Madeleine Castaing

Membrane permeation by multidrug-resistance-modulators and non-modulators: effects of hydrophobicity and electric charge.

This study was designed to test the hypothesis that lipophilic cationic drugs with only roughly similar structures mediate the reversal of multidrug‐resistance (MDR) by interacting with membrane phospholipids. The permeation properties of MDR‐modulators and non‐modulators were studied by quantifying their ability to induce the leakage of Sulphan blue through the membrane of negatively charged unilamellar liposomes.

Designing multidrug‐resistance modulators circumventing the reverse pH gradient in tumours

Multidrug‐resistant tumours often exhibit a reverse pH gradient (acid outside), as they have an acid extracellular pH (pHe) and a neutral alkaline intracellular pH (pHi). This study was designed to test the hypothesis that the ability of lipophilic drugs to mediate multidrug resistance (MDR) reversal by interacting with the membrane phospholipids may be correlated with pH in resistant tumours. The permeation properties of five MDR modulators were therefore studied at 37°C by quantifying their ability to induce the leakage of Sulfan blue through unilamellar anionic liposomes, over the range pH 6.5–7.7, and in the absence of any membrane potential (pHe = pHi). The dye leakage induced by two calcium blockers (diltiazem and verapamil) and two antiparasitic agents (thioacridine derivative and mepacrine) was found to significantly increase with the pH of the medium (P < 0.001), whereas that induced by a non‐ionic detergent (Triton X‐100) showed almost no pH‐dependent variations. This process was a cooperative one (0.8 < Hill coefficient < 8.5) and the permeation doses inducing 50% dye leakage (PD50) ranged from 1.6 to 36.0 mM. The permeation ability of the MDR modulators (log(1/PD50)) significantly increased with their octanol‐buffer distributions (logD) (slope = 0.35 ± 0.06; y intercept = 1.65 ± 0.14; P < 0.0001) and significantly decreased with their net electric charge (z) (slope = −0.48 ± 0.07; y intercept = 2.85 ± 0.08; P < 0.0001). A highly significant multiple correlation was found to exist between the variations of log(1/PD50) with those of logD and z (dlog(1/PD50)/dlogD = 0.21 ± 0.05; dlog(1/PD50)/dz = −0.34 ± 0.07; y intercept = 2.27 ± 0.17; P < 0.000001). The results provide evidence that in resistant tumours (acid pHe and neutral alkaline pHi), the MDR reversal might be enhanced by favourable drug‐membrane interactions if the modulators are designed in the form of highly lipophilic (logP ≅ 4) mono‐basic drugs with a near neutral pKa (pKa ≅ 7–8).

Effects of cholesterol on dye leakage induced by multidrug-resistance modulators from anionic liposomes.

Multidrug-resistance (MDR) in cancer cells is often associated with marked changes in the membrane cholesterol levels. To assess the cholesterol-dependence of MDR modulator efficiency in terms of the drug-membrane interactions, the ability of 5 MDR-modulators to induce the leakage of Sulphan blue through anionic liposomes was quantified at various mole fractions x(chol) of cholesterol (0-0.42). Depending on the electric charge of the drug, cholesterol modified to a large extent either the permeation dose inducing 50% dye leakage (PD(50)) or the co-operativity (h) of the permeation process. The PD(50) of Triton X-100 (non-ionic) and that of diltiazem and verapamil (mono-basic amines) varied only slightly (0.3 mM) with the cholesterol level, whereas the co-operativity increased by 1.9-2.7. On the reverse, the PD(50) of a thioacridine derivative and mepacrine (di-basic amines) increased by 4.8-7.5 mM in the cholesterol range investigated, whereas the co-operativity (h) increased slightly (0.2-0.7). In the permeation process, the rate-limiting character of the electric charge (z) of the drug is likely to be strengthened by high cholesterol levels. The results provide evidence that in resistant tumours exhibiting high cholesterol levels, the MDR might be reversed by favourable drug-membrane interactions if the modulators are designed in the form of highly lipophilic mono-basic drugs that counteract the effects of cholesterol on the membrane dipolar potential and membrane fluidity.

Synergy between verapamil and other multidrug -resistance modulators in model membranes

Various cationic lipophilic compounds can reverse the multidrug resistance of cancer cells. Possible interaction between these compounds, which are known as modulators, has been assessed by measuring leakage of Sulphan blue from anionic liposomes, induced both by verapamil alone and by verapamil in combination with diltiazem, quinine, thior idazine or clomipramine. An equation was derived to quantify the permeation doses and Hill coefficients of the drugs and mixtures between them by simultaneous fitting of the experimental data. The interaction was tested by two methods, the competition plot and the isobole method; both showed synergy between verapamil and each of diltiazem, qui nine and thioridazine. The dose factor of potentiation for verapamil determined within membranes was 4.0 ± 0.4 with diltiazem, 3.2 ± 0.4 with quinine and 2.4 ± 0.3 with thior idazine. The results suggest that the effectiveness of reversing multidrug resistance may be increased with modulators such as verapamil and diltiazem that have a much greater effect in combination than what would be expected from their effects when considered separately.

Multidrug resistance modulator interactions with neutral and anionic liposomes: membrane binding affinity and membrane perturbing activity

A variety of cationic lipophilic compounds (modulators) have been found to reverse the multidrug resistance of cancer cells. In order to determine the membrane perturbing efficacy and the binding affinity of such drugs in neutral and anionic liposomes, the leakage of Sulfan blue induced by five modulators bearing different electric charges was quantified using liposomes with and without phosphatidic acid (xEPA = 0 and 0.1), at four lipid concentrations. The binding isotherms were drawn up using the indirect method based on the dependency of the leakage rate on the modulator and the lipid concentrations. Upon inclusion of negatively charged lipids in the liposomes: (i) the binding of cationic drugs was favoured, except in a case where modulator aggregation occurred in the lipid phase; (ii) the drugs with a net electric charge greater than 1.1 displayed a greater enhancement in their potency to produce membrane perturbation; and (iii) the EPA effect on membrane permeation was due mainly to that on membrane perturbation (≥50%) and, to a lesser extent, to that on the binding affinity (≤50%). The present study provides evidence that drug‐membrane interactions are the result of a complex interplay between the structural and electrical characteristics of the drugs and those of the membranes.

Thermal Dependence of Multidrug‐resistant‐modulator Efficiency: a Study in Anionic Liposomes

This study was designed to test the hypothesis that there exists a correlation between the ability of lipophilic drugs to mediate the reversal of multidrug‐resistance (MDR) by interacting with the membrane phospholipids and the metabolic level in tissues. The permeation properties of five MDR‐modulators were studied by quantifying their ability to induce the leakage of Sulphan blue through unilamellar liposomes, over the temperature range 27–42°C.

POTASSIUM TRANSPORT IN OPOSSUM KIDNEY CELLS : EFFECTS OF NA-SELECTIVE AND K-SELECTIVE IONIZABLE CRYPTANDS, AND OF VALINOMYCIN, FCCP AND NYSTATIN

The effects of two ionizable cryptands, the Na-selective (221)C10 and the K-selective (222)C10, and of valinomycin, FCCP and nystatin on K+ fluxes in opossum kidney (OK) cells have been quantified. The Na,K-ATPase (ouabain-sensitive 86Rb influx) was stimulated by nystatin (> or = 20%), and inhibited by the other ionophores (50-80%), by barium (K-channel blocker) (61%) and by amiloride (Na entry blocker) (34%). The Vmax of the Na,K-ATPase phosphatase activity was unmodified by the ionophores, indicating the absence of direct interaction with the enzyme. The ATPi content was unmodified by the inhibitors and nystatin, but was lowered by (221)C10 (47%), (222)C10 (75%), valinomycin (72%) and FCCP (88%). Amiloride was found to partially remove the inhibition caused by (222)C10 (51%) and valinomycin (49%). Rb efflux was stimulated by nystatin (32%), unmodified by valinomycin, and was inhibited by (221)C10 (19%), (222)C10 (19%) and FCCP (10%). Barium (39%) and amiloride (32%) inhibited this efflux and, in their presence, the nystatin effect persisted, whereas that of the other ionophores vanished. At pH 6.4, the Rb efflux decreased by 14% of its value at pH 7.4, with no additional inhibition by cryptands. Cryptands are shown to inhibit the pH-sensitive K+-conductance, probably by inducing a K+-H+ exchange at the plasma membrane, and by uncoupling oxidative phosphorylation by inducing the entry of K+ and H+ (and possibly Ca2+) ions into the mitochondria.

Interactions between verapamil and neutral and acidic liposomes: effects of the ionic strength

Patients with cancer often develop major electrolyte disorders, which are aggravated by radiation therapy and chemotherapy and by the concomitant impairment of the renal function and the development of drug resistance. In addition, tumour cells have membranes with more negative charges than normal eukaryotic cells. This study was designed to test the hypothesis that the ability of the Ca(2+) blocker verapamil to mediate the reversal of multidrug resistance (MDR) by interacting with the membrane phospholipids may be correlated with the ionic strength and membrane surface potential in resistant tumours. The permeation properties of verapamil, which is the best-known MDR-modulator, were therefore studied by quantifying its ability to induce the leakage of carboxyfluorescein through unilamellar liposomes containing various mole fractions of phosphatidic acid (x(EPA)=0, 0.1 and 0.3), at four different ionic strengths (I=0.052, 0.124, 0.204 and 0.318 M). The dye leakage induced by verapamil varied greatly with I, depending on x(EPA). The permeation process was a co-operative one (1.3<Hill coefficient<3.5) and the permeation doses inducing 50% dye leakage (PD(50)) ranged between 0.2 and 1.8 mM. A highly significant multiple correlation was found to exist between the variations of log(1/PD(50)) with those of 1/ radical I and x(EPA) (dlog(1/PD(50))/d(1/ radical I)=0.15+/-0.01, dlog(1/PD(50))/dx(EPA)=2.07+/-0.08, y-intercept=2.46+/-0.03, P<0.000001). Kinetic studies on the permeation process showed that it involved two steps. The apparent rate constants of the slow and fast kinetic steps, which were driven by electrostatic and hydrophobic interactions, respectively, increased with the verapamil concentrations, depending on x(EPA). The results provide evidence that in resistant tumours (high negative membrane surface potential), the MDR reversal by verapamil might be enhanced by favourable drug-membrane interactions in patients with severe hypo-electrolytic (Na(+) and K(+)) disorders, whereas the MDR reversal might be reduced by unfavourable drug-membrane interactions in patients with severe hyper-electrolytic (Ca(2+), Na(+) and K(+)) disorders.

Na/K competitive transport selectivity of (221) C lo-cryptand: effects of pH and carrier concentration

The kinetics of the competitive transport of Na+ and K+ ions across the membrane of large unilamellar vesicles (LUV) were determined when transport was induced by (221)C10-cryptand, an ionizable mobile carrier. The experiments were performed at various pH values (7.7 and 8.7) and carrier concentrations (0.1, 0.5 and 1.0 microM) in order to quantify the effects of these parameters on the Na/K competitive transport selectivity of this mobile carrier. At any given pH and carrier concentration, the apparent affinity of (221)C10 for Na+ was higher and less dependent on the concentration of the other competing ion than that for K+. The Na/K competitive transport selectivity (SC(Na/K)) of (221)C10 increased linearly with the Na+ concentrations, decreased hyperbolically with increasing those of K+ and was independent of the pH and of the carrier concentration. In equimolecular ionic mixtures, this competitive selectivity amounted to about 1.5 and when the pH rose, the carrier selectivity for Na+ over K+ ions was enhanced by cation competition compared to transport of cations as unique substrates. Equations were established to describe the variations of the competitive transport selectivity (SC) of cryptands, and for comparison of their noncompetitive selectivity (SNC), with the ionic concentrations, the Michaelis parameters of the cations and the pH. The reaction order in Na+ (n(Na)) increased significantly with decreasing the pH and the K+ concentration. The results are discussed in terms of the structural, physico-chemical and electrical characteristics of carriers and complexes.

Transport of competing Na and K ions by (222) C10-cryptand, an ionizable mobile carrier: Effects of pH and temperature

The kinetics of the electroneutral exchange of competing sodium and potassium with protons across the membrane of large unilamellar vesicles (LUV) were determined at two pH values when transport was induced by the simultaneous presence of (222)C10-cryptand and FCCP (proton carrier) at various temperatures. The aim of the present work was to quantify the pH-dependent enthalpies of an ionizable mobile carrier affinities for competing alkali cations, and to focus on the effects of pH and temperature on the competitive transport selectivity of the carrier for K+ over Na+ ions. At any given temperature and pH, the apparent pH-dependent affinity of (222)C10 was higher for K+ than for Na+. The enthalpy of this affinity for K+ was significantly lower than that for Na+, whereas it varied similarly with the pH (delta H(KpHmK) = 32.8 and 37.0 kJ/mol, and delta H(KpHmNa) = 47.9 and 52.9 kJ/mol at pH 7.8 and 8.8, respectively). When using a kinetic model, the pH effect on these parameters was discriminated (delta H(KmK) = 37.9 kJ/mol and delta H(KmNa) = 53.9 kJ/mol). The pH-dependence of the delta H(KpHm) of the cations could therefore theoretically be shown to arise from the temperature-induced changes in the ionization of the buffer dissolved in the aqueous phases and of the amine groups of the binding cavity of the carrier. The K/Na competitive transport selectivity (Sc(K/Na)) of (222)C10 increased linearly with the K+ concentration. It decreased hyperbolically with increasing concentration of Na+ while being independent of pH at any given temperature. In equimolecular ionic mixtures, Sc(K/Na) varied from 2.2 to 3.0 when temperature rose from 20 degrees C to 35 degrees C (delta H(Sc(K/Na)) = 15.6 +/- 0.5 kJ/mol). The results are discussed in terms of the structural, physico-chemical and electrical characteristics of carriers and complexes.