We hypothesize that related biological processes are involved not only in normal morphogenesis and tissue repair, but also in pathological processes such as tumor cell invasion and bi-directional host-pathogen interactions. Previous studies by numerous laboratories including our own have identified overlapping pathological mechanisms that include changes in cell-matrix and cell-cell adhesion, migration, and associated signal transduction pathways. Understanding these processes should help to clarify mechanisms of pathogenesis and to identify novel potential targets for therapeutic intervention. In cancer metastasis, normal cell behavior is subverted by processes that modify the formation of cell adhesions, migration, and invasion. We are investigating mechanisms of invasion in hopes of identifying targets for potential therapeutic intervention in the pathogenic process, e.g., key steps and target molecules such as specific proteases or components of integrin adhesion and signaling complexes. Previous studies by numerous laboratories including our own have identified mechanisms involved in invasion that include choreographed changes in cell-matrix and cell-cell adhesion, migration, and associated signal transduction pathways. Understanding these processes may ultimately lead to novel therapeutic approaches. The coordinated roles of integrins, proteases, and signaling in these processes are not currently sufficiently clear. We are focusing on the following facets of this problem relevant to both developmental and cancer cell migration. 1. How are integrins involved in tumor cell invasion and metastasis? 2. How do invadopodia - tiny cell surface structures mediating proteolysis - initiate and function? 3. What are the patterns of tumor cell proteolytic degradation of the matrix during cell motility, and how is proteolysis used for migration? 4. What aspects of host-pathogen interaction involve related processes? A long-term collaboration with colleagues at Oncoimmunin Inc. has culminated in a highly sensitive method for detecting local proteolysis in 3D systems. The peptide sequence GPLGIAG (a substrate for multiple proteases and especially matrix metalloproteases) was inserted into an Oncoimmunin probe that fluoresces only after the peptide insert is cleaved by a protease. This proteolysis is suppressed by TIMP-2 but not TIMP-1, and we showed that this protease activity was essential for tumor cell invasion of native collagen gels. Such proteolysis is not needed for migration in a non-crosslinked collagen gel. These findings resolved a conflict in the literature by showing that tumor cell invasion through a natural endogenously crosslinked collagen matrix requires protease function, but collagen without crosslinks permits cell penetration without proteases. A second issue in the literature has concerned the site of proteolysis as cells invade, with a recent paper reporting that proteolysis occurs at the main cell body rather than at the leading edge. Using our new probe and live-cell imaging, we showed unequivocally that proteolysis is strongest by far at the leading edge. We also reported finding accumulations of collagen fibers around the cell body that can be misinterpreted as proteolysis if one uses a """"""""leaky"""""""" probe for protease detection such as DQ-collagen, which shows significant background fluorescence even without proteolytic cleavage. Nevertheless, DQ-collagen can be used successfully for determining overall levels of proteolysis by cancer cells. A collaborative study with Sameni et al. using functional live-cell imaging demonstrated that downregulating beta1 integrins suppresses the proteolysis of collagen IV by breast and prostate carcinoma cells. We previously published a description of the essential steps in formation and function of tumor cell invadopodia. The structural actin cores of invadopodia are formed first and then the protease MT1-MMP accumulates to mediate ECM degradation. Recent analyses have established roles for the exocyst and integrins in invadopodia formation by acting at specific steps in the formation of invadopodia and their subsequent function. These findings underscore the complex stepwise process of tumor invadopodia formation and function. We are developing new tools to address these questions. One approach has been to apply TIRF (total internal reflection fluorescence) microscopy to study the initial steps of formation of invadopodia by the use of fluorescent protein chimeras and immunolocalization. A second approach has been an ongoing effort to develop optimal matrices for high-resolution microscopy. Although three-dimensional matrices can be used readily to study invasion by phase-contrast microscopy, they have proven difficult to use for high-resolution immunolocalization of the many proteins involved in the formation of cell adhesion complexes, migration, and invasion. We are experimenting with various thin-film approaches to permit localization of multiple proteins at the same time in cells in a physiological matrix environment.
Showing the most recent 10 out of 16 publications
