Network biology identifies novel apoptosis-related proteins and synergistic drug combinations in breast cancer

[eng] Breast cancer is a very heterogeneous disease with a poor prognostic outcome, largely due to its resistance to current cancer therapies. The balance between cell proliferation and apoptosis plays a critical role in determining the overall growth or regression of the tumor in response to treatments. Hence, identifying treatments involved in apoptosis resistance is essential in order to find new therapeutic approaches. The heterogeneity of cancer is rarely due to abnormalities in single genes, but rather reflects the discontinuation of complex intercellular processes. Therefore, a useful way to describe and analyse this heterogeneity is the use of systems biology. This approach is based on the study of the interactions between the elements of a given system with the aim to understand its properties. Particularly, the use of protein-protein interaction networks gives a broader perspective of protein environment without losing the molecular details, providing a deeper understanding of the molecular mechanisms underlying complex pathological processes. Thus, the application of network biology can significantly help the elucidation of novel drug targets and more effective therapies. Since the molecular mechanisms that relate, for instance, changes in expression of the many passenger genes to breast cancer onset and progression are largely unknown, the first objective of this thesis was to identify the role of certain proteins in breast cancer apoptosis, which function on the disease is unknown. With this aim, we functionally characterized two breast cancer passenger genes, PSMC3IP and EPSTI1, to analyse their potential apoptotic role in breast cancer. We first explored the existence of direct physical interactions with annotated breast cancer-apoptotic genes, and subsequently, we examined several apoptotic markers to determine the effect of PSMC3IP and EPSTI1 gene expression modulation in two different human breast cancer cell lines. Overall, our results showed that PSMC3IP and EPSTI1 are able to modulate the extrinsic apoptotic pathway in estrogen receptor positive and triple negative breast cancer cell lines, highlighting them as potential therapeutic targets. Currently approved drug combinations are the result of empirical clinical experience, are not being rationally designed and cover only a small fraction of the potential therapeutic space. Therfore, our second objective was to implement two different network biology approaches in order to predict and validate novel synergistic drug combinations for the treatment of breast cancer, with greater clinical efficacy and reduced side effects. The first approach was based on the quantification of pathway crosstalk inhibition in therapeutic networks. We applied this measure to a set of antineoplastic (cancer drug combined with breast cancer drug) and compassive (non-cancer drug combined with breast cancer drug) combinations. We next validate experimentally the most promising drug combinations in several breast cancer cell lines. Finally, we determined whether the cytotoxic effect of the selected drug combinations was due to apoptosis induction or cell cycle arrest promotion. Overall, our findings highlight two different antineoplastic combinations as promising therapeutic strategies for triple negative breast cancer subtype. The second approach integrated the known human interactome with the basal gene expression measurements in hundreds of cancer cell lines, whose sensitivity to several drugs was previously determined. Subsequently, a weighted score in the protein network was defined in order to predict synergistic drug combinations for breast cancer treatment. Finally, we analysed the inhibitory effect of the most promising combinations in different breast cancer cell lines harbouring molecular alterations in PI3K/AKT pathway. Interestingly, our results revealed significant synergism of the drug combinations when compared to single-compound assays. Thus, suggesting this computational approach as a useful tool for the identification of effective and synergistic combinations for breast cancer treatment