Arlette Garnier-Suillerot

Inhibition of the P-glycoprotein- and multidrug resistance protein-mediated efflux of anthracyclines and calceinacetoxymethyl ester by PAK-104P

Multidrug resistance phenotype in mammalian cells is often correlated with overexpression of P-glycoprotein or Multidrug Resistance-Associated protein (MRP(1)). Both proteins are energy-dependent drug efflux pumps that efficiently reduce the intracellular accumulation and hence the cytotoxicity of many natural cytotoxins. Overexpression of these transporters by tumor cells is thought to be a significant factor in both intrinsic and acquired resistance to anticancer drugs. Consequently a great deal of interest is focused on identifying chemical agents that can either antagonise drug transport by these proteins or that can inhibit the proliferation of tumors cells despite the expression of these transporters. P-glycoprotein-mediated multidrug resistance is reversed by a variety of compounds, but surprisingly, few agents reverse the MRP(1)-mediated multidrug resistance. However, it has recently been shown that 2-[4-(diphenylmethyl)-1-piperazinyl]ethyl-5-(trans-4,6-dimethyl-1, 3, 2-dioxaphosphorinan-2-yl)-2, 6-dimethyl-4-(3-nitrophenyl)-3-pyridinecarboxylate P oxide (PAK-104P) was able to inhibit the P-glycoprotein and MRP(1)-mediated efflux of several compounds. Understanding of the interactions between transporters and multidrug resistance reversing agents is important in the design of more effective multidrug resistance modulators. We now examined the effect of PAK-104P on Pgp-and MRP1-mediated efflux of three anthracyclines, daunorubicin, pirarubicin, hydroxydoxorubicin and of calcein acetoxymethyl ester and calcein. Our data show that PAK-104P non-competitively inhibits the P-glycoprotein-mediated efflux of anthracycline derivatives and calcein acetoxymethyl ester with an inhibitory constant K(I)=0. 25+/-0.05 microM. PAK-104P also non-competitively inhibits the MRP(1)-mediated efflux of daunorubicin, pirarubicin, hydroxyrubicin, calcein acetoxymethyl ester and calcein. However, surprisingly, in this case the K(I) values obtained were very different ranging from 0.06 for hydroxyrubicin to 10 microM for calcein. These data strongly suggested the existence of two different mechanisms for the inhibition by PAK-104P of the MRP(1)-mediated efflux of molecules: a first mechanism, involving a low-affinity site for PAK-104P, and which would concern molecules such as calcein, cysteinyl leukotriene LCT(4) etc. whose efflux do not depend on glutathione. A second mechanism involving a high-affinity site for PAK-104P and which would concern molecules such as anthracyclines, calcein acetoxymethyl ester whose efflux depends on the presence of glutathione.

Analysis of drug transport kinetics in multidrug-resistant cells : Implications for drug action

Multidrug resistance (MDR) in model systems is known to be conferred by two different integral proteins--the 170-kDa P-glycoprotein (P-gp) and the 190-kDa multidrug resistance-associated protein (MRP1)--that pump drugs out of MDR cells. The intracellular level of a drug, which influences the drugs cytotoxic effect, is a function of the amount of drug transported inside the cell (influx) and the amount of drug expelled from the cell (efflux). One possible pharmacological approach to overcoming drug resistance is the use of specific inhibitors that enhance the cytotoxicity of known antineoplastic agents. Many compounds have been proven to be very efficient in inhibiting P-gp activity, but only some of them can inhibit MRP1. However, the clinical results obtained so far by this approach have been rather disappointing. The other likely approach is based on the design and synthesis of new non-cross-resistant drugs whose physicochemical properties favor the uptake of such drug by resistant cells. Our recent studies have shown that whereas the P-gp- and MRP1-mediated efflux of different anthracycline-based drugs may not differ considerably, their kinetics of uptake do. Thus, the high uptake of drug by cells may lead to concentrations at the cellular target site high enough to achieve the needed cytotoxicity against MDR cells. Therefore, increased drug lipophilicity might be one factor in improving drug cytotoxicity in MDR cells. In vitro studies have shown that idarubicin, an analogue of daunorub cin, is more effective than daunorubicin and doxorubicin against MDR tumor cell lines and that this increased effectiveness is related in part to the increased lipophilicity of idarubicin. Other studies have also confirmed the strong impact of lipophilicity on the uptake and retention of anthracyclines in MDR cells.

New findings in the study on the intercalation of bisdaunorubicin and its monomeric analogues with naked and nucleus DNA.

DNA is a target molecule for anthracycline anticancer drugs. We have used new anthracycline derivatives, bisdaunorubicin (WP631) and its monomeric analogues (WP700 serie), and look if there was a relation between the drug binding affinity to naked DNA and to cell nucleus in the cell with its cytotoxicity. Circular dichroism (CD) and fluorescence were used to follow the interaction of anthracycline derivatives with naked DNA and cell nuclei. WP631 interacts with DNA at two distinct stoichiometries, 6:1 and 3:1 base pair (bp)/WP631 molecule (3:1 and 1.5:1 per anthracycline rings). Monomeric daunorubicin (DNR) with its amino sugar N-bound to amino- and nitro-substituted benzyl moiety, representing p-xylenyl linker present in WP631 bisintercalator, is much more binding to DNA than DNR or WP631. These findings are supported by the study of drug binding by nuclei of K562 cells. Around 70% of WP700 intercalate to nucleus DNA in the steady-state, while only 45% of DNR intercalate DNA in the cell. The binding of WP631 by K562 cells is even less effective ( approximately 20%). WP 700 compounds, which are very similar to each other in their binding to DNA, self-association and cell accumulation, differ very distinctly in their cytotoxicity power. The most effective compounds are amino-benzyl derivatives of WP 700 series. The nitro-benzyl compounds have very low toxicity, even if they bind to DNA with similar power with that of the amino derivatives. The comparison of the all data clearly indicates no relation between cytotoxicity of the drug and its ability to intercalate DNA.

Kinetics of glutathione and daunorubicin efflux from multidrug resistance protein overexpressing small-cell lung cancer cells

The present study examined how the multidrug resistance protein (MRP1), which is an ATP-dependent anionic conjugate transporter, also mediates the transport of reduced glutathione (GSH) and the co-transport of the cationic drug, daunorubicin, with GSH in living GLC4/Adr cells. To obtain information on the affinity of GSH for the multidrug resistance protein in GLC4/Adr cells, we investigated the GSH concentration dependence of the ATP-dependent GSH efflux. The intracellular GSH concentration was modulated by preincubation of the cells with 25 microM buthionine sulfoximine, an inhibitor of GSH synthetase, for 0-24 h. The transport of GSH was related to the intracellular GSH concentration up to approximately 5 mM and then plateaued. Fitting of the obtained data according to the Michaelis-Menten equation revealed a Km of 3.4+/-1.4 mM and a Vmax of 1.5+/-0.2x10(-18) mol/cell/s. The ATP-dependent transport of GSH was inhibited by 3-([[3-(2-[7-chloro-2-quinolinyl]ethenyl)phenyl]-[(3-dimethylamino-3-oxopropyl)-thio]-methyl]thio)propanoic acid (MK571), with 50% inhibition being obtained with 1.4 microM MK571. We investigated the GSH concentration dependence of the MRP1-mediated ATP-dependent transport of daunorubicin under conditions where the transport of daunorubicin became saturated. The daunorubicin transport was related to the intracellular GSH concentration up to approximately 5 mM and then plateaued. We were therefore in the situation where GSH acted as an activator: its presence was necessary for the binding and transport of daunorubicin by MRP1. However, GSH was also transported by the multidrug resistance protein. The concentration of GSH that gave half the maximal rate of daunorubicin efflux was 2.1+/-0.8 mM, very similar to the Km value obtained for GSH. In conclusion, the rate of daunorubicin efflux, under conditions where the transport of daunorubicin became saturated, and the rate of GSH efflux determined at any intracellular concentration of GSH were very similar, yielding a 1:1 stoichiometry with respect to GSH and daunorubicin transport. These results support a model in which daunorubicin is co-transported with GSH.

Permeability of lipid bilayer to anthracycline derivatives. Role of the bilayer composition and of the temperature

The uptake of three anthracycline derivatives: doxorubicin, daunorubicin and pirarubicin, into large unilamellar vesicles (LUV) in response to a driving force provided by DNA encapsulated inside the LUV has been investigated as a function of the temperature and of the bilayers lipid composition. The kinetics of the decay of the anthracycline fluorescence in the presence of DNA-containing liposome was used to follow the diffusion of the drug through the membrane. For the three drugs, the permeability coefficient of the neutral form of the drug (P0) decreases as the amount of negatively charged phospholipid in the bilayers increases. This can be explained by the fact that the kinetics of passive diffusion of the drugs depends on the amount of neutral form embedded in the polar head group region, which decreases as the quantity of negatively charged phospholipids increases. P0 also decreases as the amount of cholesterol, that makes the bilayer more rigid, increases. The activation energies, Ea, for the passage of the neutral form of these anthracyclines through the bilayers lie within 100 +/- 15 kJ x ml-1, except for pirarubicin and doxorubicin through anionic phospholipid-rich membranes (Ea = 57 kJ x mol-1) and cholesterol-rich membranes (Ea = 167 kJ x mol-1).

CORRELATION BETWEEN THE KINETICS OF ANTHRACYCLINE UPTAKE AND THE RESISTANCE FACTOR IN CANCER CELLS EXPRESSING THE MULTIDRUG RESISTANCE PROTEIN OR THE P-GLYCOPROTEIN

Multidrug resistance (MDR) in model systems is known to be conferred by two different integral proteins, the 170-kDa P-glycoprotein (Pgp) and the 190-kDa multidrug resistance-associated protein (MRP1). One possible pharmacological approach to overcome drug resistance is the use of specific inhibitors, which enhance the cytotoxicity of known antineoplastic agents. However, while many compounds have been proven to be very efficient in inhibiting Pgp activity only some of them are able to inhibit MRP1. The other likely approach is based on the design and synthesis of new non-cross-resistant drugs with physicochemical properties favoring the uptake of the drug by the resistant cells. The intracellular drug retention influences its cytotoxic effect. The level of the intracellular drug content is a function of the amount of drug transported inside the cell (influx) and the amount of drug expelled from the cell (efflux). In this work, the kinetics of drug uptake and the kinetics of active efflux of several anthracycline derivatives in both Pgp expressing K562/Adr cells and MRP1 expressing GLC4/Adr cells was determined. Our data have shown that in both cell lines there is no correlation between the resistance factor and the kinetics of drug efflux by these pumping systems. However, a very good correlation between the resistance factor and the kinetics of drug uptake has been established in both cell lines: the resistance factor decreases when the kinetics of drug uptake increases. This work has clearly shown that when the rate of transmembrane transport of anthracycline is high enough, the efflux mediated by the protein transporter is not able to pace with it. The protein transporter essentially operates in a futile cycle and the resistance factor is tending to one. It does not mean, however, that when the resistance factor is close to one the anthracycline is not transported by the pump.

How Fe3+ binds anthracycline antitumour compounds: The myth and the reality of a chemical sphinx

The interaction of Fe3+ with several anthracycline antitumour antibiotics has been reinvestigated. Absorption and circular dichroism (CD) measurements were carried out (i) in aqueous solution and (ii) in semi-aqueous MeOH to avoid the stacking of the anthracycline molecules. The Fe3+ binding to anthracycline was dependent on the metal-to-ligand molar ratio, antibiotic concentration, ionic strength, and pH. The formation of two major Fe3(+)-anthracycline complexes, I and II, was observed for all the drugs. These species differed in their coordination modes to the anthracycline ligands. Complex I was a monomeric species, where Fe3+ was bound to the anthracycline through the {C(11)-O-; C(12) = O} chelating site. In complex II, Fe3+ was also bound through the {C(5) = O; C(6)-O-} coordination site. Thus, the antibiotic ligand was acting as a bridge between two metal ions, forming oligomeric (or polymeric) structures. The different degree of association of the anthracyclines could be responsible for the reactivity of the metal ion. In fact, complexes I and II could constitute mononuclear, binuclear or polynuclear Fe3+ species depending on the competitive kinetics of both coordination and hydrolysis of the metal ion.

Physicochemical studies of the iron(III)-carminomycin complex and evidence of the lack of stimulated superoxide production by NADH deydrogenase

Fe(III) complex of an antitumoral antibiotic carminomycin has been studied. Using potentiometric and spectroscopic measurements we have shown that carminomycin forms with Fe(III) a well-defined species in which three molecules of drug are chelated to one Fe(III) ion. This occurs with the release of one proton per molecule of drug. Magnetic susceptibility measurements suggest that six oxygen atoms are bound to iron. The stability constant is 3 X 10(34). The in vitro inhibition of P 388 leukemia cell growth by this complex compares with that of the free drug. This complex, unlike the free drug, does not catalyze the flow of electrons from NADH to molecular oxygen through NADH dehydrogenase.

The MRP1-mediated effluxes of arsenic and antimony do not require arsenic-glutathione and antimony-glutathione complex formation.

Arsenic trioxide is an effective treatment for acute promyelocytic leukemia, but resistance to metalloïd salts is found in humans. Using atomic absorption spectroscopy, we have measured the rate of uptake of arsenic trioxide and of antimony tartrate in GLC4 and GLC4/ADR cells overexpressing MRP1 and the rate of their MRP1-mediated effluxes as a function of the intracellular GSH concentration. In sensitive cells, after 1 h, a pseudosteady state is reached where intra- and extracellular concentrations of metalloid are the same. This precludes the formation, at short term, of complexes between arsenic or antimony with GSH. In resistant cells reduced intracellular accumulation of arsenic (or antimony), reflecting an increased rate of arsenic (or antimony) efflux from the cells, is observed. No efflux of the metalloid is observed in GSH depleted cells. The two metalloïds and GSH are pumped out by MRP1 with the same efficiency. Moreover for the three compounds 50% of the efflux is inhibited by 2 μM MK571. This led us to suggest that As- and Sb-containing species could be cotransported with GSH.

Preparation and cytotoxicity of cisplatin-containing liposomes

We encapsulated cisplatin into stealth pH-sensitive liposomes and studied their stability, cytotoxicity and accumulation in a human small-cell lung carcinoma cell line (GLC4) and its resistant subline (GLC4/CDDP). Since reduced cellular drug accumulation has been shown to be the main mechanism responsible for resistance in the GLC4/CDDP subline, we evaluated the ability of this new delivery system to improve cellular uptake. The liposomes were composed of dioleoylphosphatidylethanolamine (DOPE), cholesteryl hemisuccinate (CHEMS), and distearoylphosphatidylethanolamine-polyethyleneglycol 2000 (DSPE-PEG2000) and were characterized by determining the encapsulation percentage as a function of lipid concentration. Among the different formulations, DOPE/CHEMS/DSPE-PEG liposomes (lipid concentration equal to 40 mM) encapsulated cisplatin more efficiently than other concentrations of liposomes (about 20.0%, mean diameter of 174 nm). These liposomes presented good stability in mouse plasma which was obtained using a 0.24-M EDTA solution (70% cisplatin was retained inside the liposomes after 30 min of incubation). Concerning cytotoxic effects, they are more effective (1.34-fold) than free cisplatin for growth inhibition of the human lung cancer cell line A549. The study of cytotoxicity to GLC4 and GLC4/CDDP cell lines showed similar IC50 values (approximately 1.4 microM), i.e., cisplatin-resistant cells were sensitive to this cisplatin formulation. Platinum accumulation in both sensitive and resistant cell lines followed the same pattern, i.e., approximately the same intracellular platinum concentration (4.0 x 10-17 mol/cell) yielded the same cytotoxic effect. These results indicate that long-circulating pH-sensitive liposomes, also termed as stealth pH-sensitive liposomes, may present a promising delivery system for cisplatin-based cancer treatment. This liposome system proved to be able to circumvent the cisplatin resistance, whereas it was not observed when using non-long-circulating liposomes composed of phosphatidylcholine, phosphatidylserine, and cholesterol.