RESEARCH PROJECTS

	The general scope of our research is to expose mechanisms that cell- or tissue-intrinsically restrain the ability of human oncogenes to promote unscheduled cell proliferation. Our focus is in cancer biology, cell cycle, apoptosis and epithelial cell biology and we use in our research various functional genomics technologies, 2D and 3D cell-based assays and genetically engineered mouse models.

Our long standing interest in the apoptotic function of oncogenes stems from PI's grad student years in the laboratory of prof. Kari Alitalo, when we showed that activation of a single oncogene, c-Myc, can sensitize non-transformed cell lines and primary cells to the apoptotic action of death receptor ligand TNF (Klefström et al, EMBOJ 1994 and EMBOJ 1997). This finding was an important milestone in formation of the concept that mitogenic oncogenes do not only do “good” for tumor cells by promoting proliferation but also have “bad” effects (for tumor cells) by sensitizing tumor cells to apoptosis. We later showed, during my post doctoral period in the laboratory of Gerard Evan, that this phenomenon is not only restricted to c-Myc since also viral oncogenes, such as v-cyclin sensitizes cells to ligation of TNF receptor and related TRAIL and CD95 receptors (collectively death receptors) (Verschuren, Klefstrom et al Cancer Cell 2002). We believe that understanding of such intrinsic tumor suppression mechanisms will provide new possibilities to develop therapeutic modalities that are not toxic for normal cells but specifically target oncogene expressing tumor cells to apoptosis.

Cross-talk of mitochondrial and death receptor apoptosis pathways

To model therapeutic exploitation of TRAIL pathway, we first identified that the synergistic apoptosis induction by oncoproteins and death receptors involves a molecular cross-talk between the death receptor and the mitochondrial apoptosis pathways (Klefström et al J.Biol.Chem 2002). These central findings led us to search for mitochondrial c-Myc targets focusing on the Bcl-2 family. We identified the pro-apoptotic Bcl-2 family member Bak as a protein, which can be activated by c-Myc and also, that the co-activation of c-Myc-Bak axis and the TRAIL-caspase8-Bid axis together establishes a lethal caspase amplification loop that kill cells. Once a mechanistic model was established to explain the lethal interaction of c-Myc and TRAIL, it prompted us to perform several therapy modeling studies with drugs or drug-like approaches. We have now shown that the sensitizing action of c-Myc can be mechanistically phenocopied by ABT-737, a small molecule inhibitor of major anti-apoptotic Bcl-2 proteins and also, that the tumor cell targeting apoptotic action of TRAIL-TRAIL receptor complex can be short-circuited by chemical dimerization of caspase-8 (Nieminen..Klefstrom, EMBOJ 2007 ). These findings identify mitochondrial Bcl-2 proteins, especially Bak, as cental regulators of TRAIL sensitivity and show that c-Myc sensizes cells to TRAIL by coupling the mitochondrial and the death receptor pathways (reviewed in Nieminen..Klefstrom, Cell Cycle 2007). The results also demonstrate the therapeutically important proof-of concept that small molecule drugs can induce TNF/TRAIL-like selective killing of oncogene-harboring cells.

Cells united against cancer

Recently, we also stumbled into very surprising finding that the supposedly most powerful mitogenic oncogene on earth, c-Myc, completely failed to re-initiate the cell cycle in three-dimensional organotypic epithelial culture model. What was wrong? I first blamed my graduate student for her heretic views on c-Myc but she defended by showing papers from Pagliarini&Xu, Science and Brumby &Richardson, EMBO J demonstrating that at least in fly, the epithelial cell structure and cell polarity are powerful restraints of the cell cycle promoting oncogenes. This was a start for very exciting journey that led us to explore the interaction of human epithelial structure and c-Myc and we learned that LKB1, the human homologue of Drosophila polarity gene Par4, has an important role in planning the architecture that can restrain the cell cycle promoting action of c-Myc (Partanen..Klefstrom, PNAS 2007). This PNAS paper was reviewed in Nature Reviews Cancer and featured as the editor's choice in Science.

Research Highlight. Nature Reviews Cancer 7, 2007
Oncogenes: Suppressive organization

Science STKE, 18 September 2007
EDITORS' CHOICE, Cancer Biology, Preventing Transformation

Currently we are exploring in many fronts the concept that our model suggests that organized structure of breast epithelial cells serves as a natural barrier against cancer. For these studies we have generated unique in vitro tools and the figure below represents our experimental approaches based on 3D organotypic mammary cell cultures. In our 3D culture model, the oncogene c-Myc can be switched on and off at any time during the morphogenesis of mammary epithelial structures. These epithelial structures arise from single cells in specific matrix and when fully developed, they are morphologically and functionally similar to mammary acini. The mammary acini are hollow spheres of milk producing cells in the breast and also units from which breast cancer often arises. However, unlike the physiological acini, the 3D cultured acini are amenable for large-scale genetic experiments. We use recombinant lentiviruses to deliver gene activating or silencing (shRNA) DNA constructs in the 3D acinar structures and we have discovered that while c-Myc induces overproliferation and transformation (abnormally large size and distorted, non-polarized morphology) in developing acinar structures, the fully formed epithelial organization completely suppresses the cancer promoting effects of c-Myc. This organizational suppression effect required presence of LKB1. The development of organization also suppresses c-Myc-dependent apoptosis yet co-activation of c-Myc and TRAIL pathways is sufficient to kill acinar cells by apoptosis.

This research line is continuing with several different approaches including genetic shRNA screens and in vivo mouse models.