1) Mutations in human lung adenocarcinomasWill Lockwood (with the earlier help of Kreshnik Zejnullahu) has been directing the sequencing of unusual lung adenocarcinomas, with emphasis on tumors arising in patients who have never smoked, in a partnership with the NHGRI Sequencing Core and the University of British Columbia, to search for novel mutations that drive tumor development. Whole-exome sequences have been integrated with pre-existing data about single nucleotide polymorphisms, copy number variation, CGH, DNA methylation, and mRNA expression. Even in a relatively small cohort of tumors, we have identified a few novel candidate driver genes, in addition to well-established ones, and these are currently being assessed in a larger cohort of 100 lung tumor/normal pairs. Functional validation of candidates, such as alternatively spliced MET genes and novel cancer genes, is being undertaken in cell lines and in mouse models of lung adenocarcinoma. 2) Role of RNA processing in lung adenocarcinogenesis.Characterization of cancer genomes by TCGA and other teams over the past few years has revealed that cancer cells often harbor mutations in splicing signals and in genes encoding known splicing factors. These mutations are likely to contribute to neoplasia by generating abnormal RNAs and proteins, although the critical features of this putative oncogenic mechanism have not yet been defined. Dennis Fei is attempting to characterize the oncogenic properties of mutant proteins that regulate RNA splicing in some human cancers. He is focusing on a mutation (S34F) in the splicing factor U2AF1 that has been repeatedly observed in lung adenocarcinoma, myeloid dysplastic syndrome, and acute myeloid leukemia. In efforts to seek oncogenic effects of the mutant gene, he has introduced it into cultured human lung cell and myeloid lines, and he has made mice in which the mutant gene can be activated by the Cre recombinase. In these biological settings, with some help from Jovian Yu, he is now looking for changes in spliced RNAs, for changes in the development and behavior of lung and hematopoietic cells, and for effects on oncogenesis promoted by other cancer genes, such as mutant EGFR or KRAS (for lung adenocarcinoma) and drivers of myeloid dysplasias and cancers. 3)Mutual exclusivity of KRAS and EGFR mutations in human lung adenocarcinomasAlthough lung adenocarcinomas frequently contain KRAS or EGFR mutations (in about 25 and 10% of tumors, respectively), tumors containing mutations in both genes have not be observed. Statistical evaluation argues that such mutual exclusivity has not occurred by chance. Not only are the two genes thought to be acting in the same signaling pathways, which may explain a lack of additive or synergistic effects, but co-existence of mutations in both genes is likely to be detrimental and thus selected against. Will Lockwood and Arun Unni are pursuing a set of experiments in mice with inducible mutant EGFR and KRAS transgenes and in cell lines in which the two mutant genes can be expressed or introduced by viral vectors to attempt to document the predicted incompatibility and to understand and perhaps exploit the predicted detrimental effects.4)Inflammation, immune response, and lung tumorigenesis The immune system can promote or retard cancer progression in different settings. Arun Unni has chosen to better understand this interplay by using our mutant EGFR-driven mouse models of lung adenocarcinoma. He has observed early recruitment of macrophages to the lung parenchyma after induction of mutant EGFR. Experiments in various cell lines suggest that signaling by mutant EGFR is capable of driving an inflammatory response by induction of expression of cytokines. By isolating clonal lines expressing different levels of EGFR, he has been able to identify a threshold for induction of a set of factors, including cytokines that mediate immune responsiveness and a readily measured cell surface protein, PD-L1, a ligand that reduces the activity of tumor-reactive T cells. Experiments are underway to determine how these signals are relayed from activated EGFR and their significance in the progression and therapy of cancers. 5)Using insertional mutagens to define molecular events that cooperate with known lung cancer driver genePang Fan (a former post-doc, still at MSKCC) and Will Lockwood have adapted the Sleeping Beauty (SB) transposition system to generate mice that express both the SB transposase and transposon, along with mutant EGFR alleles, specifically in type 2 airway epithelial cells after doxycycline treatment. Although SB transposition does not appear to accelerate tumorigenesis by mutant EGFR, we have observed some tumors in mice lacking EGFR transgenes, and we are currently examining the SB insertion sites in tumors arising in mice with and without mutant EGFR, seeking genes that initiate or potentiate oncogenic growth. In addition, Dennis Fei has recently started using the SB system to study another important type of lung cancer arising in small (neuro-endocrine) cells.6) Effects of lineage-specific genes on development and differentiation of lung cells The gene encoding the lineage-specific transcription factor, NKX2-1/TITF-1, is known to be amplified and expressed at high levels in about 15% of human lung adenocarcinomas. Jeff Chen is using viral vectors that carry this gene to study its effects on differentiation and growth of normal and tumor-derived lung cells in culture.7) Identification of targets for growth inhibitors of human lung adenocarcinoma cell lines.Several years ago, Romel Somwar, a former fellow still at MSKCC, used high throughput screening of large chemical libraries to identify small molecules that restrict the growth of human lung cancer cell lines (Somwar et al, J.Biomol Screen 14:1176, 2009). Four products of his Lung Cancer Screen (LCS 1-4) have been selected to identify the targets of these compounds. Using a combination of affinity chromatography and analysis of gene expression patterns, he identified superoxide dismutase (SOD)-1 as the likely target for LCS-1(Somwar et al, PNAS, 2011). Using similar approaches, Will Lockwood has found that this compound inhibits the growth of diverse types of tumor cell lines and has identified a potential molecular target for LCS-3 that he is attempting to confirm in collaboration with Steve Elledges laboratory at Harvard.8) Inhibitors of BET domain-containing epigenetic factors inhibit growth of human lung adenocarcinoma cell lines.Bromodomain and extra terminal domain (BET) proteins function as epigenetic signaling factors that associate with acetylated histones and facilitate transcription of target genes. Inhibitors targeting the activity of BET proteins in hematological cancers have recently been shown to have potent anti-proliferative effects through the suppression of c-MYC and downstream target genes. However, as the epigenetic landscape of cells varies drastically depending on lineage, transcriptional co-activators such as BETs would be expected to have different targets in different cancer types. To test this hypothesis, Will Lockwood and Kreshnik Zejnullahu treated a panel of lung adenocarcinoma cell lines with the BET inhibitor JQ1 and found that a subset is acutely susceptible to BET inhibition Lockwood et al PNAS 2012 through a mechanism independent of c-MYC downregulation and dependent, at least in part, on the transcription factor FOSL1 and its targets.Collaborative projects Our group is involved in a variety of collaborations, most initiated before we arrived at the NIH and conducted largely in other laboratories. Most involve studies of lung carcinogenesis in our mouse models and have produced some of the publications in the list below. Further details are available upon request.