""""""""The ability to redirect cellular processes from proliferation to terminal differentiation offers a promising approach toward preventing cancer progression, and understanding the molecular mechanisms involved in this switch can potentially uncover important targets for early detection of cancer. We have been studying differentiation in non-small cell lung cancer cell lines by three different approaches: 1) Develop model systems of differentiation to enable the study of molecular mechanisms of differentiation control in lung carcinogenesis. We have developed a new model system of epithelial differentiation associated with terminal or reversible growth arrest using the NSCLC cell line H358 treated with the tumor promoter mezerein with or without the cell cycle inhibitor aphidicolin. Treatment of cells results in irreversible or reversible mucinous differentiation. This process is accompanied by inhibition of cdk2 kinase activity despite accumulation of cyclin E. The cell cycle inhibitor Waf1 mediates the early growth arrest. A striking shift in the make-up of the main G1 cdk complexes occurs during differentiation, with an increase in cdk5-cyclin E and cdk5-cyclin D1 complexes. Conversely, inhibition of cyclin dependent kinase activity with flavopiridol (a broad spectrum cdk inhibitor currently in Phase I clinical trials) or roscovitine (a cdk2 but not cdk4/6 inhibitor) induces differentiation. Similarly, antisense inhibition of cdk2 also induces mucinous differentiation. These data show that alterations in the levels of and association between cell cycle regulatory proteins accompanies differentiation and that targeting the cell cycle regulatory proteins can result in differentiation, thereby serving as a target for cancer therapy. Similarly, we have begun studying ligands of PPAR-gamma with respect to their ability to induce differentiation in NSCLC. PPAR-gamma, which controls adipocytic differentiation, is expressed in 10/10 NSCLC cell lines by RT-PCR and 9/10 by Western analysis. Treatment with 2 different ligands results in growth arrest and modulation of multiple differentiation markers, including increased gelsolin, p21/Waf1, and Mad along with decreased MUC1 and surfactant protein A. Clonogenic potential is greatly decreased after pre-treatment with PPAR-gamma ligands. Metalloproteinase activity is also decreased. These data suggest that ligands of PPAR-gamma are highly active in inducing differentiation in NSCLC in vitro and may be efficacious agents for therapy in vivo. 2) Study differentiation markers and genes controlling differentiation and proliferation in primary lung tumors, premalignant lesions, and cell lines to identify targets for early detection screening. We have previously shown a role for cJun, a known mediator of tumor promotion, in early carcinogenesis and a potential role as a target for early detection of lung cancer. We now examined the role of MUC1, a mucin associated protein known to be upregulated in breast cancer, in lung carcinogenesis. We showed that MUC1 is frequently expressed in primary lung cancers (50-60%) as well as in type II pneumocytes in surrounding lung obtained during surgical resection and all hyperplasias/ atypias derived from type II cells. MUC1 is down regulated in in vitro models of differentiation when cells stop proliferating and differentiate away from a type II pneumocyte lineage. These data do not support a role for MUC1 as a marker for early detection, but establish its use as an excellent marker of the type II pneumocyte lineage that can be used to follow the type II lineage during studies of differentiation and carcinogenesis. 3) Evaluating the function and regulation of the Clara cell protein CC10. Transgenic mouse model systems suggest that CC10 expression is lost early during lung carcinogenesis and CC10 mRNA is absent by Northern analysis in 8/10 NSCLC cell lines. We showed that overexpression of CC10 results in decreased clonogenic survival, decreased invasiveness, and decreased expression of metalloproteinases. The basis for this appears to be interference by CC10 with adherence of cells to fibronectin,thereby abrogating the growth/metastasis stimulatory pathway initiated when fibronectin binds to its integrin receptors. These data show that downregulation of CC10 contributes to neoplastic progression. These studies will shed light on the biology of lung cancer and enable us to identify targets for early detection and intervention as well as potentially develop strategies for enforcing terminal differentiation as a therapeutic goal during lung carcinogenesis.""""""""
