To its ability to degrade ErbB2 with its subsequent effects on ErbB1 . However, ErbB3 appears to confer resistance to this Hsp90 inhibitor’s induced radiosensitivity and thus cell lines that overexpressed ErbB3 did not exhibit PARP increased radiosensitivity when treated with Hsp90 inhibitors . Thus, as trials move forward, it is critical that tumors that overexpress ErbB3 be targeted with an additional agent affecting ErbB3 or that tumors overexpressing ErbB3 not be included in these early “proofofconcept studies.” CTerminal Inhibitors As mentioned previously, there are four domains to Hsp90, including the Nterminus, the midregion, and the C terminus . Functional studies of the C terminus have demonstrated that this is required for dimerization of Hsp90, and when inhibited leads to inability to fold the client protein and subsequent ubiquitination of the client protein and proteosomal degradation.
Supporting studies that have used recombinant forms of HSP90, where the Cterminal portion has been deleted, fail to show dimerization and manifest reduced ATPase activity, leading to lack of formation of the HSP90client protein multichaperone complex . Novobiocin was the first known Cterminal inhibitor of the C terminus of Hsp90 but had weak affinity for Carboplatin the binding site and thus was not clinically useful. However, Yu has prepared a library of analogues of novobiocin that have shown significantly greater affinity for the C terminus of Hsp90. Through the structural modification of novobiocin, our laboratory as well as others have screened more than 300 analogues of novobiocin and identified compounds with 1,000fold greater efficacy than novobiocin.
eukaryotic The first compound tested, A4, was a coumarin analogue and was found in in vitro studies to have efficacy at 1,000 to 10,000fold lower concentrations that novobiocin . In these initial studies in several different cell lines, degradation of client proteins was demonstrated yet there was no cellular toxicity seen. Further modifications to the structure of A4 have resulted in compounds with antiproliferative activity and cytotoxic effect in the midnanomolar range . In in vitro studies of several different cell lines, these compounds exert these effects, whereas at similar concentrations are nontoxic to normal renal epithelial tubular cells. Thus, these agents appear to have selective cytotoxicity in cancer cells over normal cells.
Furthermore, these analogues exhibit improved solubility, and preliminary in vivo studies of some of the compounds suggest minimal toxicity. Perhaps one of the most exciting revelations about this group of compounds is that in in vitro studies, in addition to the degradation of the client proteins, we also observed the degradation of HSF1, Hsp27, and Hsp70 as opposed to the HSP induction seen with Nterminal inhibitors. This has potentially very important in vivo implications because the HSF1mediated induction of HSPs, in particular, has been one of the mechanisms theorized to be behind some of the disappointing results in clinical trials with the Nterminal inhibitors of Hsp90. Therefore, optimistically, these agents may exhibit improved pharmacodynamic properties, decreased toxicity, and escape the resistance mechanisms seen with the Nterminal inhibitors.