Off-target drug effects due to retroactivity in signaling pathway

MICHELLE WYNN; ALEJANDRA C VENTURA; HECTOR GARCIA; SOFIA D MERAJVER

Conferencia; 2010 AACR Annual Meeting; 2010

A majority of targeted cancer therapeutics involve the specific inhibition of a molecular target in a signal transductionpathway. It is well known that targeted therapies, such as kinase inhibitors, may have effects on pathways other than thosespecifically targeted, whether by non-specific interactions or by indirect pathway cross-talk effects. The simplest view ofsignal transduction entails a cascade of molecular events initiated by the recognition of a stimulus and culminating in thechemical alteration of an effector molecule. We have recently shown that cascades can exhibit bidirectional signalpropagation without the addition of regulatory feedback connections via a phenomenon termed retroactivity. Thus,retroactivity represents another potential source of off-target effects in signaling cascades.We have extended our previous work to a computational model which allows us to characterize the significance of theupstream off-target effect via retroactivity with a series of simple signaling networks using physiologically relevantparameters values. We aim to understand the dynamics of cancer signaling by integrating systems biology based models ofsignaling dynamics into our experimental investigations of breast cancer with the intent of both directing experiments andmaking predictions. One of the signaling networks we studied takes the form of two independent signaling cascades whichhave no regulatory feedback connections but are activated by the same upstream kinase. Our results suggest that underphysiologically relevant conditions the application of an inhibitor near the bottom of one cascade can produce a significantchange in concentration of a protein in the other cascade, even without a regulatory feedback connection between the twocascades. Developing a deeper understanding of bidirectional signal propagation in signal transduction pathways will bevitally important in the effort to develop safer and more effective targeted cancer therapies. Our modeling results challengethe notion that information in cascades only flows in the cell surface-to-nucleus direction and suggests that a perturbationapplied within a cascade can produce both an upstream and a downstream effect. The notion of an upstream off-target effectis completely novel and may have significant implications if the perturbation takes the form of an inhibitory therapeutic drug.