Cellular localization domains of a rabbit and a human carboxylesterase: influence on irinotecan (CPT-11) metabolism by the rabbit enzyme.

Enzyme activation of prodrugs to improve the therapeutic index of specific anticancer agents is an attractive alternative to current chemotherapy regimens. This study addresses the potential for activating irinotecan (CPT-11) with recombinant carboxylesterases (CEs). CEs are a ubiquitous class of enzymes thought to be involved in the detoxification of xenobiotics. Their primary amino acid sequence indicates that these proteins should be localized to the endoplasmic reticulum. By PCR-mediated mutagenesis of a rabbit liver and a human alveolar macrophage CE cDNA, expression in Cos7 cells, and subsequent immunohistochemical localization, we have determined that an 18-amino acid NH2-terminal hydrophobic signal peptide is responsible for the localization of these proteins to the endoplasmic reticulum. By similar approaches, we have demonstrated that the COOH-terminal amino acids HIEL prevent secretion of the proteins from the cell. Enzymatic activity was lost by removing the NH2-terminal domain; however, active enzyme could be detected in the culture media of cells expressing the COOH-terminally truncated proteins. Secretion of CEs lacking the six COOH-terminal amino acids could be prevented with brefeldin A, confirming that these truncated enzymes were processed and released from cells by endoplasmic reticulum-mediated exocytosis. Double-truncation mutant enzymes lacking both NH2- and COOH-terminal sequences demonstrated immunostaining patterns similar to those of the NH2-terminally truncated proteins and also lacked CE activity. In all cases, metabolism of the classic esterase substrate o-nitrophenyl acetate predicted the sensitivity of cells expressing the rabbit CE to the anticancer agent CPT-11. In addition, the secreted enzyme sensitized Cos7 cells to this drug, indicating that protein association with a lipid bilayer is not required for substrate metabolism.

In situ subcellular localization of epitope-tagged human and rabbit carboxylesterases.

Carboxylesterases are a ubiquitous class of enzymes thought to be involved in xenobiotic metabolism and detoxification. Primary amino acid sequence data suggest that these proteins localize to the endoplasmic reticulum. However, since this family of proteins is highly homologous, the generation of specific reagents to monitor expression and subcellular localization has been unsuccessful. To accomplish in situ detection of a human alveolar macrophage carboxylesterase and a rabbit liver carboxylesterase, we constructed plasmids that expressed recombinant proteins containing an 11 amino acid influenza hemagglutinin tag near the C-terminus. These proteins retained carboxylesterase activity as determined by the conversion of o-nitrophenol acetate to o-nitrophenol. Following transfection of plasmids encoding these proteins into mammalian cells, cells were analyzed by both fluorescence and electron microscopy. The tagged enzymes were localized to the endoplasmic reticulum of both Cos7 monkey kidney cells and Rh30 human rhabdomyosarcoma cells. No tagged protein was detectable in the culture media. Hence, epitope tagging allowed the analysis of expression and localization of specific carboxylesterases. The methods described in this paper are, therefore, applicable to any protein, including those that are highly homologous to other candidate molecules.

Intermittent exposure of medulloblastoma cells to topotecan produces growth inhibition equivalent to continuous exposure.

Camptothecin analogues such as topotecan increase the number of covalent topoisomerase I-DNA complexes, which, in turn, have been proposed to initiate apoptosis. If induction of apoptosis by the camptothecins is, in fact, dependent on the formation of topoisomerase I-DNA complexes, then it would be of clinical relevance to identify schedules of exposure to the camptothecins that maximize the formation of these complexes but minimize the total amount of the drug administered. The time and dose dependence of topoisomerase I-DNA complex formation was determined by incubating Daoy pediatric medulloblastoma cells in vitro with topotecan at concentrations equivalent to those achievable in the plasma clinically (10, 50, or 200 nM) and measuring the number of complexes present in cells incubated for 15 min to 48 h with the drug. Regardless of the concentration of topotecan used, covalent topoisomerase I-DNA complexes were maximal within 15 min following addition of the lactone form of topotecan to the tissue culture medium. After 2 h of exposure to topotecan, complexes had decreased from maximum to approximately half of that value. Few, if any, complexes were detectable with topotecan incubations of 24-48 h. Growth inhibition studies showed that the IC50s of topotecan for the Daoy cell line (2.2 x 10(-9) M) and also for a second pediatric medulloblastoma cell line, SJ-Med3 (3.6 x 10(-9) M), exposed to topotecan 8 h daily for 5 days or continuous exposure were equivalent. The decrease in topoisomerase I-DNA complexes between 15 min and 1 h was consistent with a pH-dependent re-equilibration of topotecan to the less active hydroxyacid form of the drug. The decrease in complexes after a 2-48-h incubation with the drug was attributable neither to biological inactivation of topotecan as shown by sequential growth inhibition studies nor to a decrease in amount of topoisomerase I in the drug-treated cells. Indirect immunofluorescence labeling of topoisomerase I in Daoy cells incubated for 48 h with 10 nM topotecan showed a redistribution of nucleolar topoisomerase I. We are currently evaluating the antitumor effect of intermittent repetitive exposures to topotecan in mice bearing Daoy cells as a xenograft. The clinical utility of each effective schedule of exposure will depend on whether the therapeutic index of repetitive intermittent exposure to the drug is more or less favorable than the therapeutic index of continuous exposure.

Altered DNA topoisomerase II in multidrug resistance.

The characteristic feature of multidrug resistance (MDR) associated with drugs that interact with DNA topoisomerase II (topo II) is alterations in topo II activity or amount (at-MDR). We have characterized the at-MDR phenotype in human leukemic CEM cells selected for resistance to the topo II inhibitor, VM-26. Compared to drug-sensitive cells, the key findings are that at-MDR cells exhibit (i) decreased topo II activity; (ii) decreased drug sensitivity, activity and amount of nuclear matrix topo II; (iii) increased ATP requirement of topo II; (iv) a single base mutation in topo II resulting in a change of Arg to Gln at position 449, at the start of the motif B/nucleotide binding site; and (v) decreased topo II phosphorylation, suggesting decreased kinase or increased phosphatase activities. Recent results using single-stranded conformational polymorphism analysis reveals the presence of a mutation in the motif B/nucleotide binding site of the topo II alpha gene in CEM at-MDR cells and in another leukemic cell line selected for resistance to m-AMSA. Finally, we have observed marked changes in the nuclear distribution of topo II in cells treated with anti-topo II drugs and have also found these changes to be attenuated in drug-resistant cells. We postulate that traditional inhibitors of topo II alter the equilibrium of the strand-passing reaction such that the number of enzyme-DNA covalent complexes increases. We further suggest that when the enzyme is bound to DNA it is protected from proteolysis, thus allowing more topo II molecules to be detected.(ABSTRACT TRUNCATED AT 250 WORDS)

DNA topoisomerase II immunostaining in human leukemia and rhabdomyosarcoma cell lines and their responses to topoisomerase II inhibitors.

DNA topoisomerase II is an enzyme that affects nuclear structure and function and is the target of a number of anticancer drugs in clinical use, including teniposide (VM-26). We have used our polyclonal antisera that recognize both the M(r) 170,000 and 180,000 forms of topoisomerase II to examine the nuclear distribution of topoisomerase II in cytospin preparations of drug-sensitive (CEM) and VM-26-resistant (CEM/VM-1 and CEM/VM-1-5) human leukemic lymphoblasts. We have also examined the nuclear distribution of topoisomerase II in monolayer cultures of a human rhabdomyosarcoma (Rh30) cell line. In the absence of drug, we observed a focal "patchy" staining of nuclear topoisomerase II in all cell lines, that was especially notable in the lymphoblastic cells. Treatment of CEM and Rh30 cells with VM-26 under conditions that increase the number of covalent topoisomerase II-DNA complexes increased both the intensity and the homogeneity of nuclear topoisomerase II staining in a subpopulation of cells; focal staining was less evident after treatment with drug. These responses were roughly proportional to the concentration of VM-26 used and required only brief (approximately 25-min) incubation with drug. We also found that treatment of CEM cells with 4'-(9-acridinylamino)methanesulfon-m-anisidide similarly increased the intensity and homogeneity of nuclear topoisomerase II immunostaining. In contrast, 4'-(9-acridinylamino)methanesulfon-o-anisidide and 1-beta-D-arabinofuranosylcytosine, agents that do not inhibit topoisomerase II, did not produce this effect. Finally, the VM-26-mediated alteration in topoisomerase II staining intensity and distribution was attenuated in proportion to the degree of VM-26 resistance in the CEM/VM-1 and CEM/VM-1-5 sublines. These results appear to be related to the ability of the drug to stabilize DNA-topoisomerase covalent ("cleavable") complexes in intact cells. Our findings indicate that anti-topoisomerase II drugs, such as VM-26, have profound effects on the ability to detect topoisomerase II in the nucleus and provide a novel way of examining drug-stabilized DNA topoisomerase II complexes in intact single tumor cells.

Genetic characterization of the multidrug-resistant phenotype of VM-26-resistant human leukemic cells.

Our human T-cell leukemia line, CEM/VM-1, selected for resistance to VM-26 (teniposide), is cross-resistant to several drugs that interact with topoisomerase II, including VP-16 (etoposide), 4'-(9-acridinylamino)methanesulphon-m-anisidide, daunorubicin, and mitoxantrone. However, in contrast to cell lines exhibiting multidrug resistance (MDR) associated with overexpression of P-glycoprotein, this line is not cross-resistant to the Vinca alkaloids, is not impaired in drug accumulation, and does not overexpress the mdrl gene (Cancer Res., 47: 1297, 5455, 1987). More recently we found that nuclear extracts of these cells exhibit decreased topoisomerase II catalytic and cleavage activity, compared to the drug-sensitive line (Biochemistry, 1988). These results suggest that an alteration in topoisomerase II or a modulator of this enzyme may be responsible for this altered topoisomerase II-form of multidrug resistance (at-MDR). In the present work, we studied the somatic cell genetics of at-MDR. We produced hybrid cell lines by polyethylene glycol-mediated fusion of the CEM/VM-1 line with a hypoxanthine-guanine phosphoribosyl transferase-deficient, ouabain-resistant CEM line (CEM.AG1.OU1.5) that exhibits VM-26 sensitivity. Ten of the hybrid lines that grew in selective medium were randomly chosen for expansion and four were analyzed for both DNA content by flow cytometry and VM-26 sensitivity in a 72-h growth inhibition assay. The hybrid lines all contained approximately 2x DNA compared to unfused controls, indicating that the fusions were successful. The IC50 for VM-26 in 3 of the 4 lines was the same as that of the sensitive controls, ranging from 4.7 to 7.4 x 10(-8) M, and another was 76 x 10(-8) M. These data indicate that drug sensitivity was reconstituted by the hybridization procedure. By comparison, the VM-26 IC50 values in the CEM/VM-1 cells and CEM/VM-1 x CEM/VM-1 control "fusions" were 360 and 750 x 10(-8) M, respectively. To determine whether a topoisomerase II-mediated function was reconstituted in the hybrids, we measured drug-stimulated DNA cleavage ("cleavable complex formation"). Using 32P-labeled pBR322 DNA as substrate with nuclear extracts from drug sensitive cells, 100 microM VM-26 maximally stimulated DNA cleavage by approximately 11-fold compared to no-drug controls.(ABSTRACT TRUNCATED AT 400 WORDS)

Pharmacological, molecular, and cytogenetic analysis of "atypical" multidrug-resistant human leukemic cells.

We previously described the cross-resistance patterns and cellular pharmacology of a human leukemic cell line, CEM/VM-1, selected for resistance to the epipodophyllotoxin teniposide (M. K. Danks et al., Cancer Res., 47: 1297-1301, 1987). Compared to CEM/VLB100, which is a well characterized "classic" multidrug-resistant (MDR) cell line, the CEM/VM-1 cells display "atypical" multidrug resistance (at-MDR) in that they are cross-resistant to a wide variety of natural product antitumor drugs, except the Vinca alkaloids, and they are not impaired in their ability to accumulate radiolabeled epipodophyllotoxin. We have extended our characterization of this at-MDR cell line in the present study. In comparison to CEM/VLB100 cells, we found that CEM/VM-1 cells are not cross-resistant to either actinomycin D or colchicine. Verapamil and chloroquine, which enhance the cytotoxicity of vinblastine in CEM/VLB100 cells, had little or no ability to do so in the CEM/VM-1 cells. Membrane vesicles of the two resistant sublines were examined for overexpression of the MDR-associated plasma membrane protein (P-glycoprotein, Mr 170,000 protein, or 180,000 glycoprotein) by photoaffinity labeling with the vinblastine analogue N-(p-azido[3-125I]salicyl)-N'-beta-aminoethylvindesine. We were unable to visualize the MDR-associated protein in the CEM/VM-1 membranes with this photoaffinity probe under conditions in which the P-glycoprotein was readily seen in the membranes of CEM/VLB100 cells. Furthermore, no hybridization of the pMDR1 complementary DNA was seen in slot-blot analyses of the RNA from at-MDR cells, indicating that the mdr gene coding for P-glycoprotein is not overexpressed as is the case in the classic MDR cells. However, cytogenetic analysis indicated that the CEM/VM-1 cells contained an abnormally banded region on chromosome 13q, suggesting that a gene other than mdr may be amplified in these cells. Thus, despite the two cell lines having approximately equal degrees of resistance to epipodophyllotoxins, our data indicate that the mechanism(s) responsible for at-MDR is different from that for classic, P-glycoprotein-associated MDR.