puff drawn over 1 second was obtained at 10-second intervals. Each puff was held for 3 seconds and bubbled through the PBS for 5 seconds. One cigarette yielded an average of 45 puffs by this procedure. The aqueous CSE was diluted in culture medium before use, and the CSE solutions were prepared by the same person (M.S.) by exactly the same method, and they were used within 5 minutes after preparation.

Primary human umbilical vein endothelial cells (HUVECs) were obtained from Clonetics (San Diego, CA). Frozen cells were thawed, plated onto a 100-mm dish (Falcon^® Becton Dickinson Labware, Franklin Lakes, NJ) precoated with bovine type I collagen (Vitrogen^®; Cohesion Technologies, Palo Alto, CA), and grown in endothelial growth medium (EBM-2 containing hydrocortisone, hFGF-B, VEGF, R3-IGF-1, ascorbic acid, heparin, 2% FBS, hEGF, gentamicin, and amphotericin B; Clonetics CC-4176). When the cells reached 70% confluence, they were trypsinized with 0.1% trypsin (Gibco BRL^® Life Technologies, Inc., Grand Island, NY), and 2% EDTA (Gibco BRL^®), neutralized with soybean trypsin inhibitor (Gibco BRL^®), rinsed once with PBS, and seeded onto a new dish. Cells between passages 3 and 5 were used in the experiments. Human neutrophils from EDTA-anticoagulated venous blood specimens obtained from healthy volunteers were isolated by dextran sedimentation and centrifugation on a Histopaque gradient (without endotoxin; Sigma-Aldrich Japan, Tokyo, Japan), as previously described 12. The purity of the neutrophil populations was >95% based on May-Grunewald-Giemsa staining, and neutrophil viability determined by trypan blue dye exclusion was >98%.

HUVECs at a density of (10⁴ cells/mm²) were plated onto 96-well plates (Falcon^®), 8-well chamber slides (Lab-Tek Nalge^® Nunc International, Tokyo, Japan), or onto 60-mm plates (Falcon^®) precoated with type I collagen, and grown in endothelial growth medium. When the cells reached confluence, they were incubated for 8 hours with EBM-2 medium lacking growth factors, and then exposed to different concentrations of H₂O₂ (Wako Pure Chemical Industries, Ltd., Osaka, Japan) or CSE for 16 hours in the presence or absence of catalase (500 U/ml; Sigma-Aldrich). Cell viability after exposure to H₂O₂ or CSE was assessed by staining with 40 μg/ml of propidium iodide (Sigma-Aldrich) and spectrofluorimetry at an excitation wavelength of 530 nm and emission wavelength of 590 nm, as described previously 13.

Membrane FasL expression in HUVECs was analyzed by cell-monolayer-based spectrofluorimetry or flow cytometry. For cell monolayer-based spectrofluorimetry, HUVECs cultured on 96-well plates (Falcon^®) were fixed with cold 1% paraformaldehyde for 15 min. After rinsing with PBS, plates were blocked with blocking buffer (PBS containing 3% bovine serum albumin and 2% goat serum: Sigma-Aldrich) and incubated for 60 minutes at room temperature with monoclonal mouse anti-FasL antibody (NOK1, 8 μg/ml; Pharmingen, San Diego, CA) or isotype-matched mouse nonimmune IgG in blocking buffer. The plates were rinsed 3 times with PBS and incubated with FITC-conjugated anti-mouse IgG antibody (5 μg/ml; Southern Biotechnology Associates, Inc., Brimingham, AL) for 1 hour at room temperature. Fluorescence was measured on a Cytofluor II multiplate fluorometer (Perceptive Biosystems, Framingham, MA) at an excitation wavelength of 485 nm and emission wavelength of 530 nm.

For flow cytometry, cells were detached from 60-mm culture plates with 0.5% EDTA solution containing matrix metalloproteinase inhibitor (KB8301, 10 μg/ml; Pharmingen) and incubated for 1 hour at room temperature with anti-FasL mAb (NOK1, 8 μg/ml) or isotype-matched mouse nonimmune IgG (Dako Cytomation Japan, Tokyo, Japan) in blocking buffer. Cells were then washed 3 times with PBS and stained with FITC-conjugated anti-mouse IgG antibody for 1 hour at room temperature. FITC-labeled cells were detected using a flow cytometer (FACS) on an FL-1 channel.

Amounts of soluble Fas L in samples of the supernatant from the 96-well plate cultures of HUVECs were analyzed by using a soluble FasL enzyme-linked immunosorbent assay (ELISA) kit (R&D systems Inc., Minneapolis, MN) according to manufacturer's instructions.

The ability of FasL on endothelial cells to induce apoptosis in Fas-positive target cells was assessed by co-incubating neutrophils with HUVECs. HUVECs were grown to confluence in 96-well plates and exposed to 100 μM H₂O₂ for 16 hours. The plates were then washed 3 times with PBS, and neutrophils (10⁵ cells) suspended in EBM-2 medium lacking growth factors were added to HUVECs. The plates were centrifuged at 200 × g for 2 minutes, and after incubation at 37°C for 16 hours, non-adherent cells were collected and attached to glass slides by cytocentrifugation. The slides were fixed with a cold mixture of 60% acetone and 2% glutaraldehyde, reacted with a solution consisting of 250 μg/ml 3,3'-diaminobenzidine (Dako Cytomation) and 0.01% H₂O₂ for 10 minutes at room temperature, and stained with a Giemsa solution. Apoptotic neutrophils were identified as cells exhibiting nuclear pyknosis or chromatin condensation 12 together with brown-colored cytoplasmic staining when examined by oil immersion microscopy. Three hundred cells were scored in each experiment to determine the percentage of apoptotic neutrophils cells.

The Animal Care and Use Committee of Tokyo Women's Medical University approved the animal protocol. Thoracic aortas were obtained from male SD rats (Sankyo Labo Service Co., Tokyo, Japan) weighing 250 g, and they were cultured in the presence or absence of 100 μM H₂O₂ for 16 hours in EBM-2 medium lacking growth factors. The thoracic aortas were then fixed in 3% paraformaldehyde, and paraffin-embedded specimens were cut into 3-μm sections. Sections were deparaffinized, digested with a pepsin solution (Research Genetics, Huntsville, AL) for 5 minutes at 40°C, and exposed to 3% H₂O₂ for 20 minutes to inhibit endogenous peroxidase. After blocking with 3% bovine serum albumin and 2% goat serum, sections were incubated with polyclonal rabbit anti-FasL antibody (Wako Pure Chemical Industries, Ltd., Osaka, Japan) or non-immune rabbit IgG (Dako Cytomation) for 1 hour at room temperature. The primary antibody was reacted with biotin-conjugated anti-rabbit IgG antibody and streptavidin-horseradish peroxidase (Research Genetics). Immunoreactants were visualized with a substrate solution of 3-amino-9-ethylcarbazole (Dako Cytomation) and nuclear counterstaining with a hematoxylin solution.

All data are reported as means ± SEM. The mean values were compared by one-way analysis of variance (ANOVA) with Fisher's protected least significance difference (PLSD) post hoc analysis and the unpaired t test. A p value < 0.05 was considered significant.

Since as shown in Figure 1, exposure to H₂O₂ at concentrations of 1–100 μM and to CSE at concentrations of 0.5% and 1% did not affect HUVEC viability, we used these concentrations of H₂O₂ and CSE to assess their effect on FasL expression. The results showed that exposure of HUVECs to H₂O₂ dose-dependently increased the level of membrane FasL expression as assessed by cell-monolayer-based spectrofluorimetry (Figure 2A), and flow cytometry also showed an increased level of membrane FasL expression on HUVECs after H₂O₂ exposure (Figure 2C). Since cigarette smoke is an important source of oxidants that may affect endothelial function, we investigated whether CSE enhances membrane FasL expression by HUVECs. As shown in Figure 2B, HUVEC exposure to 0.5% and 1% CSE increased the level of membranous FasL expression. The increase was significantly inhibited in the presence of catalase, which catalyzes H₂O₂, suggesting that CSE induces FasL expression at least in part by oxidative mechanisms.

Since membrane FasL is converted to a soluble form, we measured the amount of soluble FasL released by HUVECs exposed to H₂O₂. As shown in Figure 3, HUVEC exposure to H₂O₂ increased the amount of soluble FasL in the culture medium.

To determine the functional capacity of HUVECs to induce apoptosis, we quantified Fas-positive neutrophils 14 co-cultured with HUVECs after exposure to H₂O₂. As shown in Figure 4, exposure of HUVECs to 100 μM H₂O₂ significantly promoted neutrophil apoptosis, suggesting that oxidative stress renders HUVECs cytotoxic to neutrophils.

The results obtained thus far had suggested that oxidative stress enhanced FasL expression in a HUVEC monolayer. We then investigated whether oxidative stress increases FasL expression on the endothelium of vascular tissues cultured in vitro. Since arteries have been shown to express low levels of FasL 4, we obtained thoracic aortas from rats and incubated them with or without 100 μM H₂O₂. Very weak immunoreactivity against anti-FasL was observed on the endothelium of thoracic aortas cultured with medium alone (Figure 5A), whereas culturing the thoracic aortas with H₂O₂ considerably enhanced endothelial immunoreactivity against anti-FasL (Figure 5B), suggesting that oxidative stress increases FasL expression on the endothelium of arteries.