Dihydrodiol dehydrogenase in drug resistance and sensitivity of human carcinomas.

We previously reported (UroOncology 1:165, 2001) cross-resistance and collateral-sensitivity to 2-chlorodeoxyadenosine (CldAdo) and fludarabine (FaraA), respectively, in a human renal cell carcinoma selected for resistance to 2'-deoxytubercidin (Caki-dTub). Insofar that these drugs generally demonstrate cross resistance rather than collateral sensitivity, we further examined the bases for this phenomenon. Both CldAdo and FaraA induce apoptosis, as the triphosphates, via binding to Apaf-1. In the presence of cytochrome c, this binding leads to activation of procaspase 9 to active caspase 9 that induces apoptosis through its activation of caspase 3. CldAdo and FaraA induced caspase 3 activities in wild type and Caki-dTub cell lines in a dose-dependent manner that paralleled the cross-resistance (CldAdo, 200-fold) or collateral sensitivity (FaraA, 20-fold) with regard to cell viability. The activation of caspase 3 was inhibited by the caspase 9 inhibitor, Z-LEHD-FMK, suggesting that both drugs act via the same pathway. By differential display and direct enzyme analysis, dihydrodiol dehydrogenase (DDH) was observed to be profoundly underexpressed in the Caki-dTub compared to wild-type Caki-1 cells. Stable transfection of the Caki-dTub cells with a vector encoding the enzyme led to partial reversal of the resistance to CldAdo. Resistance to cisplatin has recently been ascribed to overexpression of DDH in a human ovarian carcinoma cell line (Deng et al. in J Biol Chem 227:15035, 2002). It is tempting to speculate a mutation in the Apaf-1 nucleotide binding site that reduces (CldAdo) or increases (FaraA) toxicity in the Caki-dTub cells; however, the recent finding by others in a human ovarian carcinoma cell line suggests that DDH expression mediates the cross-resistance and perhaps, collateral-sensitivity.

Suicide gene therapy using E. coli beta-galactosidase.

PURPOSE: Suicide gene therapy offers the potential to increase the selective toxicity of antitumor agents by intratumoral expression of exogenous enzymes that convert nontoxic prodrugs to toxic products. The use of herpes simplex virus thymidine kinase with ganciclovir, and E. coli cytosine deaminase with 5-fluorocytosine are well-known examples of this approach. The purpose of this study was to investigate a novel suicide gene therapy using E. coli beta-galactosidase (beta-gal) as the prodrug-activating enzyme. Advantages of this approach include: (1) the ability to use prodrugs that are cleaved by beta-gal to agents that are known to possess activity against human solid tumors, and (2) the extensive experience gained with targeting beta-gal to specific tumors in experimental animals and in humans. METHODS: Two different structural types of anthracycline prodrugs, N-[4"-(beta- D-galactopyranosyl)-3"-nitrobenzyloxycarbonyl]daunomycin (Daun02) and N-[(4" R,S)-4"-ethoxy-4"-(1"'- O-beta- D-galactopyranosyl)butyl]daunorubicin (gal-DNC4) were investigated. The prodrugs were evaluated as substrates for beta-gal. Cytotoxicity studies of Daun02 were conducted against a murine tumor (Panc02), two human breast tumors (MCF-7 and T47D), and three human prostate tumors (PC3, DU145 and LNCAP) that had been transduced to express beta-gal. Antitumor studies of Daun02 were conducted against mouse tumor Panc02 xenografts implanted subcutaneously. RESULTS: Daun02 was a good substrate for beta-gal. By comparison, gal-DCN4 was a poor substrate. Except for PC3, the beta-gal-transduced tumors showed 3- to 60-fold increased sensitivity to Daun02 compared with mock-transduced control cells. Daunomycin was formed from Daun02 in tissue culture medium containing beta-gal-transduced cell lines but was not observed in the medium from mock-transduced controls. In vivo therapeutic studies of Daun02 against the Panc02 tumor in athymic mice showed no significant inhibition of tumor growth. Pharmacokinetic studies showed limited distribution of the prodrug beyond the vascular space. CONCLUSIONS: E. coli beta-gal may be useful as a prodrug-activating enzyme in suicide gene therapy and has the potential to increase the selective toxicity of conventional antitumor agents. Although this approach worked well against tumor cells in vitro, it was not effective against a xenograft model in vivo, apparently because of poor drug-tissue distribution.

Renal organic cation and nucleoside transport.

We previously reported that the rat organic cation transporter rOCT1 could transport the nucleoside analog deoxytubercidin (dTub) (Chen R, Nelson JA. Biochem Pharmacol 2000;60:215-9). The cationic form of dTub (dTub(+)) appeared to be the true substrate of rOCT1. We also reported that although rOCT2 is similar to rOCT1, it does not transport dTub at pH 7.4. In this study, we measured the K(m) and V(max) values of dTub(+) uptake at a reduced pH (pH 5.4) for both rOCT1 and rOCT2. The difference in substrate activity appears due, in large part, to a poor affinity of rOCT2 for dTub(+). The transport efficiency estimated by V(max)/K(m) values for rOCT2 was only 6% that of rOCT1. Chimeras constructed between rOCT1 and rOCT2 revealed that the difference in dTub binding lies within transmembrane domains 2-7. To evaluate the potential of OCT1 in the renal secretion of dTub, tissue distribution and urinary excretion of dTub in OCT1 knockout mice were measured. No significant difference was observed in renal elimination, plasma level, and tissue distribution of dTub between the knockout and the wild-type mice. Therefore, dTub is a good substrate for OCT1; however, OCT1 does not appear to be necessary for its renal secretion.