Identification of stable cytotoxic factors in the gas phase extract of cigarette smoke and pharmacological characterization of their cytotoxicity.

Smoking is a major risk factor for atherosclerotic vascular diseases, but the mechanism for its genesis is unknown. We have recently shown that the gas phase of cigarette smoke (nicotine- and tar-free cigarette smoke extract; CSE) likely to reach the systemic circulation contains stable substances which cause cytotoxicity like plasma membrane damage and cell death in cultured cells, and also that the plasma membrane damage is caused through sequential activation of protein kinase C (PKC) and NADPH oxidase (NOX) and the resulting generation of reactive oxygen species (PKC/NOX-dependent mechanism), whereas cell death is caused through PKC/NOX-dependent and -independent mechanisms. To identify these stable substances, the CSE was prepared by passing the main-stream smoke of 10 cigarettes through a Cambridge glass fiber filter, trapping of the smoke in a vessel cooled at -80°C, and subsequent dissolution in 10ml of water. The CSE was fractionated into nine fractions using reversed-phase HPLC, and each fraction was screened for cytotoxicity in cultured cells, using propidium iodide uptake assay for cell membrane damage and MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium] reduction assay for cell viability. The cytotoxicity was positive in two of the nine fractions (Fr2 and Fr5). After extraction of the active fractions into dichloromethane, GC/MS analysis identified 2-cyclopenten-1-one (CPO) in Fr5 but none in Fr2. After derivatization of the active fractions with O-(2,3,4,5,6-pentafluorobenzyl) hydroxylamine hydrochloride, GC/MS analysis identified acrolein, acetone and propionaldehyde in Fr2, and methyl vinyl ketone (MVK) in Fr5. After 4-h incubation, authentic acrolein and MVK induced concentration-dependent cytotoxicity with EC50 values of 75.9±8.2 and 47.0±8.0μM (mean±SEM; n=3), respectively, whereas acetone, propionaldehyde and CPO were without effect. However, after 24-h incubation, CPO induced concentration-dependent cytotoxicity with an EC50 value of 264.0±16.9μM (n=3). The concentrations of acrolein, MVK and CPO in the CSE were 3368±334, 2429±123 and 392.9±31.8μM (n=4), respectively, which were higher than the cytotoxic concentrations. The cytotoxicity of acrolein and MVK consisted of plasma membrane damage and decreased cell viability: the plasma membrane damage was totally prevented by treatment with an inhibitor of PKC or NOX, whereas the decreased cell viability was only partially prevented by these inhibitors. The cytotoxicity of CPO consisted only of decreased cell viability, which was totally resistant to these inhibitors. These results show that acrolein and MVK are responsible for the acute cytotoxicity of the CSE through PKC/NOX-dependent and -independent mechanisms, whereas CPO is responsible for the delayed cytotoxicity of the CSE through a PKC/NOX-independent mechanism.