Heteroaromatic quinols 4-(benzothiazol-2-yl)-4-hydroxycyclohexa-2,5-dienone (1) and 4-(1-benzenesulfonyl-1H-indol-2-yl)-4-hydroxycyclohexa-2,5-dienone (2) exhibit potent and selective antitumor activity against colon, renal, and breast carcinoma cell lines in vitro (GI 50 < 500 nmol/L). In vivo growth inhibition of renal, colon, and breast xenografts has been observed. Profound G 2 -M cell cycle block accompanied down-regulation of cdk1 gene transcription was corroborated by decreased CDK1 protein expression following treatment of HCT 116 cells with growth inhibitory concentrations of 1 or 2. The chemical structure of the quinol pharmacophore 4-(hydroxycyclohexa-2,5-dienone) suggested that these novel agents would readily react with nucleophiles in a double Michael (B-carbon) addition. Indeed, COMPARE analysis within the National Cancer Institute database revealed a number of chemically related quinone derivatives that could potentially react with sulfur nucleophiles in a similar manner and suggested that thioredoxin/ thioredoxin reductase signal transduction could be a putative target. Molecular modeling predicted covalent irreversible binding between quinol analogues and cysteine residues 32 and 35 of thioredoxin, thereby inhibiting enzyme activity. Binding has been confirmed, via mass spectrometry, between reduced human thioredoxin and 1. Microarray analyses of untreated HCT 116 cells and those exposed to either 1 (1 Mmol/L) or 2 (500 nmol/L and 1 Mmol/L) determined that of z z10,000 cancer-related genes, expression of thioredoxin reductase was up-regulated > >3-fold. Furthermore, quinols 1 and 2 inhibited insulin reduction, catalyzed by thioredoxin/thioredoxin reductase signaling in a dose-dependent manner (IC 50 < < 6 Mmol/L). Results are consistent with a mechanism of action of novel antitumor quinols involving inhibition of the small redox protein thioredoxin. (Cancer Res 2005; 65(9): 3911-9)

Telomeric integrity is required to maintain the replicative ability of cancer cells and is a target for the G-quadruplex-stabilizing drug 3,11-difluoro-6,8,13-trimethyl-8H-quino[4,3,2-kl]acridinium methosulfate (RHPS4). We report a senescent-like growth arrest in MCF-7 breast cancer cells, within 14 to 17 days, and a reduction in telomere length (from 5.2 kilobases (kb) to 4.7 and 4.3 kb after 17 days of treatment at 0.5 and 1 M, respectively). These effects occurred at noncytotoxic drug concentrations (doses Ͻ 1 M over a 14-day exposure) compatible with long-term drug dosing. The telomere length of cancer cells influences their sensitivity to growth inhibition by RHPS4: mutant (mt) human telomerase reverse transcriptase (hTERT)-expressing MCF-7 cells [short telomere restriction fragment (TRF) length, 1.9 kb; IC 50 , 0.2 M] were 10 times more sensitive to RHPS4 compared with wild-type (wt) hTERT-expressing, vector-transfected control cells (longer TRF-length 5.2 kb; IC 50 2 M) in the 5 day SRB assay. This relationship was corroborated in a panel of 36 human tumor xenografts grown in vitro showing a positive correlation between telomere length and growth inhibitory potency of RHPS4 (15-day clonogenic assay, r ϭ 0.75). These observations are consistent with loss of the protective capping status of telomeres mediated by RHPS4 G-quadruplex-stabilization, thus leading to greater susceptibility of cells with shorter telomeres. In combination studies, paclitaxel (Taxol), doxorubicin (Adriamycin), and the experimental therapeutic agent 17-(allylamino)-17-demethoxygeldanamycin, which inhibits the 90-kDa heat shock protein, conferred enhanced sensitivity in RHPS4 treated MCF-7 cells, whereas the DNA-interactive temozolomide and cisplatin antagonized the action of RHPS4. Our results support the combined use of certain classes of cytotoxic anticancer agents with RHPS4 to enhance potential clinical benefit.

The growth-inhibitory activities of an extensive series of quaternized quino[4,3,2- kl]acridinium salts against tumor cell lines in vitro have been measured and their biological properties interpreted in the light of differential binding to different DNA isoforms. Selectivity for quadruplex DNA binding and stabilization by compounds were explored through an array of methods: UV absorption and fluorescence emission spectroscopy, surface plasmon resonance, and competition dialysis. Quadruplex DNA interaction was further characterized through FRET and DNA polymerase arrest assays. Telomerase inhibition, inferred from the TRAP assay, is attributed to quadruplex stabilization, supported by the strong correlation (R(2) = 0.81) across the series between quadruplex DNA binding affinity and TRAP inhibition potency. Growth inhibition potency in the NCI60 human tumor cell line panel is more marked in compounds with greater DNA duplex binding affinity (R(2) = 0.82). Quantification of relative quadruplex and duplex binding affinity constants puts some of these ligands among the most selective quadruplex DNA interactive agents reported to date.