Antiproliferative and apoptotic activities of tosylcyclonovobiocic acids as potent heat shock protein 90 inhibitors in human cancer cells

Abstract

We evaluated whether inhibition of heat shock protein 90 (hsp90) function by novobiocin derivatives could induce the degradation of signal transducers that drive cancer cell growth and thereby promote apoptosis. Removal of the noviose moiety in novobiocin and introduction of a tosyl substituent at C-4 or C-7 coumarin nucleus provided derivatives 4TCNA and 7TCNA which compared favourably with novobiocin in MCF-7 breast cancer cells. Here we extend the antiproliferative and apoptotic properties of these analogues to a panel of cancer cell lines. Destabilization of hsp90 client proteins Raf-1, HER2, and cdk4 suggests inhibition of hsp90 chaperoning function. In HT29 colon and IGROV1 ovarian cancer cells, the growth inhibiting effect of 4TCNA and 7TCNA was consistent with the stimulation of cell death as assessed by the processing and activation of caspase 9, 8, 7 and 3 and the subsequent cleavage of poly(ADP-ribose) polymerase (PARP). In Ishikawa endometrial adenocarcinoma cells, 4TCNA also promoted apoptosis and the processing of PARP. These derivatives impacting multiple pathways involved in the neoplastic process may represent promising drugs for cancer therapy.

Introduction

Heat shock protein 90 (hsp90) is a ubiquitous molecular chaperone essential for the folding, maturation, stability and function of a wide range of client proteins [1]. Many of its substrates, namely protein kinases (HER2, Akt, raf-1, cdk4) and transcription factors (steroid hormone receptors, p53, HIF-1α) are involved in cell signalling, growth and/or survival, angiogenesis, metastasis and implicated in the pathogenesis of human cancers [2], [3]. Since inhibitors exert broad-spectrum antitumour activity as they impact multiple oncogenic pathways at the same time, hsp90 has emerged as an attractive therapeutic target in diverse cancers in which signal transduction pathways are persistently activated [4], [5]. In vitro and in vivo, the chaperone activity of hsp90 depends on ATP binding and hydrolysis [6], [7]. Natural product antibiotics of the ansamycins geldanamycin (GA), and its less toxic derivatives, 17-allylamino-17-demethoxygeldanamycin (17-AAG) and 17-(dimethylaminoethylamino)-17-demethoxygeldanamycin (17-DMAG) or the macrolide radicicol classes bind the N-terminal ATP-binding pocket of hsp90 and impair its chaperone function, resulting in depletion of many client proteins [8], [9]. Although less hepatotoxic than GA, 17-AAG is poorly soluble, difficult to formulate and undergoes extensive metabolism leading to non-target-based effects [10]. Radicicol binds to hsp90 with much higher affinity than GA and 17-AAG and is more potent at inhibiting hsp90 ATPase activity in vitro [11] but it lacks antitumour activity in animals because of its rapid metabolic inactivation. Besides these natural compounds, purine-based inhibitors of hsp90 have been developed by rational design [12].

In addition to the identification of the N-terminal ATP-binding site, the specific binding of the coumarin antibiotic novobiocin to a previously unrecognized ATP-binding domain in the C-terminal region of hsp90 has been reported [13]. Novobiocin disrupts both C- and N-terminal nucleotide binding and antagonizes the binding of GA or radicicol to hsp90 [14]. Therefore the C-terminal half of the hsp90 represents a second important approach for the development of selective inhibitors. At high concentration, novobiocin is able to interfere with the chaperone function of hsp90 and to deplete tumour cells of a series of hsp90-dependent signalling proteins [15]. Among more active analogues of novobiocin, Blagg et al. highlighted the crucial role of the noviose moiety at the 7-position of the coumarin ring for the biological activity whereas the 4-hydroxy and 8-methyl groups are not indispensable [16]. In more recent studies, 3′-descarbamoyl-4-deshydroxynovobiocin proved to be a more effective and selective hsp90 inhibitor [17]. As part of our research devoted to nonsugar coumarin analogues that target hsp90, we recently showed that removal of the noviose moiety together with introduction of a tosyl substituent at C-4 or C-7 coumarins provided 4- or 7-tosylcyclonovobiocic acid (4TCNA or 7TCNA) as lead compounds [18]. Biological evaluation of these derivatives in MCF-7 breast cancer cells revealed higher potency to induce the proteasome-mediated degradation of several known hsp90 client proteins and stronger antiproliferative activity as compared to the parent compound novobiocin. In addition, 4TCNA was a potent inducer of programmed cell death. In view of these findings, the purpose of this study was to examine the potential of novobiocin, 4TCNA and 7TCNA to affect the fate of a subset of hsp90 client proteins and to inhibit proliferation or induce apoptosis in a panel of cancer cells.