Design, Synthesis And Antitumor Activity Of Geldanamycin Derivatives

Geldanamycin (GA), a benzoquinone ansamycin antibiotic, was reported as the first heat shock protein90(Hsp90) inhibitor. GA binds to the N-terminal domain ATP binding site of Hsp90, inhibiting the chaperone activity of the protein and resulting in disruption of the Hsp90-client protein complexes. This process subsequently leads to degradation of the unchaperoned client proteins by the proteasome, leading to inhibition of cell proliferation and induction of apoptosis.The antitumor potential of GA has been recognized for a long time, clinical evaluation of GA has not proceeded due to its hepatotoxicity and poor water solubility. The structural modifications of GA have appeared a boom during the last decade. To modify17-position of GA is a well-known strategy, which yielded some less hepatotoxic derivatives, e.g.17-allylamino-17-demethoxygeldanamycin (tanespimycin,17-AAG) and its water-soluble analog17-[2-(dimethylamino)ethyl]amino-17-demethoxygeldanamycin (alvespimycin,17-DMAG). Although17-AAG showed some promise in Phase III trials for the treatment of breast cancer, the clinic trial has been terminated, indicating that novel generation of GA derivatives are urgently in demand.The thesis is divided into four chapters. In the first chapter, we reviewed recent progress in the development of Hsp90in oncogenesis is illuminated. Particular emphasis is put on inhibitors such as geldanamycin and analogues that serve as anatural product show case. Hsp90has emerged as an important target in cancer therapy and/or against pathogenic cells which elicit abnormal Hsp patterns. Competition for ATP by GA and related compounds abrogate the chaperone function of Hsp90. In this context, this account pursues three topics in detail:a) Hsp90and its biochemistry, b) Hsp90and its role in oncogenesis and c) strategies to create compound libraries of structurally complex inhibitors like GA on which SAR studies and the development of drugs that are currently in different stages of clinical testing rely.In the second chapter, design, synthesis and evaluation of GA fluorescent probes was performed. Lots of studies showed that GA-induced oxidative stress directly cause cell damage and death by oxidative degradation of proteins, which is not dependent on Hsp90. GA-induced oxidative stress mainly includes two aspects:a) GA induces cells to produce reactive oxygen species (ROS). A single electron reduction of quinone group by flavin enzyme to generate semiquinone anion, then reacts with oxygen molecules to produce superoxide anion, these super oxide can be further converted into other reactive oxygen species, b) Under normal physiological conditions, GA can react with glutathione (GSH) to form a stable thiolated GA, decreasing intracellular GSH levels, destruction of intracellular redox balance. In order to verify our hypothesis, a series of double-mechanism fluorescent probes were designed and synthesized for detecting hydrogen peroxide (H2O2):1) verify whether GA induces to produce ROS;2) to explore the location of GA in cells.In the third and fourth chapters, we design, synthesis and biological evaluation of7series of GA derivatives (146compounds). In the third chapter, GA was modified by substituted arylmethylamines, phenylethylamines and phenoxyethylamine/phenylpropylamines and synthesized17-substituted arylmethylamines (A series), phenylethylamines (B series) and phenoxyethylamine/phenylpropylamines (C series)(67compounds), respectively. Biological evaluation of all the GA derivatives were performed and followed by structure-activity relationship (SAR) discussion.The design strategy of17-arylmethylamine-17-demethoxygeldanamycin derivatives is based on scaffold hopping of17-AAG allylamine side chain group, and the small π bond (double bond) is replaced by large π bond (aromatic ring) to increase hydrophobic interaction. On MDA-MB-231cell line, most of the derivatives showed good cytotoxicities, which are comparable to that of17-AAG but less potent than that of GA. Interestingly, among the various substituted17-benzylamine derivatives of GA, varying substitutions at different positions of the benzyl group remarkably influenced their cytotoxicities against LNCaP cell line. Among them,17-(4-aminobenzylamine)-17-demethoxygeldanamycin (A21) is potent against the two cell lines. Then, we used heterocyclic rings to replace substituted benzyl groups to obtain compounds A30-33. Six-membered heterocyclic derivatives (A32-33) showed excellent antitumor activity against both cell lines. Especially, compound A33,17-((6-(trifluoromethyl)pyridin-3- yl)methylamino)-17-demethoxygeldanamycin was potent against both cell lines, which was comparable to that of17-AAG. Accordingly, compounds A21, A32-33were selected to undertake a preliminary study on their hepatotoxicities in mice. The AST and ALT levels of A33-treated mice showed no significant difference (P>0.05) with the vehicle group, suggesting that A33was low hepatotoxic. Consequently, the anti-tumor mechanism of action and drugability of A33is worthy of further study.Inspired by the above breakthroughs, twenty-six17-phenylethylaminegeldanamycins (B series) were synthesized based on bioisosterism of17-DMAG and17-AAG. However, in vitro antitumor activity of majority of B series compounds were weaker than that of A series. Among them, the antitumor activity of17-(3-trifluoromethylphenylamino)-17-demethoxygeldanamycin (B25) against LNCap cells (IC50=0.27±0.11μM) was more active than GA (IC50=0.43±0.15μM). The AST and ALT levels of B25-treated mice showed no significant difference (P>0.05) with the vehicle group, suggesting that B25was low hepatotoxic. Considering the in vitro antitumor activity and in vivo liver toxicity results, compound B25is a potential lead compound for the treatment of prostate cancer.On the other hand, we also synthesized C series compounds:17-phenylpropylamine/phenoxyethylamine-17-demethoxygeldanamycins to preliminary study of C-17substituent length of the side chain between benzene ring and GA affecting antitumor activity and toxicity of GA derivatives. In vitro antitumor activity results showed that overall activity of C series compounds has increased compared with the B series.In the fourth chapter, we decided to undertake a study of17-diamine-linked17-aroylamido-17-demethoxygeldanamycins (D, E, F and G series), because modification of17-position of GA can not only maintain the excellent antitumor activity, but also endow compounds with reduced hepatotoxicity. Moreover, in order to find a linker with proper length, four types of linkers (4-aminomethylpiperidine,1,4-butanediamine,1,6-hexanediamine, and1,8-diamino-3,6-dioxaoctane) were investigated. In the designed compounds, varieties of aroyl groups were introduced to investigate the SAR.Initially, we introduced three linkers4-aminomethylpiperidine,1,4-butanediamine,1,6-hexanediamine to synthesize the compound Dl-7, E1-7and Fl-7. Biological sceening results showed that1,6-hexanediamine linker is the best. Then1,6-hexanediamine linker was fixed, and compounds F8-39were synthesized to investigate the substitutions of the aroyl groups. All the derivatives were evaluated for their antitumor activities against MDA-MB-231cells. Among these compounds,17-(6-(3,4,5-trimethoxycinnamamido)hexylamino)-17-demethoxygeldanamycin (F36) showed the most potent cytotoxicity against MDA-MB-231(IC50=0.19±0.02μM) with the lowest hepatotoxicity (AST=181.0±23.6U/L, ALT=40.4±11.8U/L). Compared to17-AAG, F36exhibited lower hepatotoxicity in mice, higher Hsp90inhibitory activity in vitro and antitumor activity in human breast carcinoma (MDA-MB-231) xenograft nude mice.In order to fill the gap,1,8-diamino-3,6-dioxaoctane linker was introduced to the17-position of GA and aroyl groups were selectively used the substituted cinnamoyl groups (G series). Twenty-six GA derivatives were synthesized and evaluated for their cytotoxicities against MDA-MB-231cells. However, compared with F series, the antitumor activity of G series compounds were all significantly decreased.In summary, in our modification of17-position of GA,1,6-hexanediamine linker is the best and the aroyl group3,4,5-trimethoxycinnamyl group is the best. F36exhibited lower hepatotoxicity and higher antitumor activity than17-AAG in mice. This research would provide impotant reference value for the development of novel antitumor agents.