The role of cigarette smoke-induced gene expression modifications in the etiogenesis of lung cancer

Epigenetic gene silencing in the form of aberrant promoter hypermethylation is an alternative to coding region mutation for loss of gene function. Downregulation of gene transcription by promoter methylation is a crucial event in lung cancer initiation and progression. Genes involved in many aspects of normal cellular function can be silenced by methylation. Promoters of more than 30 genes have been identified as hypermethylated in lung cancer so far. Since 90% of lung cancer patients are smokers, the aim of our study is to identify and characterize genes important in lung tumorigenesis, the silencing of which is a result of hypermethylation due to the exposure to cigarette smoke condensate. Three different approaches are adopted in this study: 1. Analysis of differences in gene expression profiles during chronic treatment of normal human bronchial epithelial cells (BEAS-2B) with cigarette smoke condensate (CSC); 2. Identification of re-activated genes after 5’Azadeoxycytidine and trychostatin A treatment of several lung cancer cell lines (CALU 1, CALU 6, SW2, etc); 3. Analysis of gene expression profiles in human lung tumors (smokers vs non smokers) using laser capture microdissection (LCM) to identify differences in gene expression profiles, we are using DNA microarray U133 2.0plus affymetrix chips containing probes representing the entire human genome. Our aim is to identify common genes using those three approaches and to verify their relevance on clinical samples at the protein level. Chronic treatment with CSC causes loss of typical morphological characteristics of epithelial cells. In addition, anchorage independent growth of CSC treated cells is enhanced as compared to the control. After one month of chronic treatment of BEAS-2B cells with 20 μg/ml of CSC, 650 genes were differentially expressed as compared to the control (DMSO treated cells). Eighty five genes with unknown function and 336 known downregulated genes (fold change ≥ 2) were identified. Many of those genes are related to cell-cell interactions and cell adhesion (i.e. cadherin 1 and 3, desmocollin 2, desmoglein 3, kallikrein etc). Some of identified genes were already reported as targets of methylation in lung cancer (i.e. tissue inhibitor metalloproteinase 3, TIMP3). We identified several down-regulated transcription factors (ELF3, GATA3, ATF3). Expression of proapoptotic gene BIK (NBK) was downregulated in chronically treated BEAS-2B cells (fold change 9.2) and restored after de-methylation in three lung cancer cell lines. Gene silencing and re-activation has been verified at the protein level using western blot. Promoter hypermethylation of BIK has already been described in renal cell carcinoma and malignant glioma. Relevance of BIK silencing by promoter methylation in lung cancer is poorly understood. Our preliminary immunohistochemistry results on lung adenocarcinomas indicate presence of BIK positive and BIK negative tumors in comparison to epithelial cells of normal bronchus in the same tissue section. Protein expression within BIK positive tumors is heterogenous. Absence of BIK protein in BIK negative tumors might be due to methylation of the promoter. This hypothesis will be topic of our further investigation. Setting up of LCM of human lung tumors is still in progress. Our aim is to obtain a group of common genes using three different DNA microarray based approaches mentioned above. The role of these genes in lung tumorigenesis will be further investigated.