LY294002, an inhibitor of phosphatidylinositol-3-kinase, causes preferential induction of apoptosis in human multidrug resistant cells.

Protein kinase B (PKB), a kinase downstream of phosphatidylinositol 3-kinase (PI3-kinase) provides anti-apoptotic and survival signals via phosphorylation of various targets. Inhibiting PI3-kinase with a 12 h exposure to 10 microM LY294002 induces levels of apoptosis of 30.39+/-1.53% in the KB-V1 multidrug resistant (MDR) cell line compared to 4.54+/-1.00% in drug sensitive KB-3-1 cells (P<0.001). This occurred in conjunction with a preferential reduction in activated PKB in MDR cells. These results suggest the PI3-kinase/PKB signalling pathway is important for the survival of MDR cells and inhibition of this pathway results in the selective induction of apoptosis in MDR cells.

The preferential induction of apoptosis in multidrug-resistant KB cells by 5-fluorouracil.

It has been previously been shown that multidrug resistance may be associated with biochemical changes which increase the sensitivity of resistant cells to the induction of apoptosis by certain agents. We have shown here that 48 h exposure to 5-fluorouracil (5-FU) induces both a significantly greater proportion of apoptotic cells and much greater cleavage of the apoptosis-related protein poly-(ADP-ribose)-polymerase in the multidrug-resistant (MDR) carcinoma cell line, KB-A1, than in corresponding drug-sensitive control KB-3.1 cells. Exposure to 5-FU also reduced the level of the anti-apoptotic protein, protein kinase B, in the MDR cells, but not in the control cells.

Preferential killing of multidrug-resistant KB cells by inhibitors of glucosylceramide synthase.

This study has compared the preferential killing of three multidrug-resistant (MDR) KB cell lines, KB-C1, KB-A1 and KB-V1 by two inhibitors of glucosylceramide synthase, 1-phenyl-2-decanoylamino-3-morpholino-1-propanol (PDMP) and 1-phenyl-2-hexadecanoylamino-3-pyrrolidino-1-propanol (PPPP), to the killing produced by these compounds in the drug-sensitive cell line, KB-3-1. Both of the inhibitors caused much greater induction of apoptosis in each of the three MDR cell lines than in the drug-sensitive cell line, as judged by morphological assay and confirmed by poly-(ADP-ribose)-polymerase cleavage. The highest level of apoptosis was produced following 24-h exposure to 5 microM PPPP. This treatment produced 75.8 (+/- 7.1)%, 73.6 (+/- 9.8)% and 75.3 (+/- 6.4)% apoptotic cells in the three MDR cell lines respectively, compared to 19.0 (+/- 9.8)% in the drug-sensitive cell line. A reduction in glucosylceramide level following inhibitor treatment occurred in KB-3-1 cells as well as in the MDR cell lines, suggesting that the increased apoptotic response in the MDR cells reflected a different downstream response to changes in the levels of this lipid in these cells compared to that in the drug-sensitive cells. These results suggest that the manipulation of glucosylceramide levels may be a fruitful way of causing the preferential killing of MDR cells in vitro and possibly in vivo.

2-Deoxy-D-glucose preferentially kills multidrug-resistant human KB carcinoma cell lines by apoptosis.

The aim of this study was to determine the mechanism of cell death associated with the preferential killing of multidrug-resistant (MDR) cells by the glycolytic inhibitor 2-deoxy-D-glucose (2DG) in a range of MDR human KB carcinoma cell lines selected in different drugs. The D10 values for KB-V1, KB-C1 and KB-A1 (selected in vinblastine, colchicine and doxorubicin respectively) were 1.74, 1.04 and 0.31 mM, respectively, compared with 4.60 mM for the parental cell line (KB-3-1). The mechanism of cell death was identified as apoptosis, based on nuclear morphology, annexin V binding and poly(ADP-ribose) polymerase (PARP) cleavage. 2DG induced apoptosis in the three MDR cell lines in a dose- and time-dependent manner and did not induce necrosis. PARP cleavage was detected in KB-C1 cells within 2 h of exposure to 50 mM 2DG and slightly later in KB-A1 and KB-V1 cells. The relative levels of 2DG sensitivity did not correlate with the levels of multidrug resistance or with the reduced levels of the glucose transporter GLUT-1 in these cells. We speculate that a 2DG-stimulated apoptotic pathway in MDR KB cells differs from that in normal KB cells.

Protein kinases and multidrug resistance.

The role of protein kinases in the multidrug resistance phenotype of cancer cell lines is discussed with an emphasis on protein kinase C and protein kinase A. Evidence that P-glycoprotein is phosphorylated by these kinases is summarised and the relationship between P-glycoprotein phosphorylation and the multidrug-resistant phenotype discussed. Results showing that protein kinase C, particularly the alpha subspecies, is overexpressed in many MDR cell lines are described: this common but by no means universal finding seems to be drug- and cell line-dependent and in only in a few cases is there a direct correlation between protein kinase C activity and multidrug resistance. From co-immunoprecipitation results it is suggested that P-glycoprotein is a specific protein kinase C receptor, as well as being a substrate. Revertant experiments provide conflicting results as to a direct relationship between expression of P-glycoprotein and protein kinase C. Evidence that protein kinase A influences P-glycoprotein expression at the gene level is well documented and the mechanisms by which this occurs are becoming clarified. Results on the relationship between protein kinase C and multidrug resistance using many inhibitors and phorbol esters are difficult to interpret because such compounds bind to P-glycoprotein. In spite of huge effort, a direct involvement of protein kinase C in regulating multidrug resistance has not yet been firmly established. However, evidence that PKC regulates a Pgp-independent mechanism of drug resistance is accumulating.

The human KB multidrug-resistant cell line KB-C1 is hypersensitive to inhibitors of glycosylation.

We have previously shown that the hypersensitivity of KB MDR cells to 2-deoxy-D-glucose is associated with diminished plasma membrane GLUT-1 levels compared with parental, non-MDR cells. Here we report that MDR cells are hypersensitive to the N-linked glycosylation inhibitor tunicamycin, which induces partial inhibition of GLUT-1 glycosylation and diminishes GLUT-1-mediated transport. The effect of tunicamycin, which also enhances the hypersensitivity of MDR cells to 2-deoxy-D-glucose, could not be attributed to alterations in P-glycoprotein activity. The use of agents that act synergistically to diminish the level and activity of GLUT-1 in MDR cells may be of clinical potential.

Monensin and verapamil do not alter intracellular localisation of daunorubicin in multidrug resistant human KB cells.

The effects of monensin, verapamil and several inhibitors of membrane transport processes on the accumulation of [3H] daunorubicin by human KB-A1 cells have been investigated to determine the role of subcellular vesicular transport in the multidrug resistance phenotype. The Golgi inhibitor, brefeldin A, had no effect on drug accumulation, which suggests that vesicular transport is not a significant factor in drug resistance in these cells. KB-A1 cells were collaterally sensitive to both monensin and verapamil. Both of these compounds reduced drug efflux but did not alter subcellular distribution of daunorubicin, consistent with the view that monensin, like verapamil, acts directly on P-glycoprotein.

2-deoxy-D-glucose toxicity and transport in human multidrug-resistant KB carcinoma cell lines.

It is shown that a series of colchicine-selected multidrug-resistant (MDR) human KB carcinoma cell lines displayed increasing 2-deoxy-D-glucose collateral sensitivity, which correlated with increasing multidrug resistance. The relative resistance of MDR cell lines to 2-deoxy-D-glucose was reduced to 0.73 (KB-8-5), 0.3 (KB-8-5-11) and 0.2 (KB-C1) when compared with parental KB-3-1 (1.0). 2-Deoxy-D-glucose accumulation was found to be reduced in the MDR cell lines in a manner that correlated with 2-deoxy-D-glucose collateral sensitivity. At 30 min 2-deoxy-D-glucose accumulation was reduced to 0.61 (KB-8-5), 0.41 (KB-8-5-11) and 0.22 (KB-C1) relative to KB-3-1 uptake (1.0). The efflux of 2-deoxy-D-glucose was not significantly different between resistant and sensitive cell lines. Analysis of 2-deoxy-D-glucose uptake kinetics, by initial rate measurements, showed alterations in K(t) and J(max) for MDR when compared with KB-3-l cells. The levels of GLUT-1 facilitative transporter were found to be reduced significantly in the MDR cell lines in total cell homogenate and plasma membrane fractions by using Western blot analysis. Changes in the plasma membrane level of GLUT-1 correlated with 2-deoxy-D-glucose toxicity and uptake for MDR cell lines, where relative GLUT-i levels were reduced to 0.71 (KB-8-5), 0.43 (KB-8-5-1 1) and 0.27 (KB-Cl) relative to KB-31(1.0). It is concluded that the response of human KB MDR cells to 2-deoxy-D-glucose involved alterations in the level and activity of the facilitative glucose transporter, GLUT-1, in a manner that is associated with the degree of multidrug resistance.

Reduced daunomycin accumulation in drug-sensitive and multidrug-resistant human carcinoma KB cells following phorbol ester treatment: a potential role for protein kinase C in reducing drug influx.

We have investigated the role of protein kinase C (PKC) in the multidrug resistance (MDR) phenotype of human KB carcinoma cell lines. The PKC activator 12-O-tetradecanoylphorbol-13-acetate (TPA) reduced daunomycin accumulation in both drug-sensitive KB-3-1 and MDR KB-C1 cells in a time-dependent manner. The inactive phorbol ester 4 alpha TPA did not reduce daunomycin accumulation, and the PKC inhibitor, Ro 31-8220, reversed the TPA effect. TPA had no effect on daunomycin efflux and did not induce Pgp expression in KB-3-1 cells or alter Pgp levels in KB-C1 cells. Linear, short-term daunomycin accumulation was reduced by pretreatment with TPA, an effect that could be reversed by Ro 31-8220. The effects of TPA on PKC subspecies localisation and downregulation were also examined. TPA initially induced translocation of PKCs alpha and delta, and to a lesser extent, PKC epsilon to the membrane fraction; 8 h after TPA treatment, differential effects on downregulation of PKCs alpha and delta were observed between cell lines, although PKC epsilon was not reduced in either cell line. We therefore propose that the TPA-induced reduction in daunomycin accumulation in KB cells is due to a PKC-mediated process, which is maintained after depletion of certain PKC subspecies or is due to activation of downregulation insensitive PKC subspecies. These results suggest that PKC may regulate drug resistance by reducing drug influx in a Pgp-independent manner in KB cells. This may represent a mechanism of drug-resistance independent of, or in addition, to, Pgp-mediated drug efflux.

Gain and loss of hypersensitivity to resistance modifiers in multidrug resistant Chinese hamster ovary cells.

Some, but not all, multidrug resistant cell lines exhibit collateral hypersensitivity to resistance modifiers. We have examined the relationship between levels of P-glycoprotein expression and resistance modifier hypersensitivity in Chinese hamster ovary cell lines. We have defined a model system for the gain and loss of such sensitivity which indicates that its presence is not simply proportional to P-glycoprotein levels, but is acquired only above a certain level of P-glycoprotein expression. We also show that previously reported differences in such sensitivity between lines is not attributable to differences in genetic background of the cell lines used for selection of drug resistance.

Changes in the nature of P-glycoprotein gene transcripts during the development of multidrug resistance in CHO cells.

We have studied the nature of P-glycoprotein gene transcripts in multidrug resistant Chinese hamster ovary cell lines selected in vincristine. Using a novel, 35S based, reverse transcription polymerase chain reaction assay, we have shown that there is a considerable increase in the level of pgp2 gene transcripts, with levels rising to a hundred fold those observed in sensitive control cells. The level of pgp1 gene transcripts was raised five fold. The levels of pgp3 gene transcripts were close to the limits of detection by the assay, but were increased in the more highly resistant cell lines. An increase in the size of the P-glycoprotein gene transcript was also observed in some of the resistant cell lines.

Changes in protein kinase C subspecies protein expression and activity in a series of multidrug-resistant human KB carcinoma cell lines.

We have investigated the relationship between protein kinase C (PKC), levels of resistance and drug used for selection in a series of human KB carcinoma cell lines by comparing protein kinase C activity and PKC alpha, beta I, beta II, gamma, delta, epsilon, and zeta subspecies protein expression. PKC alpha protein expression was increased by 600% and 375% in KB-A1 and KB-C1 lines respectively over the parent KB-3-1 line; only KB-A1 cells showed increased PKC delta expression. Expression of other PKC subspecies was equal to that of KB-3-1 cells. There was considerable variation between the different KB cell lines in total cytosolic PKC activity, the KB-A1 and KB-C1 lines showing 400% and 350% increases respectively, KB-V1 and KB-8-5-11 about 180%, and KB-8-5 no increase relative to the parent KB-3-1 line. For calcium-independent PKC activity, the KB-C1 and KB-A1 lines only were increased over the KB-3-1 line. Immunoprecipitation with antisera to PKC subspecies confirmed that the increase in KB-A1 cytosolic total PKC activity was due largely to PKC alpha and partially to PKC delta. Membrane-associated PKC activity was increased by 500% and 350% in KB-A1 and KB-C1 lines respectively, by 250% and 270% in KB-V1 and KB-8-5-11, and not increased in KB-8-5 cells relative to the KB-3-1 cells. For KB-C1, KB-8-5-11, and KB-8-5 lines, which show decreasing resistance to colchicine, our results suggest a correlation between PKC and multidrug resistance in cells selected for resistance to this drug. There is no correlation between PKC and multidrug resistance for cells selected in different drugs. Our study therefore suggests that specific PKC subspecies are associated with the MDR phenotype of some KB cell lines, but that the extent of PKC involvement depends on the type of drug used for selection and its concentration.

Increase of cell viability and P-glycoprotein expression in Chinese hamster ovary cells after 24 hours' exposure to vincristine.

Cell viability and P-glycoprotein expression in Chinese hamster ovary cells have been measured following incubation in a range of vincristine concentrations for 24 h. An increase of 12% in cell viability was observed with increasing drug concentrations in a drug-sensitive cell line (E29) and a lesser increase of 5% in a multidrug-resistant cell line (VRA15). The increased cell viability corresponded to a rise of P-glycoprotein level. Slot blots showed that there was an increase in P-glycoprotein mRNA with all drug concentrations tested in E29, although there was only partial correlation of mRNA levels with P-glycoprotein levels viability.

Relationship between multidrug resistance, hypersensitivity to resistance modifiers and cell volume in Chinese hamster ovary cells.

We have previously shown that very high levels of hypersensitivity to several resistance modifiers are correlated with increasing multidrug resistance in a series of Chinese hamster ovary cell lines. We have now selected a new member of the series which is an exception to this correlation in that although it is almost twice as multidrug resistant as the cell line from which it was derived, it shows much less hypersensitivity to resistance modifiers. Level of resistance modifier hypersensitivity correlated with the level of reduction of verapamil accumulation in these cells, and with the density of P-glycoprotein, but since the selection of this cell line has involved a doubling of cell volume, it was not correlated with total amount of P-glycoprotein.

Reduced influx is a factor in accounting for reduced vincristine accumulation in certain verapamil-hypersensitive multidrug-resistant CHO cell lines.

The rates of accumulation, influx and efflux of vincristine have been examined in a series of multidrug-resistant Chinese hamster ovary cell lines which show exceptionally high levels of hypersensitivity (collateral sensitivity) to several resistance modifiers. The more highly resistant members of the series show significantly reduced levels of vincristine influx compared to the control cell line from which they were derived. It is possible that resistance modifier hypersensitivity and reduced vincristine influx may be due to a common change in membrane composition which has arisen during prolonged selection for vincristine resistance in these cell lines.

Amplification and expression of mdr genes and flanking sequences in verapamil hypersensitive hamster cell lines.

The properties of several multidrug resistant (MDR) Chinese hamster ovary (CHO) cell lines which are verapamil hypersensitive have been investigated, extending our earlier study of two such cell lines. It was observed that increasing levels of multidrug resistance are associated with increasing verapamil and nicardipine sensitivity, although the cell lines are not hypersensitive to cyclosporin A. Although there is appreciable amplification of the P-glycoprotein gene at higher levels of multidrug resistance/verapamil hypersensitivity, there is only very low or no amplification of five flanking genes, including the sorcin gene. Low levels of resistance (3-10 fold) appear to involve increased P-glycoprotein gene expression at the level of transcription. P-glycoprotein levels of the cell lines have been measured by flow cytometry using the monoclonal antibody C219, and there is a general correlation between P-glycoprotein overexpression, increased levels of the corresponding mRNA and the level of verapamil hypersensitivity.