We reported previously that Bcl-2 is paradoxically down-regulated in paclitaxel-resistant cancer cells. We reveal here that paclitaxel directly targets Bcl-2 in the loop domain, thereby facilitating the initiation of apoptosis. Molecular modeling revealed an extraordinary similarity between the paclitaxel binding sites in Bcl-2 and β-tubulin, leading us to speculate that paclitaxel could be mimetic of an endogenous peptide ligand, which binds both proteins. We tested the hypothesis that paclitaxel mimics Nur77, which, like paclitaxel, changes the function of Bcl-2. This premise was confirmed by Nur77 interacting with both paclitaxel targets (Bcl-2 and β-tubulin) and a peptide sequence mimicking the Nur77 structural region, thus reproducing the paclitaxel-like effects of tubulin polymerization and opening the permeability transition pore channel in mitochondria. This discovery could help in the development of novel anticancer agents with nontaxane skeleton as well as in identifying the clinical subsets responsive to paclitaxel-based therapy. [Cancer Res 2009;69(17):6906–14]

Histone deacetylase inhibitors (HDACis) are one of the last frontiers in pharmaceutical research. Several classes of HDACi have been identified. Although more than 20 HDACi are under preclinical and clinical investigation as single agents and in combination therapies against different cancers, just two of them were approved by the US FDA: Zolinza(®) and Istodax(®), both licensed for the treatment of cutaneous T-cell lymphoma, the latter also of peripheral T-cell lymphoma. Since HDAC enzymes act by forming multiprotein complexes (clusters), containing cofactors, the main problem in designing new HDACi is that the inhibition activity evaluated on isolated enzyme isoforms does not match the in vivo outcomes. In the coming years, the research will be oriented toward a better understanding of the functioning of these protein complexes as well as the development of new screening assays, with the final goal to obtain new drug candidates for the treatment of cancer.

A prominent mechanism of drug resistance to taxanes is the overexpression of class III β-tubulin. The seco-taxane IDN5390 was chosen for its selective activity in paclitaxel-resistant cells with an overexpression of class III β-tubulin. Moreover, the combined treatment paclitaxel/IDN5390 yielded a strong synergism, which was also evident in cell-free tubulin polymerization assays. In the presence of an anti-class III β-tubulin as a blocking antibody, tubulin polymerization induced by paclitaxel and IDN5390 was enhanced and not affected, respectively, whereas synergism was abolished, thereby indicating that IDN5390 activity is not modulated by class III β-tubulin levels. Such properties can be explained by taking into consideration the composition of class III β-tubulin paclitaxel binding site; in fact, Ser²⁷⁷ interacting with paclitaxel C group in class I is replaced by an Arginine in class III. IDN5390 that has an open and flexible C ring and an acidic α-unsaturated enol-keton moiety better fits with class III β-tubulin than paclitaxel at the binding site. Taking altogether, these findings indicate that the concomitant treatment IDN5390/paclitaxel is able to successfully target class I and III β-tubulin and the combined use of two taxanes with diverse spectrum activity against tubulin isotypes could represent a novel approach to overcome paclitaxel resistance.