ABSTRACT: To study a complex network of gene expression changes underlying c-Myc-triggered carcinogenesis in lung we performed oligonucleotide array analysis of solid lung adenocarcinomas which developed in transgenic mice due to constitutive expression of c- Myc in alveolar type II epithelial cells and compared them with the non-transgenic lung as a control. The microarray analysis yielded 162 up regulated and 301 down regulated genes. The genes were further analyzed with regard to their biological functions, to known responsiveness to c-Myc and to the presence of potential c-Myc -binding sites in their promoters. We identified inappropriate expression of a variety of genes involved in the several pathways critical for the development of lung tumor: cell cycle, cell growth, cell death, adhesion, cytoskeleton organization, invasive and angiogenic capacity. Some genes identified in this study were known to be similarly changed in human malignancies and specifically in lung tumors which makes the SPC/c-Myc-transgenic mouse model useful for investigating human lung cancer. Many other genes differently expressed in lung tumors are novel targets for mechanism based therapies and serve as useful tumor markers. New c- Myc-responsive genes and putative c-Myc -targets identified in this study contributed to the collective examination of distinct c-Myc transcriptomes that are partly cell type- and species- specific The molecular rules defining c-Myc’s activity in lung cancer remain elusive. We therefore investigated the c-Myc gene networks in a transgenic mouse model of lung cancer. Genome wide gene expression profiling applied to histological well-defined tumors revealed 162 up and 301 repressed genes and identified novel c-Myc targeted cell cycle and apoptosis genes. To better understand transcriptional responses the co-occupancy of different transcription factors (TF) were analyzed. This defined composite modules with 64, 68, 96 and 92, 88 and 96% of tumor associated down-regulated genes having TF-binding sites for either c-Myc, Klf7, Gata3 or c-Myc, Sox18 and P53, respectively. The composite modules fitted 48% and 80% of regulated genes. Likewise, 75, 96, 89 and 44, 59 and 54% of up-regulated genes in tumor or non-tumor transgenic tissue were enriched for Myc, Elf5,Cebp and c-Myc, Hbp1 and Hif1 TF-binding sites, respectively to suggest diverse and distinct TF-clusters at different stages of disease. Importantly, the genes coding for TFs were regulated as well and electrophoretic mobility shift assay (EMSA) confirmed c-Myc DNA-binding activity at targeted promoters of regulated genes. Furthermore, computational biology defined Gata3, mortalin and moesin to function in master regulatory gene networks and experimental validation of the constructed networks was achieved by EMSA, RT-PCR, Western blotting and gene reporter assay. Lastly, evidence for c-Myc DNA binding activity at targeted gene promoters was also derived from published Chip-sequence data of 7 human cell lines and the clinical significance of findings can be demonstrated by their regulation in human lung cancer. In conclusion, novel c-Myc targeted cell cycle and apoptosis genes were identified to function in gene regulatory networks. The results of the present study improve an understanding of c-Myc transforming capacity and provide a molecular rationale for the development of molecularly targeted therapies.
