Previously, we identified global epigenetic aberrations in smoking-associated oral squamous cell carcinoma (OSCC). We hypothesized that cigarette exposure triggers OSCC through alteration of the methylome of oral cells. Here we report that cigarette smoke condensate (CSC) significantly changes the genomic 5-methyldeoxycytidine content and nuclear accumulation of DNA methyltransferase 1 (DNMT1) and DNMT3A in human untransformed oral cells. By using integrated analysis of cDNA and methylation arrays of the smoking-associated dysplastic oral cell line and OSCC tumors, respectively, we identified four epigenetic targets—UCHL1, GPX3, LXN, and LDOC1—which may be silenced by cigarette. Results of quantitative methylation-specific PCR showed that among these four genes, LDOC1 promoter was the most sensitive to CSC. LDOC1 promoter hypermethylation and gene silencing followed 3 weeks of CSC treatment. LDOC1 knockdown led to a proliferative response and acquired clonogenicity of untransformed oral cells. Immunohistochemistry showed that LDOC1 was downregulated in 53.3% (8/15) and 57.1% (20/35) of premalignant oral tissues and early stage OSCCs, respectively, whereas 76.5% (13/17) of normal oral tissues showed high LDOC1 expression. Furthermore, the microarray data showed that LDOC1 expression had decreased in the lung tissues of current smokers compared with that in those of never smokers and had significantly decreased in the lung tumors of smokers compared with that in normal lung tissues. Our data suggest that CSC-induced promoter methylation may contribute to LDOC1 downregulation, thereby conferring oncogenic features to oral cells. These findings also imply a tumor suppressor role of LDOC1 in smoking-related malignancies such as OSCC and lung cancer.

trans,trans-2,4-Decadienal (tt-DDE), a lipid peroxidation product of linolieic acid, is the most abundant aldehyde identified in cooking oil fumes and is readily detectable in food products as well as in restaurant emissions. Previously, we have reported the toxicological effects of tt-DDE in vitro and in vivo. However, the metabolic pathways of tt-DDE in vivo remain unclear. In our present study, we combined liquid chromatography-mass spectrometry with triple quadrupole and time-of-flight to identify tt-DDE metabolites in the urine of mice orally administered tt-DDE. We identified two tt-DDE metabolites, 2,4-decadienoic acid and cysteine-conjugated 2,4-decadien-1-ol, in the urine of mice gavaged with tt-DDE and in human hepatoma cell cultures. The structure of 2,4-decadienoic acid was confirmed upon comparison of its tandem mass spectrometry (MS/MS) spectrum and retention time with those of synthetic standards. The moieties of cysteine and alcohol on cysteine-conjugated 2,4-decadien-1-ol were validated by treating cell cultures with stable-isotope-labeled cysteine and 4-methylpyrazole, an alcohol dehydrogenase inhibitor. The MS/MS spectra of a cysteine standard and ionized cysteine detached from cysteine-conjugated 2,4-decadien-1-ol were identical. Two metabolic pathways for the biotransformation of tt-DDE in vivo are proposed: (i) the oxidation of tt-DDE to the corresponding carboxylic acid, 2,4-decadienoic acid, in liver cells and (ii) glutathione (GHS) conjugation, GSH breakdown, and aldehyde reduction, which generate cysteine-conjugated 2,4-decadien-1-ol in both liver and lung cells. In conclusion, this platform can be used to identify tt-DDE metabolites, and cysteine-conjugated 2,4-decadien-1-ol can serve as a biomarker for assessing exposure to tt-DDE.