Impact of different K-ras mutations on tumor progression and drug sensitivity in non-small-cell lung cancer

Worldwide, lung cancer is one of the most common cause of cancer incidence and mortality. Even though there was some improvement in survival during the past decades, only 20% of patients present potentially resectable early-stage disease and the 5-year survival rate remains poor at approximately 16%. Mutations in the K-ras gene occur frequently in NSCLC and they mostly involve codons 12 and 13, causing aminoacid substitutions that present different frequencies compared to other tumor types, in particular colon cancer, in which K-ras mutations play a key role, especially as regards the outcome of drug therapies. Considering these characteristics, the laboratory wants to investigate whether different aminoacids in the same position of K-ras proteins could determine a different regulation of the genes targeted by Ras-Raf-MAPK pathway and a different impact on drug resistance in lung cancer. Clones overexpressing K-ras bearing the three most common mutations found in NSCLC patients (G12C, G12D and G12V) were generated from the human NSCLC cell line NCI-H1299. These clones were tested for their sensitivity toward a panel of agents commonly used in the treatment of NSCLC patients and the most interesting finding was represented by the lower cisplatin sensitivity of the G12C clone. So, the aim of this project consists in better understanding the molecular basis of this resistance, given that preliminary results obtained to date did not highlight any important differences in the resistance mechanisms among clones. We decided to investigate the levels of cisplatin-DNA adducts after cisplatin treatment and we highlighted that, after 24 hours from treatment start, Pt-DNA adduct levels were significantly lower in the G12C cisplatin-resistant clone compared to the others. Afterwards, we performed ɣH2AX immunofluorescence in order to investigate whether the extent of DNA damage inflicted by a fixed cisplatin concentration was comparable among K-ras mutant clones. ɣH2AX was observed in the WT, G12D, and G12V clones after 16, 24, and 48 hours from treatment, while in the G12C resistant clone ɣH2AX was almost undetectable at every time point of the experiment. Altogether, these new findings suggested that cisplatin damage was quickly resolved before DSBs formation in the resistant clone. Therefore, we decided to investigate the possible role of the nucleotide excision repair system in cisplatin resistance, but no differences were observed. Then, we hypothesized that the K-ras G12C mutation could induce the removal of platinum mono-adducts from DNA before crosslink formation, and, when we decided to evaluate the cell viability of all clones following combined treatment with cisplatin and methoxyamine, a known base excision repair inhibitor, we observed that the cell viability in the G12C clone was similar to that observed in other clones. Therefore, we hypothesized that the presence of the G12C mutation might alter BER function. Recent studies identified overexpression of BER genes possible mechanism of cisplatin resistance in tumors. We found that the DNA polymerase β gene was present in the G12C clone at higher levels compared to other clones. Unfortunately, when we tried to confirm the role played by Polβ in the response to cisplatin by analyzing the cell viability of all clones following transfection with a siRNA that blocks Polβ expression and a concomitant treatment with cisplatin, we did not see any significant differences in cell viability. However, western blot analysis of Polβ protein levels showed that protein levels were unmodified 48 hours after transfection, in spite of substantial downregulation of Polβ mRNA. This observation suggests that the protein has a long half-life and that blocking its mRNA for 48 hours is not enough to achieve significant protein depletion. In conclusion, we currently hypothesize that the K-ras G12C mutation could induce the removal of platinum mono-adducts from DNA before crosslink formation and we think that the molecular basis of this phenomenon are founded in a different functionality of the BER, even if we are still unable to define which kind of alterations are present. In the meantime, in order to confirm and extend the data generated in NCI-H1299 to another cellular model, we selected the NCI-H1437 cell line to generate three other different clones that present the three most common mutations of K-ras. What is important is that these new clones will have basal levels of K-ras, unlike the clones generated with NCI-H1299 that show overexpressed K-ras. Induction of site-specific mutation in this cell line could be obtained thanks to Zinc Finger Nucleases (ZFNs) technology, based on the creation of a donor plasmid with the desired mutation. This technology allows the insertion of the desired mutation directly into the K-ras locus, leading to the generation of mutated cells with characteristics more closely resembling to those present in human tumors.