In several solid tumors, but especially in breast cancers, overexpression of ERBB2 results in more aggressive growth, and enhanced metastatic potential and probability of emergence of drug resistance. An important aspect in this response is the interaction between ERBB2 and ERBB3. Simultaneously elevated levels of ERBB3 phosphorylation (and in many cases receptor levels) result in enhanced anti-apoptotic signaling since several PI3K binding sites reside in the kinase-deficient ERBB3. Unless ERBB2 inhibition is quantitative (which is currently not feasible), steady state levels of phospho-ERBB3 rapidly return to pre-inhibition levels after kinase inhibitor treatment through stabilization of the residual activated ERBB3, regardless of sustained and substantial inhibition of ERBB2. This seriously undercuts ERBB2 directed therapy, but it also reflects on intrinsic aspects of ERBB2-ERBB3 signaling. ERBB2 has apparently evolved to avoid stable self-association and is constitutively activated in very transient interactions. Our preliminary data suggest that ERBB2 has built-in control mechanisms to further suppress """"""""accidental activation"""""""" in such transient complexes. ERBB2 overexpression results in high levels of activated ERBB2 but at a state of spatial separation and low fraction of activation. We propose that this reflects efforts by the cancer cell to ensure a low spatial density of activated receptors, because the combination of spatial density and exclusion of ERBB3 in stable ERBB2 dimers would favor a cytotoxic signaling outcome. We have investigated this by using chimeric receptor constructs with altered intrinsic control and association behavior as well as ligand variants that differ in their degree of disruption of ERBB3 clusters. In contrast to ERBB2, self-association and orderly dissociation of clusters during activation appears to be required for ERBB3. Perturbing this system causes cell death in a manner that appears to be related to changes in cell adhesion characteristics, possibly because the cells progress towards a more pro-metastatic path without sufficient signaling to """"""""brace"""""""" against the onset of anoikis like apoptosis. The avoidance of ERBB2 clustering and maintenance of ERBB3 clusters are important in getting a balanced signaling response and are very accessible targets for deliberate system perturbation. We will evaluate the nature of cell death in different cellular contexts, adopt existing and novel methodology for direct assessment of receptor clustering and spatial proximity of activated receptors to evaluate their contribution to signaling outcomes, and determine the nature of signaling that emanates from out-of-context and cell-death promoting activation. It is our long range goal to capitalize on this understanding to enforce out-of-context activation of ERBB2 by engineered therapeutics (antibodies or aptamers). The studies proposed here are also likely to shed light on the mechanisms of currently used therapeutics (such as Herceptin) which may, albeit without mechanistic foundation, already utilize part of this response. We also expect to better understand why previous attempts to actively enforce ERBB2 mediated apoptosis have not been more successful.
Showing the most recent 10 out of 17 publications
