Investigation of the molecular mechanisms underlying the invasive phenotype in a panel of lung cancer cell lines

The research described in this thesis aimed to increase our knowledge of the molecular mechanisms by which lung cancer cells acquire the capacity to invade. The research can be divided into two main sections a) Development of genetic tools for stable, targetable transgene overexpression, b) Investigation of the role of differential expression of a group of genes, including microRNAs, in cancer invasion. DLKP mildly invasive cell line derived from a non-small cell lung carcinoma and its Adriamycin resistant variant DLKPA highly invasive were used in this study to evaluate the roles of differentially expressed mRNAs and microRNAs. We developed two targetable DLKP cell lines based on Cre-LoxP technology. These were generated by random and homologous recombination (site specific) targeting. These cell lines were developed to overcome challenges one faces with non-targetable systems including large discrepancies in target gene expression, genomic instability, and unpredictable phenotypes. A panel of thirty one DLKP single cell clones were assessed for stable and transgene overexpression, over four and a half months with regular passages and freeze thaw cycles. Many clones were found to display unstable transgene expression over time but several were identified with stable expression and at different levels. Two clones were selected based on stable expression level and on invasive phenotype, i) high expressing non invasive clone DLKP 17 and ii) low expressing invasive clone DLKP 11. Microarray profiling studies done in this lab on a panel of lung cancer cell lines with various invasive phenotypes identified a list of differentially expressed genes. GLP1R, KCNJ8 and TFPI2 genes were stably overexpressed and all induced invasion in non-invasive DLKP 17. Furthermore GLP1-R induced invasion could be reversed through siRNA induced silencing of the transgene. GLP1-R overexpression in non-invasive MCF-7 also induced cellular invasion. These findings are the first time that GLP1-R has been linked to this phenotype. MiRNA expression profiling was performed comparing low invasive DLKP (parent) and invasive DLKPA (Adriamycin selected) cell lines. Three differentially regulated miRNAs were selected for functional validation. Mir-21 and mir-27a were pro-invasive and mir-29a anti-invasive in DLKP and DLKPA cell lines. We later confirmed the impact on invasion and proliferation through stable overexpression studies with mir-29a. Pre-mir-29a overexpression in the invasive PANC-1 (pancreatic cancer line) resulted in reduced invasion suggesting that the effect of mir-29a was not cell line specific. 2D-DIGE proteomic profiling of cells transfected with mir-29a generated a list of differentially regulated protein spots, some of which were identified by MALDI-TOF and LC-MS analysis. Bioinformatic analysis revealed that the majority of these proteins were involved in cellular processes like apoptosis, proliferation, motility and differentiation. We chose GRB2, RAN, MIF and ANXA2 gene targets which were differentially downregulated due to mir-29a overexpression and performed siRNA induced knockdown in DLKPA cell line. Results revealed RAN to be anti-invasive and anti-proliferative in our cell line model. Knockdown of the other three targets did not affect the invasive or proliferative phenotype of DLKPA. GFP-RAN 3’-UTR reporter assay indicated a small (5-6%) but statistically significant reduction in GFP RAN-UTR expressing cell population when transfected with mir-29a. In conclusion we provide evidence that mir-29a may be an anti-invasive microRNA and that this effect is mediated via modulation of the expression of RAN and several other cellular proteins