Lung cancer is the most common human neoplasm in the U.S. and, increasingly, in much of the world. While smoking is known to be the major etiological factor, the causative cellular and molecular mechanisms are complex and not well understood. Currently, we are focusing on several aspects of causation, behavior and possible new treatment responsiveness of adenocarcinoma, the most common form of lung cancer. (1) The roles of the mutant and wildtype K-ras proteins. The oncogene K-ras is often mutated in adenocarcinoma of the lung (as well as other common carcinomas), but the wild-type form is tumor suppressive. An important question, then, is: why is mutant K-ras actively oncogenic? Answers to this question could aid in prevention of up to 50 percent of human lung adenocarcinomas, and an even higher percentage of cancers of the colon and pancreas. We have found that transfection of mutant K-ras in lung epithelial cells causes increases in reactive oxygen species and DNA damage. In a panel of cell lines derived from clinical human lung adenocarcinomas, DNA damage was increased relative to a nontransformed line, as measured by the comet assay for DNA strand breaks. The degree of this damage was significantly correlated with intracellular levels of superoxide, a reactive form of oxygen. K-ras mRNA is spliced into alternative forms, 4A and 4B, giving proteins which differ in C-terminal modification and thus in intracellular localization. The 4B form is the most highly expressed in adult tissues;however 4A may be upregulated in cancers. We find that in a panel of human lung cancer cell lines with mutant K-ras, amounts of the 4A protein correlate strongly with superoxide and with DNA damage. Current evidence indicates that superoxide regulates K-ras4A expression. Since tobacco smoke contains and generations reactive oxygen species including superoxide, this could be a contributing mechanism in tobacco-related lung carcinogenesis.K-ras 4B, protein, on the other hand,is unique among the ras proteins in binding specifically with calmodulin, a calcium signaling protein known to be involved in regulation of differentiation in lung. We have demonstrated that the hypervariable C-terminal region of K-ras 4B is responsible for its binding to calmodulin. Thus, we hypothesize that K-ras 4B may provide tumor suppressive functions, while mutant K-ras 4A contributes to tumorigenesis and may be upregulated by reactive oxygen species. (2) ERBB3 and AKT2 as targets for molecular therapy of lung cancer, based on siRNA treatment. The majority of human and mouse lung adenocarcinoma cell lines, but not nontransformed cells, express the ErbB3 receptor, which belongs to the epidermal growth factor receptor family. We have demonstrated that ErbB3 in the lung cancer cells signals through phosphatidylinositol 3-kinase, AKT, GSK3beta, and cyclin D1 to stimulate the cell cycle and also cell invasiveness and migration. These behaviors in cell culture can be blocked with siRNA to ErbB3 or to the several Akt isoforms. Thus, siRNA treatment may be an approach to therapy. To confirm this, it is necessary to test the siRNA on tumors transplanted in vivo into mice. ERBB3 or AKT2 siRNA markedly suppressed the growth of human lung adenocarcinoma xenografts, 60% to 80%, in four separate trials, with residual effects several weeks after the end of treatment. We have confirmed that non-specific immune system effects were not involved, and that the expected changes in the ERBB3 and AKT2 mRNAs and proteins did in fact occur. Notably, the siRNAs were administered as saline solutions. Previously it had been thought that siRNAs would be too unstable by this route to be effective. Elimination of the need for complex carriers is very hopeful, since these carriers can introduce many complications clinically. After some additional confirmatory work, we expect to introduce this siRNA approach into the pipeline for possible clinical application. (3) Ribosomal RNA as a potential therapeutic target in lung cancer. Ribosomal RNA (rRNA) is a limiting component of ribosomes and is highly regulated in its expression from its multiple gene copies. Ribosomes are necessary for cancer growth, and over-production of ribosomes may even drive cancer development. Using human lung adenocarcinoma cells, we have discovered that a noncoding (nc) RNA is transcribed from the rRNA gene complex, starting in the intergenic spacer region. This ncRNA had previously been described only in mouse fibroblasts. Thus we are the first to report it in human cells, in epithelial cells, and in cancer cells. Furthermore, in a panel of thirteen human lung epithelial cancer cell lines, the ncRNA amount (as determined by real-time PCR), correlated negatively with amount of total rRNA present in eleven of the lines. Thus it appears that the ncRNA may commonly have some sort of negative regulatory role controlling production of the rRNA and hence ribosomes. Furthermore, a single nucleotide polymorphism was present in the 5'leader region of the rRNA, with high frequency in several cell lines. This 5'leader region is key in initiating the processing of pre(45S)rRNA in the nucleus. Strikingly, the frequency of this gene polymorphism correlated negatively with amounts of ncRNA and positively with total rRNA. Although many mechanistic questions still need to be answered, it is possible that this polymorphism affects human risk of lung cancer. Efforts are underway to characterize the nc-rRNA, the nature of its molecular effects, and the possibility of using it as a target for lung cancer therapy. In both human and mouse lung cancer cells, molecular reagents that target specific regions of the nc-rRNA cause massive cell death, and the effect is much great in cancer than in non-transformed lung epithelial cells. (4) A nitric oxide-based pro-drug effects on lung adenocarcinoma. Lung cancers are usually not treated successfully, and a challenge in the development of new drugs is the fact that these cancers are heterogeneous at the cellular level. Targeting of lung cancers based on their specific characteristics has become a major goal. In collaboration with the Chemistry Section, LCC, we have tested a nitric oxide-releasing diazeniumdiolate prodrug, JS-K, against human lung adenocarcinoma cells in culture. The drug is rapidly effective in the sub-micromolar concentration range, for a subset of the cell lines. It also kills these cells as xenograft tumors in vivo. Three specific features characterize the sensitive cell lines: high endogenous reactive oxygen species;low levels of an anti-oxidant enzyme, PRX1;and low levels of an enzyme, OGG1, that repairs oxidative DNA damage. The PRX1 and OGG1 levels could be readily quantified in primary surgical specimens and so used to identify patients that might be responsive to treatment with this drug.
| Abraham, Sherwin J; Nolet, Ryan P; Calvert, Richard J et al. (2009) The hypervariable region of K-Ras4B is responsible for its specific interactions with calmodulin. Biochemistry 48:7575-83 |
| Shiao, Yih-Horng; Lupascu, Sorin T; Gu, Yuhan D et al. (2009) An intergenic non-coding rRNA correlated with expression of the rRNA and frequency of an rRNA single nucleotide polymorphism in lung cancer cells. PLoS One 4:e7505 |
