Section: New Results

Resistance to anti-cancer drugs by non-mutational mechanisms: insights from a cell-based multi-scale model of TRAIL-induced apoptosis

Participants : Virgile Andréani, Grégory Batt, François Bertaux.

The fact that tumors can acquire drug resistance by non-mutational mechanisms is increasingly gaining attention (Sharma et al., 2010; Pisco et al., 2013; Flusberg et al., 2013). Stochastic fluctuations in cellular states of different resistance and proliferative potential could play an important role in such resistance acquisition. Thus, to enable a quantitative, molecular-level understanding of those phenomena, modeling approaches that go beyond traditional, deterministic kinetic models of biological pathways are required.

An interesting and well-studied example of non-mutational resistance acquisition concerns the response of cancer cells to the agent TRAIL, a selective inducer of apoptotic cell death. In a previous work (Bertaux et al., 2014), we have developped a single-cell model of TRAIL-induced apoptosis that accounts for (1) protein-protein signaling reactions linking TRAIL exposure to commitment to apoptosis, (2) stochastic gene expression for the proteins involved in this signaling and (3) protein degradation. Under parsimonious and realistic assumptions for parameter values, fractional killing and transient resistance acquisition readily emerged from model simulations. Those two properties relating to TRAIL resistance are observed in-vitro for many different cancer cell lines.

Here, again in collaboration with Dirk Drasdo and Szymon Stoma, we investigate the long-term response of proliferating cancer cell populations repeatedly treated by TRAIL by integrating our single-cell model of TRAIL-induced apoptosis into a multi-cellular simulation framework. We predict that the long-term killing efficiency of repeated treatments is strongly reduced compared to the first treatment. A detailed analysis showed that resistance acquisition is caused mainly by the targeted degradation of activated pro-apoptotic proteins and an imbalance between the turnover of pro- and anti- apoptotic proteins. In addition, simulations of the treatment of multi-cellular spheroids suggested that limited TRAIL penetration is unlikely to be a driving cause of resistance, but that it can exacerbate the impact of cell-intrinsic resistance acquisition.